Power demand decision-making method and related device

Abstract

The invention discloses a power demand decision-making method and a related device. The method comprises the following steps: acquiring historical power data of a demand end in a target area in a historical time period; inputting the historical power data into a target power demand prediction model to obtain target power data of a demand end in the target area in a target future time period; determining n power decision schemes according to the target power data; evaluating the n power decision schemes to obtain n evaluation scores; and according to the n evaluation scores, selecting one power decision scheme from the n power decision schemes as a target power decision scheme. The method is beneficial for improving the accuracy of power demand decision.

Description

Technical Field

[0001] The present application relates to the technical field of power demand forecasting, and in particular to a power demand decision-making method and related devices. Background Art

[0002] Electricity demand forecasting refers to the forecasting of electricity demand on the demand side in the future. Through the forecasting of electricity demand on the demand side, corresponding electricity decision-making plans can be formulated, and the power generation data on the power generation side, the power load data on the power supply side, and the market power purchase data from the power market can be dispatched to optimize the allocation of power resources, reduce the redundancy of power generation, power supply and power purchase, and thus reduce the operating cost of the power grid.

[0003] The existing power decision-making method uses statistical methods to determine the probability of multiple predicted power data on the demand side in future time periods based on the historical power data on the demand side. After taking the predicted power data with the highest probability as the target power data, the power decision plan is directly determined based on the target power data, resulting in unreasonable determination of the power decision plan and poor accuracy of power demand decisions. Summary of the invention

[0004] In order to solve the above-mentioned problems existing in the prior art, the embodiments of the present application provide a power demand decision-making method and related devices, which predict the target power data in the target future time period through the historical power data of the demand side of the target area in the historical time period, and determine n power decision plans based on the target power data. By evaluating the n power decision plans, the target decision plan is selected from the n power decision plans, thereby improving the accuracy of the power demand decision.

[0005] In a first aspect, an embodiment of the present application provides a method for determining power demand, the method comprising:

[0006] Obtain historical power data of the demand side in the target area within a historical time period;

[0007] Inputting the historical power data into a target power demand forecasting model to obtain target power data of the demand side in the target area within a target future time period;

[0008] According to the target power data, n power decision schemes are determined; the power decision schemes are used to allocate the target power data of the demand side within the target future time period, the power generation data of the power generation side within the target future time period, and the market power purchase data of the power supply side within the target future time period; n is an integer greater than 1;

[0009] Evaluate the n electric power decision-making schemes to obtain n evaluation scores; each electric power decision-making scheme corresponds to an evaluation score;

[0010] According to the n evaluation scores, one power decision scheme is selected from the n power decision schemes as a target power decision scheme.

[0011] In a second aspect, an embodiment of the present application provides a power demand decision device, the device comprising:

[0012] An acquisition unit, used for acquiring historical power data of a demand side in a target area within a historical time period;

[0013] A processing unit, configured to input the historical power data into a target power demand forecasting model to obtain target power data of the demand side in the target area within a target future time period;

[0014] According to the target power data, n power decision schemes are determined; the power decision schemes are used to allocate the target power data of the demand side within the target future time period, the power generation data of the power generation side within the target future time period, and the market power purchase data of the power supply side within the target future time period; n is an integer greater than 1;

[0015] Evaluate the n electric power decision-making schemes to obtain n evaluation scores; each electric power decision-making scheme corresponds to an evaluation score;

[0016] According to the n evaluation scores, one power decision scheme is selected from the n power decision schemes as a target power decision scheme.

[0017] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the program includes instructions for executing the steps in the first aspect of the embodiment of the present application.

[0018] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute part or all of the steps described in the first aspect of the embodiment of the present application.

[0019] In a fifth aspect, an embodiment of the present application provides a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps described in the first aspect of the embodiment of the present application. The computer program product may be a software installation package.

[0020] Implementing the embodiments of the present application has the following beneficial effects:

[0021] In an embodiment of the present application, firstly, the historical power data of the demand side in the target area within the historical time period is obtained, and the historical power data is input into the target power demand prediction model to obtain the target power data of the demand side in the target area within the target future time period. Then, according to the target power data, n power decision schemes are determined. Furthermore, the n power decision schemes are evaluated to obtain n evaluation scores. Finally, according to the n evaluation scores, one power decision scheme is selected from the n power decision schemes as the target power decision scheme. Thus, by pre-training the target power demand prediction model, the accuracy of the target power data prediction can be improved. According to the target power data, n power decision schemes are determined, and the n power decision schemes are evaluated. A reasonable target power decision scheme can be selected from the n power decision schemes to improve the accuracy of the power demand decision. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0023] Figure 1 A schematic diagram of an application scenario of a power demand decision-making method provided in an embodiment of the present application;

[0024] Figure 2 A flowchart of a power demand decision method provided in an embodiment of the present application;

[0025] Figure 3 A schematic diagram of a power demand decision-making device provided in an embodiment of the present application;

[0026] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0028] The terms "first", "second", "third" and "fourth" etc. in the specification and claims of the present application and the drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0029] Reference to "embodiments" herein means that a particular feature, result, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0030] First, see Figure 1 , Figure 1 A schematic diagram of an application scenario of a power demand decision method provided in an embodiment of the present application. Figure 1 As shown, in the application scenario of the power demand decision method, the power demand decision method is applied to a power demand decision device. The power demand decision device is communicatively connected with the power generation end, the power supply end and the demand end.

[0031] In the embodiment of the present application, the power generation end can be a power station, for example, a thermal power station, a hydroelectric power station, a nuclear power station, a wind power station, a solar power station, and the like. The power generation end is mainly used to convert other types of energy into electrical energy and transmit the converted electrical energy to the power supply end. The power generation end is equipped with a smart meter, which can collect the power generation data of the power generation end in real time, and the power generation data can be stored in a database. The power supply end refers to the power supply network that transmits electrical energy from the power generation end to the demand end, mainly including substations, transmission lines, distribution lines, and the like. The power supply end also includes a power storage device, which can store the electrical energy converted by the power generation end. When the power supply end transmits the electrical energy converted by the power generation end to the demand end, if the electrical energy converted by the power generation end cannot meet the power demand of the demand end, the power supply end can transmit the stored electrical energy to the demand end to meet the power demand of the demand end. Among them, when the electrical energy converted by the power generation end and the stored electrical energy cannot meet the power demand of the demand end, the power supply end can also purchase electricity from the power market and transmit the purchased electrical energy to the demand end to meet the power demand of the demand end. The demand side refers to the electrical equipment and users that have power demand, such as residential electrical equipment and users, enterprise electrical equipment and users, etc. The demand side is equipped with smart meters, which can collect power data from the demand side in real time, and the power data can be stored in the database.

[0032] In an embodiment of the present application, the power demand decision device can be a server with data processing, data storage and communication functions, including: an application server, a database server, a virtual private server (VPS), a rack server, and the like. Among them, the power demand decision device can obtain the power generation data of the power generation end from the smart meter at the power generation end, obtain the power data of the demand end from the smart meter at the demand end, and obtain the reserve power data stored at the power supply end, and store the power generation data of the power generation end, the power data of the demand end, and the reserve power data stored at the power supply end in the memory or database of the power demand decision device. The power demand decision device can also process the above data to determine the power decision plan, so that the technical personnel of the power system can dispatch the power generation end, the power supply end, the demand end and the power market according to the power decision plan.

[0033] It should be noted that the power demand decision device needs to predict the power data of the demand side in the future time period based on the historical power data of the demand side, so as to specify the power decision plan based on the power data of the demand side in the future time period. The existing power decision method mainly uses statistical methods to determine the probability corresponding to the power data of the demand side in the future time period, and uses the power data with the highest probability as the power data of the demand side in the future time period. The accuracy of the determined power data is poor. In addition, directly determining the power decision plan based on the power data in the future time period is likely to lead to the unreasonable determination of the power decision plan, and the accuracy of the power demand decision is poor.

[0034] To this end, applied to the above scenario, in the power demand decision method provided in the embodiment of the present application, the power demand decision device obtains historical power data of the demand side in the target area within the historical time period;

[0035] The power demand decision device inputs the historical power data into the target power demand prediction model to obtain the target power data of the demand side in the target area in the target future time period;

[0036] The power demand decision device determines n power decision plans according to the target power data;

[0037] The power demand decision device evaluates n power decision schemes and obtains n evaluation scores;

[0038] The power demand decision device selects one power decision scheme from the n power decision schemes as a target power decision scheme according to the n evaluation scores.

