Abstract:In this work, we present approaches to rigorously certify $A$- and $A(\alpha)$-stability in Runge-Kutta methods through the solution of convex feasibility problems defined by linear matrix inequalities. We adopt two approaches. The first is based on sum-of-squares programming applied to the Runge-Kutta $E$-polynomial and is applicable to both $A$- and $A(\alpha)$-stability. In the second, we sharpen the algebraic conditions for $A$-stability of Cooper, Scherer, T{ü}rke, and Wendler to incorporate the Runge-Kutta order conditions. We demonstrate how the theoretical improvement enables the practical use of these conditions for certification of $A$-stability within a computational framework. We then use both approaches to obtain rigorous certificates of stability for several diagonally implicit schemes devised in the literature.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to strictly certify the A - stability and A(α) - stability of Runge - Kutta methods through solutions of Linear Matrix Inequalities (LMIs). Specifically, the authors propose two methods: 1. **Method based on sum - of - squares programming**: This method is applied to Runge - Kutta E - polynomials and is suitable for A - stability and A(α) - stability. 2. **Improved algebraic conditions**: This method strengthens the existing CSTW conditions by introducing Runge - Kutta order conditions, making them more practical in the computational framework. ### Main contributions of the paper 1. **Theoretical contributions**: - **Improvement of CSTW conditions**: The authors modify the CSTW conditions, considering the approximate order of the stability function as an exponential function, so that these conditions can strictly certify stability through computational methods. - **Sum - of - squares optimization**: A detailed method for certifying A - stability and A(α) - stability using sum - of - squares optimization is provided by testing the non - negativity of E - polynomials. 2. **Practical applications**: - **Strict certification of stability**: The authors use the above methods to strictly certify the stability of several recently designed schemes proposed in the literature, including diagonal implicit schemes of 7 - stage 4 - order, 12 - stage 5 - order, 9 - stage 6 - order and 13 - stage 8 - order. ### Paper structure - **Sections 2 and 3**: Introduce the background knowledge of Runge - Kutta methods and linear matrix inequalities. - **Section 4**: Develop linear matrix inequalities for A - stability and A(α) - stability of Runge - Kutta methods, and include the main theoretical results and examples. - **Section 5**: Construct strict certifications of A - stability and A(α) - stability through computer - aided solution of related linear matrix inequalities. - **Section 6**: Summary and outlook. ### Key concepts - **A - stability**: When $ z \in C^-=\{ z \in \mathbb{C}:\text{Re}(z)\leq0\} $, $ |W(z)|\leq1 $. - **A(α) - stability**: When $ z \in S_\alpha = \{ z \in \mathbb{C}:|\arg(-z)|\leq\alpha\} $, $ |W(z)|\leq1 $. - **E - polynomial**: $ E(y;\alpha)=|D(ye^{-i\alpha})|^2 - |N(ye^{-i\alpha})|^2 $, where $ 0 < \alpha\leq\frac{\pi}{2} $. - **Linear Matrix Inequality (LMI)**: Used to verify the non - negativity of polynomials, in the form of $ F(\eta)=P+\sum_{j = 1}^d\eta_jN_j\succeq0 $. ### Examples - **SDIRK(3, 2) method**: Shows how zero eigenvalues reduce the dimension of the convex set $ R $. - **Hammer & Hollingsworth method**: Illustrates that $ X\succeq0 $ is a necessary assumption for $ X $ having zero eigenvalues. ### Computational details - **Using CVX**: Verify the feasibility of LMI by solving semidefinite programming problems. - **Strict certification**: Convert the output of CVX into symbolic rational numbers and perform an exact symbolic LDL decomposition to ensure $ D\succeq0 $. ### Conclusion This paper provides a systematic method for strictly certifying the A - stability and A(α) - stability of Runge - Kutta methods through theoretical and practical applications, which is of great significance for numerical analysis and stability research in dynamical systems.

Algebraic Conditions for Stability in Runge-Kutta Methods and Their Certification via Semidefinite Programming

Stability theory of TASE-Runge-Kutta methods with inexact Jacobian

Necessary and sufficient conditions for strong stability of explicit Runge-Kutta methods

Â-and Î-stability of collocation Runge-Kutta methods

Stability of multistep Runge-Kutta methods for systems of functional-differential and functional equations

Hypocoercivity for Linear ODEs and Strong Stability for Runge--Kutta Methods

A Unifying Theory for Runge--Kutta-like Time Integrators: Convergence and Stability

A Symplectic Algorithm For The Stability Analysis Of Nonlinear Parametric Excited Systems

Many-Stage Optimal Stabilized Runge–Kutta Methods for Hyperbolic Partial Differential Equations

Stability Analysis of Runge-Kutta Methods for Systems U'(t)=lu(t)+mu ([T])

On Lyapunov Stability of Positive and Conservative Time Integrators and Application to Second Order Modified Patankar--Runge--Kutta Schemes

Many-Stage Optimal Stabilized Runge-Kutta Methods for Hyperbolic Partial Differential Equations

Runge-Kutta methods are stable

Learning deep Koopman operators with convex stability constraints

Very High-Order A-stable Stiffly Accurate Diagonally Implicit Runge-Kutta Methods with Error Estimators

Convergence and stability of randomized implicit two-stage Runge-Kutta schemes

Diagonally drift-implicit Runge-Kutta methods of weak order one and two for Itô SDEs and stability analysis

Algebraic Structure of the Weak Stage Order Conditions for Runge–Kutta Methods

Robust Optimization of Index-2 Differential Algebraic Equations with Guaranteed Stability under Parametric Uncertainty: Application to a Reactor–Separator–Recycle Process

Verification Methods for the Lyapunov–Krasovskii Functional Inequalities

Enforcing strong stability of explicit Runge--Kutta methods with superviscosity