A robust and efficient genetic algorithm for solving a chemical reactor problem: theory, application and convergence analysis

Abstract

Solution of the chemical reactor problem, as an optimal control problem, using continuous genetic algorithms (CGAs) is presented in this paper. The proposed approach overcomes the drawbacks of the traditional approaches in terms of lack of efficiency, lack of accuracy and lack of robustness. The solution is based on the value of the performance index and the final system state constraints. Simulation results show clearly that the new technique outperforms the existing direct and indirect methods. Based on the convergence analysis, the solution of the optimal control problem is achieved without any limitation on the nature of the problem and regardless of the CGA tuning parameters.