Interpretable Fault Diagnosis of Fiber Optic Gyroscope Based on Belief Rule Base with Projection Gradient Method

As an important type of navigation equipment in inertial navigation systems, fiber optic gyroscopes (FOG) play a very important role. Due to the complex operating environment, it is inevitably subject to failure. Therefore, it is necessary to conduct the fault diagnosis on FOG. In this paper, an interpretable fault diagnosis method using belief rule base (BRB) is proposed to provide trustworthy diagnosis results. The work of this article consists of two parts: First, BRB is constructed based on expert expertise and used as an interpretable knowledge base for fault diagnosis. The evidential reasoning (ER) algorithm is used to infer fault states. Then, to maintain the interpretability of the fault diagnosis process, the projected gradient method is proposed to optimize the initial BRB. A FOG fault diagnosis case is conducted to verify the practicability of the proposed method. The case indicates that the BRB-based FOG fault diagnosis model achieves better diagnostic accuracy and better interpretability than other models.