基于模糊逻辑的航空发动机虚拟传感器

在双通道的发动机控制系统中，未超过故障阈值的传感器慢漂移故障难以隔离，且在故障检出前影响控制系统的反馈信号。引入机栽自校正实时模型作为第三通道的虚拟传感器，并提出模糊隶属度加权和动态阈值技术以获得可靠的传感器适应值，实现双余度系统中失效传感器的检测和隔离。仿真结果表明，这种方法能可靠地改进适应信号，将由于传感器故障引起的发动机控制系统扰动抑制到0．2％以内，并能避免失控。

Aeroengine Virtual Sensor Based on Fuzzy Logic

The sensor in-range fault isolation and fault accommodation offers a challenge in dual channel engine control systems. An on-board, real-time self-tuning engine model is used to provide a third virtual sensor of the channel and to detect and isolate sensor faults. The sensor outputs are assigned as weights based on fuzzy membership function, the accommodation value is computed as a weighted average. Finally the dynamic threshold is provided to improve the reliability of the virtual sensor. Simulation shows that the method can improve the quality of the accommodation signal and reduce the induced failure transients.