A method and system for predicting and adjusting the client-side power load using big data

Abstract

The present application relates to the technical field of power management, and discloses a method and system for predicting and adjusting the power load of the user side using big data. The method includes constructing a power consumption scenario library, performing real-time data collection, generating real-time power consumption combined data, generating a first prediction result, and generating a second prediction result. The system corresponds to the method. In the present application, by distinguishing between the energy storage type and non-energy storage type user sides, the power consumption combination situations of different types of users under different power consumption scenarios are fully considered. The introduction of real-time energy storage data reflects the changes in the real-time energy storage state of the energy storage type user side during the generation process of real-time power consumption prediction data and the second prediction result, effectively making up for the deficiency of the traditional method that does not consider the energy storage factor, significantly improving the accuracy of power load prediction in complex power consumption environments, and providing strong support for the stable operation of the power grid and the reasonable allocation of resources.

Description

Technical Field

[0001] This application relates to the technical field of power management, and specifically to a method and system for predicting and adjusting the power load of the user side using big data. Background Art

[0002] With the rapid development of the power industry, power load forecasting has become a key link to ensure the stable operation of the power grid. Traditional power load forecasting technologies show the following problems when facing complex and changing user electricity consumption behaviors: The previous methods usually only analyzed based on historical electricity consumption data simply, and failed to fully consider the impact of different electricity consumption scenarios on the load. Specifically, for example, in special scenarios such as sudden weather changes, holidays, large-scale events, and charging of new energy equipment (such as new energy vehicles), the electricity consumption demand at the user side will change significantly, but the existing technologies are difficult to accurately identify and analyze these changes, resulting in large deviations in the forecasting results.

[0003] There are some methods using big data in the prior art. For example, Chinese Patent No. CN118889402A discloses a power load forecasting method and system based on big data-driven. Although it integrates regional power station data and population density data to a certain extent, in the analysis process, it is unable to deeply analyze the electricity consumption scenarios corresponding to the power load at the user side, making the analysis of the user side inaccurate. This leads to the inability to provide a reliable basis for the reasonable dispatching of the power grid and the effective allocation of power resources in practical applications, seriously affecting the safe and stable operation of the power grid and the power supply quality.

[0004] In summary, there is an urgent need for a new technical solution for predicting and adjusting the power load of the user side using big data to solve the above technical problems. Summary of the Invention

[0005] The purpose of this application is to provide a method and system for predicting and adjusting the power load of the user side using big data to solve the technical problems proposed in the above background art.

[0006] To achieve the above purpose, this application discloses the following technical solutions:

[0007] In the first aspect, this application discloses a method for predicting and adjusting the power load of the user side using big data, and the method includes:

[0008] Construct an electricity consumption scenario library based on the collected historical electricity consumption data of the user terminals; wherein, the user terminals include electricity storage user terminals and non-electricity storage user terminals, the electricity storage user terminals have the ability to store electricity, the non-electricity storage user terminals do not have the ability to store electricity, the historical electricity consumption data is used to characterize the independent electricity consumption situation of the historical user terminals, and the electricity consumption scenario library stores the electricity consumption scenarios formed when different user terminals work simultaneously, and this electricity consumption scenario is used to characterize the combined electricity consumption situation of different user terminals consuming electricity simultaneously;

[0009] Collect the real-time electricity consumption data of the user terminals, and when the user terminal is the electricity storage user terminal, collect its corresponding real-time electricity storage data; wherein, the real-time electricity consumption data is used to characterize the independent electricity consumption situation of the real-time user terminals, and the real-time electricity storage data is used to characterize the electricity storage situation of the real-time electricity storage user terminals;

[0010] Use the electricity consumption scenario library to match the electricity consumption scenario corresponding to the real-time electricity consumption data, and generate real-time electricity consumption combined data based on this electricity consumption scenario; wherein, the real-time electricity consumption combined data is used to characterize the combined electricity consumption situation of different user terminals consuming electricity simultaneously in real time;

[0011] Use a preset power load prediction algorithm, and generate a first prediction result based on the real-time electricity consumption combined data and its corresponding electricity consumption scenario; wherein, the power load prediction algorithm is used for power load prediction, and the first prediction result is the result of rough power prediction;

[0012] Generate real-time electricity consumption prediction data by using the real-time electricity storage data and the real-time electricity consumption combined data, and generate a second prediction result by using the real-time electricity consumption prediction data, the first prediction result and the power load prediction algorithm; wherein, the real-time electricity consumption prediction data is used to characterize the combined electricity consumption situation of different user terminals in real time and predicted, considering the change of the real-time electricity storage data, and the second prediction result is the result of refined power prediction adjusted considering the real-time electricity storage data.

[0013] Preferably, the construction of the electricity consumption scenario library includes:

[0014] Clean the collected electricity consumption historical data, perform the first clustering on the data after cleaning based on a preset electricity consumption pattern, add a description of the electricity consumption situation to the result after the first clustering, and then perform the second clustering to obtain an electricity consumption scenario. Use the electricity consumption scenario to build a database to obtain an electricity consumption scenario database. Among them, the electricity consumption pattern at least includes the type, quantity, and electricity consumption power range of the user side under this pattern, the description of the electricity consumption situation at least includes the electricity consumption time distribution characteristics and charge-discharge characteristics of the user side under this electricity consumption situation. The electricity consumption time distribution characteristics are used to characterize the electricity consumption time distribution of different user sides, and the charge-discharge characteristics are used to characterize the charge-discharge situation of the electricity storage type user side.

[0015] Preferably, the electricity storage type user side at least includes new energy devices composed of energy storage devices in the power grid and new energy vehicles owned by users.

[0016] Preferably, the collection of the electricity storage real-time data includes:

[0017] Use the electricity quantity collection devices preset in the energy storage device and the new energy device to collect the electricity storage real-time data of the energy storage device and the new energy device respectively.

[0018] Preferably, the matching of the electricity consumption scenario includes:

[0019] Extract the real-time mode characteristics and real-time scenario characteristics of the electricity consumption real-time data, calculate the similarity between the electricity consumption real-time data and the electricity consumption scenario based on the real-time mode characteristics and the real-time scenario characteristics, and select the electricity consumption scenario with the highest similarity as the matching result. Among them, the real-time mode characteristics at least include the characteristics corresponding to the type, quantity, and electricity consumption power range of the real-time user side, and the real-time scenario characteristics at least include the electricity consumption time distribution characteristics and charge-discharge characteristics of the real-time user side.

[0020] The extraction of the electricity consumption time distribution characteristics includes:

[0021] Based on the distribution of the electricity consumption behavior of different user sides over time within a preset prediction period, extract the electricity consumption concentration degree and electricity consumption pattern of different user sides at different times to obtain the electricity consumption time distribution characteristics.

[0022] The extraction of the charge-discharge characteristics includes:

[0023] Based on the charge-discharge behavior of the electricity storage type user side, extract the impact of the charge-discharge behavior on the overall power load and the impact on the adjustment of the power load prediction as the charge-discharge characteristics. Among them, the charge-discharge behavior at least includes the charge-discharge power, charge-discharge capacity, charge-discharge efficiency, and charge-discharge time law.

[0024] Preferably, the generation of the electricity consumption real-time combined data includes:

[0025] Determine the client involved according to the matched power consumption scenario to obtain the corresponding real-time power consumption data, and use the obtained real-time power consumption data to construct a real-time power consumption combined data matrix. The matrix elements of the real-time power consumption combined data matrix are the real-time mode features of each client. Define the real-time power consumption combined data matrix as real-time power consumption combined data and output it.

[0026] Preferably, the generation of the first prediction result includes:

[0027] Input the real-time power consumption combined data into the power load prediction algorithm, output an initial prediction result with a time series, and use the power consumption scenario to correct the initial prediction result to obtain the first prediction result; wherein, the use of the power consumption scenario to correct the initial prediction result is:

[0028] Use the initial prediction result correction formula to obtain the first prediction result. The initial prediction result correction formula is:

[0029]

[0030] Wherein, is the initial predicted value at time is the correction factor obtained based on the power consumption scenario at time . This correction factor is used to characterize the influence degree of the power consumption scenario on the real-time power consumption data. is the first predicted value at time calculated. The first prediction result is composed of the first predicted value based on the time series.

[0031] Preferably, the generation of the real-time power consumption prediction data includes:

[0032] Calculate the real-time power consumption prediction data using the real-time power consumption prediction data calculation formula. The real-time power consumption prediction data calculation formula is:

[0033]

[0034] Wherein, represents the real-time power consumption data of client at time ; represents the total number of clients in the real-time power consumption combined data corresponding to client at time and ; represents the prediction adjustment value for power supply substitution of the energy storage client based on the real-time energy storage data at time . This prediction adjustment value is obtained based on the power consumption scenario. The calculated real-time power consumption prediction value at a moment, and the real-time power consumption prediction data is composed of the real-time power consumption prediction values based on a time series.

[0035] Preferably, the generation of the second prediction result includes:

[0036] Calculating the second prediction result by using a second prediction result calculation formula, and the second prediction result calculation formula is:

[0037]

[0038] where is the first prediction value at the moment corresponding to the first prediction result, is the first prediction value at the moment, is a preset prediction period is the confidence level of the real-time power consumption prediction data obtained from the error rate generated, and this confidence level is obtained based on a preset confidence level query table. Different confidence levels of the real-time power consumption prediction data corresponding to different error rates are stored in the confidence level query table, and the preset prediction period , where is the start time corresponding to the prediction period, is the end time corresponding to the prediction period, is the real-time power consumption prediction data, is the calculated second prediction value at the moment, and the second prediction result is composed of the second prediction values based on a time series.

[0039] In a second aspect, the present application discloses a user-side power load prediction and adjustment system using big data. This system is applicable to the user-side power load prediction and adjustment method using big data as described above. The system includes an electricity consumption scenario library module, a real-time data acquisition module, a real-time combined data generation module, a first prediction module, and a second prediction module that are communicatively connected;

[0040] The electricity consumption scenario library module is configured to construct an electricity consumption scenario library based on the collected historical electricity consumption data of the user side; wherein, the user side includes a storage-type user side and a non-storage-type user side. The storage-type user side has the ability to store electricity, and the non-storage-type user side does not have the ability to store electricity. The historical electricity consumption data is used to represent the independent electricity consumption situation of the historical user side. Different electricity consumption scenarios formed when different user sides work simultaneously are stored in the electricity consumption scenario library, and this electricity consumption scenario is used to represent the combined electricity consumption situation of different user sides when they use electricity simultaneously;

[0041] The real-time data acquisition module is configured to: acquire the real-time power consumption data of the client, and when the client is the energy storage type client, acquire the corresponding real-time energy storage data; wherein, the real-time power consumption data is used to characterize the independent power consumption situation of the client in real time, and the real-time energy storage data is used to characterize the energy storage situation of the energy storage type client in real time;

[0042] The real-time combined data generation module is configured to: match the power consumption scenario corresponding to the real-time power consumption data by using the power consumption scenario library, and generate real-time power consumption combined data based on this power consumption scenario; wherein, the real-time power consumption combined data is used to characterize the combined power consumption situation of different clients consuming power simultaneously in real time;

[0043] The first prediction module is configured to: use a preset power load prediction algorithm, and generate a first prediction result based on the real-time power consumption combined data and its corresponding power consumption scenario; wherein, the power load prediction algorithm is used for power load prediction, and the first prediction result is the result of rough power prediction;

[0044] The second prediction module is configured to: generate real-time power consumption prediction data by using the real-time energy storage data and the real-time power consumption combined data, and generate a second prediction result by using the real-time power consumption prediction data, the first prediction result and the power load prediction algorithm; wherein, the real-time power consumption prediction data is used to characterize the combined power consumption situation of different clients consuming power simultaneously in real time and considering the change of the real-time energy storage data in prediction, and the second prediction result is the result of refined power prediction adjusted considering the real-time energy storage data.

[0045] Beneficial effects: The method and system for predicting and adjusting the power load of the client using big data in this application use the historical power consumption data of the client collected to construct a power consumption scenario library, and combine the real-time power consumption data and the real-time energy storage data to achieve a comprehensive analysis and accurate prediction and adjustment of the power load of the client; by distinguishing between energy storage type and non-energy storage type clients, it fully considers the power consumption combination situations of different types of users in different power consumption scenarios. The introduction of the real-time energy storage data reflects the change of the real-time energy storage state of the energy storage type client in the process of generating the real-time power consumption prediction data and the second prediction result, effectively compensating for the deficiency of the traditional method that does not consider the energy storage factor, and significantly improving the accuracy of power load prediction in a complex power consumption environment, providing strong support for the stable operation of the power grid and the reasonable allocation of resources. Description of the Drawings

[0046] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those skilled in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0047] Figure 1 The flowchart of the method for predicting and adjusting the power load of the user side using big data provided by the embodiment of the present application;

[0048] Figure 2 The structural block diagram of the system for predicting and adjusting the power load of the user side using big data provided by the embodiment of the present application. Specific embodiments

[0049] The following will clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope protected by the present application.

[0050] In this article, the term "comprising" is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such a process, method, article or device. Without further limitation, the elements defined by the statement "comprising..." do not exclude the existence of additional identical elements in the process, method, article or device comprising the said elements.

[0051] The first aspect of this embodiment discloses a method for predicting and adjusting the power load of the user side using big data as shown in Figure 1 The method includes:

[0052] Construct an electricity consumption scenario library based on the collected historical electricity consumption data of the user side; wherein, the user side includes a storage-type user side and a non-storage-type user side. The storage-type user side has the ability to store electricity, and the non-storage-type user side does not have the ability to store electricity. The historical electricity consumption data is used to represent the independent electricity consumption situation of the historical user side. The electricity consumption scenario library stores the electricity consumption scenarios formed when different user sides work simultaneously, and the electricity consumption scenario is used to represent the combined electricity consumption situation of different user sides when consuming electricity simultaneously;

[0053] Collect the real-time power consumption data of the user terminal, and when the user terminal is a storage-type user terminal, collect its corresponding real-time energy storage data; among them, the real-time power consumption data is used to characterize the independent power consumption situation of the real-time user terminal, and the real-time energy storage data is used to characterize the energy storage situation of the storage-type user terminal in real time;

[0054] Use the power consumption scenario library to match the power consumption scenario corresponding to the real-time power consumption data, and generate real-time power consumption combined data based on this power consumption scenario; among them, the real-time power consumption combined data is used to characterize the combined power consumption situation of different user terminals consuming electricity simultaneously in real time;

[0055] Use a preset power load forecasting algorithm, and based on the real-time power consumption combined data and its corresponding power consumption scenario, generate a first prediction result; among them, the power load forecasting algorithm is used for power load forecasting, and the first prediction result is the result of rough power forecasting;

[0056] Use the real-time energy storage data and the real-time power consumption combined data to generate real-time power consumption prediction data, and use the real-time power consumption prediction data, the first prediction result and the power load forecasting algorithm to generate a second prediction result; among them, the real-time power consumption prediction data is used to characterize the combined power consumption situation of different user terminals in real time and predicted, considering the change of the real-time energy storage data, and the second prediction result is the result of refined power forecasting adjusted after considering the real-time energy storage data.

[0057] Through the above, this embodiment differentiates between storage-type and non-storage-type user terminals, fully considers the power consumption combination situations of different types of users in different power consumption scenarios, and the introduction of real-time energy storage data reflects the real-time energy storage state change of the storage-type user terminal during the generation process of real-time power consumption prediction data and the second prediction result, effectively compensating for the deficiency of the traditional method that does not consider the energy storage factor, significantly improving the accuracy of power load forecasting in a complex power consumption environment, and providing strong support for the stable operation of the power grid and the rational allocation of resources.

[0058] Specifically, the construction of the power consumption scenario library includes:

[0059] Perform data cleaning on the collected historical power consumption data, perform the first clustering on the data after data cleaning based on a preset power consumption mode, perform the second clustering on the result after the first clustering after adding a power consumption situation description to obtain a power consumption scenario, and use the power consumption scenario to build a library to obtain a power consumption scenario library; among them, the power consumption mode at least includes the type, quantity and power consumption power range of the user terminal in this mode, the power consumption situation description at least includes the power consumption time distribution characteristics and charge-discharge characteristics of the user terminal in this power consumption situation, the power consumption time distribution characteristics are used to characterize the power consumption time distribution situation of different user terminals, and the charge-discharge characteristics are used to characterize the charge-discharge situation of the storage-type user terminal.

