Parallel state estimation assessment with practical data

This paper presents a parallel state estimation (PSE) implementation using a preconditioned conjugate gradient algorithm and an orthogonal decomposition-based algorithm. Preliminary tests against a commercial Energy Management System (EMS) State Estimation (SE) tool using real-world data are performed. The results show that while the preconditioned conjugate gradient algorithm can solve the SE problem faster with the help of parallel computing techniques, it might not be good for real-world data due to the large condition number of its gain matrix introduced by the wide range of measurement weights. With the help of PETSc package, the orthogonal decomposition-based PSE algorithm can achieve 5-20 times speedup comparing against the commercial EMS tool. It is very promising that the developed PSE can solve the SE problem for large power systems at the SCADA rate, to improve grid reliability.