Abstract In the field of industrial automation and intelligent manufacturing, accurate assessment of the health status of electromechanical equipment is crucial for preventing failures, improving productivity, and reducing maintenance costs. However, due to the complexity of the equipment operating environment and the uncertainty of the observed data, the existing belief rule base (BRB) -based assessment methods face the problems of exponential growth of the rule size (combinatorial rule explosion) and impaired model interpretability. To this end, this paper proposes an explainable BRB with interval structure (EBRB-I) method. The method improves belief rules by introducing interval structure and constructs a health status assessment model to alleviate the combinatorial rule explosion problem. Meanwhile, a data similarity-based interval division method is applied to optimize rule partitioning and improve model accuracy. In addition, interpretability constraints are added in the parameter optimization process to preserve the interpretability of the optimized model. In order to test the effectiveness of the model, experiments were carried out using the measured data of the EQ6BT diesel engine. Experimental results demonstrate that the EBRB-I model consistently achieves an accuracy exceeding 97% across multiple iterations and outperforms various comparison methods. Meanwhile, its rule base size is reduced by more than 33% compared to the traditional BRB model, significantly improving evaluation efficiency. Furthermore, interpretability analysis indicates that the EBRB-I model’s belief distribution aligns more closely with expert knowledge, verifying its interpretability.
Liu et al. (Thu,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: