A method and device for predicting power grid load, electronic equipment and storage medium

Abstract

The application provides a method and device for predicting power grid load, electronic equipment and storage medium. Through the application, when predicting the future load of the power grid in the future time period, multiple factors that will affect the future load of the power grid in the future time period are considered, for example, historical load, current load, historical additional electricity-related event, current additional electricity-related event, future additional electricity-related event, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity price, current electricity price and future electricity price, etc. The considered factors are more comprehensive, so the accuracy of predicting the future load of the power grid in the future time period can be improved. In addition, the process of predicting the future load of the power grid in the future time period can be free of human intervention, reducing labor costs and avoiding the problem of low prediction accuracy caused by human experience.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, apparatus, electronic device and storage medium for predicting power grid load. Background Technology

[0002] With the continuous growth of energy demand and increasing emphasis on environmental protection, the stability and efficiency of the power grid have become increasingly important. Traditional power systems mainly rely on fossil fuel power generation, which not only causes serious pollution but also suffers from low energy utilization efficiency and insufficient power generation during peak periods, leading to power shortages.

[0003] Therefore, the need for peak shaving and valley filling has been proposed.

[0004] Peak shaving and valley filling refer to the practice of energy storage systems storing excess electrical energy during periods of low grid load and then releasing the stored energy during periods of high load, thus achieving peak shaving and valley filling. This balances grid load, reduces waste of electrical resources, improves grid stability and efficiency, and increases energy utilization, which has significant economic implications. Summary of the Invention

[0005] This application discloses a method, apparatus, electronic device, and storage medium for predicting grid load.

[0006] In a first aspect, this application discloses a method for predicting grid load, comprising:

[0007] Get the historical load of the power grid in the historical time period before the current time, and get the current load of the power grid in the current time period in which the current time is located;

[0008] Obtain historical additional electricity-related events within a historical time period in the electricity consumption area supplied by the power grid, obtain current additional electricity-related events within the current time period in the electricity consumption area, and obtain future additional electricity-related events within a future time period in the electricity consumption area.

[0009] Obtain historical weather data for the power consumption area within a historical time period, obtain current weather data for the power consumption area within the current time period, and obtain future weather data for the power consumption area within a future time period;

[0010] Get historical calendar data for a historical time period, get current calendar data for the current time period, and get future calendar data for a future time period;

[0011] Obtain historical electricity prices within a historical time period in the electricity consumption area supplied by the power grid, obtain current electricity prices within the current time period in the electricity consumption area, and obtain future electricity prices within the future time period in the electricity consumption area.

[0012] Based on historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices, predict the future load of the power grid in the future time period.

[0013] Secondly, this application discloses an apparatus for predicting grid load, comprising:

[0014] The first acquisition module is used to acquire the historical load of the power grid in the historical time period before the current time, and to acquire the current load of the power grid in the current time period in which the current time is located;

[0015] The second acquisition module is used to acquire historical additional electricity-related events within a historical time period in the electricity consumption area supplied by the power grid, acquire current additional electricity-related events within the current time period in the electricity consumption area, and acquire future additional electricity-related events within a future time period in the electricity consumption area.

[0016] The third acquisition module is used to acquire historical weather data of the power consumption area within a historical time period, acquire current weather data of the power consumption area within the current time period, and acquire future weather data of the power consumption area within a future time period.

[0017] The fourth acquisition module is used to acquire historical calendar data for historical time periods, current calendar data for the current time period, and future calendar data for future time periods.

[0018] The fifth acquisition module is used to acquire historical electricity prices within a historical time period in the electricity consumption area supplied by the power grid, acquire current electricity prices within the current time period in the electricity consumption area, and acquire future electricity prices within the future time period in the electricity consumption area.

[0019] The forecasting module is used to predict the future load of the power grid in the future time period based on historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices.

[0020] Thirdly, this application discloses an electronic device comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to perform the method as described in any of the preceding aspects.

[0021] Fourthly, this application discloses a non-transitory computer-readable storage medium in which, when the instructions in the storage medium are executed by a processor of an electronic device, enable the electronic device to perform the methods described in any of the preceding aspects.

[0022] Fifthly, this application discloses a computer program product in which, when the instructions in the computer program product are executed by a processor of an electronic device, the electronic device is enabled to perform the method described in any of the preceding aspects.

[0023] The technical solution provided in this application may include the following beneficial effects:

[0024] In this application, the following are obtained: historical load of the power grid during historical time periods prior to the current moment; current load of the power grid during the current time period; historical additional power-related events within the power consumption area supplied by the power grid during historical time periods; current additional power-related events within the current time period within the power consumption area; future additional power-related events within the future time period within the power consumption area; historical weather data of the power consumption area during historical time periods; current weather data of the power consumption area during the current time period; future weather data of the power consumption area during the future time period; historical calendar data of the historical time periods; and the current time period. The system uses current calendar data to obtain future calendar data for future time periods; it obtains historical electricity prices for historical time periods within the electricity consumption area supplied by the power grid, current electricity prices for the current time period within the electricity consumption area, and future electricity prices for future time periods within the electricity consumption area; and it predicts the future load of the power grid for future time periods based on historical load, current load, historical additional electricity-related events, current additional electricity-related events, future additional electricity-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices.

[0025] This application considers multiple factors that will affect the future load of the power grid in the future time period when predicting the future load of the power grid. These factors include historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices. The consideration of factors is more comprehensive, which can improve the accuracy of predicting the future load of the power grid in the future time period. Secondly, the prediction of the future load of the power grid in the future time period can be carried out without human intervention, reducing labor costs and avoiding the problem of low prediction accuracy caused by human experience. Attached Figure Description

[0026] Figure 1 This is a flowchart of the steps of a method for predicting power grid load according to this application.

[0027] Figure 2 This is a structural block diagram of a device for predicting grid load according to this application.

[0028] Figure 3 This is a block diagram of an electronic device according to this application.

[0029] Figure 4 This is a block diagram of an electronic device according to this application. Detailed Implementation

[0030] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0031] The optimal charging and discharging strategy for energy storage systems depends not only on the current load conditions of the power grid, but also on the prediction of the future load of the power grid. By integrating the prediction of the future load of the power grid into the charging and discharging strategy of the energy storage system, more precise regulation of the power grid load can be achieved.

