Research on the Application of Multi-Layer Tree-Structured Belief Rule Base System in Aviation Engine Performance Evaluation

In aerospace, to ensure the safe operation of aircraft, as well as efficiency and maintenance costs. It is necessary to perform reliable safety checks on complex systems such as aircraft engines. The Belief Rule Base (BRB) system is an intelligent decision support system for dealing with uncertain information and simulating human reasoning processes, which combines evidential reasoning and fuzzy logic and introduces a belief framework structure with strong interpretability, and is commonly used in areas such as evaluation and diagnosis. However, the rule base of BRBs often faces the risk of combinatorial rule explosion, and the traditional hierarchical structure often lacks a common construction standard. In this paper, we introduce a BRB model (MTS-BRB) based on a multi-layer tree structure that can self-organize to construct a tree structure, thus providing a guiding scheme for modeling hierarchical BRBs and solving the combinatorial rule explosion problem. In addition, an interpretable optimization algorithm applicable to the MTS-BRB model is proposed, which ensures the post-hoc interpretability of the model to a certain extent, and the reliability of the model and the effectiveness of the optimization algorithm are demonstrated by the turbofan engine case study and the milling case study.