An intelligent management system for power systems

Abstract

The application discloses an intelligent power system management system, which comprises a power generation management module, a power distribution automation safety platform, an electric energy quality module, a dispatching management module, an energy saving management module, a power utilization service module, a power grid line inspection module and a weather forecast module. The application realizes intelligent auxiliary support for operation management decision in a more intelligent manner, improves the safety and stability of intelligent power grid system operation, and provides efficient and accurate weather forecast service, thereby guaranteeing the safe operation of the power grid line.

Description

Technical Field

[0001] This invention belongs to the field of power management, specifically relating to an intelligent management system for power systems. Background Technology

[0002] With the gradual development of smart power systems, highly flexible, data-driven power supply will gradually replace traditional static power supply. Through the mining and analysis of massive amounts of data, power production supply and demand management becomes more effective. A smart grid is a more intelligent, economical, and interactive power grid system compared to a traditional grid, leading the future development direction of the power grid system. The deployment of smart grids can effectively improve the level of power grid business services and achieve a high degree of integration of power flow, information flow, and business flow. With the development of smart grids, the automation level of the power grid system is constantly improving, and its various power, communication, and monitoring systems are becoming increasingly complex. The types and scale of application systems in all aspects of the power grid are growing exponentially, and the amount of various power grid information data is increasing geometrically. Faced with increasingly complex power grid information, power grid operation and management personnel urgently need a platform that can intuitively grasp the overall operating status of the power grid by sorting out key information from all aspects of the power grid, and provide intelligent auxiliary support for operation and management decisions.

[0003] Furthermore, with the increasing frequency of meteorological disasters, especially high temperatures, floods, and icing posing significant threats to power grid safety, numerical weather prediction technology is gradually being applied to the power industry. Extensive research and development of numerical weather prediction data has been conducted both domestically and internationally in the field of new energy. In the United States, companies such as Windlogic, 3Tier, and AWS TrueWind are developing weather forecasting systems based on real-time and historical meteorological observation data, as well as raw numerical weather data with coarse and fine grids. The German Weather Service provides regional and global numerical weather prediction data with resolutions of 7km×7km and 60km×60km, respectively. The French Meteorological Centre's regional forecast model has a spatial resolution of 9.5km×9.5km. Sweden, Denmark, Norway, Spain, Iceland, Ireland, and other countries have obtained numerical weather prediction products based on the HIRLAM model, with resolutions between 10-50km. The European Centre for Medium-Range Weather Forecasts (ECMWF) provides numerical weather predictions with a resolution of 40km×40km.

[0004] Correspondingly, with increasing demands, there is an urgent need in this field for a comprehensive smart power system that can provide multi-functional services to ensure the normal operation and maintenance of smart grids. In particular, the disaster prevention and mitigation capabilities of power grids rely not only on the grid's own adaptability but also, and more importantly, on accurate meteorological monitoring and forecast data. With the accelerated development of numerical weather prediction systems applied to new energy power plants, power big data has emerged. Consequently, the volume of data is constantly increasing, the frequency of data anomalies is rising, and data formats are becoming increasingly diverse. How to apply multi-dimensional, multi-source meteorological big data more flexibly and efficiently is also one of the challenging problems for those skilled in the art. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention proposes an intelligent management system for power systems, characterized in that the system includes at least:

[0006] The power generation management module is used to manage overall power production, specifically including the management of power supply reliability, overall voltage, overall line loss, substations, lines, monthly planned power outages, and monthly fault power outages;

[0007] A power distribution automation safety platform is used to execute safe power distribution schemes.

[0008] Power quality modules are used to protect power grid lines and transformers from overcurrent, overload, and temperature.

[0009] The dispatch management module is used to manage power dispatch.

[0010] The energy-saving management module is used to statistically analyze the energy consumption data of energy-consuming equipment and provide energy-saving solutions for the energy-consuming equipment based on electricity consumption and energy consumption data.

[0011] The electricity service module is used to provide electricity services to electricity users.

[0012] The power grid line inspection module is used for video monitoring and intelligent inspection of the power grid lines;

[0013] The weather forecast module is used to collect and process meteorological data, forecast weather conditions, and ensure the safe operation of the power grid lines.

