Recursive Approximation of Complex Behaviours With IoT-Data Imperfections

Korkut Bekiroglu; Seshadhri Srinivasan; Ethan Png; Rong Su; Constantino Lagoa

doi:10.1109/JAS.2020.1003126

Volume 7 Issue 3

Apr. 2020

IEEE/CAA Journal of Automatica Sinica

JCR Impact Factor: 15.3, Top 1 (SCI Q1)

CiteScore: 23.5, Top 2% (Q1)
Google Scholar h5-index: 77， TOP 5

Turn off MathJax

Article Contents

Article Navigation > IEEE/CAA Journal of Automatica Sinica > 2020 > 7(3): 656-667

Korkut Bekiroglu, Seshadhri Srinivasan, Ethan Png, Rong Su and Constantino Lagoa, "Recursive Approximation of Complex Behaviours With IoT-Data Imperfections," IEEE/CAA J. Autom. Sinica, vol. 7, no. 3, pp. 656-667, May 2020. doi: 10.1109/JAS.2020.1003126

Citation:

Korkut Bekiroglu, Seshadhri Srinivasan, Ethan Png, Rong Su and Constantino Lagoa, "Recursive Approximation of Complex Behaviours With IoT-Data Imperfections," IEEE/CAA J. Autom. Sinica, vol. 7, no. 3, pp. 656-667, May 2020. doi: 10.1109/JAS.2020.1003126

Citation:

PDF( 1218 KB)

Recursive Approximation of Complex Behaviours With IoT-Data Imperfections

doi: 10.1109/JAS.2020.1003126

Funds: This work was supported by the Building and Construction Authority through the NRF GBIC Program (NRF2015ENC-GBICRD001-057)

More Information

Author Bio:
Korkut Bekiroglu received the B.S. degree from the Department of Electrical and Electronics Engineering of Black Sea Technical University, Turkey, in 2005, the M.S. degree from the Northeastern University, USA, in 2010, and the Ph.D. degree from Penn State University, USA and University Park, USA, in 2015, both in the Department of Electrical Engineering. After graduation, in August 2015, he was appointed as a Postdoctoral Researcher in the Methodology Center at Pennsylvania State University, USA until 2016. He is currently an Assistant Professor of electrical engineering technology at the College of Engineering, SUNY Polytechnic Institute, USA. Before joining SUNY Polytechnic Institute, Dr. Bekiroglu was a Research Fellow working on the Green Building Innovation Cluster (GBIC) project at Nanyang Technological University, Singapore. His research interests include system identification, model predictive control, optimization, robust optimization, discrete event dynamical systems, mechatronics, building automation, signal processing, and robust control

Seshadhri Srinivasan is currently a Postdoctoral at Berkeley Education Alliance for Research in Singapore (BEARS), Singapore, working on building signal processing, smart grid data imputation, fault-diagnosis of buildings, occupancy-based energy control of HVAC systems, multi-level optimization of Building Energy Management, Singapore. He is part of the SinBerBEST1, SinBerBEST2, and GBIC project. Dr. Srinivasan serves the IEEE CSS as a Member of Standardization Committee. Currently, he is working with Prof. Kameshwar Poolla, Dept. of Mechanical Engineering, UC Berkeley on multi-level optimal control and deep learning. His research interests include mathematical optimization, cyber-physical systems, and industrial automation. He received the Ph.D. degree with National Institute of Technology-Tiruchirappalli, India on networked control systems and has been part of various projects during his different stints as a Researcher and as an Academician

Ethan Png received the M.Sc. computation degree from The University of Manchester Institute of Science and Technology (UMIST), UK, in 2004. His research interests include visible light communication, IoT, intelligent self-driving bicycles, artificial neural networks, medical robots, IC design, and condensed matter physics (semiconductors)

