Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.zaISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012 287
* To whom all correspondence should be addressed. +27 21 959-6459; fax: +27 21 959-6117; e-mail: [email protected] Received 7 August 2009; accepted in revised form 2 April 2012.
Sensitivity study of reduced models of the activated sludge process, for the purposes of parameter estimation and process optimisation: Benchmark process with ASM1
and UCT reduced biological models
S du Plessis and R Tzoneva*Department of Electrical Engineering, Cape Peninsula University of Technology, Cape Town, South Africa
Abstract
The problem of derivation and calculation of sensitivity functions for all parameters of the mass balance reduced model of the COST benchmark activated sludge plant is formulated and solved. The sensitivity functions, equations and augmented sensitivity state space models are derived for the cases of ASM1 and UCT reduced biological models. Matlab software for sensitivity function calculation and sensitivity model simulation is developed. The results are described and discussed. The behaviour of the sensitivity functions is used to determine which parameters of the reduced model need to be estimated in order to fit the reduced model behaviour to the real data for the process behaviour.
Keywords: wastewater treatment, activated sludge process, reduced model, model parameters, sensitivity function, Matlab simulation
Introduction
The problem of effective and optimal control of wastewater treatment plants has become very important in recent years due to increased populations and requirements for the quality of the effluent. The activated sludge process (ASP) is a waste-water treatment process characterised by complex nonlinear dynamics, a large number of variables, and a lack of sensors for real-time measurement of many of these variables. The above process characteristics require real-time control design and implementation strategies to be developed in order to achieve process operation which is compliant with the international standards for effluent quality. Modern optimal control design and implementation for the activated sludge process demands extensive insight into the plant’s operation, clear objectives, and knowledge about process dynamics described by an appro-priate mathematical model. Modelling is the most critical phase in the solution of any control problem because nearly all control techniques require knowledge of the dynamics of the system before control design can be attempted. This means that the primary task of any modern control design is to construct and identify a model for the system which is to be controlled.
Mathematical models of the wastewater treatment pro-cesses were developed using the first principles of conservation of mass and energy (Copp, 2002), on the basis of developed biological models, such as the first ones described by Dold et al. (1980), Henze et al. (1987) and Wentzel et al. (1992). The obtained mass-balance models describe the process-technology structure and biological reactions taking part in this structure.
Examples of such models are the COST benchmark model (Copp, 2002) and the University of Cape Town (UCT) model (Ekama and Marais, 1979; Ekama et al., 1984). The most used biological models are the IAWQ1 (Henze et al., 1987) – the model of the International Association for Water Quality called Activated Sludge Model 1 (ASM1), and the UCT model – the model developed at the UCT Department of Civil Engineering (Dold et al., 1980).
The full ASM1 and especially the UCT model present major problems in terms of their direct use for real-time simu-lation and control design purposes, because of their complexity, limitations and drawbacks, as follows:• Large number of model variables• Complex dependencies and interconnections between the
biological variables• Different time scales for the process dynamics• The control actions for the process are not included in an
explicit way in the model equations• Many variables are difficult to measure• Many model kinetic and stoichiometric parameters are dif-
ficult to determine and have uncertain values
A way to overcome these difficulties would be to simplify thecomplex model by developing reduced biological andmass-balance models with a small number of variables, while still maintaining the same characteristics as these of the original full model (Pearson, 2003). Kinetic parameters of this reduced model can be determined by development and applica-tion of different methods for parameter estimation (Holmberg and Ranta, 1982; Jeppson, 1996; Halevi et al., 1997; Noykova and Gyllenberg, 2000).
Solution of the problem of parameter estimation is based on the given mathematical model in which the parameters are not known, and on data for the process variables obtained by measurement. Many of the biological nutrient removal models are not identifiable since they have many more parameters than
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.za
ISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012288
feasible measurements (Pertev et al., 1997; Varma et al., 1999). In this case the problem can be solved if the influence of the parameters on model behaviour is studied and the non-influen-tial parameters are considered to have zero or constant known (nominal) values. The influential parameters can then be estimated on the basis of data from the measurements (Varma et al., 1999). The goal of the paper is to develop a mathematical and computational tool for determining which parameters can be accepted as known and which need to be estimated on the basis of the reduced model of the COST benchmark process (Du Plessis, 2009).
A sensitivity study of the model variables towards the model parameters provides a tool to identify the influential parameters. The sensitivity study of the dynamic model set of equations examines the changes in the model variables in response to changes in the model parameters (Smets et al., 2002; Mussafiet et al., 2002; De Pauw and Vanrolleghem, 2003). Parameters resulting in large values for the sensitivity function have to be estimated (Sato and Ohmori, 2002; Louries et al., 2008). If some parameters with small values for the sensi-tivity function could be found, the problem of parameter esti-mation could be simplified by considering these parameters as known (Holmberg and Ranta, 1982; Noykova and Gillenberg, 2000).
There are different mathematical methods for parametric sensitivity analysis, such as finite difference, direct, the Green’s function, the polynomial approximation, and so on. Most previ-ous studies have used the finite difference method (Smets et al., 2002; Plazl et al., 1999; Mussafiet at al., 2002; Varma et al., 1999; Pertev et al., 1997). This method is calculation-intensive and gives a local estimation of the sensitivity function for a given parameter. The global sensitivity method simultane-ously calculates the sensitivity functions for a large number of parameters and for a large variety of parameters. This method allows better analysis of the parameter’s influence as it provides full information about each sensitivity function. The direct method is applied for the mass balance model of the COST benchmark plant (Copp, 2002), based on ASM1 and UCT reduced biological models, developed in Du Plessis (2009). The mass balance models are used for derivation of the sensitivity functions of the model variables towards all kinetic and stoi-chiometric model parameters. As a result, a dynamic sensitivity state-space model is developed. The reduced process model is augmented with the sensitivity model in order to build a model giving possibilities for global sensitivity analysis of all model variables to all model parameters. Matlab software for sensi-tivity function calculation and global (augmented) sensitivity model simulation is developed. The results are described and discussed.
First, a definition of the sensitivity functions is given. Reduced ASM1 and UCT biological and benchmark mass balance models are then described. The augmented sensitiv-ity state-space model of the benchmark mass balance reduced model, based on the ASM1 biological model and based on the UCT biological model, is derived further. Description of Matlab software and results from simulation of the above models are given followed by discussion of the results. Finally, a summary of the results is given and their importance and applicability discussed.
Sensitivity functions, vectors and models
The behaviour of physical systems is determined by the values of their parameters. Investigation of the system response to
changes in the values of the parameters enables determina-tion of parametric sensitivity. Such analysis is important for all spheres of science and engineering, especially for design of the systems and their control (Frank, 1978; Varma et al.,1999; Fesso, 2007). The sensitivity function Xθ(t) of the nonlinear system:
(1)
where: XЄRl is the state space vector, uЄRm is the control vector, gЄRl is the nonlinear vector function of the process states, controls and parameters, and θЄRp is the vector of the parameters, is defined as (Sato & Ohmori, 2002):
(2)
In other words, the sensitivity function (Xθij) is a mathematical
description which indicates the influence of a slight change in parameter θj on the behaviour of the state variable Xi and XθЄRl. Differentiating (1) according to the vector θ gives the sensitiv-ity dynamic nonlinear equation:
(3)
The solution of the Eq. (3) describes the sensitivity func-tion behaviour, but requires data for the state-space vari-ables of the model given by Eq. (1). This means that a set of Eq. (1) and (3) has to be solved simultaneously. The set of these 2 equations can be represented as an augmented mathematical model (Tzoneva et al., 1996). The augmented sensitivity model is obtained by extension of the model state-space vector by the vector of the sensitivity func-tions, as follows:
(4)
The augmented model is used for global parametric analysis of the reduced COST benchmark mass balance model for the case of the ASM1 and UCT reduced biological models. Equations (1)-(4) are derived for each of these 2 cases.
Reduced biological and mass balance models for cost benchmark plant layout
Reduced biological models
A fundamental requirement for development of reduced models is that they contain a minimum number of state vari-ables and parameters to allow for model identification based on available online measurements and for online calculation of the process optimal control. Different types of reduced models are introduced in the literature (Moore, 1981; Glover, 1984; Safanov and Chiang, 1989; Latham and Anderson, 1985; Al-Saggaf and Franklin, 1988; Wisnewski and Doyle, 1996; Halevi et al., 1997; Beck et al., 1996; Lee at al., 2002; Jeppson, 1996) using different approaches of reduction, such as qualitative analysis, truncation methods, dominant eigen-values, and optimisation.
The existing biological models of the activated sludge
0)0(,)(),(,,, XXtuttXgtX
ttXtX
,,
,pjlitXtX
j
iij ,1,,1,),()),((
0)0(,)(
XtgtX
XtgtX
0
0
0)0(
,)(
))(),(),,((),(),(
))(),(),,(())(),(),,((
),(),(
,
XX
ttuttXgtX
tXtuttXg
tuttXg
tXtX
tX
S
S
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.zaISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012 289
process are characterised by a large number of processes and compounds, as well as a large number of kinetic and stoichio-metric parameters. These models, such as ASM1 and UCT, are very complex for use in real-time state and parameter estima-tion and process optimisation. ASM1 and UCT reduced models, used in conjunction with the benchmark mass balance model, are developed in Du Plessis (2009), applying time-scale analy-sis of the model variables’ dynamics (Dochain, 2003; Lennox et al., 2001; Stecha et al., 2005; Wang and Gawthrop, 2001). This allows decomposition of the complex models’ variables into the following time scales: • Fastest (minutes) – dynamics of physical-chemical variables
such as dissolved oxygen, pH, conductivity, redox potential• Intermediate (hours) – dynamics of utilisation of carbona-
ceous and nitrogenous substrates and the inflow flow rate and concentrations of waste materials
• Slowest (weeks or months) – dynamics of the heterotrophic and autotrophic microorganisms and slowly biodegradable biomass
The main disturbances to the activated sludge process are the load disturbances determined by the inflow rate and waste concentrations. The control of the ASP would be effective if it could overcome the effect of the inflow distur-bances. This means that the model of the process has to have dynamics within the same time scale as that of the distur-bances. That is why only dynamics of the carbonaceous and nitrogenous substrate and inflow disturbances are included in the reduced model. It is possible to neglect the equations of the full biological model describing the concentrations of the slowest variables because their dynamics are many times slower and they can thus be considered to be in a steady state. It is also possible to neglect the equation for dissolved oxygen (DO) concentration because its dynamics are approx. 10 times faster than the dynamics of the substrate concen-trations, and it can be assumed that the value of the DO is controlled and is equal to that of the required set-point. The dissolved oxygen concentration as a control variable is included in the switching functions of the process rates describing the reduced models. The above principles are used for development of both the ASM1 and UCT reduced models. At the same time the differences between them, due to the different views on the processes of adsorption and hydrolysis, are preserved. The notations of the ASM1 model are used for the process variables of both models. The nota-tions of the model parameters are kept as for the original models.
Reduced-order ASM1 biological model
Aerobic growth of heterotrophs XBH for degrading of organic matter, anoxic growth of heterotrophs for the de-nitrification process, aerobic growth of autotrophs XBA for the nitrification process and, lastly, the hydrolysis of entrapped organic nitro-gen, are the processes that characterise the dynamic behaviour of the 3 components used in describing the removal of carbon and nitrogen: soluble ammonium nitrogen SNHn; soluble nitrate nitrogen SNOn, and soluble readily-biodegradable substrate SSn concentrations. This model is described by the Peterson matrix given in Table 1. It has 4 processes and 3 compounds (vari-ables), where SOn is the concentration of dissolved oxygen and XSn is the slowly-biodegradable substrate concentration. The typical values of the model parameters for the ASM1 model are given in Table 2.
The corresponding differential equations describing the vari-ables’ rates for the n-th tank are:
(5)
(6)
(7)
Table 2ASM1 model parameters
Symbol Value ExplanationYA 0.24 Autotrophic biomass yieldYH 0.67 Heterotrophic biomass yieldiXB 0.086 Nitrogen mass per mass of COD in biomass µA 0.8 Maximum specific growth rate of autotrophic
biomassµH 6 Maximum specific growth rate of heterotro-
phic biomassKS 20 Half-saturation coefficient for heterotrophic
biomassKOH 0.2 Oxygen half-saturation coefficient for hetero-
trophic biomassKNO 0.5 Nitrate for denitrifying heterotrophsKNH 1.0 Ammonium half-saturation coefficient for
autotrophic biomassKOA 0.4 Oxygen half-saturation coefficient for auto-
trophic biomassηg 0.8 Correction factor for µH under anoxic
conditionsηh 0.4 Correction factor for hydrolysis under anoxic
conditionskh 3 Maximum hydrolysis rateKX 0.03 Half-saturation coefficient for hydrolysis of
slowly-biodegradable substrate
Table 1 The reduced-order ASM1 biological model in a matrix format
Compound i
process j Process
SS SNO SNH Pj, j = process for the n-th tank
1.Aerobic growth of heterotrophs
HY1
0 XBi BH
OOH
O
SS
SH X
SKS
SKS
2.Anoxic growth of heterotrophs
HY1
H
H
YY
86.21
XBi BHg
NONO
NO
OOH
OH
SS
SH X
SKS
SKK
SKS
ˆ
3.Aerobic growth of autotrophs
0 AY
1
AXB Yi 1
BAOOA
O
NHNH
NHA X
SKS
SKS
4. Hydrolysis 1 0 0
NONO
NO
OOH
OHh
OOH
O
BHsx
BHsBHh
SKS
SKK
SKS
XXKXXXk
nn
//
gBHOOH
OH
NONO
NO
SS
SHXBBH
OOH
O
SS
SHXBS X
SKK
SKS
SKS
iXSK
SSK
Si
NHr ˆˆ
BA
NHNH
NH
OOA
OA
AXB X
SKS
SKS
Yi 1
BA
NHNH
NH
OOA
OA
Ag
NONO
NO
OOH
OH
SS
SHBH
H
HS X
SKS
SKS
YSKS
SKK
SKSX
YY
NOr ˆ1ˆ
86.21
gBHOOH
OH
NONO
NO
SS
SH
HBH
OOH
O
SS
SH
HS X
SKK
SKS
SKS
YX
SKS
SKS
YSr ˆ1ˆ1
BH
OOH
OH
NONO
NOh
OOH
O
BHsx
BHsh X
SKK
SKS
SKS
XXKXXk
/
/
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.za
ISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012290
Reduced-order UCT biological model
The model is described by the Peterson matrix given in Table 3. It has 10 processes and 4 variables. The considered processes are: aerobic growth of XBH on SS with SNH, aero-bic growth of XBH on SS with SNO, anoxic growth of XBH on SS with SNH, anoxic growth of XBH on SS with SNO, aerobic growth of XBH on Sads with SNH, aerobic growth of XBH on Sads with SNO, anoxic growth of XBH on with SNH, anoxic growth of XBH on Sads with SNO, adsorption of XS , aerobic growth of XBA on SNH. The additional variable, representing the main concept of the enzyme reactions in the UCT model is Sads, which describes the concentration of the adsorbed slowly-biodegradable substrate. The model parameters are given in Table 4. The UCT notations for the model para-meters are used.
The corresponding equations for the variables’ rates are:
(8)
(9)
(10)
(11)
Mass-balance reduced model of the benchmark plant
The layout of the COST benchmark structure (Copp, 2002) is given in Fig. 1. The benchmark format has 2 anoxic tanks, 3 aerobic tanks and a secondary settler with 2 recycle flows (from aerobic tank 5 to the input and from the settler to the input), where: Q0 is the input flow rate, Qa is the internal recycle flow rate, Qn, n=1:5, is the output flow rate of the n-th tank, Qr is the recycle flow rate, Qe is the effluent flow rate, and Qw is the waste flow rate, Xn, n=1:5, is the vector of the waste compound concentrations in the influent and the n-th tank. The values of the flow rates and tanks volumes are given in Table 5.
