1 简介针对引力搜索算法存在的易早熟收敛、易陷入局部最优、搜索精度有待提高等缺陷提出一种混合方法优化的自适应引力搜索算法。首先 利 用 序列初始化种群增强 算法全局搜索能力其次引入 贴进度计算种群成熟度判断种群是否早熟然后引入 混沌对种群作混沌搜索变异已陷入局部最优的粒子位置最后基于早熟收敛判断因子改进引力系数并为粒子位置公式添加收缩因子促使种群加快脱离局部最优。对个不同类型的基准测试函数做仿真实验结果表明新算法能有效改善种群的早熟问题具备更好的寻优性能。2 部分代码%_________________________________________________________________________%% GSA chaotic gravitational constant %clear allP_no30;Max_iteration500;Run_no2;ElitistCheck1;All_Convergence_curveszeros(2,Max_iteration);chValue20;F_index 1;for Algorithm_num1:11for i1:Run_noif Algorithm_num1cg_curveGSA(F_index,P_no,Max_iteration,ElitistCheck,Algorithm_num,chValue);endif Algorithm_num2cg_curveGSA(F_index,P_no,Max_iteration,ElitistCheck,Algorithm_num,chValue);endif Algorithm_num3cg_curveGSA(F_index,P_no,Max_iteration,ElitistCheck,Algorithm_num,chValue);endif Algorithm_num4cg_curveGSA(F_index,P_no,Max_iteration,ElitistCheck,Algorithm_num,chValue);endif Algorithm_num5cg_curveGSA(F_index,P_no,Max_iteration,ElitistCheck,Algorithm_num,chValue);endif Algorithm_num6cg_curveGSA(F_index,P_no,Max_iteration,ElitistCheck,Algorithm_num,chValue);endif Algorithm_num7cg_curveGSA(F_index,P_no,Max_iteration,ElitistCheck,Algorithm_num,chValue);endif Algorithm_num8cg_curveGSA(F_index,P_no,Max_iteration,ElitistCheck,Algorithm_num,chValue);endif Algorithm_num9cg_curveGSA(F_index,P_no,Max_iteration,ElitistCheck,Algorithm_num,chValue);endif Algorithm_num10cg_curveGSA(F_index,P_no,Max_iteration,ElitistCheck,Algorithm_num,chValue);endif Algorithm_num11cg_curveGSA(F_index,P_no,Max_iteration,ElitistCheck,Algorithm_num,chValue);endtemp(i,:)cg_curve;endAll_Convergence_curves(Algorithm_num,:)mean(temp);endfigurefor k 1:size(All_Convergence_curves,1)semilogy(All_Convergence_curves(k,:))hold onendlegend(GSA, CGSA1, CGSA2, CGSA3, CGSA4, CGSA5, CGSA6, CGSA7, CGSA8, CGSA9, CGSA10)save resuls3 仿真结果4 参考文献[1]龚安, 吕倩, 胡长军, 康忠健, 李华昱. (2015). 基于混沌万有引力搜索算法的svm参数优化及应用. 计算机科学, 42(4), 4.​