Identification of structured state-space models. (English) Zbl 1388.93017
Summary: Identification of structured state-space (gray-box) model is popular for modeling physical and network systems. Due to the nonconvex nature of the gray-box identification problem, good initial parameter estimates are crucial for successful applications. In this paper, the nonconvex gray-box identification problem is reformulated as a structured low-rank matrix factorization problem by exploiting the rank and structured properties of a block Hankel matrix constructed by the system impulse response. To address the low-rank optimization problem, it is first transformed into a Difference-of-Convex (DC) formulation and then solved using the sequentially convex relaxation method. Compared with the classical gray-box identification methods like the Prediction-Error Method (PEM), the new approach turns out to be more robust against converging to non-global minima, as supported by a simulation study. The developed identification can either be directly used for gray-box identification or provide an initial parameter estimate for the PEM.
MSC:
| 93A30 | Mathematical modelling of systems (MSC2010) |
| 93B30 | System identification |
| 93E12 | Identification in stochastic control theory |
| 93B17 | Transformations |
| 93B40 | Computational methods in systems theory (MSC2010) |
Software:
System Identification ToolboxReferences:
| [1] | Bellman, R.; Åström, K.J., On structural identifiability, Mathematical biosciences, 7, 3, 329-339, (1970) |
| [2] | Bergamasco, M.; Lovera, M., State space model identification: from unstructured to structured models with an \(h_\infty\) approach, IFAC Proceedings volumes, 46, 2, 202-207, (2013) |
| [3] | Boyd, S., () |
| [4] | Chen, T.; Andersen, M.S.; Ljung, L.; Chiuso, A.; Pillonetto, G., System identification via sparse multiple kernel-based regularization using sequential convex optimization techniques, IEEE transactions on automatic control, 59, 11, 2933-2945, (2014) · Zbl 1360.93720 |
| [5] | Chen, T.; Francis, B., () |
| [6] | Dorf, R.C.; Bishop, R.H., Modern control systems, (2011), Pearson |
| [7] | Franklin, G.F.; Powell, J.D.; Workman, M.L., Digital control of dynamic systems, vol. 3, (1998), Addison-wesley Menlo Park |
| [8] | Hu, Y.; Zhang, D.; Ye, J.; Li, X.; He, X., Fast and accurate matrix completion via truncated nuclear norm regularization, IEEE transactions on pattern analysis and machine intelligence, 35, 9, 2117-2130, (2013) |
| [9] | Ljung, L., System identification: theory for the user, (1999), Prentice Hall New Jersey |
| [10] | Ljung, L., The system identification toolbox: the manual, (2013), The MathWorks Inc. Natick, MA, USA, 1986, Edition 8.3 2013 |
| [11] | Lu, C.; Tang, J.; Yan, S.; Lin, Z., Nonconvex nonsmooth low rank minimization via iteratively reweighted nuclear norm, IEEE transactions on image processing, 25, 2, 829-839, (2016) · Zbl 1408.94866 |
| [12] | Mercere, G.; Prot, O.; Ramos, J., Identification of parameterized gray-box state-space systems: from a black-box linear time-invariant representation to a structured one, IEEE transactions on automatic control, 59, 11, 2873-2885, (2014) · Zbl 1360.93184 |
| [13] | Parrilo, P., & Ljung, L. (2003). Initialization of physical parameter estimates, In S. W. P. van der Hof, B. Wahlberg (Ed.), Proc. 13th IFAC symposium on system identification; Parrilo, P., & Ljung, L. (2003). Initialization of physical parameter estimates, In S. W. P. van der Hof, B. Wahlberg (Ed.), Proc. 13th IFAC symposium on system identification |
| [14] | Prot, O.; Mercere, G.; Ramos, J., A null-space-based technique for the estimation of linear-time invariant structured state-space representations, IFAC Proceedings volumes, 45, 16, 191-196, (2012) |
| [15] | Qi, L.; Womersley, R.S., On extreme singular values of matrix valued functions, Journal of convex analysis, 3, 153-166, (1996) · Zbl 0876.49017 |
| [16] | Van den Hof, J., Structural identifiability of linear compartmental systems, IEEE transactions on automatic control, 43, 6, 800-818, (1998) · Zbl 0951.93023 |
| [17] | Verhaegen, M.; Verdult, V., Filtering and system identification: a least squares approach, (2007), Cambridge University Press · Zbl 1149.93001 |
| [18] | Vizer, D.; Mercere, G.; Prot, O.; Laroche, E., \(h_\infty\)-norm-based optimization for the identification of gray-box LTI state-space model parameters, Systems & control letters, 92, 34-41, (2016) · Zbl 1338.93123 |
| [19] | Wernholt, E.; Moberg, S., Nonlinear gray-box identification using local models applied to industrial robots, Automatica, 47, 4, 650-660, (2011) · Zbl 1215.93089 |
| [20] | Xie, L.-L., & Ljung, L. (2002). Estimate physical parameters by black box modeling. In Proceedings of the 21st chinese control conference; Xie, L.-L., & Ljung, L. (2002). Estimate physical parameters by black box modeling. In Proceedings of the 21st chinese control conference |
| [21] | Yu, C., Ljung, L., & Verhaegen, M. (2017). Gray box identification of state-space models using difference of convex programming. In 20th IFAC world congress; Yu, C., Ljung, L., & Verhaegen, M. (2017). Gray box identification of state-space models using difference of convex programming. In 20th IFAC world congress |
| [22] | Yu, C., Verhaegen, M., Kovalsky, S., & Basri, R. (2015). Identification of structured LTI MIMO state-space models. In 2015 54th IEEE conference on decision and control; Yu, C., Verhaegen, M., Kovalsky, S., & Basri, R. (2015). Identification of structured LTI MIMO state-space models. In 2015 54th IEEE conference on decision and control |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
