International Journal of Mechanical Engineering and Technology (IJMET)

Volume 9, Issue 5, May 2018, pp. 647–654, Article ID: IJMET_09_05_071

Available online at http://www.iaeme.com/ijmet/issues.asp?JType=IJMET&VType=9&IType=5
ISSN Print: 0976-6340 and ISSN Online: 0976-6359

© IAEME Publication Scopus Indexed

ENERGY SYSTEM ANALYSIS

Ranjith Soundararajan and Tobias Richards
Department of Resource Recovery (Sustainable Engineering)
School of Engineering, University of Boras, Sweden

ABSTRACT
In the past decades, many synthesis of heat exchanger network is proposed to
minimize the annual cost. The purpose of this thesis is to use a model to optimize the
heat exchanger network for process industry and to estimate the minimum cost
required for the heat exchanger network without compromising the energy demand by
each stream as much as possible with the help of MATLAB programming software.
Here, the optimization is done without considering stream splitting and stream
combining. The first phase involves with deriving a simple heat exchanger network
consisting of four streams i.e... Two hot streams and two cold streams required for the
heat exchanger using the traditional Pinch Analysis method. The second phase of this
work deals with randomly placing the heat exchanger network between the hot and
cold streams and calculating the minimum cost of the heat exchanger network using
genetic coding which is nothing but thousands of randomly created heat exchangers
which are evolved over series of population.
Keywords: Pinch technology, heat exchanger network, MATLAB, linear
programming, genetic algorithm
Cite this Article: Ranjith Soundararajan and Tobias Richards. Energy System
Analysis, International Journal of Mechanical Engineering and Technology, 9(5),
2018, pp. 647–654.
http://www.iaeme.com/IJMET/issues.asp?JType=IJMET&VType=9&IType=5

1. INTRODUCTION
In today’s process industry, the importance of devouring energy is increased as the power
demand for the world is increased. To reduce the energy utilization or devouring, it became
necessary to pay attention to the heat exchanger network design. This is because the rate of
consumption of fossil fuels increases and as a result the cost and emissions such as co2 etc.
Are also increased (akbarnia et al., 2009). In the process industry, there are many streams
which needed to be heated or cooled. This heating and cooling can be achieved by external
utilities such as heaters and coolers respectively. The energy analysis or integration helps to
find thermodynamically possible stream matches which is followed by introducing heat
exchangers and thereby reducing the energy consumed by other utilities (gu and vassiliadis,
2014). The energy integration or heat integration uses different approaches such as insight
based methods and optimization based methods. To calculate the energy targets the insight

http://www.iaeme.com/IJMET/index.asp 647 editor@iaeme.com

 Energy System Analysis

method uses the graphical tools such as grand composite curve whereas the optimization
method uses mathematical programming to minimize or maximize an objective function such
as cost or profit based upon the constraints related to the heat exchanger (bonhivers et al.,
2014).
The commonly used heat exchangers in industrial practice are multipass heat exchangers
as they have advantages such as easy mechanical cleaning, allowance for thermal expansion,
longer flow paths for given heat exchanger length, as well as good heat transfer coefficient. In
the last few decades’ notable research efforts on the synthesis of the heat exchanger network
(hen) have been performed. Despite that, most of the published methods for the synthesis heat
exchanger network are related to single pass heat exchangers. One of the most widely used
method is called pinch technology, which was developed by bodo linnhoff and his
collaborators at ici, union carbide, and the university of manchester. The pinch technology is
based upon first and second law of thermodynamics. It is possible to design multipass heat
exchanger network based upon the concepts of pinch technology (sun and luo, 2011). The
pinch analysis do not give any global optimal solution but it is widely used due to its
simplicity of the concepts (bonhivers et al., 2014). To enhance the rate of heat transfer
Karunamurthy et al (2017) used twisted tapes of twist ratios Y=7, Y=9 and Y=11 in the flow
passage of the heat exchanger located in the lower convective zone of salinity gradient solar
pond and improved its performance considerably. To utilize the waste heat of engine exhaust
Inakollu Bhanu Harsha et al (2018) used passive heat transfer enhancement and augmented
the rate of heat transfer. Karunamurthy et al (2014) analyzed the performance of in-pond heat
exchanger of solar pond both experimentally and computationally under laminar and turbulent
flow conditions for Reynolds number varying from 1804 to 9020.

