Power supply prediction method, system, device and storage medium based on multi-source data

Abstract

The present invention relates to the technical field of power supply prediction, and in particular to a power supply prediction method, system, device and storage medium based on multi-source data, which can improve the accuracy and reliability of power supply prediction; the method comprises: based on the historical power supply data records of each power supply unit in the area to be predicted, generating a power supply historical time series characterization matrix for the area to be predicted; the power supply historical time series characterization matrix includes the power consumption of all power consumption categories in each power supply unit; obtaining the power supply environment impact characteristics corresponding to the area where each power supply unit is located within the time period to be predicted; for the power supply environment impact characteristics of each power supply unit, in the power supply environment correlation coefficient matrix obtained by pre-analysis, mapping and extracting the environment correlation coefficient set corresponding to the power supply unit; the environment correlation coefficient set includes the correlation coefficient between the power supply environment impact characteristics and the power consumption of each power consumption category in the power supply unit.

Description

Technical Field

[0001] The present invention relates to the technical field of power supply prediction, and in particular to a power supply prediction method, system, device and storage medium based on multi-source data. Background Art

[0002] In modern power systems, with the development of the economy and the growth of population, electricity demand has become complex and changeable, which has put forward higher requirements on power supply forecasting methods. Accurate power supply forecasting is crucial for the planning, scheduling and operation of power systems. The power supply forecast not only affects the balance of supply and demand of electricity, but is also directly related to the stability, economy and reliability of the power grid.

[0003] Existing power supply forecasting methods mainly include time series analysis, regression analysis, machine learning algorithms, etc.; however, these methods usually only consider the time series characteristics of the historical electricity consumption data itself, while ignoring other external factors that may affect the power supply. They lack the ability to predict future power supply changes in advance and are difficult to adapt to the rapidly changing power market environment. Summary of the invention

[0004] To solve the above technical problems, the present invention provides a power supply prediction method, system, device and storage medium based on multi-source data, which can improve the accuracy and reliability of power supply prediction and enhance the stability, economy and reliability of the power system.

[0005] In a first aspect, the present invention provides a power supply prediction method based on multi-source data analysis, the method comprising:

[0006] Based on the historical power supply data records of each power supply unit in the area to be predicted, a power supply historical time series characterization matrix of the area to be predicted is generated; the power supply historical time series characterization matrix includes the power consumption of all power consumption categories in each power supply unit;

[0007] Obtain the power supply environment impact characteristics corresponding to the area where each power supply unit is located within the predicted time period;

[0008] According to the power supply environment impact characteristics of each power supply unit, in the power supply environment correlation coefficient matrix obtained by pre-analysis, a set of environment correlation coefficients corresponding to the power supply unit is mapped and extracted; the set of environment correlation coefficients includes the correlation coefficients between the power supply environment impact characteristics and the power consumption of each power consumption category in the power supply unit;

[0009] Based on the environment-related coefficient set corresponding to each power supply unit, the power supply historical time series representation matrix is ​​corrected for environment-related factors to obtain the power supply prediction time series representation matrix;

[0010] The power supply forecast time series characterization matrix is ​​statistically calculated to obtain the power supply of the area to be predicted within the time period to be predicted.

[0011] Furthermore, based on the historical power supply data records of each power supply unit in the area to be predicted, a method for generating a historical time series characterization matrix of power supply quantity in the area to be predicted includes:

[0012] Collect historical power supply data records of each power supply unit in the area to be predicted;

[0013] Based on a preset time segmentation rule, the historical power supply data records of each power supply unit are divided to obtain a number of historical power supply analysis windows; the time segmentation rule is to take the length of the time period to be predicted as the analysis period, and divide it into a number of time windows according to the length of the analysis period, and the time window is the same as the length corresponding to the historical power supply analysis window;

[0014] For each historical power supply analysis window, extract the power supply characteristics of each power supply unit respectively, wherein the power supply characteristics include the power consumption of each power consumption category;

[0015] The power supply characteristics of each power supply unit corresponding to several historical power supply analysis windows are serialized and adjusted in chronological order, and the power supply characteristics of the same power supply unit are aligned to obtain the historical time series representation matrix of the power supply in the area to be predicted.

[0016] Furthermore, in the power supply historical time series characterization matrix, the same row represents the power supply characteristics of each power supply unit in the same historical power supply analysis window, and the same column represents the power supply characteristics of the same power supply unit in different historical power supply analysis windows; the power supply historical time series characterization matrix is:

[0017] in, N Indicates the number of power supply units, T represents the number of time windows, C Indicates the number of electricity consumption categories, Indicates the power supply unit N In the time window T The electricity consumption of the internal electricity category C.

