Modeling and Simulation of Failures for Primary Control Surfaces

Abstract

This paper describes the results of the extension of an existing failure modeling approach for the longitudinal aircraft dynamics to the lateral-directional dynamics. In particular a general formulation is developed for the most common failure scenarios, that is actuator blockage with and without a missing portion of the control surface. The approach consists of modeling the contribution of each individual control device within the expressions for the total external forces and moments using a single "efficiency" parameter that is easily accessible during simulation. Although the methodology can be applied to any force and moment generating control surface the paper describes simulation results relative to failures for elevators/stabilators, ailerons, and rudder occurring separately and simultaneously.

Evaluation of a Fault-Tolerant Scheme in a Six-Degree-of-Freedom Motion Flight Simulator

“…The resulting selection typically reflects a compromise between maximizing the detection capabilities while minimizing the false alarm rate. While a number of FDII schemes featuring "fixed thresholds" have shown capabilities in correctly detecting and identifying actuator and sensor failures [16,17] in desktop computer simulations, they tend to be very sensitive to false alarms and/or incorrect identifications in the presence of a pilot-in-the loop. West Virginia University (WVU) researchers [19] have introduced an adaptive threshold technique based on the concept of a "Floating Limiter" (FL); this scheme eliminates the need for scheduling the failure thresholds with varying flight conditions.…”

“…Since a substantial cross coupling of aircraft dynamics is associated with an actuator failure, the occurrence of an actuator failure can be differentiated or isolated from the sensor failure scenario through the analysis of the crosscorrelation between the longitudinal and lateral-directional dynamics [16,17]. Thus, conditions for actuator/sensor failure isolation can be represented through the following Boolean relationships as follows: R pq R pq_FL or R rr R rr_FL and t t 1 ⇒ Actuator Failure (10) R pq R pq_FL or R rr R rr_F L and t t 1 ⇒ Sensor Failure (11)…”

“…While fixed threshold approaches have been used for actuator [16] and integrated [17,18] sensor/actuator failures, Perhinschi et al [19] have extended the actuator failure FDII with an adaptive threshold, resulting in a lower rate of false alarms and a relatively simpler logic. A similar effort for sensor failure [13] was performed using an adaptive threshold concept.…”

Dynamic neuro-fuzzy control design for civil aviation aircraft in intelligent landing system

“…Details of the model can be found in [16]. For the purposes of this study, the elevator and aileron control surface aerodynamic data has been split into two parts corresponding to left and right surfaces using CFD computations, similar to [17]. The aerodynamic model also contains the ground effects.…”

UAV Analytical Redundancy based fault detection of the Airspeed Sensor via Generalized Likelihood Ratio Test

“…The main advantage of these approaches is the capacity of learning specific tasks exploiting the experience derived by the continuous interaction with the system under control. In this context the authors in 24,25,26,27 the authors proposes a number of NN learning based SFD and Isolation schemes for UAV systems.…”