LPV system identification with multiple-model approach based on shifted asymmetric laplace distribution

Abstract

The robust linear parameters varying systems identification method with multiple-model approach is addressed in this paper. Various noise and outliers commonly exist in practical industrial processes and have a serious impact on data-driven system identification methods. A statistic approach is proposed in the paper where the centralised asymmetric Laplace (CAL) distribution is employed to model the noise and therefore the parameters estimation algorithm based on CAL distribution is robust to the symmetric/asymmetric noise and outliers. CAL distribution is represented as the product of exponential distribution and Gaussian distribution, and therefore the parameters estimation formulas are deduced in the expectation maximisation algorithm framework. The efficacy and the robustness of the proposed algorithm are verified by a numerical example and the continuous stirred tank reactor system experiment.

An Online Data-Driven LPV Modeling Method for Turbo-Shaft Engines

“…The least square criterion and gradient descent algorithm are adopted to optimize the parameters of the LPV model [31][32][33]. A statistical approach was used in data-driven LPV system identification in [34]. However, the previously mentioned data-driven modeling methods are inefficient because they must be carried out offline [31,32,34] or carry online identification at the cost of reduced accuracy with the stochastic gradient descent algorithm.…”

“…A statistical approach was used in data-driven LPV system identification in [34]. However, the previously mentioned data-driven modeling methods are inefficient because they must be carried out offline [31,32,34] or carry online identification at the cost of reduced accuracy with the stochastic gradient descent algorithm. For aero engines, a more engaging, newly developed data-driven modeling method is the equilibrium manifold expansion (EME) model [35][36][37].…”