A method of managing a charge of a plurality of vehicles on a site, the site including a plurality of charging points and a gateway, the gateway being connected between an electricity grid of an electricity provider, configured to provide electricity to the gateway, and the plurality of charging points, the gateway including a controller and an electrical storage device that stores electrical energy. The method, operated by the controller, includes determining the electricity need of the plurality of charging points at a given time, if the electricity need of the plurality of charging points is greater than zero, determining the electrical consumption of the site, and if the electrical consumption of the site is greater than a predetermined site limit, providing electrical energy to the plurality of charging points using electricity at least partially received from the electrical storage device.

A method of power control for a textile machine with simultaneously operated workstations and with at least one power supply unit for jointly powering the workstations, a power control unit and a spinning machine. The method enables rapid run-up of the textile machine and minimise delays and downtimes. The method comprises steps that initially involve determining the required electrical power output of each of the workstations, then determining the simultaneously executable operations on the basis of the determined required power and a maximum power output of the power supply unit. Optimised operating data of the several workstations powered jointly by a power supply unit can be obtained in order to make the best possible use of the maximum power output of a power supply unit, wherein the required electrical power output and/or the simultaneously executable operations are determined, and/or optimised operating data is obtained taking into consideration current batch data.

Disclosed herein are AI-based platforms for enabling intelligent orchestration and management of power and energy. In various embodiments, a policy and governance engine is configured to deploy a set of rules and/or policies that govern a set of energy generation, storage, and/or consumption workloads, wherein the rules and/or policies are associated with a configuration of a set of edge devices operating in local data communication with a set of energy generation facilities, energy storage facilities, energy delivery facilities or energy consumption systems. In some embodiments, upon configuration in the policy and governance engine, a policy associated with an energy generation instruction is automatically applied by at least one of the edge devices to control energy generation by at least one energy generation system that is controlled via the edge device.

Disclosed herein are AI-based platforms for enabling intelligent orchestration and management of power and energy. In various embodiments, a data processing system is configured to fuse at least one entity of an energy grid entity generation, storage, delivery or consumption grid data set with at least one entity of an off-grid energy entity generation, storage, delivery and/or consumption data set. In some embodiments, the data processing system is configured to automatically time align energy grid entity data with off-grid energy entity data. In some embodiments, the data processing system is configured to automatically collect off-grid energy entity sensor data from a set of edge devices via which a set of off-grid energy entities are controlled. In some embodiments, the data processing system automatically normalizes the energy grid entity data and the off-grid energy entity data such as to present the data according to a set of common units.

Disclosed herein are AI-based platforms for enabling intelligent orchestration and management of power and energy. In various embodiments, the AI-based platform includes a set of adaptive, autonomous data handling systems for energy data collection and transmission from a set of edge networking devices via which a set of distributed energy entities are controlled, wherein the data handling systems are trained based on a training data set to recognize a set of events and/or signals that indicate the energy patterns of the set of distributed energy resources. In some embodiments, the distributed energy entities include at least one energy generation resource, energy consuming entities, and/or energy storage resources. In various embodiments, the training data set includes historical energy generation data and/or historical energy consumption data for a set of entities similar to the entities controlled via the edge networking devices.

A system refers to actual weather data made publicly available by a first institution, and creates a model that uses a value of a weather element for each section as an input and uses a value of a renewable energy power generation amount of the area as an output based on the actual value of the weather element calculated for each section, and the actual value of the renewable energy power generation amount of the area. The system refers to weather prediction data made publicly available by the second institution, and calculates an actual value of the weather element regarding each of the plurality of sections including the area based on a prediction value of the weather element for each segment in the corresponding section, and calculates a prediction value of the renewable energy power generation amount based on the prediction value of the weather element for each section.

A utility employs a method for generating available operating reserve. Electric power consumption by at least one device serviced by the utility is determined during at least one period of time to produce power consumption data, stored in a repository. A determination is made that a control event is to occur during which power is to be reduced to one or more devices. Prior to the control event and under an assumption that it is not to occur, power consumption behavior expected of the device(s) is generated for a time period during which the control event is expected to occur based on the stored power consumption data. Additionally, prior to the control event, projected energy savings resulting from the control event, and associated with a power supply value (PSV) are determined based on the devices' power consumption behavior. An amount of available operating reserve is determined based on the projected energy savings.

An energy-saving control method includes obtaining a demand instruction configured to indicate an energy-saving operation and a performing time slot of the energy-saving operation. The energy-saving operation is determined based on load data of a power grid and is configured to reduce a load of the power grid. The method further includes transmitting a first control message configured to control an authorized device to perform the energy-saving operation in response to detecting that time reaches the performing time slot, and transmitting a second control message configured to control the authorized device to stop performing the energy-saving operation in response to detecting that time is outside the performing time slot.

A device may receive state data, actions, and rewards associated with a network of RL agents monitoring a microgrid environment, and may model the network of RL agents as a spatiotemporal representation. The device may represent interactions of the RL agents as edge attributes in the spatiotemporal representation, and may determine edge attributes, transmissibility, connectedness, and communication delay for each of the RL agents in the spatiotemporal representation. The device may determine, based on the transmissibility, the connectedness, and the communication delay, localized clusters of the RL agents, and may process the localized clusters, with a first machine learning model, to identify consensus master RL agents. The device may process the consensus master RL agents, with a second machine learning model, to identify a final master RL agent for the network of RL agents, and cause the final master RL agent to control the microgrid environment.

A transformer economizer automatically disconnects a main electric power transformer from a grid power line during low-load periods, and automatically reconnects the main transformer to the grid power line during high-load periods, to reduce low-power electricity losses incurred by the main transformer. The main transformer is therefore deenergized and a much smaller auxiliary transformer is energized during low-load periods to reduce the low-power electricity losses incurred by the main transformer. The main transformer is then automatically switched back into service during high-load periods, while the auxiliary transformer is switched out of service. This provides long-term energy, cost, and carbon footprint savings by automatically switching the large transformer's loads to a much smaller auxiliary transformer, and therefore proportionally lower losses, during light-load conditions. Transformer inrush currents are ramped and the secondary voltage remains electrically connected at all times to avoid service interruptions and switching disturbances.