[0060] With the above, this embodiment uses existing data cleaning and clustering technologies to clean and cluster the electricity consumption history data, and constructs an electricity consumption scenario library with the added electricity situation descriptions, achieving a detailed description of different electricity consumption scenarios. Its clustering operation based on the electricity consumption pattern and electricity situation description makes the classification of electricity consumption scenarios more accurate and reasonable, and can better capture the diversity and complexity of the electricity consumption behavior at the user end. This not only helps the subsequent accurate matching of electricity consumption scenarios, but also lays a foundation for considering the impact of various factors on the load during the power load forecasting process. Especially when combined with the real-time data of electricity storage, it can more accurately analyze the impact of the charging and discharging behavior of the electricity storage type user end on the overall load, thereby further improving the accuracy of power load forecasting and ensuring the reliability of the power grid operation.

[0061] Specifically, the electricity storage type user end at least includes new energy devices composed of energy storage devices in the power grid and the user's own new energy vehicles.

[0062] With the above, this embodiment realizes the effective integration and targeted analysis of electricity storage resources by clarifying that the electricity storage type user end includes new energy devices such as power grid energy storage devices and new energy vehicles. By including these common electricity storage type user ends in the research scope, during the subsequent data collection and prediction processes, the impact of electricity storage factors on the power load can be considered more comprehensively. Especially in the data collection link of real-time electricity consumption data and real-time electricity storage data, special data collection is carried out for the special properties of these electricity storage type user ends, providing an important data basis for improving the accuracy of power load forecasting, making the prediction results better adapt to the power environment with the wide application of new energy devices, and promoting the stable operation of the power grid.

[0063] Specifically, the collection of real-time electricity storage data includes:

[0064] Using the electricity quantity collection devices preset in the energy storage devices and new energy devices to collect the real-time electricity storage data of the energy storage devices and new energy devices respectively.

[0065] Through the above, this embodiment utilizes the existing power acquisition devices preset in energy storage devices and new energy devices to achieve accurate acquisition of real-time electricity storage data, thereby obtaining the electricity storage status of energy storage devices and new energy vehicles in real time, ensuring the timeliness and accuracy of real-time electricity storage data. Among them, the acquisition of new energy devices can, but is not limited to, collecting their SOC data using plugins installed in the terminal software of new energy vehicles, so as to obtain real-time energy storage real-time data of different users. It can be understood that with the popularization of new energy power generation, the popularization of energy storage devices is obvious. Further, with the popularization of user terminals with electricity storage capabilities such as new energy vehicles, the centralized charging and discharging behaviors of user terminals will have a significant impact on the power load prediction and adjustment process, and the centralized charging and discharging behaviors of electricity storage user terminals have significant time characteristics (which depend on new energy power generation methods, such as photovoltaic power generation, and also depend on the centralized charging of users for new energy devices, such as charging in the evening after work). Therefore, accurate real-time electricity storage data is crucial, as it can provide key input information for the generation of subsequent real-time electricity consumption prediction data and the second prediction result, enabling the power load prediction and adjustment to fully consider the changes in the charging and discharging capabilities of electricity storage user terminals, effectively correcting the prediction result, and thus significantly improving the accuracy of power load prediction and enhancing the power grid's ability to respond to changes in energy storage devices.

[0066] Specifically, the matching of electricity consumption scenarios includes:

[0067] Extracting the real-time mode characteristics and real-time scenario characteristics of real-time electricity consumption data, calculating the similarity between the real-time electricity consumption data and electricity consumption scenarios based on the real-time mode characteristics and real-time scenario characteristics, and selecting the electricity consumption scenario with the highest similarity as the matching result; among them, the real-time mode characteristics at least include the characteristics corresponding to the type, quantity, and electricity consumption power range of real-time user terminals, and the real-time scenario characteristics at least include the electricity consumption time distribution characteristics and charging and discharging characteristics of real-time user terminals;

[0068] The extraction of electricity consumption time distribution characteristics includes:

[0069] Based on the distribution of electricity consumption behaviors of different user terminals over time within a preset prediction period, extracting the electricity consumption concentration and electricity consumption patterns of different user terminals at different times to obtain the electricity consumption time distribution characteristics;

[0070] The extraction of charging and discharging characteristics includes:

[0071] Based on the charging and discharging behaviors of electricity storage user terminals, extracting the impact of charging and discharging behaviors on the overall power load and the impact on power load prediction and adjustment as the charging and discharging characteristics; among them, the charging and discharging behaviors at least include charging and discharging power, charging and discharging capacity, charging and discharging efficiency, and charging and discharging time rules.

[0072] Through the above, this embodiment extracts the real-time mode features and real-time scenario features of the real-time power consumption data, and calculates the similarity based on these features and the existing similarity analysis technology to match the power consumption scenarios, achieving efficient and accurate matching of power consumption scenarios. In the process of feature extraction, the detailed analysis of the power consumption time distribution features and charge-discharge features fully considers the time law of the power consumption behavior at the user end and the charge-discharge characteristics of the energy storage user end. By accurately matching the power consumption scenarios, it can provide a reliable basis for the generation of subsequent real-time combined power consumption data, and then better combine the actual power consumption situation in power load forecasting. Especially the comprehensive application of energy storage real-time data and these features further optimizes the forecasting process, improves the accuracy of power load forecasting, and enhances the pertinence of power grid regulation.

