Transformer for monitoring electrical power network

Abstract

A method of estimating historical apparent power over time for at least one transformer device in a power network in a geographic area, the method comprising: providing training data, the training data includes apparent power data over time for a plurality of transformer devices in one or more geographic areas, energy consumption data over time for a power network in a respective geographic area, and ambient temperature data over time for a respective geographic area; in one embodiment, a method includes initializing a weight of an untrained neural network, the untrained neural network including an input layer, a plurality of hidden layers, and an output layer, and iteratively updating the weight using the training data to minimize a loss function to produce a trained neural network, the trained neural network including an input layer, a plurality of hidden layers, and an output layer. The trained neural network is configured to output, from input energy consumption data and input ambient temperature data of a geographic area over a period of time, an estimated apparent power of a transformer device in the geographic area over the period of time; and inputting the energy consumption data and the ambient temperature data within the first time period in the geographic area into the trained neural network to estimate the apparent power of at least one transformer device within the first time period in the geographic area.

Images