[0014] The beneficial effects of this invention include: First, it helps staff more effectively grasp the real-time operating status of the power grid in a more intelligent way, thereby providing intelligent auxiliary support for operation management decisions; it provides real-time monitoring and defense for the safety and stability of the smart grid, improving the safety and stability of the smart grid system; and it provides efficient and accurate weather forecast services, ensuring the safe operation of power grid lines. Second, compared with existing integrated systems or methods, this invention greatly improves the availability and accuracy of meteorological data, enhances the forecast accuracy of numerical weather prediction models, and thus improves the power grid's disaster prevention and mitigation capabilities. Third, its special fusion method for multi-source meteorological data, based on the correlation matching of the multi-source meteorological data itself, provides a more comprehensive and in-depth analysis of meteorological data, greatly improving the quality and efficiency of the fused numerical weather prediction data. Furthermore, this invention integrates multi-source meteorological data with the power industry big data platform, significantly reducing the difficulties in meteorological data preprocessing, calculation, and storage. Finally, its method for dimensionality reduction of high-dimensional weather forecast data and its method for obtaining the covariance matrix greatly reduce the number of features, achieving multiple dimensionality reduction objectives. Attached Figure Description

[0015] Figure 1 This is a framework diagram of an intelligent management system for a power system according to an embodiment of the present invention. Detailed Implementation

[0016] To better understand the present invention, the method and system of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0017] To provide a comprehensive understanding of the invention, numerous specific details are set forth in the following detailed description. However, those skilled in the art will understand that the invention can be implemented without these specific details. In the embodiments, well-known methods, processes, and components are not described in detail to avoid unnecessarily complicating the embodiments.

[0018] See Figure 1 As shown, the present invention provides an intelligent management system for a power system, the system comprising at least:

[0019] The power generation management module is used to manage overall power production, specifically including the management of power supply reliability, overall voltage, overall line loss, substations, lines, monthly planned power outages, and monthly fault power outages;

[0020] A power distribution automation safety platform is used to execute safe power distribution schemes.

[0021] Power quality modules are used to protect power grid lines and transformers from overcurrent, overload, and temperature.

[0022] The dispatch management module is used to manage power dispatch.

[0023] The energy-saving management module is used to statistically analyze the energy consumption data of energy-consuming equipment and provide energy-saving solutions for the energy-consuming equipment based on electricity consumption and energy consumption data.

[0024] The electricity service module is used to provide electricity services to electricity users.

[0025] The power grid line inspection module is used for video monitoring and intelligent inspection of the power grid lines;

[0026] The weather forecast module is used to collect and process meteorological data, forecast weather conditions, and ensure the safe operation of the power grid lines.

[0027] Preferably, the power grid line inspection module specifically includes:

[0028] A video monitoring unit is used for real-time video monitoring of the power grid lines;

[0029] The inspection and data collection unit is used to collect and upload information from each inspection and monitoring point.

[0030] The integrated control unit is used to record and statistically manage the collected information.

[0031] Preferably, the electricity service module specifically includes:

[0032] The electricity sales unit is used to determine the electricity consumption of each electricity user and the corresponding electricity fee.

[0033] The electricity payment unit is used to provide payment data for each electricity user, wherein the payment data includes at least: payment balance, payment time, and payment bill;

[0034] The bill notification unit, together with the electricity payment module, is used to send payment bills to the target user's address.

[0035] Preferably, the weather forecast module specifically includes:

[0036] The data acquisition unit is used to collect multi-source meteorological data;

[0037] The processing unit is used for data processing of multi-source meteorological data;

[0038] The analysis unit is used to perform statistical analysis and prediction on the processed data;

[0039] Forecasting unit, used to provide routine forecasts of weather conditions;

[0040] Early warning unit, used to provide early warnings for extreme weather.

[0041] Preferably, the processing unit is used to process multi-source meteorological data, specifically including:

[0042] The fusion unit is used to fuse the multi-source meteorological data to form numerical weather forecast data.

[0043] The preprocessing unit is used to preprocess the numerical weather forecast data.

[0044] A filtering unit is used to filter the numerical weather forecast data by dimension.

[0045] The dimensionality reduction unit is used to reduce the dimensionality of the high-dimensional weather forecast data.

[0046] Preferably, the fusion unit is used to fuse the multi-source meteorological data to form numerical weather forecast data, specifically including:

[0047] The allocation unit is used to assign initial weights to each type of meteorological data;

[0048] Sampling units are used to sample each type of meteorological data based on a weighted distribution.

