CAJ | 학술논문

本文首先对插铣铣削力进行了理论分析，并基于正交试验方法对铣削力进行了测量试验，然后利用遗传算法对BP神经网络的权值和阈值进行优化，建立了预测铣削力的遗传神经网络模型，最后将经过遗传算法优化的BP网络与未优化的进行对比分析。结果表明，经遗传算法优化后BP网络模型预测误差明显减小，网络的计算精度和收敛速度有了显著提高。
본문수선대삽선선삭력진행료이론분석，병기우정교시험방법대선삭력진행료측량시험，연후이용유전산법대BP신경망락적권치화역치진행우화，건립료예측선삭력적유전신경망락모형，최후장경과유전산법우화적BP망락여미우화적진행대비분석。결과표명，경유전산법우화후BP망락모형예측오차명현감소，망락적계산정도화수렴속도유료현저제고。
First, plunge milling force was analyzed theoretically, and the milling force was measured based on the orthogonal experiment method. Then, the weights and thresholds of BP neural network was optimized using genetic algorithm, and the genetic neural network for predicting milling force was established. Finally, network optimized was compared and analyzed with unoptimized network. The results show that the prediction error of BP network optimized reduced significantly, and the network's accuracy and convergence rate has been improved significantly.