[0032] However, current forecasts of future grid loads are made manually based on experience, resulting in low accuracy.

[0033] Therefore, in order to solve the above problems, the technical solution of this application is proposed.

[0034] Specifically, refer to Figure 1 This paper illustrates a flowchart of a method for predicting power grid load according to this application. The method is applied to an electronic device, which may include a terminal or a server. The method includes:

[0035] In step S101, the historical load of the power grid in the historical time period before the current time is obtained, and the current load of the power grid in the current time period is obtained.

[0036] Grid load refers to the total electrical power consumed by all electricity users in the power grid (including industrial, commercial, and residential users). The size of the grid load directly reflects the electricity demand of the grid and is an important basis for grid operation and dispatch.

[0037] This application divides time into time periods, each of which can have the same duration. For example, the duration of a time period may include 1 hour, 2 hours, 3 hours, 6 hours, 12 hours, 24 hours, 36 hours, 48 ​​hours, 72 hours, 7 days, 14 days, or one month, etc. In two adjacent time periods, the end time of the previous time period is adjacent to the start time of the next time period.

[0038] The time period in which the current moment occurs is the current time period. The time period adjacent to and following the current time period is the future time period. The time period adjacent to and preceding the current time period is the historical time period.

[0039] Historical loads of the power grid in the historical time period prior to the current moment, as well as the current load of the power grid in the current time period, are used to help the power grid load forecasting model learn the periodic patterns (such as hourly, daily, weekly, and / or monthly patterns) and trends of the power grid load (especially the recent periodic patterns and trends of the power grid load).

[0040] The power grid system records and updates the load of the power grid in real time for each time period. In this way, the historical load of the power grid in the historical time period before the current moment and the current load of the power grid in the current time period can be directly obtained through the power grid system.

[0041] In step S102, historical additional electricity-related events within a historical time period in the electricity consumption area supplied by the power grid are obtained, current additional electricity-related events within the current time period in the electricity consumption area are obtained, and future additional electricity-related events within a future time period in the electricity consumption area are obtained.

[0042] The areas supplied by the power grid can include residential communities, streets, industrial parks, township administrative areas, or district / county administrative areas.

[0043] Additional electricity-related events include events that require a large amount of electricity in the short term or events that require a short period of non-use of electricity.

[0044] Events that require a large amount of electricity in the short term include sporting events, concerts, or other large gatherings, which can lead to a short-term increase in grid load.

[0045] Short-term events requiring the non-use of electricity include: power outages for maintenance or due to faults, which reduce the grid load for a short period (because electricity cannot be used), as well as factory shutdowns or power pole collapses, which also reduce the grid load for a short period.

[0046] Historical additional electricity-related events within a historical time period within the electricity consumption area supplied by the power grid, as well as current additional electricity-related events within the current time period within the electricity consumption area, are used to help the power grid load forecasting model learn the correlation between additional electricity-related events in the electricity consumption area and the power grid load of the electricity consumption area.

[0047] Historical additional electricity-related events within a historical time period in the electricity consumption area supplied by the power grid, as well as current additional electricity-related events within the current time period in the electricity consumption area, can be collected based on information such as news or notices.

[0048] In step S103, historical weather data of the power consumption area within a historical time period is obtained, current weather data of the power consumption area within the current time period is obtained, and future weather data of the power consumption area within a future time period is obtained.

[0049] Weather data includes: temperature, humidity, wind speed, wind direction, rainfall, snowfall, and sunshine intensity.

[0050] The meteorological system (e.g., meteorological bureaus, meteorological data APIs, or meteorological monitoring stations in the power grid system; meteorological data APIs include OpenWeatherMap, etc.) records weather data for each power consumption area in real time for each time period. It can obtain historical weather data for a power consumption area within a historical time period, current weather data for the current time period, and future weather data for the power consumption area in future time periods through the meteorological system. The API is the Application Programming Interface.

[0051] Temperature has a significant impact on grid load. For example, high temperatures (such as high temperatures in summer) may lead to an increase in the number of users using air conditioning for cooling, resulting in an increase in the cooling load of air conditioners, which in turn increases the grid load. Alternatively, low temperatures (such as low temperatures in winter) may lead to an increase in the number of users using air conditioning for heating, resulting in an increase in the heating load of air conditioners, which in turn increases the grid load.

[0052] Humidity, sunlight, wind speed, and wind direction can all affect the electricity demand of industrial and commercial users in an area, and may also affect the electricity consumption of users using new energy sources (such as solar and wind power). For example, if humidity, sunlight, wind speed, or wind direction prevents users of new energy sources (such as solar and wind power) from having sufficient power generation equipment (such as wind power generation or photovoltaic power generation), they will need to use electricity from the grid to make up for the demand, thereby increasing the grid load. Alternatively, if humidity, sunlight, wind speed, or wind direction prevents users of new energy sources (such as solar and wind power) from having sufficient power generation equipment, they will not need to use electricity from the grid, thereby reducing the grid load.

[0053] In addition, rain or snowfall may cause sudden changes in power grid load.

[0054] Historical weather data for the electricity consumption area within a historical time period, as well as current weather data for the electricity consumption area within the current time period, are used to help the power grid load forecasting model learn the correlation between the weather data of the electricity consumption area and the power grid load of the electricity consumption area.

[0055] In step S104, historical calendar data for historical time periods are obtained, current calendar data for the current time period is obtained, and future calendar data for future time periods are obtained.

[0056] Calendar data for a time period is used to indicate whether that time period is Monday, Tuesday, Wednesday, Thursday, Friday, Saturday, or Sunday.

[0057] Alternatively, calendar data for a specific time period can be used to indicate whether that period falls on a weekday or a holiday.

[0058] Statistical analysis of historical data reveals that the impact of weekdays and non-weekdays on the power grid load is usually different. During weekdays, the power grid has specific load characteristics, and on non-weekdays, the power grid also has specific load characteristics. For example, the power grid load during holidays is often lower than the power grid load during weekdays.

[0059] Alternatively, the load characteristics of the power grid may differ on each day of the week, including Monday, Tuesday, Wednesday, Thursday, Friday, Saturday, and Sunday.

