《遗传算法多目标函数优化》由会员分享，可在线阅读，更多相关《遗传算法多目标函数优化（6页珍藏版）》请在装配图网上搜索。
1、多目标遗传算法优化铣削正交试验结果序号vc(m/min)fz(mm/z)ae(mm)ap(mm)F(N)Ra(m)1400.020.25446.30.1742400.040.5659.70.2113400.060.758106.50.2884400.081.010123.00.3485600.020.25899.30.2526600.040.510134.00.3027600.061.04139.70.2388600.080.756130.40.2769800.020.7510255.80.29810800.041.08255.80.28811800.060.256110.40.2591280
2、0.080.54140.80.258131000.021.06314.20.240141000.040.754233.40.239151000.060.510278.60.330161000.080.258144.70.328说明：1.建立切削力和外表粗糙度模型如：1此模型你们来拟合上面有实验数据，剩下的两个方程已经是我帮你们拟合好的了 2Ra=10-0.92146vc0.14365fz0.16065ae0.047691ap0.384573变量约束围：公式1和2值越小越好，公式3值越大越好。=3.14 D=82.请将多目标优化操作过程录像同时考虑三个方程，优化出最优的自变量数值，方便我后续进展
3、修改；将能保存的所有图片与源文件发给我；将最优解多组发给我，类似于下列图黄色局部为到达的要求遗传算法的结果:vcfzaeapFRaQ69.49510.02010.59362.030144.45690.1507-133.92377.76040.080.80047.7189298.80560.3241-3058.3469.84320.04290.79237.316175.06530.2827-1383.0170.2130.02820.69457.6443117.280.2673-837.41369.49510.02010.59362.030144.45690.1507-133.923程序如下:cl
4、ear; clc;% 遗传算法直接求解多目标优化D=8;% Function handle to the fitness functionF=(X)10(3.19)*(X(1).(-0.0836).*(X(2).0.825).*(X(3).0.564).*(X(4).0.454);Ra=(X)10(-0.92146)*(X(1).0.14365).*(X(2).0.16065).*(X(3).0.047691).*(X(4).0.38457);Q=(X)-1000*2*X(1).*X(2).*X(3).*X(4)/(pi*D); nvars = 4; % Number of decision
5、variableslb = 40, 0.02,0.25, 2; % Lower boundub = 100,0.08, 1.0,10; % Upper boundA = ; b = ; % No linear inequality constraintsAeq = ; beq = ; % No linear equality constraints% 遗传算法设定约束options = gaoptimset(ParetoFraction,0.3,PopulationSize,200,Generations,300,StallGenLimit,200,TolFun,1e-100,PlotFcns
6、,gaplotpareto);% the data of Pareto1FUN=(X)F(X);Ra(X);x,fval = gamultiobj(FUN,nvars,A,b,Aeq,beq,lb,ub,options);% 开启一个新的图形figure;% 画出Pareto1plot(fval(:,1),fval(:,2),mp);% 画出网格grid on;% 设定横坐标xlabel(F);% 设定纵坐标ylabel(Ra);% 设定题目title(Pareto front 1);% the data of Pareto2FUN=(X)F(X);Q(X);x,fval = gamultio
7、bj(FUN,nvars,A,b,Aeq,beq,lb,ub,options);% 开启一个新的图形figure;% 画出Pareto2plot(fval(:,1),fval(:,2),mp);% 画出网格grid on;% 设定横坐标xlabel(F);% 设定纵坐标ylabel(Q);% 设定题目title(Pareto front 2);% the data of Pareto3FUN=(X)Ra(X);Q(X);x,fval = gamultiobj(FUN,nvars,A,b,Aeq,beq,lb,ub,options);% 开启一个新的图形figure;% 画出Pareto3plo
8、t(fval(:,1),fval(:,2),mp);% 画出网格grid on;% 设定横坐标xlabel(Ra);% 设定纵坐标ylabel(Q);% 设定题目title(Pareto front 3);% the data of Pareto (F,Ra,Q)FUN=(X)F(X);Ra(X);Q(X);x,fval = gamultiobj(FUN,nvars,A,b,Aeq,beq,lb,ub,options);% 开启一个新的图形figure;% 画出Pareto4plot3(fval(:,1),fval(:,2),fval(:,3),m.);% 画出网格grid on;% 设定横坐
9、标xlabel(F);% 设定纵坐标ylabel(Ra);% 设定竖坐标zlabel(Q);% 设定题目title(Pareto front 4);% 重新设定遗传算法options = gaoptimset(ParetoFraction,0.3,PopulationSize,5,Generations,1000,StallGenLimit,200,TolFun,1e-100,PlotFcns,gaplotpareto);% 遗传算法求解x,fval = gamultiobj(FUN,nvars,A,b,Aeq,beq,lb,ub,options);% 显示结果format shortgxfval=x,fval% 保存结果到Excelxfval=round(xfval*10000)/10000;xlswrite(data.xlsx,xfval,sheet1,A1)6 / 6
遗传算法多目标函数优化
