Apparatus and method for dynamic prediction, aggregation, and validation

Abstract

Techniques for dynamic forecasting, aggregation, and validation (120) can include circuitry configured to perform collecting data (304) indicative of power flow at a plurality of locations in an electric grid, receiving one or more parameters (332) for generating a customized forecast indicative of a predicted power flow associated with one or more of the plurality of locations for a defined period, selecting a subset of the collected data (344) that satisfies the one or more parameters, generating a model (348) for predicting power flow associated with the one or more locations in the electric grid, determining whether the model is validated (362) by determining whether a predicted power production minus a predicted loss is within a predefined range of a predicted power consumption at the one or more locations, and generating the customized forecast (364) of the predicted power flow associated with the one or more locations for the defined period.

Description

[0001] This patent application claims priority to U.S. Application No. 16 / 405,000, filed May 7, 2019, and entitled “Techniques for Dynamic Forecasting, Aggregation, and Validation,” which is incorporated by reference as if its entirety. BACKGROUND

[0002] Renewable energy encourages a decentralized approach to generating and owning power. In the future, as distributed energy resources (DERs) replace large base loads, the economic dispatch and management system (MS) of utilities and retailers will become increasingly complex. The evolving energy market includes customers that are adding new generation and storage resources and utility companies and / or third-party energy marketers that are signing contracts with different terms to sell energy to customers. However, existing forecasting models for predicting power flow (e.g., consumption and / or production) are based on a specific load or generation point and are not able to account for changes in available resources in the grid or produce predictions related to specific types of power consumers or power producers that exist in the grid. SUMMARY

[0003] In an aspect, the disclosure provides an apparatus. The apparatus includes circuitry configured to collect data indicative of power flow at a plurality of locations in a power grid. Further, the circuitry is configured to receive one or more parameters for generating a customized forecast indicative of a predicted power flow associated with one or more of the plurality of locations in the power grid for a defined period. Further, the circuitry is configured to select a subset of the collected data that satisfies the one or more parameters. The circuitry is also configured to generate, from the selected subset of the collected data, a model for predicting power flow associated with the one or more locations in the power grid, determine whether the model is validated by determining whether a predicted power production minus a predicted loss is within a predefined range of a predicted power consumption at the one or more locations in the power grid, and generate, in response to determining that the model is validated and based on the one or more parameters, the customized forecast of the predicted power flow associated with the one or more locations for the defined period.

[0004] In another aspect, the present disclosure provides a method. The method includes collecting, by an apparatus, data indicative of power flow at a plurality of locations in an electrical grid. The method also includes receiving, by the apparatus, one or more parameters for generating a customized forecast indicative of a predicted power flow associated with one or more of the plurality of locations in the electrical grid for a defined period. The method further includes selecting, by the apparatus, a subset of the collected data that satisfies the one or more parameters. The method also includes generating, by the apparatus and from the selected subset of the collected data, a model for predicting power flow in the electrical grid associated with the one or more locations. In addition, the method includes determining, by the apparatus, whether the model is validated by determining whether a predicted power production minus a predicted loss is within a predefined range of a predicted power consumption at the one or more locations in the electrical grid. Moreover, the method includes generating, by the apparatus, the customized forecast of the predicted power flow associated with the one or more locations for the defined period in response to determining that the model is validated and based on the one or more parameters.

[0005] In yet another aspect, the present disclosure provides one or more machine-readable storage media having stored thereon multiple instructions that, in response to execution, cause an apparatus to collect data indicative of power flow at a plurality of locations in an electrical grid. The instructions also cause the apparatus to receive one or more parameters for generating a customized forecast indicative of a predicted power flow associated with one or more of the plurality of locations in the electrical grid for a defined period. In addition, the instructions cause the apparatus to select a subset of the collected data. The subset satisfies the one or more parameters. Moreover, the instructions cause the apparatus to generate, from the selected subset of the collected data, a model for predicting power flow in the electrical grid associated with the one or more locations and determine whether the model is validated by determining whether a predicted power production minus a predicted loss is within a predefined range of a predicted power consumption at the one or more locations in the electrical grid. Furthermore, the instructions cause the apparatus to generate the customized forecast of the predicted power flow associated with the one or more locations for the defined period in response to determining that the model is validated and based on the one or more parameters. BRIEF DESCRIPTION OF DRAWINGS

[0006] The concepts described herein are illustrated by way of example in the accompanying drawings, in which like references indicate similar elements, and in which:

[0007] Figure 1 is a simplified block diagram of at least one embodiment of a system for aggregating data from an electrical grid, validating a prediction model for power flow in the electrical grid, and providing a customized forecast;

[0008] Figure 2 is a simplified block diagram of at least one embodiment of a prediction computing device included in Figure 1 the system of

[0009] Figures 3 to 5 is a simplified block diagram of at least one embodiment of a prediction computing device included in Figure 1 and Figure 2 is a simplified block diagram of at least one embodiment of a method that can be performed by the prediction computing device of the system of

[0010] While the concepts of the present disclosure are susceptible to various modifications and alternative forms, specific embodiments thereof have been shown by way of example in the drawings and will herein be described in detail. It should be understood, however, that there is no intent to limit the concepts of the present disclosure to the particular forms disclosed, but on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the present disclosure and the appended claims.