Figure 1Wastewater treatment plant layout of the benchmark model
gOOH
OH
OOH
O
NONO
NO
NHHA
HA
gNONO
NO
OOH
OH
OOH
O
NHHA
NH
BHBHadsSP
BHadsMP
ZHads
SKK
SKS
SKS
SKK
SKS
SKK
SKS
SKS
XXSK
XSK
Yr
1
BHadsMABHSA XSfXXK
Table 3 The reduced-order UCT biological model in a matrix format
i
j
adsS SS NHS
NOS Process rates, j for the n-th tank
1) 0 ZHY1
HZBf ,
0 BH
NHHA
NH
OOH
O
SS
SH X
SK
S
SK
S
SK
S
2) 0
ZHY1
0 HZBf ,
BHNONO
NO
NHHA
HA
OOH
O
SS
SH X
SK
S
SKK
SKS
SK
S
3) 0
ZHY1
HZBf , ZH
ZH
YY
86.21
BH
NONO
NO
NHHA
NH
OOH
OH
SS
SH X
SKS
SKS
SKK
SK
S
4) 0 ZHY1
0
HZB
ZH
ZH
fYY
,
86.21
BHNONO
NO
NHHA
HA
OOH
OH
SSH
SH X
SK
S
SKK
SKK
SK
S
5)
ZHY1
0
HZBf ,
0 BH
NHHA
NH
OOH
O
BHadsSP
BHadsMP X
SK
S
SK
S
XSK
XSK
/
/
6)
ZHY1
0 0
HZBf , BH
NONO
NO
NHHA
HA
OOH
O
BHadsSP
BHadsMP X
SK
S
SKK
SK
S
XSKXSK
./
/
7)
ZHY1
0 HZBf ,
ZH
ZH
YY
86.21
BHgNONO
NO
NHHA
NH
OOH
OH
BHadsSP
BHadsMP X
SK
S
SKS
SKK
XSKXSK
//
8)
ZHY1
0 0
HZB
ZH
ZH
fYY
,
86.21
BHgNONO
NO
NHHA
HA
OOH
OH
BHadsSP
BHadsMP X
SKS
SKK
SKK
XSKXSK
/
/
9) 1 0 0 0 BHadsMABHSA XSfXXK 10) 0 0
HZB
ZA
fY
,
1
ZAY1
BANHSA
NH
OOA
OA X
SK
S
SKS
OOH
OH
OOH
O
NONO
NO
NHHA
HA
NONO
NO
OOH
OH
OOH
O
NHHA
NH
SSH
SBHH
ZHSS
SKK
SKS
SKS
SKK
SKS
SKK
SKS
SKS
SKSX
Yr 1
gNONO
NO
OOH
OH
OOH
O
NHHA
NHBH
BHadsSP
BHadsMPHZB
NONO
NO
OOH
OH
OOH
O
NHHA
NHBH
SSH
SHHZBSNH
SKS
SKK
SKS
SKS
XXSK
XSKf
SKS
SKK
SKS
SKS
XSK
Sfr
.,
,
BAOOH
O
NHSA
NHAHZB
ZAX
SKS
SKS
fY
,
1
BHadsSP
BHadsgMP
SSH
SH
NONO
NO
NHHA
HA
OOH
OHBHHZB
ZH
ZH
BHadsSP
BHadsgMP
SSH
SH
NONO
NO
NHHA
NH
OOH
OHBH
ZH
ZH
BHadsSP
BHadsMP
SSH
SH
NHHA
HA
NONO
NO
OOH
OBHHZBSNO
XSKXSK
SKS
SKS
SKK
SKKXf
YY
XSKXSK
SKS
SKS
SKS
SKKX
YY
XSKXSK
SKS
SKK
SKS
SKSXfr
,
,
86.21
.86.2
1
BAOOH
O
NHSA
NH
ZA
A XSK
SSK
SY
OOH
OH
OOH
O
NONO
NO
NHHA
HA
NONO
NO
OOH
OH
OOH
O
NHHA
NH
SSH
SBHH
ZHSS
SKK
SKS
SKS
SKK
SKS
SKK
SKS
SKS
SKSX
Yr 1
gNONO
NO
OOH
OH
OOH
O
NHHA
NHBH
BHadsSP
BHadsMPHZB
NONO
NO
OOH
OH
OOH
O
NHHA
NHBH
SSH
SHHZBSNH
SKS
SKK
SKS
SKS
XXSK
XSKf
SKS
SKK
SKS
SKS
XSK
Sfr
.,
,
BAOOH
O
NHSA
NHAHZB
ZAX
SKS
SKS
fY
,
1
BHadsSP
BHadsgMP
SSH
SH
NONO
NO
NHHA
HA
OOH
OHBHHZB
ZH
ZH
BHadsSP
BHadsgMP
SSH
SH
NONO
NO
NHHA
NH
OOH
OHBH
ZH
ZH
BHadsSP
BHadsMP
SSH
SH
NHHA
HA
NONO
NO
OOH
OBHHZBSNO
XSKXSK
SKS
SKS
SKK
SKKXf
YY
XSKXSK
SKS
SKS
SKS
SKKX
YY
XSKXSK
SKS
SKK
SKS
SKSXfr
,
,
86.21
.86.2
1
BAOOH
O
NHSA
NH
ZA
A XSK
SSK
SY
Table 4Reduced-order UCT process parameters
with their common valuesASM1/ UCT notations
Value Explanation
YA / YZA 0.15 Autotrophic biomass yieldYH / YZH 0.666 Heterotrophic biomass yieldiXB / fzbN 0.068 Nitrogen mass per mass of COD in
biomass µA/ myA 0.5 Maximum specific growth rate for auto-
trophic biomassµH / myH 2.5 Maximum specific growth rate for hetero-
trophic biomassKS / KSH 5 Half-saturation coefficient for heterotro-
phic biomassKOH/ KOH 0.002 Oxygen half-saturation coefficient for
heterotrophic biomassKNO/ KNO 0.1 Nitrate half-saturation coefficient for
denitrifying heterotrophic biomassKSA / KNH 1 Ammonium half-saturation coefficient for
autotrophic biomassKOA / KOA 0.002 Oxygen half-saturation coefficient for
autotrophic biomassηg / nyG 0.33 Correction factor for µH under anoxic
conditionskh / KMP 1.35 Maximum specific growth rate of the het-
erotrophs when utilising adsorbed particu-late slowly-biodegradable COD
aa XQ
Q2 Q3 Q4 Q5
5X 3X
An1 An2 Ae3 Ae4 Ae5 X 1X
2X
Q1
4X
Q0 rr XQ
ee XQ
wwXQ
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.zaISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012 291
Table 5Benchmark plant specifications
Notation Value DescriptionQ0 18 446 m3/day Influent flow rateQa 55 338 m3/day Internal flow rateQr 18 446 m3/day Recirculation flow rateQw 385 m3/day
(sludge age 10 days)Waste flow rate
V1 , V2 / V3 ÷ V5 1000 m3 /1333 m3 Anoxic/Aerobic volume
The mass-balance equations describing the benchmark struc-ture for the reduced ASM1 and UCT biological models in discrete time domain are (Du Plessis, 2009):For Tank 1: where Q = Qa + Qr + Q0
(12)
For Tank n=2, 3, 4, and 5:
(13)
where: Xn=[SNHn SNOn SSn]
T and Xn=[SNHn SNOn SSn Sadsn]
T, n=1:5, are vectors of the concentrations of the variables considered in the ASM1 and UCT reduced models, and∆t=15 (min) is the sampling period. The model of the settler, considered as an ideal one, is incorporated in the above equations as:
(14)
where: for the soluble compounds, λ=1, and for the particulate compounds, λ=(Q0+Qr)/(Qr+Qw).
The above equations are the same for both the ASM1 and UCT model, as structure, flow and volume. The difference is in the description of the number of compounds and the process rates, due to the different approaches to representing the biological activities of the microorganisms in these 2 models (Wentzel et al., 1992). The variables’ rates r are described correspondingly by Eqs. (5) ÷(7) for the reduced ASM1 and by Eqs. (8)÷(11) for the reduced UCT model.
Additionally to the parameters of the reduced models, a parameter f is included in the mass balance equations for the 1st tank. This parameter multiplies the concentration of the ammonia nitrogen in order to take account of the biological ammonia not considered in the reduced models, due to neglecting of the variable SND (soluble biodegradable organic nitrogen concentration).
The sensitivity of the reduced model variables towards the model parameters is evaluated by using the theory of sensitiv-ity (Schermann and Garag-Gabin, 2005; Montgomery, 1997; Breyfogle and Breyfogle, 2003).
Augmented sensitivity mass balance model derivation for the case of ASM1 reduced model
Sensitivity functions and equation derivation
The derivations of the sensitivity functions and models are analogous for every process tank. That is why they are calcu-lated for the n-th tank, as follows:
• For the parameter f 1) For the 1st tank
(15)
(16)
(17)
2) For tanks 2 ÷ 5
(18)
(19)
(20)
dSNH,n /df=SfNH,n, dSNO,n/df=Sf
NO,n, dSS,n/df=SfS,n are the sensitivity
functions of the process variables towards the parameter f. The variables’ rate derivatives towards the process variables
are given in Appendix A. • For the parameters
1) For the 1st tank
(21)
)()()()()1()1(
ktrkSkSQVtkSkS
nNHnnnn SNHNHn
NHNH
)()()()()1()1(
ktrkSkSQVtkSkS
nNOnnnn SNONOn
NONO
)()()()()1()1(
ktrkSkSQVtkSkS
nSnnnn SSSn
SS
ktrkSkSQVtkSkS
nnnnn Sadsadsadsn
adsads
)1(
1
5XXr
fS
SSSNH
fNO
SNOSNH
fNH
SNHSNH
fNHNHo
fNHr
fNHa
fNH
SrSrSr
QSSQSQSQV
S
1,1,
1,1,1,1,1,
1,1,
1,,5,5,1
1,1
fS
SSSNO
fNO
SNOSNO
fNH
SNHSNO
fNO
fNOr
fNOa
fNO
SrSr
SrQSSQSQV
S
1,1,
1,1,1,
1,
1,1,
1,1,5,5,1
1,1
fS
SSSS
fNO
SNOSS
fNH
SNHSS
fS
fSr
fSa
fS
SrSr
SrQSSQSQV
S
1,1,
1,1,1,
1,
1,1,
1,1,5,5,1
1,1
fnS
nSSnSNH
fnNO
nSNOnSNH
fnNH
nSNHnSNH
fnNH
fnNH
n
fnNH
SrSr
SrSSQV
S
,,
,,,,
,,
,,1,,1
fnS
nSSnSNO
fnNO
nSNOnSNO
fnNH
nSNHnSNO
fnNO
fnNO
n
fnNO
SrSr
SrSSQV
S
,,
,,,
,
,,
,,1,,1
fnS
nSSnSS
fnNO
nSNOnSS
fnNH
nSNHnSS
fnS
fnS
n
fnS
SrSr
SrSSQV
S
,,
,,,
,
,,
,,1,,1
nSNHnSNH
nNH
nSNH rSr ,
,,
,
, nSNOnSNH
nNO
nSNH rSr ,
,,
,
, nSSnSNH
nS
nSNH rSr ,
,,
,
, nNHnSNO
nNH
nSNO rSr ,
,,
,
nNOnSNO
nNO
nSNO rSr ,
,,
,
, nSSnSNO
nS
nSNO rSr ,
,,
,
,0
,
,
nNH
nSS
Sr , nSNO
nSSnNO
nSS rSr ,
,,
,
, nSSnSS
nS
nSS rSr ,
,,
,
XhhgOANHNOOHSHAXBAH KkKKKKKiYY ,,,,,,,,,,,,,
1,1,1,
1,1,1,1,
1,1,
1,1,5,1
1,1
SNHSSSSNHNO
SNOSNH
NHSNHSNHNHNHraNH
rSrSr
SrQSSQQV
S
)()()(
)()()()1(
11
55
1101
ktrkQSkfSQ
kSQkSQ
VtkSkS
NHSNHNH
NHrNHaNHNH
)()()(
)()()()1(
11
55
1101
ktrkQSkSQ
kSQkSQ
VtkSkS
NOSNONO
NOrNOaNONO
)()()(
)()()()1(
11
55
1101
ktrkQSkSQ
kSQkSQ
VtkSkS
SSSS
SrSaSS
ktr
kQSkSQ
kSQkSQ
VtkSkS Sads
adsadsO
adsradsaadsads 1
1
55
111
1
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.za
ISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012292
(22)
(23)
2) For the tanks n = 2,5
(24)
(25)
(26)
where: Sθ
NH,n, Sθ
NO,n, Sθ
S,n are the sensitivity functions to the parameter q, the partial derivatives of the variable’s rates according to the variables are determined above and the derivatives of the variables’ rates according to the model parameters are given in Appendix A.
Augmented sensitivity state-space model
The sensitivity state-space model is derived for the whole plant. The vector of the sensitivity state-space consists of all sensitiv-ity functions according to the model parameters, as follows for the n-th tank:
(27)
The state process variables’ vector for the n-th tank is:
(28)
The combined vector for the process variables and sensitivity functions for the n-th tank is:
(29)
This vector is used for the augmented state-space process and sensitivity-function model derivation. The matrix of the derivatives of the variables’ rates towards the process variables is:
(30)
The vector of the derivatives of the process rates to the model parameters, Appendix A, is:
(31)
Description of the augmented model in the discrete state-space domain is given for every tank separately because of the large dimensions of the vector of the state-space.
• Augmented sensitivity state-space equation for the 1st tank in the discrete form incorporates the process variables and sensitivity functions, as follows:
(32) where:
the variables’ rates are expressed by the process rates r1(k)=C1
STρ1(k), and ρ1=[ρ11 ρ12 ρ13 … ρ1N]T is a vector of the process rates, u1(k)=SO,1(k) is the dissolved oxygen concentration in the 1st tank, considered as its control action:
(33)
The matrices A1θ are identical for all parameters. The matrix
C1Srepresents the parameters in the Peterson matrix and is:
(34)
The matrix B1S is:
(35)
The inflow variables’ concentration vector is Xiϕ=[SNHϕ SNOϕ SSϕ]T.
The vector D1S represents the derivatives of the variables’ rates
towards the model parameters.
(36)
The matrix A15S describes the internal recycle between the end
of the 5thtank and the beginning of the 1stone. The connections
1,1,1,
1,1,1,
1,
1,1,
1,1,5,1
1,1
SNOSSSSNONO
SNOSNO
NHSNHSNONONOraNO
rSrSr
SrQSSQQV
S
1,1,1,
1,1,1,
1,
1,1,
1,1,5,1
1,1
SSSSSSSNO
SNOSS
NHSNHSSSSraS
rSrSr
SrQSSQQV
S
nSNHnSnSSnSNHnNO
nSNOnSNH
nSNHnSNHnSNHnNHnNHnNH
rSrSr
SrSSQV
S
,,,
,,,,
,,
,,1,1
,1
nSNOnSnSSnSNOnNO
nSNOnSNO
nSNHnSNHnSNOnNOnNOnNO
rSrSr
SrSSQV
S
,,,,,
,,
,,
,,1,1
,1
nSSnSnSSnSSnNO
nSNOnSS
nSNHnSSnSSnSnSnS
rSrSr
SrSSQV
S
,,,
,,,
,
,,
,,1,1
,1
45
,,,,,,,,,
,,,,,,,,,,,,
,,,,,,,,,,,,
,,,,,,,,,,,,
...
...
...
R
SSSSSSSSS
SSSSSSSSSSSS
SSSSSSSSSSSS
SSSSSSSSSSSS
S
T
KnS
KnNO
KnNH
knS
knNO
knNHnSnNOnNH
nSnNOnNHKnS
KnNO
KnNH
KnS
KnNO
KnNH
KnS
KnNO
KnNH
KnS
KnNO
KnNH
KnS
KnNO
KnNHnSnNOnNHnSnNOnNH
inS
inNO
inNH
YnS
YnNO
YnNH
YnS
YnNO
YnNH
fnS
fnNO
fnNH
n
XXXhhhhhh
gggOAOAOANHNHNHNONONO
OHOHOHSSSHHHAAA
XBXBXBAAAHHH
3,,, RSSSX TnSnNOnNHn
48RSXX nnSn
33,
,,
,,
,,
,,
,,
,,
,,,
,, ;; xnSS
nSSnSNO
nSSnSNH
nSSnSSnSNO
nSNOnSNO
nSNHnSNO
nSSnSNH
nSNOnSNH
nSNHnSNH
n
n RrrrrrrrrrXr
45
,,,,,,,,,
,,,,,,,,,,,,
,,,,,,,,,,,,
,,,,,,,,,,,,
R
rrrrrrrrr
rrrrrrrrrrrr
rrrrrrrrrrrr
rrrrrrrrrrrr
r
T
KnSS
KnSNO
KnSNH
knSS
knSNO
knSNHnSSnSNOnSNH
nSSnSNOnSNHKnSS
KnSNO
KnSNH
KnSS
KnSNO
KnSNH
KnSS
KnSNO
KnSNH
KnSS
KnSNO
KnSNH
KnSS
KnSNO
KnSNHnSSnSNOnSNHnSSnSNOnSNH
inSS
inSNO
inSNH
YnSS
YnSNO
YnSNH
YnSS
YnSNO
YnSNH
fnSS
fnSNO
fnSNH
n
XXXhhhhhh
gggOAOAOANHNHNHNONONO
OHOHOHSSSHHHAAA
XBXBXBAAAHHH
kXkuXAkuXDkXB
kuXCkXkuXAkXSSS
iS
STSSS
511151111
111111111
,,,,
,,,,1
484811111 ,,,,,,,,
,,,,,,,,,,
R
KkKKKKKiYYfA
AdiagkuxAXhhgOANHNOOHS
HABAHS
331
11
11
11
1
100
010
001
RA
QVt
QVt
QVt
A
33
1,1,
1
11,1,
1,1,
1,1,
1,1,
1
11,1,
1,1,
1,1,
1,1,
1
1
1
1
1
1
R
rVQttrtr
trrVQttr
trtrrVQt
A
SSSS
SNOSS
SNHSS
SSSNO
SNOSNO
SNHSNO
SSSNH
SNOSNH
SNHSNH
48445411 0
RCC S
100
011
186.2
1
10
1
AAXB
HH
HXB
HXB
YYi
YYY
i
Yi
tC
33
01
01
01
1345
344
01
1348
11
00
00
00
,0
00
, xxS R
QVt
QVt
QVtf
BR
QVt
BRBB
B
481,1,1,311 ,,,,,,
,,,,,,,,,0R
KkKKK
KKiYYfrrrD
T
XhhgOANHNO
OHSHAXBAHSSSNOSNHS
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.zaISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012 293
are done by the vector X5S=[X5 S5
θ]∈R48:
(37)
where: all matrices A15 are identical and:
• Augmented sensitivity model equations for tanks n = 2 ÷ 5
The state-space model incorporating the process variables and sensitivity functions are derived for the n-th tank, as follows:
(38) where: ρn=[ρn1 ρn2 ρn3 … ρn4]
T
(39)
The matrices An,n-1S are:
(40)
where: An,n-1 and An,n-1
q are identical
The vector Dn
S is:
(41)
The matrices CnS are equal to C1
S.
• Augmented sensitivity state-space model of the whole plant
The model of the variables and sensitivity functions for the whole plant is described on the basis of Eqs. (32) and (38) as follows: (42)
The matrices AS∈R240x240 and CS∈ R20x240 are:
The vector of the process rates is:
(43)
The matrix BS and the vector DS are:
(44)
(45)
Model (42)-(45) is used to calculate the parametric sensitivity functions of the benchmark structure with the ASM1 reduced model.
Augmented sensitivity mass balance model derivation for the case of the UCT reduced-order model
Sensitivity function and equation derivation
The order of calculation is done in the same way as above.
• For the parameter f
1) For the 1st tank
The sensitivity functions are given by equations:
(46)
(47)
(48)
(49)
2) For the other n = 2÷5 tanks
The sensitivity functions are given by the equations:
(50)
4848151515 ,,,,,,
,,,,,,,,,,
RKkKKK
KKiYYfAAdiagA
XhhgOANHNO
OHSHAXBAHS
33
1
1
1
1515
00
00
00
R
QQVt
QQVt
QQVt
AA
ra
ra
ra
raS QQ
VtdiagA1
15
S
nSnnnn
Snnn
STn
Snnn
Sn
Sn XAkuXDkuXCkXkuXAkX 11,,,,,,,1
33
1
1
1
1515
00
00
00
R
QQVt
QQVt
QQVt
AA
ra
ra
ra
raS QQ
VtdiagA1
15
S
nSnnnn
Snnn
STn
Snnn
Sn
Sn XAkuXDkuXCkXkuXAkX 11,,,,,,,1
4848
,,,,,,,,,,,,,,,,
RKkKK
KKKiYYfAAdiagA
XhhgOANH
NOOHSHAXBAHnnSn
4545
,,
,,
,,
,,
,,
,,
,,
,,
,,
1
1
1
R
trQVttrtr
trtrQVttr
trtrtrQVt
A
nSSnSS
n
nSNOnSS
nSSnSS
nSSnSNO
nSNOnSNO
n
nSNHnSNO
nSSnSNH
nSNOnSNH
nSNHnSNH
n
n
33
100
010
001
R
QVt
QVt
QVt
A
nn
nn
nn
n
48481,1,1, ,,,,,
,,,,,,,,,,,
RKkKK
KKKiYYfAAdiagA
XhhgOANH
NOOHSHAXBAHnnnnSnn
33
1
1
1
1,1,
00
00
00
R
QVt
QVt
QVt
AA
nn
nn
nn
nnnn
33
1
1
1
1,1,
00
00
00
R
QVt
QVt
QVt
AA
nn
nn
nn
nnnn
484811,
RQVtdiagA nn
Snn
48,,,31 ,,,,,
,,,,,,,,,0 R
KkKKKKKiYYf
rrrDT
XhhgOANH
NOOHSHAXBAHnSSnSNOnSNH
Sn
kuxDXBkuxCkXAkX Si
SSTSSS ,,,,1
SS
SS
SS
SS
SS
S
AAAA
AAAA
AA
A
554
443
332
221
151
000000000000
000
S
S
S
S
S
S
CC
CC
C
C
5
4
3
2
1
00000000000000000000
20
5453525144434241
343332312423222114131211,, RkuxT
SS
SS
SS
SS
SS
S
AAAA
AAAA
AA
A
554
443
332
221
151
000000000000
000
S
S
S
S
S
S
CC
CC
C
C
5
4
3
2
1
00000000000000000000
20
5453525144434241
343332312423222114131211,, RkuxT
3240
3192
1
0
RB
B
S
S
24054321 RDDDDDDTSSSSSS
fSNH
fads
SSNH
fS
SSSNH
fNO
SNOSNH
fNH
SNHSNH
fNHNH
fNHr
fNHa
fNH
rSrSrSr
SrQSSQSQSQV
S
ads1,1,1,1,
1,1,1,
1,1,
1,1,
1,1,05,5,1
1,
1,
1
fSNO
fads
SSNO
fS
SSSNO
fNO
SNOSNO
fNH
SNHSNO
fNO
fNOr
fNOa
fNO
rSrSrSr
SrQSSQSQV
S
ads1,1,1,1,
1,1,1,
1,1,
1,1,
1,1,5,5,1
1,
1,
1
fSS
fads
SSS
fS
SSSS
fNO
SNOSS
fNH
SNHSS
fS
fSr
fSa
fS
rSrSrSr
SrQSSQSQV
S
ads1,1,1,1,
1,1,1,
1,1,
1,1,
1,1,5,5,1
1,
1,
1
fads
fads
Sads
fS
SSads
fNO
SNOads
fNH
SNHads
fads
fadsr
fadsa
fads
rSrSrSr
SrQSSQSQV
S
ads1,1,1,1,
1,1,1,
1,1,
1,1,
1,1,5,5,1
1,
1,
1
f
nSNHf
nadsnSadsnSNH
fnS
nSSnSNH
fnNO
nSNOnSNH
fnNH
nSNHnSNH
fnNH
fnNH
n
fnNH
rSrSrSr
SrSSQV
S
,,,,,
,,,
,,
,,
,,1,,1
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.za
ISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012294
(51)
(52)
(53)
• For the parameters
the equations are as follows: 1) For the 1st tank
(54)
(55)
(56)
(57)
2) For the tanks n = 2 ÷ 5
(58)
(59)
(60)
(61)
The partial derivatives of the variables’ rates are determined in Appendix B.
