基于PNN神经网络的凿岩台车电液控制系统故障诊断研究  

Rock Drill Robot’s Electro-hydraulic Control System Fault Diagnosis Using PNN Neural Network

作　　者：牛帅亭 徐巧玉[1] 张正 NIU Shuaiting;XU Qiaoyu;ZHANG Zhen(School of Mechatronics Engineering,Henan University of Science and Technology,Luoyang 471000,China;Luoyang GINGKO Technology Co.,Ltd.,Luoyang 471000,China)

机构地区：[1]河南科技大学机电工程学院,洛阳471000 [2]洛阳银杏科技有限公司,洛阳471000

出　　处：《自动化与仪表》2024年第4期31-36,共6页Automation & Instrumentation

基　　金：国家自然科学基金项目(51205108)。

摘　　要：针对凿岩台车电液控制系统故障诊断效率低的问题,该文提出一种结合故障树分析法和概率神经网络(probabilistic neural network,PNN)的故障诊断方法。首先,基于电液控制系统的结构和工作原理构建其故障树模型;然后通过对故障树模型进行定性分析,确定其最小割集和典型故障种类,以选取的典型故障种类的关键参数构建故障征兆矩阵,通过PNN神经网络对该矩阵进行训练和计算,实现对系统典型故障状态的自动识别。实验结果表明,该文方法的平均诊断时间为1.2 s,平均诊断准确率为80%,能够快速准确地定位系统故障,可满足凿岩台车电液控制系统故障诊断的工程实际需求。Aiming at the problem of low fault diagnosis efficiency of the electro-hydraulic control system of rock drill truck,this paper proposes a fault diagnosis method combining fault tree analysis and probabilistic neural network(PNN).First of all,based on the structure and working principle of the electro-hydraulic control system,a fault tree model is constructed,then the fault tree model is qualitatively analyzed to determine the minimum cut-off set and the typical fault types,and the fault sign matrix is constructed with the key parameters of the selected typical fault types,which is trained and computed by the PNN neural network,to realize the automatic identification of the typical fault states of the system.The experimental results show that the average diagnosis time of this paper’s method is 1.2 s,and the average diagnosis accuracy rate is 80%,which can quickly and accurately locate the system faults,and can satisfy the engineering practical needs of the fault diagnosis of the electro-hydraulic control system of rock drill cart.

关 键 词：凿岩台车 电液控制系统 故障树 PNN神经网络算法 故障诊断 

分 类 号：TP241[自动化与计算机技术—检测技术与自动化装置]