Optimization of Rate of Reaction Using Nature Inspired Algorithms

Abstract

Chemical reaction involves transformation of reactant into product. Rate of chemical reaction is the measure of how fast these changes are taking place. Some reactions occur very rapidly, others very slowly. For example, ionic reactions are very fast, while those taking place in water treatment plant may last up to few days. Rate of reaction can also be defined as the speed at which a reaction happens. If a chemical reaction has high rate, it shows that molecules combine at a higher rate than the reaction which has a slow rate. The rate of chemical reaction can also depend on different types of molecules that are combining. If there is low concentration of an essential element or compound, the reaction will be slower. Earlier Arrhenius equation was used to calculate the rate of reaction with some parameters. However, to know the exact reaction mechanism, actual reaction needed to be executed, which was not feasible and was time consuming. There are some simulators available for chemical kinetics which can compute the reaction rate though not in an optimal way. This thesis focuses on the optimization of rate of chemical reaction using four different nature inspired optimization algorithms, i.e, Random Algorithm, Genetic Algorithm, Differential Evolution and Particle Swarm Optimization, in order to maximize the rate of chemical reaction. In addition, these algorithms aim to get the best possible reaction rate along with the input tuning parameters. Test results show that different algorithms perform significantly better for different reactions and have different convergence rate. In this thesis, a simulator is used which is a object-oriented software tool named Cantera to calculate the rate of reaction which uses modified Arrhenius equation.