[0049] Training units are used to iteratively train weak classifiers in parallel for each type of meteorological data;

[0050] The selection unit is used to select the weak classifier corresponding to the sample with the lowest training error rate, and calculate the combined weight value associated with the selected weak classifier based on the lowest training error rate of the iteration.

[0051] The update unit is used to assign update weights to each type of meteorological data. The updated weight distribution corresponds to the iterative function with the lowest training error rate.

[0052] An iteration unit is used to repeat steps 2-2 to 2-5 for a predetermined number of iterations;

[0053] A synthesis unit is used to form a final classifier, which is the sum of selected weak classifiers weighted by updated combined weight values ​​in each iteration.

[0054] Preferably, the preprocessing unit is used to preprocess the numerical weather forecast data, specifically including: converting the numerical weather forecast data into ASCII text data and performing data quality control.

[0055] Preferably, the preprocessing unit converts the numerical weather forecast data into ASCII text data and performs data quality control, specifically including:

[0056] The identification unit is used to identify the format of numerical weather forecast data. If the data format is ASCII text format, it is directly imported into the database for data quality control; otherwise, it is converted.

[0057] The conversion unit is used to convert data into ASCII text data;

[0058] The data entry unit is used to control data quality in terms of logic, temporal continuity, and spatial consistency, thereby enabling automatic data entry.

[0059] Preferably, the filtering unit is used to filter the dimensions of the numerical weather forecast data. The principle is that the larger the information entropy value of a dimension, the greater the amount of information in the original data it contains, and it is a feature that should be retained; if the information entropy of a certain dimension is lower, it means that the dimension contains less data information and it is a feature that should be removed.

[0060] The filtering unit specifically includes:

[0061] The entropy calculation unit is used to calculate the information entropy value H(ai) of each dimension ai of the high-dimensional weather forecast data;

[0062] The preset unit is used to preset the information entropy threshold δ.

[0063] The elimination unit is used to retain dimension ai if the information entropy value H(ai) of each dimension ai of the high-dimensional weather forecast data is greater than δ; otherwise, dimension ai is eliminated.

[0064] Preferably, the dimensionality reduction unit is used to reduce the dimensionality of the high-dimensional weather forecast data, specifically including:

[0065] The centralization unit is used to subtract the mean of each dimension of the high-dimensional weather forecast data to center the sample matrix, thereby obtaining the high-dimensional weather forecast data centralization matrix X. m×n ;

[0066] The covariance calculation unit is used to calculate the dimensional covariance matrix;

[0067] The feature calculation unit is used to calculate the eigenvectors and eigenvalues ​​of the dimensional covariance matrix.

[0068] Principal component selection unit (PCU) is used to select the k eigenvectors corresponding to the k largest eigenvalues ​​as column vectors to form the eigenvector matrix V. n×k The largest k eigenvalues ​​represent the number of principal components, and the number k is determined by the contribution rate of the eigenvalues.

[0069] The dimensionality reduction calculation unit is used to calculate the dimensionality reduction result Y=V of the high-dimensional weather forecast data.n×k T ×X m×n ;

[0070] The covariance calculation unit, used to calculate the dimensional covariance matrix, specifically includes:

[0071] A sample selection unit is used to select a subset of samples from multiple dimensions from the high-dimensional weather forecast data;

[0072] An extraction unit is used to extract a d-dimensional feature vector for each sample in each sample subset, wherein the feature vector includes indices pointing to the corresponding sample and sample attributes;

[0073] The combination unit is used to combine the feature vectors of each sample subset into a d×d feature vector covariance matrix, which serves as the descriptor of the corresponding sample subset.

[0074] The eigenvector covariance matrix

[0075] Where p represents the number of samples in the subset, V represents the feature vector, μ is the feature mean, and T represents the transpose.

[0076] The distance calculation unit is used to determine the pairwise distance value between each pair of eigenvector covariance matrices to measure the similarity between sample subsets;

[0077] The distance value

[0078] Where, λ i (C1, C2), i = 1, ..., p are the generalized eigenvalues ​​of two covariance matrices C1 and C2;

[0079] The distance filtering unit is used to define distance thresholds and filter pairwise covariance distances that meet the conditions.

[0080] Constructing units are used to construct d×d-dimensional distance matrices based on pairwise covariance distance values;

[0081] The description unit is used to treat the distance matrix as the dimensional covariance matrix of high-dimensional weather forecast data.

[0082] Preferably, wherein the covariance calculation unit, used to calculate the dimensional covariance matrix, further includes:

[0083] The mapping configuration unit is used to configure N mapping functions (Mappers).

