A system with deep reinforcement learning based control determines optimal actions for major components in a commercial building to minimize operation costs while maximizing comprehensive comfort levels of occupants. An unsupervised deep Q-network method is introduced to handle the energy management problem by evaluating the influence of operation costs on comfort levels considering the environment factors at each time slot. An optimum control decision can be derived that targets both immediate and long-term goals, where exploration and exploitation are considered simultaneously.

A method for managing energy for a building includes connecting a power line of the building to a battery of a vehicle via a charger; responsive to occurrence of a power outage, supplying electric energy to the building from the battery and from an energy storage separate from the vehicle; and instructing the vehicle to recharge the battery and return to the building at a predefined time before stored energy of the energy storage falls below a predefined value.

A method for aggregating a group of EVs based on EV flexibility includes: obtaining a prediction result on a fast charging demand of each EV based on a preset model; determining a distribution area of each substation involved in an EV activity range as an aggregation area; establishing an aggregation model based on the EV fast charging demand, in which the aggregation model includes energy balance equations and various constraints; obtaining an EV fast charging load power curve and an EV energy curve of each aggregation area based on the aggregation model; determining upper and lower limits of multiple EV fast charging load power curves as a flexibility range for the EV fast charging load power, and determining upper and lower limits of multiple EV energy curves as a flexibility range for the EV energy; and aggregating a group of EVs based on the above flexibility ranges.

Disclosed is a system and a method of controlling a battery connection of an electric vehicle. The system for controlling a battery connection of an electric vehicle, the system including: a battery unit including a plurality of battery packs; a power relay assembly (PRA) for electrically connecting or disconnecting the battery unit and an inverter, and changing a connection structure of the plurality of battery packs to one of a battery unit serial mode and a battery unit parallel mode through a plurality of relays; and a controller for switching the connection structure to the battery unit serial mode through the PRA in response to a driver's request to increase motor output, and switches the connection structure to the battery unit parallel mode through the PRA in response to a driver's request to increase a cruising distance.

A refuse vehicle including a chassis, a body assembly coupled to the chassis, the body assembly defining a refuse compartment, an electric energy system configured to store power and supply power to the refuse vehicle, and a power control system configured to measure one or more electrical attributes of the refuse vehicle and determine a power profile for the refuse vehicle, the power profile describing a length of time the refuse vehicle can continue to operate based on a remaining power of the electrical energy system, and wherein the power control system controls operation of a lift assembly of the refuse vehicle based on the power profile.

Methods and systems are provided for controlling a high power output direct current to alternating current converter for a vehicle. In one example, a method may include at a vehicle-on event, automatically operating the converter in a first power output mode, and transitioning to a different mode of operation in response to a transition request being received at a controller of the vehicle. In this way, the different mode of operation may be subject to confirmation via an operator of the vehicle, which may improve operational performance of the direct current to alternating current converter.

A wireless battery management system includes a plurality of slave BMSs coupled to a plurality of battery modules in one-to-one correspondence. Each slave BMS is configured to operate in active mode and sleep mode. Each slave BMS is configured to wirelessly transmit a detection signal indicating a state of the battery module. The wireless battery management system further includes a master BMS configured to wirelessly receive the detection signal from each of the plurality of slave BMSs. The master BMS is configured to set a scan cycle and a scan duration for each of the plurality of slave BMSs based on the detection signal, and wirelessly transmit a control signal to the plurality of slave BMSs. The control signal includes a wireless balancing command indicating the scan cycle and the scan duration set for each of the plurality of slave BMSs.

An example operation includes one or more of determining an estimated arrival time of a first transport to a charging station, determining an estimated remaining stored transport energy at the estimated arrival time of the first transport, notifying the first transport to provide a portion of the determined remaining stored transport energy and when a next transport is delayed to the charging station, notifying the first transport to provide an additional portion of the determined remaining stored transport energy based on the delay.

An example operation includes one or more of providing to a charging station a distance desired to be driven by a transport and determining, by the charging station, an amount of energy to retrieve from the transport such that a residual amount of energy in the transport is equivalent to the distance desired to be driven.

A vehicle system and method includes determining that a state of charge of an energy storage assembly of a receiving vehicle is insufficient to power the receiving vehicle to an upcoming location based on a difference between the state of charge and a needed amount of energy from the energy storage assembly to power the receiving vehicle to the upcoming location. The receiving vehicle may be controlled to move to an intermediate location that includes an increased traffic area or to a first donating vehicle location of plural different donating vehicle locations. The first donating vehicle location includes a predicted upcoming location of a first donating vehicle. The receiving vehicle receives energy from the first donating vehicle to charge the energy storage assembly of the receiving vehicle while both the first donating vehicle and the receiving vehicle area moving at the intermediate location.