人工神经网络辅助汽车变速箱设计

Artificial Neural Network Aided Automotive-transmission Design

针对汽车变速箱优化设计中所存在的大计算量、多图表问题,提出应用人工神经网络进行映射辨识.根据BP 网络的在特征识别上的优势,以其为工具,通过算法改进,对设计数据进行应用以提高计算速度和编程处理能力,消除数据的“组合爆炸”现象.对具体数据、图表的实际应用表明,人工神经网络方法能够很好地表达原图表数据关系,使数据应用更加灵活、方便,数据的识别精度能够满足计算要求.

Today,the automotive industry is faced with more pressure to realize higher quality products,lower production costs and shorter lead times.The key to success is the effective design.In order to accelerate the calculating career and ability of designing program,the artificial neural network(ANN)was used to map the chart during automotive-transmission optimum design.This paper introduces the theory of back propagation neural network and analyzes it's short coming.An improved back propagation neural network was used in fea- ture recognition.The application of real data and chart shows that the ANN can express these data correctly, the identification can suffice calculating requirement with the good precision.