[0084] A reduction configuration unit is used to configure a reduction function Reducer, the input of which is the output of each Mapper;

[0085] The result substitution unit is used to substitute the summary results of the Reducer into the calculation of the dimensional covariance matrix of the high-dimensional weather forecast data.

[0086] Compared with existing technologies, this invention has the following significant advantages: First, it helps staff to more effectively grasp the real-time operating status of the power grid in a more intelligent way, thereby providing intelligent auxiliary support for operation management decisions. It also provides real-time monitoring and defense for the safety and stability of the smart grid, improving the safety and stability of the smart grid system. Furthermore, the system can provide weather forecast services, ensuring the safe operation of power grid lines. Second, compared with existing integrated systems or methods, this invention greatly improves the availability and accuracy of meteorological data, enhances the forecast accuracy of numerical weather prediction models, and thus improves the power grid's disaster prevention and mitigation capabilities. Third, its special fusion method for multi-source meteorological data, based on the inherent correlation and matching of the multi-source meteorological data, provides a more comprehensive and in-depth analysis of the meteorological data, greatly improving the quality and efficiency of the fused numerical weather prediction data. Additionally, this invention integrates multi-source meteorological data with the power industry big data platform, significantly reducing the difficulties in meteorological data preprocessing, calculation, and storage. Finally, the method for dimensionality reduction of high-dimensional weather forecast data and the method for obtaining the covariance matrix significantly reduce the number of features, achieving multiple dimensionality reduction objectives.

[0087] This description only illustrates preferred embodiments of the invention and is not intended to limit the scope, applicability, or configuration of the invention. Rather, the detailed description of the embodiments will enable those skilled in the art to practice the invention. It should be understood that appropriate changes and modifications may be made to some details without departing from the spirit and scope of the invention as defined by the appended claims.

Claims

1. An intelligent power system management system, characterized by, The system at least comprises: a power generation management module for managing overall power generation, specifically including management of power supply reliability, integrated voltage, integrated line loss, transformer substation, line, monthly planned power outage, and monthly fault power outage; a power distribution automation security platform for executing a secure power distribution scheme; a power quality module for protecting against overcurrent, overload, and temperature of power grid lines and transformers; a dispatch management module for managing power dispatch; an energy saving management module for statistically analyzing energy consumption data of energy-using equipment and providing energy saving schemes for the energy-using equipment according to power consumption and energy consumption data; a power consumption service module for providing power consumption services to power consumption units; a power grid line inspection module for video monitoring and intelligent inspection of the power grid lines; a weather forecast module for collecting and processing meteorological data, forecasting weather conditions, and ensuring safe operation of the power grid lines; The weather forecast module specifically comprises: a collection unit for collecting multi-source meteorological data; a processing unit for processing the multi-source meteorological data; an analysis unit for statistically analyzing and predicting the processed data; a forecast unit for providing regular forecasts of weather conditions; a warning unit for providing warnings for extreme weather; The processing unit is configured to process the multi-source meteorological data, specifically comprising: a fusion unit for fusing the multi-source meteorological data to form numerical weather forecast data; a preprocessing unit for preprocessing the numerical weather forecast data; a screening unit for dimension screening of the numerical weather forecast data; a dimension reduction unit for dimension reduction of high-dimensional weather forecast data; The dimension reduction unit is configured to reduce the dimension of high-dimensional weather forecast data, specifically comprising: a centering unit, configured to subtract the mean value of each dimension of the high-dimensional weather forecast data from the high-dimensional weather forecast data, to obtain a high-dimensional weather forecast data centering matrix ; a covariance calculation unit for calculating a dimension covariance matrix; a feature calculation unit for calculating eigenvectors and eigenvalues of the dimension covariance matrix; The principal component selection unit is configured to select k eigenvectors corresponding to the k largest eigenvalues as column vectors to form an eigenvector matrix ; wherein the k largest eigenvalues represent the number of principal components, and the number k is determined by the eigenvalue contribution rate. a dimension reduction computing unit configured to compute a result of dimension reduction on the high-dimensional weather forecast data ; The covariance calculation unit is configured to calculate a dimension covariance matrix, specifically comprising: a sample selection unit for selecting a plurality of dimension sample subsets from the high-dimensional weather forecast data; an extraction unit for extracting a d-dimensional eigenvector for each sample in each sample subset, wherein the eigenvector includes an index pointing to a corresponding sample and sample attribute; a combination unit for combining the eigenvectors of each sample subset into a dxd eigenvector covariance matrix as a descriptor of the corresponding sample subset; the feature vector covariance matrix ; where p represents the number of samples in the subset, V represents the feature vector, is the feature mean, and T represents the transpose; a distance calculation unit for determining a pair distance value between each pair of eigenvector covariance matrices to measure the similarity between sample subsets; the distance value ; wherein, , i = 1,... p are the generalized eigenvalues of the two covariance matrices Ci and C2; a distance screening unit for defining a distance threshold to screen the pair covariance distances that meet the conditions; a construction unit for constructing an n x n distance matrix according to the pair covariance distance values; a description unit for taking the distance matrix as a high-dimensional weather forecast data dimension covariance matrix.

