Nonlinear Recursive Estimation With Estimability Analysis for Physical and Semiphysical Engine Model Parameters

Abstract

A methodology for nonlinear recursive parameter estimation with parameter estimability analysis for physical and semiphysical engine models is presented. Orthogonal estimability analysis based on parameter sensitivity is employed with the purpose of evaluating a rank of estimable parameters given multiple sets of observation data that were acquired from a transient engine testing facility. The qualitative information gained from the estimability analysis is then used for estimating the estimable parameters by using two well-known nonlinear adaptive estimation algorithms known as extended Kalman filter (EKF) and unscented Kalman filter (UKF). The findings of this work contribute on understanding the real-world challenges which are involved in the effective implementation of system identification techniques suitable for online nonlinear estimation of parameters with physical interpretation.