During the drilling process, the non-linear contacts between the bit and the bottom hole, the drill string and the borehole wall can cause the bit’s stick-slip vibration, which will shorten the life of the bit and even endanger the safety of the drill string. The severity of stick-slip vibration of a bit can be identified by the rotary speed of a bit, the triaxial accelerations of the drill string, the wellhead torque and other parameters measured by the measuring while drilling (MWD) tools in the downhole and devices on the surface. To evaluate the level of stick-slip vibration, this paper proposes a risk assessment method of sick-slip vibration based on backpropagation neural network (BPNN). According to the time and frequency domain analysis of the data collected from simulation, the feature parameters of the time and frequency domains of signals are extracted, and then the kernel principal component analysis (KPCA) is applied to reduce dimensions. Consequently, the feature vectors can be obtained, which become the input parameters of the BPNN. Based on BPNN algorithm, the stick-slip vibration of the bit is determined, and the classification of stick-slip vibration strength is carried out. The results show that this method can effectively identify the severity of stick-slip vibration of a bit. Therefore, this method is valid to evaluate the stick-slip vibration of a bit, which will help drillers adjust the drilling parameters practically according to the severity of vibration, so as to reduce the risks of stick-slip vibration during drilling and improve the efficiency and safety of drilling operation.
- Pressure Vessels and Piping Division
Research on Risk Assessment Method of Stick-Slip Vibration of the Bit Based on BP Neural Network Algorithm
- Views Icon Views
- Share Icon Share
- Search Site
Chen, C, Zhang, S, Zhang, H, Li, X, & He, Z. "Research on Risk Assessment Method of Stick-Slip Vibration of the Bit Based on BP Neural Network Algorithm." Proceedings of the ASME 2018 Pressure Vessels and Piping Conference. Volume 7: Operations, Applications, and Components. Prague, Czech Republic. July 15–20, 2018. V007T07A026. ASME. https://doi.org/10.1115/PVP2018-84144
Download citation file: