An Improved Whale Optimization Algorithm For High Dimensional Functions Optimization And Its Application

Meta-heuristic algorithms don’t depend on function form when solving optimization problems.They have strong adaptability and are widely used in various fields.Whale optimization algorithm(WOA)is a meta-heuristic algorithm based on the social behavior of humpback whales.Compared with other meta-heuristic algorithms,WOA owns better performance.However,when solving high dimensional optimization problems,WOA is easy to fall into local optimal,the convergence speed is slow,and the solution accuracy is low.Aiming at the above shortcomings of WOA,an improved whale optimization algorithm(IWOA)based on the front center swarm and terminal individuals mutation is proposed in this paper.Firstly,a multi-scale adaptive control factor is proposed to enhance the exploitation and exploration capabilities of the algorithm in view of the fact that the linear decline of the original control factor is not conducive to global optimization.Secondly,considering that the initial population individuals generated by random method may lead to slow convergence and local optimization,this paper uses Tent chaotic mapping to generate initial population and increase the global of WOA.In addition,aiming at the exploration stage of the algorithm,an update mode integrating the front center swarm is proposed to increase the global convergence of the algorithm.Finally,the terminal individuals mutation mechanism and opposition-based learning strategy are introduced to enhance the quality and diversity of population individuals.In order to verify the methods proposed in this paper,in the aspect of numerical simulation,23 low dimensional benchmark functions and 13 high dimensional benchmark functions are selected to test the performance of IWOA.In the aspect of engineering analysis,the prediction model of weather temperature is built based on fuzzy neural network.The simulation results show that IWOA achieves excellent performance in the optimization of low and high dimensional test functions,especially in the optimization of high dimensional functions.In terms of weather temperature prediction,IWOA has the best prediction effect.Compared with other algorithms,the convergence accuracy and convergence speed of the proposed methods are improved.The boxplot experiments show that the stability of the proposed methods is improved.The ablation experiment shows that several improvement strategies are mutually reinforcing and cooperative.The Wilcoxon symbolic rank test results show that the proposed algorithm owns significant differences.