pith. sign in

arxiv: 2508.14840 · v3 · pith:L4J4HJZNnew · submitted 2025-08-20 · 🧮 math.AP · cs.SY· eess.SY

A State-Space Representation of Coupled Linear Multivariate PDEs and Stability Analysis using SDP

Declan S. Jagt , Matthew M. Peet This is my paper

Pith reviewed 2026-05-21 22:39 UTC · model grok-4.3

classification 🧮 math.AP cs.SYeess.SY
keywords partial differential equationsmultivariate PDEspartial integral equationsstability analysissemidefinite programmingboundary conditionsoperator inversion
0
0 comments X

The pith

If each univariate domain is well-posed, a simple consistency condition on lifted boundaries is necessary and sufficient for the inverse of the multivariate spatial operator D^α to exist.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper shows how to extend the conversion of single-variable PDEs into partial integral equations to the case of multiple spatial variables. It does so by treating the multivariate domain as the intersection of domains lifted from each separate variable. Under the assumption that each univariate domain meets a well-posedness requirement, the authors derive a consistency condition on the boundary data that guarantees the existence of an inverse to the full spatial differential operator. This inverse belongs to a *-algebra of partial integral operators whose kernels are polynomial and semi-separable. The resulting representation turns any linear multivariate PDE into an equivalent PIE, which in turn admits a convex stability test based on semidefinite programming.

Core claim

Representing the multivariate domain as the intersection of lifted univariate domains permits an inductive argument that a readily verified consistency condition is necessary and sufficient for the existence of an inverse to the multivariate spatial differential operator D^α = ∂_{s1}^{α1} ⋯ ∂_{sN}^{αN}. This inverse lies in a *-algebra of partial integral operators with polynomial semi-separable kernels. Consequently, any suitably well-posed linear multivariate PDE admits an equivalent partial integral equation representation, and stability of that PIE can be tested by a semidefinite program that parameterizes positive elements of the algebra.

What carries the argument

The consistency condition on the boundary traces of the lifted univariate domains, which guarantees that the inverse of D^α exists and remains inside the *-algebra of partial integral operators.

If this is right

  • Any linear multivariate PDE meeting the well-posedness hypotheses can be rewritten exactly as a partial integral equation whose coefficients are elements of the PI algebra.
  • Stability of the PDE reduces to a convex semidefinite program that searches for a positive matrix parameterizing a positive element of the PI algebra.
  • The same representation supplies explicit decay-rate bounds once the SDP is solved, as demonstrated for two-dimensional heat, wave, and plate equations.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same lifting-and-consistency technique might be applied to derive state-space forms for other classes of linear operators whose domains admit product structure.
  • Control synthesis problems for multivariate PDEs could be posed directly in the PIE algebra once the representation is available, mirroring the univariate case.
  • The computational cost of the SDP test will grow with the number of spatial variables and the degree of the polynomial kernels, suggesting a natural limit on the dimension for which the method remains practical without further approximation.

Load-bearing premise

The multivariate domain can be expressed as the intersection of domains obtained by lifting each univariate domain separately, and every univariate domain already satisfies the well-posedness condition needed for the inductive construction.

What would settle it

A concrete set of boundary conditions on a rectangle or other product domain where every univariate restriction is well-posed and the consistency condition holds, yet the operator D^α fails to possess a bounded inverse on the full domain.

