加工精度在线检测系统预行程误差预测与补偿

The prediction and compensation for the pre-travel error of on-line inspecting system based on the BP neural network

预行程误差的预测和补偿能够大大提高加工精度在线检测系统的测量精度。提出了一种基于BP神经网络的检测误差预测新方法,建立了一个基于BP神经网络的在线检测系统预行程误差预测模型,通过实验数据对该网络进行训练,并将训练好的神经网络应用到实际加工零件的误差预测和补偿。为了验证该方法的有效性,以一圆柱零件的圆度误差检测为例,对其加工精度的在线测量进行了预行程误差的预测与补偿,经与CMM检测结果的对比,说明了该方法的有效性。

The accuracy of on-line measurement systems can be greatly improved by predicting and compensating for the pre-travel error.A new pre-travel error prediction method based on the back propagation（BP）neural network is proposed.A BP neural network is designed to predict the pre-travel error of the on-line inspecting system.The neural network is then trained to predict and compensate for the pre-travel error.In order to verify the accuracy of the proposed method the results of the modified on-line measurement system are compared with the CMM inspection results for a cylindrical part.The comparison of the result shows the feasibility of this method.