Abstract—This paper proposes robust control and power

management strategies for a 6kW stationary fuel cell hybrid

power system. The system consists of two 3kW PEMFC modules,

a Li-Fe battery set and electrical components to form a parallel

hybrid power system that is designed for telecom base stations

to supply uninterruptible power during emergency power

failures. The study is carried out in three steps: the PEMFC

modules control, power management, and system integration.

First, we apply robust control to regulate the hydrogen flow

rates of the PEMFC modules to increase system stability,

performance, and efficiency. Second, we design a parallel power

train that consists of two PEMFC modules and one Li-Fe

battery set for uninterruptible power supply (UPS) requirement.

When the main power is shut down, the Li-Fe battery will

activate PEMFC modules. Then the PEMFC modules provide

steady power at low current loadings. At high loading, both

PEMFC modules and the Li-Fe battery set will simultaneously

provide electricity. Lastly, we integrate the system for

experimental verification. Based on the results, the proposed

robust control and power management are deemed effective in

improving stability, performance and efficiency of the

stationary power system.

I. INTRODUCTION

s an alternative energy source, the proton exchange

membrane fuel cell (PEMFC) has drawn much attention

because of its advantageous properties, such as low operating

temperature, high efficiency, and environmental-friendly.

Gorgun [1] described a PEM electrolyzer as a dynamic model

that considered water phenomena, electro-osmotic drag and

diffusion, and voltage ancillary. Bao et al. [2] developed a

nonlinear PEMFC model and further linearized it for linear

control design. Woo and Benziger [3] designed a

proportional-integral-derivative (PID) controller that

regulated the hydrogen flow rate for optimal performance.

Vega-Leal et al. [4] developed a multivariable system that

Manuscript received October 28, 2012. This work was supported in part

by the Taiwan National Science Council under Grant

100-2622-E-002-027-CC3

Yi-Fu Kuo is with the Mechanical Engineering Department of National Taiwan University, Taiwan. (e-mail: great.speed@msa.hinet.net)

Po-Cheng Kuo is with the Mechanical Engineering Department of

National Taiwan University, Taiwan. (e-mail: r99522827@ntu.edu.tw). Hsueh-Ju Chen is with the Mechanical Engineering Department of

National Taiwan University, Taiwan. (e-mail: phon211821@yahoo.com.tw).

Chung-Huang Yu is with the Department of Physical Therapy & Assistive Technology, National Yang-Ming University, Taipei, Taiwan (e-mail:

chyu@ym.edu.tw).

Fu-Cheng Wang is with the Mechanical Engineering Department of National Taiwan University, No.1, Sec. 4, Roosevelt Road, Taipei 10617,

Taiwan. (corresponding author phone: +886-2-33662680; fax:

+886-2-23631755; e-mail: fcw@ntu.edu.tw).

controlled the air and hydrogen flow rates to optimize net

power. Wang et al. [5–9] applied robust control that regulated

the oxygen and hydrogen flow rates to provide steady voltage

and to reduce hydrogen consumption. Andreasen et al. [10] applied proportional-integral (PI) temperature control to

achieve fuel cell system efficiency of 45 percent.

There are four main application fields for PEMFC,

namely transportation, niche markets, portable power, and

stationary power generation [11]. In this paper, we discuss the

application of stationary power, which include backup power,

auxiliary power unit (APU), uninterruptible power supply

(UPS), and combined heat & power system (CHP) [11]. Choi

et al. [12] focused on designed considerations of replacing

conventional UPS power sources by fuel-cells. Corbo et al.

[13] analyzed the application of a 20kW PEMFC stack to

both stationary power and automotive systems. Zhan et al. [14]

developed an hybrid intelligent UPS system that comprises a

PEMFC and a battery. Tao et al. [15] proposed a

line-interactive fuel-cell powered UPS system that composed

of a three-port bidirectional converter, a fuel cell, a

super-capacitor, and a grid-interfacing inverter. This paper

discusses the application of a 6kW PEMFC hybrid backup

power system for UPS of telecom base stations, where an

unexpected power disruption might cause injuries, fatalities,

serious business disruption, and data loss. Compared to

traditional UPS with combustion engines backup power,

battery-based UPS are adopted to improve system

performance and environment friendliness. However, the

operating time of batteries is limited by their capacities.

