基于一维卷积神经网络的定向钻机故障诊断专家系统设计

Design of Directional Drilling Machine Fault Diagnosis Expert System Based on One-dimensional Convolutional Neural Network

针对目前定向钻机的监测主要依靠人力,其故障的诊断主要依靠专家经验,速度缓慢,效率较低,提出基于一维卷积神经网络(1D CNN)的定向钻机故障诊断专家系统,利用1D CNN实现了推理机的设计,利用SQL Server 2005建立知识库,最后通过LabVIEW和Python混合编程开发专家系统。经模拟测试,该专家系统能快速给出故障原因、部位和处理建议,提高了定向钻机故障诊断的速度和效率,为进一步实现矿用装备的智能化奠定基础。

As the current monitoring of directional drilling rig mainly relies on manpower,and its fault diagnosis mainly relies on expert experience,which is slow and inefficient,an expert system for fault diagnosis of directional drilling rig based on 1D convolutional neural network(1D CNN)is proposed.The reasoning machine is designed by 1D CNN,and the knowledge base is established by SQL Server 2005.Finally,the expert system is developed through LabVIEW and Python mixed programming.Through the simulation test,the expert system can quickly give the fault cause,location and treatment suggestions,improve the speed and efficiency of directional drilling rig fault diagnosis,and lay a foundation for further realization of intelligent mining equipment.