[0011] References in the specification to “one embodiment,” “an embodiment,” “an illustrative embodiment,” etc. indicate that the embodiment described can include a particular feature, structure, or characteristic, but every embodiment can or can not necessarily include the particular feature, structure, or characteristic. Moreover, such phrases are not necessarily referring to the same embodiment. Further, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is submitted that it is within the purview of one of ordinary skill in the art to effect such feature, structure, or characteristic in connection with other embodiments whether or not explicit

[0012] In certain scenarios, the disclosed embodiments can be implemented by hardware, firmware, software, or any combination thereof. The disclosed embodiments can also be implemented as instructions carried by or stored on a transitory or non-transitory machine-readable (e.g., computer- readable) storage medium, which can be read and executed by one or more processors. A machine-readable storage medium can be implemented as any storage device, mechanism, or other physical structure for storing or transmitting information in a form readable by a machine, e.g., a volatile or non-volatile memory, media disc, or other media device. In the drawings, certain structural or methodical features can be shown in specific arrangements and / or order. It should be understood, however, that these specific arrangements and / or order can not be required. Rather, in some embodiments, such features can be arranged in a different manner and / or order than shown in the illustrative figures. Additionally, the inclusion of a structural or methodical feature in a particular figure does not imply that such feature is required in all embodiments, and the features can be combined with or

[0013] Reference will now be made to Figure 1, for aggregating data from a power grid 110, validating predictive models for power flow in the power grid 110, and providing customized predictions, the system 100 includes a prediction computing device 120 that communicates with components of the power grid 110 and a client computing device 122 over a network 130. The power grid 110 can include power producers, including power plants 140, 142 (e.g., combined heat and power (CHP) plants), solar power plants 144, and wind power plants. Further, in the illustrative embodiment, the power grid includes power consumers, including houses 150, 152, office buildings 154, and factories 158. Further, the power grid 110 can include a house with a home CHP 160 (e.g., a house with a device for producing combined heat and power, such as a house with micro-CHP technology). The power grid 110 can also include other devices capable of managing power flow through the power grid 110, including energy storage devices (e.g., batteries) 170, 172, 174, 176, flow control devices 180, and power quality devices 182, 184 (e.g., devices configured to maintain power at a target quality by continuously monitoring and adjusting the voltage, frequency, and / or waveform of the power). The power grid 110 can also include feeders 190, 192, 194, each of which is implemented as a location at which power from different producers can be combined and provided to different consumers in the power grid 110. In operation, as the combination and location of distributed energy resources (DERs) evolves and changes, the prediction computing device 120 enables distribution grid operators to predict loads and power generation on their networks (e.g., the power grid 10). By predicting available power generation and load obligations, distribution grid operators can operate the power grid 110 more reliably and efficiently, benefiting market participants, integrators, and individual consumers of power. Rather than relying on a predefined set of load and / or power generation points and their corresponding historical data and associated independent variables, the prediction computing device 120, in predicting (e.g., in response to a prediction request from the client computing device 122), combines historical data and corresponding independent data based on user-defined parameters common to a target subset of collected data (e.g., a subset of collected data related to a particular type of power consumer). For example, the prediction computing device 120 can produce power flow predictions for customers in a particular geographic region and / or associated with a particular type of electrical device (e.g., a solar power plant, a particular feeder, etc.).

[0014] In an illustrative embodiment, the predictive computing device 120 can generate predictions at the account (e.g., electricity consumer, such as house 150), feeder, or aggregator (e.g., multiple feeders) level upon request. In each scenario, the predictive computing device 120 utilizes collected data (e.g., historical data) indicative of power flow in the power grid 110, analyzes the data using statistical techniques and the like to determine relationships and patterns, and develops one or more models to determine (e.g., over time, considering weather changes, etc.) how much electricity is generated and consumed. The predictive computing device 120 can generate models for any account category or subset of accounts. For example, residences with similar solar installations in the same location can be considered a category. Thus, the predictive computing device 120 uses data indicative of the topology of the power grid 110 (e.g., data indicative of electrical equipment installed in the power grid 110) to illustrate (e.g., model) power losses due to the presence of electrical equipment (e.g., due to inefficiency of electrical equipment). Furthermore, the forecasting calculation device 120 verifies that a given model (e.g., a power flow model at feeder 190) considers all power flows that might affect the forecast by confirming that the forecasted power output minus losses due to known electrical equipment is within a predefined range (e.g., equal to a certain percentage (e.g., 1%) or around a certain percentage (e.g., 1%)) of the forecasted power consumption of the power consumers associated with the forecast (e.g., power consumers 150, 156 connected to feeder 190). In other words, in operation, the forecasting calculation device 120 can determine whether the model conforms to Kirchhoff's first law, which states that the current at a given node must sum to zero before providing the forecast generated by the model to the requester of the forecast (e.g., the operator of the client calculation device 122).

[0015] Now for reference Figure 2 The predictive computing device 120 can be implemented as any type of device capable of performing the functions described herein. For example... Figure 2 As shown, the illustrative predictive computing device 120 includes a computing engine 210, an input / output (110) subsystem 216, communication circuitry 218, and a data storage subsystem 222. Of course, in other embodiments, the predictive computing device 120 may include other or additional components, such as components commonly found in computers (e.g., a display, etc.). Additionally, in some embodiments, one or more of the illustrative components may be incorporated into another component or otherwise formed as part of another component.

[0016] The computing engine 210 can be implemented as any type of device or collection of devices capable of performing the various computing functions described below. In some embodiments, the computing engine 210 can be implemented as a single device, such as an integrated circuit, an embedded system, a field-programmable gate array (FPGA), a system on a chip (SOC), or other integrated system or device. Further, in some embodiments, the computing engine 210 includes or is implemented as the processor 212 and the memory 214. The processor 212 can be implemented as any type of processor capable of performing the functions described herein. For example, the processor 212 can be implemented as a microcontroller, single- or multi-core processor(s), or other processor or processing / control circuit. In some embodiments, the processor 212 can be implemented as, include, or be coupled to an FPGA, an application-specific integrated circuit (ASIC), reconfigurable hardware or hardware circuit, or other specialized hardware that facilitates performance of the functions described herein. In illustrative embodiments, the processor 212 includes a prediction logic unit 230, which can be implemented as any device or circuit (e.g., reconfigurable circuit, field-programmable gate array (FPGA), application-specific integrated circuit (ASIC), etc.) capable of offloading functions related to analyzing a set of collected data from the power grid 110 to produce a customized prediction related to a particular subset of the collected data, producing one or more models to generate the customized prediction, and validating the model(s) (e.g., determining whether the model(s) adhere to Kirchhoff’s First Law, as described above) from other functions of the processor 212. Although shown as being integrated into the processor 212, in some embodiments, the prediction logic unit 230 can be located in a different part of the prediction computing device 120 (e.g., as a discrete unit).

[0017] The main memory 214 can be implemented as any type of volatile memory (e.g., dynamic random access memory (DRAM), etc.), non-volatile memory, or data storage that is capable of performing the functions described herein. Volatile memory can be a storage medium that requires power to maintain the data state of the medium. In some embodiments, all or a portion of the main memory 214 can be integrated into the processor 212. In operation, the main memory 214 can store various software and data used during operation, such as data indicative of power flow at one or more locations in the power grid 110, one or more models used to predict power flow in the power grid 110, applications, programs, libraries, and drivers.

[0018] The compute engine 210 is communicatively coupled to other components of the prediction computing device 120 via an I / O subsystem 216, which can be implemented as circuitry and / or components to facilitate input / output operations with the compute engine 210 (e.g., with the processor 212, the prediction logic unit 230, the main memory 214) and other components of the prediction computing device 120. For example, the I / O subsystem 216 can be implemented as or otherwise include a memory controller hub, an input / output control hub, an integrated sensor hub, firmware devices, communication links (e.g., point-to-point links, bus links, wires, cables, light guides, printed circuit board traces, etc.), and / or other components and subsystems to facilitate input / output operations. In some embodiments, the I / O subsystem 216 can form a portion of a system on a chip (SoC) and be incorporated with one or more of the processor 212, the main memory 214, and other components of the prediction computing device 120 to form the compute engine 210.

[0019] The communication circuitry 218 can be implemented as any communication circuit, device, or collection thereof, capable of enabling communications over a network between the prediction computing device 120 and another device (e.g., the client computing device 122, components of the power grid 110, etc.). The communication circuitry 218 can be configured to implement communications WiMAX, etc.) using any one or more communication technology (e.g., wired or wireless communications) and associated protocols (e.g., Ethernet, Bluetooth®, Wi-Fi, tactile internet, etc.).

