摘要
将齿形系数与齿数之间的关系映射为径向基神经网络,并且利用遗传算法优化径向基神经网络隐层与输出层的权值,最后利用遗传算法合理选择交叉概率与变异概率对齿轮进行优化设计。利用本算法对二级斜齿圆柱齿轮减速器进行设计,结果表明本算法的性能优于原遗传算法且设计效率高,在机械优化设计中具有广泛的应用前景。
The relationship between The tooth shape coefficient and the number of teeth is mappted into radial basis function neural network, hidden layer and output layer weights has been optimized using genetic algorithm. Finally, ge- netic algorithm is used in reasonable selection of crossover probability and mutation probability for gear optimization de- sign. This algorithm is used to design two grade helical cylindrical gear reducer, the result shows that the algorithm per- formanee is better than the original genetic algorithm.This algorithm has wide application prospect in mechanical opti- mization design due to its high efficiency.
出处
《科技通报》
北大核心
2012年第12期96-97,175,共3页
Bulletin of Science and Technology
关键词
径向基神经网络
遗传算法
交叉变异
齿形系数
radial basis function neural network
genetic algorithm
Crossover and mutation
Tooth form factor
