CHAPTER 3 SHELL AND TUBE HEAT EXCHANGER

Transcription

1 20 CHAPTER 3 SHELL AND TUBE HEAT EXCHANGER 3.1 INTRODUCTION A Shell and Tube Heat Exchanger is usually used for higher pressure applications, which consists of a series of tubes, through which one of the fluids runs. The second fluid flows over the tubes to be heated or cooled. The set of tubes is called tube bundle, and may be composed of several types of tubes, differing in shape and design: plain, longitudinally finned, etc. Extensive research work have been done in the past, design and construction of shell and tube heat exchanger by Anantharaman (1997), design of shell and tube heat exchangers when the fouling depends on local temperature and velocity by David Butterworth (2002), changing different parameters to increase shell and tube heat exchanger performance by Lunsford (1998), studies on performance characteristics of variable area heat exchangers by Narayanan (1998), visualization and determination of local heat transfer coefficients in shell and tube heat exchangers, the effect of leakage on pressure drop and local heat transfer by Huadong Li and Volker Kott Ke (1998), determination of convective heat transfer and friction losses in helically enhanced tubes for both Newtonian and non-newtonian fluids by Rozzi and Massini (2007), developing mathematical models for the estimation of hot and cold fluid outlet temperatures as a function of flow rates and inlet temperatures for a shell and tube heat exchanger by Mandavgane (2006). But not much details are available in literature regarding performance analysis with miscible and immiscible systems in the tube side flow. Hence efforts

2 21 have been made to study the performance of shell and tube heat exchanger with miscible and immiscible systems. In this part of the thesis a shell and tube heat exchanger having dimensions, 150 mm inner shell diameter, 615 mm length and 32 tubes with square pitch was taken for the experimental studies. The experimental studies involved in the determination of outlet temperature of both cold and hot fluid for various flow rates. The Water-Water System, Kerosene-Water System, Toluene-Water System, Acetic acid-water System and Ethylene Glycol- Water System 9%, 10%, 20% and 25% composition on volume basis were used to determine the performance of shell and tube heat exchanger i.e. Overall Heat Transfer Coefficient (U 0 ), Effectiveness (ε), Shell Efficiency (η S ) and Tube Efficiency (η T ). These experimental data were used to develop NN using general regression. Further, these networks were tested with a set of testing data and then the simulated results were compared with the actual results of the testing data. 3.2 EXPERIMENTAL STUDIES The shell and tube heat exchanger under this study is made of mild steel and consists of outside shell diameter of 0.15 m and 32 inner tubes of outer diameter m with square pitch arrangement. Four baffles with 0.1 m baffle spacing are used. The experiments are carried out with 1-1 pass with both parallel and counter flow patterns with cold fluid (miscible and immiscible systems) in shell side and hot fluid (water) in tube side. The centrifugal pump of 0.5HP, water softener, rotameter and storage vessel of 100 liter capacity is installed for carrying out the experiment. The schematic diagram of 1-1 shell and tube heat exchanger is shown in Figure 3.1.

3 22 Shell outlet Tube outlet Tube inlet Shell inlet Figure 3.1 Schematic Diagram of 1-1 Shell and Tube Heat Exchanger The schematic diagram of the experimental set up of shell and tube heat exchanger with all accessories is shown in Figure 3.2. At the start of the experiment exchanger was washed with water. The stored cold fluid was pressurized to the tube side of the equipment using a centrifugal pump. The flow control was achieved by a Rotameter connected between the pump and the heat exchanger. Before switching on the heater ensure that the tank is filled with water. A control valve controls the flow of hot fluid to the tube Shell and Tube Heat Exchanger, 2 Hot Water Tank, 3 Level Indicator, 4 Drain, 5 Hot Liquid Inlet, 6 Cold Liquid Tank, 7 Hot Liquid Outlet, 8 Cold Liquid Outlet, 9 Rotameter, 10 Pump, 11 Collecting Tank Figure 3.2 Schematic Diagram of Experimental Setup of Shell and Tube Heat Exchanger

