基于一组卡尔曼滤波器信息融合的故障诊断

Application of a Bank of Kalman Filters for Aircraft Engine Sensor\Actuator Fault Diagnosis

为了能及时准确地诊断发动机的传感器和执行机构故障,文章提出了基于一组卡尔曼滤波器信息融合的方法进行故障诊断;首先根据传感器特性设计了一组滤波器用于传感器故障诊断、隔离,每个滤波器针对一个传感器进行设计;其次根据执行机构故障特性设计了一组卡尔曼滤波器进行执行机构偏差估计,从而对执行机构进行故障诊断、隔离;接着给出了传感器、执行机构信息融合的诊断方案;最后分别给传感器、执行机构添加故障进行方案验证,仿真结果得出在传感器或者执行机构任意部件出故障的情况下,该融合方法可以有效地诊断并隔离出有故障的传感器或者执行机构。

In order to detect and isolate the sensor/actuator fault the method of information fusion based on a bank of Kalman filter for fault detection and isolation（FDI）is proposed in this paper.Firstly,a bank of Kalman filters is developed for aircraft engine sensors,each of which is designed based on a specific hypothesis for detecting a specific sensor fault.Then,a bank of Kalman filters is also applied for the actuator FDI,the actuator fault is modeled as a bias.The system which utilizes a bank of Kalman filter is developed for aircraft engine sensor and actuator FDI.Finally,add the fault into sensor/actuator separately to test the method,the results indicate that the proposed FDI system is promising for reliable diagnostics of aircraft engine sensor and actuator.