[0060] Historical calendar data for historical time periods and current calendar data for the current time period are both used to help the power grid load forecasting model learn the correlation between calendar data and power grid load.

[0061] In step S105, the historical electricity price within a historical time period in the electricity consumption area supplied by the power grid is obtained, the current electricity price within the current time period in the electricity consumption area is obtained, and the future electricity price within the future time period in the electricity consumption area is obtained.

[0062] In step S106, the future load of the power grid in the future time period is predicted based on historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity price, current electricity price, and future electricity price.

[0063] In this application, a power grid load forecasting model is pre-trained. Thus, in this step, historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices can all be input into the power grid load forecasting model. This allows the power grid load forecasting model to process (e.g., through collaborative processing) the historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices to obtain the future load of the power grid in the future time period and output the future load of the power grid in the future time period. Electronic devices can obtain the future load of the power grid in the future time period output by the power grid load forecasting model.

[0064] Power grid load forecasting models can include: random forest models, XGBoost (eXtreme Gradient Boosting), LSTM (Long Short Term Memory) networks, or Transformer (a deep learning architecture based on self-attention mechanisms, originally used for natural language processing tasks such as machine translation), etc.

[0065] In this application, the following are obtained: historical load of the power grid during historical time periods prior to the current moment; current load of the power grid during the current time period; historical additional power-related events within the power consumption area supplied by the power grid during historical time periods; current additional power-related events within the current time period within the power consumption area; future additional power-related events within the future time period within the power consumption area; historical weather data of the power consumption area during historical time periods; current weather data of the power consumption area during the current time period; future weather data of the power consumption area during the future time period; historical calendar data of the historical time periods; and the current time period. The system uses current calendar data to obtain future calendar data for future time periods; it obtains historical electricity prices for historical time periods within the electricity consumption area supplied by the power grid, current electricity prices for the current time period within the electricity consumption area, and future electricity prices for future time periods within the electricity consumption area; and it predicts the future load of the power grid for future time periods based on historical load, current load, historical additional electricity-related events, current additional electricity-related events, future additional electricity-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices.

[0066] This application considers multiple factors that will affect the future load of the power grid in the future time period when predicting the future load of the power grid. These factors include historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices. The consideration of factors is more comprehensive, which can improve the accuracy of predicting the future load of the power grid in the future time period. Secondly, the prediction of the future load of the power grid in the future time period can be carried out without human intervention, reducing labor costs and avoiding the problem of low prediction accuracy caused by human experience.

[0067] In one embodiment of this application, after inputting historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices into the power grid load forecasting model, the feature extraction network in the power grid load forecasting model can extract feature vectors corresponding to historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices, respectively.

[0068] The feature extraction network can include an encoding layer and a multi-head self-attention layer. For historical loads, the encoding layer can be used to encode the historical load (e.g., one-hot encoding) to obtain the sparse vector corresponding to the historical load. Then, the multi-head self-attention layer is used to perform multi-head self-attention weighting on the sparse vector corresponding to the historical load to obtain the feature vector corresponding to the historical load.

[0069] The same applies to current load, historical additional electricity-related events, current additional electricity-related events, future additional electricity-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices, which will not be detailed here.

[0070] Encoding layers can include one-hot encoding, etc.

[0071] Multi-head self-attention layers can include, for example, Multi-head Self-attention Layers.

[0072] Then, the feature vectors corresponding to historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices are input into the aggregation network in the power grid load forecasting model. This allows the aggregation network to aggregate the feature vectors corresponding to historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices into an aggregated vector.

[0073] For example, the feature vectors corresponding to historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices can be concatenated sequentially to obtain an aggregated vector.

[0074] Alternatively, if the aforementioned feature vectors have the same dimension, the feature vectors corresponding to historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices can be averaged or max-pooled to obtain an aggregated vector.

[0075] Then, the aggregation vector is processed using the prediction network in the power grid load forecasting model to obtain the future load of the power grid in the future time period.

[0076] Alternatively, in one embodiment of this application, when predicting the future load of the power grid in a future time period based on historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices, historical interaction data between historical additional power-related events and historical load can be obtained. (By using the power grid load in the same time period and additional power-related events in the power consumption area supplied by the power grid, the power grid load prediction model can learn the correlation between the power grid load in the same time period and additional power-related events in the power consumption area supplied by the power grid during the training process, or learn the impact of additional power-related events in the power consumption area supplied by the power grid on the power grid load in the same time period. Thus, when using the power grid load prediction model to predict the future load of the power grid in a future time period, using historical interaction data between historical additional power-related events and historical load can improve the accuracy of predicting the future load of the power grid in a future time period.)

[0077] By acquiring historical interaction data between historical weather data and historical load data (using the power grid load and weather data within the power grid supply area during the same time period, the power grid load forecasting model can learn the correlation between the power grid load and weather data within the power grid supply area during the training process, or in other words, learn the impact of weather data within the power grid supply area on the power grid load during the same time period. Thus, when using the power grid load forecasting model to predict the future load of the power grid in the future time period, using historical interaction data between historical weather data and historical load can improve the accuracy of predicting the future load of the power grid in the future time period).

[0078] Acquire historical interaction data between historical calendar data and historical load. (By using grid load and calendar data within the same time period, the grid load forecasting model can learn the correlation between grid load and calendar data during the training process, or in other words, learn the impact of calendar data on grid load within the same time period. Thus, when using the grid load forecasting model to predict future grid load in future time periods, using historical interaction data between historical calendar data and historical load can improve the accuracy of predicting future grid load in future time periods.)

[0079] Acquire historical data on the interaction between historical electricity prices and historical load. (By analyzing the grid load and electricity prices within the same time period, the grid load forecasting model can learn the correlation between grid load and electricity prices during the training process. In other words, it learns the impact of electricity prices on grid load within the same time period. Thus, when using the grid load forecasting model to predict future grid load in future time periods, using historical data on the interaction between historical electricity prices and historical load can improve the accuracy of predicting future grid load in future time periods.)