1.1. Genetic algorithm:

The optimization process consists of forming an initial basic concept and improving the
concept based upon gained information to find the maximum or minimum output results. In
the optimization process the input may be a variable or functions such as objective function or
cost function or fitness function (Haupt and Haupt, 2003).
There are different types of optimization technique but the most commonly used
technique to find the global minimum or maximum is genetic optimization. In 1960, Genetic
Algorithm (GA) was invented by John Holland and later in the year 1960s and 1970s the
algorithm was developed by Holland and his students and colleagues at the University of
Michigan. Original goal was to design an algorithm to understand the phenomenon of nature
adaptation and to find ways to implement this process into computer systems (Mitchell,
1996). Generally speaking, genetic algorithms are probabilistic optimization methods in
which the concept of evolution is used to improve the solution to find better global minimum
or maximum value (Bodenhofer, October,2003).
In the genetic algorithm, to a given problem or process each individual represents a
possible solution where the individuals are determined randomly from search space. The
fitness function or objective function determines the result of the variable that needs to be
optimized to find the fitness of the solution. The next generation of population is determined
by the individuals that has the best fitness solution within the previous population. An optimal
solution is obtained for a given problem from the group of possible solutions.
(m.a.s.s.ravagnani, june, 2005)

http://www.iaeme.com/IJMET/index.asp 648 editor@iaeme.com

 Ranjith Soundararajan and Tobias Richards

2. ANALYTICAL PROCEDURES
The methodology implemented in this thesis consists of two distinct stages. In the first stage,
Pinch Analysis is used to find the pinch temperature or pinch point and in the second stage the
heat exchangers are placed between the cold and hot streams using genetic coding to find the
optimal solution that is optimal operating cost of the plant. The tool used here for
optimization is MATLAB. The below steps explains the process of energy analysis

2.1. Step 1

The above flow chart explains the steps in Pinch Analysis (Tewari et al., 2015). In the
Pinch Analysis step, the actual steam temperatures are converted to interval temperatures
using the below equations,
Hot streams Tint = Tact - ΔTmin/2 (1)
Cold streams Tint = Tact + ΔTmin/2 (2)
Where, Tint is the interval temperature, Tact is the actual temperature, ΔTmin is the
minimum temperature difference. The minimum temperature difference is assumed and
interval temperature is calculated.

http://www.iaeme.com/IJMET/index.asp 649 editor@iaeme.com

 Energy System Analysis

From appendix–1 shows the codes which are used to in the construction of hot and cold
composite curve to implement the Pinch Analysis in MATLAB. The initial heating value is
assumed to be zero. After finding the temperature intervals, there will be some duplicate
temperatures, which are arranged in the order of magnitude. Heat balance is carried out for
streams within each temperature intervals. There will be few negative values which are
eliminated by introducing some heat.
The pinch analysis is used as a function file which will be used later for heat exchanger
network synthesis. The minimum temperature difference is added manually to find the
possible pinch temperatures. A grand composite curve is then constructed from merging the
hot and the cold composite curve in one graph. From the grand composite curve the pinch
temperature is determined.

2.2. Step 2
The equation for area of heat exchanger and cost function or objective function to run the
gene coding is implemented as a function in MATLAB. The equation for calculating area is
given below,
Q = U A (ΔT)lm (3)
Q = ṁCp,h (Th,in - Th,out) = ṁCp,c (Tc,in – Tc,out) (4)
ΔTlm=[(Th,in - Th,out) + (Tc,in – Tc,out)]/[ln((Th,in - Th,out)/(Tc,in – Tc,out))] (5)
Where, Q is the heat transfer rate (W),
U is the overall heat transfer co-efficient (w/km2),
A is the surface area of the heat exchanger (m2),
(ΔT)lm is the logarithmic mean temperature difference (K),
Th,in is the inlet temperature of hot stream (K),
Th,out is the outlet temperature of hot stream (K),
Tc,in is the inlet temperature of cold stream (K),
Tc,out is the outlet temperature of cold stream (K),
ṁ = mass flow rate of fluid (kg/sec),
Cp,h and Cp,c is the specific heat capacity of hot and cold fluid (J/kg.K).
To calculate the area, heat transfer rate is calculated using eq-15. After finding the heat
transfer rate, it is substituted in eq-14 followed by the logarithmic mean temperature
difference and overall heat transfer co-efficient. The above equations are then implemented as
a function so that they can be used by the gene coding for optimization.
The objective function is the cost equation which is also used as a function file along with
area in MATLAB. The cost function used is as follows,
Cost of heat exchanger = A1 + A2 [area]A3 $/yr (6)
Where A1, A2 and A3 are constants in the cost function.
Capital cost = cost of heat exchangers + operating cost of heat exchangers + cost of hot
utilities + cost of cold utilities $/yr (7)

