基于RVM-ANFIS的燃气轮机气路故障诊断

Gas turbin fault diagnosis based on RVM-ANFIS

中文关键词: 燃气轮机  故障诊断  相关向量机  自适应模糊推理系统

英文关键词: gas turbine  fault diagnosis  relevance vector machine  adaptive neural-fuzzy inference system

中文摘要:

      为了提高燃气轮机气路故障诊断的准确率和效率，本文采用相关向量机（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，本文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。

英文摘要:

      In order to improve the accuracy and efficiency of gas turbine fault diagnosis, the correlation vector machine (RVM) is used to classify the compressor, turbine blades and combustion chamber in the gas path of the gas turbine at first, 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, compared with methods such as support vector machine（SVM）and BP neural network can identify faults more accurately and quickly.