[0073] Specifically, the generation of real-time combined power consumption data includes:

[0074] Determine the user ends involved according to the matched power consumption scenarios to obtain the corresponding real-time power consumption data, use the obtained real-time power consumption data to construct a real-time combined power consumption data matrix, where the matrix elements of the real-time combined power consumption data matrix are the real-time mode features of each user end, and define the real-time combined power consumption data matrix as real-time combined power consumption data and output it.

[0075] Through the above, this embodiment uses the matched power consumption scenarios to determine the user ends and constructs a real-time combined power consumption data matrix, achieving effective integration and structured representation of the real-time power consumption data of different user ends. According to the power consumption scenarios, accurately screen out the real-time power consumption data of relevant user ends and organize them in matrix form, making the data more convenient to process in subsequent power load forecasting algorithms. In this process, the real-time mode features, as matrix elements, fully reflect the power consumption situation of the user ends. Combining with the energy storage real-time data, it can more comprehensively consider the influence of energy storage user ends, thereby improving the input quality of power load forecasting algorithms and ultimately enhancing the accuracy of power load forecasting and ensuring the stability of power grid operation.

[0076] Specifically, the generation of the first prediction result includes:

[0077] Input the real-time combined power consumption data into the power load forecasting algorithm, output an initial prediction result with a time series, and use the power consumption scenarios to correct the initial prediction result to obtain the first prediction result; where using the power consumption scenarios to correct the initial prediction result is:

[0078] Use the initial prediction result correction formula to obtain the first prediction result, and the initial prediction result correction formula is:

[0079]

[0080] Where is The initial predicted value at a moment, is the correction factor obtained based on the electricity consumption scenario at a moment, and this correction factor is used to characterize the influence degree of the electricity consumption scenario on the real-time electricity consumption data. is the calculated first predicted value at a moment, and the first prediction result is composed of the first predicted value based on the time series.

[0081] It should be noted that the power load prediction algorithm in this embodiment can be any existing algorithm that utilizes real-time electricity consumption combined data. This algorithm aims to obtain the initial predicted value, and the purpose of this design in this embodiment is to enhance the applicability of this embodiment.

[0082] Furthermore, the correction factor in this embodiment can be a function of time regarding the influence of the electricity consumption scenario on the real-time electricity consumption data obtained based on regression analysis, so as to construct the correction factor and use it to correct the initial prediction result.

[0083] Through the above, this embodiment uses the electricity consumption scenario to correct the initial prediction result, realizing the optimization of the prediction result based on the electricity consumption scenario information. By introducing the correction factor, the influence of the electricity consumption scenario on the real-time electricity consumption data is fully considered, making the first prediction result able to more accurately reflect the actual power load situation. In the process of calculating the correction factor, various characteristics of the electricity consumption scenario are combined, complementing the real-time electricity storage data, and further improving the consideration of the influencing factors of the power load. This correction mechanism effectively improves the reliability of the first prediction result, laying a foundation for generating a more accurate second prediction result subsequently, thereby enhancing the accuracy of the power load prediction and facilitating the reasonable dispatching of the power grid.

[0084] Specifically, the generation of real-time electricity consumption prediction data includes:

[0085] Calculating the real-time electricity consumption prediction data using the real-time electricity consumption prediction data calculation formula, and the real-time electricity consumption prediction data calculation formula is:

[0086]

[0087] Wherein, represents the real-time electricity consumption data of the user side at a moment, represents the total number of user sides in the real-time electricity consumption combined data corresponding to the user side at a moment and , represents the prediction adjustment value for power supply by the electricity storage user side obtained based on the real-time electricity storage data at a moment, and this prediction adjustment value is obtained based on the electricity consumption scenario. is the calculated​ The real-time power consumption prediction value at a moment, and the real-time power consumption prediction data is composed of the real-time power consumption prediction values based on a time series.

[0088] It should be noted that, in this embodiment, The power storage type user terminal obtained based on the real-time power storage data at a moment for acting as the prediction adjustment value of electricity can be a function of time regarding the influence of the real-time power storage data on the real-time power consumption data obtained based on regression analysis. It can be understood that, in a specific example, during the centralized charging of new energy devices, it will cause the growth of real-time power consumption data; based on this, this embodiment constructs a prediction adjustment value and uses it for the generation of real-time power consumption prediction data.

[0089] Through the above, this embodiment calculates the real-time power consumption prediction data by using the real-time power storage data and the real-time combined power consumption data, and realizes the real-time prediction of the power load considering the power storage factor. In the calculation formula, through the prediction adjustment value obtained based on the real-time power storage data, it can accurately reflect the influence of the charging and discharging of the power storage type user terminal at different moments on the overall power load. Combining the user terminal power consumption information in the real-time combined power consumption data, it comprehensively considers the changes in the power consumption behaviors of various user terminals. This calculation method makes the real-time power consumption prediction data more in line with the actual power consumption situation, provides important intermediate data for the generation of the subsequent second prediction result, significantly improves the accuracy of the power load prediction, and ensures the stable power supply of the power grid.