[0018] Furthermore, the method for obtaining the environment correlation coefficient set includes:

[0019] Calculate the correlation coefficient between each power supply environment impact characteristic and the power consumption of each power consumption category;

[0020] The calculated correlation coefficients are constructed into a power supply environment correlation coefficient matrix according to the corresponding relationship between the power supply environment impact characteristics and the power consumption categories;

[0021] According to the power supply environment impact characteristics of the area where each power supply unit is located in the time period to be predicted, the corresponding row is found in the power supply environment correlation coefficient matrix, so as to obtain the environment correlation coefficient of the power supply unit;

[0022] For each power supply unit, the correlation coefficients between all corresponding environmental characteristics and power consumption categories are extracted to form an environmental correlation coefficient set.

[0023] Furthermore, the method for performing environment-related correction on the power supply historical time series characterization matrix includes:

[0024] Quantify the power supply environment impact characteristics and the corresponding environment correlation coefficient set of each power supply unit within the time period to be predicted;

[0025] Based on the quantified power supply environmental impact characteristics and the corresponding environmental correlation coefficient set, calculate the impact of each environmental characteristic on the power consumption of each power consumption category;

[0026] The calculated environmental characteristic impact is added to the power consumption of the corresponding power supply unit and power consumption category in the power supply history time series characterization matrix to achieve environment-related correction.

[0027] Furthermore, a method for performing statistical calculation on a power supply prediction time series characterization matrix includes:

[0028] Integrate the data in the power supply forecast time series characterization matrix;

[0029] Perform time series analysis on the integrated data to identify trends, seasonal patterns, cyclical and random fluctuations in the data;

[0030] The power supply of the entire area to be predicted is calculated by weighted summation of the prediction data of each power supply unit and power consumption category in the prediction time series representation matrix;

[0031] After obtaining the prediction results, the prediction results are evaluated for errors;

[0032] Output the verified prediction results.

[0033] Furthermore, the power supply environment influencing characteristics include meteorological conditions, socio-economic factors, user behavior, power market factors, power system status and geographic location information.

[0034] On the other hand, the present application also provides a power supply prediction system based on multi-source data analysis, the system comprising:

[0035] A data collection and preprocessing module generates a power supply history time series representation matrix for the area to be predicted based on the historical power supply data records of each power supply unit in the area to be predicted; the power supply history time series representation matrix includes the power consumption of all power consumption categories in each power supply unit;

[0036] The power supply environment feature extraction module obtains the power supply environment impact features corresponding to the area where each power supply unit is located within the predicted time period;

[0037] An environmental correlation coefficient mapping module maps and extracts a set of environmental correlation coefficients corresponding to each power supply unit from a power supply environment correlation coefficient matrix obtained by pre-analysis, with respect to the power supply environment impact characteristics of each power supply unit; the set of environmental correlation coefficients includes correlation coefficients between the power supply environment impact characteristics and the power consumption of each power consumption category in the power supply unit;

[0038] An environment-related correction module performs environment-related correction on the power supply historical time series representation matrix based on the environment-related coefficient set corresponding to each power supply unit to obtain a power supply prediction time series representation matrix;

[0039] The power supply prediction module performs statistical calculations on the power supply prediction time series representation matrix to obtain the power supply of the area to be predicted within the time period to be predicted.

[0040] In a third aspect, the present application provides an electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, and the computer program, when executed by the processor, implements the steps of any one of the above methods.

[0041] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps in any one of the above-mentioned methods when executed by a processor.

[0042] Compared with the prior art, the present invention has the following beneficial effects: the method not only takes into account the time series characteristics of historical power supply data, but also incorporates the characteristics of the power supply environment. The comprehensive analysis method can more comprehensively reflect the complexity and variability of power demand, thereby improving the accuracy and reliability of the prediction;

[0043] By introducing the power supply environment correlation coefficient matrix and making environment-related corrections to the power supply historical time series representation matrix based on it, the possible impact on future power supply can be considered in advance, thereby enhancing the ability to predict future power supply changes;

[0044] It can respond flexibly to the rapidly changing power market environment; by continuously updating the power supply environment correlation coefficient matrix and correction model, it can ensure that the prediction results are always close to the actual situation and maintain the accuracy and effectiveness of the prediction; accurate power supply forecasting helps the power system to better balance supply and demand, optimize resource allocation, and reduce the demand for spare capacity, thereby improving the stability, economy and reliability of the power grid;

[0045] This method provides a powerful decision support tool for planners, dispatchers and operators of power systems. Through accurate power supply forecasting, it is possible to better formulate power generation plans, adjust the grid structure and optimize operation strategies to improve the overall performance and benefits of the power system.

