基于知识规则挖掘的风电机组故障识别与解释性分析  

FAULT IDENTIFICATION AND INTERPRETATIVE ANALYSIS OF WIND TURBINE BASED ON KNOWLEDGE RULE MINING

作　　者：钱小毅 孙天贺 王宝石[1] 韩玥[1] Qian Xiaoyi;Sun Tianhe;Wang Baoshi;Han Yue(Key Laboratory of Power Grid Energy Saving and Control,Shenyang Institute of Engineering,Shenyang 110036,China)

机构地区：[1]沈阳工程学院,电网节能与控制重点实验室,沈阳110036

出　　处：《太阳能学报》2023年第8期379-385,共7页Acta Energiae Solaris Sinica

基　　金：辽宁省教育厅基本科研项目(LJKQZ2021085,LJKQZ2021086);辽宁省自然科学基金(2022-BS-222)。

摘　　要：针对传统风电机组故障识别方法精度难以保证且缺乏解释性的问题,提出一种以模糊规则分类系统为框架的风电机组故障识别与解释性分析方案。离线阶段通过启发式学习生成代表性故障规则,采用多种群量子进化算法实现故障规则寻优,提高对故障的识别精度;在线阶段定义近邻规则竞争策略,在此基础上提出基于故障规则后处理策略的故障解释性分析方案,实现潜在故障概率排序以及解释性的关键异常征兆表达。以兆瓦级风电机组常见的10种故障数据进行仿真分析。结果表明,所提方法可有效提高故障识别精度,同时提供可靠的故障解释性结果。Aiming at the problems that the accuracy of traditional wind turbine fault identification methods is difficult to guarantee and lacks interpretability,a scheme for wind turbine fault identification and explanatory analysis based on the fuzzy rule classification system is proposed.In the offline stage,representative fault rules are generated through heuristic learning,and the multi-population quantum evolutionary algorithm is used to optimize fault rules and improve the accuracy of fault identification.In the online stage,a competition strategy for neighbor rules is defined,and a post-processing strategy based on fault rules is constructed.On this basis,the fault interpretive analysis scheme realizes the ranking of potential fault probabilities and the explanatory expression of key abnormal symptoms.The simulation analysis is carried out with the data of 10 common faults of megawatt wind turbines.The results show that the proposed method can effectively improve the fault identification accuracy while provide reliable fault interpretability results.

关 键 词：风电机组 数据挖掘 故障诊断 模糊规则分类系统 故障概率 解释性分析 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置] TM614[自动化与计算机技术—控制科学与工程]