Sensor Fault Detection for the Attitude Measurement System of Rotary Steerable Tools

The attitude measurement of drilling tools is essential to the control of rotary steering drilling tools. However, sensor faults seriously affect its measurement accuracy. Therefore, this article focuses on the problem of sensor fault detection under drilling disturbances in the attitude measurement system, ensuring the reliable operation of the measurement system. Firstly, the attitude measurement system including accelerometers and gyroscope is modeled as a linear system with unknown inputs, measurement noise and parameter uncertainties. These bounded disturbances are described by zonotopes. Secondly, a set membership unknown input observer generating residual in finite frequency domain is proposed. In order to make the residual sensitive to faults and robust to disturbances, design conditions for observer gain are given based on the P-radius criterion and generalized KYP lemma. Then, based on whether the boundary of the residual cross zero, it is determined whether sensor faults have occurred in the system. Finally, the effectiveness of the proposed method was verified through experiments, and the results show that the method successfully detected accelerometer and gyroscope faults.