Lyapunov methods for input-to-state stability of time-varying evolution equations

Basic Information

• Paper ID: 2510.14057
• Title: Lyapunov methods for input-to-state stability of time-varying evolution equations
• Authors: Rahma Heni, Andrii Mironchenko, Fabian Wirth, Hanen Damak, Mohamed Ali Hammami
• Classification: math.OC (Optimization and Control)
• Publication Date: October 15, 2025 (arXiv preprint)
• Paper Link: https://arxiv.org/abs/2510.14057

Abstract

This paper establishes that (local) input-to-state stability ((L)ISS) and integral input-to-state stability (iISS) of time-varying infinite-dimensional abstract systems can be guaranteed by the existence of corresponding Lyapunov functions. In particular, input-to-state stability of linear time-varying control systems in Hilbert spaces with bounded input operators is discussed. For a class of time-varying semilinear evolution equations, a construction method for non-coercive LISS/iISS Lyapunov functions is proposed. The validity of the obtained results is verified through two examples.

Research Background and Motivation

Research Problems

This paper primarily addresses the input-to-state stability analysis of time-varying infinite-dimensional systems, specifically:

1. How to characterize ISS and iISS properties of time-varying infinite-dimensional systems through Lyapunov function methods
2. How to construct non-coercive Lyapunov functions for analyzing time-varying systems with unbounded operators
3. How to extend classical ISS theory from time-invariant systems to the time-varying case

Significance Analysis

1. Theoretical Significance: ISS theory is one of the core concepts in robust nonlinear control and is crucial for analyzing system robustness against external disturbances
2. Practical Applications: Many real-world systems (such as distributed parameter systems described by partial differential equations) are inherently time-varying and require corresponding theoretical support
3. Technical Challenges: Analysis of time-varying systems is considerably more complex than time-invariant systems, particularly when system operators are unbounded

Limitations of Existing Methods

1. Insufficient Time-Varying System Theory: Compared to ISS theory for time-invariant infinite-dimensional systems, research on the time-varying case is relatively limited
2. Difficulty in Handling Unbounded Operators: For time-varying systems with unbounded operators A(t), there are almost no existing ISS and iISS results
3. Complex Lyapunov Function Construction: The construction of ISS Lyapunov functions in time-varying systems is more difficult than in the time-invariant case

Core Contributions

1. Established Lyapunov Characterization of ISS/iISS for Time-Varying Infinite-Dimensional Systems: Proved that the existence of corresponding Lyapunov functions is equivalent to ISS/iISS properties of the system
2. Proposed Construction Method for Non-Coercive Lyapunov Functions: For linear time-varying systems with unbounded operators, provided explicit construction of non-coercive ISS Lyapunov functions
3. Extended Analysis Framework for Semilinear Systems: Established construction methods for LISS/iISS Lyapunov functions for time-varying semilinear evolution equations
4. Refined Well-Posedness Theory for Time-Varying Systems: Provided sufficient conditions for well-posedness of time-varying semilinear evolution equations
5. Provided Concrete Application Examples: Verified the validity of theoretical results through the Kuramoto-Sivashinsky equation and heat equation

Methodology Details

Problem Formulation

Study the stability of abstract control systems Σ = (X, U, φ), where:

• X: state space (normed linear space)
• U: input value set
• φ: transition mapping

Specifically, consider time-varying semilinear evolution equations:

ẋ(t) = A(t)x(t) + Ψ(t, x(t), u(t)), t ≥ t₀ ≥ 0
x(t₀) = x₀

Core Theoretical Framework

1. ISS/iISS Definitions

Input-to-State Stability (ISS): There exist β ∈ KL, γ ∈ K such that

‖φ(t, t₀, x₀, u)‖ₓ ≤ β(‖x₀‖ₓ, t - t₀) + γ(‖u‖ᵤ)

Integral Input-to-State Stability (iISS): There exist α ∈ K∞, μ ∈ K, β ∈ KL such that

‖φ(t, t₀, x₀, u)‖ₓ ≤ β(‖x₀‖ₓ, t - t₀) + α(∫ᵗₜ₀ μ(‖u(s)‖ᵤ)ds)

2. Lyapunov Function Definition

Non-Coercive ISS Lyapunov Function: A continuous function V : ℝ₊ × D → ℝ₊ satisfying:

• V(t, 0) = 0
• 0 < V(t, x) ≤ α₂(‖x‖ₓ), ∀x ∈ D{0}
• ‖x‖ₓ ≥ κ(‖u‖ᵤ) ⟹ V̇ᵤ(t, x) ≤ -μ(V(t, x))

where the Lie derivative is defined as:

V̇ᵤ(t, x) := lim sup[h→0⁺] (1/h)[V(t+h, φ(t+h, t, x, u)) - V(t, x)]

Technical Innovations

1. Stability Characterization of Evolution Families

Established the equivalence between uniform exponential stability of evolution families {W(t, s)} and system ISS properties:

Theorem 4.5: For linear systems ẋ = A(t)x + B(t)u, the following conditions are equivalent:

• System is ISS
• System is 0-UGAS
• System is iISS
• Evolution family is uniformly asymptotically stable
• Evolution family is uniformly exponentially stable

2. Non-Coercive Lyapunov Function Construction

For uniformly exponentially stable evolution families, an explicit non-coercive ISS Lyapunov function was constructed:

V(t, x) = ∫^∞ₜ ‖W(τ, t)x‖²ₓ dτ

and proved to satisfy the dissipation inequality:

V̇ᵤ(t, x) ≤ -‖x‖²ₓ + (ηk²)/(2w)‖x‖²ₓ + (k²)/(2ηw)‖B‖²∞‖u(t)‖²ᵤ

3. Treatment of Semilinear Systems

Extended results from linear systems to the semilinear case through local linear boundedness assumption (H2) and bilinear growth assumption (H3).

