改进D-S证据理论在机床状态监测中的应用研究

Application Research of Improved D-S Evidence Theory in Machine Tool Condition Monitoring

针对传统机床状态监测中单一传感器准确率不高,实时性较差的问题,提出了一种多传感器D-S证据理论融合方法进行状态监测。该方法首先利用多个传感器感知机床状态参数,同时为了减少数据收集过程中的干扰,引入了分批估计获得传感器可信度,并对获得的证据进行第一次修正;其次通过计算证据体偏离度来获得证据的修正系数,并对获得的证据进行第二次修正;最后利用证据理论对修正后的证据进行融合,完成决策和信息处理过程,得到机床当前工作状态。通过实验验证了改进方法可以提高状态监测的准确率和鲁棒性,并能有效解决证据冲突问题。

In order to solve the problems of low accuracy and poor real-time of single sensor in traditional machine condition monitoring,a multi-sensor D-S evidence theory fusion method is proposed for condition monitoring.In order to reduce the interfer⁃ence in the process of data collection,a batch estimation is introduced to obtain the credibility of the sensor,and the evidence ob⁃tained is corrected for the first time.Secondly,the correction coefficient of the evidence is obtained by calculating the deviation de⁃gree of the evidence body,and the evidence obtained is corrected for the second time.Finally,the revised evidence is fused by us⁃ing evidence theory to complete the decision-making and information processing process,and the current working state of the ma⁃chine tool is obtained.Experiments show that the improved method can improve the accuracy and robustness of condition monitor⁃ing,and can effectively solve the problem of evidence conflict.