[0039] It can be seen that, applied to the above scenario, the power demand decision device can obtain the historical power data of the demand side in the historical time period, and input the historical power data into the target power demand prediction model to obtain the target power data of the demand side in the target future time period, thereby improving the accuracy of power demand prediction. According to the target power data, n power decision schemes can be determined, and by evaluating the n power decision schemes, a target power decision scheme can be selected from the n power decision schemes. Since the target power decision scheme is determined by evaluating the n power decision schemes, compared with directly determining the power decision scheme, the target power decision scheme is more reasonable, thereby improving the accuracy of power demand decision.

[0040] See also Figure 2 , Figure 2 The flowchart of a power demand decision method provided in the embodiment of the present application is shown in FIG. The method is applied to the power demand decision device described above. Figure 2 As shown, the method includes but is not limited to the following steps:

[0041] 201: Obtain historical power data of the demand side in the target area within a historical time period.

[0042] In an embodiment of the present application, the target area is the area to which the power supply end transmits electric energy from the power generation end, and the demand end is located in the target area. The historical time period can be determined according to the category of power forecasting. For example, when the category of power forecasting is short-term forecasting, the historical time period can be the time period from the day before the current moment to the current moment. When the category of power forecasting is medium-term forecasting, the historical time period can be the time period from the month before the current moment to the current moment, or the time period from the quarter before the current moment to the current moment. When the category of power forecasting is long-term forecasting, the historical time period can be the time period from the year before the current moment to the current moment. This application does not specifically limit the category of power forecasting.

[0043] In an embodiment of the present application, historical power data may include: power load data, voltage, current, temperature, humidity, and other data. The power demand decision device may obtain historical power data of the demand side in a historical time period from smart meters at the power generation end and the demand side, databases of the power system, meteorological stations, etc. The acquisition method may include: Application Programming Interface (API) interface, database synchronization, file transfer, etc. Optionally, historical power data may also be obtained through standard communication protocols and interfaces, such as network communication Modbus protocol, IEC 61850 interface, OPC UA interface, etc. The historical power data may be error checked through the above communication protocols and interfaces to remove erroneous data in the historical power data.

[0044] 202: Inputting historical power data into a target power demand forecasting model to obtain target power data of the demand side in the target area in a target future time period.

[0045] In an embodiment of the present application, the target power demand forecasting model is obtained by pre-training a preset power demand forecasting model. The target future time period can be determined according to the category of the power demand forecast. For example, when the category of the power demand forecast is a short-term forecast, the target future time period can be the time period between the current moment and one day after the current moment. When the category of the power demand forecast is a medium-term forecast, the target future time period can be the time period between the current moment and one month after the current moment, or the time period between the current moment and one quarter after the current moment. When the category of the power demand forecast is a long-term forecast, the target future time period can be the time period between the current moment and one year after the current moment. This application does not make specific restrictions on this.

[0046] It should be noted that before the historical power data is input into the target power demand forecasting model, the historical power data needs to be preprocessed. Among them, data preprocessing may include: data cleaning, filtering, and format conversion. Specifically, the power demand decision device obtains the first processed data by removing duplicate data from the historical power data. Then, the power demand decision device determines the average value of the historical power data, and fills the missing values ​​in the historical power data with the average value, thereby completing data cleaning and obtaining the second processed data. Next, the power demand decision device filters the second processed data to reduce the noise and interference in the second processed data to obtain the third processed data. Further, the power demand decision device performs format conversion on the third processed data to convert the format of the third processed data into a format that can be input into the target power demand forecasting model to obtain the target historical power data. Optionally, data preprocessing may also include: range checking and logic checking, which are not limited in this application.

[0047] Then, the power demand decision device extracts features from the target historical power data, extracts features that are helpful for power demand forecasting, and obtains multiple data features corresponding to the historical power data. The multiple data features are input into the target power demand forecasting model. Exemplarily, the method may also include:

[0048] Obtaining a preset power demand forecasting model;

[0049] Acquire a first training power data set and a plurality of first actual power data;

[0050] Inputting the first training power data set into a preset power demand prediction model to obtain a plurality of first expected power data;

[0051] Determine a mean square error between a plurality of first expected power data and a plurality of first actual power data, and use the mean square error as a model loss;

[0052] If the model loss is less than or equal to the preset loss, the preset power demand forecasting model is determined as the target power demand forecasting model;

[0053] If the model loss is greater than the preset loss, the model parameters of the preset power demand forecasting model are adjusted to obtain a candidate power demand forecasting model; a second training power data set and multiple second actual power data are obtained; the candidate power demand forecasting model is trained according to the second training power data set and multiple second actual power data to obtain a target power demand forecasting model.

[0054] In the embodiment of the present application, the first training power data set includes historical power data of the demand side in multiple time periods, and the multiple first actual power data correspond one-to-one with the historical power data in multiple time periods. The multiple first expected power data correspond one-to-one with the multiple first actual power data.

[0055] Among them, the preset power demand forecasting model can be a neural network model based on a long short-term memory network (Long short-term memory, LSTM) structure. Among them, the preset power demand forecasting model includes: an input layer, multiple LSTM layers, multiple fully connected layers, an output layer and an activation function. Among them, the number of nodes in the input layer is equal to the number of data features corresponding to the historical power data, and is used to input multiple data features corresponding to the historical power data. Multiple LSTM layers are used to perform long-term and short-term memory training on multiple data features and transfer the data features. Each LSTM layer is connected to a fully connected layer, and the data features are passed to the next LSTM layer through the activation function (Rectified Linear Unit, ReLU) of the fully connected layer, and finally the activation function of the output layer outputs the expected power data. The activation function of the output layer can be a linear activation function, and the number of nodes in the output layer is equal to the number of data features corresponding to the expected power data. Optionally, the preset power demand forecasting model can also be a convolutional neural network model (Convolutional Neural Network, CNN). The embodiment of the present application only takes the preset power demand forecasting model as a neural network model based on the LSTM structure as an example for explanation, and does not specifically limit the preset power demand forecasting model.

[0056] Specifically, the power demand decision device obtains a preset power demand prediction model, a first training power data set and a plurality of first actual power data. Then, the historical power data in a plurality of time periods in the first training power data set are respectively input into the preset power demand prediction model, and the data features corresponding to the historical power data in each of the plurality of time periods are passed through the input layer, the plurality of LSTM layers and the output layer, and the data features corresponding to the first expected power data corresponding to the historical power data are output, and the data features corresponding to the plurality of first expected power data are obtained, thereby obtaining the plurality of first expected power data according to the data features corresponding to the plurality of first expected power data.

[0057] It can be understood that the time length between the future time period corresponding to the first expected power data and the time period of the historical power data corresponding to the first expected power data is equal to the time length between the target future time period and the historical time period.

[0058] Further, the power demand decision device obtains all first expected power data values ​​corresponding to each first expected power data in a plurality of first expected power data to obtain a plurality of first expected power data values, and obtains all first actual power data values ​​corresponding to each first actual power data in a plurality of first actual power data to obtain a plurality of first actual power data values. Among them, the plurality of first expected power data values ​​correspond to the plurality of first actual power data values ​​one by one. The square value of the difference between each first expected power data value in the plurality of first expected power data values ​​and the first actual power data value corresponding to the first expected power data value is determined to obtain a plurality of first square values. Then, the power demand decision device determines the ratio of the sum of the plurality of first square values ​​to the sum of the plurality of first actual power data values ​​to obtain the mean square error of the plurality of first expected power data and the plurality of first actual power data, and uses the mean square error as the model loss of the preset power demand prediction model.

[0059] Optionally, the model loss of the preset power demand prediction model can also be the total variance, explained variance, residual variance, etc. of multiple first expected power data values ​​and multiple first actual power data values. This application only takes the model loss as the mean square error as an example for illustration.

[0060] If the model loss of the preset power demand forecasting model is less than or equal to the preset loss, it means that the prediction accuracy of the preset power demand forecasting model meets the prediction requirements, and the power demand decision device determines the preset power demand forecasting model as the target power demand forecasting model.

