A common enclosure includes a housing, inverter input connectors and an inverter output connector coupled to the housing, a common DC bus mechanically coupled to the housing and electrically coupled to the inverter input connectors, a common AC bus mechanically coupled to the housing and electrically coupled between the inverter output connector and a power grid connector, a controller mechanically coupled to the housing and electrically coupled to the common DC and AC buses, local controllers coupled to the inverters, decentralized controllers coupled to the local controllers, and a centralized controller in communication with the local controllers. The decentralized controllers generate decentralized control signals for the local controllers based on measured voltages and currents of the electrical power grid and the inverters. The centralized controller transmits centralized control signals to the local controllers to maintain a constant voltage on the common DC bus based on a predicted DC load.

Disclosed herein are systems and methods for energy management. A system (e.g., a vehicle) includes energy storage units that include a supercapacitor and an electrochemical battery. The system includes a communication interface that receives an indication of a requested process to be powered using at least a subset of the plurality of energy storage units. The system includes an energy controller that tracks historical power draw from the energy storage units over time in power tracking data, and that identifies a power draw for the requested process based on the power tracking data. The energy controller switches between a first configuration and a second configuration for the requested process based on the identified power draw for the requested process. The first configuration draws power from the electrochemical battery and disconnecting from the supercapacitor, while the second configuration draws power from the supercapacitor and disconnecting from the electrochemical battery.

Systems and methods may use a low speed controller in addition to an economic optimizer to achieve long-term objectives without significantly disrupting or destabilizing an electrical system. Specific long-term objectives include maximizing a capacity factor incentive and regulating battery degradation, but the methods and systems herein can be extended to more long-term objectives. A low speed controller can adjust one or more parameters of a cost function based on the relation between the projected state of the electrical system and the one or more parameters to effectuate a change to the electrical system to attempt to comply with the long-term objective.

Certain aspects of the present disclosure relate to a local energy management system (LEMS) at local mixed power generating sites for providing grid services and grid service applications. The LEMS generally serves as a local power control agent for facilitating energy management at the local site level by controlling and leveraging a plurality of local assets deployed at the local site, and combining a plurality of generated power from each site which acts as its own virtual power plant for delivering grid services to the grid. In addition, the LEMS has the ability to effectively handle and fulfill energy and electrical objectives of the grid services, including regulation or demand response objectives from the grid, by conveying operational set points that control the power charge and discharge at each local asset in order to meet those objectives.

A power management device for intelligently splitting and controlling a high power outlet is described. The device includes a housing, power input, outlets, power sensor, and a controller. The housing has an internal compartment configured to hold components of the device and an exterior surface. The device has a plurality of outlets on the exterior surface of the housing. Each outlet is configured to connect to an appliance. Each outlet has a power sensor configured to sense current draw and/or power use at the outlet. The controller of the device is contained within the internal compartment of the housing and monitors the usage of each outlet via readings from the power sensors. The controller determines, based on the monitored usage and appliance parameters, one or more outlets to provide current to, and causes current to be provided to the determined outlet(s).

An electric power management system is a system that performs an exchange of electric power with an electric power system of an electric power company that is a counterparty of the exchange of the electric power, and includes a plurality of the vehicles, each including a battery, and a server that manages an exchange of the electric power between the battery of each of the vehicles and the electric power system. The server manages the exchange of the electric power for each vehicle group in which the vehicles are bundled, and configures the vehicle groups in advance such that distributions of the electric power supply and demand characteristics of the batteries of the vehicles included in the vehicle groups are the same or similar.

An example operation includes one or more of determining a modified use of energy by an entity, based on a current environmental factor, and providing an energy offset to the entity based on the modified use. The offset derives an alternate benefit to overcome a negative balance by another entity.

A power prediction system includes a battery removably mounted on an electric power device using electric power, a charging device configured to charge the battery, and a power prediction device configured to predict an amount of electric power capable of being supplied by the charging device to outside of the charging device through machine learning on the basis of usage information indicating at least one of the usage state and the usage environment of the charging device.

Embodiments implement non-intrusive load monitoring using ensemble machine learning techniques. A first trained machine learning model configured to disaggregate target device energy usage from source location energy usage and a second trained machine learning model configured to detect device energy usage from source location energy usage can be stored, where the first trained machine learning model is trained to predict an amount of energy usage for the target device and the second trained machine learning model is trained to predict when a target device has used energy. Source location energy usage over a period of time can be received, where the source location energy usage includes energy consumed by the target device. An amount of disaggregated target device energy usage over the period of time can be predicted, using the first and second trained machine learning models, based on the received source location energy usage.

The present invention pertains to an energy management device 50 of a microgrid S1 which is associated with a power system 1 and is provided with a power storage device 15, the energy management device calculating a target value of received power of the microgrid, the target value optimizing the utilization efficiency of the energy of the microgrid S1, on the basis of power demand prediction in the microgrid S1 by taking, as constraint conditions, the upper and lower limits of the received power of the microgrid S1 and the upper and lower limits of output power of the power storage device.