Abstract:The higher-order guaranteed lower eigenvalue bounds of the Laplacian in the recent work by Carstensen, Ern, and Puttkammer [Numer. Math. 149, 2021] require a parameter $C_{\mathrm{st},1}$ that is found $\textit{not}$ robust as the polynomial degree $p$ increases. This is related to the $H^1$ stability bound of the $L^2$ projection onto polynomials of degree at most $p$ and its growth $C_{\rm st, 1}\propto (p+1)^{1/2}$ as $p \to \infty$. A similar estimate for the Galerkin projection holds with a $p$-robust constant $C_{\mathrm{st},2}$ and $C_{\mathrm{st},2} \le 2$ for right-isosceles triangles. This paper utilizes the new inequality with the constant $C_{\mathrm{st},2}$ to design a modified hybrid high-order (HHO) eigensolver that directly computes guaranteed lower eigenvalue bounds under the idealized hypothesis of exact solve of the generalized algebraic eigenvalue problem and a mild explicit condition on the maximal mesh-size in the simplicial mesh. A key advance is a $p$-robust parameter selection.
What problem does this paper attempt to address?
### What problem does this paper attempt to solve?
This paper aims to solve the problem of Guaranteed Lower Eigenvalue Bounds (GLB) for computing the eigenvalues of the Laplace operator. Specifically, it proposes a new Adaptive Hybrid High - Order Method to directly compute the lower bounds of these eigenvalues.
#### Main problem background
1. **Limitations of existing methods**:
- In previous literature, it is relatively easy to compute the upper bounds of eigenvalues, but it is more difficult to directly compute the lower bounds.
- In particular, when using high - order polynomials, existing methods rely on a parameter \( C_{st,1} \), and this parameter becomes less robust as the order \( p \) of the polynomial increases (\( C_{st,1} \propto (p + 1)^{1/2} \)).
2. **Advantages of the new method**:
- The new method introduces a \( p \)-robust constant \( C_{st,2} \), and \( C_{st,2} \leq 2 \) for an isosceles right triangle.
- Using this more robust constant, an improved Hybrid High - Order (HHO) eigenvalue solver is designed, which can directly compute the lower bounds of eigenvalues.
3. **Objectives**:
- To provide a method that can accurately solve the generalized algebraic eigenvalue problem under ideal conditions.
- To ensure that the maximum mesh size under simplified assumptions satisfies certain conditions, so as to compute the lower bounds of eigenvalues.
4. **Key progress**:
- **\( p \)-robust parameter selection**: Ensure stability even under high - order polynomials.
- **A priori best approximation and improved \( L^2 \) error estimate**: Provide better theoretical guarantees.
- **Reliable a posteriori error control**: Efficient error estimation can be achieved without additional stabilization terms.
5. **Practical applications**:
- The paper verifies the effectiveness of the new method through numerical experiments, shows its superior performance in computer benchmark tests, and provides numerical evidence of the optimal high - order empirical convergence rate.
In summary, the core problem of this paper is to develop a new HHO method that is robust to high - order polynomials to directly compute the lower bounds of the eigenvalues of the Laplace operator, and to verify its effectiveness and superiority through theoretical analysis and numerical experiments.