4 23 The overhead tank was first filled with water and the heater was switched on and heated till temperature reaches 90 o C (T 1 ). The inlet of cold fluid was also noted and then the cold fluid valve was opened. The pump was started and the required water flow rate of the cold fluid was fixed using Rotameter. The hot fluid inlet valve was opened till the steady state has been reached. The flow rate of the hot fluid, the outlet temperature of the hot fluid (T 2 ) and cold fluid (t 2 ) was measured. The flow rate of the cold fluid was changed and new steady state reached. The above procedure was repeated for the all set of readings. The experimental set up of 1-1 shell and tube heat exchanger is shown in Figure 3.3. Figure 3.3 Experimental Setup of Shell and Tube Heat Exchanger 3.3 EXPERIMENTAL DATA FOR PARALLEL FLOW The experimental data for parallel flow shell and tube heat exchanger for Water-Water system, Immiscible and Miscible Systems are shown in the Tables 3.1 to 3.5 (The model calculation for 9% Ethylene Glycol Water system is given in Appendix 1).

5 Table 3.1 Water-Water System Temperature Cold Fluid (Water) Mass Flow Rate Reynolds Number Hot Fluid (Water) Nusselt Number Temperature Mass Flow Reynolds Number Nusselt Number Shell Heat Transfer Coefficient Tube Overall Capacity Rate Ratio No. of Transfer Units Effectiveness Efficiency K kg/s K kg/s W/m 2 K W/m 2 K W/m 2 K % % % t 1 t 2 m c N Re N Nu T 1 T 2 m h N Re N Nu h o h i U o R c NTU ε η S η T Shell Tube 24

6 Table 3.2 Kerosene-Water (Immiscible) System Temperature Cold Fluid (Kerosene) Hot Fluid (Water) Heat Transfer Coefficient Mass Reynolds Nusselt Mass Reynolds Nusselt Shell Tube Flow Temperature Overall Number Number Flow Number Number Rate Capacity Rate Ratio No. of Transfer Units Effectiveness Efficiency K kg/s K kg/s W/m 2 K W/m 2 K W/m 2 K % % % t 1 t 2 m c N Re N Nu T 1 T 2 m h N Re N Nu h o h i U o R c NTU ε η S η T 9% Kerosene Water System % Kerosene Water System % Kerosene Water System % Kerosene Water System Shell Tube 25

7 Table 3.3 Toluene-Water (Immiscible) System Temperature Cold Fluid (Toluene) Hot Fluid (Water) Heat Transfer Coefficient Mass Flow Reynolds Nusselt Temperature Mass Reynolds Nusselt Shell Tube Overall Number Number Flow Number Number Rate Capacity Rate Ratio No. of Transfer Units Effectiveness Efficiency K kg/s K kg/s W/m 2 K W/m 2 K W/m 2 K % % % t 1 t 2 m c N Re N Nu T 1 T 2 m h N Re N Nu h o h i U o R c NTU ε η S η T 9% Toluene Water System % Toluene Water System % Toluene Water System % Toluene Water System Shell Tube 26

8 Table 3.4 Acetic acid-water (Miscible) System Temperature Cold Fluid ( Acetic Acid) Hot Fluid (Water) Heat Transfer Coefficient Mass Reynolds Nusselt Mass Reynolds Nusselt Shell Tube Flow Temperature Overall Number Number Flow Number Number Rate Capacity Rate Ratio No. of Transfer Units Effectiveness Efficiency K kg/s K kg/s W/m 2 K W/m 2 K W/m 2 K % % % t 1 t 2 m c N Re N Nu T 1 T 2 m h N Re N Nu h o h i U o R c NTU ε η S η T 9% Acetic Acid Water System % Acetic Acid Water System % Acetic Acid Water System % Acetic Acid Water System Shell Tube 27

