PNNL

Abstract

The growing trend of solar photovoltaic systems within electric power distribution networks can potentially introduce challenges such as reverse power flow and overvoltage issues. This makes accurate knowledge of and the ability to validate rooftop photovoltaic system location and power generation valuable to distribution system operators. In modern power distribution systems, accurate information on the location status, and settings of all controllable assets mitigates operational risk and supports coordination of grid-edge devices. This paper presents a state estimation algorithm for distribution systems that can accurately estimate photovoltaic system injections. Specifically, the algorithm 1) detects unregistered rooftop photovoltaics by estimating the injections of hypothetical systems at behind-the-meter locations, 2) identifies inverter settings for unregistered systems, and 3) detects incorrect inverter settings for registered systems. The algorithm is implemented as a state estimator federate of the Open Energy Data Initiative Systems Integration platform. The proposed method was shown to correctly identify unregistered photovoltaic systems and power factor settings using and combination of course forecasts and simulated smart meter and substation measurements in demonstrations with the IEEE 4-bus and IEEE 13-bus, and IEEE 123-bus test systems under various scenarios.