A Robust Fault Diagnosis Method for Carrier Pitch Channel Control System

Abstract

The invention discloses a robust fault diagnosis method of a carrier pitching channel control system. The method includes the following steps that state space description of the pitching channel control system is built, and unknown input disturbance of the pitching channel control system is obtained; unknown input disturbance constraint conditions for meeting stability of the system are determined; fault diagnosis observer robust performance indexes are determined; a robust fault diagnosis observer is constructed for the pitching channel control system with unknown input disturbance and faults; fault decision is made by comparing the relation between the residual error output by the robust fault diagnosis observer and a threshold value. The fault diagnosis observer is constructed through the robust technology, and the robustness of the fault diagnosis method is improved.

Description

technical field [0001] The invention relates to a fault diagnosis method, in particular to a robust fault diagnosis method for a carrier pitch channel control system. Background technique [0002] System fault diagnosis methods are broadly classified into two categories: knowledge-based and system model-based methods. The advantage of the knowledge-based fault diagnosis method is that it does not rely on the system analysis model but requires a large amount of system working condition information for fault reasoning and judgment. But its disadvantage is that it is difficult to separate and estimate faults, especially inconvenient for online diagnosis of faults, and it is ineffective in judging minor faults. With the strong coupling of the system and the improvement of its reliability requirements, the fault diagnosis method based on knowledge cannot meet the high performance index requirements. The improvement of modeling theory and control theory has promoted the developm...

AI Technical Summary

Problems solved by technology

However, most of the methods have the defects of poor algorithm robustness, high diagnostic false alarm rate, poor convergence and insufficient analysis of disturbances.

Images