[0046] In summary, the power supply forecasting method based on multi-source data analysis has significant advantages and broad application prospects in modern power systems; it can not only improve the accuracy and reliability of power supply forecasting, but also enhance the stability, economy and reliability of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is a flow chart of the present invention;

[0048] Figure 2 is a flow chart of a method for generating a historical time series representation matrix of power supply in a region to be predicted;

[0049] Figure 3 It is a structural diagram of the power supply prediction system based on multi-source data analysis. DETAILED DESCRIPTION

[0050] In the description of this application, those skilled in the art should know that this application can be implemented as a method, an apparatus, an electronic device, and a computer-readable storage medium. Therefore, this application can be specifically implemented in the following forms: complete hardware, complete software (including firmware, resident software, microcode, etc.), a combination of hardware and software. In addition, in some embodiments, this application can also be implemented in the form of a computer program product in one or more computer-readable storage media, and the computer-readable storage medium contains computer program code.

[0051] The above-mentioned computer-readable storage medium may adopt any combination of one or more computer-readable storage media. Computer-readable storage media include: electrical, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or devices, or any combination of the above. More specific examples of computer-readable storage media include: portable computer disks, hard disks, random access memories, read-only memories, erasable programmable read-only memories, flash memories, optical fibers, optical disc read-only memories, optical storage devices, magnetic storage devices, or any combination of the above. In the present application, computer-readable storage media can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device, or device.

[0052] The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws.

[0053] The present application describes the provided methods, devices, and electronic devices through flowcharts and / or block diagrams.

[0054] It should be understood that each box in the flowchart and / or block diagram and the combination of boxes in the flowchart and / or block diagram can be implemented by computer-readable program instructions. These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device to produce a machine, and these computer-readable program instructions are executed by a computer or other programmable data processing device to produce a device that implements the functions / operations specified by the boxes in the flowchart and / or block diagram.

[0055] These computer-readable program instructions may also be stored in a computer-readable storage medium that enables a computer or other programmable data processing device to work in a specific manner. In this way, the instructions stored in the computer-readable storage medium produce an instruction device product including functions / operations specified in the blocks in the flowchart and / or block diagram.

[0056] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device, so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby enabling the instructions executed on the computer or other programmable data processing apparatus to provide a process for implementing the functions / operations specified in the blocks in the flowchart and / or block diagram.

[0057] The present application is described below in conjunction with the drawings in the present application.

[0058] Embodiment 1: Figure 1 to Figure 2 As shown, the power supply prediction method based on multi-source data analysis of the present invention specifically includes the following steps:

[0059] S1. Based on the historical power supply data records of each power supply unit in the area to be predicted, generate a power supply historical time series characterization matrix for the area to be predicted; the power supply historical time series characterization matrix includes the power consumption of all power consumption categories in each power supply unit;

[0060] A method for generating a historical time series representation matrix of power supply quantity in a to-be-predicted area based on historical power supply data records of each power supply unit in the to-be-predicted area includes:

[0061] S11. Collect historical power supply data records of each power supply unit in the forecast area; these data come from the database and monitoring system of the power company, including daily, weekly, monthly and annual power consumption data; the data covers multiple time periods so as to capture the seasonality, cyclicality and long-term trends of power demand; clean the collected raw data, including removing duplicate records, filling missing values, identifying and correcting outliers; the cleaned data should be complete, accurate and consistent;

[0062] S12, determining an analysis period according to the length of the time period to be predicted, dividing the historical power supply data records according to the analysis period, and each period constitutes a historical power supply analysis window;

[0063] S13. For each historical power supply analysis window, extract power supply characteristics from the historical power supply data records of each power supply unit; these characteristics include the total power consumption, average power consumption, peak power consumption and power consumption volatility of each power consumption category; for each power supply unit, in each historical power supply analysis window, calculate the total power consumption of different power consumption categories and record it;

[0064] S14, serialize and adjust the power supply characteristics of each power supply unit in each historical power supply analysis window in chronological order to ensure the continuity and consistency of the time series data; at the same time, align the power supply characteristics of the same power supply unit and put them into the same time series matrix to facilitate subsequent data analysis and model training, and obtain the historical time series representation matrix of the power supply of the area to be predicted;

[0065] In the power supply historical time series characterization matrix, the same row represents the power supply characteristics of each power supply unit in the same historical power supply analysis window, and the same column represents the power supply characteristics of the same power supply unit in different historical power supply analysis windows; the power supply historical time series characterization matrix is:

[0066] in, N Indicates the number of power supply units, T represents the number of time windows, C Indicates the number of electricity consumption categories, Indicates the power supply unit NIn the time window T The electricity consumption of the internal electricity category C.

