Integration of heat pumps in industrial processes

Pergamon Heat Recovery Systems & CHP Vol. 14, No. 3, pp. 287-296, 1994

Elsevier Science Ltd Printed in Great Britain

0890-4332/94 $7.00 + .00

I N T E G R A T I O N O F H E A T P U M P S IN I N D U S T R I A L

PROCESSES

ERIK WALLIN a n d THORE BERNTSSON

Chahners University of Technology, Department of Heat and Power Technology, S-412 96 Gothenburg, Sweden

(Received 2 September 1993)

Abstract--In spite of several technical and economic advantages, the number of heat pumps in industry is still very low compared to those for house heating. There are several reasons for this; one of the important ones being a lack of knowledge of how to find good, economic applications with the aid of process-integration principles. With the aid of these principles, the appropriate design in terms of heat pump type, size and heat source and sink temperatures can be identified. In doing that, the characteristics of both the industrial process and the heat pump must be taken into account. For the process the pinch temperature, the shape of the composite curves and the number of heat exchangers in the system are the most important factors. For the heat pump, the possible COPs that can be achieved and the ratio of heat to heat sink/heat from heat source are the most important factors, in addition to investment costs, energy prices etc. Methods for optimization of the main parameters in a grassroot design and for finding the most appropriate designs in a retrofit situation have been developed. With the aid of such methods, the potential for heat pumping in industry can be shown to be higher than earlier anticipated. Studies in real plants have verified this.

1. INTRODUCTION

Compared with those for house heating, heat pumps in industry have, in many cases, the following advantages [1]:

high COPs (coefficient of performance) due to possible small temperature lifts and/or high temperature level;

long annual operation time;

relatively low investment cost (short distances between heat sink and heat source);

heat source production and heat sink demand occur simultaneously.

In spite of this, the number of heat pumps in industry is still very low. Even in situations when the payback period would be acceptable, this technology is used only in some special applications. Some important reasons for this are:

lack of good hardware in some types of application;

lack of experimental and demonstration plants in different types of industry;

lack of knowledge of combination process technology/heat pump technology in industry, consulting companies, etc.

One other important reason, however, is the lack of knowledge in industry how to find good, economic applications with the aid of process-integration principles. Most heat pumps installed so far in industry are used in certain unit operations, such as drying and evaporation, and are therefore not process-integrated. By a systematic search for appropriate heat sources and heat sinks in the whole industrial plant, the true potential for different types of industrial heat pumps can be shown to be considerably higher than earlier anticipated.

2. PRINCIPLES OF PROCESS INTEGRATION WITH PINCH TECHNOLOGY

To integrate a heat pump, a good knowledge of the process is necessary. In this respect pinch technology is a very powerful tool in process analysis, see Linnhoff et aL [2].

287

288 E. WALLIN and T. BERNTSSON

In pinch technology analysis the process streams are characterized by their start and final temperatures and by their heat flow rate, i.e. the mass flow multiplied by the heat capacity. Streams requiring cooling are called hot streams, whatever their absolute temperatures are, and streams requiring heating are called cold streams.

One of the most important features of pinch technology is that it is possible to identify a temperature (the pinch) in a process which divides the stream system into two separate parts. In the part above the pinch there is a net heat deficit and heat must be added to the system by a hot utility. If a cold utility is applied above the pinch it follows that the demand for the hot utility will increase by the same amount. Thus, valuable heat is just removed by cooling. Conversely, in the part below the pinch there is an excess of heat that must be removed from the system by a cold utility. Any heat added below the pinch must also be removed. Hence, in a well-designed process no cold utility should be used above the pinch and no hot utility below the pinch.

Composite curves are useful tools in pinch technology. These consist of two curves representing the composite hot and cold streams. The curves are constructed by adding the heat content of all hot and cold streams respectively in each temperature interval. The accumulated heat content can then be plotted on a temperature/heat content diagram (Fig. 1). It is possible to exchange heat from the hot streams to the cold streams where the two curves overlap horizontally, and the vertical distance between the curves represents the temperature difference in the heat exchange. Where the two curves do not overlap, external heating or cooling by plant utilities is necessary.

