基于LASSO-RBFNN的采煤机液压系统故障诊断研究  

Research on Fault Diagnosis of Shearer Hydraulic System Based on LASSO-RBFNN

作　　者：张海波 刘昊 Zhang Haibo;Liu Hao(Licun Coal Mine,Lu’an Chemical Group,Changzhi 046000,China;College of Electrical and Control Engineering,Liaoning Technical University,Huludao 125105,China)

机构地区：[1]潞安化工集团李村煤矿,山西长治046000 [2]辽宁工程技术大学电气与控制工程学院,辽宁葫芦岛125105

出　　处：《煤矿机械》2024年第1期160-162,共3页Coal Mine Machinery

摘　　要：针对采煤机液压系统故障诊断精度不高的问题,提出一种套索(LASSO)算法与径向基函数神经网络(RBFNN)相结合的故障诊断模型。首先利用LASSO算法去除液压系统中冗余特征,筛选关键故障特征,减少模型过拟合风险;故障特征筛选后确定RBFNN拓扑结构,将采煤机液压系统故障数据输入模型中,进行故障诊断;最后将LASSO-RBFNN模型诊断结果与RBFNN模型和BP神经网络模型诊断结果进行对比。试验结果表明,该模型可用更短的网络训练时间得到较高的故障诊断精度。Aiming at the issue of low accuracy in fault diagnosis of the shear hydraulic system,proposed a fault diagnosis model that combines the least absolute shrinkage and selection operator(LASSO)algorithm with radial basis function neural networks(RBFNN).Firstly the LASSO algorithm was employed to eliminate redundant features in the hydraulic system,so as to select crucial fault characteristics and reduce the risk of overfitting in the model.After selecting the filtered fault characteristics,the RBFNN topology was determined to incorporate the fault data of the shearer hydraulic system into the model for fault diagnosis.Finally,the diagnostic results of the LASSO-RBFNN model were compared with those of the RBFNN model and the BP neural network model.The experimental results demonstrate that the proposed model achieves shorter network training time while obtaining higher fault diagnosis accuracy.

关 键 词：LASSO算法 RBFNN 采煤机液压系统 故障诊断 

分 类 号：TD421.6[矿业工程—矿山机电]