[0080] Acquire historical interaction data among historical weather data, historical calendar data, and historical load data. (By using power grid load, calendar data, and weather data within the power grid's supply area during the same time period, the power grid load forecasting model can learn the correlation between these three factors during training. In other words, it learns the combined impact of weather and calendar data within the power grid's supply area on the power grid load during the same time period. Therefore, when using the power grid load forecasting model to predict future power grid load in future time periods, utilizing historical interaction data among these three factors can improve the accuracy of future load predictions.)

[0081] By acquiring historical data on the interaction between historical weather data, historical additional power-related events, and historical load (through the power grid load, additional power-related events within the power grid supply area, and weather data within the power grid supply area during the same time period, the power grid load forecasting model can learn the correlation between these three factors during training. In other words, it learns the combined impact of weather data and additional power-related events within the power grid supply area on the power grid load during the same time period. Thus, when using the power grid load forecasting model to predict the future load of the power grid in future time periods, the historical interaction data between historical weather data, historical additional power-related events, and historical load can improve the accuracy of predicting the future load of the power grid in future time periods).

[0082] By acquiring historical interaction data among historical calendar data, historical additional power-related events, and historical load (using the power grid load, additional power-related events within the power grid supply area, and calendar data within the same time period, the power grid load forecasting model can learn the correlation among these three factors during training; or, in other words, learn the combined impact of additional power-related events and calendar data within the power grid supply area on the power grid load within the same time period. Thus, when using the power grid load forecasting model to predict the future load of the power grid in future time periods, using historical interaction data among historical calendar data, historical additional power-related events, and historical load can improve the accuracy of predicting the future load of the power grid in future time periods).

[0083] In addition, it retrieves the current interaction data between the current additional electricity-related events and the current load. It also retrieves the current interaction data between the current weather data and the current load, the current calendar data and the current load, the current electricity price and the current load, and the current interaction data among the current weather data, current calendar data, and current load. The specific effects can be seen in the aforementioned example descriptions, which will not be detailed here.

[0084] In one example, the feature extraction network in the power grid load forecasting model extracts feature vectors corresponding to historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices. Then, the feature interaction network in the power grid load forecasting model can be used to obtain historical interaction data between historical additional power-related events and historical load. This includes data on historical interaction between historical weather data and historical load, historical calendar data and historical load, historical electricity prices and historical load, and historical interaction data among historical weather data, historical calendar data, and historical load. Acquire historical interaction data among historical calendar data, historical additional power-related events, and historical load.

[0085] For example, the feature interaction network calculates the product between the feature vector corresponding to the historical additional power-related event and the feature vector corresponding to the historical load, and uses it as the historical interaction relationship data between the historical additional power-related event and the historical load.

[0086] The feature interaction network calculates the product between the feature vector corresponding to historical weather data and the feature vector corresponding to historical load, and uses it as the historical interaction relationship data between historical weather data and historical load.

[0087] The feature interaction network calculates the product between the feature vector corresponding to the historical calendar data and the feature vector corresponding to the historical load, and uses it as the historical interaction relationship data between the historical calendar data and the historical load.

[0088] The feature interaction network calculates the product between the feature vector corresponding to historical electricity prices and the feature vector corresponding to historical loads, and uses it as the historical interaction relationship data between historical electricity prices and historical loads.

[0089] The feature interaction network calculates the product of the feature vectors corresponding to historical weather data, historical calendar data, and historical load, and uses this product as the historical interaction relationship data among the three.

[0090] The feature interaction network calculates the product of the feature vectors corresponding to historical weather data, historical additional power-related events, and historical load, and uses this product as the historical interaction relationship data among historical weather data, historical additional power-related events, and historical load.

[0091] The feature interaction network calculates the product of the feature vectors corresponding to historical calendar data, historical additional power-related events, and historical loads, and uses this product as the historical interaction relationship data among the three: historical calendar data, historical additional power-related events, and historical loads.

[0092] The feature interaction network in the power grid load forecasting model can be used to obtain the current interaction data between current additional power-related events and current load. This includes: obtaining the current interaction data between current weather data and current load; obtaining the current interaction data between current calendar data and current load; obtaining the current interaction data between current electricity price and current load; obtaining the current interaction data among current weather data, current calendar data, and current load; and obtaining the current interaction data among current calendar data, current additional power-related events, and current load.

[0093] For example, the feature interaction network calculates the product between the feature vector corresponding to the current additional power-related event and the feature vector corresponding to the current load, and uses it as the current interaction relationship data between the current additional power-related event and the current load.

[0094] The feature interaction network calculates the product between the feature vector corresponding to the current weather data and the feature vector corresponding to the current load, and uses it as the current interaction relationship data between the current weather data and the current load.

[0095] The feature interaction network calculates the product between the feature vector corresponding to the current calendar data and the feature vector corresponding to the current load, and uses it as the current interaction relationship data between the current calendar data and the current load.

[0096] The feature interaction network calculates the product between the feature vector corresponding to the current electricity price and the feature vector corresponding to the current load, and uses it as the current interaction data between the current electricity price and the current load.

[0097] The feature interaction network calculates the product of the feature vector corresponding to the current weather data, the feature vector corresponding to the current calendar data, and the feature vector corresponding to the current load, and uses it as the current interaction relationship data among the current weather data, current calendar data, and current load.

[0098] The feature interaction network calculates the product of the feature vector corresponding to the current weather data, the feature vector corresponding to the current additional power-related event, and the feature vector corresponding to the current load, and uses it as the current interaction relationship data among the current weather data, the current additional power-related event, and the current load.

[0099] The feature interaction network calculates the product of the feature vector corresponding to the current calendar data, the feature vector corresponding to the current additional power-related event, and the feature vector corresponding to the current load, and uses it as the current interaction relationship data among the current calendar data, the current additional power-related event, and the current load.

[0100] The product mentioned above is also an eigenvector.

[0101] All of the aforementioned feature vectors have the same dimension.

[0102] Then, based on future additional power-related events, future weather data, future calendar data, future electricity prices, various historical interaction data (all historical interaction data obtained above), and various current interaction data (all current interaction data obtained above), the future load of the power grid in the future time period can be predicted.

