Improvement of Grasshopper Optimization Algorithm Based on Chaotic Mapping and Levy Flight

A recently developed population-intelligent optimization method, the grasshopper optimization algorithm (GOA) has been effectively used in numerous real-world applications. However, this algorithm has flaws like poor convergence speed and propensity to enter local optima. A grasshopper optimization algorithm based on chaos initialization and Levy flying (CLGOA) was thus presented to address these problems with grasshopper optimization methods and to enhance their performance. First, the population is initialized using Sine chaotic mapping, which makes it more uniform; Second, adjust the position-determining parameter c and use it as the weight for the ideal location to balance the algorithm's capacities for development and exploration; In order to help the algorithm exit the local region and locate the global optimal value throughout the optimization phase, the Levy flight mechanism is introduced last. By selecting 6 test functions (divided into unimodal and multimodal functions) to evaluate the performance of the algorithm, and comparing the average and variance of 30 consecutive experimental results with other algorithms, it is demonstrated that CLGOA has strong optimization performance. The suggested algorithm CLGOA's excellent optimization capabilities were demonstrated, strengthening the algorithm's robustness and convergence capabilities.