Therefore, in this study, we integrate a PEMFC hybrid

backup power for continuous operation of UPS systems.

This paper is organized as follows: section II discusses the

dynamics of a 3kW PEMFC modules. Section III applies H∞

robust control to regulate the hydrogen flow rates of the

PEMFC to improve system performance. Section IV

proposes a 6kW PEMFC hybrid stationary power system that

satisfies the UPS requirements for telecom base stations.

Section V demonstrates system integration and experimental

verification. Lastly, we draw conclusions in Section VI.

II. PEMFC SYSTEM DESCRIPTION AND IDENTIFICATION

The proposed 6kW PEMFC UPS system consists of two

independent M-FieldTM

3kW PEMFC models. We first

consider an air-breathing 3kW PEMFC module, as shown in

Fig. 1, which is equipped with a hydrogen inlet and an outlet.

In addition, four 50W fans are used for air-cooling and

supplying oxygen. The stack is operated in a dead-end mode,

i.e., the hydrogen outlet is closed except for purging, in order

Control and Power Management of a Hybrid Stationary Fuel Cell

System

Yi-Fu Kuo, Po-Chen Kuo, Hsueh-Ju Chen, Chung-Huang Yu, and Fu-Cheng Wang

A

Proceedings of the 2013 IEEE/SICE InternationalSymposium on System Integration, Kobe InternationalConference Center, Kobe, Japan, December 15-17,

SP1-I.3

978-1-4799-2625-1/13/$31.00 ©2013 IEEE 144

to maintain necessary reactant in the system. During

experiments, the valve is purged for 0.5 s at an interval of 50 s.

The specifications of this PEMFC module are illustrated in

Table I.

Fig. 1. The M-Field 3kW PEMFC module.

TABLE I

STACK SPECIFICATIONS OF A M-FIELDTM

3KW PEMFC [16].

Number of cells 80

Maximum power DC 3.2kW (at DC75A) Input voltage 24VDC

Output voltage 42–80VDC

Size 80×46 ×40 cm^3

weight 40 kg Fuel inlet temperature -15→55 oC

Fuel inlet pressure 136kpa (0.36barg)

Air inlet temperature 10→50 oC

From the system point of view, the PEMFC module can

be depicted as the following single-input multi-output

system [7]:

2

( ) ,H

IG s N

V

(1)

where 2HN is the hydrogen flow rate and ( )G s is a 2×1

transfer function matrix, which represents the PEMFC

dynamics, where we can regulate the fuel cell voltage (V )

and current ( I ) outputs by controlling the hydrogen input.

Because most electrical devices require a steady voltage

supply, we can assume the output impedance as Ze (i.e., V=

Ze× I), and simplify Eq. (1) as a single-input single-output

(SISO) model:

2

( ) ,HV G s N (2)

whose output voltage can be controlled by regulating the

hydrogen flow rate [6]. To decide the transfer function

( )G s of (2), we generate chirp signals to control the

hydrogen flow rates, and measure the output voltage signals.

Applying the subspace state space system identification

(N4SID) algorithms [17, 18], we then obtain the state-space

models of the PEMFC. Based on the characteristics of the

PEMFC, we set the operating conditions as 59V, 55.8V,

53.6V and 50.8V with four different current loads: 20A,

30A, 40A, and 50A. Considering the system variations, we

repeat the experiments twice at each load and illustrate the

obtained system transfer functions in Table II. These

transfer functions will be used for robust control design in

Section III. TABLE II

TRANSFER FUNCTIONS OF THE PEMFC.

20A01 20A022 2

2.355 0.357 2.342 0.464920A : ,

0.7334 0.00096 0.7971 0.00126

s sG G

s s s s

30A01 30A022 2

1.764 0.06668 1.851 0.419530A : ,

0.6097 0.00026 0.6816 0.00016

s sG G

s s s s

40A01 40A022 2

2.072 0.5781 2.021 0.180240A : ,

1.239 0.003012 0.9912 0.00094

s sG G

s s s s

50A01 50A022 2

2.059 0.1174 2.192 0.23950A : ,

1.319 0.0007 1.471 0.00147

s sG G

s s s s

III. ROBUST CONTROL DESIGN

In this section, we introduce robust control algorithms

and design a H∞ controller for the PEMFC module. Suppose a

nominal plant 0G has a normalised left coprime factorization

of 1

0G M N , where ,M N RH and * *MM NN I .