[0103] For example, feature vectors corresponding to future additional electricity-related events, future weather data, future calendar data, future electricity prices, historical interaction data, and current interaction data are aggregated into an aggregate vector. Then, a prediction network is used to process the aggregate vector to obtain the future load of the power grid in the future time period.

[0104] In another embodiment of this application, when predicting the future load of the power grid in a future time period based on future additional power-related events, future weather data, future calendar data, future electricity prices, various historical interaction relationship data, and various current interaction relationship data, attention weights corresponding to each historical interaction relationship data can be obtained, and the corresponding historical interaction relationship data can be weighted according to the attention weights corresponding to each historical interaction relationship data to obtain the historical weighted interaction relationship data corresponding to each historical interaction relationship data.

[0105] For example, the "historical interaction data between historical additional power-related events and historical loads" is weighted according to the attention weights corresponding to the "historical interaction data between historical additional power-related events and historical loads" to obtain the "historical weighted interaction data between historical additional power-related events and historical loads".

[0106] Based on the attention weights corresponding to the "historical interaction data between historical weather data and historical load", the "historical interaction data between historical weather data and historical load" is weighted to obtain the "historical weighted interaction data between historical weather data and historical load".

[0107] Based on the attention weights corresponding to the "historical interaction data between historical calendar data and historical workload", the "historical interaction data between historical calendar data and historical workload" is weighted to obtain the "historical weighted interaction data between historical calendar data and historical workload".

[0108] Based on the attention weight corresponding to the "historical interaction data between historical electricity prices and historical load", the "historical interaction data between historical electricity prices and historical load" is weighted to obtain the "historical weighted interaction data between historical electricity prices and historical load".

[0109] Based on the attention weights corresponding to the historical interaction data among historical weather data, historical calendar data, and historical load, the historical interaction data among historical weather data, historical calendar data, and historical load is weighted to obtain the historical weighted interaction data among historical weather data, historical calendar data, and historical load.

[0110] Based on the attention weights corresponding to the historical interaction data among historical weather data, historical additional power-related events, and historical load, the historical interaction data among historical weather data, historical additional power-related events, and historical load is weighted to obtain the historical weighted interaction data among historical weather data, historical additional power-related events, and historical load.

[0111] Based on the attention weights corresponding to the historical interaction data among historical calendar data, historical additional power-related events, and historical load, the historical interaction data among historical calendar data, historical additional power-related events, and historical load is weighted to obtain the historical weighted interaction data among historical calendar data, historical additional power-related events, and historical load.

[0112] Obtain the attention weight corresponding to each current interaction relationship data. Based on the attention weight corresponding to each current interaction relationship data, weight the corresponding current interaction relationship data to obtain the current weighted interaction relationship data for each current interaction relationship data.

[0113] For example, the "current interaction data between current additional power-related events and current load" is weighted according to the attention weight corresponding to the "current interaction data between current additional power-related events and current load" to obtain the "current weighted interaction data between current additional power-related events and current load".

[0114] The current interaction data between current weather data and current load is weighted according to the attention weight corresponding to the current interaction data between current weather data and current load to obtain the current weighted interaction data between current weather data and current load.

[0115] The current interaction data between the current calendar data and the current load is weighted according to the attention weight corresponding to the current interaction data between the current calendar data and the current load, resulting in the current weighted interaction data between the current calendar data and the current load.

[0116] The current interaction data between current electricity price and current load is weighted according to the attention weight corresponding to the current interaction data between current electricity price and current load to obtain the current weighted interaction data between current electricity price and current load.

[0117] Based on the attention weights corresponding to the current interaction data between the current weather data, current calendar data, and current load, the current interaction data between the current weather data, current calendar data, and current load is weighted to obtain the current weighted interaction data between the current weather data, current calendar data, and current load.

[0118] Based on the attention weights corresponding to the current interaction data between the current weather data, current additional power-related events, and current load, the current interaction data between the current weather data, current additional power-related events, and current load is weighted to obtain the current weighted interaction data between the current weather data, current additional power-related events, and current load.

[0119] Based on the attention weights corresponding to the current interaction data between the current calendar data, the current additional power-related events, and the current load, the current weighted interaction data between the current calendar data, the current additional power-related events, and the current load is weighted to obtain the current weighted interaction data between the current calendar data, the current additional power-related events, and the current load.

[0120] Based on future additional power-related events, future weather data, future calendar data, future electricity prices, various historical weighted interaction data (all historical weighted interaction data mentioned above), and various current weighted interaction data (all current weighted interaction data mentioned above), the future load of the power grid in the future time period is predicted.

[0121] For example, aggregated data can be obtained by combining future additional power-related events, future weather data, future calendar data, future user prices, various historical weighted interaction data, and various current weighted interaction data. The future load on the power grid within a future time period can then be predicted based on this aggregated data.

[0122] In this application, the power grid load forecasting model also includes an attention network, which can be used to obtain the attention weights corresponding to the feature vectors of each historical interaction data.

[0123] For example, for any historical interaction data point, the corresponding feature vector can be weighted according to the attention weights, resulting in a weighted feature vector. This weighted feature vector is then input into the prediction network. The same applies to every other historical interaction data point, and also to every current interaction data point.

[0124] For example, the feature vectors corresponding to future additional electricity-related events, future weather data, future calendar data, future electricity prices, various historical weighted interaction data, and various current weighted interaction data are aggregated to obtain an aggregated vector. Then, a prediction network is used to process the aggregated vector to obtain the future load of the power grid in the future time period.

[0125] Prediction networks include logistic regression functions, normalized exponential functions, fully connected layers, or activation functions, such as sigmoid, softmax, or ReLU.

[0126] The predictive network can first use an MLP (Multilayer Perceptron) to process the aggregated vector to obtain an intermediate vector, then use the activation function tanh to process the intermediate vector to obtain an activation vector, and then use Softmax, Sigmoid or ReLU to process the activation vector to obtain the future load of the power grid in the future time period.

[0127] For two eigenvectors, if the two eigenvectors have the same dimension, calculating the product between the two eigenvectors includes calculating the vector product or cross product between the two eigenvectors.

[0128] The attention mechanism allows us to obtain the contribution of each historical interaction data point and each current interaction data point to the future load of the power grid in the future time period.

