An Improved Reptile Search Algorithm with Ghost Opposition-Based Learning for Global Optimization Problems

Abstract In 2021, a meta-heuristic algorithm, Reptile Search Algorithm (RSA), was proposed. RSA mainly simulates the cooperative predatory behavior of crocodiles. Although has fast convergence speed, due to influence crocodile predation mechanism, if algorithm falls into local optimum in early stage, will probably be unable jump out optimum, resulting poor comprehensive performance. Because shortcomings RSA, introducing escape operator can effectively improve crocodiles' ability explore space and generate new crocodiles replace Benefiting from adding restart strategy, when optimal solution is no longer updated, algorithm's improved by randomly initializing crocodile. Then joining Ghost opposition-based learning balance IRSA's exploitation exploration, Improved with Opposition-based Learning for Global Optimization Problem (IRSA) To verify performance IRSA, we used nine famous optimization algorithms compare IRSA twenty-three standard benchmark functions CEC2020 test functions. The experiments show that good robustness, solve six classical engineering problems, thus proving its effectiveness solving practical problems.