read the original abstract

Physical processes evolving in both time and space are often modeled using Partial Differential Equations (PDEs). Recently, it has been shown how stability analysis and control of coupled PDEs in a single spatial variable can be more conveniently performed using an equivalent Partial Integral Equation (PIE) representation. The construction of this PIE representation is based on an analytic expression for the inverse of the spatial differential operator, $\partial_s^{d}$, on the domain defined by boundary conditions. In this paper, we show how this univariate representation may be extended inductively to multiple spatial variables by representing the domain as the intersection of lifted univariate domains. Specifically, we show that if each univariate domain is well-posed, then there exists a readily verified consistency condition which is necessary and sufficient for existence of an inverse to the multivariate spatial differential operator, $D^\alpha=\partial_{s_1}^{\alpha_1}\cdots\partial_{s_N}^{\alpha_N}$, on the PDE domain. Furthermore, we show that this inverse is an element of a $*$-algebra of Partial Integral (PI) operators defined by polynomial semi-separable kernels. Based on this operator algebra, we show that the evolution of any suitably well-posed linear multivariate PDE may be described by a PIE, parameterized by elements of the PI algebra. A convex computational test for PDE stability is then proposed using a positive matrix parameterization of positive PI operators, and software (PIETOOLS) is provided which automates the process of representation and stability analysis of such PDEs. This software is used to analyze stability of 2D heat, wave, and plate equations, obtaining accurate bounds on the rate of decay.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript extends the univariate PIE representation of linear PDEs to the multivariate setting. It represents the PDE domain as the intersection of lifted univariate domains, shows that well-posedness of each univariate domain plus a consistency condition is necessary and sufficient for the existence of an inverse to the multivariate operator D^α, proves that this inverse lies in a *-algebra of partial integral operators with polynomial semi-separable kernels, reformulates the PDE evolution as a PIE, and derives an SDP-based convex test for stability that is automated in PIETOOLS. Numerical results are given for 2D heat, wave, and plate equations.

Significance. If the domain-representation and consistency-condition claims are valid, the work supplies a systematic route from multivariate PDEs to a state-space PIE form whose stability can be checked by semidefinite programming. The accompanying software and concrete decay-rate bounds on standard examples constitute a practical contribution that builds directly on prior univariate PIE theory.

major comments (2)
  1. [Section 3] Section 3 (inductive construction): the central claim that the multivariate domain equals the intersection of lifted univariate domains is load-bearing for both necessity and sufficiency of the consistency condition. For non-product or non-rectangular domains, or when boundary conditions couple the variables non-separably, the lifted intersection can omit cross-boundary constraints; in such cases the consistency condition may hold while D^α still fails to be invertible on the true domain because its kernel on the lifted space is larger.
  2. [Section 3 or 4] Theorem/Proposition establishing necessity and sufficiency (presumably in Section 3 or 4): the manuscript should state explicitly whether the consistency condition is formulated only for rectangular product domains or whether additional trace-matching or commutativity requirements are imposed to handle general boundary compatibilities; without this clarification the necessity claim is not fully supported.
minor comments (2)
  1. [Abstract] Abstract: the phrase 'readily verified consistency condition' would be clearer if a short description or explicit algebraic form of the condition were supplied.
  2. [Numerical examples] Numerical section: the precise rectangular domains, boundary conditions, and polynomial degrees used for the 2D heat/wave/plate examples should be listed so that the reported decay bounds can be reproduced from the PIETOOLS code.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful reading of the manuscript and for highlighting important points regarding the scope of the domain representation and the consistency condition. We address each major comment below and have revised the manuscript to clarify the assumptions under which our results hold.

read point-by-point responses

  1. Referee: [Section 3] Section 3 (inductive construction): the central claim that the multivariate domain equals the intersection of lifted univariate domains is load-bearing for both necessity and sufficiency of the consistency condition. For non-product or non-rectangular domains, or when boundary conditions couple the variables non-separably, the lifted intersection can omit cross-boundary constraints; in such cases the consistency condition may hold while D^α still fails to be invertible on the true domain because its kernel on the lifted space is larger.

    Authors: We agree that the inductive construction and the necessity/sufficiency of the consistency condition rely on the multivariate domain being exactly the intersection of the lifted univariate domains. This equivalence holds when the spatial domain is a rectangular product of intervals and the boundary conditions are separable with respect to each variable. For non-rectangular domains or non-separable boundary conditions, cross-boundary constraints may indeed be omitted, and the consistency condition alone would not guarantee invertibility on the true domain. We have revised Section 3 to explicitly restrict the claims to rectangular product domains with separable boundary conditions and have added a remark noting that extensions to more general domains lie outside the present scope. revision: yes

  2. Referee: [Section 3 or 4] Theorem/Proposition establishing necessity and sufficiency (presumably in Section 3 or 4): the manuscript should state explicitly whether the consistency condition is formulated only for rectangular product domains or whether additional trace-matching or commutativity requirements are imposed to handle general boundary compatibilities; without this clarification the necessity claim is not fully supported.

