基于CNN-LSTM的钻井泵液力端故障诊断方法研究  

Study on CNN-LSTM Based Fault Diagnosis Method for Drilling Pump Fluid Ends

作　　者：单代伟[1] 朱骅 张芳芳[1] SHAN Daiwei;ZHU Hua;ZHANG Fangfang(Sichuan Honghua Petroleum Equipment Co.,Lt,Sichuan Province Guanghan 608300,China;State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation Chengdu University of Technology,Sichuan Province Chengdu 610059,China)

机构地区：[1]四川宏华石油设备有限公司,四川广汉608300 [2]成都理工大学油田气藏地质及开发工程国家重点实验室,四川成都610059

出　　处：《内蒙古石油化工》2024年第3期29-34,共6页Inner Mongolia Petrochemical Industry

摘　　要：钻井泵液力端工作环境复杂,容易发生故障,传统故障诊断方法难以满足钻井现场需求。针对五缸式钻井泵,开展了基于深度神经网络的钻井泵液力端故障诊断研究,设计了CNN-LSTM故障诊断模型结构,研究了LSTM对故障诊断模型性能影响。结果表明,提出的CNN-LSTM模型实现了钻井泵液力端多种工况下9类故障快速准确诊断,通过引入LSTM结构,将故障诊断准确率提升了7.85%,达到了97.67%。因此提出的CNN-LSTM故障诊断模型可为钻井现场提供一种高效准确的钻井泵液力端故障诊断方法。Under complex working conditions,it is easy to lead to a failure at the drilling pump fluid end.Traditional fault diagnosis methods are difficult to meet the requirements of the drilling process.In this paper,aiming at the fault diagnosis of five-cylinder drilling pumps,a deep neural network-based fluid end fault diagnosis research was carried out,a CNN-LSTM fault diagnosis model structure was designed,and the effect of LSTM on the performance of the fault diagnosis model was investigated.The results show that the CNN-LSTM model proposed in this paper realizes fast and accurate diagnosis of 9 types of faults under multiple working conditions at the drilling pump fluid end,and by applying the LSTM structure,the accuracy of fault diagnosis model is improved by 7.85%to 97.67%.The CNN-LSTM fault diagnosis model proposed in this paper provides an efficient and accurate fault diagnosis method for the drilling pump fluid end of the drilling process.

关 键 词：钻井泵液力端 故障诊断 振动信号 CNN-LSTM 

分 类 号：TE926[石油与天然气工程—石油机械设备]