Constraint handling strategies in Genetic Algorithms application to optimal batch plant design

Abstract

Optimal batch plant design is a recurrent issue in Process Engineering, which can be formulated as a Mixed Integer Non-Linear Programming (MINLP) optimisation problem involving specific constraints, which can be, typically, the respect of a time horizon for the synthesis of various products. Genetic Algorithms constitute a common option for the solution of these problems, but their basic operating mode is not always wellsuited to any kind of constraint treatment: if those cannot be integrated in variable encoding or accounted for through adapted genetic operators, their handling turns to be a thorny issue. The point of this study is thus to test a few constraint handling techniques on a mid-size example in order to determine which one is the best fitted, in the framework of one particular problem formulation. The investigated methods are the elimination of infeasible individuals, the use of a penalty term added in the minimized criterion, the relaxation of the discrete variables upper bounds, dominancebased tournaments and, finally, a multiobjective strategy. The numerical computations, analysed in terms of result quality and of computational time, show the superiority of elimination technique for the former criterion only when the latter one does not become a bottleneck. Besides, when the problem complexity makes the random location of feasible space too difficult, a single tournament technique proves to be the most efficient one.

A robust learning based evolutionary approach for thermal-economic optimization of compact heat exchangers

“…In Eqs. (28) and (29), r, a and b are the number of automaton actions, the reward parameter and the penalty parameter, respectively. More details on the different varieties of LA are presented by Thathachar and Sastry [40].…”

“…The static penalty function method employed in all previous works has the drawback of depending on the proper selection of the penalty parameters [28][29][30][31]. Even worse, the penalty parameters are problem-dependent and have to be defined with a trial-and-error process that could be time-consuming and error-prone.…”

Dimensional and Layout Optimization Design of Multistage Gear Drives Using Genetic Algorithms

“…Many researchers in the field of computer algorithms have put forward numerous methods to conduct this transformation and construct various types of fitness functions [17][18][19]. However, when facing up a specific problem, some of these methods may work well, while others may not [20]. is fact gives us an eagerness to figure out which one performs the best in our problem.…”

Performance Analysis of Optimization Methods in PSE Applications

“…the variable coding is carried out with the so-called weighting box. This coding procedure is classically used for continuous variables such as presented in previous studies (Ponsich et al, 2006a). The variable variation range is discretized according to some desired accuracy and a technique similar to binary coding is used to represent the resulting reduced variables.…”

“…Finally, the minimum ratio B i /T is L represents the productivity of the plant Prod i for each product i. The complete model and equations will not be presented here, but the reader should report to Ponsich et al (2006a) to get the detailed formulation.…”