Hardware versus lumped cross-monitoring in smart multilane torque-summed electromechanical actuators

Abstract

High integrity in multilane smart electromechanical actuation systems is achieved through hardware redundancy and intelligent monitoring. Hardware redundancy is associated with the inclusion of repeated hardware of key components that are likely to fail. Intelligent monitoring is associated with the inclusion of a fault detection and fault isolation (FDI) system that is capable of correctly detecting, identifying, isolating, and replacing failed components. The main aim of this paper is to address two cross-monitoring techniques (lumped and threephase) in a single-type torque-summed architecture. In the lumped cross-monitoring technique, all lanes will be represented by their lumped models, while in hardware cross-monitoring, three-phase equivalents will model each motor. The latter is a more accurate representation and allows for more detailed motor and power conditioner failure-related tests. The analysis is based on a four-lane actuation system capable of driving aerodynamic and inertial loads (with two lanes failed) of an aileron control surface similar to that of the Sea Harrier.The paper will start by giving an overview of the system, by describing:(a) the fault detection and fault isolation (FDI) system; (b) the associated simulation graphical Monte Carlo (SGMC) as a threshold setting method; (c) for completeness, brief yet key models of the brushless DC motor, the aerodynamic load, and a small aircraft.The paper will then proceed to develop a threshold setting strategy by examining peak lane disparities due to inherent system randomness. The largest failure transient envelopes after failure isolations are then identified for different flight conditions, and their impact on the aircraft response in roll will then be assessed. The paper will conclude by showing that both actuator and aircraft requirements were met despite the use of two different cross-monitoring techniques.