基于改进灰色神经网络模型的板料成形缺陷预测研究

为了准确预测和减少板料成形过程中可能出现的缺陷,提出了一种改进的灰色神经网络预测模型。该模型利用BP神经网络辅助灰色预测模型进行预测。其中,灰色模型进行粗预测,神经网络模型修正其误差,再通过寻找最佳权值以优化灰色模型中微分所对应的背景值,进而得到精度更高的灰色神经网络模型。以国际著名板料成形数值模拟会议NUMISHEET'93的方盒件拉深为例,运用改进的灰色神经网络模型,预测其拉裂和起皱。结果表明,改进的灰色神经网络模型具有很高的预测精度,相比于未改进的灰色神经网络模型,预测结果更加准确和稳定。

Research on Defect Prediction in Sheet Metal Forming Based on Improved Gray Neural Network Model

In order to predict accurately and reduce the possible defects in the sheet metal forming process, an improved gray neural network prediction model was proposed. In this model, BP neural network model was used to assist the grey prediction model, that is: the grey model made a coarse firstly,and the errors were corrected by the neural network, after that a new gray neural network model with higher accuracy was obtained through finding the best weight values to optimize the background value corresponding to the differential in the grey model. Taking the drawing of square boxes as an example, which is from the internationally famous sheet metal forming numerical simulation conference NUMISHEET'93, the crack and wrinkle is predicted by applying the improved gray neural network model. Compared with the gray neural network model which is not improved, the results show that the improved gray neural network model has higher prediction accuracy, and the prediction results are more stable.