Assume a perturbed system G can be expressed as:

1

M NG M N

with N M

, and

,M N RH . Because coprime factorization of a system is

not unique, we can define the gap between a nominal plant

0G and a perturbed plant G as [19]:

The smallest value of N M

which

perturbs 0G into G , is called the gap between

0G and G , and is denoted as 0 ,G G .

Therefore, we can select a nominal plant 0G for the PEMFC

system, which minimizes the maximum gap between 0G and

the perturbed plants G , as follows:

0

0 0arg min max , , .i

i iG G

G G G G

(3)

Consider the system transfer functions of Table II, we

select 30A01G as the nominal plant, i.e.,

0 30A01G G , which

gives 0 , 0.2199iG G for all iG . The gap can be regarded

as the maximum system variation of the PEMFC, due to the

changes in operating conditions, such as temperature,

humidification, and current loads. A closed-loop system with

a perturbed system G and a controller K can be represented

as Fig. 3(a) and rearranged as Fig. 3(b). Therefore, using the

Small-Gain Theorem [20], the system is internally stable for

all perturbations [ ]N M with

if and only

if:

1

0 0

1.

KI G K I G

I

(4)

978-1-4799-2625-1/13/$31.00 ©2013 IEEE 145

(a) A closed-loop system with G and K.

(b) Rearrangement for the Small-Gain Theorem.

Fig. 3. The closed-loop control system.

Define the system’s stability margin 0 ,b G K as [18]:

-1

-1

0 0 0, - ,K

b G K I G K I GI

(5)

the closed-loop system is internally stable for all uncertainties

[ ]N M with

if and only if its stability

margin 0 ,b G K . Hence, the objective of controller

synthesis is to design a controller K for the nominal plant 0G

such that 0 , 0.2199b G K . Using the H loop-shaping

design techniques [21], as shown in Fig. 4, we set the

weighting functions as:

1 2

0.3 0.06, 1

sW W

s

,

to shape the plant as 2 0 1sG W G W . The weighting functions

are selected and iteratively verified by simulations and

experiments, in order to provide a high-gain at the

low-frequency range to improve system performance, and a

low-gain at the high-frequency range for robust stability.

Following the aforementioned procedures, the standard H

robust controller is designed as:

2

2

1.203 0.5017 0.0575,

0.5456 0.0692

s sK s

s s

which gives a stability bound of 0.5679. Because the

stability margin is greater than the system gap

0 , 0.2199iG G , the system stability is guaranteed. The

weighted controller 1 2K W K W is then implemented for

the PEMFC system, as illustrated in Fig. 4(b). We

implement the designed controller, and illustrate the test

results in Table III, where the hydrogen consumption is

reduced. The root mean square error (RMSE) is effectively

controlled within a reasonable range (<0.05V) by the

designed robust controller.

(a) Shape the open-loop plant 2 0 1sG W G W .

(b) Implemente the weighted controller 1 2K W K W .

Fig. 4. Loop-shaping design procedures.

TABLE III COMPARISON OF THE OPEN- AND CLOSED-LOOP CONTROL.

H2 consumption (L) in 5 min.

RMSE(V)

Closed-loop

Open-Loop Closed-Loop Improved (%)

20A 76.95 71.29 7.36% 0.0363

30A 104.44 98.60 5.59% 0.0475

40A 133.99 128.96 3.75% 0.0304

50A 162.92 156.96 3.66% 0.0272

IV. POWER MANAGEMENT

In this section, we design the power management

strategies for the 6kW stationary PEMFC hybrid system. A

classical UPS is shown in Fig. 5(a), where the main 220VAC

grid power is converted (by the switch module rectifier; SMR)

into 48VDC for the stations supplied, and charges the battery

set. When the grid power is shut down, the backup battery set

is used for uninterrupted operation. However, the operating

time of the classical UPS system is limited due to the limited

capacity of the battery set. Therefore, we construct a PEMFC

978-1-4799-2625-1/13/$31.00 ©2013 IEEE 146

UPS system, as illustrated in Fig. 5(b), which completes the

battery set with a 6kW PEMFC for continuous operation as

long as the hydrogen is continuously supplied.

The detailed structure of the 6kW PEMFC hybrid backup

power system is shown in Fig. 6, in which the parameters are:

Q : PEMFC module hydrogen flow rate.

