Abstract:This work constructs the first-ever sixth-order exponential Runge--Kutta (ExpRK) methods for the time integration of stiff parabolic PDEs. First, we leverage the exponential B-series theory to restate the stiff order conditions for ExpRK methods of arbitrary order based on an essential set of trees only. Then, we explicitly provide the 36 order conditions required for sixth-order methods and present convergence results. In addition, we are able to solve the 36 stiff order conditions in both their weak and strong forms, resulting in two families of sixth-order parallel stages ExpRK schemes. Interestingly, while these new schemes require a high number of stages, they can be implemented efficiently similar to the cost of a 6-stage method. Numerical experiments are given to confirm the accuracy and efficiency of the new schemes.
What problem does this paper attempt to address?
This paper aims to solve the time - integration problem of stiff systems, especially the time - integration of rigid parabolic partial differential equations (PDEs). Specifically, the authors constructed the first sixth - order exponential Runge - Kutta (ExpRK) method for the time - integration of stiff initial - value problems in the form of \(u'(t)=A u(t)+g(t, u(t)) = F(t, u(t))\). Such problems usually stem from the spatial discretization of parabolic partial differential equations, such as diffusion - reaction problems.
### Main problems
1. **Construction of high - order methods**: The existing ExpRK methods can only reach up to fourth - order at most, while the goal of this paper is to construct a sixth - order method.
2. **Satisfaction of stiffness conditions**: In order to ensure the high - order accuracy of the method, a large number of stiffness conditions need to be satisfied. These conditions increase significantly as the order of the method increases, and a sixth - order method needs to satisfy 36 stiffness conditions.
3. **Efficient implementation**: Although the sixth - order method requires more stages, the authors designed these methods so that they can be efficiently implemented at a cost similar to that of a six - stage method by parallel computing multiple independent internal stages.
### Solutions
- **Theoretical basis**: Using the exponential B - series theory, the stiffness conditions of the ExpRK method of any order were re - formulated, and the 36 stiffness conditions required for the sixth - order method were explicitly provided.
- **Condition solving**: By weakening one condition, these 36 stiffness conditions were successfully solved, and two new families of sixth - order parallel - stage ExpRK methods were constructed.
- **Numerical experiments**: The accuracy and efficiency of the new methods were verified through numerical experiments, demonstrating the superior performance of these methods in practical applications.
### Key contributions
- **New stiffness conditions**: For the first time, the 36 stiffness conditions required for the sixth - order ExpRK method were explicitly provided.
- **Construction methods**: Two new families of sixth - order rigid ExpRK methods, named š“š”šššŗš¼šš·š» and š“š”šššŗš¼šš·šŗ respectively, were constructed.
- **Efficient implementation**: These methods were designed so that they can be efficiently implemented at a cost similar to that of a six - stage method while maintaining high accuracy.
### Mathematical formulas
- **Exponential function**:
\[
\varphi_k(z)=\int_0^1 e^{(1 - \theta)z}\frac{\theta^{k - 1}}{(k - 1)!}d\theta,\quad k\geq1
\]
- **Stiffness conditions**:
\[
\sum_{i = 2}^s b_i(hA)\frac{c_i^{|\tau|- 1}}{(|\tau|- 1)!}=\varphi_{|\tau|}(hA)\quad\text{for all }\tau\in T_1\text{ with }|\tau|\leq p
\]
\[
\sum_{i = 2}^s b_i(hA)g^{(m)}(u)(\mathcal{S}_i(\tau_1)(u),\ldots,\mathcal{S}_i(\tau_m)(u)) = 0\quad\text{for all }\tau=[\tau_1,\ldots,\tau_m]\in T_2\text{ with }|\tau|\leq p
\]
Through these methods and theories, the authors successfully constructed the first sixth - order exponential Runge - Kutta method and verified its effectiveness and efficiency in practical applications.