9 Table 3.5 Ethylene Glycol-Water (Miscible) System Temperature Cold Fluid (Ethylene Glycol) Hot Fluid (Water) Heat Transfer Coefficient Mass Reynolds Nusselt Mass Reynolds Nusselt Shell Tube Flow Temperature Overall Number Number Flow Number Number Rate Capacity Rate Ratio No. of Transfer Units Effectiveness Efficiency K kg/s K kg/s W/m 2 K W/m 2 K W/m 2 K % % % t 1 t 2 m c N Re N Nu T 1 T 2 m h N Re N Nu h o h i U o R c NTU ε η S η T 9% Ethylene Glycol Water System % Ethylene Glycol Water System % Ethylene Glycol Water System % Ethylene Glycol Water System Shell Tube 28

10 EXPERIMENTAL DATA FOR COUNTER CURRENT FLOW The experimental data for counter current flow shell and tube heat exchanger for Water-Water system, Immiscible and Miscible Systems are shown in the Tables 3.6 to RESULTS AND DISCUSSION In this part of the thesis, the performance characteristics of shell and tube heat exchanger for parallel flow and counter current flow i.e. Overall Heat Transfer Coefficient in terms of Nusselt number for cold and hot fluid, Effectiveness, Efficiency for cold fluid side and hot fluid side related with respect to Reynolds number of the cold fluid are illustrated from Figures 3.4 to Effect of Flow Rate of the Cold Fluid In the Figures 3.4 to 3.8 and Figures 3.24 to 3.28, Nusslet number (Cold) versus Reynolds number (Cold) are plotted which shows that heat transfer coefficients increases with increase in the mass flow rate of cold fliud. This is because increase in flowrate of cold fluid increases the Reynolds number and Nusselt number which inturn increases the individual heat transfer coefficients for hot fluid. The increase in these heat transfer coefficients will increase the overall heat transfer coefficient. In the Figures 3.9 to 3.13 and Figures 3.34 to 3.38, Nusselt number of the hot fluid versus Reynolds number of the cold fluid are plotted, which shows that the Nusselt number of the hot fluid decreases diminutively with the increase in Reynolds number of the cold fluid except for miscible liquids in counter current flow pattern.

11 Table 3.6 Water-Water System Temperature Cold Fluid (Water) Mass Flow Rate Reynolds Number Nusselt Number Hot Fluid (Water) Temperature Mass Flow Reynolds Number Nusselt Number Shell Heat Transfer Coefficient Tube Overall Capacity Rate Ratio No. of Transfer Units Effectiveness Efficiency K kg/s K kg/s W/m 2 K W/m 2 K W/m 2 K % % % t 1 t 2 m c N Re N Nu T 1 T 2 m h N Re N Nu h o h i U o R c NTU ε η S η T Shell Tube 30

12 Table 3.7 Kerosene-Water (Immiscible) System Temperature Cold Fluid (Kerosene) Hot fluid (Water) Heat transfer coefficient Mass Mass Reynolds Nusselt Reynolds Nusselt Shell Tube Flow Temperature Flow Overall Number Number Number Number rate rate Capacity Rate Ratio No. of Transfer Units Effectiveness Efficiency K kg/s K kg/s W/m 2 K W/m 2 K W/m 2 K % % % t 1 t 2 m c N Re N Nu T 1 T 2 m h N Re N Nu h o h i U o R c NTU ε η S η T 9% Kerosene Water System % Kerosene Water System % Kerosene Water System % Kerosene Water System Shell Tube 31

