A Novel Student Psychological Based Optimization Algorithm with Multiple Strategies

To tackle the low convergence accuracy and vulner-ability to local optima problems of the SPBO algorithm, a novel student psychology based optimization (NSPBO) algorithm is introduced, integrating different enhancement strategies. Firstly, a wavelet mutation strategy is implemented to improve the guiding function of the optimal individual in the algorithm's search process, thus enhancing the algorithm's local optimization capability. Secondly, a weighted center and Cauchy mutation mechanism are integrated into the randomly enhanced student individuals to maintain diversity among student individuals and avoid the algorithm from getting stuck in local optima prematurely. Finally, a region-based learning search strategy is applied to the entire population to boost the algorithm's search accuracy and expedite the search process. Experimental tests were conducted on eighteen standard test functions to compare the performance of the proposed algorithm with other intelligent optimization algorithms. The findings indicate that the proposed algorithm exhibits exceptional optimization performance, excelling in both accuracy and convergence speed.