Information digitization system for power grid power supply
By designing a digital information system for power supply in the power grid and using deep learning algorithms for data identification and analysis, the problem of lack of real-time intelligent management of power grid business data is solved, intelligent data arrangement and classification is realized, and power supply reliability and power consumption service quality are improved.
Patent Information
- Application Number
- CN202411740473.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-29
- Publication Date
- 2025-03-21
AI Technical Summary
The existing power grid business data lacks real-time intelligent management, resulting in data being unable to be effectively processed and transmitted, and cannot respond quickly to business needs, affecting power supply reliability and power service quality.
Design a digital information system for power supply in power grids, including central control module, intelligent identification module, information arrangement and classification module, data verification and supplementary module, power consumption prediction module and business push module, use deep learning algorithms to identify and analyze data, realize intelligent arrangement and classification of data, and pass integrated verification and supplementation of government and enterprise data.
It realizes intelligent identification and analysis of power grid data, improves data processing and transmission efficiency, can quickly respond to business needs, improve power supply reliability and power consumption service quality, and provide accurate business push and power consumption prediction.
Smart Images
Figure CN119670964A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of power distribution network management, and in particular to an information digitization system for power supply of a power grid. Background Art
[0002] As an advanced productive force and basic industry, the power industry has played an important role in promoting the development of the national economy and social progress. The power grid business is an efficient and fast energy transmission channel and an optimized configuration platform. It is a key link in the sustainable development of energy and electricity. It plays an important hub role in the modern energy supply system and is an important way to ensure national energy security. The power grid business can provide effective data and information support for power operations through information and intelligent means, thereby improving power supply reliability and power service quality. The existing power grid business data lacks real-time intelligent management of business data analysis and operation, resulting in the inability to effectively process, analyze and transmit data after the power grid data is collected, and the inability to respond quickly to business needs is not conducive to the sustainable growth of the business. Summary of the invention
[0003] The object of the present invention is to provide an information digitization system powered by a power grid to solve the problems raised in the above background technology.
[0004] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a power grid-powered information digitization system, comprising a central control module, the central control module is connected to an intelligent identification module, an information arrangement and classification module, a data verification and supplement module, a power consumption prediction module and a business push module, the central control module and the intelligent identification module, the information arrangement and classification module, the data verification and supplement module, the power consumption prediction module, and the business push module are bidirectionally connected.
[0005] Preferably, the intelligent identification module includes a data acquisition unit and a data analysis unit, the data acquisition unit is used to acquire the power grid operation data required by the system, and the data analysis unit uses a deep learning algorithm to identify and analyze the power grid operation data.
[0006] Preferably, the data analysis unit includes a preprocessing subunit, a model creation subunit, an extraction subunit and a calculation subunit;
[0007] The preprocessing subunit cleans the data collected by the data acquisition unit, processes missing values and abnormal values, and extracts features that are helpful in describing the state of the power grid;
[0008] The model creation subunit is used to construct a deep learning model to extract the correlation of power grid data;
[0009] The extraction subunit is used to extract the correlation characteristics of power grid data after the model training is completed;
[0010] The computing subunit obtains the low-dimensional representation of the data and calculates the correlation matrix.
[0011] Preferably, the information arrangement and classification module arranges and classifies the data after obtaining the correlation between the power grid data information using the deep learning algorithm, and the information arrangement and classification module includes a visualization unit, a cluster analysis unit, a classification evaluation unit, a result application unit and a cluster visualization unit;
[0012] The visualization unit is used to visually view the distribution of the correlation data through data visualization. The correlation features can be mapped to a two-dimensional or three-dimensional space through a scatter plot or a heat map to observe the distribution of the scatter plot data points. If the correlation matrix is large, it can be drawn into a heat map to observe the correlation between different data points.
[0013] The cluster analysis unit is used to arrange and classify the correlation features using a clustering algorithm;
[0014] The classification evaluation unit evaluates the quality of the clustering result;
[0015] The result application unit performs anomaly detection, pattern recognition and system monitoring on the clustering results;
[0016] The cluster visualization unit visualizes the clustering results to better understand the data distribution.