Experimental Setup

Application Examples

1. Kuramoto-Sivashinsky Equation

Consider the controlled time-varying KS equation:

∂x/∂t = -∂⁴x/∂z⁴ - ϱ∂²x/∂z² - μ(t)x + (x|sin(t)|)/(1 + e^(-zt)x²)u

under homogeneous Dirichlet boundary conditions.

2. Controlled Heat Equation

Consider the system:

∂x/∂t = ν∂²x/∂z² + R(t)x + ωsin(z)x + u

where R(t) is a family of bounded linear operators.

Analysis Methods

1. Operator Theory Analysis: Utilize operator semigroup theory to analyze system well-posedness
2. Lyapunov Function Construction: Construct concrete Lyapunov functions based on theoretical results
3. Stability Verification: Verify stability conditions through Lie derivative calculations

Experimental Results

Main Results

1. iISS Property of KS Equation

Proposition 6.1: When ϱ < 4π², the KS equation (43)-(44) is iISS.

The proof constructs a Lyapunov function:

V(t, x) = ln(1 + Z(t, x))
Z(t, x) = (1 + e^(-t))∫₀¹ x²(z)dz

2. ISS Property of Heat Equation

For the controlled heat equation, when the condition:

r + ω < νπ²/ℓ²

is satisfied, the system is ISS, where r = supt≥0 ‖R(t)‖.

Theoretical Verification

Both examples verify that:

1. The theoretically constructed Lyapunov functions indeed satisfy the corresponding dissipation inequalities
2. The stability conditions' relationship with physical parameters aligns with intuition
3. The method is applicable to different types of partial differential equations

Related Work

Development of ISS Theory

1. Classical ISS Theory: Sontag introduced the ISS concept in the late 1980s, followed by development of a complete theoretical framework
2. Infinite-Dimensional Extension: In recent years, ISS theory has been extended to infinite-dimensional systems, primarily for time-invariant cases
3. Time-Varying Systems: ISS theory for time-varying infinite-dimensional systems is relatively limited; this paper fills this gap

Lyapunov Methods

1. Coercive Functions: Traditional methods primarily considered coercive Lyapunov functions
2. Non-Coercive Functions: Non-coercive Lyapunov function theory has been developed recently and is more suitable for infinite-dimensional systems
3. Construction Methods: This paper provides systematic construction methods, particularly for the time-varying case

Conclusions and Discussion

Main Conclusions

1. Theory Refinement: Established complete Lyapunov characterization of ISS/iISS for time-varying infinite-dimensional systems
2. Method Innovation: Proposed systematic construction methods for non-coercive Lyapunov functions
3. Application Verification: Verified the practical utility of the theory through concrete PDE examples

Limitations

1. Assumption Conditions: The assumptions on nonlinear terms are relatively strong (local Lipschitz continuity, etc.)
2. Construction Complexity: In certain cases, Lyapunov function construction remains relatively complex
3. Computational Complexity: Verifying stability conditions in practical applications may involve complex calculations

Future Directions

1. Boundary Disturbances: Extend to ISS and iISS analysis of time-varying nonlinear parabolic PDEs with boundary disturbances
2. Network Systems: Apply results to time-varying infinite-dimensional network systems
3. Stochastic Systems: Consider ISS theory for stochastic time-varying infinite-dimensional systems

In-Depth Evaluation

Strengths

1. Significant Theoretical Contribution: First systematic establishment of ISS Lyapunov theory for time-varying infinite-dimensional systems
2. Strong Method Innovation: The construction method for non-coercive Lyapunov functions has important theoretical value
3. High Mathematical Rigor: Proofs are rigorous and theoretical results are complete
4. Clear Practical Value: PDE examples demonstrate the practical application value of the method

Weaknesses

1. Limited Application Scope: The assumptions on system classes are relatively restrictive
2. Computational Complexity: Computational complexity of certain theoretical results in practical applications is high
3. Insufficient Numerical Verification: Lacks large-scale numerical experiments for verification

Impact

1. Academic Value: Provides important theoretical tools for stability analysis of time-varying infinite-dimensional systems
2. Application Prospects: Has significant application potential in the field of distributed parameter system control
3. Theoretical Advancement: Promotes extension of ISS theory to more general system classes

Applicable Scenarios

1. Distributed Parameter Systems: Applicable to control systems described by time-varying PDEs
2. Robust Control Design: Provides theoretical foundation for robust controller design for time-varying systems
3. Stability Analysis: Provides systematic methods for stability analysis of complex time-varying systems

References

The paper cites 50 important references covering classical and recent research results in ISS theory, evolution equations, operator semigroup theory, and related fields, reflecting the authors' deep understanding and comprehensive grasp of the relevant areas.