Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
156 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Backstepping Neural Operators for $2\times 2$ Hyperbolic PDEs (2312.16762v3)

Published 28 Dec 2023 in math.OC, cs.LG, and math.AP

Abstract: Deep neural network approximation of nonlinear operators, commonly referred to as DeepONet, has proven capable of approximating PDE backstepping designs in which a single Goursat-form PDE governs a single feedback gain function. In boundary control of coupled PDEs, coupled Goursat-form PDEs govern two or more gain kernels-a PDE structure unaddressed thus far with DeepONet. In this paper, we explore the subject of approximating systems of gain kernel PDEs for hyperbolic PDE plants by considering a simple counter-convecting $2\times 2$ coupled system in whose control a $2\times 2$ kernel PDE system in Goursat form arises. Engineering applications include oil drilling, the Saint-Venant model of shallow water waves, and the Aw-Rascle-Zhang model of stop-and-go instability in congested traffic flow. We establish the continuity of the mapping from a total of five plant PDE functional coefficients to the kernel PDE solutions, prove the existence of an arbitrarily close DeepONet approximation to the kernel PDEs, and ensure that the DeepONet-approximated gains guarantee stabilization when replacing the exact backstepping gain kernels. Taking into account anti-collocated boundary actuation and sensing, our $L2$-Globally-exponentially stabilizing (GES) approximate gain kernel-based output feedback design implies the deep learning of both the controller's and the observer's gains. Moreover, the encoding of the output-feedback law into DeepONet ensures semi-global practical exponential stability (SG-PES). The DeepONet operator speeds up the computation of the controller gains by multiple orders of magnitude. Its theoretically proven stabilizing capability is demonstrated through simulations.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (47)
  1. O. M. Aamo, “Leak detection, size estimation and localization in pipe flows,” IEEE Transactions on Automatic Control, vol. 61, no. 1, pp. 246–251, 2016.
  2. U. J. F. Aarsnes, F. Di Meglio, S. Evje, and O. M. Aamo, “Control-oriented drift-flux modeling of single and two-phase flow for drilling,” in Dynamic Systems and Control Conference, vol. 46209.   American Society of Mechanical Engineers, 2014, p. V003T37A003.
  3. H. Anfinsen, M. Diagne, O. M. Aamo, and M. Krstic, “An adaptive observer design for n+1𝑛1n+1italic_n + 1 coupled linear hyperbolic PDEs based on swapping,” IEEE Transactions on Automatic Control, vol. 61, no. 12, pp. 3979–3990, 2016.
  4. H. Anfinsen, M. Diagne, O. M. Aamo, and M. Krstić, “Estimation of boundary parameters in general heterodirectional linear hyperbolic systems,” Automatica: A journal of IFAC the International Federation of Automatic Control, no. 79, pp. 185–197, 2017.
  5. J. Auriol and F. Di Meglio, “Two-sided boundary stabilization of heterodirectional linear coupled hyperbolic pdes,” IEEE Transactions on Automatic Control, vol. 63, no. 8, pp. 2421–2436, 2017.
  6. J. Auriol and F. D. Meglio, “Minimum time control of heterodirectional linear coupled hyperbolic PDEs,” Automatica, vol. 71, pp. 300–307, 2015.
  7. G. Bastin and J.-M. Coron, “Further results on boundary feedback stabilisation of 2×\times× 2 hyperbolic systems over a bounded interval,” IFAC Proceedings Volumes, vol. 43, no. 14, pp. 1081–1085, 2010.
  8. ——, “On boundary feedback stabilization of non-uniform linear 2×2222\times 22 × 2 hyperbolic systems over a bounded interval,” Systems & Control Letters, vol. 60, no. 11, pp. 900–906, 2011.
  9. L. Bhan, Y. Shi, and M. Krstic, “Neural operators for bypassing gain and control computations in PDE backstepping,” arXiv preprint arXiv:2302.14265, 2023.
  10. M. Burkhardt, H. Yu, and M. Krstic, “Stop-and-go suppression in two-class congested traffic,” Automatica, vol. 125, p. 109381, 2021.
  11. T. Chen and H. Chen, “Universal approximation to nonlinear operators by neural networks with arbitrary activation functions and its application to dynamical systems,” IEEE Transactions on Neural Networks, vol. 6, no. 4, pp. 911–917, 1995.
