The invention discloses a credible intelligent fault diagnosis method for a ship power engine, which belongs to the technical field of fault detection and comprises the following steps: step 1, acquiring fault data of the ship power engine; step 2, screening important features of the fault data, extracting fault features of the data set obtained in the step 1, sorting feature importance through a method based on Spearman correlation coefficients, and selecting features with high importance; 3, constructing an interpretable initial model, and reasoning a confidence rule base through an evidence reasoning algorithm; step 4, designing a balance optimization strategy oriented to accuracy and interpretability dual metrics; 5, the model is optimized, an improved projection covariance matrix adaptive evolution strategy (P-CMA-ES) capable of balancing interpretability and accuracy is adopted for optimization, high fault diagnosis precision can be kept, meanwhile, the interpretability of the model is remarkably enhanced, and the balance problem between the accuracy and the interpretability is solved.
本发明公开了一种可信的船舶动力引擎智能故障诊断方法,属于故障检测技术领域,包括如下步骤:步骤一,获取船舶动力引擎故障数据;步骤二,对故障数据进行筛选重要特征,对步骤一中所获取的数据集进行故障特征提取,通过基于Spearman相关系数的方法进行特征重要性排序,选择重要性较高的特征;步骤三,构建可解释的初始模型,通过证据推理算法进行置信规则库的推理;步骤四,设计面向准确性和可解释性双度量的平衡优化策略;步骤五,对模型进行优化,采用可以平衡可解释性和准确性的改进的投影协方差矩阵自适应进化策略(P‑CMA‑ES)进行优化,本发明中,能够在保持高故障诊断精度的同时,显著增强模型的可解释性,解决了准确性与可解释性之间的平衡问题。
Trusted ship power engine intelligent fault diagnosis method
一种可信的船舶动力引擎智能故障诊断方法
16.05.2025
Patent
Elektronische Ressource
Chinesisch
DOAJ | 2017
|Intelligent ship fault diagnosis system and method based on big data
Europäisches Patentamt | 2025
|Intelligent diagnosis method for load fault of ship vibration isolation device
Europäisches Patentamt | 2024
|