[0020] The illustrative communication circuitry 218 includes a network interface controller (NIC) 220. The NIC 220 can be implemented as one or more add-in boards, daughterboards, network interface cards, controller chips, chipsets, or other devices that the prediction computing device 120 can use to connect with another device. In some embodiments, the NIC 220 can be implemented as part of a system on a chip (SoC) that includes one or more processors or on a multi-chip package that also includes one or more processors. In some embodiments, the NIC 220 can include a local processor (not shown) and / or a local memory (not shown), both of which are local to the NIC 220. In these embodiments, the local processor of the NIC 220 can perform one or more functions of the processor 212. Additionally or alternatively, in these embodiments, the local memory of the NIC 218 can be integrated at a board level, a socket level, a chip level, and / or other level into one or more components of the prediction computing device 120.

[0021] The data storage subsystem 222 can be implemented as any type of device configured for short-term or long-term data storage, such as memory devices and circuitry, memory cards, hard disk drives, solid-state drives, or other data storage devices. In an illustrative embodiment, the data storage subsystem includes data collected from the power grid 110 indicating power flow at multiple locations over time, weather data indicating weather at these locations in the power grid 110 over time, data indicating the location and type of electrical equipment present in the power grid 110, and data indicating the electricity producers and consumers present in the power grid 110 (e.g., defining the topology of the power grid 110).

[0022] Client computing device 122 may have the same characteristics as Figure 2 The components described in the reference to predictive computing device 120 are similar to those in the reference to predictive computing device 120. The descriptions of those components in predictive computing device 120 also apply to the descriptions of the components in client computing device 122, except that client computing device 122 may not include predictive logic unit 230 in the illustrative embodiment. Furthermore, it should be understood that either predictive computing device 120 or client computing device 122 may include other components, sub-components, and devices typically found in computing devices that have not been discussed above with reference to predictive computing device 120 and are not discussed herein for clarity. Similarly, devices 140, 142, 144, 146, 150, 152, 154, 156, 160, 170, 172, 174, 176, 180, 182, 184, 190, 192, and 194 in power grid 110 may include components similar to those in predictive computing device 120 and client computing device 122.

[0023] Predictive computing device 120, client computing device 122, and devices 140, 142, 144, 146, 150, 152, 154, 156, 160, 170, 172, 174, 176, 180, 182, 184, 190, 192, 194 exemplarily communicate via network 130, which can be implemented as any type of wired or wireless communication network capable of transmitting data, including global networks (e.g., the Internet), local area networks (LANs) or wide area networks (WANs), cellular networks (e.g., Global System for Mobile Communications (GSM), 3G, Long Term Evolution (LTE), Global Microwave Access Interoperability (WiMAX), etc.), digital subscriber line (DSL) networks, cable networks (e.g., coaxial networks, fiber optic networks, etc.), or any combination thereof.

[0024] Now for reference Figure 3In operation, the prediction computing device 120 can perform the method 300 for aggregating data from a power grid (e.g., the power grid 110), validating a prediction model for power flow in the power grid 110, and providing customized predictions (e.g., to the client computing device 122). The method 300 begins at block 302. In block 302, the prediction computing device 120 determines whether dynamic prediction is enabled (e.g., whether to perform the remainder of the method 300). As such, the prediction computing device 120 can determine that dynamic prediction is enabled in response to determining that the prediction computing device 120 has received a request to enable dynamic prediction (e.g., from the client computing device 122), in response to determining that the prediction computing device 120 is equipped with the prediction logic 230, and / or based on other factors. Regardless, in response to determining that dynamic prediction is enabled, the method 300 proceeds to block 304. In block 304, the prediction computing device 120 collects power production data, which can be implemented as any data indicative of power flow at a plurality of locations in a power grid (e.g., the power grid 110). As such, as shown in block 306, the prediction computing device 120 can collect power production data, which can be implemented as any data indicative of an amount of power produced over time at locations in the power grid 110 (e.g., at the power plants 140, 142, the solar power plant 144, and the wind power plant 146, among others). As shown in block 308, the prediction computing device 120 also collects power consumption data, which can be implemented as any data indicative of an amount of power consumed over time at locations in the power grid 110 (e.g., at the houses 150, 152, the office building 154, and the factory 158). Additionally, as shown in block 310, in illustrative embodiments, the prediction computing device 120 collects data from one or more feeders 190, 192, 194 (e.g., data indicative of power flow into the feeder, data indicative of power flow out of the feeder, and data indicative of power loss due to inefficiencies of electrical equipment associated with the feeder).

