Parallel Computation of Transient Stability Using Symplectic Gauss Method and GPU

Baijian Wen,Wenxin Guo,Jiayi Hu,Fangzong Wang,Lei Zhang
DOI: https://doi.org/10.1109/MACE.2012.509
2012-01-01
Abstract:In this paper, a novel parallel algorithm is proposed for power system transient stability computation. The proposed algorithm uses the s-stage 2s-order symplectic Gauss method to convert the differential-algebraic system simultaneously at s time points into a set of nonlinear algebraic equations, and the algebraic system is then solved using Newton's method. For the solution of the linear equations involved in Newton's process, a matrix factorization method is used to decompose the whole computational task into two parts: the first is fully parallelizable-in-time, and the second is solved using GMRES method combined with the multifrontal complete LU factorization based preconditioner, which is well suited for parallelization spatially. The proposed algorithm has shown fast convergence rate, and is suited for the GPU based computing. For test, the proposed algorithm has been implemented on a single GPU based computer, the results on 3 example power systems show the proposed algorithm achieves great computational efficiency relative to the traditional CPU computing.
What problem does this paper attempt to address?