CAJ | 학술논문

通过利用DEFORM-3D软件对斜齿轮精密锻造过程进行有限元模拟，获得了损伤因子的分布特点。运用Matlab软件建立BP神经网络损伤因子预测模型，分析了不同锻造条件对损伤因子的影响，并对不同锻造条件下的损伤因子进行预测。利用检测样本对训练的神经网络进行验证，结果表明，BP神经网络预测损伤因子与模拟结果吻合较好，预测结果误差较小，预测精度满足实际应用要求。
통과이용DEFORM-3D연건대사치륜정밀단조과정진행유한원모의，획득료손상인자적분포특점。운용Matlab연건건립BP신경망락손상인자예측모형，분석료불동단조조건대손상인자적영향，병대불동단조조건하적손상인자진행예측。이용검측양본대훈련적신경망락진행험증，결과표명，BP신경망락예측손상인자여모의결과문합교호，예측결과오차교소，예측정도만족실제응용요구。
The precision forging finite element model for helical gear was established by DEFORM -3D to obtain the distri-bution characteristics of damage factor .The damage factor prediction model of BP neural network was established by MATLAB to predict influence of different forging conditions on the damage factor .The trained neural network was validated using test sam-ples.The results show that the predicted results agree well with the simulated ones .The differences of prediction results exhibit low value;the predicted precision satisfies the request of industry .The trained BP neural network could be used to analyze the effect of different forging conditions on damage factor .