Test Research Of HJT517G Tricone Bit Fault Diagnosis

Tricone bit is an important tool in the petroleum well drilling, whose performance has a direct impact on the drilling quality, drilling efficiency and drilling cost. However, because of the complexity of drill bit in the well bottom, it is difficult to accurately judge its operating performance. If the drill bit is taken out ahead of time, it will cause waste of manpower and resources. And if too late, it will seriously affect the drilling efficiency. In order to accurately determine the damage situation and the performance of tricone bit, it is important to carry on the reliability analysis and the failure diagnosis research. Taking tricone bit as the research object, this article adopts the methods of theoretical and experimental research to study the vibration signal feature extraction, the neural network state identification, and the reliability analysis of HJT517G tricone bit.Firstly, based on the service condition investigation of HJT517G tricone bit in some oil field and the work principle of HJT517G tricone bit, this paper analyses its failure modes and failure reason. Secondly, we design and make a set of tricone bit loading and simulating experiment system, and gather the vibration signal of normal roller bit, roller bit tooth losing and roller bit bearing fault. Based on the time domain analysis, the frequency range analysis, wavelet and wavelet packet analysis, as well as the double spectrum analysis, we obtain the failure characteristic parameter of the vibration signal. Thirdly, based on BP neural network, the BP network model is established and the numbers of network input, output layer, hidden layers and hidden nodes is determined. And then based on the study, we compile a tricone bit fault diagnosis system for the status identification of different bits by the LabWindows/CVI and matlab software. Finally, based on the probability statistics of normal working hours of HJT517G three drill bit in the different stratum, a reliability analysis system is compiled to analyze the probability distribution rule by the Visual Basic language. The results show that the fault diagnosis system and the reliability analysis system can provide an important theoretical basis for the determination of the trip time, and can provide important theory and practice foundation for the use and management of tricone bit.