[0061] If the model loss of the preset power demand forecasting model is greater than the preset loss, it means that the prediction accuracy of the preset power demand forecasting model does not meet the prediction requirements, and the power demand decision device will adjust the model parameters of the preset power demand forecasting model to obtain a candidate power demand forecasting model. For example, the model parameters can be adjusted by a gradient descent method, including: Adaptive Moment Estimation (Adam), adaptive learning rate optimization (Root Mean Square Propagation, RMSprop) and other methods. Then, a second training power data set and a plurality of second actual power data are obtained, and the second training power data set is input into the preset power demand forecasting model to obtain a plurality of second expected power data. According to the plurality of second expected power data and the plurality of second actual power data, the model loss of the candidate power demand forecasting model is determined until the model loss of the candidate power demand forecasting model is less than or equal to the preset loss, and the candidate power demand forecasting model with a model loss less than or equal to the preset loss is used as the target power demand forecasting model.

[0062] Optionally, the target power demand prediction model can also be tested by testing the power data set to determine the accuracy, recall rate, etc. of the target power demand prediction model, and optimize the target power demand prediction model based on the accuracy or recall rate of the target power demand prediction model. This application is not limited to this.

[0063] Thus, by training the preset power demand forecasting model with the first training power data set and a plurality of first actual power data, the model loss of the preset power demand forecasting model can be determined, and when the model loss is less than or equal to the preset loss, the preset power demand forecasting model is determined to be the target power demand forecasting model. When the model loss is greater than the preset loss, the model parameters are adjusted to obtain a candidate power demand forecasting model, and the candidate power demand forecasting model is trained with the second training power data set and a plurality of second actual power data to obtain a target power demand forecasting model, thereby improving the accuracy of the target power demand forecasting model in predicting the target power data based on the historical power data, and further improving the accuracy of the power demand decision.

[0064] Based on this, the power demand decision device inputs multiple data features corresponding to the historical power data of the demand side in the target area in the historical time period into the target power demand prediction model, and can obtain multiple data features corresponding to the target power data of the demand side in the target future time period, thereby obtaining the target power data. Among them, the processing method of multiple data features in the target power demand prediction model is similar to the processing method of multiple data features in the preset power demand prediction model, which will not be repeated here.

[0065] Optionally, the power demand decision device may also determine the confidence level or probability corresponding to the target power data to indicate the prediction accuracy of the target power data through the confidence level or probability.

[0066] 203: Determine n power decision plans according to the target power data.

[0067] In the embodiment of the present application, the power decision scheme is used to allocate the target power data of the demand side in the target future time period, the power generation data of the power generation side in the target future time period, and the market power purchase data of the power supply side in the target future time period. n is an integer greater than 1. The power demand decision device can formulate multiple scenarios according to the target power data to determine n power decision schemes.

[0068] Exemplarily, determining n power decision schemes according to target power data may include:

[0069] Obtain i historical demand scenarios and j emergency plan scenarios corresponding to the target area;

[0070] Determine the time periods corresponding to i historical demand scenarios in the target future time period, and obtain i time periods;

[0071] According to i historical demand scenarios, determine the power data corresponding to each time period in i time periods to obtain i power data;

[0072] Determine i power generation data of the power generation end within i time period according to i power data;

[0073] According to i power data and i power generation data, determine the market power purchase data of the power supply end in i time period;

[0074] Determine a first power decision-making plan according to i power data within i time periods, i power generation data, and market power purchase data within i time periods;

[0075] Obtain n-1 combinations of j emergency plan scenarios, and obtain n-1 emergency plan scenario combinations;

[0076] Determine the time period corresponding to each of the n-1 plan scenario combinations in the target future time period, and obtain n-1 time periods;

[0077] Determine the power data corresponding to each time period in n-1 time periods according to n-1 plan scenario combinations, and obtain n-1 power data;

[0078] Determine the power generation data corresponding to each power data in the n-1 power data to obtain n-1 power generation data;

[0079] Determine the market power purchase data within n-1 time periods based on n-1 power data and n-1 power generation data;

[0080] Determine an electricity decision plan based on each of the n-1 electricity data, the power generation data corresponding to the electricity data, the market power purchase data in the n-1 time periods, and the i electricity data in the i time periods, and obtain n-1 second electricity decision plans;

[0081] The first power decision scheme and n-1 second power decision schemes are taken as n power decision schemes.

[0082] In the embodiment of the present application, i and j are both integers greater than 1. The number of permutations and combinations of j contingency scenarios is greater than or equal to n-1. Historical demand scenarios may include: electricity consumption for industrial production, rising temperatures, daily lighting electricity consumption, etc. Contingency scenarios may include: changes in electricity prices, electricity consumption for large-scale events, etc. It is understandable that under different historical demand scenarios and contingency scenarios, the power demand on the demand side may be different.

[0083] Specifically, the power demand decision device obtains i historical demand scenarios and j contingency scenarios corresponding to the target area. Then, the time periods corresponding to the i historical demand scenarios in the target future time period are determined to obtain i time periods. It should be noted that the power demand decision device can use the historical demand scenarios corresponding to the target area to replace the demand scenarios in the target future time period. Therefore, the power demand decision device needs to determine the time periods corresponding to the i historical demand scenarios in the target future time period to obtain i time periods.

[0084] Then, the power demand decision device determines the power consumption of the power-consuming equipment in each of the i historical demand scenarios, and determines the power data corresponding to each of the i time periods according to the power consumption of the power-consuming equipment in each of the i historical demand scenarios, and obtains i power data. According to the i power data, i power generation data of the power generation end in the i time period can be determined, and the i power data correspond to the i power generation data one by one, wherein the corresponding relationship between the power data and the power generation data can be preset according to the actual test results.

[0085] Furthermore, the power demand decision device determines the market power purchase data of the power supply end in i time periods based on the i power data and the i power generation data. For example, the power demand decision device can determine the corresponding electric energy surplus value of each time period in the i time periods of the power supply end based on the reserve power data, i power data and i power generation data of the power supply end at the current moment. If the electric energy surplus value is a positive value, it means that there is surplus electric energy in the time period corresponding to the electric energy surplus value. If the electric energy surplus value is a negative value, it means that the electric energy supply is insufficient in the time period corresponding to the electric energy surplus value, and it is necessary to purchase electricity from the power market, so as to determine the market power purchase data in the time period corresponding to the electric energy surplus value based on the electric energy surplus value. In this way, the market power purchase data of the power supply end in i time periods can be determined.

[0086] Finally, according to the i power data, i power generation data and market power purchase data in the i time periods, the first power decision-making plan can be determined.

[0087] Next, the power demand decision device obtains n-1 combinations of j plan scenarios to obtain n-1 plan scenario combinations. It should be noted that this application does not limit the number of plan scenarios in the plan scenario combination, and the number of permutations and combinations of j plan scenarios is greater than or equal to n-1. The power demand decision device determines the time period corresponding to each plan scenario combination in the n-1 plan scenario combination in the target future time period to obtain n-1 time periods. Each of the n-1 time periods includes at least one sub-time period.

[0088] Furthermore, the power demand decision device determines the power consumption of all power-consuming devices in each of the n-1 plan scenario combinations, and determines the power data corresponding to each of the n-1 time periods according to the power consumption of all power-consuming devices in each of the n-1 plan scenario combinations, and obtains n-1 power data. Then, the power generation data corresponding to each of the n-1 power data is determined, and the n-1 power generation data corresponding to the n-1 plan scenario combination is obtained. Based on the n-1 power data and the n-1 power generation data, the market power purchase data of the power supply end in the n-1 time period can be determined.

[0089] Based on this, a power decision scheme can be determined according to each power data in the n-1 power data, the power generation data corresponding to the power data, the market power purchase data in n-1 time periods, and the i power data in i time periods, thereby obtaining n-1 second power decision schemes. The power demand decision device uses the first power decision scheme and the n-1 second power decision schemes as the above n power decision schemes.

[0090] It can be seen that the power demand decision device can associate the target power data with i historical demand scenarios, determine i power data in i time periods of the target future time period, and determine the power generation data and market power purchase data in the i time period based on the i power data, thereby determining the first power decision plan. By combining j plan scenarios, n-1 plan scenario combinations can be determined, and n-1 power data in the n-1 time period corresponding to the n-1 plan scenario combinations can be determined. Based on the n-1 power data, the power generation data and market power purchase data in the n-1 time period can be determined, thereby determining n-1 second power decision plans, and obtaining n power decision plans. Since each power decision plan considers multiple scenarios, a power decision plan selected from the n power decision plans is more reasonable, which improves the accuracy of power demand decision.