[0090] Specifically, the generation of the second prediction result includes:

[0091] Calculating the second prediction result by using the second prediction result calculation formula, and the second prediction result calculation formula is:

[0092]

[0093] Where, is the first prediction value at the moment corresponding to the first prediction result, is the confidence level of the real-time power consumption prediction data obtained from the error rate generated by the preset prediction period, and this confidence level is obtained based on a preset confidence level query table. Different error rates corresponding to the confidence levels of the real-time power consumption prediction data are stored in the confidence level query table, and the preset prediction period , where, , is the start moment corresponding to the prediction period, is the end moment corresponding to the prediction period, is the real-time power consumption prediction data, is the calculated second prediction value at the

[0094] With the above, this embodiment uses the first prediction result, real-time electricity consumption prediction data, and confidence based on the error rate to calculate the second prediction result, achieving a refined adjustment of the power load prediction result. By setting the confidence of the real-time electricity consumption prediction data based on the empirical values well-known to those skilled in the art, according to the preset confidence query table, combined with the error rate generated during the prediction period, the weights of the first prediction result and the real-time electricity consumption prediction data in the final prediction can be reasonably balanced. The fusion of the real-time electricity storage data in the real-time electricity consumption prediction data further enhances the accuracy of the second prediction result. This calculation method makes full use of various data information, effectively improves the accuracy of power load prediction, and provides a reliable decision-making basis for the efficient operation of the power grid.

[0095] The second aspect of this embodiment discloses a user-side power load prediction adjustment system using big data as shown in Figure 2 The system is applicable to the user-side power load prediction adjustment method using big data as above. The system includes an electricity consumption scenario library module, a real-time data acquisition module, a real-time combined data generation module, a first prediction module, and a second prediction module that are communicatively connected;

[0096] The electricity consumption scenario library module is configured to construct an electricity consumption scenario library based on the collected historical electricity consumption data of the user side; among them, the user side includes a storage-type user side and a non-storage-type user side. The storage-type user side has the ability to store electricity, and the non-storage-type user side does not have the ability to store electricity. The historical electricity consumption data is used to represent the independent electricity consumption situation of the historical user side. The electricity consumption scenario library stores the electricity consumption scenarios formed when different user sides work simultaneously. The electricity consumption scenario is used to represent the combined electricity consumption situation of different user sides when using electricity simultaneously;

[0097] The real-time data acquisition module is configured to: collect the real-time electricity consumption data of the user side, and when the user side is a storage-type user side, collect its corresponding real-time electricity storage data; among them, the real-time electricity consumption data is used to represent the independent electricity consumption situation of the real-time user side, and the real-time electricity storage data is used to represent the electricity storage situation of the real-time storage-type user side;

[0098] The real-time combined data generation module is configured to: use the electricity consumption scenario library to match the electricity consumption scenario corresponding to the real-time electricity consumption data, and generate real-time combined electricity consumption data based on the electricity consumption scenario; among them, the real-time combined electricity consumption data is used to represent the combined electricity consumption situation of different user sides when using electricity simultaneously in real time;

[0099] The first prediction module is configured to: use a preset power load prediction algorithm, and based on the real-time combined electricity consumption data and its corresponding electricity consumption scenario, generate a first prediction result; among them, the power load prediction algorithm is used to perform power load prediction, and the first prediction result is the result of rough power prediction;

[0100] The second prediction module is configured to: generate real-time electricity consumption prediction data by using real-time electricity storage data and real-time combined electricity consumption data, and generate a second prediction result by using the real-time electricity consumption prediction data, the first prediction result, and an electric power load prediction algorithm; wherein, the real-time electricity consumption prediction data is used to represent the combined electricity consumption situation of different user terminals in real time and predicted, considering the change of real-time electricity storage data, and the second prediction result is the result of refined electric power prediction adjusted after considering the real-time electricity storage data.

[0101] It should be noted that the user terminal electric power load prediction and adjustment system using big data in this embodiment corresponds to the aforementioned user terminal electric power load prediction and adjustment method using big data. Therefore, the content not specifically described in the user terminal electric power load prediction and adjustment system using big data in this embodiment, such as but not limited to function definition, working principle, and technical effect, etc., can refer to the description of the aforementioned user terminal electric power load prediction and adjustment method using big data, and will not be elaborated herein.

[0102] In summary, the user terminal electric power load prediction and adjustment method and system using big data in this embodiment construct an electricity consumption scenario library by using the collected historical electricity consumption data of the user terminal, and combine real-time electricity consumption data and real-time electricity storage data to achieve a comprehensive analysis and accurate prediction and adjustment of the user terminal electric power load; by distinguishing between electricity storage type and non-electricity storage type user terminals, fully considering the combined electricity consumption situation of different types of users in different electricity consumption scenarios, the introduction of real-time electricity storage data reflects the real-time change of the electricity storage state of the electricity storage type user terminal in the process of generating real-time electricity consumption prediction data and the second prediction result, effectively compensating for the deficiency of the traditional method that does not consider the electricity storage factor, significantly improving the accuracy of electric power load prediction in a complex electricity consumption environment, and providing strong support for the stable operation of the power grid and the reasonable allocation of resources.

[0103] In the embodiments provided in the present application, it should be understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, code, or any appropriate combination thereof. For hardware implementation, the processor can be implemented in one or more of the following units: application specific integrated circuit (ASIC), digital signal processor (DSP), digital signal processing device (DSPD), programmable logic device (PLD), field programmable gate array (FPGA), processor, controller, microcontroller, microprocessor, or other electronic units designed to implement the functions described herein, or a combination thereof. For software implementation, part or all of the processes of the embodiments can be completed by instructing the relevant hardware through a computer program. When implemented, the above program can be stored in a computer-readable storage medium or transmitted as one or more instructions or codes on a computer-readable storage medium. The computer-readable storage medium includes computer storage media and communication media, where the communication media includes any medium that facilitates the transfer of a computer program from one place to another. The storage media can be any available medium accessible by a computer. The computer-readable storage medium can include, but is not limited to, RAM, ROM, EEPROM, CD-ROM, or other optical disk storage, magnetic disk storage media, or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer.