Rong Su received the bachelor of engineering degree from University of Science and Technology of China, and master of applied science degree, and Ph.D. degree from University of Toronto, Canada, respectively. He was affiliated with University of Waterloo, Canada, and Technical University of Eindhoven, Netherlands before he joined Nanyang Technological University, Singapore in 2010. Dr. Su’s research interests include multi-agent systems, discrete-event system theory, model-based fault diagnosis, operation planning and scheduling with applications in flexible manufacturing, intelligent transportation, human-robot interface, power management, and green building. In the aforementioned areas he has more than 110 journal and conference publications and 2 patents, and has been involved in several projects sponsored by Singapore National Research Foundation (NRF), Singapore Agency of Science, Technology and Research (A*STAR), Singapore Ministry of Education (MoE), Singapore Civil Aviation Authority (CAAS), and Singapore Economic Development Board (EDB). Prof. Su is a Senior Member of IEEE, and an Associate Editor for Transactions of the Institute of Measurement and Control, Journal of Control and Decision, and Journal of Discrete Event Dynamic Systems: Theory and Applications. He is also the Chair of the Technical Committee on Smart Cities in the IEEE Control Systems Society, and a Board Member of the School of EEE at NTU

Constantino Lagoa received the B.S. and M.S. degrees from the Instituto Superior Tecnico, Technical University of Lisbon, Portugal, in 1991 and 1994, respectively, and the Ph.D. degree from the University of Wisconsin at Madison, USA, in 1998. He joined the Electrical Engineering Department of Pennsylvania State University, University Park, USA, in August 1998, where he currently holds the position of Professor. He has a wide range of research interests including robust optimization and control, chance constrained optimization, controller design under risk specifications, system identification, control of computer networks, and discrete event dynamical systems. Dr. Lagoa is currently an Associate Editor of Automatica and was an Associate Editor for IEEE Transactions on Automatic Control and IEEE Transactions on Control Systems Technology. He is a Member of the IFAC Technical Committee on Robust Control and has served as a Member of the International Program Committee of several conferences
Corresponding author: K. Bekiroglu is with the College of Engineering, SUNY Polytechnic Institute, Utica 13503, NY, USA (e-mail: korkut.bekiroglu@sunypoly.edu)
¹ The imperfect measurements can result from a variety of reasons which include hardware aspects such as sensor failures, noise, faults, etc., or network induced imperfections such as latencies, packet dropout, quantization, and others.
² The set of poles can be defined based on priori such as rise time, overshoot, etc. Only the stable poles are chosen for this paper, reasonable one for HVAC systems, but can include unstable poles.
³ Discrete linear systems with distinct poles have two facts: i) the same poles appear in the impulse response and the initial condition response; ii) the number of distinct poles used in these two is equal to the order of the system. Hence, sparsifying the impulse response of the system (i.e., identifying the system of lowest order) is equivalent to minimizing the number of unique poles that are used in the total response, i.e., the response that includes both the response to initial conditions and the zero state response to the input.
⁴ Justification of Hankel norm normalization can be found in [16], [26].
Received Date: 2020-01-04
Revised Date: 2020-02-21
Accepted Date: 2020-03-05

Available Online: 2020-03-24

Abstract

Abstract

This paper presents an approach to recursively estimate the simplest linear model that approximates the time-varying local behaviors from imperfect (noisy and incomplete) measurements in the internet of things (IoT) based distributed decision-making problems. We first show that the problem of finding the lowest order model for a multi-input single-output system is a cardinality (ℓ₀) optimization problem, known to be NP-hard. To solve the problem a simpler approach is proposed which uses the recently developed atomic norm concept and the modified Frank-Wolfe (mFW) algorithm is introduced. Further, the paper computes the minimum data-rate required for computing the models with imperfect measurements. The proposed approach is illustrated on a building heating, ventilation, and air-conditioning (HVAC) control system that aims at optimizing energy consumption in commercial buildings using IoT devices in a distributed manner. The HVAC control application requires recursive thermal dynamical model updates due to frequently changing conditions and non-linear dynamics. We show that the method proposed in this paper can approximate such complex dynamics on single-board computers interfaced to sensors using unreliable communication channels. Real-time experiments on HVAC systems and simulation studies are used to illustrate the proposed method.
- Adaptability,
- distributed decision systems,
- imperfect measurements,
- internet of things (IoT),
- low order model identification

FullText(HTML)

¹ The imperfect measurements can result from a variety of reasons which include hardware aspects such as sensor failures, noise, faults, etc., or network induced imperfections such as latencies, packet dropout, quantization, and others.
² The set of poles can be defined based on priori such as rise time, overshoot, etc. Only the stable poles are chosen for this paper, reasonable one for HVAC systems, but can include unstable poles.
³ Discrete linear systems with distinct poles have two facts: i) the same poles appear in the impulse response and the initial condition response; ii) the number of distinct poles used in these two is equal to the order of the system. Hence, sparsifying the impulse response of the system (i.e., identifying the system of lowest order) is equivalent to minimizing the number of unique poles that are used in the total response, i.e., the response that includes both the response to initial conditions and the zero state response to the input.
⁴ Justification of Hankel norm normalization can be found in [16], [26].

