CPC G01R 22/063 (2013.01) [G06N 7/01 (2023.01); G06Q 50/06 (2013.01)] | 18 Claims |
1. A method performed by an electric meter of an end user in a power system, the method comprising:
measuring, at the electric meter, voltage data associated with use of electrical power by the end user, wherein the electric meter is positioned at a location of the end user, and wherein the end user receives the electrical power from a power station of the power system via a transmission system of the power system;
analyzing, at the electric meter, the voltage data to detect a critical voltage characteristic, by—
decomposing historical measurements of the voltage data into components with different time scales,
training a plurality of machine learning models, by training a separate machine learning model for each time scale based on the components with that time scale,
generating predicted voltage profiles using the trained machine learning models, and
determining whether the measured voltage data differs from any of the predicted voltage profiles by a predetermined threshold;
estimating, at the electric meter, an electrical margin of the end user based on the voltage data; and
controlling, via the electric meter, an electrical power source other than the power station to change an active power supply to the end user and controlling an electric appliance to change an active power consumption of the end user to increase the electrical margin of the end user independent from any adjustment of the electrical power supplied to the end user from the power station, wherein the electrical power source and the electric appliance are local to the end user in the power system.
|