Augmented sensitivity state-space model
The augmented sensitivity state-space model is formed in the same way as above. It is based on the model of the plant extended with the model of the sensitivity functions. The vector of the sensitivity functions consists of all sensitivity functions according to the model parameters, as follows for the n-th tank:
(62)
The process state-space vector for the n-th tank is:
(63)
The augmented vector for the process variables and sensitivity functions is: (64)
This vector is used to describe the augmented model of the process variables and sensitivity functions. The matrix of the process variables’ rates derivatives towards the process variables is:
(65)
The vector of the variables’ rate derivatives towards the para-meters can be represented in the following way (Appendix B):
(66)
The augmented process and sensitivity state-space model is described for every tank separately because the large dimen-sion of the vector of state-space and the number of tanks creates the need for a large amount of space for describing the model matrices.
• The state-space model for Tank 1 in discrete form is:
(67)
where: ρ1=[ρ11 ρ12 ρ13 … ρ1,10]T is a vector of the process rates
(68) The matrices A1
θare identical for all model parameters:
The matrix C1S=[C1 010x68]∈R10x72 represents the parameters
from the Peterson matrix:
f
nSNOf
nadsnSadsnSNO
fnS
nSSnSNO
fnNO
nSNOnSNO
fnNH
nSNHnSNO
fnNO
fnNO
n
fnNO
rSrSrSr
SrSSQV
S
,,,
,,,
,,,
,
,,
,,1,,1
fnSS
fnads
nSadsnSS
fnS
nSSnSS
fnNO
nSNOnSS
fnNH
nSNHnSS
fnS
fnS
n
fnS
rSrSrSr
SrSSQV
S
,,,
,,,
,,,
,
,,
,,1,,1
fnSads
fnads
nSadsnSads
fnS
nSSnSads
fnNO
nSNOnSads
fnNH
nSNHnSads
fnads
fnads
n
fnads
rSrSrSr
SrSSQV
S
,,,
,,,
,,,
,
,,
,,1,,1
SASPMPgOANANOOHSHAZBHZAZH KKKKKKKKfYY ,,,,,,,,,,,,,
1,1,1,1,1,
1,1,1,
1,1,
1,1,
1,1,5,1
1,1
SNHadsSadsSNHS
SSSNHNO
SNOSNH
NHSNHSNHNHNHraNH
rSrSrSr
SrQSSQQV
S
1,1,1,
1,1,1,
1,1,1,
1,
1,1,
1,1,5,1
1,1
SNOadsSadsSNOS
SSSNONO
SNOSNO
NHSNHSNONONOraNO
rSrSrSr
SrQSSQQV
S
1,1,1,
1,1,1,
1,1,1,
1,
1,1,
1,1,5,1
1,1
SSadsSadsSSS
SSSSNO
SNOSS
NHSNHSSSSraS
rSrSrSr
SrQSSQQV
S
1,1,1,
1,1,1,1,1,
1,1,
1,1,
1,1,5,1
1,1
adsadsSadsadsS
SSadsNO
SNOads
NHSNHadsadsadsraads
rSrSrSr
SrQSSQQV
S
nSNHnadsnSadsnSNHnS
nSSnSNHnNO
nSNOnSNH
nNHnSNHnSNHnNHnNH
nnNH
rSrSrSr
SrSSQV
S
,,,,,
,,,
,,
,,
,,1,,1
nSNOnadsnSadsnSNOnS
nSSnSNOnNO
nSNOnSNO
nNHnSNHnSNOnNOnNO
nnNO
rSrSrSr
SrSSQV
S
,,,
,,,
,,,
,
,,
,,1,,1
nSSnadsnSads
nSSnSnSSnSSnNO
nSNOnSS
nNHnSNH
nSSnSnSn
nS
rSrSrSr
SrSSQV
S
,,,
,,,
,,,
,
,,
,,1,,1
nSadsnadsnSadsnSadsnS
nSSnSadsnNO
nSNOnSads
nNHnSNHnSadsnadsnads
nnads
rSrSrSr
SrSSQV
S
,,,
,,,
,,,
,
,,
,,1,,1
68
,,,,,,,,
,,,,,,,,,,,,
,,,,,,,,,,,,
,,,,,,,,,,,,
,,,,,,,,,,,,
,,,,,,,,,,,,
...
...
...
...
...
R
SSSSSSSS
SSSSSSSSSSSS
SSSSSSSSSSSS
SSSSSSSSSSSS
SSSSSSSSSSSS
SSSSSSSSSSSS
S
T
Knads
KnS
KnNO
KnNH
fnads
fnS
fnNO
fnNH
Knads
KnS
KnNO
KnNH
Knads
KnS
KnNO
KnNH
Knads
KnS
KnNO
KnNH
nadsnSnNOnNHK
nadsKnS
KnNO
KnNH
Knads
KnS
KnNO
KnNH
Knads
KnS
KnNO
KnNH
Knads
KnS
KnNO
KnNH
Knads
KnS
KnNO
KnNH
nadsnSnNOnNHnadsnSnNOnNHfnads
fnS
fnNO
fnNH
Ynads
YnS
YnNO
YnNH
Ynads
YnS
YnNO
YnNH
fnads
fnS
fnNO
fnNH
n
AAAAMAMAMAMA
SASASASASPSPSPSPMPMPMPMP
ggggOAOAOAOANANANANA
NONONONOOHOHOHOHSSSS
HHHHAAAAZBHZBHZBHZBH
ZAZAZAZAZHZHZHZH
4,,,, RSSSSX TnadsnSnNOnNHn
72RSXXT
nnSn
44
,,
,,
,,
,,
,,
,,
,,
,,
,,
,,
,,
,,
,,
,,
,,
,,
x
T
nSadsnSads
nSSnSads
nSNOnSads
nSNHnSads
nSadsnSS
nSSnSS
nSNOnSS
nSNHnSS
nSadsnSNO
nSSnSNO
nSNOnSNO
nSNHnSNO
nSadsnSNH
nSSnSNH
nSNOnSNH
nSNHnSNH
n
n R
rrrrrrrrrrrrrrrr
Xr
(
68
,,,,,,,,
,,,,,,,,,,,,
,,,,,,,,,,,,
,,,,,,,,,,,,
,,,,,,,,,,,,
,,,,,,,,,,,,
...
...
...
...
...
R
rrrrrrrr
rrrrrrrrrrrr
rrrrrrrrrrrr
rrrrrrrrrrrr
rrrrrrrrrrrr
rrrrrrrrrrrr
r
T
Knads
KnS
KnNO
KnNH
fnads
fnS
fnNO
fnNH
Knads
KnS
KnNO
KnNH
Knads
KnS
KnNO
KnNH
Knads
KnS
KnNO
KnNH
nadsnSnNOnNHKnads
KnS
KnNO
KnNH
Knads
KnS
KnNO
KnNH
Knads
KnS
KnNO
KnNH
Knads
KnS
KnNO
KnNH
Knads
KnS
KnNO
KnNH
nadsnSnNOnNHnadsnSnNOnNHfnads
fnS
fnNO
fnNH
Ynads
YnS
YnNO
YnNH
Ynads
YnS
YnNO
YnNH
fnads
fnS
fnNO
fnNH
n
AAAAMAMAMAMA
SASASASASPSPSPSPMPMPMPMP
ggggOAOAOAOANANANANA
NONONONOOHOHOHOHSSSS
HHHHAAAAZBHZBHZBHZBH
ZAZAZAZAZHZHZHZH
kXkuXAkuXD
kXBkuXCkXkuXAkXSSS
STSSSS
51115111
1111111111
,,,,
,,,,1
7272111 ,,,,,,
,,,,,,,,,,,
R
KfKKKKKKKKfYYf
AAdiagAAMASASPMPgOA
NANOOHSHAZBHZAZHS
441
11 }1{
RQVtdiagA
44
1,1,
1
11,1,
1,1,
1,1,
1,1,
1,1,
1
11,1,
1,1,
1,1,
1,1,
1,1,
1
11,1,
1,1,
1,1,
1,1,
1,1,
1
1
1
1
1
1
1
R
rVQttrtrtr
trrVQ
ttrtr
trtrrVQttr
trtrtrrVQ
t
A
SadsSads
SSSads
SNOSads
SNHSads
SadsSS
SSSS
SNOSS
SNHSS
SadsSNO
SSSNO
SNOSNO
SNHSNO
SadsSNH
SSSNH
SNOSNH
SNHSNH
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.zaISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012 295
(69)
(70) The inflow vector is: (71)
The vector D1Srepresents the derivatives of the variables’ rates
towards the model parameters:
(72)
The matrix A15Sdescribes the internal recycle between the end
of the 5th tank and the beginning of the 1st one. The vector X5Sis
formed in the same way as the vector X1S : X5
S=[X5 S5θ]T∈R72
where:
(73)
• Derivation of the equations for tanks n = 2 ÷ 5
The return-flow dynamics for all other tanks are identical, thus the models will have the same structure, as follows:
(74)
where: ρn=[ρn1 ρn2 ρn3 … ρn,10]
T
The matrix Anθ has the same structure for all parameters θ and
for all tanks n=2:5.
The matrices An,n-1S are:
(75)
for all parameters θ.
The matrices CnS are equal to C1
S with the same coefficients.
The vector DnS is:
(76) • Augmented sensitivity state-space model of the whole plant
Combining all equations for the tanks, the full model is:
(77)
where:
The augmented model (77) is used for calculation of the sensitivity functions of the COST benchmark plant reduced mass-balance model for the case of the UCT reduced biologi-cal model. The models (42) and (77) are nonlinear because of the nonlinear rate expressions in the matrices AS and vectors r and DS. The dissolved oxygen concentration is considered as a control input for these models and it appears in the rate expres-sions forming AS, r, and DS. Matlab programs are developed for parameter sensitivity calculations using equations (42) and (77). A matrix/vector representation of the models allows simplifica-tion of the software code and reduction of time for calculations. Sensitivity analysis
The sensitivity analysis of the wastewater treatment model variables towards the model parameters is done in order to determine which parameters have to be estimated for the cor-responding reduced models during the real-time operation and control of the process.
Software for the sensitivity analysis
Software programs are developed in the Matlab environment
4101
00111000
1086.2
10
1086.2
1100100
0186.2
10
0186.2
1010010
*
R
YfY
YfYY
YYY
f
YfYf
YfYY
YYY
f
YfYf
tC
ZAZBHZH
ZHZBHZH
ZH
ZHZH
ZHZBH
ZHZBH
ZHZBH
ZHZBHZH
ZH
ZHZH
ZHZBH
ZHZBH
ZHZBH
,468
144
1472
1
11
RBRBRBB
BS
467
1
0
1
0....
000
x
VtQ
B
, )(1
0
1
0
1
0
1
01 V
tQVtQ
VtQ
VtQfdiagB
SadsNOT
adsSNONH XSSRSSSSX ,0,41
721,1,1,1,411 ,,,,,,,
,,,,,,,,,,0R
KfKKKKKKKKfYYfrrrr
DT
AMASASPMPgOANA
NOOHSHAZBHZAZHadsSSSNOSNHS
7272151515 ,,,,,,,,
,,,,,,,,,
R
KfKKKKKKKKfYYf
AAdiagAT
AMASASPMPgOANANO
OHSHAZBHZAZHS
44
115
xra RVQQtdiagA
6868
115
xra RVQQtdiagA
7272
115
xraS RVQQtdiagA
kXAkuXDkuXCkXkuXAkX S
nSnnnn
Snnn
TSn
Snnn
Sn
Sn 11,,,,,,,1
kXAkuXDkuXCkXkuXAkX S
nSnnnn
Snnn
TSn
Snnn
Sn
Sn 11,,,,,,,1
7272
,,,,,,,,,,,,,,,,
,
R
KfKKKKKKKKfYYf
AAdiagAT
AMASASPMPgOANANO
OHSHAZBHZAZHnn
Sn
441
n
nn Q
VtdiagA
44
,,
,,
,,
,,
,,
,,
,,
,,
,,
,,
,,
,,
,,
,,
,,
,,
1
1
1
1
R
rVQ
ttrtrtr
trrVQ
ttrtr
trtrrVQ
ttr
trtrtrrVQ
t
A
nSadsnSads
n
nnSSnSads
nSNOnSads
nSNHnSads
nSadsnSS
nSSnSS
n
nnSNOnSS
nSNHnSS
nSadsnSNO
nSSnSNO
nSNOnSNO
n
nnSNHnSNO
nSadsnSNH
nSSnSNH
nSNOnSNH
nSNHnSNH
n
n
n
AMASASPMPgOANANO
OHSHAZBHZAZHnnnnSnn KfKKKKKK
KKfYYfAAdiagA
,,,,,,,,,,,,,,,,, ,1,1,
1,
(
4411,
4411, , x
n
nnn
x
n
nnn R
VtQdiagAR
VtQdiagA
,
727211,
x
n
nSnn R
VtQdiagA
72,,,,41
,,,,,,,,,,,,,,,,,0
RKfKKKKKKKK
fYYfrrrrD
T
AMASASPMPgOANANOOHSH
AZBHZAZHnadsnSSnSNOnSNHSn
SS
SSi
SSTSSSS
XX
kuxDkXBkuxCkXAkX
00
,,,,,1
360360
554
443
332
221
151
000000000000
000
R
AAAA
AAAA
AA
A
SS
SS
SS
SS
S
S
4360
4288
1
0
RB
B
S
S
36050
54321 ,,,, RCCCCCdiagC SSSSSS , 36054321 RDDDDDD TSSSSSS
50
51059
5857565554535251410494847
4645444342413103938373635
3433323121029282726252423
2221110191817161514131211
R
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.za
ISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012296
for the considered 2 cases:• Calculation of the sensitivity functions of the benchmark
process based on ASM1 reduced biological model: programs BASM1S.m, rateASM1.m
• Calculation of the sensitivity functions of the bench-mark process based on UCT biological model: programs BUCTS1.m, rateBUCTS.m
For each of the considered cases the software consists of:• Main program for input of the nominal process parameters,
initial conditions, average values of the biomass concen-trations, calculation of the process model matrices and organisation of the calculation algorithm
• Sub-programs for calculation of the process rates, sensitiv-ity functions and formation of the sensitivity model state-space and rate matrices and vectors.
The algorithm of the calculation is given in Fig. 2a for the main programme and in Fig. 2b for the sub-programmes.
XS= 0.05*[82.135*ones(1,K);76.386*ones(1,K);64.855*ones (1,K);55.694*ones(1,K);49.306*ones(1,K)]
The vector of the dissolved oxygen concentration is:
U=[0.2;0.2;2.0;2.29;1.91].
The results from the simulations of the sensitivity functions for Tank 1 and Tank 5 are given in Figs. 3 to 8. The minimum or maximum values of the sensitivity functions are given in Table 6 and Table 7.
Results for the UCT reduced-order biological model
The simulation is done for the parameters given in Table 4 and for the inflow average concentration given by the vector Xϕ:
Xϕ =[31.56; 0.0;69.5;202.3].
The values for the biomass and the slowly biodegradable sub-strate are given by the vectors:
XBH = 0.02*[2551.76*ones(1,K);2553.38*ones(1,K);2557.13* ones(1,K);2559.18*ones(1,K);2559.34*ones(1,K)]; XBA = 0.02*[148.389*ones(1,K);148.309*ones(1,K);148.941*ones(1,K);149.527*ones(1,K);149.797*ones(1,K)]; XS = 0.02*[82.135*ones(1,K);76.386*ones(1,K);64.855*ones (1,K);55.694*ones(1,K);49.306*ones(1,K)].
The vector of the dissolved oxygen concentration is:
U=[0.2;0.2;2.0;2.29;1.91].
The results from the simulations of the sensitivity functions for Tank 1 and Tank 5 are given in Figs. 9 to 16. The minimum or maximum values of the sensitivity functions are given in Table 6 and Table 7.
Discussion of the results
The minimum or maximum values of the sensitivity functions are shown in the tables for the different process variables and parameters. These values can be analysed as follows.
Sensitivity functions simulation for the benchmark process model based on the reduced ASM1 biological model
The sensitivity functions of SNH1 for almost all parameters (without KOH and KNO) display the same type of behaviour. They start from a zero value (in some cases with a small delay), grow in a positive or negative direction to a maximum/minimum value and then reduce to a steady-state value. The maximum/minimum value is in the time interval between the 5th and 8th hour from the beginning of the simulation period. The steady-state value is achieved at around the 15th hour from the beginning of the simulation period. The maximum values are obtained for parameters f,YA, KS,KOH,KNO,KNH,KOA,KX. The great-est maximum is for parameter YA (490), followed by KNH (100) and KOA (80). The smallest minimums are for parameters iXB (-300), μA (-200), and μH (-45). The behaviour of the sensitivity functions of SNH5 for Tank 5 has the same characteristics as for Tank 1, but the maximum and minimum values of these func-tions are greater in absolute values, as for example YA (498) and
Calculation process mass balance model matrices
k=0
Call subroutine for sensitivity model matrices calculation
Sensitivity model simulation
k<K
Simulation of the sensitivity functions and process behavior.Plot, Stop
k=k+1
No
Yes
Input data for model parameters, initial conditions, biomass and inflow values
Global parameters
Calculation of the process rates for every tank
Calculation of derivatives of the process rates towards the process variables
Calculation of derivatives of process rates towards the model parameters for every tank
Calculation of the sensitivity state space model matrices and vectors
Figure 2b: Flow chart of the subprogram
Figure 2a: Flow chart of the main program for sensitivity functions calculation
Figure 2aFlow chart of the main program for
sensitivity functions calculation
Figure 2bFlow chart of the subprogram
Results for the ASM1 reduced biological model
Simulation is done for the parameters given in Table 3 and for the inflow average concentration given by the vector Xθ:
Xθ=[31.56; 0.0;69.5;202.3].
