PCA-based preprocessing method of electronic data in Power Grid

Abstract

The electronic data which are acquired respectively by SCADA (Supervisory Control and Data Acquisition) and WAMS (wide-area measurement system) have different structures and characters, therefore these two kinds of data could not be merged easily as well as used conveniently. This problem makes online analysis in SASOPG (Stabilization and Security of Power Grid) difficult. Aiming the problem, this paper proposes a PCA-based preprocessing method of electronic data in power grid. The method is based on a two-stage state estimation algorithm optimized by PCA. By using linear/non-linear mixed state estimation algorithm, the origin data from SCADA/WAMS are transformed into state values. To accelerate the state estimation restrained rapidly, a data value optimized algorithm of principal component analysis is provided, and finally the basic message of online analysis is obtained. The result shows that the processing method merges the electronic data well, and it also has higher execution efficiency.