Another often-used tool is the Grand Composite Curve (GCC). As is shown in section 4, it can be used for a good assessment of different heat-pump opportunities. It is constructed as the net heat needed (heat needed in all cold s t reams--heat available in all hot streams) in each temperature interval. As in the composite curves, a certain ATmin is subtracted from all hot streams corresponding to a necessary temperature difference in the heat exchange process. An example of a GCC is shown in Fig. 2. At the pinch temperature the net heat needed is zero. Above the pinch the distance between the ordinate and the curve represents the net heat needed at each temperature level and the distance at the highest temperature corresponds to the hot utility to the system. Below the pinch the horizontal distance represents the net heat available at each temperature level.

3. PROCESS I N T E G R A T I O N OF HEAT PUMPS

In Fig. 3, three ways of integrating a heat pump are shown. From the figure it is obvious that a heat pump must be integrated in such a way that the heat source is situated where there is an excess of heat, i.e. below the pinch, and the heat sink where there is a need for heat, i.e. above the pinch. In fact, the existence of a pinch makes a heat pump always thermodynamically feasible, as a certain amount of cooling below the pinch and heating above it would remain, even after the most intelligent heat-exchanging arrangement. However, in practice the technical and economic

1 4 0 -

M i n l m u ~ ~ .

hot N 120- utility ~o < > ~

~0 100- .~d' 0

8 0 - O') o 1 ) ..,I

f ~ . . . . . . . . . /

1 t / / /

~ o l d J . . • - ut.Bitv., neag exohange.

>

&H

-.s

¢1 4 0 -

E I - - 2 0 - Q L . . . . .

I I I I I 0 " 1 0 ' 1 "10 ' 2 " 1 0 ' 3 "10 ' 4"10" 5"10"

Heat Load (kW)

Fig. 1. Composite curves. Fig. 2. A typical example of a Grand Composite Curve.

~,mtm

Integration of heat pumps

II +,- tmlaRtaro

Fig. 3. Placement relative to the pinch of a heat pump.

289

constraints limit the actual potential for this technology. Hence the temperature level of the pinch in a process is a crucial parameter when assessing the opportunities for heat pumping.

The consequences when integrating a heat pump into a process can be analysed with the aid of the composite curves (Fig. 4). When taking heat to the heat pump below the pinch, the heat available for process heat exchanging decreases. This also means that the vertical distance between the curves decreases below the starting temperature of the heat source stream(s). Hence, due to this decrease in driving force, the area needed for process heat exchanging must become larger and possibly more heat exchanger units must also be added. The influence on the heat exchanger network obviously increases the closer to the pinch the heat source temperature is.

At each temperature level, the theoretically highest amount of heat that can be used as heat source corresponds either to the horizontal distance between the composite curves at that temperature (in reality a certain ATe. must be accounted for) or, if a new pinch is created at a lower temperature, the heat amount creating this new pinch. If one extracts more heat, say Qadd, it means that the composite curves must be moved from each other and hence that the hot utility increases with the same amount, i.e. Q,da. Therefore, this case means a need for a larger heat pump without any additional heat saving and consequently it can never be of interest. The same discussion as above can, of course, also be held for the situation above the pinch.

By process integrating the heat pump, instead of using it from the cold utility temperature to the hot utility one, it will, in most cases, by necessity become smaller but, conversely, it will be economic due to the smaller temperature lift and hence higher COP.

The approach for finding the most appropriate heat-pump installation in a given plant depends on whether it is a grassroot or a retrofit design problem. In a grassroot design, an optimization of the most important parameters can be done, whereas in a retrofit situation, the practical constraints make it necessary to have another approach. These aspects are discussed below. First, however, some important characteristics for heat pumps in a process integration context are discussed.