[0104] Finally, it should be noted that the above are only the preferred embodiments of the present application and are not intended to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for predicting and adjusting user-side power load using big data, characterized in that: The method includes: A power usage scenario library is constructed based on the collected power usage history data of user terminals; wherein the user terminals include power storage user terminals and non-power storage user terminals, the power storage user terminals have power storage capabilities, and the non-power storage user terminals do not have power storage capabilities. The power usage history data is used to represent the historical independent power usage of the user terminals. The power usage scenario library stores power usage scenarios formed when different user terminals operate simultaneously, and the power usage scenarios are used to represent the combined power usage of different user terminals at the same time. Collecting real-time electricity consumption data of the user terminal, and when the user terminal is the electricity storage type user terminal, collecting its corresponding real-time electricity storage data; wherein the real-time electricity consumption data is used to represent the real-time independent electricity consumption of the user terminal, and the real-time electricity storage data is used to represent the real-time electricity storage status of the electricity storage type user terminal; Matching the power usage scenarios corresponding to the real-time power usage data using the power usage scenario library, and generating real-time power usage combination data based on the power usage scenarios; wherein the real-time power usage combination data is used to represent the real-time combined power usage of different user terminals at the same time; Using a preset power load forecasting algorithm, and based on the real-time combined power consumption data and the corresponding power consumption scenario, generating a first forecasting result; wherein the power load forecasting algorithm is used to perform power load forecasting, and the first forecasting result is a result of extensive power forecasting; Generating the first prediction result includes: The real-time combined power consumption data is input into the power load forecasting algorithm, an initial forecast result having a time series is output, and the initial forecast result is corrected using the power consumption scenario to obtain the first forecast result; wherein, the correction of the initial forecast result using the power consumption scenario is: The first prediction result is obtained by using the initial prediction result correction formula, and the initial prediction result correction formula is: in, for The initial prediction value at time t, for A correction factor based on the electricity usage scenario at the time, the correction factor is used to characterize the degree of influence of the electricity usage scenario on the real-time electricity usage data, For the calculated a first prediction value at a time instant, wherein the first prediction result is formed by the first prediction value based on a time series; generating real-time power consumption forecast data using the real-time power storage data and the real-time power consumption combined data, and generating a second forecast result using the real-time power consumption forecast data and the first forecast result; wherein the real-time power consumption forecast data is used to represent the real-time, forecasted combined power consumption of different user terminals, taking into account the changes in the real-time power storage data, and the second forecast result is the result of a refined power forecast adjusted after taking into account the real-time power storage data; The generation of the real-time electricity consumption prediction data includes: The real-time electricity consumption forecast data calculation formula is used to calculate the real-time electricity consumption forecast data. The real-time electricity consumption forecast data calculation formula is: in, Indicates Moment User Terminal Real-time data on electricity consumption, express Moment User Terminal The total number of user terminals in the corresponding real-time combined electricity consumption data and , Indicates The predicted adjustment value of the electricity storage user terminal based on the real-time data of electricity storage is obtained at all times, and the predicted adjustment value is obtained based on the electricity usage scenario. For the calculated The real-time electricity consumption prediction value at the time, wherein the real-time electricity consumption prediction data is composed of the real-time electricity consumption prediction value based on a time series; Generating the second prediction result includes: The second prediction result is calculated using the second prediction result calculation formula, which is: in, The first prediction result corresponds to The first predicted value at time t, The preset forecast period The confidence level of the real-time electricity consumption forecast data obtained based on the error rate generated is obtained based on a preset confidence level query table. The confidence level query table stores the confidence levels of the real-time electricity consumption forecast data corresponding to different error rates, and the preset forecast period ,in, is the starting time corresponding to the prediction period, is the end time corresponding to the prediction period, The real-time electricity consumption forecast data, For the calculated The second prediction value at the time instant, the second prediction result is composed of the second prediction value based on the time series.

2. The method for predicting and adjusting user-side power load using big data according to claim 1, characterized in that: The construction of the electricity usage scenario library includes: The collected historical electricity usage data is cleaned, and a first clustering is performed on the cleaned data based on a preset electricity usage pattern. A second clustering is performed after adding a description of the electricity usage situation to the result of the first clustering to obtain an electricity usage scenario. A database is built using the electricity usage scenario to obtain an electricity usage scenario database; wherein, the electricity usage pattern at least includes the type, number and power range of the user terminals under the mode, and the electricity usage situation description at least includes the electricity usage time distribution characteristics and charge and discharge characteristics of the user terminals under the electricity usage situation. The electricity usage time distribution characteristics are used to characterize the electricity usage time distribution situation of different user terminals, and the charge and discharge characteristics are used to characterize the charge and discharge situation of the energy storage user terminals.

3. The method for predicting and adjusting user-side power load using big data according to claim 1, characterized in that: The power storage user terminal at least includes energy storage equipment in the power grid and new energy equipment consisting of new energy vehicles owned by the user.

4. The method for predicting and adjusting user-side power load using big data according to claim 3, characterized in that: The collection of real-time data of electricity storage includes: The real-time power storage data of the energy storage device and the new energy device are collected using the power collection devices preset in the energy storage device and the new energy device.