13 Table 3.8 Toluene-Water (Immiscible) System Temperature Cold Fluid (Toluene) Hot Fluid (Water) Heat Transfer Coefficient Mass Mass Reynolds Nusselt Reynolds Nusselt Shell Tube Flow Temperature Flow Overall Number Number Number Number rate rate Capacity Rate Ratio No. of Transfer Units Effectiveness Efficiency K kg/s K kg/s W/m 2 K W/m 2 K W/m 2 K % % % t 1 t 2 m c N Re N Nu T 1 T 2 m h N Re N Nu h o h i U o R c NTU ε η S η T 9% Toluene Water System % Toluene Water System % Toluene Water System % Toluene Water System Shell Tube 32

14 Table 3.9 Acetic acid-water (Miscible) System Temperature Cold Fluid (Acetic Acid ) Hot Fluid (Water) Heat Transfer Coefficient Mass Mass Reynolds Nusselt Reynolds Nusselt Shell Tube Flow Temperature Flow Overall Number Number number Number rate Rate Capacity Rate Ratio No. of Transfer Units Effectiveness Efficiency K kg/s K kg/s W/m 2 K W/m 2 K W/m 2 K % % % t 1 t 2 m c N Re N Nu T 1 T 2 m h N Re N Nu h o h i U o R c NTU ε η S η T 9%Acetic Acid Water System %Acetic Acid Water System %Acetic Acid Water System %Acetic Acid Water System Shell Tube 33

15 Table 3.10 Ethylene Glycol-Water (Miscible) System Cold Fluid ( Ethylene Glycol) Hot Fluid (Water) Heat Transfer Coefficient Mass Reynolds Nusselt Mass Reynolds Nusselt Shell Tube Flow Temperature Overall Number Number Flow Number Number Rate Temperature Capacity Rate Ratio No. of Transfer Units Effectiveness Efficiency K kg/sec K kg/sec W/m 2 K W/m 2 K W/m 2 K % % % t 1 t 2 m c N Re N Nu T 1 T 2 m h N Re N Nu h o h i U o R c NTU ε η S η T 9% Ethylene Glycol Water System % Ethylene Glycol Water System % Ethylene Glycol Water System % Ethylene Glycol Water System Shell Tube 34

16 35 The graphs plotted aganist Efficiency and Effectiveness versus flow rate of the cold fluid implies that increase in the flow rate decreases the shell side efficiency and increases the tube side efficiency of the heat exchanger which are illustrated from Figures 3.19 to 3.23 and Figures 3.14 to 3.18 respectively. The efficiency of shell side is the ratio of the quantity of heat removed from a fluid to the maximum heat removed. If flowrate increases the heat removal decreases so shell side efficiency decreases, where as the tube side flowrate is kept constant the heat removal increases so the tube side efficiency increases. If the flow rate of cold fluid increases the capacity rate ratio increases which in turn decreases the NTU, so Effectiveness decreaes with increaing Reynolds number (Cold) for Parallel flow heat echanger. In counter current flow heat exchanger, for immiscible systems, the effectiveness decreases with increasing the mass flow rate of cold fluid. For miscible systems, if the mass flow rate of cold fluid increases, the capacity rate ratio decreases which in turn increases the NTU, so effectiveness increases with increasing Reynolds number (Cold) Effect of Variation of Composition From the Tables 3.1 to 3.10, the overall heat transfer coefficient for 9 % to 25% of Water-Water System, Kerosene-Water System, Toluene-Water System, Acetic acid-water System and Ethylene Glycol-Water System on volume basis almost increases with increasing in the mole fraction of system used for constant flow rate of the hot fluid. This is because decrease in the concentration of water increases the heat capacity of the tube side fluid and hence the heat transferred. Increase in composition decreases the tube outlet temperature because decrease in the concentration increases the specific heat value.