[0129] For example, different attention weights can be obtained for each historical interaction data, and different attention weights can be obtained for each current interaction data. For example, the weight of historical interaction data with greater contribution is appropriately higher, so that historical interaction data with more positive influence contributes more to the future load of the power grid in the future time period, and the weight of historical interaction data with less positive influence is appropriately lower, so that historical interaction data with less positive influence contributes less to the future load of the power grid in the future time period.

[0130] Furthermore, different attention weights can be obtained for each current interaction data. For example, the weight of current interaction data with a greater contribution is appropriately higher, so that current interaction data with a greater positive impact contributes more to the future load of the power grid in the future time period, while the weight of current interaction data with a less positive impact is appropriately lower, so that current interaction data with a less positive impact contributes less to the future load of the power grid in the future time period.

[0131] Thus, this application can use the attention weights corresponding to each historical interaction data to weight each historical interaction data separately, so as to help improve the accuracy of predicting the future load of the power grid in the future time period through the attention mechanism. For example, the attention mechanism can make full use of each historical interaction data according to the degree of contribution. Similarly, the application can use the attention weights corresponding to each current interaction data to weight each current interaction data separately, so as to help improve the accuracy of predicting the future load of the power grid in the future time period through the attention mechanism. For example, the attention mechanism can make full use of each current interaction data according to the degree of contribution, thereby improving the accuracy of predicting the future load of the power grid in the future time period.

[0132] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions involved are not necessarily required by this application.

[0133] Reference Figure 2 This application illustrates an apparatus for predicting grid load, comprising:

[0134] The first acquisition module 11 is used to acquire the historical load of the power grid in the historical time period before the current time, and to acquire the current load of the power grid in the current time period in which the current time is located;

[0135] The second acquisition module 12 is used to acquire historical additional electricity-related events within a historical time period in the electricity consumption area supplied by the power grid, acquire current additional electricity-related events within the current time period in the electricity consumption area, and acquire future additional electricity-related events within a future time period in the electricity consumption area.

[0136] The third acquisition module 13 is used to acquire historical weather data of the power consumption area in a historical time period, acquire current weather data of the power consumption area in the current time period, and acquire future weather data of the power consumption area in a future time period.

[0137] The fourth acquisition module 14 is used to acquire historical calendar data for historical time periods, current calendar data for the current time period, and future calendar data for future time periods.

[0138] The fifth acquisition module 15 is used to acquire historical electricity prices within a historical time period in the electricity consumption area supplied by the power grid, acquire current electricity prices within the current time period in the electricity consumption area, and acquire future electricity prices within the future time period in the electricity consumption area.

[0139] The prediction module 16 is used to predict the future load of the power grid in the future time period based on historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity price, current electricity price, and future electricity price.

[0140] In one optional implementation, the prediction module includes:

[0141] The first acquisition unit is used to acquire historical interaction data between historical additional power-related events and historical loads.

[0142] The second acquisition unit is used to acquire historical interaction data between historical weather data and historical load.

[0143] The third acquisition unit is used to acquire historical interaction data between historical calendar data and historical workload.

[0144] The fourth acquisition unit is used to acquire historical data on the interaction between historical electricity prices and historical load.

[0145] The fifth acquisition unit is used to acquire historical interaction data among historical weather data, historical calendar data, and historical load data.

[0146] The sixth acquisition unit is used to acquire historical data on the interaction between historical weather data, historical additional power-related events, and historical load data.

[0147] The seventh acquisition unit is used to acquire historical interaction data among historical calendar data, historical additional power-related events, and historical load.

[0148] The eighth acquisition unit is used to acquire the current interaction relationship data between the current additional power-related event and the current load;

[0149] The ninth acquisition unit is used to acquire the current interaction relationship data between the current weather data and the current load;

[0150] The tenth acquisition unit is used to acquire the current interaction relationship data between the current calendar data and the current workload;

[0151] The eleventh acquisition unit is used to acquire data on the current interaction relationship between the current electricity price and the current load.

[0152] The twelfth acquisition unit is used to acquire the current interaction relationship data between current weather data, current calendar data, and current load.

[0153] The thirteenth acquisition unit is used to acquire the current interaction data among current weather data, current additional power-related events, and current load.

[0154] The fourteenth acquisition unit is used to acquire the current interaction relationship data among the current calendar data, current additional power-related events, and current load.

[0155] The forecasting unit is used to predict the future load of the power grid in the future time period based on future additional power-related events, future weather data, future calendar data, future electricity prices, various historical interaction data, and various current interaction data.

[0156] In one optional implementation, the prediction unit includes:

[0157] The first weighting subunit is used to obtain the attention weights corresponding to each historical interaction data, and to weight the corresponding historical interaction data according to the attention weights corresponding to each historical interaction data to obtain the historical weighted interaction data corresponding to each historical interaction data.

[0158] The second weighting subunit is used to obtain the attention weights corresponding to each current interaction relationship data; and to weight the corresponding current interaction relationship data according to the attention weights corresponding to each current interaction relationship data to obtain the current weighted interaction relationship data corresponding to each current interaction relationship data.

[0159] The forecasting subunit is used to predict the future load of the power grid in the future time period based on future additional power-related events, future weather data, future calendar data, future user prices, various historical weighted interaction data, and various current weighted interaction data.

[0160] In one optional implementation, the prediction subunit is specifically used to: aggregate future additional power-related events, future weather data, future calendar data, future user prices, various historical weighted interaction relationship data, and various current weighted interaction relationship data to obtain aggregated data; and predict the future load of the power grid in the future time period based on the aggregated data.

[0161] In this application, the following are obtained: historical load of the power grid during historical time periods prior to the current moment; current load of the power grid during the current time period; historical additional power-related events within the power consumption area supplied by the power grid during historical time periods; current additional power-related events within the current time period within the power consumption area; future additional power-related events within the future time period within the power consumption area; historical weather data of the power consumption area during historical time periods; current weather data of the power consumption area during the current time period; future weather data of the power consumption area during the future time period; historical calendar data of the historical time periods; and the current time period. The system uses current calendar data to obtain future calendar data for future time periods; it obtains historical electricity prices for historical time periods within the electricity consumption area supplied by the power grid, current electricity prices for the current time period within the electricity consumption area, and future electricity prices for future time periods within the electricity consumption area; and it predicts the future load of the power grid for future time periods based on historical load, current load, historical additional electricity-related events, current additional electricity-related events, future additional electricity-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices.

