Aircraft parameter identification for application within a fault-tolerant flight control system.

摘要

A parameter identification study was conducted to identify a detailed aircraft mathematical model for application within a fault-tolerant flight control system that aims to detect, identify, and accommodate for sensor and actuator failures. Specifically, a mathematical model was identified under nominal conditions for two aircraft platforms, and a model was developed for one platform under actuator failure conditions. These models are to be used in flight control law design and to account for actuator failures on the primary control surfaces for one of the research platforms. In order to accurately model the aircraft behavior following a control surface failure, the effects of an individual surface on the aircraft dynamics was estimated. Since an individual control surface deflection---for example in the event of a locked actuator---causes a coupling between the longitudinal and lateral-directional dynamics, additional terms were identified in the state space and stability and control derivative mathematical models. These models were derived from measured flight data acquired from pilot and automated computer-injected maneuvers under both nominal and failure conditions. From this analysis, the stability and control derivatives were extracted to determine the aerodynamic forces and moments on each aircraft. These aerodynamics were next introduced into a simulation environment to validate the accuracy of the identified mathematical models. A Data Compendium (DATCOM) -- based analysis was conducted in order to provide a means of comparison of the models obtained through the parameter identification study and to provide constraints on parameter optimization. Finally, a confidence interval analysis was conducted to determine the reliability of the estimated values. Several simulation studies were conducted to validate the accuracy of the models for each research platform, focusing on both nominal and primary control surface failure conditions where applicable. The model outputs were compared to the measured flight data from the two respective research platforms to validate the accuracy of the estimated parameters.