- - - With lns t ld led h l a t pump Original o o m p o s i t e ourves T

j j / / 7 ,+ - - - W i t h r u s t l e d

heat pump Orll~aal oomposlte OUlW~|

j f s Sf

I I

• # /

AH All

Fig. 4. Composite curves with and without installed heat pump for two cases of composite curves.


4. SOME C H A R A C T E R I S T I C S OF D I F F E R E N T HEAT PUMP TYPES

As has been shown, it is of high importance for a process integration study to include a detailed knowledge on the characteristics of both the heat pump and the industrial process. Therefore, some technical characteristics and constraints for the most often discussed types of heat pumps for industrial application are presented in this section.

One factor of importance when integrating a heat pump is the relation between the heat sink and heat source amounts for each type. This ratio is below called q. The two most important characteristics for a heat pump type from a process integration point of view are, therefore, the range of pinch temperature levels, in which it can operate, and the q-value. Of course, economic factors (energy prices and investment costs) also influence the proper choice of type.

The electrically driven compression heat pump cannot, with the working fluids used today, operate at condensing temperatures above approximately 120-130°C. Thus the pinch temperature must be lower than 120°C. Furthermore, to be economic the COP must probably be higher than at least 3, which limits the temperature difference between the condenser and evaporator. This also implies that the composite curves must allow the heat to be extracted and delivered not too far from the pinch, q = 1.2-1.5 (see also ref. [3]).

The mechanical vapour recompression heat pump can operate at relatively low condensing temperatures, 60-80°C, if the temperature lift is extremely low. For probably the most common case in process industry, production of process steam at 120°C or above, the heat source temperature must normally be higher than approximately 80°C for economic and construction- related reasons. Thus, in such cases the pinch temperature should be higher than, say, 100°C for this type. The temperature lift for the compressor involved should normally be relatively modest, which implies that the composite curves must allow heat to be extracted and delivered not too far from the pinch. COPs are normally high, 5 to at least 20, and q = 1.1-1.3 (see also ref. [4]).

The heat transformer can operate up to 150°C with a maximum temperature lift of about 50~C. The pinch temperature should therefore be between 60 and 130°C to fit the heat transformer. This type of heat pump operates without electricity and is capable of upgrading half of a heat source to a higher temperature. Heat sinks, which are half the size of the heat source, are therefore ideal (q = 0.5).

The absorption heat pump can operate up to 100°C with the working fluids used today, which implies that the pinch temperature must be lower than, say, 80°C to fit this type. The heat pump is driven by primary heat and with a constant COP of approximately 1.3 (if the boiler losses are included) and q = 2.7 (see also ref. [3]).

Obviously, the suitability for each type of heat pump in a given process varies with the energy characteristics of the process. This can easily be studied in a Grand Composite Curve. In Fig. 5 "ideal" forms of Grand Composite Curves for different types of heat pumps are shown. Hence, the Grand Composite Curve can be used to make an assessment of suitable type(s) of heat pump in a given industrial process.

5. I M P O R T A N T P A R A M E T E R S

As is obvious from the discussion in section 3, there is normally a range of technically possible sizes, as well as condensing and evaporation temperatures, in a given industrial process. In such cases a high COP should be possible to achieve, as the condensing and evaporation temperatures can be chosen close to the pinch. This means, however, that the size by necessity must normally be small (Fig. 4).

Another important aspect is the necessary changes in the heat exchanger network (HEN) when the hot utility (QH) shall be decreased. A decrease in Q , can be achieved by increased heat exchanging (both enlargements of existing units and totally new units in an existing plant), or by the use of a heat pump, or by a combination of these two techniques. In the first case the global ATmi, (AT at the pinch) between the composite curves (Fig. 4) is reduced, whereas in the second one it remains constant. Also in this case, however, there is a need for improvement of the HEN, normally both in terms of more area and more units (at least in a retrofitting situation). The reason is that a heat pump reduces the driving forces for heat exchanging at temperatures above the

Integration of heat pumps 291

T T b) l l l l w w .M imi

o) ~ heat pump

All

,l) Heat t n m l i w

lit

i

ill

lit

T o) llitlot ~ l i l

#, % / X

I l I !