References(26)

References

[1]	J. Liu and H. Xiao, “Distributed output feedback model predictive control for a team of coupled linear subsystems,” IET Control Theory & Applications, vol. 11, no. 11, pp. 1807–1812, 2017. [Online]. Available: https://digital-library.theiet.org/content/journals/10.1049/iet-cta.2016.0751
[2]	L. D. Xu, W. He, and S. Li, “Internet of things in industries: a survey,” IEEE Trans. Industrial Informatics, vol. 10, no. 4, pp. 2233–2243, 2014. doi: 10.1109/TII.2014.2300753
[3]	H. Dibowski, J. Ploennigs, and K. Kabitzsch, “Automated design of building automation systems,” IEEE Trans. Industrial Electronics, vol. 57, no. 11, pp. 3606–3613, 2010. doi: 10.1109/TIE.2009.2032209
[4]	Y.-L. Chang and C.-C. Tsai, “Adaptive generalised predictive temperature control for air conditioning systems,” IET Control Theory & Applications, vol. 5, no. 6, pp. 813–822, 2011. [Online]. Available: https://digital-library.theiet.org/content/journals/10.1049/iet-cta.2010.0085
[5]	D. Li, Y. Zhou, G. Hu, and C. J. Spanos, “Handling incomplete sensor measurements in fault detection and diagnosis for building hvac systems,” IEEE Trans. Automation Science and Engineering, pp. 1–14, 2019. doi: 10.1109/TASE.2019.2948101
[6]	P. Van Overschee and B. De Moor, “N4SID: subspace algorithms for the identification of combined deterministic-stochastic systems,” Automatica, vol. 30, no. 1, pp. 75–93, 1994. doi: 10.1016/0005-1098(94)90230-5
[7]	G. Mercère, M. Lovera, L. Bako, and S. Lecoeuche, “Recursive subspace identification of Hammerstein models based on least squares support vector machines,” IET Control Theory & Applications, vol. 3, no. 9, pp. 1209–1216, 2009. [Online]. Available: https://digital-library.theiet.org/content/journals/10.1049/iet-cta.2008.0339
[8]	B. Recht, M. Fazel, and P. A. Parrilo, “Guaranteed minimum-rank solutions of linear matrix equations via nuclear norm minimization,” SIAM Review, vol. 52, no. 3, pp. 471–501, 2010. doi: 10.1137/070697835
[9]	M. Sznaier, M. Ayazoglu, and T. Inanc, “Fast structured nuclear norm minimization with applications to set membership systems identification,” IEEE Trans. Automatic Control, vol. 59, no. 10, pp. 2837–2842, 2014. [Online]. Available: http://ieeexplore.ieee.org/articleDetails.jsp?arnumber=6778049
[10]	Z. Liu and L. Vandenberghe, “Interior-point method for nuclear norm approximation with application to system identification,” SIAM J. Matrix Analysis and Applications, vol. 31, no. 3, pp. 1235–1256, 2009.
[11]	K. Bekiroglu, S. Srinivasan, E. Png, R. Su, K. Poolla, and C. Lagoa, “An internet of things compliant model identification methodology for smart buildings,” in Proc. IEEE 56th Annual Conf. Decision and Control, 2017, pp. 4440–4445.
[12]	P. Van Dooren, K. Gallivan, and P.-A. Absil, “H₂-optimal model reduction with higher-order poles,” SIAM J. Matrix Analysis and Applications, vol. 31, pp. 2738, 2010. doi: 10.1137/080731591
[13]	B. Yilmaz, K. Bekiroglu, C. Lagoa, and M. Sznaier, “A randomized algorithm for parsimonious model identification,” IEEE Trans. Automatic Control, vol. 63, no. 2, pp. 532–539, 2018. doi: 10.1109/TAC.2017.2723959
[14]	D. L. Donoho, “Compressed sensing,” IEEE Trans. Information Theory, vol. 52, pp. 1289–1306, 2006. doi: 10.1109/TIT.2006.871582