    Authors: The consistency condition is formulated exclusively for rectangular product domains; no additional trace-matching or commutativity requirements are imposed beyond the well-posedness of each univariate domain. We have revised the statement of the relevant theorem (now clearly labeled in Section 3) to make this domain assumption explicit and have included a short discussion of why the necessity proof does not extend without further conditions to non-product or non-rectangular settings. revision: yes

Circularity Check

0 steps flagged

Minor self-citation to univariate PIE foundation; multivariate consistency condition and inductive extension are independently derived

full rationale

The paper's central result extends the univariate inverse construction (cited as 'recently shown') to the multivariate case via an inductive representation of the domain as the intersection of lifted univariate domains plus a consistency condition on the inverses. This consistency condition is presented as a new, readily verified necessary-and-sufficient criterion rather than being fitted or defined in terms of the target inverse. The univariate base is external to the present derivation (even if self-cited), and the multivariate claims do not reduce to a renaming, self-definition, or parameter fit within this paper. No load-bearing step collapses by construction to its own inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central construction rests on well-posed univariate domains and the algebraic closure of the PI operator set; no free parameters or new physical entities are introduced.

axioms (1)
  • domain assumption Each univariate domain is well-posed.
    Invoked as the base case for the inductive multivariate construction in the abstract.

pith-pipeline@v0.9.0 · 5838 in / 1134 out tokens · 35798 ms · 2026-05-21T22:39:45.517149+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Verifying Well-Posedness of Linear PDEs using Convex Optimization

    math.AP 2026-04 unverdicted novelty 6.0

    Reformulating Lumer-Phillips conditions via PIE representation turns well-posedness verification for linear PDEs into convex optimization problems that bound exponential growth rates.

Reference graph

Works this paper leans on

62 extracted references · 62 canonical work pages · cited by 1 Pith paper

  1. [1]

    U. G. Abdulla. Generalized Newton–Leibniz formula and th e embedding of the Sobolev functions with dominating mixed smoothness into H¨ older spaces.AIMS Mathematics, 8(9):20700–20717, 2023

    work page 2023

  2. [2]

    Ahmadi, G

    M. Ahmadi, G. V almorbida, and A. Papachristodoulou. Diss ipation inequalities for the analysis of a class of PDEs. Automatica, 66:163–171, 2016

    work page 2016

  3. [3]

    Ahmadi, G

    M. Ahmadi, G. V almorbida, D. Gayme, and A. Papachristodou lou. A framework for input–output analysis of wall-bounded shear flows. Journal of Fluid Mechanics, 873:742–785, 2019

    work page 2019

  4. [4]

    M. S. Andersen, J. Dahl, and L. V andenberghe. Cvxopt documentation, 2021

    work page 2021

  5. [5]

    M. ApS. The MOSEK optimization toolbox for MATLAB manual. V ersion 1 0.0.34, 2023. URL http://docs.mosek.com/latest/toolbox/index.html

    work page 2023

  6. [6]

    Bensoussan, G

    A. Bensoussan, G. Da Prato, M. C. Delfour, and S. K. Mitter. Representation and control of infinite dimensional systems, volume 2. Springer, 2007

    work page 2007

  7. [7]

    S. I. Chernyshenko, P . Goulart, D. Huang, and A. Papachris todoulou. Polynomial sum of squares in fluid dynamics: a review with a look ahead. Philosophical Transactions of the Royal Society A: Mathema tical, Physical and Engineering Sciences , 372(2020):20130350, 2014

    work page 2020

  8. [8]