! Y

Ill

Fig. 5. "Ideal" Grand Composite Curves for different types of heat pump.

condensing one, as well as below the evaporation one (Fig. 4). Hence, in order to achieve the desired reduction in QH, i.e. equalling the heat pump condenser output, this reduction in driving forces must be compensated for by improved heat exchanging. The investment costs for such heat exchanging vary with position of the heat pump and the total number of streams (i.e. with the original number of heat exchanger units). As is obvious from Fig. 4, the additional heat exchanger area needed increases with the size of the heat pump, with decreasing condensing temperature and with increasing evaporation temperature. The number of additional units needed for this area is a function of the number of streams influenced by the heat pump introduction and increases with the same parameters as above, as well as, in addition, with the total number of streams and heat exchangers in the system.


The influence on the HEN by a heat pump introduction, discussed above, is crucial to take into account when economic heat pump opportunities in a given industrial process are investigated.

Improved heat exchanging (reduction in ATtar,) and heat pumping are normally competing technologies for energy conservation in industry. However, they should be seen as two possibilities, which when combined can be a better solution than any of the two acting alone. This means that an improvement of the HEN, i.e. a certain reduction in ATtar,, before a heat pump is introduced can, in many cases, be the most optimal solution. Therefore, ATmi . should be seen as one of the main design variables when integrating a heat pump.

As is obvious from the discussion above (see also e.g. Wallin et al. [7]) the most important parameters to take into account are:

investment costs for pure heat exchanging;

COP for the heat pump;

investment costs for the heat pump and for necessary changes in the HEN (area and additional units).

The main design variables then are:

global ATmi n ;

type of the heat pump;

size of the heat pump;

condensing temperature;

evaporation temperature.

The most appropriate choice of these variables in a given industry depends on the economic criterion used for the evaluation, e.g. payback period (PBP), annual profit (at a given annuity factor), maximum allowed investment cost, or any combination of these. In a grassroot design, when all the parameters discussed above, including the whole heat-exchanger network, can be designed freely, all the main design variables can be optimized in detail [6, 7].

In the far more common situation, integration of a heat pump in a retrofitting situation in an existing plant, the fact that the HEN already exists and the numerous practical constraints this causes, normally makes it impossible to implement the results of a theoretical optimization. In this case another procedure must be used. A methodology, developed at our department, is presented below.

6. M E T H O D O L O G Y FOR A S S E S S I N G HEAT-PUMP OPPORTUNITIES IN RETROFITTING SITUATIONS

Because of the great number of technical and practical constraints in an existing industrial plant, the basic principle of the methodology is to produce a "map" of all economically feasible opportunities for heat-pump integration and then compare these results with the actual constraints. Opportunities, which are both practically possible and economically feasible, can thus be identified. In this way, a number of connected interesting values for all the main design parameters for a given type of heat pump can be obtained. The methodology, in which all these parameters are included, is presented below.

With the aid of a computer program a scanning procedure is performed, in which all thermodynamically possible heat-pump installations are assessed economically. For a given global AT~n, Tco,d and Tevap the payback period (PBP) for the range of thermodynamically possible heat-pump sizes for a given type is calculated. This procedure is repeated for a number of different T~,,a values until the whole possible range for this parameter has been scanned. The same is done for Tevap and, finally, the whole procedure is repeated for a number of decreasing global AT,~, values. For each point the investment cost for the heat pump itself and for the heat exchanger network (total area and number of units needed for (i) decreasing the global ATmi, and


(ii) compensating for the temperature driving forces when a heat pump is introduced), as well as the heat pump COP, are calculated. The PBP can then be calculated as:

where

and

Investment cost PBP = ( 1 )

AOperating cost

Investment cost = / c o s t HXareatincr.Hx) + / c o s t HXarea~Hp)

+ / c o s t unitstmcr.Hx) + / c o s t unitstHp> + /cost HP

AOperating cost = Acost HUti,cr Hx) + Acost HU~Hp) -- Drive energy costtMp).