[0067] In this step, by collecting and processing the historical power supply data records of each power supply unit, including power consumption data of different time periods, this method can fully capture the seasonality, periodicity and long-term trends of power demand; it helps to more accurately predict future changes in power supply and improve prediction accuracy; the data cleaning process ensures that the data used to generate the time series characterization matrix is ​​complete, accurate and consistent; high-quality data is the basis for accurate prediction and can reduce prediction errors caused by data problems; in each historical power supply analysis window, not only the total power consumption of each power consumption category is extracted, but also multi-dimensional features such as average power consumption, peak power consumption and power consumption volatility are calculated; these features provide rich information, It helps to have a deeper understanding of the changing rules of power demand and provide powerful input for subsequent prediction models; the power supply characteristics of each power supply unit are serialized and adjusted in chronological order, and the power supply characteristics of the same power supply unit are aligned to form a power supply historical time series representation matrix; this matrix-form data structure is convenient for data analysis, feature engineering and model training; it enables the prediction model to better capture patterns and trends in time series data, and improve the generalization ability and prediction accuracy of the prediction model; the method in step S1 generates a high-quality power supply historical time series representation matrix by comprehensively collecting, processing and analyzing historical power supply data records, providing strong data support for subsequent power supply prediction.

[0068] S2. Obtain the power supply environment impact characteristics corresponding to the area where each power supply unit is located within the time period to be predicted;

[0069] In step S2, factors that may affect the power supply need to be collected and sorted out; these factors are usually multi-source heterogeneous data, and they have different degrees of influence on the change of power supply; the power supply environment influencing characteristics include:

[0070] Meteorological conditions: Temperature changes significantly affect electricity demand. In extreme weather conditions, the use of air conditioning and heating equipment will increase or decrease significantly. High humidity will cause people to feel hotter, which will increase the use of cooling equipment. For wind farms, wind speed directly affects power generation. Rainfall will affect people's outdoor activities, which will indirectly affect electricity demand during certain periods. For photovoltaic power generation systems, the intensity of sunlight determines the power generation efficiency.

[0071] Socioeconomic factors: Electricity usage patterns vary on different types of days; sporting events, concerts, or other large events can increase electricity demand in specific areas; periods of economic growth or recession have a greater impact on industrial and commercial electricity usage;

[0072] User behavior: Different types of users have different electricity usage patterns; the extent to which users adopt energy-saving measures; how users respond to dynamic electricity prices;

[0073] Power market factors: Changes in electricity prices directly affect users' electricity consumption behavior; the balance of supply and demand in the power market, including spare capacity and power generation costs;

[0074] Power system status: real-time power generation of different types of power plants; power flow on transmission lines; the impact of equipment failures and planned maintenance on power supply;

[0075] Geographical location information: the impact of climate conditions and geographical characteristics of different geographical locations on electricity demand; the differences in electricity consumption patterns between densely populated areas and sparsely populated areas.

[0076] In this step, by considering diversified and comprehensive influencing factors such as meteorological conditions, socio-economic factors, user behavior, power market factors, power system status and geographic location information, the power supply forecasting model can more comprehensively capture the laws and trends of power demand changes, thereby significantly improving the accuracy and reliability of the forecast; in the face of the rapidly changing power market environment, this step can flexibly adjust the influencing factors included in the forecasting model to adapt to new market conditions and user needs; it helps to maintain the effectiveness and practicality of the forecasting model; by accurately predicting the power supply, the power system can reasonably plan, dispatch and operate to achieve optimal allocation of power resources; it helps to reduce power waste, improve energy utilization efficiency, and promote the sustainable development of the power industry; the accuracy of power supply forecasting is directly related to the stability of the power grid; by considering multiple influencing factors, the forecasting model can identify potential supply and demand imbalance risks in advance, provide timely warnings and response measures for the power system, thereby ensuring the safe and stable operation of the power grid; step S2 not only improves the accuracy and adaptability of power supply forecasting by comprehensively collecting and sorting out the influencing characteristics of the power supply environment, but also promotes the optimal allocation of power resources, improves system stability and enhances decision-making support capabilities.

