基于BP神经网络的新零件材料消耗定额预测方法研究

Study on Prediction of Material Consumption of New Parts based on BP Neural Network Technology

为精确地预测熔模铸造中新零件材料消耗定额,采用了BP神经网络的方法进行建模。在分析影响各工序零件材料消耗主要因素的基础上,确定了BP神经网络模型的特征参数,并根据实际情况确定了输入层和隐含层的神经元个数,从而确定了模型的结构。用试验数据对模型结构进行训练,最终建立了一个用于新零件材料消耗定额预测的BP神经网络模型。

In order to precisely predict the material consumption of new parts in investment casting, BP neural network method is used to establish a prediction model, Through determining characteristic parameters of the BP neural network model by analyzing the main factors affecting the material consumption of new parts in every processing stage, and the nerve cell numbers of the input layer and the hidden layer according to the practical conditions, the structure of the model is established. The model is trained using the experimental results. A BP neural network model to predict the material consumption of new parts is established successfully.