Global-in-time existence and uniqueness of classical solutions to the unsteady initial-boundary value problem for the four-velocity planar Broadwell model in a rectangular domain

1. Problem Statement

The paper addresses the mathematical analysis of the unsteady initial-boundary value problem (IBVP) for the four-velocity planar Broadwell model, a specific discrete kinetic equation derived from the Boltzmann equation.

• Governing Equations: The model describes the evolution of particle number densities $N_i(t, x, y)$ for four discrete velocities in a 2D rectangular domain $\Omega = [a_1, b_1] \times [a_2, b_2]$. The system consists of four coupled semilinear hyperbolic partial differential equations (PDEs):
  $$\begin{aligned} \partial_t N_1 + c \partial_x N_1 &= Q(N) \\ \partial_t N_2 + c \partial_y N_2 &= -Q(N) \\ \partial_t N_3 - c \partial_y N_3 &= -Q(N) \\ \partial_t N_4 - c \partial_x N_4 &= Q(N) \end{aligned}$$
  where the collision term is $Q(N) = 2cS(N_2 N_3 - N_1 N_4)$, with $c, S > 0$ being constants.
• Domain and Conditions: The problem is defined on the space-time cylinder $[0, +\infty) \times \Omega$. It includes:
  • Initial Conditions: Specified densities $N_i(0, x, y) = N_i^0(x, y)$.
  • Boundary Conditions: Inflow conditions are prescribed on the boundaries where characteristics enter the domain (e.g., $N_1$ at $x=a_1$, $N_2$ at $y=a_2$, etc.).
  • Compatibility: Initial and boundary data must satisfy compatibility conditions at the corners of the domain at $t=0$.
• Objective: To prove the global-in-time existence and uniqueness of classical solutions (solutions with continuous first-order partial derivatives) that are non-negative and bounded, assuming sufficiently small initial and boundary data.

2. Methodology

The authors employ a fixed-point approach combined with characteristic methods and Schauder's Fixed Point Theorem. The methodology proceeds in several stages:

A. Reformulation as Fixed-Point Problems

The nonlinear system is reformulated into an equivalent fixed-point problem.

1. Regularization: A parameter $\sigma > 0$ is introduced to transform the system into a form amenable to integration along characteristics. The system is rewritten as:
   $$\partial_t N_i + \vec{u}_i \cdot \nabla N_i + \sigma \rho(N) N_i = Q_i^\sigma(N)$$
   where $\rho(N) = \sum N_i$ and $Q_i^\sigma$ are modified source terms.
2. Integral Representation: Using the method of characteristics, the solution is expressed as an integral operator $T^\sigma$ (and a related operator $T$). The solution at any point $(t, x, y)$ is determined by integrating the source terms along the characteristic lines back to either the initial time or the boundary.
   • The operators $T$ and $T^\sigma$ map a space of continuous functions to itself.
   • A fixed point of these operators corresponds to a solution of the original IBVP.

B. Functional Analysis Framework

The authors define appropriate Banach spaces and norms:

• Space: $C(P, \mathbb{R})^4$, where $P$ is the space-time domain.
• Norms: They utilize the sup-norm $\|\cdot\|_\infty$ and a stronger norm $\|\cdot\|_1$ that includes bounds on the first-order partial derivatives ($\partial_t, \partial_x, \partial_y$).
• Convex Sets: They construct a closed, convex, and bounded subset $M_R$ of the function space, defined by a bound $R$ on the norm $\|N\|_1$.

C. Key Technical Steps

1. Continuity: Proving that the operators $T$ and $T^\sigma$ are continuous with respect to the chosen norms.
2. Compactness: Using the Arzelà-Ascoli theorem, the authors demonstrate that the operators map bounded sets into relatively compact sets. This relies on establishing uniform bounds on the derivatives of the solutions, ensuring equicontinuity.
3. Invariance: Proving that for sufficiently small data and a sufficiently large parameter $\sigma$, the operators map the set $M_R$ into itself ($T(M_R) \subset M_R$). This involves deriving specific inequalities involving the data bounds ($q$) and the time interval length ($\tau' - \tau$).
4. Uniqueness: Proving uniqueness via a contraction argument on small time intervals, which is then extended globally by a "step-by-step" continuation.

3. Key Contributions and Results

A. Main Theorems

1. Uniqueness (Theorem 2.1): The solution to the IBVP, if it exists, is unique.
2. Local Existence (Theorem 2.2): For a bounded time interval $[\tau, \tau']$, if the initial/boundary data are sufficiently small (satisfying specific inequalities involving constants $\mu, \lambda, \delta$) and $\sigma$ is large enough, there exists a unique non-negative classical solution with bounded derivatives.
3. Global Existence (Theorem 2.3): By iteratively extending the local solution, the authors prove the existence of a unique global-in-time classical solution on $[0, +\infty) \times \Omega$, provided the initial and boundary data are small enough.

B. Technical Innovations

• Extension to Multidimensional Unsteady Case: While 1D and stationary 2D cases were previously studied, this work successfully addresses the non-stationary (time-dependent) multidimensional case for the Broadwell model.
• Handling of Boundary Interactions: The paper rigorously handles the complex geometry of characteristics interacting with the rectangular boundaries, defining the solution via piecewise integral formulas (distinguishing between points reached from initial data vs. boundary data).
• Positivity Preservation: The construction ensures that the solution remains non-negative, a physical necessity for particle densities, by choosing $\sigma > 2cS$.

4. Significance

• Mathematical Rigor: The paper fills a gap in the literature regarding the well-posedness of multidimensional discrete kinetic models. It provides a rigorous proof for the existence of classical solutions (not just weak or distributional solutions), which is crucial for numerical reliability.
• Methodological Framework: The use of fixed-point theorems (Schauder) combined with characteristic integration offers a robust framework that can be adapted to other discrete velocity models (e.g., models with more velocities or different collision kernels).
• Foundation for Simulation: By establishing existence, uniqueness, and regularity (bounded derivatives), the results justify the use of numerical methods for simulating gas flows in this regime, ensuring that the discrete models behave predictably.
• Physical Relevance: The preservation of non-negativity and the handling of inflow boundaries make the results directly applicable to modeling real gas flows in confined geometries.

5. Conclusion

The authors successfully established that the unsteady four-velocity planar Broadwell model in a rectangular domain admits a unique global classical solution for small data. The proof relies on transforming the PDE system into a fixed-point problem, proving the compactness and continuity of the associated operators, and using a step-by-step extension to cover the infinite time horizon. This work extends previous 1D and stationary results to a more complex, time-dependent multidimensional setting.