Synthesis of Optimum Controllable Heat Exchanger Networks using Genetic Algorithms

The last few decades have seen considerable research into the synthesis of optimal heat exchanger networks (HENs). Process integration has paved the way for designs that are optimal from the point of view of energy recovery. Subsequent optimization of net exchanger area in such designs results in a network that is optimal from the point of view of utility consumption and exchanger area. Of late, attempts have been made to incorporate controllability considerations into the synthesis of optimal HENs. The modern HEN synthesis problem can therefore be stated as the simultaneous optimization of a HEN design from the point of view of Energy, Area, and Control. Structural or static controllability has been considered in the design optimization of HENs in the past. The effect of disturbances in case of operating HENs, however, remains to be considered. This is one of the important motivations behind this project. The final aim is to achieve simultaneous optimization of the HEN design over all the aspects stated above. A novel search procedure, Genetic Algorithms (GAs), is proposed for the synthesis and optimal design of HENs. The search domain being huge and the behavior of the objective function being rather noisy, the search procedure used should be such that it is not affected by the presence of a large number of local optima. The main reason for using this search method is its relative immunity to the nature of the objective function. A basic GA has been developed for analysis. Various modifications in this basic GA have been made to suit the present problem. Details of the implementation of various design criteria for the search are discussed along with the results from the preliminary application procedure. Methods for further improving the search procedure are suggested.