  12. J.-M. Coron, B. d’Andréa Novel, and G. Bastin, “A Lyapunov approach to control irrigation canals modeled by saint-venant equations,” in 1999 European control conference (ECC).   IEEE, 1999, pp. 3178–3183.
  13. J.-M. Coron, L. Hu, and G. Olive, “Finite-time boundary stabilization of general linear hyperbolic balance laws via fredholm backstepping transformation,” Automatica, vol. 84, 2017.
  14. J.-M. Coron, L. Hu, G. Olive, and P. Shang, “Boundary stabilization in finite time of one-dimensional linear hyperbolic balance laws with coefficients depending on time and space,” Journal of Differential Equations, vol. 271, pp. 1109–1170, 2021.
  15. C. Curró, D. Fusco, and N. Manganaro, “A reduction procedure for generalized riemann problems with application to nonlinear transmission lines,” Journal of Physics A Mathematical & Theoretical, vol. 44, no. 33, p. 335205, 2011.
  16. B. Deng, Y. Shin, L. Lu, Z. Zhang, and G. E. Karniadakis, “Approximation rates of deeponets for learning operators arising from advection-diffusion equations,” Neural Networks, vol. 153, pp. 411–426, 2022.
  17. F. Di Meglio, G.-O. Kaasa, N. Petit, and V. Alstad, “Model-based control of slugging: advances and challenges,” IFAC Proceedings Volumes, vol. 45, no. 8, pp. 109–115, 2012.
  18. F. Di Meglio, R. Vazquez, and M. Krstic, “Stabilization of a system of n+1𝑛1n+1italic_n + 1 coupled first-order hyperbolic linear PDEs with a single boundary input,” IEEE Transactions on Automatic Control, vol. 58, no. 12, pp. 3097–3111, 2013.
  19. F. Di Meglio, R. Vazquez, M. Krstic, and N. Petit, “Backstepping stabilization of an underactuated 3×\times× 3 linear hyperbolic system of fluid flow equations,” in 2012 American Control Conference (ACC).   IEEE, 2012, pp. 3365–3370.
  20. A. Diagne, G. Bastin, and J.-M. Coron, “Lyapunov exponential stability of 1-D linear hyperbolic systems of balance laws,” Automatica, vol. 48, no. 1, pp. 109–114, 2012.
  21. A. Diagne, M. Diagne, S. Tang, and M. Krstic, “Backstepping stabilization of the linearized Saint-Venant-Exner model,” Automatica, vol. 76, pp. 345–354, 2017.
  22. M. Diagne, S.-X. Tang, A. Diagne, and M. Krstic, “Control of shallow waves of two unmixed fluids by backstepping,” Annual Reviews in Control, vol. 44, pp. 211–225, 2017.
  23. S. Fan, M. Herty, and B. Seibold, “Comparative model accuracy of a data-fitted generalized Aw-Rascle-Zhang model,” Networks and Heterogeneous Media, vol. 9, no. 2, pp. 239–268, 2014.
  24. P. Goatin, “The Aw-Rascle vehicular traffic flow model with phase transitions,” Mathematical and Computer Modelling, 2006.
  25. M. Gugat, M. Dick, and G. Leugering, “Gas flow in fan-shaped networks: Classical solutions and feedback stabilization,” SIAM Journal on Control and Optimization, vol. 49, no. 5, pp. 2101–2117, 2011.
  26. J. D. Halleux, C. Prieur, J. M. Coron, B. dÄndréa Novel, and G. Bastin, “Boundary feedback control in networks of open channels,” Automatica, vol. 39, no. 8, pp. 1365–1376, 2003.
  27. L. Hu, F. Di Meglio, R. Vazquez, and M. Krstic, “Control of homodirectional and general heterodirectional linear coupled hyperbolic PDEs,” IEEE Transactions on Automatic Control, vol. 61, no. 11, pp. 3301–3314, 2016.
  28. L. Hu, R. Vazquez, F. D. Meglio, and M. Krstic, “Boundary exponential stabilization of 1-dimensional inhomogeneous quasi-linear hyperbolic systems,” SIAM Journal on Control and Optimization, vol. 57, no. 2, pp. 963–998, 2019.
