基于自适应神经-模糊推理系统模糊信息融合的采煤机截齿磨损在线监测

Online Monitoring of Shearer's Pick Wear Based on ANFIS Fuzzy Information Fusion

为实现对截齿截割过程中磨损程度的实时精确在线监测,分别测试和提取不同磨损程度的截齿在截割过程中的振动信号、声发射信号和温度信号,建立不同磨损程度截齿截割信号的多特征样本数据库,根据最小模糊度优化模型计算求解各特征信号的最优模糊隶属度函数,采用自适应神经-模糊推理系统多维模糊神经网络方法实现多传感特征信息的决策融合,输出置信度和权重较高的截齿磨损量融合结果。通过随机测试实验对融合系统进行验证,结果表明,基于ANFIS模糊信息融合的截齿磨损监测系统辨识度较高,测试结果最大误差在6.5%以内,系统具有良好的融合效果以及较高的测试精度。

In order to realize the realtime and accurate online monitoring of the wear degree in the cutting processes,the vibration signals,acoustic emission signals and temperature signals of different wear degrees were tested and extracted,and the multi feature sample databases of different wear degrees to the cutting signals were established.The optimal fuzzy membership function for each characteristic signal was calculated by the minimum ambiguity optimization model,and the method of the ANFIS multidimensional fuzzy neural network was adopted to realize the fusion of multi sensor feature informations,then the fusion results of the output confidence and weight were higher.According to the results of the random experiments of the fusion system,the identification degree of the cutting wear monitoring system based on ANFIS fuzzy information fusion is high,and the maximum error of the test results is less than 6.5%,and the results show that the system has good fusion effect and higher test accuracy.