    R. F. Curtain and H. Zwart. An Introduction to Infinite-Dimensional Linear Systems The ory. Springer-V erlag, Berlin, Heidelberg, 1995. ISBN 0387944753

    work page 1995

  9. [9]

    L. C. Evans. Partial differential equations, volume 19. American Mathematical Society, 2022

    work page 2022

  10. [10]

    Fantuzzi and I

    G. Fantuzzi and I. Tobasco. Sharpness and non-sharpness of occupation measure bounds for integral variational problems. arXiv eprint:2207.13570, 2022

    work page arXiv 2022

  11. [11]

    Fridman, S

    E. Fridman, S. Nicaise, and J. V alein. Stabilization of s econd order evolution equations with unbounded feedback with time-dependent delay. SIAM Journal on Control and Optimization , 48(8):5028–5052, 2010

    work page 2010

  12. [12]

    Fuentes, D

    F. Fuentes, D. Goluskin, and S. Chernyshenko. Global sta bility of fluid flows despite transient growth of energy. Physical Review Letters, 128(20):204502, 2022

    work page 2022

  13. [13]

    D. F. Gayme, B. J. McKeon, B. Bamieh, A. Papachristodoulo u, and J. C. Doyle. Amplification and nonlinear mechanisms in plane Couette flow. Physics of Fluids, 23(6):065108, 2011

    work page 2011

  14. [14]

    Gohberg, S

    I. Gohberg, S. Goldberg, and M. A. Kaashoek. Classes of Linear Operators , volume 1. Springer Basel AG,

    work page

  15. [15]

    P . J. Goulart and S. Chernyshenko. Global stability anal ysis of fluid flows using sum-of-squares. Physica D: Nonlinear Phenomena, 241(6):692–704, 2012

    work page 2012

  16. [16]

    Hajjej and H

    Z. Hajjej and H. Zhang. Exponential stability of a Kirchh off plate equation with structural damping and internal time delay. Symmetry, 16(11):1427, 2024

    work page 2024

  17. [17]

    Henrion, M

    D. Henrion, M. Infusino, S. Kuhlmann, and V . Vinnikov. In finite-dimensional moment-SOS hierarchy for nonlinear partial differential equations. arXiv eprint:2305.18768, 2023

    work page arXiv 2023

  18. [18]

    D. S. Jagt and M. M. Peet. A PIE representation of coupled l inear 2D PDEs and stability analysis using LPIs. In 2022 American Control Conference (ACC), pages 1659–1666. IEEE, 2022. REPRESENTA TION OFND PDES 35

    work page 2022

  19. [19]

    M. R. Jovanovi´ c. From bypass transition to flow control and data-driven turbulence modeling: an input–output viewpoint. Annual Review of Fluid Mechanics , 53:311–345, 2021

    work page 2021

  20. [20]

    M. R. Jovanovi´ c and B. Bamieh. Componentwise energy amp lification in channel flows. Journal of Fluid Mechanics, 534:145–183, 2005

    work page 2005

  21. [21]

    Kang and E

    W. Kang and E. Fridman. Sampled-data control of 2-D Kuram oto-Sivashinsky equation. IEEE Transactions on Automatic Control, 67(3):1314–1326, 2021

    work page 2021

  22. [22]

    W. Kang, E. Fridman, and S. Zhuk. Sampled-data observer f or 2D Navier-Stokes equation. In 2019 IEEE 58th Conference on Decision and Control (CDC) , pages 1201–1206, 2019

    work page 2019

  23. [23]

    Korda, D

    M. Korda, D. Henrion, and J. B. Lasserre. Moments and conv ex optimization for analysis and control of nonlinear PDEs. In Handbook of Numerical Analysis , volume 23, pages 339–366. Elsevier, 2022

    work page 2022

  24. [24]

    Krstic and A

    M. Krstic and A. Smyshlyaev. Backstepping boundary cont rol – a tutorial. In 2007 American Control Conference, pages 870–875. IEEE, 2007

    work page 2007

  25. [25]