The calculation procedure for the heat-exchanger area and the number of units needed is complex and is based on "traditional" pinch technology and methods developed at our department. It would be to go too far to discuss this in detail here. For more information, see Wallin and Berntsson [10].

With the aid of the PBP calculations, the most economic alternatives when the other common economic criteria, i.e. annual profit and maximum allowed investment cost, are used. For a given PBP, the annual profit can be calculated according to:

( ' ) Annual profit = p - ~ - a x Investment cost (2)

where a is the annuity factor. The most important part of the methodology is the approach for evaluation of all the PBP

calculation. This is presented in the next section.

7. E V A L U A T I O N OF D I F F E R E N T HEAT PUMP I N T E G R A T I O N O P P O R T U N I T I E S

The evaluation is done in three steps:

Identification of the most economic global ATmi n value;

For this ATmi,, identification of the most feasible combinations of size, Tcond and Tevap;

By comparison with actual practical constraints, identification of one or a few design parameter combinations on which the real design calculations should be based.

The identification of the most economic global ATmin value(s) is done by plotting all the scanning procedure results in a diagram with the total investment cost (nominator in the PBP expression) on the ordinate and the gross annual profit (denominator in the PBP expression) on the abscissa. A constant PBP value can then be shown as a straight line from origo and a line for a given annual profit as a straight line with the annuity factor as derivative and the interception with the abscissa giving the value of the annual profit.

An example of such a diagram is presented in Fig. 6. The different heat pump opportunities are shown with the global ATmin as a parameter. As a comparison, the opportunities for heat exchanging only (i.e. with no heat pump at all in the system) are also shown. For the sake of presenting the methodology, three different curves of this kind are included in Fig. 6 (in reality of course only one curve can exist). If the heat exchanging opportunities are as in case C in the figure, heat exchanging would always be more economic than heat pumping. On the contrary, for the A case, where the costs for pure heat exchanging quickly reaches infinity, heat pumping should normally be considered with no, or only a small, decrease of the global ATmi.. Finally, in the B case a combination of heat exchanging and heat pumping should lead to the most economic design. In the figure the global ATmi,, giving the lowest payback period or the highest annual profit with a maximally allowed investment cost (or any other subcriterion), can be identified easily. Sometimes the areas for different ATmi n a r e superimposed on each other. In such cases the following detailed evaluation must be done for a couple of chosen ATtain values.

HKS 14/3--F

294 E. WALLIN and T, BERNTSSON

PBP=4 Year PI~P=3 Year

A B C . \

/

i ' i ' ' i 0 .0 0 2 0.4 0 6 0.8 1.0 1.2 1 4

Annual operating cost decrease (MM$)

Fig. 6. Economic heat-pumping and heat-exchanging opportunities.

As has been shown, with the type of diagram shown in Fig. 6 a number of important conclusions can be drawn, namely:

overall economic heat pump opportunities when the competing technique of improved heat exchanging is taken into account;

possible range of PBPs for energy conservation with heat pumps;

most economic value(s) of the global ATmi..

The diagram in Fig. 6 does not say anything about the most economic combinations of size, Tco.d and Tcwp at a given AT.i.. This must therefore be evaluated separately in the following way.