17 36 In the Figures 3.25 to 3.28 and Figures 3.50 to 3.53 plotted exchanger effectiveness versus flow rate of the cold fluid for 9% to 25% on volume basis for various miscible and immiscible systems, at constant flow rate of the hot fluid, show that the effectiveness increases with increase in the mole fraction of systems. From the Figures 3.20 to 3.23 and Figures 3.45 to 3.48, we observed that the efficiency increases with increasing in the mole fraction of miscible and immiscible systems for both parallel and counter current flow heat exchanger Nusselts Number (Cold) Vs Parallel Flow The Reynolds number (Cold) for different composition (9%, 10%, 20% and 25% on volume basis) of immiscible (Kerosene-Water, Toluene- Water) and miscible systems (Acetic acid-water, Ethylene Glycol-Water) is plotted against Nusselt number (Cold) to find the effect of varying flow rate and composition of cold side fluid on heat transfer coefficient of cold side fluid under parallel flow condition in the Figures 3.4 to 3.8. Figure 3.4 Nusselt Number (Cold) Vs Water-Water System Figure 3.5 Nusselt Number (Cold) Vs Kerosene-Water System

18 37 Figure 3.6 Nusselt Number (Cold) Vs Toluene-Water System Figure 3.7 Nusselt Number (Cold) Vs Ethylene Glycol-Water System Figure 3.8 Nusselt Number (Cold) Vs Acetic acid-water System Nusselts Number (Hot) Vs Parallel Flow The Reynolds number (Cold) for different composition (9%, 10%, 20% and 25% on volume basis) of immiscible (Kerosene-Water, Toluene- Water) and miscible system (Acetic acid-water, Ethylene Glycol-Water) is plotted against Nusselt number (Hot) to indicate the effect of varying flow rate and composition of cold side on heat transfer coefficient of hot side under parallel flow condition in the Figures 3.9 to 3.13.

19 38 Figure 3.9 Nusselt Number (Hot) Vs Water-Water System Figure 3.10 Nusselt Number (Hot) Vs Kerosene-Water System Figure 3.11 Nusselt Number (Hot) Vs Toluene-Water System Figure 3.12 Nusselt Number (Hot) Vs Ethylene Glycol-Water System Figure 3.13 Nusselt Number (Hot) Vs Acetic acid-water System Hot Efficiency Vs Parallel Flow The Reynolds number (Cold) for different composition (9%, 10%, 20% and 25% on volume basis) of immiscible (Kerosene-Water, Toluene-

20 39 Water) and miscible system (Acetic acid-water, Ethylene Glycol-Water) is plotted against efficiency of hot side fluid to find the effect of varying flow rate and composition of cold side on efficiency of hot side under parallel flow condition in the Figures 3.14 to Figure 3.14 Hot Efficiency Vs Water-Water System Figure 3.15 Hot Efficiency Vs Kerosene-Water System Figure 3.16 Hot Efficiency Vs Toluene-Water System Figure 3.17 Hot Efficiency Vs Ethylene Glycol-Water System Figure 3.18 Hot Efficiency Vs Acetic acid-water System

21 Cold Efficiency Vs Parallel Flow The Reynolds number (Cold) for different composition (9%, 10%, 20% and 25% on volume basis) of immiscible (Kerosene-Water, Toluene- Water) and miscible system (Acetic acid-water, Ethylene Glycol-Water) is plotted against efficiency of cold side fluid to find the effect of varying flow rate and composition of hot side on efficiency of cold side under parallel flow condition in the Figures 3.19 to Figure 3.19 Cold Efficiency Vs Water-Water System Figure 3.20 Cold Efficiency Vs Kerosene-Water System Figure 3.21 Cold Efficiency Vs Toluene-Water System Figure 3.22 Cold Efficiency Vs Ethylene Glycol-Water System

22 41 Figure 3.23 Cold Efficiency Vs Acetic acid -Water System Effectiveness Vs Parallel Flow The variation in Reynolds number (Cold) for different composition (9%, 10%, 20% and 25% on volume basis) of immiscible (Kerosene-Water, Toluene-Water) and miscible system (Acetic acid-water, Ethylene Glycol- Water) on effectiveness to indicate the effect of varying flow rate and composition of cold side on effectiveness under parallel flow condition is shown in Figures 3.24 to Figure 3.24 Effectiveness Vs Reynolds Number (Cold) for Water- Water System Figure 3.25 Effectiveness Vs Reynolds Number (Cold) for Kerosene- Water System