[0025] As represented by block 312, the prediction computing device 120 stores metadata (e.g., tags) indicative of attributes of the source of the collected data data in association with the collected data. For example, as represented by block 314, the prediction computing device 120 can store metadata indicative of a location in the power grid 110 at which a received data set was generated (e.g., by associating an internet protocol address of a device in the power grid 110 that sent data to the prediction computing device 120 with corresponding location data, which can be implemented as geographic coordinates or other identifier indicative of a location within the power grid 110). As represented by block 316, the prediction computing device 120 can store metadata indicative of a type of electrical device (e.g., transformer, power quality device, flow control device, etc.) associated with a location at which collected data was generated. As represented by block 318, the prediction computing device 120 can store metadata indicative of an electrical device that generated power (e.g., data indicative of a particular device type associated with power production at a particular location). For example, as represented by block 320, the prediction computing device 120 can store metadata indicative of a photovoltaic cell(s) associated with a location (e.g., a location of the solar power plant 144). Similarly, as represented by block 322, the prediction computing device 120 can store metadata indicative of a wind turbine(s) associated with a location (e.g., a location of the wind power plant 146). As represented by block 324, the prediction computing device 120 can store metadata indicative of one or more energy storage devices associated with one or more locations (e.g., locations of the energy storage devices 170, 172, 174, 176). As represented by block 326, in illustrative embodiments, the prediction computing device 120 stores metadata indicative of an electrical device that consumes power (e.g., a location of a house 150, 152, a factory 158, an office building 154, 156, etc.). Further, in illustrative embodiments, the prediction computing device 120 stores metadata indicative of a device (transformer, power quality device, flow control device, etc.) that causes a loss of power associated with a feeder (e.g., a loss that can be considered when summing power flow at a feeder). Additionally, the prediction computing device 120 can store weather data (e.g., temperature, atmospheric conditions, wind speed and direction, duration and intensity of sunlight, etc.) associated with a location in the power grid 110, as represented by block 330. In some embodiments, the prediction computing device 120 can collect additional data, including data indicative of a configuration of the power grid (e.g., network topology), a capacity of the power grid (e.g., nameplate rating(s)), a state of the power grid (e.g., circuit breaker settings), and / or expert assessments of the power grid (e.g., maintenance records). Subsequently, the method 300 proceeds to Figure 4Box 332. In box 332, the prediction computing device 120 (e.g., from the client computing device 122) receives one or more parameters that can be used to generate a custom prediction indicating the predicted power flow in the power grid (e.g., one or more specific portions of the power grid 110) over a defined time period.

[0026] Now for reference Figure 4 When receiving one or more parameters, the forecasting calculation device 120 may receive parameters indicating one or more locations (e.g., geographical areas) in the power grid to be involved in the forecast, as shown in box 334. As shown in box 336, the forecasting calculation device 120 may additionally or alternatively receive parameters indicating one or more types of electrical equipment to be involved in the forecast (e.g., specifically regarding the forecast of energy generated and consumed by solar power plant 144). As shown in box 338, the forecasting calculation device 120 may receive parameters indicating one or more electricity consumers to be involved in the forecast (e.g., specifically regarding the forecast of electricity generated for and consumed by house 150 and office building 156). As shown in box 340, the forecasting calculation device 120 may receive parameters indicating one or more electricity producers to be involved in the forecast (e.g., specifically regarding the forecast of electricity generated by power plants 140, 144, 146). Additionally or alternatively, the prediction computing device 120 may receive parameters indicating the one or more feeders to be involved in the prediction (e.g., a prediction specifically concerning the power supplied to and consumed by feeder 190), as shown in box 342. Subsequently, in box 344, the prediction computing device 120 selects a subset of the collected data (e.g., from box 304) that satisfies the parameters(s)(e.g., from box 332). Thus, as shown in box 346, the prediction computing device 120 selects from the collected data a subset associated with metadata matching the parameters(s)(e.g., other data containing keywords or indicating the parameters(s)). Method 300 then proceeds to... Figure 5 Box 348. In box 348, the predictive computing device 120 generates one or more models (e.g., each model is a mathematical relationship between independent variables (e.g., time) and dependent variables (e.g., electricity production and consumption)) from a selected subset of collected data to predict power flow in the power grid 110.