    Lasiecka and R

    I. Lasiecka and R. Triggiani. Exact controllability and uniform stabilization of Kirchoff plates with boundary control only on ∆w— Σ and homogeneous boundary displacement. Journal of Differential Equations , 93(1): 62–101, 1991

    work page 1991

  26. [26]

    Lasiecka and R

    I. Lasiecka and R. Triggiani. Control theory for partial differential equations: Contin uous and approximation theories, volume 1. Cambridge University Press, 2000

    work page 2000

  27. [27]

    B. K. Lieu, M. R. Jovanovi´ c, and S. Kumar. Worst-case amp lification of disturbances in inertialess Couette flow of viscoelastic fluids. Journal of Fluid Mechanics, 723:232–263, 2013

    work page 2013

  28. [28]

    Liu and D

    C. Liu and D. F. Gayme. Input-output inspired method for p ermissible perturbation amplitude of transitional wall-bounded shear flows. Physical Review E, 102(6):063108, 2020

    work page 2020

  29. [29]

    Liu and D

    C. Liu and D. F. Gayme. Structured input–output analysis of transitional wall-bounded flows. Journal of Fluid Mechanics, 927:A25, 2021

    work page 2021

  30. [30]

    Liu and S

    Z. Liu and S. Zheng. Exponential stability of the Kirchho ff plate with thermal or viscoelastic damping. Quarterly of Applied Mathematics , 55(3):551–564, 1997

    work page 1997

  31. [31]

    J. Lofberg. Y ALMIP: A toolbox for modeling and optimizat ion in MA TLAB. In 2004 IEEE international conference on robotics and automation (IEEE Cat. No. 04CH37 508), pages 284–289, 2004

    work page 2004

  32. [32]

    Magron and C

    V . Magron and C. Prieur. Optimal control of linear pdes us ing occupation measures and SDP relaxations. IMA Journal of Mathematical Control and Information , 37(1):159–174, 2020

    work page 2020

  33. [33]

    Mahawattege and R

    R. Mahawattege and R. Triggiani. Fluid–structure inter action with Kelvin–V oigt damping: Analyticity, spectral analysis, exponential decay. Applied Mathematics & Optimization , 84(Suppl 2):1821–1863, 2021

    work page 2021

  34. [34]

    Meyer and M

    E. Meyer and M. M. Peet. Stability analysis of parabolic l inear PDEs with two spatial dimensions using lyapunov method and SOS. In 2015 54th IEEE Conference on Decision and Control (CDC), pages 1884–1890, 2015

    work page 2015

  35. [35]

    A. A. Moghadam, I. Aksikas, S. Dubljevic, and J. F. Forbes . Boundary optimal (LQ) control of coupled hyperbolic PDEs and ODEs. Automatica, 49(2):526–533, 2013

    work page 2013

  36. [36]

    Papachristodoulou and M

    A. Papachristodoulou and M. M. Peet. On the analysis of sy stems described by classes of partial differential equations. In 2006 45th IEEE Conference on Decision and Control (CDC) , pages 747–752. IEEE, 2006

    work page 2006

  37. [37]

    Papachristodoulou, J

    A. Papachristodoulou, J. Anderson, G. V almorbida, S. Pr ajna, P . Seiler, P . Parrilo, M. M. Peet, and D. Jagt. SOSTOOLS version 4.00 sum of squares optimization toolbox f or MA TLAB, 2021

    work page 2021

  38. [38]

    P . A. Parrilo. Structured semidefinite programs and semialgebraic geomet ry methods in robustness and optimization. California Institute of Technology, 2000

    work page 2000

  39. [39]

    Riesmeier and F

    M. Riesmeier and F. Woittennek. Late lumping of transfor mation-based feedback laws for boundary control systems. arXiv eprint:2211.01238, 2022

    work page arXiv 2022

  40. [40]