The steady-state conditions of the biomass and the slowly-biodegradable substrate are given by the vectors:
XBH = 0.05*[2551.76*ones(1,K);2553.38*ones(1,K); 2557.13*ones(1,K);2559.18*ones(1,K);2559.34*ones(1,K)];XBA= 0.05*[148.389*ones(1,K);148.309*ones(1,K);148.941* ones(1,K);149.527*ones(1,K);149.797*ones(1,K)];
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.zaISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012 297
Table 6
Maximum or minimum values of sensitivity functions for the benchmark process model based on the reduced ASM1 and UCT biological models
P a r a me t e r s
Variables
Maximum or minimum values of the sensitivity functions ASM1 - Tank1 UCT - Tank 1
1NHS 1NOS
1SS 1NHS
1NOS 1SS
1adsS
f / f 0.18 0.195 -0.03 35 4E-6 -5E-11 -2E-9
HY /ZHY -1 -16 120 -0.8 0.75, -4 38 4000
AY /ZAY 490 -400 1.8, -1.9 275 35 -1.1E-9 -0.6E-7
XBi /ZBHf -300 -390 75 -1900 -0.12 0.55E-9 4E-8
A /A -200
5 200 -10
-0.75, 0.95 -500 4 -0.5
-2E-9, 0.5E-9
-0.7E-7
H /H -45 -95 -220 -39 -0.3 -550 1.2E-9
SK /SK 1 7.5 5 0.05, -0.45 0.015 33 -35E-12
OHK / OHK 8 -18 -90 190 -0.85 0.025 7.5
NOK /NOK 1.2 7 -12.5 2E-3, -1E-3 3E-3 0.06 0.037
NHK /NHK 100
-10 -98 10
0.3, -0.4 110 4, -0.5 3.5E-3 0.35
OAK /OAK 80, -5 -60, 9 0.2, -0.25 500 1.9, 0.5 -1.5E-9 -0.6E-7
g /g -4.8 -14 1.95 -0.5 -0.6 5E-11 -3.8
h -4.5 -13 1.96 - - - -
hk /MPK -0.6 -1 55 95 2.75 -3E-11 -2000
XK 2 2.5 -180 - - - - SPK - - - 105 -0.2, 0.49 -38E-10 700, -250 SAK - - - 1.8, -0.2 -3.5, 0.05 1.5E-9 0.7E-7 mAf - - - 5E-4 -0.35 3E-12 300 AK - - - 0.02 -17 1.5E-10 1.2E4
Table 7 Maximum or minimum values of sensitivity functions for the benchmark process model
based on the reduced ASM1 and UCT biological models –Tank 5 P a r a me t e r s
Variables
Maximum or minimum values of the sensitivity functions
ASM1 - Tank 5 UCT - Tank 5
5NHS 5NOS
5SS 5NHS
5NOS 5SS
5adsS
f / f 0.14 0.25 -0.025 35 25E-6 -3E-11 -0.9E-9
HY /ZHY -1.6 -18 150 -0.95 0.5, -4 45 4900
AY /ZAY 498
-10 -495, 50 0.8, -0.9 295 35 -1E-9 -0.6E-7
XBi /ZBHf -330 -500 45 -2200 -0.12, 0.1 3.9E-10 4E-8
A /A -250, 20 220, -10 -0.5, 0.4 -500 5, -0.3 -1E-9, 0.2E-9 -0.8E-7
H /H -58 -120 -220 -39 -0.3 -700 1.2E-9
SK /SK 1.2 9.5 5.5 -0.4 0.015 35 -38E-10
OHK /OHK 6 -20, 3 -50 190 -0.98 0.015 7
NOK /NOK 1 7.5 -8 2E-3, -1E-3 3E-4 0.06 0.037
NHK /NHK 120, -5 -120, 5 0.2, -0.25 120 -120 3.5E-3 0.38
OAK /OAK 85, -5 -75, 10 0.15, -0.15 500 2, -0.4 -0.9E-9 -0.6E-7
g /g -5.5 -18 1.1 -0.5 -0.7 3.6E-11 -3.8
h -0.38 -055 23 - - - -
hk /MPK -0.8 -1.3 60 105 3.5 -2E-11 -2500
XK 2.5 3.9 -190 - - - - SPK - - - 125 -0.2, 0.49 -2.5E-11 750, -780 SAK - - - 2.5, -0.5 -4.8, 0.5 1.1E-9 0.7E-7 mAf - - - 6E-4 -0.49 3E-12 350 AK - - - 0.027 -18 1.5E-10 1.8E4
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.za
ISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012298
0 20 40 60 80 100 1200
5
10Process variable Snh1
SN
H1
discrete time k0 20 40 60 80 100 120
0
0.1
0.2Sensitivity function Snh1f
SN
H1f
discrete time,k
0 20 40 60 80 100 120-2
-1
0Variable Snh1YH
SN
H1Y
H
discrete time,k0 20 40 60 80 100 120
-500
0
500Variable Snh1YA
SN
H1Y
A
discrete time,k
0 20 40 60 80 100 120-400
-200
0Sensitivity function Snh1iXB
SN
H1i
XB
discrete time,k0 20 40 60 80 100 120
-200
0
200Sensitivity function Snh1muA
SN
H1m
uA
discrete time,k
0 20 40 60 80 100 120
-40
-20
0Sensitivity function Snh1muH
SN
H1m
uH
discrete time,k0 20 40 60 80 100 120
0
1
2Sensitivity function Snh1KS
SN
H1K
S
discrete time,k
0 20 40 60 80 100 1200
5
10Sensitivity function Snh1KOH
SN
H1K
OH
discrete time,k0 20 40 60 80 100 120
0
1
2Sensitivity function Snh1KNO
SN
H1K
NO
discrete time,k
0 20 40 60 80 100 120-100
0
100Sensitivity function Snh1KNH
SN
H1K
NH
discrete time,k0 20 40 60 80 100 120
-100
0
100Sensitivity function Snh1KOA
SN
H1K
OA
discrete time,k
0 20 40 60 80 100 120
-4
-2
0Sensitivity function Snh1etag
SN
H1e
tag
discrete time,k0 20 40 60 80 100 120
-4
-2
0Sensitivity function Snh1etah
SN
H1e
tah
discrete time,k
0 20 40 60 80 100 120-1
-0.5
0Sensitivity function Snh1Kh
SN
H1K
h
discrete time,k0 20 40 60 80 100 120
0
2
4Sensitivity function Snh1Kx
SN
H1K
x
discrete time,k
0 20 40 60 80 100 120-10
0
10Process variable Snh5
SN
H5
discrete time k0 20 40 60 80 100 120
0
0.1
0.2Sensitivity function Snh5f
SN
H5f
discrete time,k
0 20 40 60 80 100 120-2
-1
0Variable Snh5YH
SN
H5Y
H
discrete time,k0 20 40 60 80 100 120
-500
0
500Variable Snh5YA
SN
H5Y
A
discrete time,k
0 20 40 60 80 100 120-400
-200
0Sensitivity function Snh5iXB
SN
H5i
XB
discrete time,k0 20 40 60 80 100 120
-500
0
500Sensitivity function Snh5muA
SN
H5m
uA
discrete time,k
0 20 40 60 80 100 120-100
-50
0Sensitivity function Snh5muH
SN
H5m
uH
discrete time,k0 20 40 60 80 100 120
0
1
2Sensitivity function Snh5KS
SN
H5K
S
discrete time,k
0 20 40 60 80 100 1200
5
10Sensitivity function Snh5KOH
SN
H5K
OH
discrete time,k0 20 40 60 80 100 120
0
1
2Sensitivity function Snh5KNO
SN
H5K
NO
discrete time,k
0 20 40 60 80 100 120-200
0
200Sensitivity function Snh5KNH
SN
H5K
NH
discrete time,k0 20 40 60 80 100 120
-100
0
100Sensitivity function Snh5KOA
SN
H5K
OA
discrete time,k
0 20 40 60 80 100 120
-4
-2
0Sensitivity function Snh5etag
SN
H5e
tag
discrete time,k0 20 40 60 80 100 120
-0.4
-0.2
0Sensitivity function Snh5etah
SN
H5e
tah
discrete time,k
0 20 40 60 80 100 120-1
-0.5
0Sensitivity function Snh5Kh
SN
H5K
h
discrete time,k0 20 40 60 80 100 120
0
2
4Sensitivity function Snh5Kx
SN
H5K
x
discrete time,k
0 20 40 60 80 100 1200
5
10Process variable SNO1
SN
O1
discrete time k0 20 40 60 80 100 120
0
0.1
0.2Sensitivity function SNO1f
SN
O1f
discrete time,k
0 20 40 60 80 100 120-20
-10
0Variable SNO1YH
SN
O1Y
H
discrete time,k0 20 40 60 80 100 120
-500
0
500Variable SNO1YA
SN
O1Y
A
discrete time,k
0 20 40 60 80 100 120
-400
-200
0Sensitivity function SNO1iXB
SN
O1i
XB
discrete time,k0 20 40 60 80 100 120
-200
0
200Sensitivity function SNO1muA
SN
O1m
uA
discrete time,k
0 20 40 60 80 100 120-100
-50
0Sensitivity function SNO1muH
SN
O1m
uH
discrete time,k0 20 40 60 80 100 120
0
5
10Sensitivity function SNO1KS
SN
O1K
S
discrete time,k
0 20 40 60 80 100 120-20
-10
0Sensitivity function SNO1KOH
SN
O1K
OH
discrete time,k0 20 40 60 80 100 120
0
5
10Sensitivity function SNO1KNO
SN
O1K
NO
discrete time,k
0 20 40 60 80 100 120-100
0
100Sensitivity function SNO1KNH
SN
O1K
NH
discrete time,k0 20 40 60 80 100 120
-100
0
100Sensitivity function SNO1KOA
SN
O1K
OA
discrete time,k
0 20 40 60 80 100 120-20
-10
0Sensitivity function SNO1etag
SN
O1e
tag
discrete time,k0 20 40 60 80 100 120
-20
-10
0Sensitivity function SNO1etah
SN
O1e
tah
discrete time,k
0 20 40 60 80 100 120-2
-1
0Sensitivity function SNO1Kh
SN
O1K
h
discrete time,k0 20 40 60 80 100 120
0
2
4Sensitivity function SNO1Kx
SN
O1K
x
discrete time,k
0 20 40 60 80 100 1200
10
20Process variable SNO5
SN
O5
discrete time k0 20 40 60 80 100 120
0
0.2
0.4Sensitivity function SNO5f
SN
O5f
discrete time,k
0 20 40 60 80 100 120-20
-10
0Variable SNO5YH
SN
O5Y
H
discrete time,k0 20 40 60 80 100 120
-500
0
500Variable SNO5YA
SN
O5Y
A
discrete time,k
0 20 40 60 80 100 120
-400
-200
0Sensitivity function SNO5iXB
SN
O5i
XB
discrete time,k0 20 40 60 80 100 120
-500
0
500Sensitivity function SNO5muA
SN
O5m
uA
discrete time,k
0 20 40 60 80 100 120-200
-100
0Sensitivity function SNO5muH
SN
O5m
uH
discrete time,k0 20 40 60 80 100 120
0
5
10Sensitivity function SNO5KS
SN
O5K
S
discrete time,k
0 20 40 60 80 100 120-20
0
20Sensitivity function SNO5KOH
SN
O5K
OH
discrete time,k0 20 40 60 80 100 120
0
5
10Sensitivity function SNO5KNO
SN
O5K
NO
discrete time,k
0 20 40 60 80 100 120-200
0
200Sensitivity function SNO5KNH
SN
O5K
NH
discrete time,k0 20 40 60 80 100 120
-100
0
100Sensitivity function SNO5KOA
SN
O5K
OA
discrete time,k
0 20 40 60 80 100 120-20
-10
0Sensitivity function SNO5etag
SN
O5e
tag
discrete time,k0 20 40 60 80 100 120
-1
-0.5
0Sensitivity function SNO5etah
SN
O5e
tah
discrete time,k
0 20 40 60 80 100 120-2
-1
0Sensitivity function SNO5Kh
SN
O5K
h
discrete time,k0 20 40 60 80 100 120
0
2
4Sensitivity function SNO5Kx
SN
O5K
x
discrete time,k
Figure 6Process variable SNO5 and its sensitivity functions towards the model
parameters for the case of ASM1 reduced biological model
Figure 3Process variable SNH1 and its sensitivity functions towards the model
parameters for the case of ASM1 reduced biological model
Figure 4Process variable SNH5 and its sensitivity functions towards the model
parameters for the case of ASM1 reduced biological model
Figure 5Process variable SNO1 and its sensitivity functions towards the model
parameters for the case of ASM1 reduced biological model
initially as positive or negative but then at steady-states there is a change in their sign.
The behaviour of the sensitivity functions of the process variable SNOn, n=1, n=5, is similar to that of the variable SNHn,
KNH (120) for maximum and iXB (330), μA (-250) and μH (-58) for minimum values. The trajectories of the sensitivity functions are only positive or only negative for most of the parameters. The trajectories for the parameters YA, μA, KNH and KOA develop
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.zaISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012 299
n=1, n=5, but the dynamics of the sensitivity functions are slower, except for parameters YA, μA, KOH KNH, and KOA. The maximum or minimum values of the sensitivity functions are reached in a time interval between 15 and 20 h. The maximum
sensitivity is towards parameters YA (-400), iXB (-390), μA (200), KNH (-98), μH (-95) and KOA (-60) for the 1st tank, and YA (-495), iXB (-500), μA (220), KNH (-120), μH (-120) and KOA (-70) for the 5th tank.
