An Improved Version of the SIPO Algorithm with Fast Convergence Speed

According to the utilization of evolutionary algorithms such as the Inclined Planes System Optimization (IPO) algorithm in various fields of engineering science, attempts to improve the performance of these algorithms have always been considered. The standard version of IPO algorithm is based on the dynamic movement of objects on inclined planes without friction. There is also a simplified version of this algorithm to reduce the complexity of equations called Simplified Inclined Planes System Optimization (SIPO). The efficiency of the standard version of IPO and the simplified IPO version (SIPO) is proven in solving many optimization problems. However, due to the strong exploitation, this algorithm suffers from weak exploration. Hence, this leads to low convergence rates and in real applications can get stuck in the local optimum. In this paper, a uniform mutation operator is applied to the SIPO algorithm to make more effective exploration of the latent information contained in the population of solutions. The measurements of the proposed method is first taken on 10 known standard benchmark functions. Then, the performance is compared to the results obtained from the standard IPO algorithm and the SIPO version. Also, the proposed method is utilized in training Adaptive Neuro Fuzzy Inference System (ANFIS) classifier and the results obtained from using IPO, SIPO, Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) algorithms in the training of this classifier are compared. The results indicate that the proposed algorithm, in both the experiments of standard benchmark functions and in the training of ANFIS classifier, is able to achieve better performance in comparison to the mentioned methods and has a faster convergence speed.