Identification of hybrid and linear parameter‐varying models via piecewise affine regression using mixed integer programming

Correspondence Manas Mejari, IDSIA, Dalle Molle Institute for Artificial Intelligence SUPSI-USI, 6928 Manno, Switzerland. Email: manas.mejari@idsia.ch Summary This article presents a two-stage algorithm for piecewise affine (PWA) regression. In the first stage, a moving horizon strategy is employed to simultaneously estimate the model parameters and to classify the training data by solving a small-size mixed-integer quadratic programming problem. In the second stage, linear multicategory separation methods are used to partition the regressor space. The framework of PWA regression is adapted to the identification of PWA AutoRegressive with eXogenous input (PWARX) models as well as linear parameter-varying (LPV) models. The performance of the proposed algorithm is demonstrated on an academic example and on two benchmark experimental case studies. The first experimental example concerns modeling the placement process in a pick-and-place machine, while the second one consists in the identification of an LPV model describing the input-output relationship of an electronic bandpass filter with time-varying resonant frequency.

[1]  Alberto Bemporad,et al.  An Accelerated Dual Gradient-Projection Algorithm for Embedded Linear Model Predictive Control , 2014, IEEE Transactions on Automatic Control.

[2]  Roland Tóth,et al.  LPV model order selection in an LS-SVM setting , 2013, 52nd IEEE Conference on Decision and Control.

[3]  Roland Tóth,et al.  An SDP approach for l0-minimization: Application to ARX model segmentation , 2013, Autom..

[4]  Alberto Bemporad,et al.  Regularized moving-horizon piecewise affine regression using mixed-integer quadratic programming , 2017, 2017 25th Mediterranean Conference on Control and Automation (MED).

[5]  Fabien Lauer,et al.  On the complexity of piecewise affine system identification , 2015, Autom..

[6]  Alberto Bemporad,et al.  A Numerically Robust Mixed-Integer Quadratic Programming Solver for Embedded Hybrid Model Predictive Control , 2018 .

[7]  René Vidal,et al.  Identification of Hybrid Systems: A Tutorial , 2007, Eur. J. Control.

[8]  Tyrone L. Vincent,et al.  Nonparametric methods for the identification of linear parameter varying systems , 2008, 2008 IEEE International Conference on Computer-Aided Control Systems.

[9]  Vihangkumar V. Naik,et al.  Embedded Mixed-Integer Quadratic Optimization using Accelerated Dual Gradient Projection , 2017 .

[10]  Kiyotsugu Takaba,et al.  Identification of piecewise affine systems based on statistical clustering technique , 2004, Autom..

[11]  Alberto Bemporad,et al.  Identification of hybrid and linear parameter varying models via recursive piecewise affine regression and discrimination , 2016, 2016 European Control Conference (ECC).

[12]  Romain Postoyan,et al.  Proceedings of the European Control Conference , 2015 .

[13]  Laurent Bako,et al.  Identification of switched linear systems via sparse optimization , 2011, Autom..

[14]  Alberto Bemporad,et al.  Identification of piecewise affine systems via mixed-integer programming , 2004, Autom..

[15]  Stéphane Lecoeuche,et al.  A recursive identification algorithm for switched linear/affine models , 2011 .

[16]  Yves Rolain,et al.  Benchmark Data on a Linear Time- and Parameter-varying system* , 2015 .

[17]  Manfred Morari,et al.  A clustering technique for the identification of piecewise affine systems , 2001, Autom..

[18]  BakoLaurent.  Identification of switched linear systems via sparse optimization , 2011 .

[19]  W. P. M. H. Heemels,et al.  A Bayesian approach to identification of hybrid systems , 2004, IEEE Transactions on Automatic Control.

[20]  Leo Breiman,et al.  Hinging hyperplanes for regression, classification, and function approximation , 1993, IEEE Trans. Inf. Theory.

[21]  Henrik Ohlsson,et al.  Identification of switched linear regression models using sum-of-norms regularization , 2013, Autom..

[22]  Roland Tóth,et al.  LPV system identification under noise corrupted scheduling and output signal observations , 2015, Autom..

[23]  Alberto Bemporad,et al.  Observability and controllability of piecewise affine and hybrid systems , 2000, IEEE Trans. Autom. Control..

[24]  Alberto Bemporad,et al.  A Convex Feasibility Approach to Anytime Model Predictive Control , 2015, ArXiv.

[25]  Alberto Bemporad,et al.  Piecewise affine regression via recursive multiple least squares and multicategory discrimination , 2016, Autom..

[26]  Michel Verhaegen,et al.  Subspace identification of Bilinear and LPV systems for open- and closed-loop data , 2009, Autom..

[27]  O. Mangasarian,et al.  Multicategory discrimination via linear programming , 1994 .

[28]  Manfred Morari,et al.  Multi-Parametric Toolbox 3.0 , 2013, 2013 European Control Conference (ECC).

[29]  A. Garulli,et al.  A survey on switched and piecewise affine system identification , 2012 .

[30]  Wei Xing Zheng,et al.  Model structure learning: A support vector machine approach for LPV linear-regression models , 2011, IEEE Conference on Decision and Control and European Control Conference.

[31]  Alberto Bemporad,et al.  Control of systems integrating logic, dynamics, and constraints , 1999, Autom..

[32]  A. Juloski,et al.  Data-based hybrid modelling of the component placement process in pick-and-place machines , 2004 .

[33]  Bassam Bamieh,et al.  Identification of linear parameter varying models , 1999, Proceedings of the 38th IEEE Conference on Decision and Control (Cat. No.99CH36304).

[34]  Constantino M. Lagoa,et al.  Set membership identification of switched linear systems with known number of subsystems , 2015, Autom..

[35]  Bart De Schutter,et al.  Equivalence of hybrid dynamical models , 2001, Autom..

[36]  Alberto Bemporad,et al.  A bounded-error approach to piecewise affine system identification , 2005, IEEE Transactions on Automatic Control.