Embodiments generate machine learning predictions to discover target device energy usage. One or more trained machine learning models configured to discover target device energy usage from source location energy usage can be stored. Multiple instances of source location energy usage over a period of time can be received for a given source location. Using the trained machine learning model, multiple discovery predictions for the received instances of source location energy usage can be generated, the discovery predictions comprising a prediction about a presence of target device energy usage within the instances of source location energy usage. And based on the multiple discovery predictions, an overall prediction about a presence of target device energy usage within the given source location's energy usage over the period of time can be generated.

A system for transferring energy based on predicted outages is described herein.

Operational planning of energy storage systems using batteries, e.g., Lithium-Ion batteries, is disclosed. A method of operating at least one server node includes: obtaining one or more load profiles associated with one or more interfacing modes of a battery energy storage system with an electrical utility distribution system, and predicting one or more degradations of the battery energy storage system, the one or more degradations being associated with operating the battery energy storage system in the one or more interfacing modes, the one or more degradations being predicted using an aging model of batteries of the battery energy storage system, the aging model being based on the one or more load profiles.

The energy management system is an energy management system for a microgrid including at least either one of an electrical load and a renewable energy power-generation system, and an energy storage device, the energy management system including: a prediction unit that predicts, at a fixed time interval, at least either one of an electricity demand and an amount of electricity generated from renewable energy, and an electricity fee, within a predetermined period; and an optimization unit that performs optimization for an optimum charging/discharging plan for the energy storage device by using a prediction result from the prediction unit and in consideration of uncertainty in prediction.

A localized smart energy management system comprises a plurality of controllable loads, at least one intermittent energy source, a selectively connectable dispatchable energy source, and optionally an energy storage system. A method for balancing power production and power consumption of such localized smart energy management systems in real time comprises performing a coarse-grained optimization in a first layer of a hierarchical optimization structure to generate a predicted schedule, based on long-term load demand profiles and long-term power generation profiles. A second layer iteratively refines the predicted schedule upon receiving a new forecast of a short-term power generation profile for the at least one intermittent energy source.

The present disclosure provides a method and a system for estimating capacity and usage pattern of behind-the-meter energy storage in electric networks. Conventional techniques on estimating an effective capacity of behind-the-meter energy storage of a consumer, in presence of distributed energy generation units is limited, computationally intensive and provide inaccurate prediction. The present disclosure provides an accurate estimate of the effective capacity and usage pattern of behind-the-meter energy storage of a target consumer utilizing data samples received from a utility in presence of one or more distributed energy generation units, using an energy balance equation with less computation and accurate prediction. Based on accurate estimation of the effective capacity and usage pattern, the utility may plan for proper infrastructure to meet power demands of the consumers.

A method and apparatus for tertiary control with over-current protection. In one embodiment, the method comprises calculating at least one unconstrained optimal net intertie target for an area of a power network; calculating, for each resource within the area, optimal scheduled current to achieve the at least one unconstrained optimal net intertie target; calculating, using the optimal scheduled currents and a plurality of stress coefficients, net scheduled current for each power line segment within the area; comparing the net scheduled currents to corresponding stress thresholds to identify any stress violations; reducing, when the comparing step identifies one or more stress violations, the optimal scheduled current for one or more resources contributing to the one or more stress violations; and calculating, when the comparing step identifies the one or more stress violations, updated optimal scheduled current for one or more resources not contributing to the one or more stress violations.

There is described a method of reserving a capacity of one or more energy storage devices. The method includes forecasting, based on past electricity demand of a site, future electricity demand of the site over a future time period. The method further includes determining a forecasting error between the forecasted future electricity demand and an actual electricity demand of the site over the future time period. The method further includes adjusting, based on the forecasting error, a target state of charge (SOC) of one or more energy storage devices. The method further includes reserving, based on the adjusted target SOC, a capacity of the one or more energy storage devices.

This disclosure discusses systems, methods, and techniques for producing and utilizing a capability region of one or more monitored equipment. To do so, an intelligent electronic device (IED) may access a data set of one or more known performance characteristics of the monitored equipment. The known performance characteristics are based on, or dependent of, one or more variables. The IED may also access a constraint library with geometric primitives. Then, the IED may analyze the data set and may produce the capability region using the geometric primitive. The IED may compare an operating point of the monitored equipment to the capability region of the monitored equipment. Based on the comparison, the IED may implement a control action.

A building system for building data point prediction, the building system comprising one or more memory devices configured to store instructions, that, when executed by one or more processors, cause the one or more processors to receive first building data for a building data point of a building and generate training data, the training data comprising a probability distribution sequence comprising a first probability distribution for the building data point. The instructions cause the one or more processors to train a prediction model based on the training data, receive second building data for the building data point, and predict, for one or more time-steps into the future, one or more second probability distributions with the second building data based on the prediction model, each of the one or more second probability distributions being a probability distribution for the building data point at one of the one or more time-steps.