Abstract: The disclosure describes three complimentary, synergistically interacting, and yet individually capable techniques for detecting electrical vehicle charging activities. In one example, convolutional neural network techniques find “edges” or points of significant change in electricity consumption. Time-series of electricity consumption are examined, and temperature is considered to normalize for changes in heating, ventilation, and air conditioning consumption. In an example, a time-series of electrical-consumption data of a service site is obtained over a time-range. The time-series of electrical-consumption data is converted into a time-series of consumption-change data. Temperature data may be associated with terms of the time-series of consumption-change data to thereby create input data for a machine-learned algorithm over the time-range. The input data is provided to a machine-learned model.

Abstract: A method, apparatus, and system for identifying electrical phases connected to electricity meters are disclosed. Voltage time series data of electricity meters are collected over a preselected collection time period, and three initial kernels representing three line-to-neutral phases are generated based on voltage correlations of meter-to-meter combinations. Three new kernels are then generated based on correlation values calculated for each of the three initial kernels with each electricity meter, and electricity meters are clustered into three groups based on average correlation values associated with each electricity meter. Six new kernels representing six phases are then formed based on the average correlation value associated with each electricity meter, and a predicted phase is assigned to each electricity meter based on correlation values of the electricity meter with each of the six new kernels based on the voltage time series data.

Abstract: A method, apparatus, and system for identifying electrical phases connected to electricity meters are disclosed. Voltage time series data of electricity meters are collected over a preselected collection time period, and three initial kernels representing three line-to-neutral phases are generated based on voltage correlations of meter-to-meter combinations. Three new kernels are then generated based on correlation values calculated for each of the three initial kernels with each electricity meter, and electricity meters are clustered into three groups based on average correlation values associated with each electricity meter. Six new kernels representing six phases are then formed based on the average correlation value associated with each electricity meter, and a predicted phase is assigned to each electricity meter based on correlation values of the electricity meter with each of the six new kernels based on the voltage time series data.