[0162] This application considers multiple factors that will affect the future load of the power grid in the future time period when predicting the future load of the power grid. These factors include historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices. The consideration of factors is more comprehensive, which can improve the accuracy of predicting the future load of the power grid in the future time period. Secondly, the prediction of the future load of the power grid in the future time period can be carried out without human intervention, reducing labor costs and avoiding the problem of low prediction accuracy caused by human experience.

[0163] Optionally, this application also provides an electronic device, including: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the various processes of the above method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0164] This application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the various processes of the above-described method embodiments and achieves the same technical effects. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0165] Figure 3 This is a block diagram illustrating an electronic device 800. For example, the electronic device 800 may be a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness equipment, personal digital assistant, etc.

[0166] Reference Figure 3 The electronic device 800 may include one or more of the following components: a processing component 802, a memory 804, a power supply component 806, a multimedia component 808, an audio component 810, an input / output (I / O) interface 812, a sensor component 814, and a communication component 816.

[0167] Processing component 802 typically controls the overall operation of electronic device 800, such as operations associated with display, telephone calls, data communication, camera operation, and recording operations. Processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the methods described above. Furthermore, processing component 802 may include one or more modules to facilitate interaction between processing component 802 and other components. For example, processing component 802 may include a multimedia module to facilitate interaction between multimedia component 808 and processing component 802.

[0168] Memory 804 is configured to store various types of data to support the operation of device 800. Examples of this data include instructions for any application or method operating on electronic device 800, contact data, phonebook data, messages, images, videos, etc. Memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0169] Power supply component 806 provides power to various components of electronic device 800. Power supply component 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to electronic device 800.

[0170] Multimedia component 808 includes a screen that provides an output interface between the electronic device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may not only sense the boundaries of touch or swipe actions but also monitor the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 808 includes a front-facing camera and / or a rear-facing camera. When the device 800 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0171] Audio component 810 is configured to output and / or input audio signals. For example, audio component 810 includes a microphone (MIC) configured to receive external audio signals when electronic device 800 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 804 or transmitted via communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.

[0172] I / O interface 812 provides an interface between processing component 802 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0173] Sensor assembly 814 includes one or more sensors for providing status assessments of various aspects of electronic device 800. For example, sensor assembly 814 may monitor the on / off state of device 800, the relative positioning of components such as the display and keypad of electronic device 800, changes in position of electronic device 800 or a component of electronic device 800, the presence or absence of user contact with electronic device 800, the orientation or acceleration / deceleration of electronic device 800, and temperature changes of electronic device 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 814 may also include an accelerometer, gyroscope, magnetometer, pressure sensor, or temperature sensor.

[0174] Communication component 816 is configured to facilitate wired or wireless communication between electronic device 800 and other devices. Electronic device 800 can access wireless networks based on communication standards, such as WiFi, carrier networks (such as 2G, 3G, 4G, or 5G), or combinations thereof. In one exemplary embodiment, communication component 816 receives broadcast signals or broadcast operation information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 816 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0175] In an exemplary embodiment, the electronic device 800 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the methods described above.

[0176] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 804 including instructions, which can be executed by a processor 820 of an electronic device 800 to perform the above-described method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0177] Figure 4This is a block diagram of an electronic device 1900 shown in this application. For example, the electronic device 1900 can be provided as a server.

[0178] Reference Figure 4 The electronic device 1900 includes a processing component 1922, which further includes one or more processors, and memory resources represented by memory 1932 for storing instructions, such as application programs, that can be executed by the processing component 1922. The application programs stored in memory 1932 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 1922 is configured to execute instructions to perform the methods described above.

[0179] Electronic device 1900 may also include a power supply component 1926 configured to perform power management of electronic device 1900, a wired or wireless network interface 1950 configured to connect electronic device 1900 to a network, and an input / output (I / O) interface 1958. Electronic device 1900 can operate on an operating system stored in memory 1932, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, or similar.

[0180] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0181] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0182] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

[0183] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this application can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0184] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0185] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0186] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0187] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0188] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0189] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for predicting power grid load, characterized in that, include: Get the historical load of the power grid in the historical time period before the current time, and get the current load of the power grid in the current time period in which the current time is located; Obtain historical additional electricity-related events within a historical time period in the electricity consumption area supplied by the power grid, obtain current additional electricity-related events within the current time period in the electricity consumption area, and obtain future additional electricity-related events within a future time period in the electricity consumption area. Obtain historical weather data for the power consumption area within a historical time period, obtain current weather data for the power consumption area within the current time period, and obtain future weather data for the power consumption area within a future time period; Get historical calendar data for a historical time period, get current calendar data for the current time period, and get future calendar data for a future time period; Obtain historical electricity prices within a historical time period in the electricity consumption area supplied by the power grid, obtain current electricity prices within the current time period in the electricity consumption area, and obtain future electricity prices within the future time period in the electricity consumption area. Based on historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices, the future load of the power grid in the future time period is predicted, specifically including the following steps: Obtain the attention weights corresponding to each historical interaction data, and weight the corresponding historical interaction data according to the attention weights corresponding to each historical interaction data to obtain the historical weighted interaction data corresponding to each historical interaction data. Obtain the attention weight corresponding to each current interaction relationship data; weight the corresponding current interaction relationship data according to the attention weight corresponding to each current interaction relationship data to obtain the current weighted interaction relationship data corresponding to each current interaction relationship data; Based on future additional power-related events, future weather data, future calendar data, future electricity prices, various historical weighted interaction data, and various current weighted interaction data, the future load of the power grid in the future time period is predicted.

