Identification of errors-in-variables model with observation outliers based on Minimum-Covariance-Determinant

• Published in: 2007 American Control Conference pp. 134 - 139
• Main Author: ALMutawa, J.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.07.2007
• Subjects: Cities and towns; Earthquakes; Error correction; errors-in-variables model; Gaussian noise; Least squares approximation; Linear regression; Minerals; minimum covariance determinant; outliers; Petroleum; random search algorithm; State-space methods; Subspace system identification; System identification
• ISBN: 9781424409884; 1424409888
• ISSN: 0743-1619
• DOI: 10.1109/ACC.2007.4282931

In this paper, we develop a subspace system identification algorithm for the errors-in-variables (EIV) model subject to observation noise with outliers. By using the minimum covariance determinant (MCD), we identify and delete the outliers, and then apply the classical EIV subspace system identification algorithms to get state space models. In order to solve the MCD problem for the EIV model we propose a random search algorithm. The proposed algorithm has been applied to a heat exchanger data.