[0077] S3. For the power supply environment impact characteristics of each power supply unit, in the power supply environment correlation coefficient matrix obtained by pre-analysis, map and extract the environment correlation coefficient set corresponding to the power supply unit; the environment correlation coefficient set includes the correlation coefficient between the power supply environment impact characteristics and the power consumption of each power consumption category in the power supply unit;

[0078] Using the power supply environment impact characteristics, mapping and searching are performed in the power supply environment correlation coefficient matrix; for each power supply unit and each power supply environment impact characteristic, the corresponding rows and columns are found in the matrix, so as to extract the correlation coefficient between the characteristic and the power consumption of each power consumption category in the power supply unit; these correlation coefficients constitute an environmental correlation coefficient set, which reflects the potential trend of the power consumption change of each power consumption category under the current power supply environment impact characteristics;

[0079] For each power supply unit, multiple sets of environment-related coefficients corresponding to power supply environment impact characteristics are obtained; these sets are integrated to form a comprehensive coefficient set that can fully reflect the power supply environment impact;

[0080] The method for obtaining the environment correlation coefficient set includes:

[0081] Calculate the correlation coefficient between each power supply environment impact feature and the power consumption of each power consumption category, and these correlation coefficients will constitute the elements of the power supply environment correlation coefficient matrix;

[0082] The calculated correlation coefficients are constructed into a power supply environment correlation coefficient matrix according to the corresponding relationship between the power supply environment impact characteristics and the power consumption categories; each row of the matrix represents a power supply environment impact characteristic, each column represents a power consumption category, and the elements in the matrix are the corresponding correlation coefficients;

[0083] According to the power supply environment impact characteristics of the area where each power supply unit is located in the time period to be predicted, the corresponding row is searched in the power supply environment correlation coefficient matrix, so as to obtain the environment correlation coefficient set of the power supply unit;

[0084] For each power supply unit, the correlation coefficients between all corresponding environmental characteristics and power consumption categories are extracted to form an environmental correlation coefficient set.

[0085] In this step, by considering the correlation between the power supply environment impact characteristics and the power consumption of the power consumption category, and making environmental correlation corrections accordingly, the impact of external environmental changes on power supply can be more accurately reflected; the limitation of traditional prediction methods that only rely on the time series characteristics of historical power consumption data is overcome, making the prediction results closer to the actual situation and improving the prediction accuracy; modern power systems face a rapidly changing power market environment, including fluctuations in various external factors such as weather, economy, and policy; by constructing a power supply environment correlation coefficient matrix and dynamically adjusting the environmental correlation coefficient set, the prediction model can flexibly respond to changes in the external environment, enhancing the adaptability of the prediction model This step improves the prediction accuracy and robustness, provides a more reliable reference for power system decision makers, helps optimize power resource allocation, balance supply and demand, and improve grid stability and economy; this step achieves refined management of power supply forecasts by distinguishing different power consumption categories and power supply units; helps the power system to more accurately grasp the power consumption characteristics of different regions and different user groups, so as to formulate more targeted management strategies and service plans; S3 step achieves precise and dynamic adjustment of power supply forecasts by constructing and applying the power supply environment correlation coefficient matrix, which can improve the operation efficiency of the power system and ensure the safety of power supply.

[0086] S4, based on the environment-related coefficient set corresponding to each power supply unit, perform environment-related correction on the power supply historical time series representation matrix to obtain the power supply prediction time series representation matrix;

[0087] A method for performing environment-related correction on a power supply history time series characterization matrix includes:

[0088] Quantify the power supply environment impact characteristics and the corresponding environment correlation coefficient set of each power supply unit in the time period to be predicted, and convert them into comparable values;

[0089] According to the quantified power supply environment impact characteristics and the corresponding environmental correlation coefficient set, calculate the impact of each environmental characteristic on the power consumption of each power consumption category; multiply the quantified value of the environmental characteristic by the corresponding correlation coefficient to obtain the specific impact of the environmental characteristic on the power consumption of each power consumption category;

[0090] The calculated environmental characteristic impact is added to the power consumption of the corresponding power supply unit and power consumption category in the power supply history time series representation matrix to achieve environmental related correction;

[0091] After the environment-related correction, part of the data in the power supply historical time series characterization matrix is ​​updated to reflect the actual impact of external environmental factors on the power supply. The updated matrix is ​​the power supply prediction time series characterization matrix.