  29. I. Karafyllis and M. Krstic, “Spill-free transfer and stabilization of viscous liquid,” IEEE Transactions on Automatic Control, vol. 67, no. 9, pp. 4585–4597, 2022.
  30. I. Karafyllis, F. Vokos, and M. Krstic, “Output-feedback control of viscous liquid–tank system and its numerical approximation,” Automatica, vol. 149, p. 110827, 2023.
  31. M. Krstic, L. Bhan, and Y. Shi, “Neural operators of backstepping controller and observer gain functions for reaction-diffusion PDEs,” arXiv preprint arXiv:2303.10506, 2023.
  32. P.-O. Lamare and N. Bekiaris-Liberis, “Control of 2×2222\times 22 × 2 linear hyperbolic systems: Backstepping-based trajectory generation and PI-based tracking,” Systems & Control Letters, vol. 86, pp. 24–33, 2015.
  33. Z. Li, N. Kovachki, K. Azizzadenesheli, B. Liu, K. Bhattacharya, A. Stuart, and A. Anandkumar, “Fourier neural operator for parametric partial differential equations,” arXiv preprint arXiv:2010.08895, 2020.
  34. ——, “Neural operator: Graph kernel network for partial differential equations,” arXiv preprint arXiv:2003.03485, 2020.
  35. X. Litrico and V. Fromion, “Boundary control of hyperbolic conservation laws using a frequency domain approach,” Automatica, vol. 45, no. 3, pp. 647–656, 2009.
  36. L. Lu, P. Jin, and G. E. Karniadakis, “Deeponet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators,” arXiv preprint arXiv:1910.03193, 2019.
  37. L. Lu, P. Jin, G. Pang, Z. Zhang, and G. E. Karniadakis, “Learning nonlinear operators via deeponet based on the universal approximation theorem of operators,” Nature Machine Intelligence, vol. 3, no. 3, pp. 218–229, 2021.
  38. V. D. S. Martins, M. Rodrigues, and M. Diagne, “A multi-model approach to saint-venant equations: A stability study by lmis,” International Journal of Applied Mathematics and Computer Science, vol. 22, no. 3, pp. 539–550, 2012.
  39. J. Qi, J. Zhang, and M. Krstic, “Neural operators for delay-compensating control of hyperbolic PIDEs,” arXiv preprint arXiv:2307.11436, 2023.
  40. T. S. Rabbani, F. Di Meglio, X. Litrico, and A. M. Bayen, “Feed-forward control of open channel flow using differential flatness,” IEEE Transactions on control systems technology, vol. 18, no. 1, pp. 213–221, 2009.
  41. C. Sagert, F. Di Meglio, M. Krstic, and P. Rouchon, “Backstepping and flatness approaches for stabilization of the stick-slip phenomenon for drilling,” IFAC Proceedings Volumes, vol. 46, no. 2, pp. 779–784, 2013, 5th IFAC Symposium on System Structure and Control. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S1474667016302026
  42. S. Tang and M. Krstic, “Sliding mode control to the stabilization of a linear 2×2222\times 22 × 2 hyperbolic system with boundary input disturbance,” in 2014 American Control Conference, 2014, pp. 1027–1032.
  43. R. Vazquez and M. Krstic, “Bilateral boundary control of one-dimensional first-and second-order pdes using infinite-dimensional backstepping,” in 2016 IEEE 55th Conference on Decision and Control (CDC).   IEEE, 2016, pp. 537–542.
  44. R. Vazquez, M. Krstic, and J.-M. Coron, “Backstepping boundary stabilization and state estimation of a 2×2222\times 22 × 2 linear hyperbolic system,” in 2011 50th IEEE Conference on Decision and Control and European Control Conference, 2011, pp. 4937–4942.
  45. S. Wang, M. Diagne, and M. Krstic, “Deep learning of delay-compensated backstepping for reaction-diffusion PDEs,” arXiv preprint arXiv:2308.10501, 2023.
  46. C.-Z. Xu and G. Sallet, “Exponential stability and transfer functions of processes governed by symmetric hyperbolic systems,” ESAIM: Control, Optimisation and Calculus of Variations, vol. 7, pp. 421–442, 2002.
  47. H. Yu and M. Krstic, “Traffic congestion control for aw–rascle–zhang model,” Automatica, vol. 100, pp. 38–51, 2019.
Citations (1)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now