Abstract:When solving partial differential equations on scattered nodes using the Radial Basis Function-generated Finite Difference (RBF-FD) method, one of the parameters that must be chosen is the stencil size. Focusing on Polyharmonic Spline RBFs with monomial augmentation, we observe that it affects the approximation accuracy in a particularly interesting way - the solution error oscillates under increasing stencil size. We find that we can connect this behaviour with the spatial dependence of the signed approximation error. Based on this observation we are able to introduce a numerical quantity that could indicate whether a given stencil size is locally optimal. This work is an extension of our ICCS 2023 conference paper.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is how to select an appropriate stencil size to optimize the approximation accuracy when using the Radial Basis Function - Generated Finite Difference (RBF - FD) method to solve Partial Differential Equations (PDEs). Specifically, the author focuses on the situation where Polyharmonic Spline Radial Basis Functions (Polyharmonic Spline RBFs) are used in combination with polynomial enhancement, and observes that as the stencil size increases, the solution error exhibits oscillatory behavior. The goal of the paper is to understand the reasons for this oscillatory behavior and its quantitative characteristics, thereby predicting the optimal stencil size for a given problem, and improving the accuracy of the method without changing the discretization set or increasing the order of the method.
### Main research contents:
1. **Error oscillation phenomenon**: Through numerical experiments, the author finds that as the stencil size increases, the maximum absolute error \( e_{\text{max}}^{\text{poiss}}(n) \) and the average absolute error \( e_{\text{avg}}^{\text{poiss}}(n) \) exhibit oscillatory behavior, with multiple local minima and maxima. These oscillations are not random but show some regularity.
2. **Spatial dependence of point errors**: Further analysis shows that the oscillatory behavior of the error is related to the sign change of point errors. At the maximum of the error, the error signs are the same throughout the domain; while at the minimum of the error, the error signs in different parts of the domain may be different.
3. **Error indicator**: Based on the above observations, the author introduces a new quantity \( \delta N_{\pm}^{\text{poiss}} \), which is used to describe the difference in the number of positive and negative error nodes. This quantity can be used as an indicator of error sign change and is helpful for predicting local minima of the error.
4. **Influence of different parameters**: The paper also explores the influence of different parameters on the error oscillatory behavior, including different discretization methods, boundary conditions, domain shapes, differential operators, and the choice of analytical solutions. The results show that although these parameters will affect the specific error oscillation patterns, the oscillatory behavior still exists in general.
### Research significance:
- **Improving accuracy**: By predicting the optimal stencil size, the accuracy of the RBF - FD method can be improved without increasing the computational complexity.
- **Theoretical explanation**: Understanding the mechanism of error oscillatory behavior is helpful for in - depth understanding of the intrinsic properties of the RBF - FD method.
- **Practical applications**: Finally, the paper discusses the application of this method in practical problems, such as determining the steady - state temperature distribution of a radiator.
### Conclusion:
Through detailed numerical experiments and theoretical analysis, the paper reveals the oscillatory behavior of the error with the change of stencil size in the RBF - FD method, and proposes an effective error indicator \( \delta N_{\pm}^{\text{poiss}} \). This finding not only helps to optimize the performance of the RBF - FD method, but also provides a theoretical basis for further research.