[0092] More specifically, the method for quantifying the impact characteristics of the power supply environment is as follows:

[0093] Meteorological conditions: For example, temperature changes can be expressed directly by using the Celsius value recorded by the thermometer, and humidity can also be expressed as a percentage;

[0094] Socioeconomic factors: such as holidays, which can be represented by binary variables (1 for holidays and 0 for not);

[0095] User behavior: For example, the degree of adoption of energy-saving measures, which can be expressed as a percentage to indicate the proportion of users taking energy-saving measures;

[0096] Electricity market factors: such as electricity price changes, which can be directly expressed in monetary units (such as RMB yuan / kWh);

[0097] Power system status: such as power generation, which can be expressed in kilowatt-hours (kWh);

[0098] Geographic location information: such as population density, which can be expressed as the number of people per square kilometer.

[0099] In order to ensure that the above features of different dimensions can be compared on the same scale, the numerical features need to be normalized. Normalization methods include MinMax Scaling and Z-Score normalization.

[0100] Through the above quantification process, the non-numerical power supply environment impact characteristics can be converted into numerical data to facilitate mathematical calculations and model analysis. The quantified data can be further used to construct an environmental correlation coefficient matrix and perform environmental correlation corrections, thereby improving the accuracy and reliability of power supply prediction.

[0101] In this step, by introducing the power supply environment impact characteristics and their correlation coefficients with power consumption, the impact of external factors on power supply can be considered more comprehensively, thereby improving the accuracy of the prediction results; it helps the power system to grasp future power demand more accurately and provide strong support for planning, scheduling and operation; traditional power supply prediction methods often only rely on the time series characteristics of historical power consumption data, which is difficult to adapt to the rapidly changing power market environment; and the environment-related correction method can flexibly respond to changes in various external factors, enhancing the adaptability and robustness of the prediction model; accurate power supply prediction helps power system managers to better formulate power generation plans, optimize resource allocation and reduce energy waste; the environment-related correction method provides decision makers with a more reliable basis by providing more accurate prediction results, improving decision-making efficiency and scientificity; the stability of power supply is an important guarantee for the safe and stable operation of the power system; through environment-related correction, future power supply demand can be predicted more accurately, so that measures can be taken in advance to deal with possible supply and demand imbalances, which helps to maintain the stability and reliability of the power system.

[0102] S5. Statistically calculate the power supply prediction time series characterization matrix to obtain the power supply of the area to be predicted within the time period to be predicted;

[0103] In step S5, the power supply forecast time series representation matrix after environment-related correction is converted into a specific and quantifiable power supply forecast value, providing a direct basis for the planning, scheduling and operation of the power system;

[0104] The method for performing statistical calculation on a power supply prediction time series characterization matrix comprises:

[0105] Integrate the data in the power supply forecast time series representation matrix; the matrix contains the predicted power consumption of multiple power supply units and different power consumption categories at multiple time points; in order to obtain the power supply forecast value of the entire area to be predicted, it is necessary to summarize these data according to the regional or time dimension;

[0106] Perform time series analysis on aggregated data to identify trends, seasonal patterns, periodicity, and random fluctuations in the data; through time series analysis, it is possible to predict the power supply at a certain point in the future and evaluate the confidence interval of the prediction;

[0107] The power supply of the entire area to be predicted is calculated by weighted summation of the prediction data of each power supply unit and power consumption category in the prediction time series representation matrix;

[0108] After obtaining the forecast results, the forecast results are evaluated for errors; this is achieved by comparing with historical data and calculating the forecast error;

[0109] The verified forecast results are output; the output results include the forecast value, forecast time range, and related instructions and suggestions for reference and decision-making by power system managers.

[0110] In this step, by integrating the data in the power supply forecast time series characterization matrix and summarizing them according to the region or time dimension, the inconsistency and redundancy problems between multi-source data are effectively solved, ensuring the comprehensiveness and accuracy of the forecast results; time series analysis is performed on the aggregated data to identify trends, seasonal patterns, periodicity, random fluctuations and other characteristics in the data; this not only improves the adaptability and robustness of the forecast model, but also provides more refined support for future power supply forecasts; the power supply of the entire area to be predicted is calculated by weighted summation of the forecast data of each power supply unit and power consumption category in the forecast time series characterization matrix; full consideration is given to the role of different power supply units and power consumption categories in the power system. The actual contribution and importance ensure the rationality and fairness of the prediction results; after obtaining the prediction results, the error assessment is carried out by comparing with historical data and calculating the prediction error; it not only helps to verify the accuracy and reliability of the prediction results, but also provides valuable reference for subsequent model optimization and improvement; the verified prediction results are output in a clear and unambiguous manner, including the prediction value, prediction time range and related instructions and suggestions; the output results not only provide intuitive and comprehensive information support for power system managers, but also provide a strong decision-making basis for the planning, scheduling and operation of the power system; the S5 step realizes the accurate prediction and evaluation of power supply by statistically calculating the power supply prediction time series characterization matrix.