When a AT,.i, value has been chosen, calculation results at this value are presented in diagrams with T ~ vs T~,p on the ordinate and abscissa, respectively, with the size as a parameter at constant value of the PBP, Diagrams at some different values of PBP should be produced. Two examples of such diagrams are presented in Fig. 7. They are the results of calculations for a set of conditions for the MVR type of heat pump. Details for the conditions chosen can be found in Wallin and Berntsson [11]. The results for two PBP values, 3.5 and 2.6 years, are shown. As can be seen in

A 1.50-

+E 14s-

.~ 140-

~135-

8 130-

3 ~ 12S-

~ 120- 'E

PBP=3.5 Years Heat pump sizes in MW

E "o

2 J v

®

"E o

i i i i i i i ,~ 120 115 110 t0S 10o 95 90

N e t w o r k w m p e n l t u r e , e v a p o r a t o r s i d e ( C e n t i g r a d e s )

150-

145-

140-

135-

130-

120-

11S,

125

PBP=2.6 Years Heat pump sizes in MW

8

8

i i i r i i - - q 12o 11s 11o lO5 10o 9s 9o

N e t w o r k t e m p e r a t u r e , e v p a o r a t o r s i d e ( C e n t i g r a d e s )

Fig. 7. Net work temperature for heat-pump integration, condenser side versus evaporator side with the size as a parameter for two payback periods.


the 3.5 years case, many combinations of size, Too~a and To,ap give the same PBP. The ones at the upper-right part of the diagram have, relatively speaking, high operating costs and a low investment cost (high temperature lift, giving a low COP and a small influence on the heat exchanger network), whereas the ones at the lower left have low operating costs and high investment cost (low temperature lift, high influence on the heat exchanger network). This means that the combinations at the lower left have a higher annual profit, due to a higher investment cost [see equation (2)], whereas the ones at the upper right influence a smaller number of heat exchanger units and should therefore create fewer problems with installation, flexibility and control.

Moving to a lower PBP means that larger sizes appear in the diagram and at high temperature lifts only. The reasons for this are that the temperature lifts needed in this example, which result in no, or only a small, influence on the HEN, still give reasonable COPs and, hence, a low PBP and that these low PBPs are impossible to achieve for the smaller sizes, due to their higher specific investment cost.

For the conditions given in the example, a higher temperature lift gives opportunities both for decreasing the PBP and for increasing the size. These results are common also in other conditions. Hence, in many cases there should be more advantages than drawbacks when placing a heat pump at a relatively high temperature lift. However, sometimes it is necessary to decrease the temperature lift, if reasonable PBPs and/or annual profits are to be achieved.

The consequences for the PBP of a change in temperature lift for a given heat pump size depends on the situation. An increase in temperature lift means:

an increase in PBP, if the total annual heat pump cost is determined mainly by the operating one;

a decrease in PBP, if the total cost is determined mainly by the investment one.

The discussions above have been done for one example of heat pump type only. Different types have different operating characteristics and different operating and investment costs, which, of course, will result in different integration possibilities. One extreme example would be the Heat Transformer (Absorption Heat Pump, Type II), which has practically no operating cost at all. However, the principles and the methodological approach discussed in connection with the diagrams can be used for all types of heat pump. The method should be an excellent tool for making comparisons between different heat pump types.

With the aid of the PBP calculations in the scanning procedure, more detailed conclusions about the most economic designs can be drawn, also when the annual profit and/or the maximum investment cost level are used as criteria, Wallin and Berntsson [11].

From the discussion above, the following general conclusions can be drawn from the type of diagrams shown in Fig. 7:

There are normally many combinations of size and condensing and evaporation temperatures, which result in the same payback period;

The most economic sets of combinations can be identified in the diagrams presented;

With these diagrams any heat pump installation can be economically evaluated and be compared with practical constraints;

Although not always totally true, normally as large a heat pump as possible at a specified demand for payback period should be chosen, due to its high annual profit;

For a heat pump of a given size, the same payback period can, in many cases, be achieved at both a high temperature lift (low HEN investment costs, high operating costs) and a low one (high HEN investments costs, low annual operating costs). The high temperature lift case normally gives small practical problems (flexibility and control aspects, changes in the HEN) but a somewhat lower annual profit.