    Shivakumar

    S. Shivakumar. Analysis, Estimation, and Control of Partial Differential Equations Using Partial Integral Equation Representation. PhD thesis, Arizona State University, 2024

    work page 2024

  41. [41]

    Shivakumar, A

    S. Shivakumar, A. Das, S. Weiland, and M. Peet. Extension of the partial integral equation representation to GPDE input-output systems. arxiv eprint:2205.03735, 2022

    work page arXiv 2022

  42. [42]

    Shivakumar, A

    S. Shivakumar, A. Das, S. Weiland, and M. Peet. Extension of the partial integral equation representation to GPDE input-output systems. IEEE Transactions on Automatic Control, pages 1–16, 2024. 36 D. S. JAGT AND M. M. PEET

    work page 2024

  43. [43]

    Shivakumar, D

    S. Shivakumar, D. Jagt, D. Braghini, A. Das, Y . Peet, and M . Peet. PIETOOLS 2024: User manual, 2025

    work page 2024

  44. [44]

    Smyshlyaev and M

    A. Smyshlyaev and M. Krstic. Backstepping observers for a class of parabolic PDEs. Systems & Control Letters, 54(7):613–625, 2005

    work page 2005

  45. [45]

    J. F. Sturm. Using sedumi 1.02, a MA TLAB toolbox for optim ization over symmetric cones. Optimization methods and software, 11(1-4):625–653, 1999

    work page 1999

  46. [46]

    V almorbida, M

    G. V almorbida, M. Ahmadi, and A. Papachristodoulou. Con vex solutions to integral inequalities in two- dimensional domains. In 2015 54th IEEE Conference on Decision and Control (CDC) , pages 7268–7273, 2015

    work page 2015

  47. [47]

    V azquez and M

    R. V azquez and M. Krstic. Boundary control and estimatio n of reaction–diffusion equations on the sphere under revolution symmetry conditions. International Journal of Control, 92(1):2–11, 2019

    work page 2019

  48. [48]

    V´ azquez, E

    R. V´ azquez, E. Tr´ elat, and J.-M. Coron. Control for fast and stable laminar-to-high-Reynolds-numbers transfer in a 2D Navier-Stokes channel flow. Discrete and Continuous Dynamical Systems-Series B , 10(4):925–956, 2008

    work page 2008

  49. [49]

    V azquez, J

    R. V azquez, J. Auriol, F. Bribiesca-Argomedo, and M. Krs tic. Backstepping for Partial Differential Equations. arXiv eprint:2410.15146, 2024

    work page arXiv 2024

  50. [50]

    Wang and L

    J. Wang and L. Hu. Solving low-rank semidefinite programs via manifold optimization. Journal of Scientific Computing, 104(1):33, 2025

    work page 2025

  51. [51]

    C. Xu, E. Schuster, R. V azquez, and M. Krstic. Stabilizat ion of linearized 2D magnetohydrodynamic channel flow by backstepping boundary control. Systems & control letters, 57(10):805–812, 2008

    work page 2008

  52. [52]

    H. Y u, K. Kashima, and J. I. Imura. Stability analysis of 2 -dimensional fluid flow based on sum of squares relaxation. In 2008 SICE Annual Conference, pages 3321–3326. IEEE, 2008

    work page 2008

  53. [53]

    Zayats, E

    M. Zayats, E. Fridman, and S. Zhuk. Global observer desig n for Navier-Stokes equations in 2D. In 2021 60th IEEE Conference on Decision and Control (CDC) , pages 1862–1867, 2021

    work page 2021

  54. [54]

    Zheng, G

    Y . Zheng, G. Fantuzzi, A. Papachristodoulou, P . Goulart , and A. Wynn. Chordal decomposition in operator- splitting methods for sparse semidefinite programs. Mathematical Programming, 180(1):489–532, 2020

    work page 2020

  55. [55]

