基于AHP-RBR的燃气轮机压气机故障诊断方法研究

Study on Gas Turbine Compressor Fault Diagnosis Method Based on AHP-RBR

燃气轮机压气机在高转速的工作环境下运行经常会发生故障,严重时会造成燃气轮机非计划停机。为了诊断压气机的故障类型,提出基于AHP(层次分析法)的RBR(规则推理)故障诊断方法。该方法针对规则推理模型不能确定各故障征兆参数对压气机故障类型影响权重的问题,采用了层次分析法对各故障征兆参数进行分析,利用专家知识、经验和压气机故障案例分析,综合确定相对客观的各故障征兆参数权重,有效提高了规则推理故障诊断方法的准确性。仿真结果表明:与RBR模型相比,AHP-RBR模型准确率更高,可为燃气轮机压气机的故障诊断提供方法参考。

Gas turbine compressor failure often occurs in high-speed operation.In severe cases,it may cause an unplanned shutdown of the gas turbine.To diagnose the type of compressor malfunctions,an RBR(rule-based reasoning)fault diagnosis method based on AHP(analytic hierarchy process)is proposed.As the RBR model cannot determine the influence weight of each fault symptom parameter on the compressor fault type,the AHP is adopted to analyze the weight of each fault symptom parameter.By use of expertise,experience and compressor failure case analysis,the relatively objective weight of each failure symptom parameter is determined,and the accuracy of the rule-based reasoning fault diagnosis method is improved.The simulation results show that AHP-RBR,in contrast to RBR,is of higher accuracy and can provide a reference for fault diagnosis of the gas turbine compressor.