多策略融合改进的白鲸优化算法

Improved beluga whale optimization algorithm with multi-strategy integration

针对白鲸优化算法(BWO)存在的寻优性能易受初值影响,迭代后期易陷入局部最优等问题,提出了一种多策略融合改进的白鲸优化算法(PSBWO).该算法首先引入PWLCM映射对初值进行混沌初始化,以提升种群的多样性,增强全局搜索能力.然后利用融合高斯-柯西变异构建了自适应共栖生物搜寻算子,并在每一次迭代结束后对种群进行动态扰动,以此提高算法跳出局部极值的概率.为了验证PSBWO算法的有效性,采用10个基准测试函数对其进行了测试,结果表明PSBWO算法的性能优于原白鲸算法,且具有良好的鲁棒性.因此,PSBWO算法在解决实际工程问题中具有良好的应用潜力.

Aiming at the problems that the optimization performance of Beluga whale optimization algorithm(BWO)is easy to be affected by the initial value and fall into the local optimum in the late iteration,a multi-strategy fusion improved Beluga whale optimization algorithm(PSBWO)is proposed.Firstly,PWLCM mapping is introduced to initialize the initial chaotic values to improve the diversity of the population and enhance the global search capability.Secondly,by fusing Gauss-Cauchy variation,an adaptive commensal search operator is constructed to dynamically perturb the population after each iteration to improve the probability of the algorithm jumping out of the local extreme value.In order to verify the effectiveness of PSBWO algorithm,10 benchmark test functions are used to test it,and the results show that the performance of PSBWO algorithm is better than that of the original Beluga whale algorithm,and it has good robustness.Therefore,PSBWO algorithm has good application potential in solving practical engineering problems.