针对将交互式遗传算法应用到服装设计中产生的人的疲劳问题,提出利用神经网络来逼近适应度函数。给出了以GA操作产生的每代最佳个体初步作为神经网络径向基网络函数的中心值并结合相似距离值,利用K-Means求出径向基网络的各参数以逼近适应度函数。在服装设计系统应用中取得了良好的效果。
To the problem of human fatigue in interactive genetic algorithm applied to fashion design, using neural network approach to fitness function of GA is proposed. The best individuals produced of every GA generations are preliminary used as the central values of the radial basis function networks neural network. RBF network parameters are solved by K-Means combined with value of similar distance. The fitness function of GA is approximated by artificial neural network. It has achieved good results in fashion design.