A Model - Based Jamming Detection Method for Electro-Hydrostatic Actuators

Abstract

A n anomaly or a fault, which may be treated as insignificant in a complicated engineering system, like an electro–hydraulic flight control actuator, can cause considerable performance degradation and deteriorating effects or even worse it may end up with a catastrophic system failure. One of the problems commonly observed in air vehicles is called jamming. Jamming is the type of failure where an actuator connected to a flight control surface permanently gets stuck at a certain position and does not move anymore. This might cause a loss of control of the air vehicle causing severe consequences. Although, there are several methods to detect jamming in the literature, still there is not a universally accepted solution. In this paper, a novel threshold-based fault detection and diagnosis method is proposed. The method is based on the variance of time rate of change of disturbance load and the average tracking error. When these two quantities are above a certain corresponding threshold, jamming is detected. The method is tested on an experimental setup. The experiments show that the detection performance of the method is satisfactory and the detection is performed under one second for all the test cases

Keywords

• fault detection and diagnosis
• disturbance observer
• eha
• jamming
• flight control actuation