[0091] 204: Evaluate n power decision-making schemes and obtain n evaluation scores.

[0092] In the embodiment of the present application, each power decision scheme corresponds to an evaluation score. The power demand decision device evaluates n power decision schemes to determine the evaluation score of each power decision scheme, and then selects a reasonable power decision scheme from the n power decision schemes according to the evaluation score of each power decision scheme.

[0093] Exemplarily, evaluating n power decision-making schemes to obtain n evaluation scores may include:

[0094] Obtain the third power decision-making plan;

[0095] Divide the target future time period into k time periods;

[0096] Determine the power data of the demand side in each of k time periods in the third power decision-making scheme to obtain k power data;

[0097] According to the k power data, determining the power generation data of the power generation end in each of the k time periods in the third power decision-making scheme, and obtaining k power generation data;

[0098] Get the current reserve power data;

[0099] According to the reserve power data, k power generation data and k power data, m market power purchase data are determined;

[0100] An evaluation score corresponding to a third electricity decision-making scheme is determined based on the k power generation data, the k power data and the m market power purchase data.

[0101] In the embodiment of the present application, the third power decision scheme is any one of n power decision schemes; k is an integer greater than 1. m is a positive integer less than or equal to k.

[0102] Specifically, the power demand decision device obtains a third power decision scheme from n power decision schemes. Then, the target future time period is divided into k time periods. The time length of each of the k time periods can be preset according to the determination accuracy of the evaluation score. The power demand decision device determines the power data of the demand side in each of the k time periods in the third power decision scheme, and based on the k power data, the power generation data of the power generation side in each of the k time periods can be determined to obtain k power generation data.

[0103] Furthermore, the power demand decision device obtains the reserve power data at the current moment, and the reserve power data may be the electric energy reserved by the power supply end. According to the reserve power data, k power generation data and k power data, the remaining electric energy within k time periods may be determined. If the remaining electric energy is a positive value, it indicates that there is a surplus of electric energy within the time period corresponding to the remaining electric energy. The power supply end may store the remaining electric energy and update the reserve power data. If the remaining electric energy is a negative value, it indicates that the reserve power data and the power generation data within the time period corresponding to the remaining electric energy cannot meet the power data of the demand end, and it is necessary to purchase electricity from the power market. The power demand decision device may obtain the market power purchase data within the time period. Based on this, m market power purchase data may be determined.

[0104] Finally, the evaluation score corresponding to the third power decision-making scheme can be determined based on the k power generation data, k power data and m market power purchase data. Thus, the n evaluation scores corresponding to the n power decision-making schemes can be determined according to the above method.

[0105] It can be seen that the power demand decision device can divide the target future time period into k time periods, and determine the k power data of the demand side in the third power decision scheme in the n power decision schemes in the k time periods, and determine the power generation data of the power generation side in the k time periods according to the k power data, and obtain k power generation data. Then, according to the reserve power data, k power generation data and k power data at the current moment, m market power purchase data can be determined. Finally, according to the k power generation data, k power data and m market power purchase data, the evaluation score corresponding to the third power decision scheme is determined to determine n evaluation scores. As a result, each power decision scheme is only evaluated from the perspectives of the power generation end, the power supply end, the demand end and the power market, and a reasonable power decision scheme can be determined according to the determined n evaluation scores, thereby improving the accuracy of power demand decision.

[0106] Exemplarily, determining the evaluation score corresponding to the third power decision-making scheme according to k power generation data, k power data, and m market power purchase data may include:

[0107] Obtaining the power generation data threshold of the power generation end, and obtaining the power data threshold of the power demand end;

[0108] Determine a plurality of power generation data values ​​corresponding to each of the k power generation data;

[0109] Determine the scores corresponding to the power generation data whose values ​​among the k power generation data are all less than or equal to the power generation data threshold as the first power generation score, and obtain a first power generation scores;

[0110] Determine that there is power generation data greater than a power generation data threshold value among multiple power generation data values ​​in the k power generation data, and obtain b power generation data;

[0111] According to a plurality of power generation data values ​​and a power generation data threshold corresponding to each power generation data in the b power generation data, a first attenuation factor corresponding to the power generation data is determined to obtain b first attenuation factors;

[0112] Determining b second power generation scores according to the b first attenuation factors;

[0113] Determine multiple power data values ​​corresponding to each power data in the k power data;

[0114] Determine the scores corresponding to the power data whose power data values ​​are all less than or equal to the power data threshold value among the k power data as the first power usage score, and obtain c first power usage scores;

[0115] Determine that there is power data greater than a power data threshold value among a plurality of power data values ​​in the k power data, and obtain d power data;

[0116] According to a plurality of power data values ​​corresponding to each power data in the d power data, a second attenuation factor corresponding to the power data is determined to obtain d second attenuation factors;

[0117] Determining d second electricity usage scores according to the d second attenuation factors;

[0118] According to the m market electricity purchasing data, m electricity purchasing cost scores are determined;

[0119] Based on a first power generation score, b second power generation score, c first electricity consumption score, d second electricity consumption score and m electricity purchase cost scores, an evaluation score corresponding to the third electricity decision-making scheme is determined.

[0120] In the embodiment of the present application, a is an integer greater than or equal to 0 and less than k. b is an integer greater than or equal to 0 and less than k, and a+b=k. Each power generation data corresponds to a first attenuation factor. c is an integer greater than or equal to 0 and less than k. d is an integer greater than or equal to 0 and less than k, and c+d=k. Each power data corresponds to a second attenuation factor.

[0121] Specifically, the power demand decision device first obtains the power generation data threshold of the power generation end, and obtains the power data threshold of the demand end. The power generation data threshold is the maximum power generation corresponding to the power generation end, and the power data threshold is the maximum power data value that the power supply end can provide to the demand end. It should be noted that the power generation data of each power decision scheme in the n power decision schemes may have a power generation data value greater than the power generation data threshold, causing the power generation end to work overloaded. Similarly, there may be a power data value greater than the power data threshold in the power data, causing the power supply end to supply electric energy to the demand end in overload. Therefore, the power demand decision device can evaluate the third power decision scheme based on the power generation data value corresponding to the power generation data and the power data value corresponding to the power data.

[0122] In a specific implementation, the power demand decision device determines multiple power generation data values ​​corresponding to each of the k power generation data. Determine that the score corresponding to the power generation data whose multiple power generation data values ​​are all less than or equal to the power generation data threshold value among the k power generation data is the first power generation score, and obtain a first power generation score. Among them, the first power generation score can be preset. Then, determine that there is power generation data greater than the power generation data threshold value among the multiple power generation data values ​​among the k power generation data, and obtain b power generation data. Determine the difference between the multiple power generation data values ​​corresponding to each of the b power generation data and the power generation data threshold value, and obtain multiple first difference values ​​corresponding to each power generation data. According to the sum of the multiple first difference values ​​corresponding to each of the b power generation data, determine the first attenuation factor corresponding to the power generation data, and obtain b first attenuation factors. The first attenuation factor represents the rate of decrease of the second power generation score as the power generation data value increases, and the corresponding relationship between the sum of the multiple first difference values ​​and the first attenuation factor can be preset.

[0123] According to the b first attenuation factors, b second power generation scores can be determined. Optionally, the second power generation score is equal to (s3-α*e), where s3 represents the first power generation score, α represents the first attenuation factor, and e represents the sum of multiple first differences corresponding to the first attenuation factor.

[0124] Further, the power demand decision device determines multiple power data values ​​corresponding to each power data in the k power data. Determine the score corresponding to the power data whose multiple power data values ​​in the k power data are all less than or equal to the power data threshold as the first power consumption score, and obtain c first power consumption scores. Among them, the first power consumption score can be set in advance. Then, determine that there is power data greater than the power data threshold among the multiple power data values ​​in the k power data, and obtain d power data. Determine the difference between the multiple power data values ​​corresponding to each power data in the d power data and the power data threshold, and obtain multiple second difference values ​​corresponding to each power data. According to the sum of the multiple second difference values ​​corresponding to each power data in the d power data, determine the second attenuation factor corresponding to the power data, and obtain d second attenuation factors.