2. The system of claim 1, wherein, The power grid line inspection module specifically comprises: a video monitoring unit for real-time video monitoring of the power grid lines; an inspection collection unit for collecting information of each inspection monitoring point and uploading the information. The integrated control unit is configured to record and statistically manage the collected information.

3. The system of claim 1, wherein, The electricity service module specifically includes: The electricity selling unit is configured to determine the electricity consumption of each electricity consuming unit and the electricity charge corresponding to the electricity consumption; The electricity payment unit is configured to provide payment data of each electricity consuming unit, wherein the payment data at least includes: payment balance, payment time, and payment bill; The bill notification unit is configured to send the payment bill to a target user address.

4. The system of claim 1, wherein, The fusion unit is configured to fuse the multi-source weather data to form numerical weather prediction data, and specifically includes: The distribution unit is configured to distribute an initial weight to each type of weather data; The sampling unit is configured to sample each type of weather data sample based on the distribution of the weight; The training unit is configured to iteratively train a weak classifier for each type of weather data in parallel; The selection unit is configured to select a weak classifier corresponding to a sample with the lowest training error rate, and calculate a combined weight value associated with the selected weak classifier according to the lowest training error rate of the iteration; The update unit is configured to distribute an updated weight to each type of weather data, and the updated weight distribution corresponds to an iterative function of the lowest training error rate; The iteration unit is configured to repeatedly execute the steps of the sampling unit, the training unit, the selection unit, and the update unit for a predetermined number of iterations; The integration unit is configured to form a final classifier, which is the sum of the selected weak classifiers weighted by the updated combined weight value at each iteration.

5. The system of claim 1, wherein, The preprocessing unit is configured to perform data preprocessing on the numerical weather prediction data, specifically including: converting the numerical weather prediction data into ASCII text data and performing data quality control.

6. The system of claim 5, wherein, The preprocessing unit converts the numerical weather prediction data into ASCII text data and performs data quality control, specifically including: The identification unit is configured to perform format identification on the numerical weather prediction data, if the data format is ASCII text format data, then directly import the database for data quality control, otherwise, perform conversion; The conversion unit is configured to convert the data into ASCII text data; The warehousing unit is configured to perform data quality control from the aspects of logicality, time continuity, and spatial consistency, and realize automatic warehousing.

7. The system of claim 1, wherein, The screening unit is configured to perform dimension screening on the numerical weather prediction data, and the principle is that the greater the information entropy value of the dimension, the greater the amount of information contained in the original data, which belongs to the feature that should be retained; If the information entropy of a certain dimension is lower, it means that the amount of data information contained in the dimension is less, which belongs to the feature that should be excluded; The screening unit specifically includes: An entropy value calculation unit is configured to calculate an information entropy value H(a i ) of each dimension a of the high-dimensional weather forecast data. i ) of each dimension a of the high-dimensional weather forecast data. A pre-setting unit is configured to pre-set an information entropy threshold value ; A pruning unit is configured to retain a dimension a i if an information entropy value H(a i ) of each dimension a of the high-dimensional weather forecast data is greater than i ; otherwise, the dimension a i is pruned.

8. The system of claim 7, wherein, The covariance calculation unit is configured to calculate the dimension covariance matrix, and further includes: The mapping configuration unit is configured to configure N mapping functions Mapper; The reduction configuration unit is configured to configure a reduction function Reducer, and the input of the Reducer is the output result of each Mapper; The result substitution unit is configured to substitute the summary result of the Reducer into the calculation of the high-dimensional weather prediction data dimension covariance matrix. ​

Images