5. The method for predicting and adjusting user-side power load using big data according to claim 1, characterized in that: The matching of the electricity usage scenario includes: extracting real-time pattern features and real-time scenario features of the real-time electricity usage data, calculating the similarity between the real-time electricity usage data and the electricity usage scenario based on the real-time pattern features and the real-time scenario features, and selecting the electricity usage scenario with the highest similarity as a matching result; wherein the real-time pattern features include at least features corresponding to the type, number, and power range of the real-time user terminals, and the real-time scenario features include at least features corresponding to the real-time electricity usage time distribution and charging and discharging characteristics of the real-time user terminals; The extraction of the electricity consumption time distribution characteristics includes: Based on the temporal distribution of electricity consumption behavior of different user terminals within a preset prediction period, the power consumption concentration and power consumption pattern at different user terminals at different times are extracted to obtain the power consumption time distribution characteristics; The extraction of the charge and discharge characteristics includes: Based on the charging and discharging behavior of the energy storage user terminal, the impact of the charging and discharging behavior on the overall power load and the impact of the power load forecast adjustment are extracted to obtain the charging and discharging characteristics; wherein, the charging and discharging behavior at least includes charging and discharging power, charging and discharging capacity, charging and discharging efficiency and charging and discharging time pattern.

6. The method for predicting and adjusting user-side power load using big data according to claim 5, characterized in that: The generation of the real-time combined electricity consumption data includes: According to the matched electricity usage scenario, the user terminals involved are determined to obtain the corresponding real-time electricity usage data, and the obtained real-time electricity usage data is used to construct a real-time electricity usage combination data matrix. The matrix elements of the real-time electricity usage combination data matrix are the real-time mode characteristics of each user terminal. The real-time electricity usage combination data matrix is defined as the real-time electricity usage combination data and output.

7. A user-side power load forecasting and adjustment system using big data, the system being applicable to the user-side power load forecasting and adjustment method using big data according to any one of claims 1 to 6, characterized in that: The system includes a power usage scenario library module, a real-time data acquisition module, a real-time combined data generation module, a first prediction module and a second prediction module that are communicatively connected; The power usage scenario library module is configured to construct a power usage scenario library based on the collected power usage history data of the user terminals; wherein the user terminals include power storage user terminals and non-power storage user terminals, the power storage user terminals have power storage capabilities, and the non-power storage user terminals do not have power storage capabilities, the power usage history data is used to represent the historical independent power usage of the user terminals, and the power usage scenario library stores power usage scenarios formed when different user terminals operate simultaneously, and the power usage scenarios are used to represent the combined power usage of different user terminals at the same time; The real-time data acquisition module is configured to: collect the real-time power consumption data of the user terminal, and when the user terminal is the power storage user terminal, collect the corresponding real-time power storage data; wherein the real-time power consumption data is used to represent the real-time independent power consumption of the user terminal, and the real-time power storage data is used to represent the real-time power storage status of the power storage user terminal; The real-time combined data generation module is configured to: use the power usage scenario library to match the power usage scenario corresponding to the real-time power usage data, and generate real-time power usage combined data based on the power usage scenario; wherein the real-time power usage combined data is used to represent the real-time combined power usage of different user terminals at the same time; The first prediction module is configured to generate a first prediction result using a preset power load prediction algorithm based on the real-time combined power consumption data and the corresponding power consumption scenario; wherein the power load prediction algorithm is used to perform power load prediction, and the first prediction result is a result of extensive power prediction; Generating the first prediction result includes: The real-time combined power consumption data is input into the power load forecasting algorithm, an initial forecast result having a time series is output, and the initial forecast result is corrected using the power consumption scenario to obtain the first forecast result; wherein, the correction of the initial forecast result using the power consumption scenario is: The first prediction result is obtained by using the initial prediction result correction formula, and the initial prediction result correction formula is: in, for The initial prediction value at time t, for A correction factor based on the electricity usage scenario at the time, the correction factor is used to characterize the degree of influence of the electricity usage scenario on the real-time electricity usage data, For the calculated a first prediction value at a time instant, wherein the first prediction result is formed by the first prediction value based on a time series; The second prediction module is configured to generate real-time power consumption prediction data using the real-time power storage data and the real-time power consumption combined data, and generate a second prediction result using the real-time power consumption prediction data and the first prediction result; wherein the real-time power consumption prediction data is used to represent the real-time, predicted combined power consumption of different user terminals, taking into account the changes in the real-time power storage data, and the second prediction result is the result of a refined power prediction adjusted after taking into account the real-time power storage data; The generation of the real-time electricity consumption prediction data includes: The real-time electricity consumption forecast data calculation formula is used to calculate the real-time electricity consumption forecast data. The real-time electricity consumption forecast data calculation formula is: in, Indicates Moment User Terminal Real-time data on electricity consumption, express Moment User Terminal The total number of user terminals in the corresponding real-time combined electricity consumption data and , Indicates The predicted adjustment value of the electricity storage user terminal based on the real-time data of electricity storage is obtained at all times, and the predicted adjustment value is obtained based on the electricity usage scenario. For the calculated The real-time electricity consumption prediction value at the time, wherein the real-time electricity consumption prediction data is composed of the real-time electricity consumption prediction value based on a time series; Generating the second prediction result includes: The second prediction result is calculated using the second prediction result calculation formula, which is: in, The first prediction result corresponds to The first predicted value at time t, The preset forecast period The confidence level of the real-time electricity consumption forecast data obtained based on the error rate generated is obtained based on a preset confidence level query table. The confidence level query table stores the confidence levels of the real-time electricity consumption forecast data corresponding to different error rates, and the preset forecast period ,in, is the starting time corresponding to the prediction period, is the end time corresponding to the prediction period, The real-time electricity consumption forecast data, For the calculated The second prediction value at the time instant, the second prediction result is composed of the second prediction value based on the time series.

Images