[0027] Now for reference Figure 5 When generating one or more models, the predictive computing device 120 can be used for a selected subset of the collected data (e.g., in...). Figure 4The selected subset of the collected data (e.g., the subset selected in block 344) identifies trends in power flow, as indicated by block 350. In addition, as indicated by block 352, the prediction computing device 120 can identify, for the selected subset of data, the impact of weather on power flow (e.g., when the temperature deviates from a reference temperature by a particular amount, power production and consumption increase). In the illustrative embodiment, the prediction computing device 120 performs validation of any models that have been generated, as indicated by block 354. In this way, as indicated by block 356, the prediction computing device 120 applies Kirchhoff's first law to determine whether a predicted power production (e.g., a prediction made by a model of the amount of power that will be produced) minus a predicted loss (e.g., a prediction of the amount of power that will be lost due to inefficiencies of known electrical equipment in the power grid 110) is within a predefined range (e.g., about 1%) of a predicted power consumption amount at one or more locations in the power grid 100 (e.g., one or more locations in the power grid 110 that are related to the subset of collected data). For example, as indicated by block 358, if one or more feeders are associated with the selected parameters (e.g., if power production and consumption data associated with one or more feeders are represented in the subset of collected data), the prediction computing device 120 can apply Kirchhoff's first law to the one or more feeders in the power grid 110. In block 360, if multiple models are generated, the prediction computing device 120 can identify one of the generated models that provides the most accurate prediction of power flow based on historical power flow represented in the collected data. In other words, the prediction computing device 120 uses each model to predict power production and consumption data for a previous time period (in which actual power production and consumption data are known) and determine the accuracy of the model's prediction of the actual power production and consumption amounts.

[0028] In block 362, the prediction computing device 120 determines a subsequent course of action based on whether at least one validated model is available (e.g., validated using operations associated with block 354). If so, the method 300 proceeds to block 364. In block 364, the prediction computing device 120 uses the validated model to generate a prediction of predicted power flow (e.g., predicted production and predicted consumption) for a defined time period (e.g., a future time period for which the requested prediction is to be generated) based on the parameters (e.g., the parameters from block 332). In this way, as indicated by block 366, if multiple validated models are available, the prediction computing device 120 generates the prediction using the model that provides the most accurate prediction from historical power flow represented in the collected data (e.g., the model identified as the most accurate in block 360). After the prediction is generated, the prediction computing device 120 can send data indicative of the prediction to the requesting device (e.g., the client computing device 122), e.g., over the network 130, to provide the prediction to the requesting device.

[0029] Referring back to block 362, if the prediction computing device 120 determines that no validated model is available, the method 300 instead branches to block 368. In block 368, the prediction computing device 120 generates an error message indicating that the collected data is erroneous or incomplete. For example, the prediction computing device 120 can generate an error message indicating that the collected data related to the electrical devices present in the power grid is erroneous or incomplete (e.g., indicating that the collected data for the topology of the power grid 10 is missing data for an electrical device that is present and causes a loss). In response, an operator of the prediction computing device 120 can provide the missing data to the prediction computing device 120 to enable the prediction computing device 120 to generate a model that satisfies the validation process of block 354 (e.g., a model that satisfies Kirchhoff’s first law). Subsequently, after generating the prediction in block 364 or after generating the error message in block 368, the method 300 can return to block 302, in which the prediction computing device 120 can determine whether to continue enabling dynamic predictions. Figure 3

[0030] While certain illustrative embodiments have been described in detail above, such description is to be considered as exemplary only, with the scope of the disclosure being indicated by the appended claims, and understood that changes and modifications can be resorted to without departing from the spirit of the disclosure. Due to the various features and aspects of the apparatus, systems, and methods described herein, the disclosure has numerous advantages. It should be noted that alternative embodiments of the apparatus, systems, and methods of the present disclosure can not include all of the features described, yet still benefit from at least some of the advantages of the features. One of ordinary skill in the art can readily devise their own implementations of apparatus, systems, and methods that incorporate one or more of the features of the present disclosure without departing from the spirit of the disclosure.​

Claims

1. An apparatus comprising: The circuit is configured as follows: Collect data indicating power flow at multiple locations within the power grid; Receive one or more parameters for generating a custom forecast, the custom forecast indicating the predicted power flow associated with one or more locations in a plurality of locations in the power grid within a defined time period; Select a subset of the collected data, wherein the subset satisfies one or more of the parameters; From a selected subset of the collected data, a model is generated for predicting power flows in the power grid associated with the one or more locations; The validation of the model is determined by whether the predicted power output minus the predicted losses falls within a predefined range of predicted power consumption at one or more locations within the power grid; and In response to determining that the model has been validated and based on the one or more parameters, the customized prediction is generated, which indicates the predicted power flow associated with the one or more locations within the defined time period.

2. The apparatus according to claim 1, wherein, Collecting data indicating power flow includes collecting data indicating power output from at least one of a plurality of locations in the power grid.

3. The apparatus according to claim 1, wherein, Collecting data indicating power flow includes collecting data indicating power consumption from at least one of a plurality of locations in the power grid.

