A Fault Detection and Separation Method for UAV Longitudinal Flight Control System Based on Nonlinear Adaptive Observer

Abstract

The invention relates to a fault detection and separation method for a longitudinal flight control system of an unmanned aerial vehicle based on a nonlinear adaptive observer. Based on the longitudinal nonlinear model of the UAV, the invention designs a group of nonlinear adaptive observers for the faults of actuators and sensors of the longitudinal flight control system, and transforms the fault separation problem into a model matching problem. Using the idea of ​​contribution analysis method for reference, the fault estimation value and fault direction obtained by each adaptive observer are used to construct a contribution function and standardize it, and the optimal matching model is determined by selecting the largest normalized contribution function. Finally, the size of the maximum normalized contribution function and the threshold are compared, and the results of fault detection and separation are obtained at the same time. The invention provides a new reference index for model matching by referring to the contribution analysis method, and at the same time uses the analytical model to design a nonlinear adaptive observer to obtain a more reliable fault estimation value and fault direction, and overcomes the problem that the fault direction is difficult to determine in the data-driven method insufficient.

Description

technical field [0001] The invention relates to a fault detection and separation method of a longitudinal flight control system of an unmanned aerial vehicle based on a nonlinear adaptive observer, which is used for real-time detection and separation of actuator and sensor faults in the longitudinal flight control system of an unmanned aerial vehicle, and belongs to unmanned aerial vehicles The technical field of system fault diagnosis. Background technique [0002] The research on fault diagnosis technology of complex control system is of great significance to improve the safety and reliability of the system. Taking the UAV flight control system as an example, real-time fault detection and separation provides important technical support for improving the reliability and survivability of the UAV system. At the same time, modern complex control systems usually present the characteristics of strong nonlinearity, multi-source interference, and difficult modeling. However, the...

AI Technical Summary

Problems solved by technology

However, the disadvantage of data-driven fault isolation methods is that the fault direction is difficult to define precisely

At the same time, because the system is inevitably affected by many factors such as external interference, modeling error, and measurement noise, the traditional model matching method based on the residual evaluation function is in short supply.

On the other hand, in this type of system, once a component fails, its failure mechanism will become very complicated due to the dynamic characteristics of the system and the existence of closed-loop control

Roughly selecting the fault direction using a pure data-driven method will inevitably lead to unreliable fault separation results

[0008] To sum up, the existing fault separation methods cannot effectively solve the problem of fault detection and separation of a class of complex nonlinear control systems such as UAV flight control systems, and this topic still needs further research.

Images