[0017] Preferably, the data verification and supplementation module is used for data resource sharing and integration between the government and enterprises, which can provide strong support for the verification and supplementation of power grid electricity consumption data;
[0018] Establish a data sharing platform using data verification and supplementation modules to allow data exchange between governments and enterprises, formulate data sharing policies, standards and processes, and develop API interfaces to achieve automatic data transmission and update;
[0019] Government data may include: data used to analyze residential electricity demand, grid layout and electricity demand analysis, and reflect corporate electricity demand; corporate data may include: real-time or regularly collected electricity consumption data, grid load, power supply quality, etc.;
[0020] The data verification and supplementation module is also used to verify the power grid electricity consumption data by using the integration of government and enterprise data. It needs to conduct consistency checks, compare the electricity consumption data provided by the government and enterprises, check whether there are inconsistencies, and verify data integrity, using government data to fill in missing values or abnormal values in enterprise data;
[0021] The data verification and supplementation module uses government data to supplement corporate electricity consumption data: uses economic and population trend data provided by the government to predict changes in electricity consumption, and combines GIS data to analyze electricity demand in different regions;
[0022] The integrated data is deeply analyzed through the data verification and supplementation module to obtain more accurate electricity demand forecasts and provide data support for power grid planning, energy efficiency management, demand response, etc.
[0023] Preferably, the electricity consumption prediction module selects a suitable prediction model based on the electricity consumption data obtained by the intelligent recognition module according to the data characteristics, then trains the model using historical data, then uses the trained model to predict the test set, and uses mean square error and root mean square error indicators to evaluate the model performance.
[0024] Preferably, the service push module builds a user portrait based on the user's electricity consumption behavior, and pushes personalized services according to the user portrait and electricity consumption prediction results.
[0025] Compared with the prior art, the present invention has the following beneficial effects:
[0026] The information digitization system for power grid power supply proposed in the present invention can intelligently identify relevant information in the power grid, arrange and classify it according to the degree of correlation, use the integration of government and enterprise data to verify and supplement it, realize the authenticity of the data, and then push its business based on the customer's electricity consumption information. According to the principle of deep learning, the electricity consumption habits of a single customer are predicted and judged, thereby realizing accurate business provision. In addition, the digital transformation and digital power grid construction technology route can provide customers with a good user experience and provide customers with a safe and reliable power supply service guarantee architecture. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 It is a schematic diagram of the system structure of the present invention.
[0028] Figure 2 It is a schematic diagram of the structure of the intelligent recognition module of the present invention.
[0029] Figure 3 It is a structural schematic diagram of the data analysis unit of the present invention.
[0030] Figure 4-5 This is an example diagram of a power grid data correlation model extracted using the deep learning model of the present invention.
[0031] Figure 6 This is an example diagram of the correlation feature extraction model of the present invention.
[0032] Figure 7 This is an example diagram of the correlation calculation model of the present invention.
[0033] Figure 8 It is a schematic diagram of the information arrangement and classification module structure of the present invention.
[0034] Fig. 9 This is a visualization unit process view of the present invention.
[0035] Fig.10 This is a clustering diagram of the clustering analysis unit of the present invention.
[0036] Fig.11 Evaluate the clustering model graph for the classification evaluation unit of the present invention.
[0037] Fig.12 This is a prediction model diagram of the electricity consumption prediction module of the present invention.
[0038] Fig.13 This is a diagram of the evaluation model of the electricity consumption prediction module of the present invention. DETAILED DESCRIPTION
[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0040] See also Figures 1 to 13 The present invention provides a technical solution: a power grid power supply information digitization system, including a central control module, the central control module is connected to an intelligent identification module, an information arrangement and classification module, a data verification and supplement module, a power consumption prediction module and a business push module, and the central control module and the intelligent identification module, the information arrangement and classification module, the data verification and supplement module, the power consumption prediction module, and the business push module are bidirectionally connected.
[0041] The intelligent identification module includes a data acquisition unit and a data analysis unit. The data acquisition unit is used to acquire the power grid operation data required by the system, and the data analysis unit uses a deep learning algorithm to identify and analyze the power grid operation data.
[0042] The data analysis unit includes a preprocessing subunit, a model creation subunit, an extraction subunit and a calculation subunit;
[0043] The preprocessing subunit cleans the data collected by the data acquisition unit, processes missing values and abnormal values, and extracts features that are helpful in describing the state of the power grid, such as statistical features, time domain features, frequency domain features, etc., and then scales the data to a small interval, such as [0,1] or a mean of 0 and a variance of 1;
[0044] The model creation subunit is used to build a deep learning model to extract the correlation of power grid data. It can be an autoencoder unsupervised learning algorithm, which can learn a low-dimensional representation of the data, such as Figure 4 As shown, or CNN, NN, LSTM, etc.; for time series data, RNN or LSTM can be used to capture long-term dependencies in the time series, such as Figure 5 As shown,
[0045] The extraction subunit is used to extract the correlation features of power grid data after the model training is completed. For the autoencoder, the correlation features can be obtained from the encoded layer. For CNN and RNN / LSTM, the features can be obtained from the last hidden layer, such as Figure 6 As shown;
[0046] The computing subunit obtains the low-dimensional representation of the data and calculates the association matrix, for example, using cosine similarity or Euclidean distance, such as Figure 7 shown.
