A Simplified and Efficient Gravitational Search Algorithm for Unconstrained Optimization Problems

Abstract

Aiming at the shortcomings that the gravitational search algorithm (GSA) is easy to fall into the local optima, this paper proposes a simplified gravitational search algorithm (SGSA). This improved gravitational search algorithm has the characteristics of faster optimization process and better convergence accuracy for solving unconstrained optimization problems. In the search process, SGSA discards the velocity and only performs the particles' position update including the particles acceleration. Ten benchmark functions are used to verify the performance of the SGSA algorithm, and the experimental results show that SGSA is better than the other four approaches with different improvement strategies for most cases.