Moving-Horizon Guaranteed Parameter Estimation

Abstract

This paper is concerned with guaranteed parameter estimation of non-linear dynamic systems in a context of bounded measurement error. The problem consists of approximating the set of all possible parameter values such that the predicted values of plant outputs match their corresponding measurements within prescribed error bounds. Efficient algorithms are studied for bounding the set of guaranteed parameter estimates, where, in order to enhance the solution procedure, we investigate a novel method using principles of moving-horizon estimation. The principle of the method lies in selection of a subset of available measurements, which are then used for on-line calculations. The crucial part of the method is the selection procedure of these measurement points, where we propose three different methods. We apply the proposed methodology to a case study of a membrane process. The proposed approach is found to significantly reduce the computational burden, in terms of CPU time, as compared to state-of-the-art approaches.