4. The apparatus according to claim 1, wherein, Collecting data indicating power flow includes collecting data indicating power flow from one or more feeders of the power grid.

5. The apparatus according to claim 1, wherein, The collection of data indicating power flow includes metadata associated with the collected data, which stores attributes indicating the source of the collected data.

6. The apparatus according to claim 5, wherein, The stored metadata includes metadata indicating the location within the power grid.

7. The apparatus according to claim 5, wherein, The stored metadata includes metadata that indicates the type of electrical equipment.

8. The apparatus according to claim 7, wherein, The storage of metadata indicating the type of electrical equipment includes the storage of metadata indicating the electrical equipment that generates electricity.

9. The apparatus according to claim 7, wherein, The storage includes metadata indicating the type of electrical equipment, including metadata indicating the electrical equipment that consumes power.

10. The apparatus according to claim 7, wherein, The storage includes metadata indicating the type of electrical equipment, including metadata indicating electrical equipment that has lost power.

11. The apparatus according to claim 1, wherein, The circuit is also configured to store weather data that indicates the data associated with the power flow in the power grid.

12. The apparatus according to claim 1, wherein, Receiving one or more parameters includes receiving one or more parameters indicating one or more locations in the power grid to be involved in the customized prediction.

13. The apparatus according to claim 1, wherein, Receiving one or more parameters includes receiving one or more types of electrical equipment to be involved in the customized prediction.

14. The apparatus according to claim 1, wherein, Receiving one or more parameters includes receiving one or more electricity consumers to be involved in the customized forecast.

15. The apparatus according to claim 1, wherein, Receiving one or more parameters includes receiving one or more power producers involved in the customized forecast, or receiving one or more feeders involved in the customized forecast of the power grid.

16. The apparatus according to claim 1, wherein, Selecting a subset of the collected data that satisfies one or more of the parameters includes selecting a subset of data associated with metadata that matches one or more of the parameters.

17. The apparatus according to claim 1, wherein, The generated models include identifying trends in power flow for a selected subset of the collected data, or identifying one or more effects of weather on power flow for a selected subset of the collected data.

18. The apparatus according to claim 1, wherein, The circuit is also configured to collect data indicating the following: the configuration of the power grid, the capacity of the power grid, one or more environmental conditions, the status of the power grid, or an expert assessment of the power grid.

19. A method comprising: The device collects data indicating power flow at multiple locations within the power grid; The device receives one or more parameters for generating a customized forecast, the customized forecast indicating a predicted power flow associated with one or more locations among a plurality of locations in the power grid within a defined time period; The device selects a subset of the collected data, wherein the subset satisfies one or more parameters; Using the device and a selected subset of collected data, a model is generated for predicting power flows in the power grid associated with the one or more locations. The device is used to determine whether the model is validated by identifying whether the predicted power output minus the predicted losses falls within a predefined range of the predicted power consumption at one or more locations in the power grid; and The device, in response to determining that the model has been validated and based on the one or more parameters, generates the customized prediction, which indicates the predicted power flow associated with the one or more locations within the defined time period.

20. The method according to claim 19, wherein, Collecting data indicating power flow includes collecting data indicating power output from at least one of a plurality of locations in the power grid.

21. The method according to claim 19, wherein, Collecting data indicating power flow includes collecting data indicating power consumption from at least one of a plurality of locations in the power grid.

22. The method according to claim 19, wherein, Collecting data indicating power flow includes collecting data indicating power flow from one or more feeders of the power grid.

23. The method according to claim 19, wherein, The collection of data indicating power flow includes metadata associated with the collected data, which stores attributes indicating the source of the collected data.

24. The method according to claim 23, wherein, The stored metadata includes metadata indicating the location within the power grid.

25. The method according to claim 23, wherein, The stored metadata includes metadata that indicates the type of electrical equipment.

26. The method of claim 25, wherein, The storage of metadata indicating the type of electrical equipment includes the storage of metadata indicating the electrical equipment that generates electricity.

27. The method according to claim 25, wherein, The storage includes metadata indicating the type of electrical equipment, including metadata indicating the electrical equipment that consumes power.

28. The method according to claim 25, wherein, The storage includes metadata indicating the type of electrical equipment, including metadata indicating electrical equipment that has lost power.

29. The method of claim 19, further comprising storing weather data that indicates the data relating to the power flow in the power grid.

30. The method according to claim 19, wherein, Receiving one or more parameters includes receiving one or more parameters indicating one or more locations in the power grid to be involved in the customized prediction.

31. The method according to claim 19, wherein, Receiving one or more parameters includes receiving one or more types of electrical equipment to be involved in the customized prediction.