[0111] Embodiment 2: Figure 3 As shown, the power supply prediction system based on multi-source data analysis of the present invention specifically includes the following modules:

[0112] A data collection and preprocessing module generates a power supply history time series representation matrix for the area to be predicted based on the historical power supply data records of each power supply unit in the area to be predicted; the power supply history time series representation matrix includes the power consumption of all power consumption categories in each power supply unit;

[0113] The power supply environment feature extraction module obtains the power supply environment impact features corresponding to the area where each power supply unit is located within the predicted time period;

[0114] An environmental correlation coefficient mapping module maps and extracts a set of environmental correlation coefficients corresponding to each power supply unit from a power supply environment correlation coefficient matrix obtained by pre-analysis, with respect to the power supply environment impact characteristics of each power supply unit; the set of environmental correlation coefficients includes correlation coefficients between the power supply environment impact characteristics and the power consumption of each power consumption category in the power supply unit;

[0115] An environment-related correction module performs environment-related correction on the power supply historical time series representation matrix based on the environment-related coefficient set corresponding to each power supply unit to obtain a power supply prediction time series representation matrix;

[0116] The power supply prediction module performs statistical calculations on the power supply prediction time series representation matrix to obtain the power supply of the area to be predicted within the time period to be predicted.

[0117] The system not only takes into account the time series characteristics of historical power supply data itself, but also integrates multiple power supply environment influencing features. The fusion of multi-source data enables the prediction model to more comprehensively capture various factors affecting power supply, thereby improving the accuracy and robustness of the prediction.

[0118] Through the environment-related coefficient mapping module and the environment-related correction module, the system can accurately calculate and apply the impact of external environmental factors on power supply for each power supply unit and power consumption category, making the prediction results closer to the actual situation and enhancing the prediction model's ability to predict future power supply changes in advance;

[0119] Since the system takes into account a variety of external factors and makes dynamic adjustments through the correlation coefficient matrix, it can better adapt to the rapidly changing power market environment. The system can maintain the accuracy and effectiveness of the prediction by updating the correlation coefficient and correcting the model;

[0120] Accurate power supply forecasting helps the power system to better plan, dispatch and operate; by predicting the changing trend of power demand in advance, the power system can reasonably arrange power generation plans, optimize resource allocation and reduce the demand for backup capacity, thereby improving the stability, economy and reliability of the power grid;

[0121] The system can continuously improve prediction accuracy and efficiency, providing strong support for the digital transformation and intelligent upgrading of the power industry;

[0122] In summary, by integrating multiple data sources and environmental factors, the system can more accurately predict power supply, which can not only help power companies better manage resources but also improve the overall efficiency and reliability of the power grid.

[0123] The various variations and specific embodiments of the power supply prediction method based on multi-source data analysis in the aforementioned embodiment 1 are also applicable to the power supply prediction system based on multi-source data analysis in this embodiment. Through the aforementioned detailed description of the power supply prediction method based on multi-source data analysis, those skilled in the art can clearly know the implementation method of the power supply prediction system based on multi-source data analysis in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here.