[0125] According to the d second attenuation factors, d second power usage scores may be determined. Optionally, the second power usage score is equal to (s4-β*u), where s4 represents the first power usage score, β represents the second attenuation factor, and u represents the sum of multiple second differences corresponding to the second attenuation factor.

[0126] Furthermore, the power demand decision-making device can determine m power purchase cost scores based on m market power purchase data, wherein each market power purchase data corresponds to a power purchase cost score, and the correspondence between the market power purchase data and the power purchase cost score can be pre-set, and the larger the power purchase amount corresponding to the market power purchase data, the lower the power purchase cost score, and the smaller the power purchase amount corresponding to the market power purchase data, the higher the power purchase cost score.

[0127] Finally, the power demand decision device can determine the evaluation score corresponding to the third power decision scheme based on a first power generation scores, b second power generation scores, c first power consumption scores, d second power consumption scores and m power purchase cost scores, thereby determining n evaluation scores corresponding to n power decision schemes according to the above scheme.

[0128] It can be seen that the power demand decision device can determine a first power generation score and b second power generation scores according to the multiple power generation data values ​​corresponding to each power generation data in the k power generation data and the power data threshold of the power generation end. According to the multiple power data values ​​corresponding to each power data in the k power data and the power data threshold of the demand end, c first power consumption scores and d second power consumption scores can be determined. According to m market power purchase data, m power purchase cost scores can be determined. Thus, based on a first power generation score, b second power generation score, c first power consumption score, d second power consumption score and m power purchase cost score, the evaluation score corresponding to the third power decision scheme is determined. Therefore, the evaluation score corresponding to the power decision scheme can be determined by evaluating the power generation data at the power generation end, the power data at the demand end and the market power purchase data, so as to select a reasonable power decision scheme and improve the accuracy of power demand decision.

[0129] Exemplarily, determining the evaluation score corresponding to the third power decision scheme based on a first power generation score, b second power generation score, c first power consumption score, d second power consumption score and m power purchase cost score may include:

[0130] Determine a total power generation score according to a first power generation score and b second power generation scores;

[0131] Determine a total electricity usage score based on c first electricity usage scores and d second electricity usage scores;

[0132] According to the m electricity purchase cost scores, a total electricity purchase cost score is determined;

[0133] The evaluation score corresponding to the third electricity decision-making plan is determined based on the total power generation score, the total power consumption score and the total power purchase cost score.

[0134] Among them, the total score of electricity purchase cost is equal to (a / k*s1+c / k*s2+s3-m / k*s3), s1 represents the total score of power generation, s2 represents the total score of electricity consumption, and s3 represents the total score of electricity purchase cost.

[0135] Specifically, the power demand decision device first determines the sum of a first power generation scores and b second power generation scores to obtain a total power generation score. Then, the power demand decision device determines the sum of c first power consumption scores and d second power consumption scores to obtain a total power consumption score. Next, the power demand decision device determines the sum of m power purchase cost scores to obtain a total power purchase cost score.

[0136] Furthermore, the power demand decision device determines a / k as the first weight corresponding to the total power generation score, determines c / k as the second weight corresponding to the total power consumption score, and determines (1-m / k) as the third weight corresponding to the total power purchase cost score. The total power generation score, the first weight, the total power consumption score, the second weight, the total power purchase cost score and the third weight are weighted and summed to obtain an evaluation score corresponding to the third power decision scheme. According to the above method, the evaluation score of each of the n power decision schemes is determined in turn to obtain n evaluation scores.

[0137] Therefore, the power demand decision device can determine the evaluation score of each of the n power decision plans by taking the weighted sum of the total power generation score, the total power consumption score and the total power purchase cost score, and thus select a reasonable power decision plan from the n power decision plans by evaluating the power generation end, the demand end and the market power purchase data of each power decision plan, thereby improving the accuracy of the power selection decision.

[0138] 205: According to the n evaluation scores, select one power decision scheme from the n power decision schemes as a target power decision scheme.

[0139] In an embodiment of the present application, the power demand decision device can select one power decision plan from n power decision plans as the target power decision plan based on n evaluation scores, the risk levels corresponding to the n power decision plans, and the total power purchase cost score.

[0140] Exemplarily, according to n evaluation scores, selecting one power decision scheme from n power decision schemes as a target power decision scheme may include:

[0141] Determine at least one evaluation score among the n evaluation scores whose evaluation score is greater than or equal to the preset evaluation score, obtain the power decision plan corresponding to the at least one evaluation score, and obtain at least one fourth power decision plan;

[0142] Determine the power generation risk level of each fourth power decision scheme according to the total power generation score corresponding to each fourth power decision scheme in at least one fourth power decision scheme, and obtain at least one power generation risk level;

[0143] Determine the power consumption risk level of each fourth power decision scheme according to the total power consumption score corresponding to each fourth power decision scheme in at least one fourth power decision scheme, and obtain at least one power consumption risk level;

[0144] Determining at least one target risk level according to at least one power generation risk level and at least one power consumption risk level;

[0145] Determine at least one fourth power decision plan whose target risk level is less than or equal to a preset risk level as a fifth power decision plan, and obtain at least one fifth power decision plan;

[0146] A target electricity decision scheme is determined according to a total score of electricity purchase costs corresponding to each fifth electricity decision scheme in at least one fifth electricity decision scheme.

[0147] Specifically, the power demand decision device can determine at least one evaluation score whose evaluation score is greater than or equal to the preset evaluation score from the n evaluation scores, and obtain the power decision plan corresponding to the at least one evaluation score to obtain at least one fourth power decision plan. Among them, the preset evaluation score can be preset according to the accuracy of the determination of the power decision plan. It should be noted that the higher the evaluation score, the fewer the power generation data values ​​exceeding the power generation data threshold in the power decision plan corresponding to the evaluation score, and / or, the fewer the power data values ​​exceeding the power data threshold, and / or, the less the market power purchase data, the higher the rationality of selecting the power decision plan.

[0148] Then, according to the total power generation score corresponding to each fourth power decision scheme in at least one fourth power decision scheme, the power generation risk level of each fourth power decision scheme is determined. Specifically, the power generation risk level of each fourth power decision scheme can be determined according to the total power generation score corresponding to the b second power generation scores corresponding to each fourth power decision scheme, and at least one power generation risk level is obtained. The mapping relationship between the total power generation score corresponding to the b second power generation scores and the power generation risk level can be preset according to the actual test results.

[0149] Similarly, the power demand decision device can determine the power risk level of each fourth power decision scheme according to the total power score corresponding to each fourth power decision scheme in at least one fourth power decision scheme. Specifically, the power risk level of each fourth power decision scheme can be determined according to the total power score corresponding to the d second power scores corresponding to each fourth power decision scheme, and at least one power generation risk level can be obtained. The mapping relationship between the total power score corresponding to the d second power scores and the power risk level can be preset according to the actual test results.

[0150] Among them, at least one power generation risk level corresponds to at least one power consumption risk level. The power demand decision device can determine at least one target risk level based on at least one power generation risk level and at least one power consumption risk level, and one power generation risk level and one power consumption risk level determine one target risk level. The mapping relationship between the power generation risk level, the power consumption risk level and the target risk level can be pre-set according to the actual test results. The power demand decision device determines that the power decision plan with a target risk level less than or equal to the preset risk level in at least one fourth power decision plan is the fifth power decision plan, and obtains at least one fifth power decision plan. It should be noted that the target risk level is less than or equal to the preset risk level, indicating that the risk of power equipment failure at the power generation end and the demand end of the power decision plan is small, and it is more reasonable to select the power decision plan.

[0151] Then, according to the total score of the electricity purchase cost corresponding to each fifth electricity decision plan in at least one fifth electricity decision plan, the market electricity purchase data corresponding to each fifth electricity decision plan can be determined, and the electricity demand decision device can use the fifth electricity decision plan with the lowest electricity purchase amount corresponding to the market electricity purchase data as the target electricity decision plan to reduce the operating cost of the power grid system.

