基于RVM-ANFIS的燃气轮机气路故障诊断  

Gas turbine fault diagnosis based on RVM-ANFIS

作　　者：茅大钧 涂雷 臧旭东 李光 MAO Dajun;TU Lei;ZANG Xudong;LI Guang(School of Automation Engineering,Shanghai University of Electric Power,Shanghai 200090,China;Shanghai HuadianFengxian Thermal Power Limited Campany,Shanghai 201403,China)

机构地区：[1]上海电力大学自动化工程学院,上海200090 [2]上海华电奉贤热电有限公司,上海201403

出　　处：《热科学与技术》2021年第3期292-296,共5页Journal of Thermal Science and Technology

基　　金：中国华电集团有限公司2019年度重点科技资助项目(CHDKJ19-01-80);上海市“科技创新行动计划”地方院校能力建设专项资助项目(19020500700)。

摘　　要：为了提高燃气轮机气路故障诊断的准确率和效率,采用相关向量机(RVM)先对燃气轮机气路中的压气机、涡轮叶片和燃烧室进行故障划分。用自适应神经模糊推理系统(ANFIS)进一步对故障进行分类。实验结果表明,方法有很强的学习能力和特征提取能力,与支持向量机(SVM)、BP神经网络相比,能更加准确、快速地识别故障。In order to improve the accuracy and efficiency of gas turbine fault diagnosis, the relevance vector machine(RVM) is used to classify the compressor, turbine blades and combustion chamber in the gas path of the gas turbine, then adaptive neural-fuzzy inference system(ANFIS) is utilized to identify the faults further. Experimental results show that the proposed method has a strong ability of learning and extracting features, and it can identify faults more accurately and quickly compared with methods such as support vector machine(SVM) and BP neural network.

关 键 词：燃气轮机 故障诊断 相关向量机 自适应模糊推理系统 

分 类 号：TK478[动力工程及工程热物理—动力机械及工程]