0 20 40 60 80 100 120-1
0
1x 10
4 Process variable Snh5
SN
H5
discrete time k0 20 40 60 80 100 120
0
20
40Sensitivity function Snh5f
SN
H5f
discrete time,k
0 20 40 60 80 100 120-1
-0.5
0Variable Snh5YZH
SN
H5Y
ZH
discrete time,k0 20 40 60 80 100 120
0
200
400Variable Snh5YZA
SN
H5Y
ZA
discrete time,k
0 20 40 60 80 100 120-4000
-2000
0Sensitivity function Snh5fZBH
SN
H5f
ZB
H
discrete time,k0 20 40 60 80 100 120
-1000
-500
0Sensitivity function Snh5muA
SN
H5m
uA
discrete time,k
0 20 40 60 80 100 120
-40-20
0Sensitivity function Snh5muH
SN
H5m
uH
discrete time,k0 20 40 60 80 100 120
-1
0
1Sensitivity function Snh5KS
SN
H5K
S
discrete time,k
0 20 40 60 80 100 1200
100
200Sensitivity function Snh5KOH
SN
H5K
OH
discrete time,k0 20 40 60 80 100 120
-2
0
2x 10
-3 Sensitivity function Snh5KNO
SN
H5K
NO
discrete time,k
0 20 40 60 80 100 1200
100
200Sensitivity function Snh5KNH
SN
H5K
NH
discrete time,k0 20 40 60 80 100 120
0
500
1000Sensitivity function Snh5KOA
SN
H5K
OA
discrete time,k
0 20 40 60 80 100 120-1
-0.5
0Sensitivity function Snh5etag
SN
H5e
tag
discrete time,k0 20 40 60 80 100 120
0
100
200Sensitivity function Snh5KMP
SN
H5K
MP
discrete time,k
0 20 40 60 80 100 1200
100
200Sensitivity function Snh5KSP
SN
H5K
SP
discrete time,k0 20 40 60 80 100 120
-5
0
5Sensitivity function Snh5KSA
SN
H5K
SA
discrete time,k
0 20 40 60 80 100 1200
0.5
1x 10
-3 Sensitivity function Snh5fMA
SN
H5f
MA
discrete time,k0 20 40 60 80 100 120
0
0.02
0.04Sensitivity function Snh5KA
SN
H5K
A
discrete time,k
0 20 40 60 80 100 1200
5
10Process variable SS1
SS
1
discrete time k0 20 40 60 80 100 120
-0.04
-0.02
0Sensitivity function SS1f
SS
1f
discrete time,k
0 20 40 60 80 100 1200
100
200Variable SS1YH
SS
1YH
discrete time,k0 20 40 60 80 100 120
-2
0
2Variable SS1YA
SS
1YA
discrete time,k
0 20 40 60 80 100 1200
50
100Sensitivity function SS1iXB
SS
1iX
B
discrete time,k0 20 40 60 80 100 120
-1
0
1Sensitivity function SS1muA
SS
1muA
discrete time,k
0 20 40 60 80 100 120-400
-200
0Sensitivity function SS1muH
SS
1muH
discrete time,k0 20 40 60 80 100 120
0
5
10Sensitivity function SS1KS
SS
1KS
discrete time,k
0 20 40 60 80 100 120-100
-50
0Sensitivity function SS1KOH
SS
1KO
H
discrete time,k0 20 40 60 80 100 120
-20
-10
0Sensitivity function SS1KNO
SS
1KN
O
discrete time,k
0 20 40 60 80 100 120-0.5
0
0.5Sensitivity function SS1KNH
SS
1KN
H
discrete time,k0 20 40 60 80 100 120
-0.5
0
0.5Sensitivity function SS1KOA
SS
1KO
A
discrete time,k
0 20 40 60 80 100 1200
1
2Sensitivity function SS1etag
SS
1eta
g
discrete time,k0 20 40 60 80 100 120
0
1
2Sensitivity function SS1etah
SS
1eta
h
discrete time,k
0 20 40 60 80 100 1200
50
100Sensitivity function SS1Kh
SS
1Kh
discrete time,k0 20 40 60 80 100 120
-200
-100
0Sensitivity function SS1Kx
SS
1Kx
discrete time,k
0 20 40 60 80 100 1200
10
20Process variable SS5
SS
5
discrete time k0 20 40 60 80 100 120
-0.04
-0.02
0Sensitivity function SS5f
SS
5f
discrete time,k
0 20 40 60 80 100 1200
100
200Variable SS5YH
SS
5YH
discrete time,k0 20 40 60 80 100 120
-1
0
1Variable SS5YA
SS
5YA
discrete time,k
0 20 40 60 80 100 1200
50Sensitivity function SS5iXB
SS
5iX
B
discrete time,k0 20 40 60 80 100 120
-1
0
1Sensitivity function SS5muA
SS
5muA
discrete time,k
0 20 40 60 80 100 120-400
-200
0Sensitivity function SS5muH
SS
5muH
discrete time,k0 20 40 60 80 100 120
0
5
10Sensitivity function SS5KS
SS
5KS
discrete time,k
0 20 40 60 80 100 120-100
-50
0Sensitivity function SS5KOH
SS
5KO
H
discrete time,k0 20 40 60 80 100 120
-10
0
10Sensitivity function SS5KNO
SS
5KN
O
discrete time,k
0 20 40 60 80 100 120-0.5
0
0.5Sensitivity function SS5KNH
SS
5KN
H
discrete time,k0 20 40 60 80 100 120
-0.2
0
0.2Sensitivity function SS5KOA
SS
5KO
A
discrete time,k
0 20 40 60 80 100 1200
1
2Sensitivity function SS5etag
SS
5eta
g
discrete time,k0 20 40 60 80 100 120
0
20
40Sensitivity function SS5etah
SS
5eta
h
discrete time,k
0 20 40 60 80 100 1200
50
100Sensitivity function SS5Kh
SS
5Kh
discrete time,k0 20 40 60 80 100 120
-200
-100
0Sensitivity function SS5Kx
SS
5Kx
discrete time,k
0 20 40 60 80 100 120-5000
0
5000Process variable Snh1
SN
H1
discrete time k0 20 40 60 80 100 120
0
20
40Sensitivity function Snh1f
SN
H1f
discrete time,k
0 20 40 60 80 100 120-1
-0.5
0Variable Snh1YZH
SN
H1Y
ZH
discrete time,k0 20 40 60 80 100 120
0
200
400Variable Snh1YZA
SN
H1Y
ZA
discrete time,k
0 20 40 60 80 100 120-2000
-1000
0Sensitivity function Snh1fZBH
SN
H1f
ZB
H
discrete time,k0 20 40 60 80 100 120
-400-200
0Sensitivity function Snh1muA
SN
H1m
uA
discrete time,k
0 20 40 60 80 100 120-40
-20
0Sensitivity function Snh1muH
SN
H1m
uH
discrete time,k0 20 40 60 80 100 120
-0.5
0
0.5Sensitivity function Snh1KS
SN
H1K
S
discrete time,k
0 20 40 60 80 100 1200
100
200Sensitivity function Snh1KOH
SN
H1K
OH
discrete time,k0 20 40 60 80 100 120
-5
0
5x 10
-3 Sensitivity function Snh1KNO
SN
H1K
NO
discrete time,k
0 20 40 60 80 100 1200
100
200Sensitivity function Snh1KNH
SN
H1K
NH
discrete time,k0 20 40 60 80 100 120
0
500
1000Sensitivity function Snh1KOA
SN
H1K
OA
discrete time,k
0 20 40 60 80 100 120
-0.4-0.2
0Sensitivity function Snh1etag
SN
H1e
tag
discrete time,k0 20 40 60 80 100 120
0
50
100Sensitivity function Snh1KMP
SN
H1K
MP
discrete time,k
0 20 40 60 80 100 1200
100
200Sensitivity function Snh1KSP
SN
H1K
SP
discrete time,k0 20 40 60 80 100 120
-2
0
2Sensitivity function Snh1KSA
SN
H1K
SA
discrete time,k
0 20 40 60 80 100 1200
0.5
1x 10
-3 Sensitivity function Snh1fMA
SN
H1f
MA
discrete time,k0 20 40 60 80 100 120
0
0.02
0.04Sensitivity function Snh1KA
SN
H1K
A
discrete time,k
Figure 10Process variable SNH5 and its sensitivity functions towards the model
parameters for the case of UCT reduced biological model
Figure 7Process variable SS1 and its sensitivity functions towards the model
parameters for the case of ASM1 reduced biological model
Figure 8Process variable SS5 and its sensitivity functions towards the model
parameters for the case of ASM1 reduced biological model
Figure 9Process variable SNH1 and its sensitivity functions towards the model
parameters for the case of UCT reduced biological model
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.za
ISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012300
The trajectories of the sensitivity functions of the variable SSn, n=1, n=2, are characterised by the slowest dynamics and smallest maximum and minimum values. The maximum values
for the 1st tanks are for parameters YH (120), iXB (75), and kh (60), and the minimum values are for parameters μH (-220), KX (-180), and KOH (-90). For the 5th tank the corresponding values are YH
0 20 40 60 80 100 1200
5000Process variable SNO1
SN
O1
discrete time k0 20 40 60 80 100 120
0
5x 10
-6 Sensitivity function SNO1f
SN
O1f
discrete time,k
0 20 40 60 80 100 120-5
0
5Variable SNO1YZH
SN
O1Y
ZH
discrete time,k0 20 40 60 80 100 120
0
20
40Variable SNO1YZA
SN
O1Y
ZA
discrete time,k
0 20 40 60 80 100 120-0.2
0
0.2Sensitivity function SNO1fZBH
SN
O1f
ZB
H
discrete time,k0 20 40 60 80 100 120
-5
0
5Sensitivity function SNO1muA
SN
O1m
uA
discrete time,k
0 20 40 60 80 100 120-0.4
-0.2
0Sensitivity function SNO1muH
SN
O1m
uH
discrete time,k0 20 40 60 80 100 120
0
0.01
0.02Sensitivity function SNO1KS
SN
O1K
Sdiscrete time,k
0 20 40 60 80 100 120-1
-0.5
0Sensitivity function SNO1KOH
SN
O1K
OH
discrete time,k0 20 40 60 80 100 120
0
2
4x 10
-3 Sensitivity function SNO1KNOS
NO
1KN
O
discrete time,k
0 20 40 60 80 100 120-5
0
5Sensitivity function SNO1KNH
SN
O1K
NH
discrete time,k0 20 40 60 80 100 120
-2
0
2Sensitivity function SNO1KOA
SN
O1K
OA
discrete time,k
0 20 40 60 80 100 120-1
-0.5
0Sensitivity function SNO1etag
SN
O1e
tag
discrete time,k0 20 40 60 80 100 120
0
2
4Sensitivity function SNO1KMP
SN
O1K
MP
discrete time,k
0 20 40 60 80 100 120-0.5
0
0.5Sensitivity function SNO1KSP
SN
O1K
SP
discrete time,k0 20 40 60 80 100 120
-5
0
5Sensitivity function SNO1KSA
SN
O1K
SA
discrete time,k
0 20 40 60 80 100 120-0.4
-0.2
0Sensitivity function SNO1fMA
SN
O1f
MA
discrete time,k0 20 40 60 80 100 120
-20
-10
0Sensitivity function SNO1KA
SN
O1K
A
discrete time,k
0 20 40 60 80 100 1200
5000
10000Process variable SNO5
SN
O5
discrete time k0 20 40 60 80 100 120
0
2
4x 10
-6 Sensitivity function SNO5f
SN
O5f
discrete time,k
0 20 40 60 80 100 120-5
0
5Variable SNO5YZH
SN
O5Y
ZH
discrete time,k0 20 40 60 80 100 120
0
20
40Variable SNO5YZA
SN
O5Y
ZA
discrete time,k
0 20 40 60 80 100 120-0.2
0
0.2Sensitivity function SNO5fZBH
SN
O5f
ZB
H
discrete time,k0 20 40 60 80 100 120
-5
0
5Sensitivity function SNO5muA
SN
O5m
uA
discrete time,k
0 20 40 60 80 100 120-0.4
-0.2
0Sensitivity function SNO5muH
SN
O5m
uH
discrete time,k0 20 40 60 80 100 120
0
0.01
0.02Sensitivity function SNO5KS
SN
O5K
S
discrete time,k
0 20 40 60 80 100 120-1
0
1Sensitivity function SNO5KOH
SN
O5K
OH
discrete time,k0 20 40 60 80 100 120
0
2
4x 10
-3 Sensitivity function SNO5KNO
SN
O5K
NO
discrete time,k
0 20 40 60 80 100 120-200
-100
0Sensitivity function SNO5KNH
SN
O5K
NH
discrete time,k0 20 40 60 80 100 120
-5
0
5Sensitivity function SNO5KOA
SN
O5K
OA
discrete time,k
0 20 40 60 80 100 120-1
-0.5
0Sensitivity function SNO5etag
SN
O5e
tag
discrete time,k0 20 40 60 80 100 120
-5
0
5Sensitivity function SNO5KMP
SN
O5K
MP
discrete time,k
0 20 40 60 80 100 120-0.5
0
0.5Sensitivity function SNO5KSP
SN
O5K
SP
discrete time,k0 20 40 60 80 100 120
-5
0
5Sensitivity function SNO5KSA
SN
O5K
SA
discrete time,k
0 20 40 60 80 100 120
-0.4-0.2
0Sensitivity function SNO5fMA
SN
O5f
MA
discrete time,k0 20 40 60 80 100 120
-20
-10
0Sensitivity function SNO5KA
SN
O5K
A
discrete time,k
0 20 40 60 80 100 1200
10
20Process variable SS1
SS
1
discrete time k0 20 40 60 80 100 120
-0.5
0x 10
-10 Sensitivity function SS1f
SS
1f
discrete time,k
0 20 40 60 80 100 1200
20
40Variable SS1YZH
SS
1YZ
H
discrete time,k0 20 40 60 80 100 120
-2
-1
0x 10
-9 Variable SS1YZA
SS
1YZ
A
discrete time,k
0 20 40 60 80 100 1200
0.5
1x 10
-9 Sensitivity function SS1fZBH
SS
1fZ
BH
discrete time,k0 20 40 60 80 100 120
-5
0
5x 10
-9 Sensitivity function SS1muA
SS
1muA
discrete time,k
0 20 40 60 80 100 120-1000
-500
0Sensitivity function SS1muH
SS
1muH
discrete time,k0 20 40 60 80 100 120
0
20
40Sensitivity function SS1KS
SS
1KS
discrete time,k
0 20 40 60 80 100 1200
0.02
0.04Sensitivity function SS1KOH
SS
1KO
H
discrete time,k0 20 40 60 80 100 120
0
0.05
0.1Sensitivity function SS1KNO
SS
1KN
O
discrete time,k
0 20 40 60 80 100 1200
2
4x 10
-3 Sensitivity function SS1KNH
SS
1KN
H
discrete time,k0 20 40 60 80 100 120
-2
-1
0x 10
-9 Sensitivity function SS1KOA
SS
1KO
A
discrete time,k
0 20 40 60 80 100 1200
5x 10
-11 Sensitivity function SS1etag
SS
1eta
g
discrete time,k0 20 40 60 80 100 120
-4
-2
0x 10
-11 Sensitivity function SS1KMP
SS
1KM
P
discrete time,k
0 20 40 60 80 100 120
-4-20
x 10-11 Sensitivity function SS1KSP
SS
1KS
P
discrete time,k0 20 40 60 80 100 120
0
1
2x 10
-9 Sensitivity function SS1KSA
SS
1KS
A
discrete time,k
0 20 40 60 80 100 1200
2
4x 10
-12 Sensitivity function SS1fMAS
S1f
MA
discrete time,k0 20 40 60 80 100 120
0
1
2x 10
-10 Sensitivity function SS1KA
SS
1KA
discrete time,k
0 20 40 60 80 100 1200
2
4Process variable SS5
SS
5
discrete time k0 20 40 60 80 100 120
-4
-2
0x 10
-11 Sensitivity function SS5f
SS
5f
discrete time,k
0 20 40 60 80 100 1200
50Variable SS5YZH
SS
5YZ
H
discrete time,k0 20 40 60 80 100 120
-1
-0.5
0x 10
-9 Variable SS5YZA
SS
5YZ
A
discrete time,k
0 20 40 60 80 100 1200
2
4x 10
-10 Sensitivity function SS5fZBH
SS
5fZ
BH
discrete time,k0 20 40 60 80 100 120
-2
0
2x 10
-9 Sensitivity function SS5muA
SS
5muA
discrete time,k
0 20 40 60 80 100 120-1000
-500
0Sensitivity function SS5muH
SS
5muH
discrete time,k0 20 40 60 80 100 120
0
20
40Sensitivity function SS5KS
SS
5KS
discrete time,k
0 20 40 60 80 100 1200
0.01
0.02Sensitivity function SS5KOH
SS
5KO
H
discrete time,k0 20 40 60 80 100 120
0
0.05
0.1Sensitivity function SS5KNO
SS
5KN
O
discrete time,k
0 20 40 60 80 100 1200
2
4x 10
-3 Sensitivity function SS5KNH
SS
5KN
H
discrete time,k0 20 40 60 80 100 120
-1
-0.5
0x 10
-9 Sensitivity function SS5KOA
SS
5KO
A
discrete time,k
0 20 40 60 80 100 1200
2
4x 10
-11 Sensitivity function SS5etag
SS
5eta
g
discrete time,k0 20 40 60 80 100 120
-4
-2
0x 10
-11 Sensitivity function SS5KMP
SS
5KM
P
discrete time,k
0 20 40 60 80 100 120-4
-2
0x 10
-11 Sensitivity function SS5KSP
SS
5KS
P
discrete time,k0 20 40 60 80 100 120
0
1
2x 10
-9 Sensitivity function SS5KSA
SS
5KS
A
discrete time,k
0 20 40 60 80 100 1200
2
4x 10
-12 Sensitivity function SS5fMA
SS
5fM
A
discrete time,k0 20 40 60 80 100 120
0
1
2x 10
-10 Sensitivity function SS5KA
SS
5KA
discrete time,k
Figure 14rocess variable SS5 and its sensitivity functions towards the model
parameters for the case of UCT reduced biological model
Figure 11Process variable SNO1 and its sensitivity functions towards the model
parameters for the case of UCT reduced biological model
Figure 12Process variable SNO5 and its sensitivity functions towards the model
parameters for the case of UCT reduced biological model
Figure 13Process variable SS1 and its sensitivity functions towards the model
parameters for the case of UCT reduced biological model
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.zaISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012 301
On the basis of the above it can be concluded that model behaviour is sensitive to parameters YA, YH, iXB, μA, KNH, μH, and KX. The sensitivity of variables SNHn and SNOn, n=1, n=5, is higher than that of variables SSn, n=1, n=5, for parameters YA, iXB, μA, and KNH. The sensitivity of the variable SSn, n=1, n=5, is greater for parameters YH, μH, and KX. Some of these para-meters can be selected for parameter estimation in order to fit the reduced model to the process data.
Sensitivity function simulation for the benchmark process model based on the reduced UCT biological model
The sensitivity functions of the variable SNHn, n=1, n=5, behave exponentially, with the exception of the sensitivity function for parameter KNO, which displays a small overshoot and then is reduced to a low steady-state value. The steady state maxi-mum/minimum values of the sensitivity functions for the 1st tank are f (35), YZA (275), iXB (-1900), μH (-39), KOH (190), KNH (110), KOA (500), μA (-500), KMP (95), and KSP (105). For the 5th tank the corresponding values are f (35), YZA (295), iXB (-1900), μH (-39), KOH (190), KNH (120), KOA (500), μA (-500), KMP (105), and KSP (125). It can be seen that the sensitivity functions have a time delay and their values are greater for the 5th tank.
For the sensitivity functions of variable SNOn, n=1, n=5, a couple of overshoots at the beginning are then followed by a slow approach to the steady state. The extrema of the sensitiv-ity functions for the parameters are as follows: YZA (35), KNH (4), KSA (-17) for the 1st tank; and YZA (35), KNH (-120), KSA (-18) for the 5th tank. It can be seen that the values of the sensitivity functions in this case are relatively smaller in comparison with the corresponding ones for the first 4 tanks. It is only the influ-ence of parameter KNH which is stronger in the 5th tank.
For variable SSn, n=1, n=5, the sensitivity function displays very low values, and many overshoots for the first 5 hours, with time delays of these trajectories for the 5th tank. After this the trajectories approach steady state. The extrema of the sensitivity functions for the parameters are as follows: YZH (38), μH (-550), and KS (33) for the 1st tank; and YZH (45), μH (-700), and KS (35) for the 5th tank. The variable SSn, n=1, n=5, is not sensitive to the parameters to which otherprocess variables are sensitive.
The behaviour of the sensitivity functions of variable Sadsn, n=1, n=5, is similar to that for variable SSn, n=1, n=5. The extrema of the sensitivity functions for the parameters are as follows: YZH (4 000), KOH (7.5), KMP (2 000), KSP (700/-250), fmA (300), and KA (12 000) for the 1st tank, and YZH (4900), KOH (7.5), KMP (2 500), KSP (750/-780), fmA (350), and KA (18 000) for the 5th tank. This variable is very sensitive to the above parameters and extremely insensitive to the rest of the parameters.
Based on the results obtained it can be concluded that the process variables are sensitive to different parameters. Some of these parameters, such as YZH, KMP, KSP, KA, μH, iXB, and KOA, can be selected for estimation in order to fit the reduced model to the process data. The reduced UCT model variables have very low sensitivity to some of the parameters and very high sensitivity to the rest of the parameters. It is difficult to find a group of parameters which result in high values of the sensitiv-ity functions for all variables of the model.