[0124] In addition, the present application also provides an electronic device, including a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor. The transceiver, the memory, and the processor are respectively connected via a bus. When the computer program is executed by the processor, each process of the above-mentioned method for controlling output data is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0125] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A power supply prediction method based on multi-source data analysis, characterized in that: The method comprises: Based on the historical power supply data records of each power supply unit in the area to be predicted, a power supply historical time series characterization matrix of the area to be predicted is generated; the power supply historical time series characterization matrix includes the power consumption of all power consumption categories in each power supply unit; Obtain the power supply environment impact characteristics corresponding to the area where each power supply unit is located within the predicted time period; According to the power supply environment impact characteristics of each power supply unit, in the power supply environment correlation coefficient matrix obtained by pre-analysis, a set of environment correlation coefficients corresponding to the power supply unit is mapped and extracted; the set of environment correlation coefficients includes the correlation coefficients between the power supply environment impact characteristics and the power consumption of each power consumption category in the power supply unit; Based on the environment-related coefficient set corresponding to each power supply unit, the power supply historical time series representation matrix is ​​corrected for environment-related factors to obtain the power supply prediction time series representation matrix; Perform statistical calculations on the power supply forecast time series characterization matrix to obtain the power supply of the area to be forecasted within the forecast time period; The method for obtaining the environment correlation coefficient set includes: Calculate the correlation coefficient between each power supply environment impact characteristic and the power consumption of each power consumption category; The calculated correlation coefficients are constructed into a power supply environment correlation coefficient matrix according to the corresponding relationship between the power supply environment impact characteristics and the power consumption categories; According to the power supply environment impact characteristics of the area where each power supply unit is located in the time period to be predicted, the corresponding row is found in the power supply environment correlation coefficient matrix, so as to obtain the environment correlation coefficient of the power supply unit; For each power supply unit, the correlation coefficients between all corresponding environmental characteristics and power consumption categories are extracted to form an environmental correlation coefficient set; A method for performing environment-related correction on a power supply history time series characterization matrix includes: Quantify the power supply environment impact characteristics and the corresponding environment correlation coefficient set of each power supply unit within the time period to be predicted; Based on the quantified power supply environmental impact characteristics and the corresponding environmental correlation coefficient set, calculate the impact of each environmental characteristic on the power consumption of each power consumption category; The calculated environmental characteristic impact is added to the power consumption of the corresponding power supply unit and power consumption category in the power supply history time series representation matrix to achieve environmental related correction; A method for generating a historical time series representation matrix of power supply quantity in a to-be-predicted area based on historical power supply data records of each power supply unit in the to-be-predicted area includes: Collect historical power supply data records of each power supply unit in the area to be predicted; Based on a preset time segmentation rule, the historical power supply data records of each power supply unit are divided to obtain a number of historical power supply analysis windows; the time segmentation rule is to take the length of the time period to be predicted as the analysis period, and divide it into a number of time windows according to the length of the analysis period, and the time window is the same as the length corresponding to the historical power supply analysis window; For each historical power supply analysis window, extract the power supply characteristics of each power supply unit respectively, wherein the power supply characteristics include the power consumption of each power consumption category; The power supply characteristics of each power supply unit corresponding to several historical power supply analysis windows are serialized and adjusted in chronological order, and the power supply characteristics of the same power supply unit are aligned to obtain a historical time series representation matrix of the power supply of the area to be predicted; In the power supply historical time series characterization matrix, the same row represents the power supply characteristics of each power supply unit in the same historical power supply analysis window, and the same column represents the power supply characteristics of the same power supply unit in different historical power supply analysis windows; the power supply historical time series characterization matrix is: ; in, N Indicates the number of power supply units, T represents the number of time windows, C Indicates the number of electricity consumption categories, U N,T,C Indicates the power supply unit N In the time window T The electricity consumption of internal electricity category C; The method for performing statistical calculation on a power supply prediction time series characterization matrix comprises: Integrate the data in the power supply forecast time series characterization matrix; Perform time series analysis on the integrated data to identify trends, seasonal patterns, cyclical and random fluctuations in the data; The power supply of the entire area to be predicted is calculated by weighted summation of the prediction data of each power supply unit and power consumption category in the prediction time series representation matrix; After obtaining the prediction results, the prediction results are evaluated for errors; Output the verified prediction results.

2. The power supply prediction method based on multi-source data analysis according to claim 1, characterized in that: The power supply environment influencing characteristics include meteorological conditions, socio-economic factors, user behavior, power market factors, power system status and geographic location information.

3. A power supply prediction system based on multi-source data analysis, characterized in that: The system is applied to the power supply prediction method based on multi-source data analysis as claimed in claim 1, and the system comprises: A data collection and preprocessing module generates a power supply history time series representation matrix for the area to be predicted based on the historical power supply data records of each power supply unit in the area to be predicted; the power supply history time series representation matrix includes the power consumption of all power consumption categories in each power supply unit; The power supply environment feature extraction module obtains the power supply environment impact features corresponding to the area where each power supply unit is located within the predicted time period; An environmental correlation coefficient mapping module maps and extracts a set of environmental correlation coefficients corresponding to each power supply unit from a power supply environment correlation coefficient matrix obtained by pre-analysis, with respect to the power supply environment impact characteristics of each power supply unit; the set of environmental correlation coefficients includes correlation coefficients between the power supply environment impact characteristics and the power consumption of each power consumption category in the power supply unit; An environment-related correction module performs environment-related correction on the power supply historical time series representation matrix based on the environment-related coefficient set corresponding to each power supply unit to obtain a power supply prediction time series representation matrix; The power supply prediction module performs statistical calculations on the power supply prediction time series representation matrix to obtain the power supply of the area to be predicted within the time period to be predicted.

4. An electronic device for power supply prediction based on multi-source data analysis, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, characterized in that: When the computer program is executed by the processor, the steps in the method according to any one of claims 1 to 2 are implemented.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps in the method according to any one of claims 1 to 2 are implemented.

Images