, : PEMFC module output voltage and current.

, : Battery set output voltage and current.

, :DC DC converter input voltage and current.

,

j

FCj FCj

BATT BATT

Cij Cij

Coj Coj

j

V I j

V I

V I j

V I

: DC DC converter output voltage and current.

, : Balance of plant BOP voltage and current.BOPj BOPj

j

V I

where j=1,2.

(a) The classical UPS structure.

(b) The PEMFC UPS structure.

Fig. 5. The UPS system structure.

Fig. 6. The 6kW PEMFC hybrid power system.

TABLE IV

System hardware.

System parts Company/

type

Specification

PEMFC modules M-Field/

LPH8020

As illustrated in Table 1.

Li-Fe battery set A123/ ANR26650

Nominal voltage nV =52.8V,

Nominal capacity nQ =23Ah

DC buses Nell/

NKJ240

Maximum reverse voltage RPMV

=600V,

Maximum forward current fI =240A

DC-DC converters PSP/

SN:0020-D

Input: 45~85V(3kW)

Ouput1: 51V(adjustable) Ouput2: 24V

Load AMREL/

PLW9k

Mode: CC,CV,CP,CR

Pmax:9kW

The hybrid system consists of three parts: two PEMFC

modules (the main source, where module-1 is a new module

and module-2 is a two-year-old module), a Li-Fe battery set

(auxiliary source) and electrical components. The power

management strategies can be described as follow:

1. When the grid power is available, the station loading is

supplied by SMR 48VDC output and the battery set is

charged by charger.

2. If the grid power is shut down, the battery set , BATTI

will provide electricity ( , , , ) for station

loadings and start up BOPs ( , , , ) to

activate the PEMFC modules.

3. After the PEMFC modules reach steady states (normally

within 1 minute), they will provide electricity

( , , , ) in parallel with the battery set: At low

current loadings ( > and > ), the PEMFC

modules provide electricity alone. At high current loadings

( = = ), both the PEMFC modules and battery set

provide electricity for station loads.

4. When the station load is low (<35A), the PEMFC

module-1 will charge the battery set directly until its voltage

is equal or higher than 54V, considering the discharging

curve of the battery set of Fig. 7. The charging method is

near to constant current charge.

5. For safety, the security protection will be activated by the

following four conditions:

(1) gas pressure protection: the input pressure of hydrogen

flow should be greater than 0.1 barg and less than 0.5

barg.

(2) temperature protection: the temperature of the

input/output flow should be less than 50 C / 70 C .

(3) PEMFC module output protection:

( 45 85, 0< 60FCj FCjV I ).

(4) Battery protection: ( 46 54BATTV ).

Table IV illustrates the system hardware: two M-field

PEMFC modules as main sources, an A123 Li-Fe battery set

as the auxiliary source, DC buses and DC-DC converters for

power managements.

978-1-4799-2625-1/13/$31.00 ©2013 IEEE 147

Fig. 7. The discharging curves of the battery set.

V. EXPERIMENT

In this section, we will integrate the aforementioned system,

as shown in Fig. 8, for experimental verification and

calculate the system efficiency. Refer to Fig. 6, we define

the system energy levels as follow:

1: ,

2 : ,

3 : ,

,

e

s

e

s

e

s

e

s

e

s

t

FCj FCj FCjt

t

BATT BATT BATTt

t

Cij Cij Cijt

t

Coj Coj Cojt

t

BOPj BOPj BOPjt

Level E V I dt

E V I dt

Level E V I dt

Level E V I dt

E V I dt

(6)

where j=1,2. Note that Level 1 includes the output energy of

two PEMFC modules and the Li-Fe Battery set. Level 2

comprises the input energy of all DC-DC converters. Level 3

contains the output energy of all DC-DC converters. Using

the experiment parameters of Table V, we show the

experimental responses in Fig. 9, and illustrate the statistic

data in Table VI.

Fig. 8 Integration of 6kW PEMFC hybrid backup power system.

TABLE V

EXPERIMENT PARAMETERS.

Operating parameter

Input

Pressure 5 barg

Temperature 25 o C

Voltage 24V

Output Purge volume 3000L(open loop) Load-meter 40A→60A→80A→100A, (100s each)

Measure signal

NI DAQ

sampling

time=0.01s

Flow rate: Q j( j=1,2)

Level 1: , , , FCj FCj BATT BATTV I V I ( j=1,2)

Level 2: , Cij CijV I ( j=1,2)

Level 3: , , , Coj Coj BOPj BOPjV I V I ( j=1,2)

Fig. 9 System current and power response.

