Description

Global optimization helps to find the optimal value of a problem. The optimization process involves exploration and exploitation in order to search the solution space. High exploration capability is required to handle the multi-modal search spaces and high exploitation helps in accelerating the convergence speed of agents towards the global optima. Gravitational Search Algorithm (GSA) is a very promising heuristic algorithm that is inspired by physics and it is based on the principle of mass attraction. It has been widely used to find the global optima for the real world problems. GSA has the advantages of inherent powerful exploration capability and higher diversification power that makes it less prone to local minima falling. Furthermore, GSA uses an elitism strategy to store the best candidate solutions. Hence, it prevents optimal agents from local minima interference. However, GSA faces the issues of premature convergence and entrapment in local minima while solving complex problems. The issues of standard Gravitational Search Algorithm have been addressed by the proposed algorithms, namely CPSOGSA (Constriction Coefficient based Particle Swarm Optimization and Gravitational Search Algorithm) and LCGSA (Levy Flight and Chaos Theory-based Gravitational Search Algorithm). CPSOGSA helps to resolve the premature convergence issue of standard GSA. Similarly, LCGSA increases the exploitation rate and resolve local minima entrapment problem of the standard GSA.