Fully implicit particle-in-cell algorithms for kinetic simulation of plasmas [Slides]
- Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
We have demonstrated, for the first time, a fully implicit, fully nonlinear electrostatic PIC formulation that features: (1) Exact charge conservation (via a novel particle mover strategy). (2) Exact energy conservation (no particle self-heating or self-cooling). (3) Adaptive particle orbit integrator to control errors in momentum conservation. The approach has been shown to be free of CFL and finite-grid numerical instabilities. As a result, the method is able to take time steps many times larger than explicit, and resolutions many times coarser. Central to our implementation is the concept of particle enslavement. We have generalized formulation to use spatial adaptivity via mapped coordinates. he method has much potential for efficiency gains vs. explicit in long-time-scale applications, with the CPU speedup scaling as $$(kΛ_D)-^{(d+1)}/N_{FE}$$. Minimize the number of nonlinear function evaluations $$N_{FE}$$ for given $Δt$, $$Δx \Rightarrow$$ preconditioning! We have formulated and implemented a very efficient moment-based preconditioner. We have ported the algorithm to GPU architectures. We have generalized the algorithm to non-radiative electromagnetic regimes (Darwin model), where, in addition to charge and energy, we also conserve canonical momenta.
- Research Organization:
- Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
- Sponsoring Organization:
- USDOE Laboratory Directed Research and Development (LDRD) Program; USDOE National Nuclear Security Administration (NNSA); USDOE Office of Science (SC), Fusion Energy Sciences (FES)
- DOE Contract Number:
- AC52-06NA25396
- OSTI ID:
- 1063911
- Report Number(s):
- LA-UR--13-21249
- Country of Publication:
- United States
- Language:
- English
Similar Records
Energy-Charge-conserving, Implicit, Electrostatic Particle-in-Cell Algorithm
Exact charge and energy conservation in implicit PIC with mapped computational meshes