Applications of Polynomial Neural Networks to FDIE and Reconfigurable Flight Control

Fault detection, isolation, and estimation (FDIEI functions and reconfiguration strategres for Right control systems present major technical chaIImges, primarily because of uncertainties resulting from limited observability and an almost unlimited variety of malfunction and damage scenarios. This paper deals primarily with a portion of the probIem, i.e., global FDIE for single impairments of control effectors. Reference is also made to reconfiguration strategies. Polynomial neural networks are synthesized using a constrained error criterion to obtain pairwise discrimination between impaired and no-fail conditions and isolation h e e n impairment classes. The pairwise discrimi~tora~re thm combined in a form of voting logic Polynomial netwarks are also synthesized to obtain estimates of the amount af effector impairment. The Algorithm for Synthesis of Polynomial Networks (ASEN) arrd related methods are used to create the networks, which are highorder, linear or nonlinear, analytic, multivariate functions of the in-flight observables. This paper outlines the design procedure, including database preparation, extraction of waveform features, network synthesis techniques, and the architecture of the FDIE system that has been studied for the Control Reconfigurable Combat Aircraft. Representative performance results are prwvided.