[0152] It can be seen that by obtaining the power decision plan corresponding to at least one evaluation score greater than or equal to the preset evaluation score, at least one fourth power decision plan can be obtained, and the power generation risk level of each fourth power decision plan can be determined by the total power generation score of each fourth power decision plan. The power consumption risk level of each fourth power decision plan is determined by the total power consumption score of each fourth power decision plan, so as to determine at least one target risk level according to at least one power generation risk level and at least one power consumption risk level. Then, the power decision plan with a target risk level less than or equal to the preset risk level is taken as the fifth power decision plan, and at least one fifth power decision plan is obtained. Finally, the target power decision plan is determined according to the total power purchase cost score corresponding to each fifth power decision plan in at least one fifth power decision plan. Thus, by evaluating the score, risk and cost of the power decision plan, a reasonable power decision plan can be determined as the target power decision plan, thereby improving the accuracy of the power demand decision.

[0153] Exemplarily, the method may further include:

[0154] Obtain the average power supply rate of the power supply end;

[0155] Determine multiple power load values ​​of the power supply end within a target future time period according to the average power supply rate and the target power data;

[0156] Performing curve fitting on multiple power load values ​​to obtain a first power load curve segment corresponding to a target future time period;

[0157] Extracting a power load curve segment whose power load value is higher than a preset power load value from the first power load curve segment to obtain at least one second power load curve segment;

[0158] Determine a tangent slope corresponding to the power load value in each second power load curve segment in at least one second power load curve segment to obtain at least one slope;

[0159] When a target slope greater than a preset slope exists in at least one slope, warning information corresponding to the target slope is generated.

[0160] The average power supply rate of the power supply end is the power conversion rate of the power supply end transmitting power to the demand end. The power load value is the total load charge value of the power supply end.

[0161] Specifically, the power demand decision device first obtains the average power supply rate of the power supply end from the database. Then, multiple power data values ​​corresponding to the target power data are determined. The products of the multiple power data values ​​corresponding to the target power data and the average power supply rate are used as multiple power load values ​​of the power supply end in the target future time period.

[0162] Further, a curve fitting is performed on multiple power load values ​​to obtain a first power load curve segment corresponding to the target future time period. A power load curve segment having a power load value higher than a preset power load value is extracted from the first power load curve segment to obtain at least one second power load curve segment. It can be understood that the curve segment length of each second power load curve segment is different, and the tangent slope corresponding to each second power load curve segment is also different. The tangent slope of each second power load curve segment can reflect the rate of change of the power load value.

[0163] Based on this, the power demand decision device determines the tangent slope corresponding to the power load value in each second power load curve in at least one second power load curve segment to obtain at least one slope. When there is a target slope greater than the preset slope in at least one slope, it indicates that the power load value at the demand end changes abnormally, and the power demand decision device generates warning information corresponding to the target slope. Therefore, when the power load value changes abnormally, corresponding warning information can be generated to help users monitor the power system abnormally and improve the stability of the power system.

[0164] Optionally, the warning information corresponding to the target slope can be sent to the user interaction interface and displayed through the user interaction interface to indicate the warning information to the user. Optionally, the user interaction interface can display the second power load curve segment corresponding to the warning information to show the user the changing trend of the power load value. Among them, the display content in the user interaction interface can be pre-set according to user needs. Among them, the user interaction interface will also display the above-mentioned target power decision plan, and the user needs to obtain access rights through a pre-set verification password to view the target power decision plan.

[0165] In summary, in an embodiment of the present application, firstly, the historical power data of the demand side in the target area within the historical time period is obtained, and the historical power data is input into the target power demand prediction model to obtain the target power data of the demand side in the target area within the target future time period. Then, according to the target power data, n power decision schemes are determined. Furthermore, the n power decision schemes are evaluated to obtain n evaluation scores. Finally, according to the n evaluation scores, one power decision scheme is selected from the n power decision schemes as the target power decision scheme. Thus, by pre-training the target power demand prediction model, the accuracy of the target power data prediction can be improved. According to the target power data, n power decision schemes are determined, and the n power decision schemes are evaluated. A reasonable target power decision scheme can be selected from the n power decision schemes to improve the accuracy of the power demand decision.

[0166] See also Figure 3 , Figure 3 The schematic diagram of a power demand decision device provided in an embodiment of the present application is shown in FIG. 300 . The power demand decision device 300 may be the power demand decision device of any of the above embodiments. The power demand decision device 300 includes an acquisition unit 301 and a processing unit 302 .

[0167] An acquisition unit 301 is used to acquire historical power data of a demand side in a target area within a historical time period;

[0168] The processing unit 302 is used to input the historical power data into the target power demand prediction model to obtain the target power data of the demand side in the target area in the target future time period;

[0169] According to the target power data, n power decision schemes are determined; the power decision schemes are used to allocate the target power data of the demand side in the target future time period, the power generation data of the power generation side in the target future time period, and the market power purchase data of the power supply side in the target future time period; n is an integer greater than 1;

[0170] Evaluate n power decision plans and obtain n evaluation scores; each power decision plan corresponds to one evaluation score;

[0171] According to the n evaluation scores, one power decision scheme is selected from the n power decision schemes as a target power decision scheme.

[0172] In a possible embodiment, in determining n power decision solutions according to the target power data, the processing unit 302 is specifically configured to:

[0173] Obtain i historical demand scenarios and j emergency plan scenarios corresponding to the target area; i and j are both integers greater than 1; the number of permutations and combinations of j emergency plan scenarios is greater than or equal to n-1;

[0174] Determine the time periods corresponding to i historical demand scenarios in the target future time period, and obtain i time periods;

[0175] According to i historical demand scenarios, determine the power data corresponding to each time period in i time periods to obtain i power data;

[0176] Determine i power generation data of the power generation end within i time period according to i power data;

[0177] According to i power data and i power generation data, determine the market power purchase data of the power supply end in i time period;

[0178] Determine a first power decision-making plan according to i power data within i time periods, i power generation data, and market power purchase data within i time periods;

[0179] Obtain n-1 combinations of j emergency plan scenarios, and obtain n-1 emergency plan scenario combinations;

[0180] Determine the time period corresponding to each of the n-1 plan scenario combinations in the target future time period, and obtain n-1 time periods;

[0181] Determine the power data corresponding to each time period in n-1 time periods according to n-1 plan scenario combinations, and obtain n-1 power data;

[0182] Determine the power generation data corresponding to each power data in the n-1 power data to obtain n-1 power generation data;

[0183] Determine the market power purchase data within n-1 time periods based on n-1 power data and n-1 power generation data;

[0184] Determine an electricity decision plan based on each of the n-1 electricity data, the power generation data corresponding to the electricity data, the market power purchase data in the n-1 time periods, and the i electricity data in the i time periods, and obtain n-1 second electricity decision plans;

[0185] The first power decision scheme and n-1 second power decision schemes are taken as n power decision schemes.

[0186] In a possible embodiment, in evaluating n power decision-making schemes to obtain n evaluation scores, the processing unit 302 is specifically configured to:

[0187] Obtaining a third power decision-making scheme; the third power decision-making scheme is any one of the n power decision-making schemes;

[0188] Divide the target future time period into k time periods; k is an integer greater than 1;

[0189] Determine the power data of the demand side in each of k time periods in the third power decision-making scheme to obtain k power data;

[0190] According to the k power data, determining the power generation data of the power generation end in each of the k time periods in the third power decision-making scheme, and obtaining k power generation data;

[0191] Get the current reserve power data;

[0192] According to the reserve power data, k power generation data and k power data, m market power purchase data are determined; m is a positive integer less than or equal to k;

[0193] An evaluation score corresponding to a third electricity decision-making scheme is determined based on the k power generation data, the k power data and the m market power purchase data.

