Optimization of electrochemical reactors using genetic algorithms

Electrochemical reactor optimization using Genetic Algorithms (GAs) has been attempted in the present work. The objectives have been focused to determine the (i) optimal design parameters that maximize the yield of the product under specified conditions and (ii) optimal current density that minimizes the operating cost of the reactor. As a vehicle to do so, a reaction mechanism is considered in which the reactant is electrochemically reduced to a desired product and further reduced to an undesired product. Both, batch and continuous reactors have been considered for performance evaluation and simulation has been done at various kinetic parameters. To illustrate the potential utility of genetic search and to justify the use of GAs for this type of optimization problem, we begin our search for optimality with usual algorithms like Exhaustive search, Fibonacci search and Golden section search techniques. The comparative results of these techniques and experimental results show that GAs find optimal reactor cost and product yield, that is also found to agree with the reactors used in industries and in the reported literature. As a result, the need to obtain a good initial guess can be eliminated also with less number of generations to reach optimum level even for a large design problem.