    S. Zhuk, M. Zayats, and E. Fridman. Detectability and glo bal observer design for 2D Navier-Stokes equations with uncertain inputs. Automatica, 153:111043, 2023. REPRESENTA TION OFND PDES 37 A. A Bijection Between the PDE Domain and Fundamental State Space In Section 4, an explicit expression was derived for the inverse of the multivariat e differential o...

    work page 2023

  56. [56]

    TDδ u = u for any u ∈ D

    work page

  57. [57]

    Proof For the first statement, we note that, by Lem

    Dδ Tv = v for any v ∈ Ln 2. Proof For the first statement, we note that, by Lem. 20, we have ∂ δ j+1 s j+1 · · ·∂ δN sN u ∈ D1 ∩ · · · ∩D j ⊆ D j for all u ∈ D and j ∈ { 1 : N}. Since, by Cor. 3, T j∂ δ j s j u = u for all u ∈ D j, it follows that for every u ∈ D, TDδ u = TN · · ·T2(T1∂ δ1 s1 )(∂ δ2 s2 · · ·∂ δN sN u) = TN · · ·T3(T2∂ δ2 s2 )(∂ δ3 s3 · · ·...

    work page

  58. [58]

    Unfortunately, admissibility of the domains Di is insufficient to guarantee that this is indeed the case. Indeed, as shown in S ubsection 4.2 for Ω = [0,1]2 and δ = (d,d), a unique inverse to ∂ d x ∂ d y : D1 ∩ D2 → Ln 2[[0,1]2] may not exist if the boundary conditions defining D1 and D2 impose conflicting constraints at any of the corners of the domain, i.e...

    work page

  59. [59]

    As such, a right-inverse to ∂ d x ∂ d y : D1 ∩ D2 → Ln 2[[0,1]2] exists only if the operators T1 and T2 commute

    In order to prove that consistency of the PDE domains is also sufficient for ∂ d x ∂ d y : D1 ∩ D2 → Ln 2[[0,1]2] to be right-invertible, we remark that the operators T = T2T1 and ˜T = T1T2 both already satisfy ∂ d x ∂ d y T = ∂ d x ∂ d y ˜T = In, though Tv /∈ D1 and ˜Tv /∈ D2 for general v ∈ Ln 2[[0,1]2]. As such, a right-inverse to ∂ d x ∂ d y : D1 ∩ D2 ...

    work page

  60. [60]

    Proof For the case n = 1, the result follows by inspection

    Then T,A ∈ ΠΠΠn× n N [Ω]. Proof For the case n = 1, the result follows by inspection. In particular, by definition, T = ∏N i=1 Ti and A = ∑⃗0≤ α≤ δ M[AAAα ] ∏N i=1 Ai, αi , where the operators Ti and Ai,αi are of the form of the operators T and A in Cor. 2, respectively. By definition, then, we find that Ti and Ai,αi are univariate 3-PI operators, wherefore ...

    work page

  61. [61]

    By linearity of the inner product, it follows that if R := ∑M i=1 ΠN [R j], then ⟨u,Rv⟩L2 = M ∑ i=1 ⟨ u,ΠN[R j]v ⟩ L2 = M ∑ i=1 ⟨ ΠN[R∗ j ]u,v ⟩ L2 = ⟨R∗u,v⟩L2 . □ C. Exponential PIE to PDE Stability of Classical PDE Examples In this appendix, we verify exponential stability of the reaction-dif fusion equation from Subsec. 7.2.1 and the wave equation from...

    work page

  62. [62]

    In particular, note that uxxyy(0,x,y) = ∞ ∑ m,n=1 bm,nµ m(λm,n − µ m) sin (√ µ mx ) sin (√ λm,n − µ my ) , utxxyy (0,x,y) = − ∞ ∑ m,n=1 [ κbm,n − am,n √ λm,n ] µ m(λm,n − µ m) sin (√ µ mx ) sin (√ λm,n − µ my ) . By orthogonality of the spatial basis functions, it follows then th at     [ uxxyy(0) utxxyy (0) ]    2 L2 = LxLy 4 ∞ ∑ m,n=1 [ b2 m,n +...

    work page