A PARALLEL-IN-SPACE ALGORITHM FOR POWER SYSTEM TRANSIENT STABILITY SIMULATION
Chao Hong
Abstract
Chao Hong
Abstract
This paper presents a parallel in space algorithm for power system transient stability analysis suitable for implementations on message passing multiprocessors. A network partitioning scheme based on the subdivision of the factorization path tree of the network matrix is proposed and a parallel algorithm for solving sparse network matrix equations has been developed by arranging the matrix into a Bordered Blocked Diagonal Form (BBDF). The parallelism in space transient stability analysis algorithm has been implemented on a message passing multiprocessor. Test results on a large power system are presented to show that the proposed algorithm can significantly speedup transient stability simulations.
OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
This paper presents a parallel in space algorithm for power system transient stability analysis suitable for implementations on message passing multiprocessors. A network partitioning scheme based on the subdivision of the factorization path tree of the network matrix is proposed and a parallel algorithm for solving sparse network matrix equations has been developed by arranging the matrix into a Bordered Blocked Diagonal Form (BBDF). The parallelism in space transient stability analysis algorithm has been implemented on a message passing multiprocessor. Test results on a large power system are presented to show that the proposed algorithm can significantly speedup transient stability simulations.
Key concepts: Parallel computing, Transient (computer programming), Computer science, Parallel algorithm, Multiprocessing, Speedup, Diagonal, Stability (learning theory)