[0194] In a possible embodiment, in determining the evaluation score corresponding to the third power decision-making scheme according to the k power generation data, the k power data and the m market power purchase data, the processing unit 302 is specifically configured to:

[0195] Obtaining the power generation data threshold of the power generation end, and obtaining the power data threshold of the power demand end;

[0196] Determine a plurality of power generation data values ​​corresponding to each of the k power generation data;

[0197] Determine the scores corresponding to the power generation data whose multiple power generation data values ​​are all less than or equal to the power generation data threshold value among the k power generation data as the first power generation score, and obtain a first power generation scores; a is an integer greater than or equal to 0 and less than k;

[0198] Determine that there is power generation data greater than the power generation data threshold value among multiple power generation data values ​​in the k power generation data, and obtain b power generation data; b is an integer greater than or equal to 0 and less than k, a+b=k;

[0199] According to a plurality of power generation data values ​​and a power generation data threshold corresponding to each of the b power generation data, a first attenuation factor corresponding to the power generation data is determined to obtain b first attenuation factors; each power generation data corresponds to one first attenuation factor;

[0200] Determining b second power generation scores according to the b first attenuation factors;

[0201] Determine multiple power data values ​​corresponding to each power data in the k power data;

[0202] Determine the scores corresponding to the power data whose multiple power data values ​​are all less than or equal to the power data threshold value among the k power data as the first power usage score, and obtain c first power usage scores; c is an integer greater than or equal to 0 and less than k;

[0203] Determine that there is power data greater than the power data threshold value among multiple power data values ​​in the k power data, and obtain d power data; d is an integer greater than or equal to 0 and less than k, c+d=k;

[0204] According to the multiple power data values ​​corresponding to each power data in the d power data, determine the second attenuation factor corresponding to the power data, and obtain d second attenuation factors; each power data corresponds to one second attenuation factor;

[0205] Determining d second electricity usage scores according to the d second attenuation factors;

[0206] According to the m market electricity purchasing data, m electricity purchasing cost scores are determined;

[0207] Based on a first power generation score, b second power generation score, c first electricity consumption score, d second electricity consumption score and m electricity purchase cost scores, an evaluation score corresponding to the third electricity decision-making scheme is determined.

[0208] In a possible embodiment, in determining the evaluation score corresponding to the third power decision scheme based on a first power generation score, b second power generation score, c first power consumption score, d second power consumption score and m power purchase cost score, the processing unit 302 is specifically used to:

[0209] Determine a total power generation score according to a first power generation score and b second power generation scores;

[0210] Determine a total electricity usage score based on c first electricity usage scores and d second electricity usage scores;

[0211] According to the m electricity purchase cost scores, a total electricity purchase cost score is determined;

[0212] According to the total power generation score, the total power consumption score and the total power purchase cost score, the evaluation score corresponding to the third power decision-making plan is determined; among which, the total power purchase cost score is equal to (a / k*s1+c / k*s2+s3-m / k*s3), s1 represents the total power generation score, s2 represents the total power consumption score, and s3 represents the total power purchase cost score.

[0213] In a possible embodiment, in selecting one power decision scheme from the n power decision schemes as the target power decision scheme according to the n evaluation scores, the processing unit 302 is specifically configured to:

[0214] Determine at least one evaluation score among the n evaluation scores whose evaluation score is greater than or equal to the preset evaluation score, obtain the power decision plan corresponding to the at least one evaluation score, and obtain at least one fourth power decision plan;

[0215] Determine the power generation risk level of each fourth power decision scheme according to the total power generation score corresponding to each fourth power decision scheme in at least one fourth power decision scheme, and obtain at least one power generation risk level;

[0216] Determine the power consumption risk level of each fourth power decision scheme according to the total power consumption score corresponding to each fourth power decision scheme in at least one fourth power decision scheme, and obtain at least one power consumption risk level;

[0217] Determining at least one target risk level according to at least one power generation risk level and at least one power consumption risk level;

[0218] Determine at least one fourth power decision plan whose target risk level is less than or equal to a preset risk level as a fifth power decision plan, and obtain at least one fifth power decision plan;

[0219] A target electricity decision scheme is determined according to a total score of electricity purchase costs corresponding to each fifth electricity decision scheme in at least one fifth electricity decision scheme.

[0220] In a possible embodiment, the processing unit 302 is further configured to:

[0221] Obtain the average power supply rate of the power supply end;

[0222] Determine multiple power load values ​​of the power supply end within a target future time period according to the average power supply rate and the target power data;

[0223] Performing curve fitting on multiple power load values ​​to obtain a first power load curve segment corresponding to a target future time period;

[0224] Extracting a power load curve segment whose power load value is higher than a preset power load value from the first power load curve segment to obtain at least one second power load curve segment;

[0225] Determine a tangent slope corresponding to the power load value in each second power load curve segment in at least one second power load curve segment to obtain at least one slope;

[0226] When a target slope greater than a preset slope exists in at least one slope, warning information corresponding to the target slope is generated.

[0227] In a possible embodiment, the processing unit 302 is further configured to:

[0228] Obtaining a preset power demand forecasting model;

[0229] Acquire a first training power data set and a plurality of first actual power data; the first training power data set includes historical power data of the demand side in a plurality of time periods, and the plurality of first actual power data correspond one to one with the historical power data in the plurality of time periods;

[0230] Inputting the first training power data set into a preset power demand prediction model to obtain a plurality of first expected power data; the plurality of first expected power data corresponds one-to-one to the plurality of first actual power data;

[0231] Determine a mean square error between a plurality of first expected power data and a plurality of first actual power data, and use the mean square error as a model loss;

[0232] If the model loss is less than or equal to the preset loss, the preset power demand forecasting model is determined as the target power demand forecasting model;

[0233] If the model loss is greater than the preset loss, the model parameters of the preset power demand forecasting model are adjusted to obtain a candidate power demand forecasting model; a second training power data set and multiple second actual power data are obtained; the candidate power demand forecasting model is trained according to the second training power data set and multiple second actual power data to obtain a target power demand forecasting model.

[0234] See also Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 4 As shown, the electronic device 400 includes a transceiver 401, a processor 402 and a memory 403. They are connected via a bus 404. The memory 403 is used to store computer programs and data, and can transmit the data stored in the memory 403 to the processor 402. The electronic device 400 can be the power demand decision device 300.

[0235] The processor 402 is used to read the computer program in the memory 403 and perform the following operations:

[0236] Obtain historical power data of the demand side in the target area within a historical time period;

[0237] Inputting historical power data into a target power demand forecasting model to obtain target power data of the demand side in the target area in a target future time period;

[0238] According to the target power data, n power decision schemes are determined; the power decision schemes are used to allocate the target power data of the demand side in the target future time period, the power generation data of the power generation side in the target future time period, and the market power purchase data of the power supply side in the target future time period; n is an integer greater than 1;

[0239] Evaluate n power decision plans and obtain n evaluation scores; each power decision plan corresponds to one evaluation score;

[0240] According to the n evaluation scores, one power decision scheme is selected from the n power decision schemes as a target power decision scheme.

[0241] The above mainly introduces the scheme of the embodiment of the present application from the perspective of the execution process of the method side. It is understandable that the electronic device 400 includes a hardware structure and / or software module corresponding to each function in order to realize the above functions. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the embodiment provided herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present application.

[0242] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. The computer program is executed by a processor to implement part or all of the steps of any one of the methods described in the above method embodiments.

[0243] An embodiment of the present application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute part or all of the steps of any one of the methods recorded in the above method embodiments.

[0244] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present application.

[0245] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0246] In the several embodiments provided in the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the device or unit can be electrical or other forms.

[0247] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0248] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of a software program module.

[0249] If the integrated unit is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or CD-ROM and other media that can store program codes.

[0250] A person skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable memory, and the memory can include: a flash drive, a read-only memory (English: Read-Only Memory, abbreviated as: ROM), a random access memory (English: Random Access Memory, abbreviated as: RAM), a magnetic disk or an optical disk, etc.

[0251] The embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for making decisions on power demand, characterized in that: include: Obtain historical power data of the demand side in the target area within a historical time period; Inputting the historical power data into a target power demand forecasting model to obtain target power data of the demand side in the target area within a target future time period; Determining n power decision-making schemes according to the target power data; The power decision-making scheme is used to allocate the target power data of the demand side in the target future time period, the power generation data of the power generation side in the target future time period, and the market power purchase data of the power supply side in the target future time period; n is an integer greater than 1; Evaluate the n power decision-making schemes to obtain n evaluation scores; Each power decision-making scheme corresponds to an evaluation score; According to the n evaluation scores, one power decision scheme is selected from the n power decision schemes as a target power decision scheme.