TABLE VI Statistic data of Fig. 9 [12].

Energy (kJ) Closed-loop

Level 1

1FCE 780.80

2FCE 561.45

BATTE 336.09

Level 2 1CiE 1037.06

2CiE 620.11

Level 3

1CoE 924.79

2CoE 527.00

1BOPE 32.80

2BOPE 21.74

(%)e 89.75

.(%)hyd 77.48

(%)total 73.59

For quantitative comparison, we define several system

efficiencies as in the following. First, we define the net power

efficiency (%)e as:

1 2 1 2

1 2

3(%) 100%,

1

Co Co BOP BOP

e

FC FC BATT

E E E ELevel

Level E E E

(7)

which represents the percentage of utilized energy from Level

1 to Level 3. Second, we define the hydrogen efficiency as

[8]:

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6

48

50

52

54

56

Time(hrs)

Voltage(V

)

Discharge curve Li-Fe Batteries 48V23Ah with BMS

0 2 4 6 8 10 12 14 16 18

48

50

52

54

56

Capacity(Ah)

Voltage(V

)0.5C

1C

1.5C

2C

Load-meter

FC

Module-2

BreakerConverter2

Lead-Acid Batteries

Diode

ComputerLi-Fe

Batteries

FC

Module-1

Converter1

0 50 100 150 200 250 300 350 4000

20

40

60

80

Level 1

Current(A)-Time(s) plot

0 50 100 150 200 250 300 350 4000

20

40

60

80

Level 2

0 50 100 150 200 250 300 350 4000

20

40

60

80

Level 3

time(s)

0 50 100 150 200 250 300 350 4000

1

2

3

4Power(kW)-Time(s) plot

0 50 100 150 200 250 300 350 4000

1

2

3

4

0 50 100 150 200 250 300 350 4000

1

2

3

4

time(s)

Ci1

Ci2

FC1

FC2

BATT

Co1

Co2

BOP1

BOP2

978-1-4799-2625-1/13/$31.00 ©2013 IEEE 148

2

.

2

(%) 100%,Th

hyd Exp

H

H (8)

where 2

ThH and 2

ExpH are the theoretical and practical

hydrogen consumption, respectively. The theoretical

hydrogen consumption 2

ThH is defined as [9]:

2 ,2

Th I n tH R

F

(9)

in which I is the current load, n is the number of cells, t is the

time interval, F is the Faraday constant (F=96,485 C/mole )

and R is the mass of per mole hydrogen (R= 2.0158 g/mole ).

Lastly, we define the total system efficiency as:

1 2 .( )total e hydr r

(10)

1 21 2

1 2 1 2

, BATT FC FC

FC FC BATT FC FC BATT

E E Er r

E E E E E E

(11)

where 1r and 2r are the battery energy ratio and the PEMFC

energy ratio, respectively, of Level 1. As shown in Table VI,

we achieve a net power efficiency of about =89.75%e and a

hydrogen efficiency .hyd 77.48% by the closed-loop

control strategies, which makes the total system efficiency

total 73.59%.

VI. CONCLUDING REMARKS

This paper has demonstrated the control and power

management of a 6kW stationary fuel cell hybrid system. We

first modeled and controlled a 3kW PEMFC module, and then

designed the PEMFC hybrid power system for UPS for

telecommunication stations. This system was integrated for

experimental verification, and shown to improve the system

efficiencies. Based on the results, the proposed robust control

and power management was deemed effective for the

PEMFC stationary system.

REFERENCES

[1] H. Gorgun, “Dynamic modelling of a proton exchange membrane (PEM) electrolyzer”, International Journal of Hydrogen Energy, vol.

31, no. 1, pp. 29–38, 2006.

[2] C. Bao, M.G. Ouyang, B.L. Yi, “Modeling and control of air stream

and hydrogen flow with recirculation in a PEM fuel cell system – I. Control-oriented modeling”, International Journal of Hydrogen

Energy, vol. 31, no. 13, pp. 1879–1896, 2006.

[3] C.H. Woo, J.B. Benziger, “PEM fuel cell current regulation by fuel

feed control”, Chemical Engineering Science, vol. 62, no. 4, pp. 957–68, 2007.

