基于混合模糊聚类的火力发电机组故障智能诊断方法研究  

Research on Intelligent Diagnosis of Fossil-Fuel Power Station Faults Based on Hybrid Fuzzy Clustering

作　　者：翁忠华 高明翔 吴博伦 项大永 卢兵 WENG Zhonghua;GAO Mingxiang;WU Bolun;XIANG Dayong;LU Bing(Zhejiang Zheneng Wenzhou Power Generation Co.,Ltd.,Wenzhou,Zhejiang 325600,China)

机构地区：[1]浙江浙能温州发电有限公司,浙江温州325600

出　　处：《自动化应用》2025年第2期69-71,共3页Automation Application

摘　　要：火力发电机组的运行条件多变,固定的阈值难以适应所有情况,导致诊断精度低,为此,提出基于混合模糊聚类的火力发电机组故障智能诊断方法。针对火力发电机组的复杂故障信息进行全面处理,运用混合模糊聚类技术深入挖掘并提取故障特征,通过精细计算各样本对各故障类别的隶属度,实现对发电机组故障的智能、精准诊断。实验结果表明,基于混合模糊聚类的火力发电机组故障智能诊断方法在不同情境下的平均正确诊断率高达98.8%,充分验证了该方法在处理复杂故障模式时的稳定性和高效性。The running conditions of the fossil-fuel power station are variable,and the fixed threshold can not be adapted to all conditions,which leads to low diagnostic accuracy.Therefore,an intelligent fossil-fuel power station fault diagnosis method based on hybrid fuzzy clustering is proposed.The complex fault information of the fossil-fuel power station was comprehensively processed,and the hybrid fuzzy clustering technology was used to mine and extract the fault features,intelligent and accurate diagnosis of electric generator faults.The experimental results show that the average correct diagnosis rate of the intelligent fossil-fuel power station fault diagnosis method based on hybrid fuzzy clustering is as high as 98.8%,the stability and efficiency of the proposed method in dealing with complex failure modes are fully verified.

关 键 词：混合模糊聚类 火力发电机组 故障诊断 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]