[15]	V. Chandrasekaran, B. Recht, P. A. Parrilo, and A. S. Willsky, “The convex geometry of linear inverse problems,” Foundations of Computational Mathematics, vol. 12, no. 6, pp. 805–849, 2012. doi: 10.1007/s10208-012-9135-7
[16]	P. Shah, B. N. Bhaskar, G. Tang, and B. Recht, “Linear system identification via atomic norm regularization,” in Proc. IEEE 51st Conf. Decision and Control, 2012, pp. 6265–6270.
[17]	A. Tewari, P. K. Ravikumar, and I. S. Dhillon, “Greedy algorithms for structurally constrained high dimensional problems,” Advances in Neural Information Processing Systems, pp. 882–890, 2011.
[18]	J.-F. Cai, E. J. Candès, and Z. Shen, “A singular value thresholding algorithm for matrix completion,” SIAM J. Optimization, vol. 20, no. 4, pp. 1956–1982, 2010.
[19]	R. K. Lim and R. W. Longman, “State-space system identification with identified Hankel matrix,” Department of Mechanical and Aerospace Engineering Technical Report No. 3045, no. 3045, pp. 1–36, 1998.
[20]	I. Boz, K. Bekiroglu, M. Solaimanian, P. Tavassoti-Kheiry, C. Lagoa, T.-k. Pexbouhan, and C. Lagoa, “Validation of model order assumption and noise reduction method for the impact resonance testing of asphalt concrete,” J. Nondestructive Evaluation, vol. 36, no. 3, pp. 58, 2017. doi: 10.1007/s10921-017-0436-2
[21]	L. Ljung, “System identification,” in Signal Analysis and Prediction. Springer, 1998, pp. 163–173.
[22]	S. Baldi, S. Yuan, P. Endel, and O. Holub, “Dual estimation: constructing building energy models from data sampled at low rate,” Applied Energy, vol. 169, pp. 81–92, 2016.
[23]	E. Png, S. Srinivasan, K. Bekiroglu, J. Chaoyang, R. Su, and K. Poolla, “An internet of things upgrade for smart and scalable heating, ventilation and air-conditioning control in commercial buildings,” Applied Energy, vol. 239, pp. 408–424, 2019.
[24]	N. Radhakrishnan, S. Srinivasan, R. Su, and K. Poolla, “Learning based hierarchical distributed HVAC scheduling with operational constraints,” IEEE Trans. Control Systems Technology, vol. 26, no. 5, pp. 1892–1900, 2017. doi: 10.1109/TCST.2017.2728004
[25]	A. Aswani, N. Master, J. Taneja, V. Smith, A. Krioukov, D. Culler, and C. Tomlin, “Identifying models of HVAC systems using semiparametric regression,” in Proc. American Control Conf., 2012, pp. 3675–3680.
[26]	K. Bekiroglu, B. Yilmaz, C. Lagoa, and M. Sznaier, “Parsimonious model identification via atomic norm minimization,” in Proc. European Control Conf., pp. 2392–2397, 2014.

Supplements(0)

Cited By

Proportional views

Proportional views

通讯作者: 陈斌, bchen63@163.com

1.
沈阳化工大学材料科学与工程学院沈阳 110142

Figures(4) / Tables(3)

Get Citation

PDF

XML

Article Metrics

Article views (1530) PDF downloads(88)

Highlights

Parsimonious recursive system update of complex non-linear and time-varying behaviors.
Identifiability minimum data requirement for low order systems.
Sparse system identification in IoT setting.

Recursive Approximation of Complex Behaviours With IoT-Data Imperfections

doi: 10.1109/JAS.2020.1003126

Abstract

References

Proportional views

Catalog

通讯作者: 陈斌, bchen63@163.com

Article Metrics

Highlights

Export File

Citation

Format

Content