基于切削区域温度数据的刀具磨损预测

Tool Wear Prediction Based on Noise Reduction Processing for Cutting Region Temperature Data

刀具磨损预测是制造业中至关重要的问题,提前预测刀具的磨损,并及时进行更换,能够降低生产成本,提高生产效率。选择切削区域温度数据来预测刀具磨损,同时考虑到加工过程中切削屑的脱落会影响数据的采集,设计了降噪算法来去除切削屑的干扰。具体而言,首先,设计了基于帧差法的降噪算法;之后,构建了卷积长短时记忆网络预测刀具磨损;最后,通过实验对方法的有效性进行验证。实验结果表明降噪算法能够有效地去除切削屑产生的噪声,提出的网络模型相比传统的BP神经网络模型预测精度有所提高,不同工况下的预测结果均方根误差平均降低了0.0171。

Tool wear prediction is a crucial issue in the manufacturing industry.Predicting tool wear in advance and replacing it in a timely manner can reduce production costs and improve production efficiency.This article selects cutting area temperature data to predict tool wear,while considering the impact of cutting chips during the data collection process,a noise reduction algorithm is designed to remove the interference of cutting chips.Specifically,we constructed a convolutional long and short term memory neural network to extract features from temperature data in the cutting area and predict tool wear.The shedding of cutting chips can generate noise,and we use the idea of frame difference method to remove the influence of cutting chips.Finally,the effectiveness of the method was verified through experiments.The experimental results show that the noise reduction algorithm can effectively remove the noise generated by cutting chips,and The proposed network model has improved prediction accuracy compared to the traditional BP neural network model,and the average root mean square error of prediction results under different working conditions has decreased by 0.0171.