摘要
针对当前采煤机故障诊断系统诊断技术落后、无法精准确定故障位置及诊断结果片面等问题,提出了传统参数诊断法与神经网络相融合的智能故障诊断方法。通过对采煤机实时状态参数进行分析,保证诊断的及时性,同时借助深度残差网络强大的特征提取能力对故障部位进行分析,细化诊断结果,快速确定故障位置。2种方法相互融合,全面提高了系统的诊断效果。通过实验验证,参数诊断法和深度残差网络的准确率分别达到了100%和99.7%。基于该方法开发了采煤机关键零部件智能融合故障诊断系统,实现了采煤机关键零部件的在线监测和故障诊断,提高了采煤机故障诊断的智能化程度。
Aiming at the problems of backward diagnosis technology of the current shearer fault diagnosis system, inaccurate fault location and one-sided diagnosis results, an intelligent fault diagnosis method combining traditional parameter diagnosis method and neural network was proposed. The realtime state parameters of the shearer were analyzed to ensure the timeliness of diagnosis, and at the same time analyzed the fault location with the help of the powerful feature extraction ability of the deep residual network, refined the diagnosis result and quickly determined the fault location. The two methods are integrated each other to comprehensively improve the system diagnosis effect. Through experimental verification, the accuracy of parameter diagnosis method and deep residual network reached 100% and 99.7% respectively. Based on this method, an intelligent fusion fault diagnosis system for the key parts of shearer was developed, which realizes the online monitoring and fault diagnosis of the key parts of shearer, and improves the intelligent level of the fault diagnosis of the shearer.
作者
王萌
丁华
Wang Meng;Ding Hua(College of Mechanical and Vehicle Engineering,Taiyuan University of Technology,Taiyuan 030024,China;Shanxi Key Laboratory of Fully Mechanized Coal Mining Equipment,Taiyuan 030024,China)
出处
《煤矿机械》
2021年第6期165-169,共5页
Coal Mine Machinery
基金
山西省重点研发项目(201903D121064)。
关键词
故障诊断
采煤机
参数诊断
深度残差网络
系统
智能融合
fault diagnosis
shearer
parametric diagnosis
deep residual network
system
intelligent fusion
作者简介
王萌(1994-),河南南阳人,硕士研究生,研究方向:采煤机关键零部件故障诊断方法与系统,电子信箱:wangm_14@163.com;通讯作者:丁华,电子信箱:dinghua2002@163.com.
