Appliance energy estimation based on voltage sensing

A method, comprising:

• determining, based at least in part on first voltage data measured by a plug-in sensor plugged into an electrical outlet of an electrical circuit at a location, a resistance at a line wire of a service drop between a service panel of the location and a transformer that provides power to the location through the service drop;
• determining, based at least in part on second voltage data measured by the plug-in sensor plugged into the electrical outlet, an input voltage to the service panel;
• generating based at least in part on the resistance at the line wire of the service drop and the input voltage to the service panel, a machine learning (“ML”) input signal for the location;
• providing the ML input signal to an ML model trained to generate a current signal of an appliance;
• and
• receiving, from the ML model, a current signal indictive of a current of the appliance for which the ML model was trained, the current signal indicating the current of the appliance operating at the location.

A computer-implemented method, comprising:

• determining, based on a first plurality of voltage measurements taken by a plug-in sensor plugged into an electrical outlet of an electrical circuit at a location, a resistance at a line wire of a service drop between a service panel of the location and a transformer that provides power to the location;
• determining, based on a second plurality of voltage measurements taken by the plug-in sensor plugged into the electrical outlet, an input voltage into the service panel;
• and
• determining, based at least in part on the resistance at the line wire and the input voltage into the service panel, a current signal of an appliance operating at the location.