This power supply and demand planning device includes: an output range calculation unit that calculates the output range of a power generator that satisfies a plurality of restriction conditions; a power generation output range output calculation unit that calculates the power generator output in a single cross section on the basis of the calculated output range calculated; and a past specified cross section output correction unit that calculates a target output in the single cross section when a restriction condition violation occurs in the power generator output in the calculated single cross section calculated, and in order to eliminate a restriction condition violation, corrects the output range and the power generator output in the single cross section and a past cross section further in the past than the single cross section so that the power generator output in the single cross section becomes the target output.

A power management system, comprises a power generating unit, a power output unit to distribute the electrical power generated by the power generating unit to a household and to a receiving unit, different from the household, wherein the receiving unit is a battery and/or a power grid, a grid power output unit to output electrical power supplied from a power grid to the household and/or to the receiving unit, a condition requirement setting unit to receive condition requirement data and a time period after which the receiving unit has to satisfy the required condition, a prediction data input unit to receive prediction data that indicates a prediction of the electrical power generated by the power generating unit over the time period, a control unit that is adapted to receive the condition requirement data from the condition requirement setting unit and the prediction data from the prediction data input unit.

The present disclosure is directed to systems and methods for economically optimal control of an electrical system. Some embodiments employ generalized multivariable constrained continuous optimization techniques to determine an optimal control sequence over a future time domain in the presence of any number of costs, savings opportunities (value streams), and constraints. Some embodiments also include control methods that enable infrequent recalculation of the optimal setpoints. Some embodiments may include a battery degradation model that, working in conjunction with the economic optimizer, enables the most economical use of any type of battery. Some embodiments include techniques for load and generation learning and prediction. Some embodiments include consideration of external data, such as weather.

Embodiments implement non-intrusive load monitoring using machine learning. A trained convolutional neural network (CNN) can be stored, where the CNN includes a plurality of layers, and the CNN is trained to predict disaggregated target device energy usage data from within source location energy usage data based on training data including labeled energy usage data from a plurality of source locations. Input data can be received including energy usage data at a source location over a period of time. Disaggregated target device energy usage can be predicted, using the trained CNN, based on the input data.

In one aspect, a method to determine a capacity of a microgrid includes applying a current test load to the microgrid and measuring a current through an energy storage device, the current indicating a charging status of the energy storage device based on a current load being applied to the microgrid through activated power outlets being served by the microgrid and the current test load, the energy storage device being integrated with the microgrid. The method also includes, responsive to a determination that the measured current based on the current load being applied to the microgrid and the current test load indicates that the energy storage device is discharging, determining the capacity of the microgrid, wherein the capacity is the current load being applied to the microgrid through activated power outlets and a test load applied to the microgrid immediately preceding the current test load.

System, method, and interface for visualized resource allocation and algorithms for the reallocation of resources to achieve a goal. The system analyses an initial state of resource allocation, a cost function for undesirable resources, and a set of potential incremental improvements, each with an associated cost, and determines a step-wise path of applying the incremental improvements to achieve an ultimate resource-allocation goal in an economically feasible way. Simultaneously, a user interface depicts the state of the allocation at the beginning, at the end, and along the path, allowing an intuitive understanding of how the goal will be achieved.

Systems and methods for operating a hybrid power system are disclosed. A controller may perform operations, including: obtaining a load forecast; obtaining a power availability forecast and an energy cost forecast for each power asset group; performing at least one prospective optimization to determine scheduled active power commands for the groups that optimize a total operating cost; tracking an on-line load; tracking an on-line power availability and an on-line energy cost for the groups; performing at least one on-line optimization to determine on-line active power commands for the groups that (i) account for variance between the load forecast and the on-line load, (ii) account for variance between the power availability forecast and the on-line power availability, and (iii) optimize the total operating cost; and operating the groups based on the scheduled active power commands and the on-line active power commands.

Systems and methods for operating a hybrid power system are disclosed. A controller may perform operations, including: obtaining load data for the hybrid power system; obtaining power availability data and energy cost data for each power asset in each power asset group of a plurality of power asset groups; and determining active power commands for each power asset by performing at least one optimization, such that the determined active power commands optimize a total operating cost, wherein: the at least one optimization is based on at least one cost function that accounts for asset degradation, asset maintenance cost, asset operation efficiency cost, and the energy cost data; and the at least one optimization is constrained by a plurality of constraints based on the load data, the power availability data, and characteristics of the power assets; and operating each power asset based on the determined active power commands.

An embedded power supply apparatus is partially buried in an enclosed structure, is configured to provide a first DC voltage to a plurality of electronic devices, and includes a first power conversion circuit, a plurality of switch circuits, a human-machine interface module and a control circuit. The first power conversion circuit is configured to convert an input AC voltage into the first DC voltage and provide the first DC voltage to the switch circuits. The switch circuits each is configured to selectively transmit the first DC voltage to a corresponding electronic device of the electronic devices according to a corresponding first control signal of a plurality of first control signals. The control circuit is configured to receive a second control signal generated by the human-machine interface module, and generate the first control signals to the switch circuits according to the second control signal.

A non-intrusive load identification method based on the Power Fingerprint characteristics of the load is provided. The method includes: S1, collecting Power Fingerprint characteristic data of several loads of the same type; S2, after preprocessing Power Fingerprint characteristic data of load, establishing convolution neural network based on attention mechanism to learn load characteristics; S3, using sliding time window algorithm to realize load switching event detection, In order to extract the change of electrical data of user bus before and after the switching event, the non-intrusive load identification problem is converted into the single load identification problem; S4, the load identification is realized, and the extracted electrical information features of single load are identified using the trained model. The provided fingerprint feature recognition model can identify and separate the unique load Power Fingerprint feature information, and realize load identification, to solve the practical problem of non-intrusive load identification in complex scenes.