2.3. Step 3
The final step involves in forming a gene coding system to optimize the cost and reduce the
energy usage. It is done by combining the pinch analysis along with the cost estimation in the
gene coding. First the program is written in such a way that it should randomly place the heat
exchanger between the hot stream from cold streams. The constraint used is that cold stream
temperature is less than the hot stream temperature. The next step is to allot the number of

http://www.iaeme.com/IJMET/index.asp 650 editor@iaeme.com

 Ranjith Soundararajan and Tobias Richards

heat exchangers and number of solutions to be saved. After allotting the number of heat
exchangers and number of solutions, only possible random connections between the streams
are made. After arranging the heat exchangers, the order of streams is corrected and the
boundary conditions such as the lower limit and upper limits are set. And initial guesses are
made. The matlab function ‘fmincon’ is used to find the minimum cost based upon the
placement of heat exchangers. In this case the gene coding selects the individual streams
based upon the cost function and place the heat exchanger between the hot and cold streams.
In this particular case, cross-over is not included.

3. RESULTS AND DISCUSSION

To test the coding designed, a simple case study was chosen from Gundersen model. The
data’s from Gundersen model is represented in the below table – 1,

Table 1 problem data for case study (Escobar and Trierweiler, 2013)
2
Streams Tin (⁰C) Tout (⁰C) ṁCp (kW/⁰C) H (kW/m ⁰C)
H1 270 160 18 1
H2 220 60 22 1
C1 50 210 20 1
C2 160 210 50 1
CU 15 20 - 1
HU 250 250 - 1
A
Cost of heat exchanger = A1 + A2 [area] 3 $/yr
A1 = 4000; A2 = 500; A3 = 0.83
Cost of cold utility = 20 ($/kW.yr)
Cost of hot utility = 200 ($/kW.yr)
Number of heat exchangers = 6
The above values or data were implemented on program and it was compared with the
original solution.

Figure 1 Hot and cold composite curve obtained from MATLAB for the above data.

http://www.iaeme.com/IJMET/index.asp 651 editor@iaeme.com

 Energy System Analysis

Figure 2 Grand composite curve obtained from MATLAB for the above data.
From the data in table – 1, the hot and cold composite curve and the grand composite
curve is constructed. The idea is that the user input stream data and the program calculates the
32 minimum energy demand for cooling and heating as well as identifies the pinch point. In
fig - 16 the x-axis heat load is plotted against the y-axis interval temperature to form the hot
cold composite curve. The pinch point obtained for this case study is 165 ⁰C.

Figure 3 Actual placement of heat exchangers from reference model. (Escobar and Trierweiler, 2013).

Figure 4 Obtained placement of heat exchanger from MATLAB

http://www.iaeme.com/IJMET/index.asp 652 editor@iaeme.com

 Ranjith Soundararajan and Tobias Richards

Table 2 Comparing costs and utility demand:

Cost and Utility Reference model Obtained model
Operating cost ($/yr) 128000 128000
Min cold utility (kW) 400 400
Min hot utility (kW) 600 600
Capital cost ($/yr) 248387.2 270700
After the grand composite curve and the pinch temperature is found, the program is
entering the gene coding stage where it’s find the best possible match between the streams
and finds the optimal heat exchanger cost. From figure – 4 it can be see that the placement of
heat exchangers is different between the obtained one and the actual reference model. From
the table – 2, it can be noted that the operating cost and the hot and cold utility demand
remains the same for both actual and obtained model. There is some difference in the capital
cost between the actual and the obtained because the actual network model considered the
steam splitting and steam combining in their heat exchanger network design. The obtained
program cannot handle steam splitting and steam combining.

Figure 5 A graph plotted between the ΔTmin and the capital cost.
From figure – 5 it can be noted a graph is plotted between change in ΔTmin and capital
cost to find the effects of ΔTmin on the cost. From the graph, as ΔTmin increases the cost
increases but below ΔTmin = 6 the cost is constant or there are only minor changes in the
value of cost. The reason why a cost increase is not seen for small values of ΔTmin in this
optimization is that the program optimizes with the target of minimizing the cost with a
temperature difference of at least ΔTmin, but it does not require the specific temperature
difference to be achieved.