2. The method according to claim 1, characterized in that, The method of predicting the future load of the power grid in a future time period based on historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices includes: Obtain historical interaction data between historical additional power-related events and historical loads; Obtain historical interaction data between historical weather data and historical load; Obtain historical interaction data between historical calendar data and historical workload; Obtain historical data on the interaction between historical electricity prices and historical load; Acquire historical interaction data among historical weather data, historical calendar data, and historical load data; Acquire historical data on the interaction relationships among historical weather data, historical additional power-related events, and historical load data. Acquire historical interaction data among historical calendar data, historical additional power-related events, and historical load; Obtain data on the current interaction between the current additional power-related events and the current load; Obtain the current interaction data between current weather data and current load; Retrieve the current interaction data between the current calendar data and the current workload; Obtain data on the current interaction between current electricity price and current load; Obtain the current interaction data among current weather data, current calendar data, and current load. Acquire current interaction data among current weather data, current additional power-related events, and current load; Retrieve the current interaction data among the current calendar data, current additional power-related events, and current load; Based on future additional power-related events, future weather data, future calendar data, future electricity prices, historical interaction data, and current interaction data, the future load of the power grid in the future time period is predicted, specifically including the following steps: Obtain the attention weights corresponding to each historical interaction data, and weight the corresponding historical interaction data according to the attention weights corresponding to each historical interaction data to obtain the historical weighted interaction data corresponding to each historical interaction data. Obtain the attention weight corresponding to each current interaction relationship data; weight the corresponding current interaction relationship data according to the attention weight corresponding to each current interaction relationship data to obtain the current weighted interaction relationship data corresponding to each current interaction relationship data; Based on future additional power-related events, future weather data, future calendar data, future electricity prices, various historical weighted interaction data, and various current weighted interaction data, the future load of the power grid in the future time period is predicted.

3. The method according to claim 2, characterized in that, The prediction of future grid load over a future time period based on future additional power-related events, future weather data, future calendar data, future electricity prices, various historical weighted interaction data, and various current weighted interaction data includes: Aggregated data is obtained by aggregating future additional electricity-related events, future weather data, future calendar data, future electricity prices, various historical weighted interaction data, and various current weighted interaction data. Predict future load on the power grid in the future based on aggregated data.

4. A device for predicting power grid load, characterized in that, include: The first acquisition module is used to acquire the historical load of the power grid in the historical time period before the current time, and to acquire the current load of the power grid in the current time period in which the current time is located; The second acquisition module is used to acquire historical additional electricity-related events within a historical time period in the electricity consumption area supplied by the power grid, acquire current additional electricity-related events within the current time period in the electricity consumption area, and acquire future additional electricity-related events within a future time period in the electricity consumption area. The third acquisition module is used to acquire historical weather data of the power consumption area within a historical time period, acquire current weather data of the power consumption area within the current time period, and acquire future weather data of the power consumption area within a future time period. The fourth acquisition module is used to acquire historical calendar data for historical time periods, current calendar data for the current time period, and future calendar data for future time periods. The fifth acquisition module is used to acquire historical electricity prices within a historical time period in the electricity consumption area supplied by the power grid, acquire current electricity prices within the current time period in the electricity consumption area, and acquire future electricity prices within the future time period in the electricity consumption area. The prediction module is used to predict the future load of the power grid in the future time period based on historical load, current load, historical additional power-related events, current additional power-related events, future additional power-related events, historical weather data, current weather data, future weather data, historical calendar data, current calendar data, future calendar data, historical electricity prices, current electricity prices, and future electricity prices; it obtains the attention weights corresponding to each historical interaction data, and weights the corresponding historical interaction data according to the attention weights of each historical interaction data to obtain the historical weighted interaction data corresponding to each historical interaction data; Obtain the attention weights corresponding to each current interaction relationship data; The current interaction data is weighted according to the attention weights corresponding to each current interaction data, resulting in the current weighted interaction data for each current interaction data. Based on future additional power-related events, future weather data, future calendar data, future electricity prices, historical weighted interaction data, and current weighted interaction data, the future load of the power grid in the future time period is predicted.

5. The apparatus according to claim 4, characterized in that, The prediction module includes: The first acquisition unit is used to acquire historical interaction data between historical additional power-related events and historical loads. The second acquisition unit is used to acquire historical interaction data between historical weather data and historical load. The third acquisition unit is used to acquire historical interaction data between historical calendar data and historical workload. The fourth acquisition unit is used to acquire historical data on the interaction between historical electricity prices and historical load. The fifth acquisition unit is used to acquire historical interaction data among historical weather data, historical calendar data, and historical load data. The sixth acquisition unit is used to acquire historical data on the interaction between historical weather data, historical additional power-related events, and historical load data. The seventh acquisition unit is used to acquire historical interaction data among historical calendar data, historical additional power-related events, and historical load. The eighth acquisition unit is used to acquire the current interaction relationship data between the current additional power-related event and the current load; The ninth acquisition unit is used to acquire the current interaction relationship data between the current weather data and the current load; The tenth acquisition unit is used to acquire the current interaction relationship data between the current calendar data and the current workload; The eleventh acquisition unit is used to acquire data on the current interaction relationship between the current electricity price and the current load. The twelfth acquisition unit is used to acquire the current interaction relationship data between current weather data, current calendar data, and current load. The thirteenth acquisition unit is used to acquire the current interaction data among current weather data, current additional power-related events, and current load. The fourteenth acquisition unit is used to acquire the current interaction relationship data among the current calendar data, current additional power-related events, and current load. The forecasting unit is used to predict the future load of the power grid within a future time period based on future additional power-related events, future weather data, future calendar data, future electricity prices, various historical interaction data, and various current interaction data; obtain the attention weights corresponding to each historical interaction data, and weight the corresponding historical interaction data according to the attention weights to obtain the historical weighted interaction data for each historical interaction data; obtain the attention weights corresponding to each current interaction data, and weight the corresponding current interaction data according to the attention weights to obtain the current weighted interaction data for each current interaction data; and predict the future load of the power grid within a future time period based on future additional power-related events, future weather data, future calendar data, future electricity prices, various historical weighted interaction data, and various current weighted interaction data.

6. The apparatus according to claim 5, characterized in that, The prediction unit is specifically used to: aggregate future additional power-related events, future weather data, future calendar data, future electricity prices, various historical weighted interaction relationship data, and various current weighted interaction relationship data to obtain aggregated data; and predict the future load of the power grid in the future time period based on the aggregated data.

7. An electronic device, characterized in that, include: A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the method as described in any one of claims 1 to 3.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method as described in any one of claims 1 to 3.

Images