Conclusion
The objective of this study was to describe the development of a sensitivity analysis direct method for developing a reduced
0 20 40 60 80 100 120-500
0
500Process variable Sads1
Sad
s1
discrete time k0 20 40 60 80 100 120
-4
-2
0x 10
-9 Sensitivity function Sads1f
Sad
s1f
discrete time,k
0 20 40 60 80 100 1200
2000
4000Variable Sads1YZH
Sad
s1Y
ZH
discrete time,k0 20 40 60 80 100 120
-1
-0.5
0x 10
-7 Variable Sads1YZA
Sad
s1Y
ZA
discrete time,k
0 20 40 60 80 100 1200
2
4x 10
-8 Sensitivity function Sads1fZBH
Sad
s1fZ
BH
discrete time,k0 20 40 60 80 100 120
-1
-0.5
0x 10
-7 Sensitivity function Sads1muA
Sad
s1m
uA
discrete time,k
0 20 40 60 80 100 1200
1
2x 10
-9 Sensitivity function Sads1muH
Sad
s1m
uH
discrete time,k0 20 40 60 80 100 120
-4
-2
0x 10
-11 Sensitivity function Sads1KS
Sad
s1K
S
discrete time,k
0 20 40 60 80 100 1200
5
10Sensitivity function Sads1KOH
Sad
s1K
OH
discrete time,k0 20 40 60 80 100 120
0
0.02
0.04Sensitivity function Sads1KNO
Sad
s1K
NO
discrete time,k
0 20 40 60 80 100 1200
0.2
0.4Sensitivity function Sads1KNH
Sad
s1K
NH
discrete time,k0 20 40 60 80 100 120
-1
-0.5
0x 10
-7 Sensitivity function Sads1KOA
Sad
s1K
OA
discrete time,k
0 20 40 60 80 100 120-4
-2
0Sensitivity function Sads1etag
Sad
s1et
ag
discrete time,k0 20 40 60 80 100 120
-2000
-1000
0Sensitivity function Sads1KMP
Sad
s1K
MP
discrete time,k
0 20 40 60 80 100 120-1000
0
1000Sensitivity function Sads1KSP
Sad
s1K
SP
discrete time,k0 20 40 60 80 100 120
0
0.5
1x 10
-7 Sensitivity function Sads1KSA
Sad
s1K
SA
discrete time,k
0 20 40 60 80 100 1200
200
400Sensitivity function Sads1fMA
Sad
s1fM
A
discrete time,k0 20 40 60 80 100 120
0
1
2x 10
4 Sensitivity function Sads1KA
Sad
s1K
A
discrete time,k
0 20 40 60 80 100 120-500
0
500Process variable Sads5
Sad
s5
discrete time k0 20 40 60 80 100 120
-1
-0.5
0x 10
-9 Sensitivity function Sads5f
Sad
s5f
discrete time,k
0 20 40 60 80 100 1200
5000Variable Sads5YZH
Sad
s5Y
ZH
discrete time,k0 20 40 60 80 100 120
-1
-0.5
0x 10
-7 Variable Sads5YZA
Sad
s5Y
ZA
discrete time,k
0 20 40 60 80 100 1200
5x 10
-8 Sensitivity function Sads5fZBH
Sad
s5fZ
BH
discrete time,k0 20 40 60 80 100 120
-1
-0.5
0x 10
-7 Sensitivity function Sads5muA
Sad
s5m
uA
discrete time,k
0 20 40 60 80 100 1200
1
2x 10
-9 Sensitivity function Sads5muH
Sad
s5m
uH
discrete time,k0 20 40 60 80 100 120
-4
-2
0x 10
-11 Sensitivity function Sads5KS
Sad
s5K
S
discrete time,k
0 20 40 60 80 100 1200
5
10Sensitivity function Sads5KOH
Sad
s5K
OH
discrete time,k0 20 40 60 80 100 120
0
0.05Sensitivity function Sads5KNO
Sad
s5K
NO
discrete time,k
0 20 40 60 80 100 1200
0.5Sensitivity function Sads5KNH
Sad
s5K
NH
discrete time,k0 20 40 60 80 100 120
-1
-0.5
0x 10
-7 Sensitivity function Sads5KOA
Sad
s5K
OA
discrete time,k
0 20 40 60 80 100 120
-4-20
Sensitivity function Sads5etag
Sad
s5et
ag
discrete time,k0 20 40 60 80 100 120
-4000
-2000
0Sensitivity function Sads5KMP
Sad
s5K
MP
discrete time,k
0 20 40 60 80 100 120-1000
0
1000Sensitivity function Sads5KSP
Sad
s5K
SP
discrete time,k0 20 40 60 80 100 120
0
0.5
1x 10
-7 Sensitivity function Sads5KSA
Sad
s5K
SA
discrete time,k
0 20 40 60 80 100 1200
200
400Sensitivity function Sads5fMA
Sad
s5fM
A
discrete time,k0 20 40 60 80 100 120
0
1
2x 10
4 Sensitivity function Sads5KA
Sad
s5K
A
discrete time,k
Figure 15Process variable SSads1 and its sensitivity functions towards the model
parameters for the case of UCT reduced biological model
Figure 16Process variable SSads5 and its sensitivity functions towards the model
parameters for the case of UCT reduced biological model
(150), iXB (45), kh (60), μH (-220), KX (-190) and KOH (-50). The maximum and minimum values of the sensitivity functions in this case are for different parameters than those in the 2 vari-ables, SNHn and SNon, considered above.
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.za
ISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012302
model of the activated sludge process, characterised by a large number of parameters and insufficient data measurements. Derivation of sensitivity functions and augmented sensitiv-ity state-space models for the reduced mass balance COST benchmark model, based on reduced ASM1 and UCT biologi-cal models, was presented. Matlab software was developed and simulations provided. The results enable one to answer the question as to which of the parameters need to be estimated in order to fit the reduced model behaviour to the data. Selection of the candidate parameters for estimation was based on a com-parison of the minimum or maximum values of the correspond-ing sensitivity functions. The shapes of the sensitivity func-tions obtained for all variables and parameters of Tank 1 for the benchmark process model, based on the reduced ASM1 or reduced UCT biological models, are similar to the correspond-ing functions for Tank 5. The difference is in the minimum or maximum values of the sensitivity functions, which are greater for Tank 5 . Another difference is that the sensitivity function trajectories for the 5th tank have a time delay and are slower. Different combinations of parameters selected by the sensitiv-ity analysis were estimated and the trajectories of the estimated model showed a good fit to the measured data. The methods used, algorithms, Matlab software and results for parameter estimation are not discussed in this paper. The reduced models, if properly calibrated, can predict the process behaviour, with small errors. These models are usually developed for real-time control design as a response to inflow disturbances. Reduced model applications for detailed studies of process interaction between variables, kinetic parameters and output performance will not be as successful as the application of the full models.
The main contribution of the study reported herein is that it provides a quantitative basis for classification of the model parameters. This helps to address the problems of complexity and uncertainty during the process of parameter estimation for nonlinear models. The proposed structure of the augmented sensitivity model is general and modular which permits its application to any wastewater treatment process and to any other nonlinear model. The vector/matrix structure of the model allows effective use of Matlab software and rapid solu-tion of the simulation sensitivity problem. This means that the proposed structure of the augmented model assures global sensitivity analysis without increasing the computation burden.
The developed algorithms and software can be used for different applications: to simplify models, to investigate robust-ness of the proposed parameter estimation results, to analyse different scenarios for the plant operation, to determine the region of the parameter space for which the output is minimum or maximum, to discover interactions between the input factors (variables and parameters), to discover issues not anticipated at the beginning of the investigation, and to use in real-time joint solution of the problem of parameter estimation, control design and implementation, as a part of an adaptive control strategy.
Acknowledgements
This study was funded by KISC NRF Grant No. 63667 ’Coordinated and joint research activities in wastewater treatment monitoring, modelling, estimation and control in real-time’.”
References
AL-SAGGAF UM and FRANKLIN GF (1988) Model reduction via balanced realizations: an extension and frequency weighting tech-niques. IEEE Trans. Autom. Control AC-33 (7) 687-692.
BECK C, DOYLE J AND GLOVER K (1996) Model reduction of multi-dimensional and uncertain systems. IEEE Trans. Autom. Control 41 (10) 1466-1477.
BREYFOGLE FW III (2003) Implementing Six Sigma: Smart Solution using Statistical Methods (2nd edn.). John Wiley and Sons, New York.
COPP J (2002) COST Action 624 - The Cost Simulation Benchmark: Description and Simulation Manual. Office for Official Publications of the European Union, Luxembourg.
DE PAUW D and VANROLLEGHEM P (2003) Practical aspects of sensitivity analysis for dynamic models. URL: http://www.bio-math.rug.ac.be (Accessed 12 June 2009).
DOCHAIN D (2003) State and parameter estimation in the chemi-cal and biochemical processes: a tutorial. J. Process Control 13 801-818.
DOLD P, EKAMA G and MARAIS G (1980) A general model for the activated sludge process. Prog. Water Technol. 12 47-77.
DU PLESSIS SC (2009) The investigation of the process parameters and developing a mathematical model for purposes of control design and implementation of the wastewater treatment process. D.Tech. thesis, Cape Peninsula University of Technology, Cape Town.
EKAMA G, MARAIS G, PITMAN A, KEAY, G, BUCHAN L, GERBER A and SMOLLEN M (1984) Theory, Design and Operation of Nutrient Removal Activated Sludge Process. WRC Report No. TT 16/84. Water Research Commission, Pretoria.
EKAMA G and MARAIS G (1979) Dynamic behaviour of the acti-vated sludge process. J. Water Pollut. Control Fed. 51 534-556.
FESSO A (2007) Statistical sensitivity analysis and water quality. In: Wymer I (ed.) Statistical Framework for Water Quality Criteria and Monitoring. Wiley, New York.
FRANK P (1978) Introduction to System Sensitivity Theory. Academic Press, London.
GLOVER K (1984) All optimal Hankel-norm approximations of linear multivariable systems and their L∞ error bounds. Int. J. Control 39 (6) 1115-1193.
HALEVI Y, ZLOCHEVSKY A and GILAT T (1997) Parameter-dependent model order reduction. Int. J. Control 66 (3) 463-485.
HENZE M, GRADY J, GUJER W, MARAIS G and MATSUO T (1987) Activated sludge model No. 1. IAWPRC Scientific and Technical Report No. 1, IAWPRC task group on mathematical modeling for design and operation of biological wastewater treat-ment, London.
HOLMBERG A and RANTA J (1982) Procedures for parameter and state estimation of microbial growth process models. Automatica 18 181-193.
JEPPSSON U (1996) Modeling Aspects of Wastewater Treatment Processes. Thesis, Lund University, Lund, Sweden.
LATHAM A and ANDERSON BDO (1985) Frequency-weighted opti-mal Hankel-norm approximation of stable transfer functions. Syst. Control Lett. 5 229-236.
LENNOX B, MONTAGUE GA, HIDEN HG, KORNFELD G and GOULDING PR (2001) Process monitoring of industrial fed-batch fermentation. Biotechnol. Bioeng. 74 (2) 125-135.
LOURIES A, ZCELY, CSIKOSZ-NAGY, TURANYI T and NOVAK B (2008) Analysis of a budding yeast cell cycle model using the shapes of local sensitivity functions. Int. J. of Chem. Kinet. 40 (11) 710-720.
MONTGOMERY DC (1997) The Use of Statistical Process Control and Design of Experiments in Product and Process Improvement. IIE Trans. 24 479.
MOORE BC (1981) Principal component analysis in linear system: Controllability, observability and model reduction. IEEE Trans. Autom. Control AC-26 17-32.
MUSSATI M, GERNAEY K, GANI R and JØRGENSEN S (2002) Computer aided model analysis and dynamic simulation of a waste-water treatment plant. Clean Technol. Environ Polic. 4 100-114.
NOYKOVA N and GYLLENBERG M (2000) Sensitivity analysis and parameter estimation in a model of anaerobic waste water treatment processes with substrate inhibition. Bioprocess Eng. 23 343-349.
PERTEV C, TURKER M and BERBER R (1997) Dynamic modeling, sensitivity analysis and parameter estimation of industrial yeast
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.zaISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012 303
fermenters. Comp. Chem. Eng. 21 739-744.PLAZL I, PIPUŠ G, DROLKA M, and KOLOINI T (1999) Parametric
sensitivity and evaluation of a dynamic model for Single-stage wastewater treatment plant. Acta Chim. Slov. 46 (2) 289-300.
SAFANOV MG and CHIANG RY (1989) A Schur method for bal-anced-truncation method reduction. IEEE Autom. Control AC-34 729-733.
SATO J and OHMORI H (2002) Sensitivity analysis and parameter identification of wastewater treatment based on Activated Sludge Model No. 1 (ASM 1). Proceedings of SICE 2002, 5-7 August 2002, Osaka. 434-439.
SCHERMANN PS and GARAG-GABIN W (2005) Process Gain, Time Lags and Reaction Curves. In: Liptak BG (ed.) Instrument Engineer’s Handbook: Process Control and Optimization. Pergamon Press, New York.
SMETS I, BERNAERTS K, SUN J, MARCHAL K, VANDER-LEYDEN J and VAN IMPE J (2002) Sensitivity function-based model reduction. A bacterial gene expression case study. Biotech. Bioeng. 80 195-200.
STECHA J, CEPAK M, PEKAR J and PACHNER D (2005) System Parameter estimation using p-norm minimization. Preprints from Proceedings of the 16th IFAC World Congress, 4-8 July 2005, Prague, Czech Republic.
TSONEVA R, POPCHEV I and PATARINSKA T (1996) Minimum time sensitive control of nonlinear time delay systems with slowly varying parameters. Proc. 13th IFAC World Congress, 1 April 1997, San Francisco. E 371-376.
VARMA A, MORBIDELLI M and WU H (1999) Parametric Sensitivity in Chemical Systems. Cambridge University Press, Cambridge.
WANG L and GAWTHROP PJ (2001) On the estimation of continuous time transfer functions. Int. J. Control 74 889-904.
WENTZEL M, EKAMA G and MARAIS G (1992) Processes and modelling of nitrification, denitrification and biological excess phosphorus removal systems. Water Sci. Technol. 25 (6) 59-82.
WISNEWSKI PA and DOYLE III FJ (1996) A reduced model approach to estimation and control of a Kamyr digester. Comput. Chem. Eng. 20 1053-1058.
Appendix AThe derivatives of the variables’ rates for ASM1 reduced biological model towards the model variables and parameters are determined as follows:
1. The variables’ rate derivatives towards the process variables:
• for the variable’s rate rSNH,n
• for the variable’s rate rSNO,n
• for the variable’s rate rSS,n
2. The variables’ rate derivatives towards the model para-meters. The partial derivatives of the process rates according to the model coefficients
are as follows:
• for the variable’s rate rSNH,n
nSNHnSNH
nNHNH
NH
nOOA
nOBAA
AXB
nNH
nSNH rSK
KSK
SX
Yi
Sr ,
,2,,
,
,
, ˆ1
nSNOnSNH
nNONO
NOgBH
nOOH
OH
nSS
SHXB
nNO
nSNH rSK
KX
SKK
SKS
iSr ,
,2,,,,
, ˆ
nSSnSNH
nSS
S
nOOH
nOBHHXB
nS
nSNH rSKK
SKS
XiSr ,
,2,,
,
,
, ˆ
nNHnSNO
nNHNH
NH
nOOA
nOBAA
AnNH
nSNO rSK
KSK
SX
YSr ,
,2,,
,
,
, ˆ1
nNOnSNO
nNONO
NO
nOOH
OH
nSS
nSgHBH
H
H
nNO
nSNO rSKK
SKK
SKS
XYY
Sr ,
,2,,,
,
,
, ˆ86.2
1
nSSnSNO
nSS
S
nNONO
nNO
nOOH
OHgHBH
H
H
nS
nSNO rSKK
SKS
SKK
XYY
Sr ,
,2,,
,
,,
, ˆ86.2
1
0,
,
nNH
nSS
Sr
nSNO
nSSnNONO
NO
nOOH
OHBH
nBHnSX
nBHnShh
nSS
nSH
HnNO
nSS rSK
KSK
KXXXK
XXk
SKS
YSr ,
,2,,,,
,,
,
,
,
, ˆ1
nSSnSS
nSS
S
nOOH
OH
nNONO
nNO
nOOH
nOnBHH
HnS
nSS rSK
KSK
KSK
SSK
SX
YSr ,
,,,,
,
,
,,
,
, ˆ1
0, fnSNHr 0, HY
nSNHr , 0, hnSNHr
, 0, hknSNHr , 0, XK
nSNHr
nBA
nNHNH
nNH
nOOA
nOA
A
YnSNH X
SK
S
SK
S
Yr A
,,
,
,
,2, ˆ1
nBAnNHNH
nNH
nOOA
nOA
gnOOH
OH
nNONO
nNO
nOOH
nO
nSS
nSnBHH
inSNH
XSK
SSK
S
SKK
SKS
SKS
SKS
Xr XB
,,
,
,
,
,,
,
,
,
,
,,,
ˆ
ˆ
nBAnNHNH
nNH
nOOA
nO
AXBnSNH X
SKS
SKS
Yir A
,,
,
,
,ˆ,
1
nOOH
OH
nNONO
nNOg
nOOH
nO
nSS
nSnBHXBnSNH SK
KSK
S
SK
S
SK
SXir H
,,
,
,
,
,
,,
ˆ,
nOOH
OH
nNONO
nNOg
nOOH
nO
nSS
nSnBHHXB
KnSNH SK
KSK
SSK
S
SK
SXir S
,,
,
,
,2
,
,,, ˆ
nNONO
nNOg
nOOH
nO
nSS
nS
nBHHXB
KnSNH SK
S
SK
S
SKS
Xir OH
,
,2
,
,
,
,
,, 1ˆ
nOOH
OH
nNONO
nNO
nSS
nSgnBHHXB
KnSNH SK
KSK
SSK
SXir NO
,2
,
,
,
,,, ˆ
nBAnNHNH
nNH
nOOA
nOA
AXB
KnSNH X
SK
S
SK
S
Yir NH
,2,
,
,
,, ˆ1
BAnNHNH
nNH
nOOA
nOA
AXB
KnSNH X
SKS
SK
SY
ir OA
,
,2
,
,, ˆ1
nBHnOOH
OH
nNONO
nNO
nSS
nSHXBnSNH X
SKK
SK
S
SK
Sir g
,,,
,
,
,, ˆ
XhhgOANHNOOHSHAXBAH KkKKKKKiYYf ,,,,,,,,,,,,,,
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.za
ISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012304
• for the variable’s rate rSNO,n • for the variable’s rate rSS,n
2,,
,
,,
,,, 86.2
1ˆHnNOnNO
nNO
nOOH
OH
nSS
nSnBHgH
YnSNO YSK
S
SKK
SK
SXr H
AA Y
nSNHnBAnNHNH
nNH
nOOA
nOA
A
YnSNO rX
SK
S
SK
S
Yr ,,
,
,
,
,2, ˆ1
0, f
nSNOr , 0, XBinSNOr 0, h
nSNOr , 0, hnSNOr , 0, hk
nSNOr , 0, XKnSNOr
nBAnNHNH
nNH
nOOA
nO
AnSNO X
SK
S
SK
S
Yr A
,,
,
,
,ˆ,
1
nOOH
OH
nNOnNO
nNO
nOOH
nO
nSS
nSnBHH
H
YnSS SK
KSK
SSK
SSK
SX
Yr H
,,,
,
,
,
,
,,2, ˆ1
gnNONO
nNO
nOOH
OH
nSS
nSnBH
H
HnSNO SK
S
SKK
SK
SX
YY
r H
,
,
,,
,,
ˆ, 86.2
1
g
nNONO
nNO
nOOH
OH
nSS
nSHnBH
H
HKnSNO SK
SSK
KSK
SX
YY
r S
,
,
,,
,,, ˆ
86.21
gnNONO
nNO
nOOH
nO
nSS
nSHnBH
H
HKnSNO SK
S
SK
SSK
SX
YY
r OH
,
,2
,
,
,
,,, ˆ
86.21
nOOH
OH
nNONO
nNO
nSS
nSgnBHH
H
HKnSNO SK
K
SK
S
SK
SX
YY
r NO
,2,
,
,
,,, ˆ
86.21
nBAnNHNH
nNH
nOOA
nOA
A
KnSNO X
SK
S
SK
S
Yr NH
,2,
,
,
,, ˆ1
BAnNHNH
nNH
nOOA
nOA
A
KnSNO X
SKS
SKS
Yr OA
,
,2
,
,, ˆ1
nBH
nOOH
OH
nNONO
nNO
nSS
nSH
H
HnSNO X
SKK
SK
S
SK
S
YY
r g,
,,
,
,
,, ˆ
86.21
Appendix B
The derivatives of the variables’ rates for the UCT reduced biological model towards the model variables and parameters are determined as follows:
nOOH
OH
nNOnNO
nNO
nOOH
nO
nSS
nSnBHH
H
YnSS SK
KSK
SSK
SSK
SX
Yr H
,,,
,
,
,
,
,,2, ˆ1
0, f
nSSr , 0, AYnSSr
, 0, XBinSSr
, 0ˆ, AnSSr
, 0, NHKnSSr
, 0, OAKnSSr
, 0, gnSSr
nOOH
OH
nNONO
nNO
nOOH
nO
nSS
nSnBH
HnSS SK
KSK
SSK
S
SKS
XY
r H
,,
,
,
,
,
,,
ˆ,
1
nOOH
OH
nNONO
nNO
nOOH
nO
nSS
nSnBHH
H
KnSS SK
KSK
SSK
S
SK
SX
Yr S
,,
,
,
,2
,
,,, ˆ1
nNONO
nNOh
nOOH
nO
nBHnSX
nBHnS
nBHhnNONO
nNO
nOOH
nO
nSS
nSnBHH
H
KnSS
SKS
SK
SXXK
XX
XkSK
S
SK
SSK
SX
Yr OH
,
,2
,
,
,,
,,
,,
,2
,
,
,
,,,
1.
