Data Processing Consideration and Model Validation in Flight Vehicle System Identification

Abstract

There are four steps in system identification. Data Processing constitutes the first and most essential step. In this paper an overview of the flight data processing for reaching a sound set of data is presented. It includes the analysis of the types of data available, the method of exclusion of outliers and noise, bias corrections, and filtering of disturbances. Filtering include time domain and frequency domain processing. On the other hand, model validation is considered the final step for aircraft identification. This was accomplished for an innovative model of elevator hinge moment (EHM) in a turboprop aircraft equipped with a mechanical control system. Here, optimization of the identification design has been achieved by iteratively estimating the unknown model parameters.