32. The method according to claim 19, wherein, Receiving one or more parameters includes receiving one or more electricity consumers to be involved in the customized forecast.

33. The method according to claim 19, wherein, Receiving one or more parameters includes receiving one or more power producers involved in the customized forecast, or receiving one or more feeders involved in the customized forecast of the power grid.

34. The method according to claim 19, wherein, Selecting a subset of the collected data that satisfies one or more of the parameters includes selecting a subset of data associated with metadata that matches one or more of the parameters.

35. The method according to claim 19, wherein, The generated models include identifying trends in power flow for a selected subset of the collected data, or identifying one or more effects of weather on power flow for a selected subset of the collected data.

36. The method of claim 19 further comprises collecting data indicating the following: the configuration of the power grid, the capacity of the power grid, one or more environmental conditions, the status of the power grid, or an expert assessment of the power grid.

37. One or more machine-readable storage media, including a plurality of instructions stored thereon, the plurality of instructions being responsive to execution, causing the apparatus to: Collect data indicating power flow at multiple locations within the power grid; Receive one or more parameters for generating a custom forecast, the custom forecast indicating the predicted power flow associated with one or more locations in a plurality of locations in the power grid within a defined time period; Select a subset of the collected data, wherein the subset satisfies one or more of the parameters; From a selected subset of the collected data, a model is generated for predicting power flows in the power grid associated with the one or more locations; The validation of the model is determined by whether the predicted power output minus the predicted losses falls within a predefined range of predicted power consumption at one or more locations within the power grid; and In response to determining that the model has been validated and based on the one or more parameters, the customized prediction is generated, which indicates the predicted power flow associated with the one or more locations within the defined time period.

38. One or more machine-readable storage media according to claim 37, wherein, Collecting data indicating power flow includes collecting data indicating power output from at least one of a plurality of locations in the power grid.

39. One or more machine-readable storage media according to claim 37, wherein, Collecting data indicating power flow includes collecting data indicating power consumption from at least one of a plurality of locations in the power grid.

40. One or more machine-readable storage media according to claim 37, wherein, Collecting data indicating power flow includes collecting data indicating power flow from one or more feeders of the power grid.

41. One or more machine-readable storage media according to claim 37, wherein, The collection of data indicating power flow includes metadata associated with the collected data, which stores attributes indicating the source of the collected data.

42. The one or more machine-readable storage media according to claim 41, wherein, The stored metadata includes metadata indicating the location within the power grid.

43. The one or more machine-readable storage media according to claim 41, wherein, The stored metadata includes metadata that indicates the type of electrical equipment.

44. The one or more machine-readable storage media according to claim 43, wherein, The storage of metadata indicating the type of electrical equipment includes the storage of metadata indicating the electrical equipment that generates electricity.

45. One or more machine-readable storage media according to claim 43, wherein, The storage includes metadata indicating the type of electrical equipment, including metadata indicating the electrical equipment that consumes power.

46. ​​The one or more machine-readable storage media according to claim 43, wherein, The storage includes metadata indicating the type of electrical equipment, including metadata indicating electrical equipment that has lost power.

47. One or more machine-readable storage media according to claim 37, wherein, In response to being executed, the plurality of instructions also cause the device to store weather data that indicates the data relating to the power flow in the power grid.

48. The one or more machine-readable storage media according to claim 37, wherein, Receiving one or more parameters includes receiving one or more parameters indicating one or more locations in the power grid to be involved in the customized prediction.

49. One or more machine-readable storage media according to claim 37, wherein, Receiving one or more parameters includes receiving one or more types of electrical equipment to be involved in the customized prediction.

50. One or more machine-readable storage media according to claim 37, wherein, Receiving one or more parameters includes receiving one or more electricity consumers to be involved in the customized forecast.

51. One or more machine-readable storage media according to claim 37, wherein, Receiving one or more parameters includes receiving one or more power producers involved in the customized forecast, or receiving one or more feeders involved in the customized forecast of the power grid.

52. The one or more machine-readable storage media according to claim 37, wherein, Selecting a subset of the collected data that satisfies one or more of the parameters includes selecting a subset of data associated with metadata that matches one or more of the parameters.

53. The one or more machine-readable storage media according to claim 37, wherein, The generated models include identifying trends in power flow for a selected subset of the collected data, or identifying one or more effects of weather on power flow for a selected subset of the collected data.

54. The one or more machine-readable storage media according to claim 37, wherein, In response to being executed, the plurality of instructions also cause the device to collect data indicating the following: the configuration of the power grid, the capacity of the power grid, one or more environmental conditions, the state of the power grid, or an expert assessment of the power grid.

Images