Higher Order Force Gradient Symplectic Algorithms

Abstract

We show that a recently discovered fourth order symplectic algorithm, which requires one evaluation of force gradient in addition to three evaluations of the force, when iterated to higher order, yielded algorithms that are far superior to similarly iterated higher order algorithms based on the standard Forest-Ruth algorithm. We gauge the accuracy of each algorithm by comparing the step-size independent error functions associated with energy conservation and the rotation of the Laplace-Runge-Lenz vector when solving a highly eccentric Kepler problem. For orders 6, 8, 10 and 12, the new algorithms are approximately a factor of $10^3$, $10^4$, $10^4$ and $10^5$ better.

23 pages, 10 figures

Cited by in corpus (10)

• Gradient Symplectic Algorithms for Solving the Schroedinger Equation with Time-Dependent Potentials
• On the construction of high-order force gradient algorithms for integration of motion in classical and quantum systems
• Forward Symplectic Integrators and the Long Time Phase Error in Periodic Motions
• The physics of symplectic integrators: perihelion advances and symplectic corrector algorithms
• Gradient Symplectic Algorithms for Solving the Radial Schrodinger Equation
• Polynomial Filtered HMC -- an algorithm for lattice QCD with dynamical quarks
• Speeding up HMC with better integrators
• Force Gradient Integrators
• An efficient time-stepping scheme for ab initio molecular dynamics simulations
• Forward and non-forward symplectic integrators in solving classical dynamics problems