[0047] The information arrangement and classification module arranges and classifies the data after obtaining the correlation between the power grid data information using the deep learning algorithm. The information arrangement and classification module includes a visualization unit, a cluster analysis unit, a classification evaluation unit, a result application unit and a cluster visualization unit.
[0048] The visualization unit is used to visually view the distribution of correlation data through data visualization. The correlation features can be mapped to two-dimensional or three-dimensional space through scatter plots or heat maps to observe the distribution of scatter plot data points. If the correlation matrix is large, it can be drawn into a heat map to observe the correlation between different data points. The process is as follows: Fig. 9 As shown;
[0049] The clustering analysis unit is used to arrange and classify the association features using a clustering algorithm. First, select a clustering algorithm. Common clustering algorithms include K-means, hierarchical clustering, DBSCAN, etc. Then determine the number of clusters. You can use the elbow rule, silhouette coefficient, etc. to determine the optimal number of clusters. Then perform clustering to cluster the association features, such as Fig.10 As shown;
[0050] The classification evaluation unit evaluates the quality of clustering results. The following indicators can be used: silhouette coefficient: measures the consistency and separation within the cluster; Calinski-Harabasz index: measures the separation of clusters and the compactness within the cluster; Davies-Bouldin index: based on the ratio of intra-class scatter and inter-class distance, using the silhouette coefficient distance, such as Fig.11 As shown;
[0051] The result application unit performs anomaly detection, pattern recognition and system monitoring on the clustering results. Anomaly detection is used to find that points far from the center in the clustering results may be abnormal points. Pattern recognition is used to identify typical patterns in power grid data for prediction or optimization. System monitoring is used to monitor the dynamic changes of clustering results to detect changes in power grid status.
[0052] The clustering visualization unit visualizes the clustering results to better understand the data distribution.
[0053] The Python language for visualizing clustering results is:
[0054] plt.scatter(encoded_data[:,0],encoded_data[:,1],c=clusters,cmap='viridis',marker='o',alpha=0.6)
[0055] plt.xlabel('Feature 1')
[0056] plt.ylabel('Feature 2')
[0057] plt.title('Clustering Results')
[0058] plt.show()
[0059] Through the above steps, the correlation between power grid data information can be effectively arranged and classified, and further applied to the monitoring, maintenance and optimization of the power grid.
[0060] The power consumption prediction module uses the power consumption data obtained by the intelligent recognition module, the user's power consumption, power consumption time, power consumption behavior and other data, and removes outliers, fills in missing values, standardizes data formats, and performs feature extraction and conversion. According to the characteristics of the data, it selects a suitable prediction model, such as linear regression, time series analysis, machine learning model (such as random forest, neural network), etc., and then uses historical data to train the model. For example, Fig.12 , and then use the trained model to predict the test set, and use indicators such as mean square error (MSE) and root mean square error (RMSE) to evaluate the model performance, such as Fig.13 shown.
[0061] The service push module builds a user profile based on the user's electricity consumption behavior and pushes personalized services based on the user profile and electricity consumption prediction results. For example, the following Python language:
[0062] #Assume push service
[0063] def push_business(user_id,predicted_usage):
[0064] if predicted_usage>threshold_high:
[0065] return "Push energy saving suggestions"
[0066] elif predicted_usage <threshold_low:
[0067] return "Push promotional activities"
[0068] else:
[0069] return "No special push"
[0070] # Apply prediction results
[0071] data['business_push']=data.apply(lambdarow:
[0072] push_business(row['user_id'],row['predicted_usage']),axis=1)
[0073] Services are pushed to users through emails, text messages, applications, etc., and user feedback on the pushed content is collected to optimize the model and push strategy. The model is retrained regularly with the latest data to maintain prediction accuracy and adjust the business push strategy based on user feedback and market changes.
[0074] Through the above steps, electricity consumption forecasting and business push can be effectively carried out based on user electricity consumption information, thereby improving user satisfaction and the service level of power companies.