.1ˆ1
nOOH
OH
nNONO
nNO
nBHnSX
nBHnS
hnBHhnOOH
OH
nNONO
nNO
nSS
nSnBHH
H
KnSS
SKK
SK
SXXK
XX
XkSK
KSK
SSK
SX
Yr NO
,2
,
,
,,
,,
,,
2,
,
,
,,,
.
.ˆ1
nBHnOOH
OH
nNONO
nNO
nOOH
nO
nBHnSX
nBHnShnSS X
SKK
SKS
SKS
XXKXX
kr h,
,,
,
,
,
,,
,,,
nBHnOOH
OH
nNONO
nNOh
nOOH
nO
nBHnSX
nBHnSknSS X
SKK
SKS
SKS
XXKXX
r h,
,,
,
,
,
,,
,,,
2,,
,,,
,,
,
,
,,
nBHnSX
nBHnSnBH
nOOH
OH
nNONO
nNOh
nOOH
nOh
KnSS XXK
XXX
SKK
SKS
SKS
kr X
1. The process rate derivatives towards the process variables • for the variable’s rate rSNH,n
• for the variable’s rate rSNO,n
BAnNHSA
SA
nOOA
nOAZBH
ZA
BHnadsSP
nBHnadsgMP
nSS
nSH
nNONO
nNO
nOOH
OH
nBHnadsSP
nBHnadsMP
nSS
nSH
nOOH
nO
nNHNA
NABHZBH
nSNHnSNH
XSKK
SKS
fY
XSKXS
KSK
SSK
SSK
K
XSKXS
KSK
SSK
S
SKK
Xfr
2,,
,
,
,,
,
,
,
,
,
,,
,,
,
,
,
,2
,
,,
1
][
BHnadsSP
BHnadsMPg
nSS
nSH
nNHNA
nNH
nOOH
OH
nNONO
NOBHZBH
nSNOnSNH
XSKXS
KSK
S
SKS
SKK
SKKXfr
,
,
,
,
,
,
,2
,
,,
.
.
nNONO
nNO
nOOH
OH
nOOH
nO
nNHNA
nNHBH
nSS
SHZBH
nSSnSNH SK
SSK
KSK
SSK
SX
SKKfr
,
,
,,
,
,
,2
,
,,
nNONO
nNO
nOOH
OHg
nOOH
nO
nNHNA
nNHBH
nBHnadsSP
nBHSPMPZBH
nSadsnSNH
SKS
SKK
SKS
SKS
XXSK
XKKfr
,
,
,,
,
,
,2
,,
,,,
.
.
BAnNHSA
SA
nOOA
nOAZBH
ZA
BHnadsSP
nBHnadsgMP
nSS
nSH
nNONO
nNO
nOOH
OH
nBHnadsSP
nBHnadsMP
nSS
nSH
nOOH
nO
nNHNA
NABHZBH
nSNHnSNH
XSKK
SKS
fY
XSKXS
KSK
SSK
SSK
K
XSKXS
KSK
SSK
S
SKK
Xfr
2,,
,
,
,,
,
,
,
,
,
,,
,,
,
,
,
,2
,
,,
1
][
BHnadsSP
BHnadsMPg
nSS
nSH
nNHNA
nNH
nOOH
OH
nNONO
NOBHZBH
nSNOnSNH
XSKXS
KSK
S
SKS
SKK
SKKXfr
,
,
,
,
,
,
,2
,
,,
.
.
nNONO
nNO
nOOH
OH
nOOH
nO
nNHNA
nNHBH
nSS
SHZBH
nSSnSNH SK
SSK
KSK
SSK
SX
SKKfr
,
,
,,
,
,
,2
,
,,
nNONO
nNO
nOOH
OHg
nOOH
nO
nNHNA
nNHBH
nBHnadsSP
nBHSPMPZBH
nSadsnSNH
SKS
SKK
SKS
SKS
XXSK
XKKfr
,
,
,,
,
,
,2
,,
,,,
.
.
BAnNOOA
nNO
nNHSA
SAA
ZA
nBHnadsSP
nBHnadsgMP
nSS
nSHnBH
nNONO
nNO
nNHNA
NA
nOOH
OHZBH
nBHnadsSP
nBHnadsMP
nSS
nSHnBH
nNONO
nNO
nNHNA
NA
nOOH
nOZBH
nSNHnSNO
XSK
S
SKK
Y
XSKXS
KSK
SX
SKS
SKK
SKK
f
XSKXS
KSK
SX
SKS
SKK
SKS
fr
,
,2
,
,,
,,
,
,,
,
,2
,,
,,
,,
,
,,
,
,2
,,
,,,
.1
nNHNA
NAZBH
ZH
ZH
nNHNA
nNH
ZH
ZH
nBHnadsSP
nBHnadsgMP
nSS
nSHnBH
nNONO
NO
nOOH
OH
nBHnadsSP
nBHnadsMP
nSS
nSHnBH
nNONO
NO
nNHNA
NA
nOOH
nOZBH
nSNOnSNO
SKK
fYY
SKS
YY
XSKXS
KSK
SX
SKK
SKK
XSKXS
KSK
SX
SKK
SKK
SKS
fr
,,
,
,,
,,
,
,,2
,,
,,
,,
,
,,2
,,,
,,,
86.21
86.21
.
.
nNHNA
NA
nOOH
OHZBH
ZH
ZH
nNHNA
nNH
nOOH
OH
ZH
ZH
nNHNA
NA
nOOH
nOZBHnBH
nNONO
nNO
nSS
SH
nSSnSNO
SKK
SKKf
YY
SKS
SKK
YY
SKK
SKS
fXSK
SSKKr
,,
,
,
,
,,
,,
,
,2
,
,,
86.21
86.21
[
]86.2
1
86.21.[
.
,,
,
,
,,,
,
,,
,2
,,
gnNHNA
NA
nOOH
OHZBH
ZH
ZH
gnNHNA
nNH
nOOH
OH
ZH
ZH
nNHNA
NA
nOOH
nOZBH
nBHnNONO
nNO
BHSPSP
BHSPMP
nSadsnSNO
SKK
SKKf
YY
SKS
SKK
YY
SKK
SKS
f
XSK
SXKK
XKKr
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.zaISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012 305
• for the variable’s rate rSS,n
2. The variables’ rate derivatives towards the model para-meters. The partial derivatives of the variables’ rates according to the model coefficients
are as follows:
• for the variable’s rate rSNH,n
• for the variable’s rate rSNO,n
BAnNOOA
nNO
nNHSA
SAA
ZA
nBHnadsSP
nBHnadsgMP
nSS
nSHnBH
nNONO
nNO
nNHNA
NA
nOOH
OHZBH
nBHnadsSP
nBHnadsMP
nSS
nSHnBH
nNONO
nNO
nNHNA
NA
nOOH
nOZBH
nSNHnSNO
XSK
S
SKK
Y
XSKXS
KSK
SX
SKS
SKK
SKK
f
XSKXS
KSK
SX
SKS
SKK
SKS
fr
,
,2
,
,,
,,
,
,,
,
,2
,,
,,
,,
,
,,
,
,2
,,
,,,
.1
nNHNA
NAZBH
ZH
ZH
nNHNA
nNH
ZH
ZH
nBHnadsSP
nBHnadsgMP
nSS
nSHnBH
nNONO
NO
nOOH
OH
nBHnadsSP
nBHnadsMP
nSS
nSHnBH
nNONO
NO
nNHNA
NA
nOOH
nOZBH
nSNOnSNO
SKK
fYY
SKS
YY
XSKXS
KSK
SX
SKK
SKK
XSKXS
KSK
SX
SKK
SKK
SKS
fr
,,
,
,,
,,
,
,,2
,,
,,
,,
,
,,2
,,,
,,,
86.21
86.21
.
.
nNHNA
NA
nOOH
OHZBH
ZH
ZH
nNHNA
nNH
nOOH
OH
ZH
ZH
nNHNA
NA
nOOH
nOZBHnBH
nNONO
nNO
nSS
SH
nSSnSNO
SKK
SKKf
YY
SKS
SKK
YY
SKK
SKS
fXSK
SSKKr
,,
,
,
,
,,
,,
,
,2
,
,,
86.21
86.21
[
]86.2
1
86.21.[
.
,,
,
,
,,,
,
,,
,2
,,
gnNHNA
NA
nOOH
OHZBH
ZH
ZH
gnNHNA
nNH
nOOH
OH
ZH
ZH
nNHNA
NA
nOOH
nOZBH
nBHnNONO
nNO
BHSPSP
BHSPMP
nSadsnSNO
SKK
SKKf
YY
SKS
SKK
YY
SKK
SKS
f
XSK
SXKK
XKKr
nNONO
nNOnBH
nNHNA
NA
nOOH
nO
nSS
nSH
ZH
nSNHnSS SK
SX
SKK
SKS
SKS
Yr
,
,,2
,,
,
,
,,, 11
nNHNA
NA
nOOH
OH
nNHNA
nNH
nOOH
OH
nNHNA
NA
nOOH
nO
nBHnNONO
NO
nSS
nSH
ZH
nSNOnSS
SKK
SKK
SKS
SKK
SKK
SKS
XSK
KSK
SY
r
,,,
,
,
,,
,
,2,,
,,,
.
.1
nNHNA
NA
nNHNA
nNH
NONO
NO
OOH
OH
NONO
NO
nNHNA
NA
nNHNA
nNH
nOOH
nO
SS
S
ZH
nSSnSS
SKK
SKS
SKS
SKK
SKS
SKK
SKS
SKS
SKK
Yr
,,
,
,,
,
,
,
,, 2
1
0,, nSadsnSSr
nNONO
nNOnBH
nNHNA
NA
nOOH
nO
nBHnadsSP
nBHnadsMP
ZH
nSNHnSads SK
SX
SKK
SKS
XSKXS
KY
r,
,,2
,,
,
,,
,,,, 11
nNHNA
NA
nOOH
OH
nNHNA
nNH
nOOH
OH
nNHNA
NA
nOOH
nO
nBHnNONO
NO
nBHnadsSP
nBHnadsMP
ZH
nSNOnSads
SKK
SKK
SKS
SKK
SKK
SKS
XSK
KXSK
XSK
Yr
,,,
,
,
,,
,
,2,,,
,,,,
.
.1
0,
, nSSnSadsr
nNHNA
NA
nNHNA
nNHnBHg
nNONO
nNO
nOOH
OH
nBHnadsSP
nBHSPMP
ZHnNONO
nNO
nNHNA
NA
nNHNA
nNH
nBHnOOH
nO
nBHnadsSP
nBHSPMP
ZH
nSadsnSads
SKK
SKS
XSK
SSK
K
XSK
XKK
YSKS
SKK
SKS
XSK
S
XSK
XKK
Yr
,,
,,
,
,
,
2,,
,
,
,
,,
,
,,
,2
,,
,,,
.
.1.
.1
SASPMPgOANANOOHSHAZBHZAZH KKKKKKKKfYYf ,,,,,,,,,,,,,,
,0, fnSNHr
nBAnOOA
nO
nNHSA
nNHA
ZA
YnSNH
YnSNH X
SKS
SKS
Yrr ZAZH
,,
,
,
,,,
1,0
nBAnOOA
nO
nNHSA
nNHA
nBHnadsSP
nBHnadsgMP
nSS
nSH
nBHnNONO
nNO
nNHNA
nNH
nOOH
OH
nBHnadsSP
nBHnadsMP
nSS
nSHnBH
nNHNA
nNH
nOOH
nOfnSNH
XSK
SSK
S
XSKXS
K
SKS
XSK
SSK
SSK
K
XSKXS
KSK
SX
SKS
SKS
r ZBH
,,
,
,
,
,,
,,
,
,
,,
,
,
,
,
,,
,,
,
,,
,
,
,
,,
nNONO
nNO
nOOH
OH
nOOH
nO
nNHNA
nNHnBH
nSS
nSZBHnSNH SK
SSK
KSK
SSK
SX
SKS
fr H
,
,
,,
,
,
,,
,
,,
nBAnOOA
nO
nNHSA
nNHZBH
ZAnSNH X
SKS
SKS
fY
r A,
,
,
,
,,
1
nNONO
nNO
nOOH
OH
nOOH
nOnBH
nNHNA
nNH
nSS
nSHZBH
KnSNH SK
SSK
KSK
SX
SKS
SKS
fr S
,
,
,,
,,
,
,2
,
,,
nBHnadsSP
nBHnadsgMP
nSS
nSH
nBHnNONO
nNO
nNHNA
nNH
nOOH
nOZBH
nBHnadsSP
nBHnadsMP
nSS
nSHnBH
nNHNA
nNH
nOOH
nOZBH
KnSNH
XSKXS
KSK
S
XSK
SSK
S
SK
Sf
XSKXS
KSK
SX
SKS
SK
Sfr OH
,,
,,
,
,
,,
,
,
,2
,
,
,,
,,
,
,,
,
,2
,
,,
.
.
nBHnadsSP
nBHnadsgMP
nSS
nSHnBH
nNONO
nNO
nNHNA
nNH
nOOH
OHZBH
KnSNH XSK
XSK
SKS
XSK
SSK
SSK
Kfr NO
,,
,,
,
,,2
,
,
,
,
,,
nBHnadsSP
nBHnadsgMP
nSS
nSH
nBHnNONO
nNO
nNHNA
nNH
nOOH
OHZBH
nBHnadsSP
nBHnadsMP
nSS
nSHnBH
nNHNA
nNH
nOOH
nOZBH
KnSNH
XSKXS
KSK
S
XSK
S
SK
SSK
Kf
XSKXS
KSK
SX
SK
SSK
Sfr NA
,,
,,
,
,
,,
,2
,
,
,
,,
,,
,
,,2
,
,
,
,,
.
.
nBAnOOA
nO
nNHSA
nNHAZBH
ZA
KnSNH X
SKS
SKS
fY
r OA,2
,
,
,
,,
1
ZBHnBHnNONO
nNO
nNHNA
nNH
nOOH
OH
nBHnadsSP
nBHnadsMPnSNH fX
SKS
SKS
SKK
XSKXS
Kr g,
,
,
,
,
,,,
,,,
g
nNONO
nNO
nOOH
OH
nOOH
nOZBHnBH
nNHNA
nNH
nBHnadsSP
nBHnadsKnSNH SK
SSK
KSK
SfX
SKS
XSKXS
r MP ,
,
,,
,,
,
,
,,
,,,
nBHnNONO
nNO
nOOH
OH
nOOH
nO
nNHNA
nNH
nBHnadsSP
nBHnadsMPZBH
KnSNH X
SKS
SKK
SKS
SKS
XSKXS
Kfr SP,
,
,
,,
,
,
,
,,
,,,
nBAnOOA
nO
nNHSA
nNHAZBH
ZA
KnSNH X
SKS
SKS
fY
r SA,
,
,2
,
,,
1
0, fnSNOr , 0, MAf
nSNOr , 0, AKnSNOr , 0, XK
nSNOr
nNHNA
NA
nNHNA
nNH
nBHnadsSP
nBHnadsgMP
nSS
nSHnBH
nNONO
nNO
nOOH
OH
ZH
YnSNO
SKK
SKS
XSKXS
KSK
SX
SKS
SKK
Yr ZH
,,
,
,,
,,
,
,,
,
,
,2,
.
.86.2
1
,0, f
nSNHrnBA
nOOA
nO
nNHSA
nNHA
ZA
YnSNH
YnSNH X
SKS
SKS
Yrr ZAZH
,,
,
,
,,,
1,0
nBAnOOA
nO
nNHSA
nNHA
nBHnadsSP
nBHnadsgMP
nSS
nSH
nBHnNONO
nNO
nNHNA
nNH
nOOH
OH
nBHnadsSP
nBHnadsMP
nSS
nSHnBH
nNHNA
nNH
nOOH
nOfnSNH
XSK
SSK
S
XSKXS
K
SKS
XSK
SSK
SSK
K
XSKXS
KSK
SX
SKS
SKS
r ZBH
,,
,
,
,
,,
,,
,
,
,,
,
,
,
,
,,
,,
,
,,
,
,
,
,,
nNONO
nNO
nOOH
OH
nOOH
nO
nNHNA
nNHnBH
nSS
nSZBHnSNH SK
SSK
KSK
SSK
SX
SKS
fr H
,
,
,,
,
,
,,
,
,,
nBAnOOA
nO
nNHSA
nNHZBH
ZAnSNH X
SKS
SKS
fY
r A,
,
,
,
,,
1
nNONO
nNO
nOOH
OH
nOOH
nOnBH
nNHNA
nNH
nSS
nSHZBH
KnSNH SK
SSK
KSK
SX
SKS
SKS
fr S
,
,
,,
,,
,
,2
,
,,
nBHnadsSP
nBHnadsgMP
nSS
nSH
nBHnNONO
nNO
nNHNA
nNH
nOOH
nOZBH
nBHnadsSP
nBHnadsMP
nSS
nSHnBH
nNHNA
nNH
nOOH
nOZBH
KnSNH
XSKXS
KSK
S
XSK
SSK
S
SK
Sf
XSKXS
KSK
SX
SKS
SK
Sfr OH
,,
,,
,
,
,,
,
,
,2
,
,
,,
,,
,
,,
,
,2
,
,,
.
.
nBHnadsSP
nBHnadsgMP
nSS
nSHnBH
nNONO
nNO
nNHNA
nNH
nOOH
OHZBH
KnSNH XSK
XSK
SKS
XSK
SSK
SSK
Kfr NO
,,
,,
,
,,2
,
,
,
,
,,
nBHnadsSP
nBHnadsgMP
nSS
nSH
nBHnNONO
nNO
nNHNA
nNH
nOOH
OHZBH
nBHnadsSP
nBHnadsMP
nSS
nSHnBH
nNHNA
nNH
nOOH
nOZBH
KnSNH
XSKXS
KSK
S
XSK
S
SK
SSK
Kf
XSKXS
KSK
SX
SK
SSK
Sfr NA
,,
,,
,
,
,,
,2
,
,
,
,,
,,
,
,,2
,
,
,
,,
.
