一种引入复合形法的改进引力搜索算法

针对引力搜索算法求解复杂问题时搜索精度不高、易出现早熟收敛问题,提出一种引入复合形法的改进引力搜索算法。该算法在寻优初期利用引力搜索算法进行全局搜索,同时对引力系数进行改进,以提高全局收敛速度;在寻优后期,当算法出现早熟收敛现象时,引入复合形法,利用复合法较强的局部搜索能力,帮助种群快速跳出局部最优解。通过5个标准测试函数验证了改进算法的可行性和有效性。与标准引力搜索算法、基于权值的引力搜索算法、记忆性引力搜索算法相比,该算法具有更高的收敛精度和更快的收敛速度。

Improved Gravitational Search Algorithm with Complex Methods

Aiming at overcoming the low precision and premature convergence issues of the gravitational search algorithm （GSA）,an improved GSA with complex methods is developed.In the early period of the search process,a GSA is used to globally search and the gravitational coefficients of the GSA is changed to improve the global convergence.In the later stage of the search process,the complex methods are applied to avoid the premature convergence.The strong local search ability of the complex methods is used to help the population jump out of the local optimal solution.Through the case study of the five standard functions,the results show that the improved algorithm is feasible and efficient in nonlinear optimization.The improved algorithm has higher convergence precision and faster convergence speed compared with the standard GSA,weighted GSA and memory GSA.