Thermodynamic Optimization of Three-Fluid Cross-Flow Heat Exchanger Using GA and PSO Heuristics

Abstract

Optimization technique for a three-fluid cross-flow plate-fin heat exchanger with offset strip fins is developed, considering significant variables subjected to given constraints. The objective of optimization is focused on maximizing the hot fluid effectiveness and minimizing the number of entropy generation units in the three-fluid cross-flow plate-fin heat exchanger. Governing equations for a cross-cocurrent flow arrangement of a three-fluid heat exchanger are solved using finite element method for given boundary conditions. The obtained mean exit fluid temperatures from FEM are used to determine the rate of entropy generation and hot fluid effectiveness of the heat exchanger. Results obtained from the optimization of a reduced model of two-fluid cross-flow heat exchanger is compared with previously published results, thus serving as a validation of the optimization technique. Geometric parameters of the heat exchanger are varied to get the optimum results. The present investigation uses two different heuristics, namely Genetic algorithm (GA) and Particle Swarm Optimization (PSO) to find the optimum design values based on the objectives. Minimization of number of entropy generation units is treated to be a single objective function and optimum solutions are determined using both GA and PSO, both giving a function value of 0.0534786. Similarly, considering maximization of the hot fluid effectiveness, the optimum function value from both techniques is found to be 0.99997095. The results obtained from both the methods are compared and provides almost identical design values, with PSO taking lesser time for execution. Also, considering minimization of number of entropy generation units and maximization of hot fluid effectiveness of the heat exchanger as objective functions, multi-objective optimization is performed using GA alone and the multiple results obtained has been illustrated as a pareto-front.