23 42 Figure 3.26 Effectiveness Vs Reynolds Number (Cold) for Toluene- Water System Figure 3.27 Effectiveness Vs Reynolds Number (Cold) for Ethylene Glycol-Water System Figure 3.28 Effectiveness Vs Reynolds Number (Cold) for Acetic acid-water System Nusselt Number (Cold) Vs Counter Current Flow The Reynolds number (Cold) for Water-Water system and different composition (9%, 10%, 20% and 25% on volume basis) of immiscible (Kerosene-Water, Toluene-Water) and miscible system (Acetic acid-water, Ethylene Glycol-Water) are plotted against Nusselt number of cold side to find the effect of varying flow rate and composition of cold side on heat transfer coefficient of cold side under counter current flow condition in the Figures 3.29 to 3.33.

24 43 Figure 3.29 Nusselt Number (Cold) Vs Water-Water System Figure 3.30 Nusselt Number (Cold) Vs Kerosene-Water System Figure 3.31 Nusselt Number (Cold) Vs Toluene-Water System Figure 3.32 Nusselt Number (Cold) Vs Ethylene Glycol-Water System Figure 3.33 Nusselt Number (Cold) Vs Acetic acid-water System

25 Nusselt Number (Hot) Vs Counter Current Flow The effect of mass flow rate of cold fluid in terms of Reynolds number (Cold) for different composition (9%, 10%, 20% and 25% on volume basis) of immiscible (Kerosene-Water, Toluene-Water) and miscible system (Acetic acid-water, Ethylene Glycol-Water) is plotted against Nusselt number (Hot) to indicate the effect of varying flow rate and composition of cold side on heat transfer coefficient of hot side under counter current flow condition in the Figures 3.34 to Figure 3.34 Nusselt Number (Hot) Vs Water-Water System Figure 3.35 Nusselt Number (Hot) Vs Kerosene-Water System Figure 3.36 Nusselt Number (Hot) Vs Toluene-Water System Figure 3.37 Nusselt Number (Hot) Vs Ethylene Glycol-Water System

26 45 Figure 3.38 Nusselt Number (Hot) Vs Acetic acid-water System Hot Efficiency Vs Counter Current Flow The Reynolds number (Cold) for different composition (9%, 10%, 20% and 25% on volume basis) of immiscible (Kerosene-Water, Toluene- Water) and miscible system (Acetic acid-water, Ethylene Glycol-Water) is plotted against efficiency of hot side to find the effect of varying flow rate and composition of cold side on efficiency of hot side under counter current flow condition in the Figures 3.39 to Figure 3.39 Hot Efficiency Vs Water-Water System Figure 3.40 Hot Efficiency Vs Kerosene-Water System

27 46 Figure 3.41 Hot Efficiency Vs Toluene-Water System Figure 3.42 Hot Efficiency Vs Ethylene Glycol-Water System Figure 3.43 Hot Efficiency Vs Acetic acid-water System Cold Efficiency Vs Counter Current Flow The Reynolds number (Cold) for different composition (9%, 10%, 20% and 25% on volume basis) of immiscible (Kerosene-Water, Toluene- Water) and miscible system (Acetic acid-water, Ethylene Glycol-Water) is plotted against efficiency of cold side to find the effect of varying flow rate and composition of hot side on efficiency of cold side under counter current flow condition in the Figures 3.44 to 3.48.

