基于扩展卡尔曼滤波的燃气轮机性能参数估计与故障诊断  

Prediction of the Performance Parameters of a Gas Turbine and Diagnosis of Its Faults Based on the Extended Kalman Wave Filtration

作　　者：康维国[1] 蒋东翔[1] 

机构地区：[1]清华大学热能工程系电力系统及发电设备控制和仿真国家重点实验室,北京100084

出　　处：《热能动力工程》2015年第5期702-707,820-821,共6页Journal of Engineering for Thermal Energy and Power

摘　　要：将扩展卡尔曼滤波方法应用在分轴燃气轮机性能参数状态估计和热力故障诊断领域。根据实际燃气轮机建立非线性热力系统模型,根据模型计算数据建立包含燃气轮机压气机、高压涡轮与动力涡轮等部件健康参数的线性SSM(状态空间模型),设计得到EKF(扩展卡尔曼滤波器);利用所设计EKF对突变型和渐变型各6种燃气轮机典型热力故障部件的健康参数进行估计。结果表明:用扩展卡尔曼滤波器对分轴燃气轮机性能参数状态估计非常及时和准确,适用于分轴燃气轮机实时状态监测和故障诊断中,并且对性能参数的渐变所引发故障的跟踪更为及时,诊断更为准确。The extended Kalman wave filtration method was applied in the domain of predicting the performance parameters and states of a gas turbine and diagnosing its thermal faults. According to a real gas turbine,a model for nonlinear thermal systems was established and based on the data calculated by using the model,a linear state and space model for gas turbines including the healthy parameters of such components as the compressor,HP turbine and power turbine of a gas turbine. On this basis,an extended Kalman filter was designed. By utilizing the extended Kalman filter thus designed,the healthy parameters of the components relating to six typical thermal faults of a gas turbine in sudden change and gradual change type respectively were estimated to realize an accurate estimation and diagnosis of the thermal faults of the split-shaft gas turbine. It has been found that to use the extended Kalman filter to predict the state of the performance parameters of the split-shaft gas turbine is very appropriate and accurate,applicable for real-time state monitoring and fault diagnosis of split-shaft gas turbines. To this end,the track of the faults triggered by gradual changes of the performance parameters is on time and the diagnosis is more accurate.

关 键 词：分轴燃气轮机 扩展卡尔曼滤波 性能参数 状态估计 热力故障诊断 

分 类 号：TK471[动力工程及工程热物理—动力机械及工程]