8. EXPERIENCES AND ONGOING WORK

Studies of process integration of heat pumps have been done in, e.g., the U.S., the U.K. and Sweden. The general results from such studies are that the economic potential for heat pumping


in industry can be shown to be clearly higher than earlier anticipated. The most comprehensive study has been performed in the U.S.A., [9]. In 1984 the U.S. Department of Energy initiated a project in which 26 different industrial processes were examined and of these 10 were selected, primarily on the basis of heat-pump potential. In these 10 processes good opportunities for heat pumping (electrically driven mechanical-vapour recompression) were found in at least 8, all with PBPs less than, or around, 3 years. This result was reached in spite of the fact that each process was first modified for economically optimal heat exchange. Savings with heat exchange were less than those with heat pumping, which gives a good illustration of the discussion around Fig. 6.

In subsequent studies by both U.S. DOE and EPRI, the same types of results have in principle been found, also when including other types of heat pump.

Based on encouraging results like these, an international cooperation in this area has started within lEA, called Advanced Heat Pumps, Annex 21: Global Env ironmenta l Benefi ts o f Industrial

H e a t Pumps , with at least 8 countries participating. The main aims of this Annex are:

To assess the impact on the greenhouse effect using IHPs;

To disseminate present knowledge of IHPs to potential industrial users;

To provide computerized tools for assessing the technical and economic opportunities for different types of industry;

To assess the economic potential for IHPs in different types of industry in each participating country.

In this work the methods for both grassroot design and retrofit situations described above will be used.

9. C O N C L U S I O N S

With the aid of process integration technologies, the potential for heat pumping in industry can be shown to be clearly higher than earlier anticipated. Methods for optimization of the main parameters in a grassroot design and for finding the most appropriate designs in a retrofit situation are available. Results from studies on real plants have verified the strength and appropriateness of these methods.

R E F E R E N C E S

1. T. Berntsson, Future prospects for industrial heat pumps in Europe. Proc. 1987 lEA Heat Pump Conference, Orlando, FL, pp. 303-324 (1987).

2. B. Linnhoff, D. W. Townsend, D. Boland, G. F. Hewitt, B. E. A. Thomas, A. R. Guy and R. H. Marsland, A Users Guide on Process Integration for the Effcient Use of Energy, 1. Chem. E., Rugby, U.K. (1982).

3. T. Berntsson, in Heat Pumps, Electricity--Efficient End-Use and New Generation Technologies and their Planning Implications (edited by T. B. Johansson, B. Bodlund and R. H. Williams), pp. 173-216. Lund University Press (1989).

4. K. Munch Berntsson, Mechanical Vapour Recompression--lnternational State-of-the-Art, Report from Energiteknisk Analys, Department of Heat and Power Technology, CTH (1991) (in Swedish).

5. K. Munch Berntsson, T. Berntsson, P. ,~, Franck, P. Holmberg and E. Wallin, Heat Transformers in Industrial Processes, CADDET Analyses Ser. No. 2 (1989).

6. S. M. Ranade and M. O. Sullivan, Optimal integration of industrial heat pumps, lchemE Symp. Ser. No. 109 pp. 303- 325 (1988).

7. E. Wallin, P./~,. Franck and T. Berntsson, Heat pumps in industrial processes--An optimization methodology. Heat Recovery Systems and ClIP 10, 437-446 (1990).

8. P. Holmberg, System studies and optimization of the single-stage absorption heat transformer cycle. Licentiate thesis, Department of Heat and Power Technology, Chalmers University of Technology, Sweden (1988).

9. P. Scheihing, Experiences with process integration of heat-pumps within complex industrial processes, in Proc. lEA Workshop on Process Integration (edited by T. Berntsson), Gothenburg, Sweden (1992).

10. E. Wallin and T. Berntsson, Heat Exchanger Network Consequences by Combined Heat Pumping and Improved Heat Exchanging. To be published.

11. E. WaUin and T. Berntsson, Process Integration of Heat Pumps in Retrofitting Situations. To be published.

Documents

Integration of heat pumps in industrial processes