.
nBAnOOA
nO
nNHSA
nNHAZBH
ZA
KnSNH X
SKS
SKS
fY
r OA,2
,
,
,
,,
1
ZBHnBHnNONO
nNO
nNHNA
nNH
nOOH
OH
nBHnadsSP
nBHnadsMPnSNH fX
SKS
SKS
SKK
XSKXS
Kr g,
,
,
,
,
,,,
,,,
g
nNONO
nNO
nOOH
OH
nOOH
nOZBHnBH
nNHNA
nNH
nBHnadsSP
nBHnadsKnSNH SK
SSK
KSK
SfX
SKS
XSKXS
r MP ,
,
,,
,,
,
,
,,
,,,
nBHnNONO
nNO
nOOH
OH
nOOH
nO
nNHNA
nNH
nBHnadsSP
nBHnadsMPZBH
KnSNH X
SKS
SKK
SKS
SKS
XSKXS
Kfr SP,
,
,
,,
,
,
,
,,
,,,
nBAnOOA
nO
nNHSA
nNHAZBH
ZA
KnSNH X
SKS
SKS
fY
r SA,
,
,2
,
,,
1
nBAnOOA
nO
nNHSA
nNHA
ZH
YnSNO X
SK
SSK
SY
r ZA,2
,
,
,
,2, 86.2
1
nBHnadsSP
nBHnadsMP
nSS
nSHnBH
nNONO
nNO
nNHNA
NA
nOOH
nOfnSNO XSK
XSK
SKS
XSK
SSK
KSK
Sr ZBH
,,
,,
,
,,
,
,
,,
,,
nNHNA
NA
nNHNA
nNHnBH
nNONO
nNO
nOOH
OH
nSS
nS
ZH
ZH
nOOH
OH
nOOH
nO
nBHnNONO
nNO
nNHNA
NA
nSS
nSZBHnSNO
SKK
SKS
XSK
S
SKK
SKS
YY
SKK
SKS
XSK
SSK
KSK
Sfr H
,,
,,
,
,
,,
,
,,
,
,,
,
,,
,,
.
.86.2
1.
.2
nBAnOOA
nO
nNHSA
nNH
ZAnSNO X
SKS
SK
SY
r A,
,
,2
,
,,
1
nNHNA
NA
nNHNA
nNHnBH
nNONO
nNO
nOOH
OH
nSS
nSH
ZH
ZH
nOOH
OH
nOOH
nOnBH
nNONO
nNO
nNHNA
NA
nSS
nSHZBH
KnSNO
SKK
SKS
XSK
SSK
KSK
SYY
SKK
SKS
XSK
SSK
KSK
Sfr S
,,
,,
,
,
,2
,
,
,,
,,
,
,
,2
,
,,
86.21
gnBHnadsSP
nBHnadsMPnBH
nNONO
nNO
nNHNA
NA
nOOH
nOZBH
nBHnadsSP
nBHnadsgMP
nSS
nSHnBH
nNONO
nNO
nOOH
OH
ZH
ZHKnSNO
XSKXS
KXSK
SSK
KSK
Sf
XSKXS
KSK
SX
SKS
SKK
YY
r OH
1
86.21
,,
,,,
,
,
,2
,
,
,,
,,
,
,,
,
,2
,,
nBHnadsSP
nBHnadsgMP
nSS
nSHnBH
nNOOH
nNO
nNHNA
NA
nOOH
OHZBH
ZH
ZH
nBHnadsSP
nBHnadsgMP
nSS
nSHnBH
nNONO
nNO
nNHNA
nNH
nOOH
OH
ZH
ZH
nBHnadsSP
nBHnadsMP
nSS
nSHnBH
nNONO
nNO
nNHNA
NA
nOS
nOZBH
KnSNO
XSKXS
KSK
SX
SKS
SKK
SKKf
YY
XSKXS
KSK
SX
SKS
SKS
SKK
YY
XSKXS
KSK
SX
SKS
SKK
SKS
fr NO
,,
,,
,
,,2
,
,
,,
,,
,,
,
,,2
,
,
,
,
,
,,
,,
,
,,2
,
,
,,
,,
86.21
86.21
nBHnadsSP
nBHnadsgMP
nSS
nSHnBH
nNONO
nNO
nNHNA
nNH
nOOH
OHZBH
nBHnadsSP
nBHnadsMP
nSS
nSHnBH
nNONO
nNO
nNHNA
nNH
nOS
nOZBH
KnSNO
XSKXS
KSK
SX
SKS
SK
SSK
Kf
XSKXS
KSK
SX
SKS
SK
SSK
Sfr NA
,,
,,
,
,,
,
,2
,
,
,
,,
,,
,
,,
,
,2
,
,
,
,,
nBAnOOA
nO
nNHSA
nNHA
ZA
KnSNO X
SKS
SK
SY
r OA,
,
,2
,
,,
1
nBHnNONO
nNO
nNHNA
NA
nOOH
OH
nBHnadsSP
nBHnadsMPZBH
nNHNA
NA
nNHNA
nNHnBH
nNONO
nNO
nOOH
OH
nBHnadsSP
nBHnadsMP
ZH
ZHnSNO
XSK
SSK
KSK
KXSK
XSKf
SKK
SKS
XSK
SSK
KXSK
XSK
YY
r g
,,
,
,,,,
,,
,,
,,
,
,
,,,
,,, 86.2
1
nNHNA
NA
nNHNA
nNHnBHg
nNONO
nNO
nOOH
OH
nBHnadsSP
nBHnads
ZH
ZH
nOOH
OHg
nOOH
nOnBH
nNONO
nNO
nNHNA
NA
nBHnadsSP
nBHnadsZBH
KnSNO
SKK
SKS
XSK
SSK
KXSK
XSYY
SKK
SKS
XSK
SSK
KXSK
XSfr MP
,,
,,
,
,
,,,
,,
,,
,,
,
,
,,,
,,,
86.21
nNHNA
NA
nNHNA
nNHnBHg
nNONO
nNO
nOOH
OH
nBHnadsSP
nBHnads
ZH
ZH
nOOH
OHg
nOOH
nOnBH
nNONO
nNO
nNHNA
NA
nBHnadsSP
nBHnadsMPZBH
KnSNO
SKK
SKS
XSK
SSK
KXSK
XSYY
SKK
SKS
XSK
SSK
KXSK
XSKfr SP
,,
,,
,
,
,,,
,,
,,
,,
,
,
,,,
,,,
86.21
nBAnOOA
nO
nNHSA
nNHA
ZA
KnSNO X
SKS
SK
SY
r SA,
,
,2
,
,,
1
http://dx.doi.org/10.4314/wsa.v38i2.15Available on website http://www.wrc.org.za
ISSN 0378-4738 (Print) = Water SA Vol. 38 No. 2 April 2012ISSN 1816-7950 (On-line) = Water SA Vol. 38 No. 2 April 2012306
• for the variable’s rate rSS,n
nBAnOOA
nO
nNHSA
nNHA
ZH
YnSNO X
SK
SSK
SY
r ZA,2
,
,
,
,2, 86.2
1
nBHnadsSP
nBHnadsMP
nSS
nSHnBH
nNONO
nNO
nNHNA
NA
nOOH
nOfnSNO XSK
XSK
SKS
XSK
SSK
KSK
Sr ZBH
,,
,,
,
,,
,
,
,,
,,
nNHNA
NA
nNHNA
nNHnBH
nNONO
nNO
nOOH
OH
nSS
nS
ZH
ZH
nOOH
OH
nOOH
nO
nBHnNONO
nNO
nNHNA
NA
nSS
nSZBHnSNO
SKK
SKS
XSK
S
SKK
SKS
YY
SKK
SKS
XSK
SSK
KSK
Sfr H
,,
,,
,
,
,,
,
,,
,
,,
,
,,
,,
.
.86.2
1.
.2
nBAnOOA
nO
nNHSA
nNH
ZAnSNO X
SKS
SK
SY
r A,
,
,2
,
,,
1
nNHNA
NA
nNHNA
nNHnBH
nNONO
nNO
nOOH
OH
nSS
nSH
ZH
ZH
nOOH
OH
nOOH
nOnBH
nNONO
nNO
nNHNA
NA
nSS
nSHZBH
KnSNO
SKK
SKS
XSK
SSK
KSK
SYY
SKK
SKS
XSK
SSK
KSK
Sfr S
,,
,,
,
,
,2
,
,
,,
,,
,
,
,2
,
,,
86.21
gnBHnadsSP
nBHnadsMPnBH
nNONO
nNO
nNHNA
NA
nOOH
nOZBH
nBHnadsSP
nBHnadsgMP
nSS
nSHnBH
nNONO
nNO
nOOH
OH
ZH
ZHKnSNO
XSKXS
KXSK
SSK
KSK
Sf
XSKXS
KSK
SX
SKS
SKK
YY
r OH
1
86.21
,,
,,,
,
,
,2
,
,
,,
,,
,
,,
,
,2
,,
nBHnadsSP
nBHnadsgMP
nSS
nSHnBH
nNOOH
nNO
nNHNA
NA
nOOH
OHZBH
ZH
ZH
nBHnadsSP
nBHnadsgMP
nSS
nSHnBH
nNONO
nNO
nNHNA
nNH
nOOH
OH
ZH
ZH
nBHnadsSP
nBHnadsMP
nSS
nSHnBH
nNONO
nNO
nNHNA
NA
nOS
nOZBH
KnSNO
XSKXS
KSK
SX
SKS
SKK
SKKf
YY
XSKXS
KSK
SX
SKS
SKS
SKK
YY
XSKXS
KSK
SX
SKS
SKK
SKS
fr NO
,,
,,
,
,,2
,
,
,,
,,
,,
,
,,2
,
,
,
,
,
,,
,,
,
,,2
,
,
,,
,,
86.21
86.21
nBHnadsSP
nBHnadsgMP
nSS
nSHnBH
nNONO
nNO
nNHNA
nNH
nOOH
OHZBH
nBHnadsSP
nBHnadsMP
nSS
nSHnBH
nNONO
nNO
nNHNA
nNH
nOS
nOZBH
KnSNO
XSKXS
KSK
SX
SKS
SK
SSK
Kf
XSKXS
KSK
SX
SKS
SK
SSK
Sfr NA
,,
,,
,
,,
,
,2
,
,
,
,,
,,
,
,,
,
,2
,
,
,
,,
nBAnOOA
nO
nNHSA
nNHA
ZA
KnSNO X
SKS
SK
SY
r OA,
,
,2
,
,,
1
nBHnNONO
nNO
nNHNA
NA
nOOH
OH
nBHnadsSP
nBHnadsMPZBH
nNHNA
NA
nNHNA
nNHnBH
nNONO
nNO
nOOH
OH
nBHnadsSP
nBHnadsMP
ZH
ZHnSNO
XSK
SSK
KSK
KXSK
XSKf
SKK
SKS
XSK
SSK
KXSK
XSK
YY
r g
,,
,
,,,,
,,
,,
,,
,
,
,,,
,,, 86.2
1
nNHNA
NA
nNHNA
nNHnBHg
nNONO
nNO
nOOH
OH
nBHnadsSP
nBHnads
ZH
ZH
nOOH
OHg
nOOH
nOnBH
nNONO
nNO
nNHNA
NA
nBHnadsSP
nBHnadsZBH
KnSNO
SKK
SKS
XSK
SSK
KXSK
XSYY
SKK
SKS
XSK
SSK
KXSK
XSfr MP
,,
,,
,
,
,,,
,,
,,
,,
,
,
,,,
,,,
86.21
nNHNA
NA
nNHNA
nNHnBHg
nNONO
nNO
nOOH
OH
nBHnadsSP
nBHnads
ZH
ZH
nOOH
OHg
nOOH
nOnBH
nNONO
nNO
nNHNA
NA
nBHnadsSP
nBHnadsMPZBH
KnSNO
SKK
SKS
XSK
SSK
KXSK
XSYY
SKK
SKS
XSK
SSK
KXSK
XSKfr SP
,,
,,
,
,
,,,
,,
,,
,,
,
,
,,,
,,,
86.21
nBAnOOA
nO
nNHSA
nNHA
ZA
KnSNO X
SKS
SK
SY
r SA,
,
,2
,
,,
1
• for the variable’s rate rads,n
0, fnSSr , 0, ZAY
nSSr , 0, ZBHfnSSr , 0, A
nSSr
]
[1
,,
,
,
,
,
,
,
,,
,
,
,,
,
,2,
nNHNA
NA
nNHNA
nNH
nNONO
nNO
nOOH
OH
nNONO
nNO
nNHNA
NA
nNHNA
nNH
nOOH
nOnBH
nSS
nSH
ZH
YnSS
SKK
SKS
SKS
SKK
SKS
SKK
SKS
SKS
XSK
SY
r ZH
]
[1
,,
,
,
,
,
,
,
,,
,
,
,,
,
,,
nNHNA
NA
nNHNA
nNH
nNONO
nNO
nOOH
OH
nNONO
nNO
nNHNA
NA
nNHNA
nNH
nOOH
nOnBH
nSS
nS
ZAnSS
SKK
SKS
SKS
SKK
SKS
SKK
SKS
SKS
XSK
SY
r H
]
[1
,,
,
,
,
,
,
,
,,
,
,
,,2
,
,,
nNHNA
NA
nNHNA
nNH
nNONO
nNO
nOOH
OH
nNONO
nNO
nNHNA
NA
nNHNA
nNH
nOOH
nOnBH
nSS
nSH
ZH
KnSS
SKK
SKS
SKS
SKK
SKS
SKK
SKS
SKS
XSK
SY
r S
nNONO
nNO
nNHNA
nNHnBH
nOOH
nO
nSS
nSH
ZH
KnSS SK
SSK
SX
SKS
SKS
Yr OH
,
,
,
,,2
,
,
,
,, 11
]
[1
,,
,
,
,,
,,2
,
,
,
,,
nNHNA
NA
nNHNA
nNH
nOOH
OH
nNHNA
NA
nOOH
nOnBH
nNONO
nNO
nSS
nSH
ZH
KnSS
SKK
SKS
SKK
SKK
SKS
XSK
SSK
SY
r NO
nNONO
nNO
nOOH
nOnBH
nNHNA
nNH
nSS
nSH
ZH
KnSS SK
SSK
SX
SKS
SKS
Yr NA
,
,
,
,,2
,
,
,
,, 11
0, OAK
nSSr , 0, gnSSr
, 0, MPKnSSr , 0, SPK
nSSr , 0, SAKnSSr , 0, MAf
nSSr, 0, AK
nSSr
0, fnSadsr 0, OAK
nSadsr , 0, ZAYnSadsr , 0, ZBHf
nSadsr , 0, HnSadsr , 0, A
nSadsr , 0, SKnSadsr
]
[1
,,
,
,
,
,
,
,
,,
,
,
,,
,,
,,2
nNHNA
NA
nNHNA
nNH
nNONO
nNO
nOOH
OHg
nNONO
nNO
nNHNA
NA
nNHNA
nNH
nOOH
nOnBH
nBHnadsSP
nBHnadsMP
ZH
YSads
SKK
SKS
SKS
SKK
SKS
SKK
SKS
SKS
XXSK
XSK
Yr ZH
nNONO
nNO
nNHNA
nNHnBH
nOOH
nO
nBHnadsSP
nBHnadsMP
ZH
KSads SK
SSK
SX
SKS
XSKXS
KY
r OH
,
,
,
,,2
,
,
,,
,, 11
]
1
,,
,
,
,,
,,2
,
,
,,
,,
nNHNA
NA
nNHNA
nNH
nOOH
OHg
nNHNA
NA
nOOH
nOnBH
nNONO
nNO
nBHnadsSP
nBHnadsMP
ZH
KSads
SKK
SKS
SKK
SKK
SKS
XSK
SXSK
XSK
Yr NO
nNONO
nNO
nOOH
nOnBH
nNHNA
nNH
nBHnadsSP
nBHnadsMP
ZH
KSads SK
SSK
SX
SKS
XSKXS
KY
r NA
,
,
,
,,2
,
,
,,
,, 11
nNHNA
NA
nNHNA
nNHnBHg
nNONO
nNO
nOOH
OH
nBHnadsSP
nBHnadsMP
ZHSads SK
KSK
SX
SKS
SKK
XSKXS
KY
r g
,,
,,
,
,
,,,
,,1
]
[1
,,
,
,
,
,
,
,
,,
,
,
,,
,,
,,
nNHNA
NA
nNHNA
nNH
nNONO
nNO
nOOH
OHg
nNONO
nNO
nNHNA
NA
nNHNA
nNH
nOOH
nOnBH
nBHnadsSP
nBHnads
ZH
KSads
SKK
SKS
SKS
SKK
SKS
SKK
SKS
SKS
XXSK
XSY
r MP
]
[1
,,
,
,
,
,
,
,
,,
,
,
,,2
,,
,,
nNHNA
NA
nNHNA
nNH
nNONO
nNO
nOOH
OHg
nNONO
nNO
nNHNA
NA
nNHNA
nNH
nOOH
nOnBH
nBHnadsSP
nBHnads
ZH
KSads
SKK
SKS
SKS
SKK
SKS
SKK
SKS
SKS
XXSK
XSY
r SP
0, SAK
nSadsr , nBHnSA
KnSads XXKr OA
,,, , nBHnadsMAnBHnSK
nSads XSfXXr A,,,,,
0, fnSSr , 0, ZAY
nSSr , 0, ZBHfnSSr , 0, A
nSSr
]
[1
,,
,
,
,
,
,
,
,,
,
,
,,
,
,2,
nNHNA
NA
nNHNA
nNH
nNONO
nNO
nOOH
OH
nNONO
nNO
nNHNA
NA
nNHNA
nNH
nOOH
nOnBH
nSS
nSH
ZH
YnSS
SKK
SKS
SKS
SKK
SKS
SKK
SKS
SKS
XSK
SY
r ZH
]
[1
,,
,
,
,
,
,
,
,,
,
,
,,
,
,,
nNHNA
NA
nNHNA
nNH
nNONO
nNO
nOOH
OH
nNONO
nNO
nNHNA
NA
nNHNA
nNH
nOOH
nOnBH
nSS
nS
ZAnSS
SKK
SKS
SKS
SKK
SKS
SKK
SKS
SKS
XSK
SY
r H
]
[1
,,
,
,
,
,
,
,
,,
,
,
,,2
,
,,
nNHNA
NA
nNHNA
nNH
nNONO
nNO
nOOH
OH
nNONO
nNO
nNHNA
NA
nNHNA
nNH
nOOH
nOnBH
nSS
nSH
ZH
KnSS
SKK
SKS
SKS
SKK
SKS
SKK
SKS
SKS
XSK
SY
r S
nNONO
nNO
nNHNA
nNHnBH
nOOH
nO
nSS
nSH
ZH
KnSS SK
SSK
SX
SKS
SKS
Yr OH
,
,
,
,,2
,
,
,
,, 11
]
[1
,,
,
,
,,
,,2
,
,
,
,,
nNHNA
NA
nNHNA
nNH
nOOH
OH
nNHNA
NA
nOOH
nOnBH
nNONO
nNO
nSS
nSH
ZH
KnSS
SKK
SKS
SKK
SKK
SKS
XSK
SSK
SY
r NO
nNONO
nNO
nOOH
nOnBH
nNHNA
nNH
nSS
nSH
ZH
KnSS SK
SSK
SX
SKS
SKS
Yr NA
,
,
,
,,2
,
,
,
,, 11
0, OAK
nSSr , 0, gnSSr
, 0, MPKnSSr , 0, SPK
nSSr , 0, SAKnSSr , 0, MAf
nSSr, 0, AK
nSSr