2. The method according to claim 1, characterized in that The step of determining n power decision schemes according to the target power data includes: Obtain i historical demand scenarios and j emergency plan scenarios corresponding to the target area; i and j are both integers greater than 1; the number of permutations and combinations of the j emergency plan scenarios is greater than or equal to n-1; Determine the time periods corresponding to the i historical demand scenarios in the target future time period to obtain i time periods; According to the i historical demand scenarios, determine the power data corresponding to each time period in the i time periods to obtain i power data; Determine i power generation data of the power generation end within the i time period according to the i power data; Determine the market power purchase data of the power supply terminal within the i time period according to the i power data and the i power generation data; Determine a first power decision-making plan according to the i power data within the i time periods, the i power generation data and the market power purchase data within the i time periods; Obtain n-1 combinations of the j emergency plan scenarios to obtain n-1 emergency plan scenario combinations; Determine the time period corresponding to each of the n-1 plan scenario combinations in the target future time period to obtain n-1 time periods; Determine the power data corresponding to each time period in n-1 time periods according to the n-1 plan scenario combinations, and obtain n-1 power data; Determine the power generation data corresponding to each power data in the n-1 power data to obtain n-1 power generation data; Determine the market power purchase data within the n-1 time period according to the n-1 power data and the n-1 power generation data; Determine an electricity decision plan according to each of the n-1 electricity data, the power generation data corresponding to the electricity data, the market power purchase data in the n-1 time periods, and the i electricity data in the i time periods, and obtain n-1 second electricity decision plans; The first power decision scheme and the n-1 second power decision schemes are used as the n power decision schemes.

3. The method according to claim 1 or 2, characterized in that: The n power decision-making schemes are evaluated to obtain n evaluation scores, including: Obtaining a third power decision scheme; the third power decision scheme is any one of the n power decision schemes; Divide the target future time period into k time periods; k is an integer greater than 1; Determine the power data of the demand side in each of the k time periods in the third power decision-making scheme to obtain k power data; According to the k power data, determining the power generation data of the power generation end in each of the k time periods in the third power decision-making scheme to obtain k power generation data; Get the current reserve power data; Determine m market power purchase data according to the reserve power data, the k power generation data and the k power data; m is a positive integer less than or equal to k; An evaluation score corresponding to the third electricity decision-making scheme is determined according to the k power generation data, the k power data and the m market power purchase data.

4. The method according to claim 3, characterized in that The step of determining the evaluation score corresponding to the third power decision-making scheme according to the k power generation data, the k power data and the m market power purchase data includes: Acquiring a power generation data threshold of the power generation end, and acquiring a power data threshold of the power demand end; Determine a plurality of power generation data values ​​corresponding to each of the k power generation data; Determine that scores corresponding to power generation data whose values ​​among the k power generation data are all less than or equal to the power generation data threshold are first power generation scores, and obtain a first power generation scores; a is an integer greater than or equal to 0 and less than k; Determine that there is power generation data greater than the power generation data threshold value among the multiple power generation data values ​​in the k power generation data, and obtain b power generation data; b is an integer greater than or equal to 0 and less than k, a+b=k; According to a plurality of power generation data values ​​corresponding to each of the b power generation data and the power generation data threshold, determining a first attenuation factor corresponding to the power generation data, and obtaining b first attenuation factors; each power generation data corresponds to one first attenuation factor; Determining b second power generation scores according to the b first attenuation factors; Determine a plurality of power data values ​​corresponding to each power data in the k power data; Determine the scores corresponding to the power data whose values ​​among the k power data are all less than or equal to the power data threshold as the first power usage score, and obtain c first power usage scores; c is an integer greater than or equal to 0 and less than k; Determine that there is power data greater than the power data threshold value among the multiple power data values ​​in the k power data, and obtain d power data; d is an integer greater than or equal to 0 and less than k, c+d=k; According to the multiple power data values ​​corresponding to each power data in the d power data, determine the second attenuation factor corresponding to the power data, and obtain d second attenuation factors; each power data corresponds to one second attenuation factor; Determining d second electricity usage scores according to the d second attenuation factors; Determining m electricity purchase cost scores according to the m market electricity purchase data; Based on the a first power generation scores, the b second power generation scores, the c first electricity usage scores, the d second electricity usage scores and the m electricity purchase cost scores, an evaluation score corresponding to the third electricity decision-making scheme is determined.

5. The method according to claim 4, characterized in that The determining, based on the a first power generation scores, the b second power generation scores, the c first power consumption scores, the d second power consumption scores and the m power purchase cost scores, an evaluation score corresponding to the third power decision scheme includes: Determining a total power generation score according to the a first power generation scores and the b second power generation scores; Determining a total electricity usage score according to the c first electricity usage scores and the d second electricity usage scores; Determining a total electricity purchase cost score according to the m electricity purchase cost scores; The evaluation score corresponding to the third electricity decision-making scheme is determined according to the total power generation score, the total power consumption score and the total power purchase cost score; wherein the total power purchase cost score is equal to (a / k*s1+c / k*s2+s3-m / k*s3), s1 represents the total power generation score, s2 represents the total power consumption score, and s3 represents the total power purchase cost score.

6. The method according to claim 5, characterized in that The selecting, according to the n evaluation scores, one power decision scheme from the n power decision schemes as a target power decision scheme comprises: Determine at least one evaluation score among the n evaluation scores whose evaluation score is greater than or equal to a preset evaluation score, obtain an electric power decision plan corresponding to the at least one evaluation score, and obtain at least one fourth electric power decision plan; Determine the power generation risk level of each fourth power decision scheme according to the total power generation score corresponding to each fourth power decision scheme in the at least one fourth power decision scheme, and obtain at least one power generation risk level; Determine the power consumption risk level of each fourth power decision scheme according to the total power consumption score corresponding to each fourth power decision scheme in the at least one fourth power decision scheme, and obtain at least one power consumption risk level; Determining at least one target risk level according to the at least one power generation risk level and the at least one power consumption risk level; Determine the power decision scheme whose target risk level is less than or equal to the preset risk level in the at least one fourth power decision scheme as the fifth power decision scheme, and obtain at least one fifth power decision scheme; The target electricity decision scheme is determined according to the total score of electricity purchase costs corresponding to each fifth electricity decision scheme in the at least one fifth electricity decision scheme.

7. The method according to claim 1 or 2, characterized in that: The method further comprises: Obtaining an average power supply rate of the power supply end; Determining a plurality of power load values ​​of the power supply end within the target future time period according to the average power supply rate and the target power data; Performing curve fitting on the multiple power load values ​​to obtain a first power load curve segment corresponding to the target future time period; Extracting a power load curve segment whose power load value is higher than a preset power load value from the first power load curve segment to obtain at least one second power load curve segment; Determine a tangent slope corresponding to the power load value in each second power load curve segment in the at least one second power load curve segment to obtain at least one slope; When a target slope greater than a preset slope exists in the at least one slope, warning information corresponding to the target slope is generated.

8. The method according to claim 1 or 2, characterized in that: The method further comprises: Obtaining a preset power demand forecasting model; Acquire a first training power data set and a plurality of first actual power data; the first training power data set includes historical power data of the demand end in a plurality of time periods, and the plurality of first actual power data correspond one-to-one to the historical power data in a plurality of time periods; Inputting the first training power data set into the preset power demand prediction model to obtain a plurality of first expected power data; the plurality of first expected power data corresponds one-to-one to the plurality of first actual power data; determining a mean square error between the plurality of first expected power data and the plurality of first actual power data, and using the mean square error as a model loss; If the model loss is less than or equal to the preset loss, determining the preset power demand forecasting model as the target power demand forecasting model; If the model loss is greater than the preset loss, the model parameters of the preset power demand forecasting model are adjusted to obtain a candidate power demand forecasting model; a second training power data set and a plurality of second actual power data are obtained; the candidate power demand forecasting model is trained according to the second training power data set and the plurality of second actual power data to obtain the target power demand forecasting model.

9. A power demand decision device, characterized in that: The device comprises: An acquisition unit, used for acquiring historical power data of a demand side in a target area within a historical time period; A processing unit, configured to input the historical power data into a target power demand forecasting model to obtain target power data of the demand side in the target area within a target future time period; According to the target power data, n power decision schemes are determined; the power decision schemes are used to allocate the target power data of the demand side within the target future time period, the power generation data of the power generation side within the target future time period, and the market power purchase data of the power supply side within the target future time period; n is an integer greater than 1; Evaluate the n electric power decision-making schemes to obtain n evaluation scores; each electric power decision-making scheme corresponds to an evaluation score; According to the n evaluation scores, one power decision scheme is selected from the n power decision schemes as a target power decision scheme.

10. An electronic device, characterized in that: include: A processor, a memory, a communication interface, and one or more programs; The one or more programs are stored in the memory and configured to be executed by the processor, the programs comprising instructions for executing the steps in the method according to any one of claims 1 to 8.

Images