[0075] The information digitization system for power grid power supply proposed in the present invention can intelligently identify relevant information in the power grid, arrange and classify it according to the degree of correlation, use the integration of government and enterprise data to verify and supplement it, realize the authenticity of the data, and then push its business based on the customer's electricity consumption information. According to the principle of deep learning, the electricity consumption habits of a single customer are predicted and judged, thereby realizing accurate business provision. In addition, the digital transformation and digital power grid construction technology route can provide customers with a good user experience and provide customers with a safe and reliable power supply service guarantee architecture.
[0076] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. An information digitization system for power grid power supply, including a central control module, characterized in that: The central control module is connected to an intelligent identification module, an information arrangement and classification module, a data verification and supplement module, a power consumption prediction module and a business push module. The central control module and the intelligent identification module, the information arrangement and classification module, the data verification and supplement module, the power consumption prediction module and the business push module are bidirectionally connected.
2. The information digitization system powered by a power grid according to claim 1, characterized in that: The intelligent identification module includes a data acquisition unit and a data analysis unit. The data acquisition unit is used to acquire the power grid operation data required by the system, and the data analysis unit uses a deep learning algorithm to identify and analyze the power grid operation data.
3. The information digitization system powered by a power grid according to claim 2, characterized in that: The data analysis unit includes a preprocessing subunit, a model creation subunit, an extraction subunit and a calculation subunit; The preprocessing subunit cleans the data collected by the data acquisition unit, processes missing values and abnormal values, and extracts features that are helpful in describing the state of the power grid; The model creation subunit is used to construct a deep learning model to extract the correlation of power grid data; The extraction subunit is used to extract the correlation characteristics of power grid data after the model training is completed; The computing subunit obtains the low-dimensional representation of the data and calculates the correlation matrix.
4. The information digitization system powered by a power grid according to claim 1, characterized in that: The information arrangement and classification module arranges and classifies the data after obtaining the correlation between the power grid data information using the deep learning algorithm. The information arrangement and classification module includes a visualization unit, a cluster analysis unit, a classification evaluation unit, a result application unit and a cluster visualization unit; The visualization unit is used to visually view the distribution of the correlation data through data visualization. The correlation features can be mapped to a two-dimensional or three-dimensional space through a scatter plot or a heat map to observe the distribution of the scatter plot data points. If the correlation matrix is large, it can be drawn into a heat map to observe the correlation between different data points. The cluster analysis unit is used to arrange and classify the correlation features using a clustering algorithm; The classification evaluation unit evaluates the quality of the clustering result; The result application unit performs anomaly detection, pattern recognition and system monitoring on the clustering results; The cluster visualization unit visualizes the clustering results to better understand the data distribution.
5. The information digitization system powered by a power grid according to claim 1, characterized in that: The data verification and supplementation module is used for data resource sharing and integration between the government and enterprises, which can provide strong support for the verification and supplementation of power grid electricity consumption data; Establish a data sharing platform using data verification and supplementation modules to allow data exchange between governments and enterprises, formulate data sharing policies, standards and processes, and develop API interfaces to achieve automatic data transmission and update; Government data may include: data used to analyze residential electricity demand, grid layout and electricity demand analysis, and reflect corporate electricity demand; corporate data may include: real-time or regularly collected electricity consumption data, grid load, power supply quality, etc.; The data verification and supplementation module is also used to verify the power grid electricity consumption data by using the integration of government and enterprise data. It needs to conduct consistency checks, compare the electricity consumption data provided by the government and enterprises, check whether there are inconsistencies, and verify data integrity, using government data to fill in missing values or abnormal values in enterprise data; The data verification and supplementation module uses government data to supplement corporate electricity consumption data: uses economic and population trend data provided by the government to predict changes in electricity consumption, and combines GIS data to analyze electricity demand in different regions; The integrated data is deeply analyzed through the data verification and supplementation module to obtain more accurate electricity demand forecasts and provide data support for power grid planning, energy efficiency management, demand response, etc.
6. The information digitization system powered by a power grid according to claim 1, characterized in that: The electricity consumption prediction module selects a suitable prediction model based on the electricity consumption data obtained by the intelligent recognition module according to the data characteristics, then trains the model using historical data, and then uses the trained model to predict the test set, and uses the mean square error and root mean square error indicators to evaluate the model performance.
7. The information digitization system powered by a power grid according to claim 1, characterized in that: The service push module builds a user profile based on the user's electricity consumption behavior, and pushes personalized services according to the user profile and electricity consumption prediction results.