Efficient screening method for branch contingency selection of overloads

In this paper, an efficient algorithm for repetitive sparse simultaneous linear equations solution is proposed. It features fast power systems analysis based on a multiprocessor pipeline, as well as efficient pipeline mapping on a hypercube architecture. A major characteristic of the proposed multiprocessor pipeline approach is the optimization of processor load balance (over 95% active time of processors). Achieved speedups are close to linear, while the system efficiency remains within the range of 95% to 97%. The communication overheads are neglectable and the execution times are calculated on the basis of the computing operations. The proposed algorithm has been successfully applied for power mismatch calculations arising in the solution phase of fast decoupled power flow (FDPF) which involves typical repeated sparse power network computations. The experiments supported the approach for suboptimal pipeline mapping on a higher order hypercube, with almost linear speed-up