[4] A.P. Vega-Leal, F.R. Palomo, F. Barragan, C. Garcia, J.J. Brey,

“Design of control systems for portable PEM fuel cells”, Journal of Power Sources, vol. 169, no. 1, pp. 194–197, 2007.

[5] F.C. Wang, Y.P. Yang, C.W. Huang, H.P. Chang, H.T. Chen, “System

identification and robust control of a portable proton exchange

membrane full-cell system”, Journal of Power Sources, vol. 165, no. 2,

pp. 704–712, 2007.

[6] F.C. Wang, H.T. Chen, Y.P. Yang, J.Y. Yen, “ Multivariable robust

control of a proton exchange membrane fuel cell system”, Journal of Power Sources, vol. 177, no. 2, pp. 393–403, 2008.

[7] F.C. Wang, H.T. Chen, “Design and implementation of fixed-order

robust controllers for a proton exchange membrane fuel cell system”,

International Journal of Hydrogen Energy, vol. 34, no.6, pp. 2705–2717, 2009.

[8] .F.C. Wang, C.C. Ko, “Multivariable Robust PID Control for a

PEMFC System”, International Journal of Hydrogen Energy, vol. 35,

no.19, pp. 10437–10445, 2010. [9] F.C. Wang, L.H. Lin and M.C. Chou, “Multivariable Robust Control

for a 500W Self-Humidified PEMFC System”, European Journal of

Control, vol. 17, no.4, pp. 429–441, 2011. [10] S.J. Andreasen, L. Ashworth, I.N.M. Remon, S.K. Kaer, “Directly

connected series coupled HTPEM fuel cell stacks to a Li-ion battery

DC bus for a fuel cell electrical vehicle”, International Journal of

Hydrogen Energy, vol. 33, no. 23, pp. 7137–7145, 2008.

[11] C.A. Cottrell, S.E. Grasman, M. Thomas, K.B. Martin, J.W. Sheffield, ”Strategies for stationary and portable fuel cell markets”

International Journal of Hydrogen Energy, vol. 36, no.13, pp. 7969–7975, 2011.

[12] W. Choi, J.W. Howze, P. Enjeti, “Fuel-cell powered uninterruptible power supply systems: Design considerations” Journal of Power

Sources, vol. 157, no. 1, pp. 311–317, 2006.

[13] P. Corbo , F. Migliardini, O. Veneri, “Experimental analysis and

management issues of a hydrogen fuel cell system for stationary and mobile application” Energy Conversion and management vol.48 no. 8

pp. 2365–2374, 2007.

[14] Y. Zhan, Y. Guo, J. Zhu, H. Wang, “Intelligent uninterruptible power supply system with back-up fuel cell/battery hybrid power

source” Journal of Power Sources, vol. 179, no. 2, pp. 745–753,2008.

[15] H. Tao, J.L. Duarte, M.A.M. Hendrix, “Line-Interactive UPS Using

a Fuel Cell as the Primary Source” IEEE Transactions on Industrial Electronics, vol.55, no.8, pp.3012–3021, 2008.

[16] Available on http://www.m-field.com.tw/product2.php.

[17] I. Goethals, K. Pelckmans, J.A.K. Suykens, B. De Moor, “Subspace

identification of Hammerstein systems using least squares support vector machines”, IEEE Transactions on Automatic Control, vol. 50,

no. 10, pp. 1509–1519, 2005.

[18] P. Van Overschee, B. De Moor, “N4SID-Subspace algorithms for

the identification of combined deterministic-stochastic system”, Automatica, vol. 30, no. 1, pp.75–93, 1994.

[19] T.T. Georgiou, M.C. Smith, “Optimal Robustness in the Gap

Metric”, IEEE Transactions on Automatic Control, vol. 35, No.6, pp. 673–686, 1990.

[20] K. Glover, D.C. McFarlane, Robust Controller Design Using

Normalized Coprime Factor Plant Descriptions, Springer-Verlag,

New York, 1989. [21] D. McFarlane, K. Glover, “A loop-shaping design procedure using

Hinf synthesis”, IEEE Transactions on Autom Control, vol. 37, no. 6,

pp. 759–769, 1992.

978-1-4799-2625-1/13/$31.00 ©2013 IEEE 149