28 47 Figure 3.44Cold Efficiency Vs Water-Water System Figure 3.45 Cold Efficiency Vs Kerosene-Water System Figure 3.46Cold Efficiency Vs Toluene-Water System Figure 3.47 Cold Efficiency Vs Ethylene Glycol-Water System Figure 3.48 Cold Efficiency Vs Acetic acid-water System

29 Effectiveness Vs Counter Current Flow The variation in Reynolds number (Cold) for different composition (9%, 10%, 20% and 25% on volume basis) of immiscible (Kerosene-Water, Toluene-Water) and miscible system (Acetic acid-water, Ethylene Glycol- Water) on effectiveness to indicate the effect of varying flow rate and composition of cold side on effectiveness under counter current flow condition is shown in Figures 3.49 to Figure 3.49Effectiveness Vs Reynolds Number (Cold) for Water- Water System Figure 3.50 Effectiveness Vs Reynolds Number (Cold) for Kerosene- Water System Figure 3.51Effectiveness Vs Reynolds Number (Cold) for Toluene- Water System Figure 3.52 Effectiveness Vs Reynolds Number (Cold) for Ethylene Glycol-Water System

30 49 Figure 3.53 Effectiveness Vs Reynolds Number (Cold) for Acetic acid-water System 3.6 SIMULATION USING ARTIFICIAL NEURAL NETWORK Introduction The General Regression Neural Network (GRNN) is a universal approximator for smooth functions, so it should be able to solve any smooth function-approximation problem given enough data. General regression neural networks perform regression where the target variable is continuous. The description of GRNN is given below Input Layer A layer of neurons that receives information from external sources, and passes this information to the network for processing. These may be either sensory inputs or signals from other systems outside the one being modeled Hidden Layer A layer of neurons that receives information from the input layer and processes them in a hidden way. It has no direct connections to the

31 50 outside world (inputs or outputs). All connections from the hidden layer are to other layers within the system Output Layer out of the system. A layer of neurons that receives processed and sends output signals Bias Acts on a neuron like offset. The function of the bias is to provide a threshold for the activation of neurons. The bias input is connected to each of the hidden and output neurons in a network Input-Output Mapping The input/output mapping of a network is established according to the weights and activation functions of their neurons in input, hidden and output layers. The number of input neurons corresponds to the number of input variables in the network, and the number of output neurons is the same as the number of desired output variables. The number on neurons in the hidden layer(s) depends on the particular neural network application Pattern Classifier The general multilayer neural network classifier structure for the simulation is shown in Figure The actual ANN classifier structure used for this problem is shown in Figure The ANN has 7 neurons in the input layer and 8 neurons in the output layer. The 8 neurons in the output layer can handle 256 outputs and will be sufficient for this work. The number of

32 51 neurons in the hidden layer is 29. So the ANN structure boils down to 7:29:8. The ANN is adaptively trained to update the weights and the bias by gradient descent method by mean square error performance. Figure 3.54 General Multilayer Neural Network Classifier The classifier structure for this work and the training pattern for 1000 epochs are shown in Figure In classifier the first block indicates the input layer comprising of 7 neurons, the central block indicates the hidden layer comprising of 29 neurons and the last block indicates the output layer comprising of 8 neurons respectively. The blocks between the input layer and the middle layer indicate the weight factor (1W {1, 1} associated with input node and bias input (b{1}acts as a neuron like offset. The blocks between the input layer and output layer indicates the weight factor (LW {2, 1}) associated with the hidden layer and Bias input (b{2}) acts as a neuron like offset.

33 52 Figure 3.55 Pattern Classifier for Shell and Tube Heat Exchanger 3.7 COMPARISON OF SIMULATION OUTPUT WITH THE EXPERIMENTAL DATA FOR SHELL AND TUBE HEAT EXCHANGER The simulation is carried out using ANN to predict Nusselts number of the cold fluid (N Nu ), effectiveness (ε), cold side efficiency (η c ) and hot side efficiency (η h ) for the Shell and Tube Heat Exchanger. A comparison is made to show the performance characteristics of heat exchanger of the simulation output using ANN with the experimental data which is given in Tables 3.11 and The table indicates that the simulation results are very well agreed with the experimental data.