Implementing a PMU Based Linear State Estimator on a Single Board Computer

Abstract

Smart Grid Technology is essential in ensuring resilient and reliable operation of power grids. Phasor Measurement Units (PMUs) provide tremendous increase in quantity and quality of data for Wide Area Monitoring Systems (WAMS) and provide increased situational awareness in power grids. Additionally, sufficient PMU penetration can enable the use of linear state estimators (LSEs), which are currently limited due to insufficient PMU deployment. However, PMU deployment is rapidly growing and LSE implementations may be feasible in the future. This study investigates the development and performance of a LSE in a real-world, small-scale laboratory environment. Specifically, a LSE algorithm was developed for a 7-bus, 1000:1 scale, smart grid testbed containing real analog hardware, eight PMUs, and a WAMS. The LSE was written in Python and benchmarked on two computing platforms: a traditional x86 architecture, and an ARM based single board computer (SBC). The results of this study show that a Raspberry Pi 3B+ single board computer is capable of real-time LSE with 8 PMUs reporting at a rate of 30 measurements per second. A LSE implementation on this scale, and computing platform, is suitable for distributed state estimation applications.