4. CONCLUSION
The most studied problems in the process synthesis is the heat exchanger network synthesis
but finding a feasible solution even for a small-scale problem has been troublesome. In this
work, the main optimization for the heat exchanger network and reduction of cost is carried
out using genetic algorithm or gene coding. The coding method implemented is tested on a
reference model which is simple that is consisting of four streams were chosen and the
obtained results were compared with the actual one. In this synthesis, the algorithm can
evolve towards the optimum solution by implementing a good starting point to start the
program. The method used in this work cannot handle steam splitting and steam combining
thus a perfect optimal solution was not found or the capital cost was not equal to the actual

http://www.iaeme.com/IJMET/index.asp 653 editor@iaeme.com

 Energy System Analysis

reference model. As the coding can handle complex cases such as increase in number of heat
exchangers and number of streams, it can be used in finding optimal solution for process
plants containing multiple heat exchangers and multiple stream.

REFERENCES
[1] Akbarnia, M., Amidpour, M. & Shadaram, A. 2009. A new approach in pinch technology
considering piping costs in total cost targeting for heat exchanger network. Chemical
Engineering Research and Design, 87, 357-365.
[2] Bonhivers, J.-C., Srinivasan, B. & Stuart, P. R. 2014. New analysis method to reduce the
industrial energy requirements by heat-exchanger network retrofit: Part 1 – Concepts.
Applied Thermal Engineering.
[3] Escobar, M. & Trierweiler, J. O. 2013. Optimal heat exchanger network synthesis: A case
study comparison. Applied Thermal Engineering, 51, 801-826.
[4] GU, K. & Vassiliadis, V. S. 2014. Limitations in using Euler's formula in the design of
heat exchanger networks with Pinch Technology. Computers & Chemical Engineering,
68, 123-127.
[5] Sun, L. & Luo, X. 2011. Synthesis of multipass heat exchanger networks based on pinch
technology. Computers & Chemical Engineering, 35, 1257-1264.
[6] Tewari, K., Agrawal, S. & Kumar Arya, R. 2015. Generalized Pinch Analysis Scheme
Using MATLAB. Chemical Engineering & Technology, 38, 530-536.
[7] Haupt, R. L. & Haupt, S. E. 2003. Practical Genetic Algorithms.
[8] Mitchell, M. 1996. An introduction to genetic algorithm.
[9] Bodenhofer, U. October, 2003. Genetic Algorithm: Theory and Applications.
[10] Ravagnani, M.A.S.S., Silva, A.P., Arroyo, P.A. & Constantino, A.A. 2005. Heat
exchanger network synthesis and optimization using genetic algorithm. Applied Thermal
Engineering, 25, 1003-1017.
[11] Karunamurthy K, Nayan Singh, Vivek Singh, Experimental Study of In – Pond Heat
Exchanger Provided with Twisted Tapes, Asian Journal of Pharmaceutical and Clinical
Research, 2017, pp. 467-470.
[12] Inakollu Bhanu Harsha, Kasyap Addepalli, K Karunamurthy, Heat Transfer Augmentation
on IC Engine Waste Heat Recovery System, International Journal of Mechanical
Engineering and Technology, 2018, Vol. 9(1), pp. 924-933.
[13] K. Karunamurthy, K Murugumohan Kumar, S Suresh, Performance Analysis of an In-
Pond Heat Exchanger of a Salt Gradient Solar Pond, Applied Solar Energy, 2014, Vol.
50(2), pp.84-89.
[14] M. Thirupathaiah, P. Venkata Prasad and V. Ganesh, Analysis of Various Compensation
Devices For Power Quality Improvement In Wind Energy System. International Journal
of Electrical Engineering & Technology, 7(3), 2016, pp. 25–39
[15] Neeraj Sharma, Jimmy Kansal, Ashwagosha Ganju, Off-Grid Hybrid Renewable Energy
System Sizing for High Altitude Cold Deserts, International Journal of Advanced
Research in Engineering and Technology (IJARET), Volume 4, Issue 7, November -
December 2013, pp. 101-108
[16] K.Rama Lingeswara Prasad, Dr.K.Chandra Sekhar, A New Sensor less Control Strategy
for a Variable Speed PMSG Wind Energy System Connected to Grid, International
Journal of Electrical Engineering & Technology (IJEET), Volume 4, Issue 5, September –
October (2013), pp. 146-154
[17] Anto Joseph, Nagarajan, Antony Mary, A Multi Converter Based Pure Solar Energy
System with High Efficiency MPPT Controller, International Journal of Electrical
Engineering & Technology (IJEET), Volume 4, Issue 4, July-August (2013), pp. 205-212

http://www.iaeme.com/IJMET/index.asp 654 editor@iaeme.com