A power information management method based on building user portraits in the data center

Abstract

The present invention provides a method for power information management based on building user portraits in a data middle platform, comprising the following steps: unifying the data interaction platform to collect various types of power business system data collected by terminal devices into the power data middle platform; creating a label application scenario based on the company's core business of marketing, production and operation; extracting the characteristic values of user-related data in the label application scenario, and generating corresponding user labels based on the characteristic values; building a report scenario based on the user labels, including building user portraits, label group analysis, label life cycle monitoring, label online and offline management information, and label classification statistical information; establishing a data label library in the data middle platform, and providing label data services to the outside through the data middle platform. This improves business coverage and scenario practicality, enhances data labeling capabilities, enhances data analysis capabilities, and improves the information service level of power grid enterprises.

Description

Technical Field

[0001] The present invention relates to the technical field of power information management methods, and in particular to a power information management method for constructing user portraits based on a data middle platform. Background Art

[0002] With the rapid development of power grid technology, power information data is becoming increasingly complex, and a comprehensive power information management method is urgently needed. In recent years, the construction of the power system data center has been gradually improved. "Promoting the construction of enterprise-level tag libraries" and "deepening big data applications and value mining to create a 'data supermarket'" have also become key tasks for power grid companies. The construction of the data center can achieve the aggregation and unified management of power business system data, and provide unified and efficient data resource services and data analysis services for the company inside and outside. However, the current business coverage of the data center is not diverse and practical. The existing tag services of the tag component are inconsistent with the data center as a unified export of data sharing services. The existing tag statistical analysis has few dimensions, relatively simple scenarios, and is not very practical. Summary of the Invention

[0003] The present invention aims to at least solve the technical problems existing in the prior art, such as the complexity of power information data processing, the non-unified power data sharing service export, the imperfect data resource service and data analysis service, the diversified coverage and low practicality of data middle-end business, the few dimensions of existing label statistical analysis, the relatively single scenario, and the low practicality.

[0004] To this end, the first aspect of the present invention provides a power information management method for building user portraits based on a data middle platform.

[0005] The present invention provides a power information management method for building a user profile based on a data center, comprising the following steps:

[0006] S1. Unified data interaction platform, which aggregates various power business system data collected by terminal devices to the power data center;

[0007] S2. Create label application scenarios based on the company's core business of marketing, production and operation;

[0008] S3. In the tag application scenario, extract characteristic values of user-related data and generate corresponding user tags according to the characteristic values;

[0009] S4. Construct a report scenario based on the user tags, including constructing user portraits, tag group analysis, tag lifecycle monitoring, tag online and offline management information, and tag classification statistics;

[0010] S5. Establish a data label library in the data middle platform, and the data label library provides label data services to the outside world through the data middle platform.

[0011] The present invention proposes a power information management method based on building user portraits in a data center.

[0012] According to the above technical solution of the present invention, a power information management method based on building a user profile on a data middle platform may also have the following additional technical features:

[0013] In the above technical solution, the specific process of generating user tags in S3 is as follows:

[0014] S31. Data preparation: Synchronize data tables from the source database to DWS through DAYU for use in label generation rules and model algorithms, and complete data processing, storage, calculation, and integration.

[0015] S32, wide table creation: Complete the mapping with the business data table through wide table configuration, and uniformly process the same intermediate result data used by different tags into wide table data;

[0016] S33. Tag management: Extract wide table data and create tags based on the basic information and business logic rules of tag configuration.

[0017] In the above technical solution, the tag management includes the following steps:

[0018] S331. Configure basic information and business logic rules for tags, where the basic information includes tag name, directory, business meaning, and implementation logic.

[0019] S332, label update: Modify the basic information of the label and optimize the label configuration according to business needs. At the same time, you can deactivate the label and stop label publishing according to business status and business needs;

[0020] S333, Group Analysis: Complete preliminary analysis of data tags, store and classify according to tag attributes;

[0021] S334. Label publishing: The label value data obtained through analysis and calculation is published in the system for users to query and operate.

[0022] Furthermore, the method for constructing the user portrait in S4 is as follows:

[0023] S41. Tag analysis: Analyze and perform combined queries on tag values through profiling and group filtering functions;

[0024] S42. Tag portrait: Generate tag application information such as tag portrait, group analysis, etc. required for the tag application scenario.

[0025] Furthermore, the data tag library in S5 includes tag value information, tag application information and tag management information, the tag value information includes attribute tags, fact tags, model tags and composite tags, the tag application information includes analysis subject data, portrait data and clustering data, and the tag management information includes business scenario configuration data, portrait application configuration data, tag evaluation analysis data, tag directory data, tag rules, algorithm models, tag no data, tag service no data and tag sharing configuration data.

[0026] Furthermore, the data middle platform is configured with an API interface, and the data middle platform API provides tag data services to the outside world. The tag management information also includes tag service API information.

[0027] Furthermore, the data middle platform includes a source layer, a sharing layer and an analysis layer. The source layer obtains the data source, which includes customer data, operation and maintenance data and financial data. The sharing layer stores the data source in the corresponding data area and passes it to the analysis layer. The analysis layer analyzes and processes the data and establishes a data label library.

[0028] Furthermore, the upper-layer data middle-station analysis layer and the lower-layer data middle-station analysis layer are connected to each other through SG-UEP.

[0029] Furthermore, the data center is provided with a BI tool, and the report scenario is displayed and monitored using the BI tool to achieve visual operation.

[0030] In summary, due to the adoption of the above technical solutions, the beneficial effects of the present invention are: data is centralized to the data center for easy management, the export of power data sharing services is unified, the tag application scenarios are improved, the practicality and coverage of tag scenarios are enhanced, the support capabilities of tag business applications are improved, and through customer tag profiling, data statistical analysis capabilities are improved, the entire tag process is monitored and displayed, and data sharing services are optimized. Constructing multi-dimensional monitoring business scenarios promotes lean management of company data, greatly improving the company's overall management efficiency, improving internal data information management efficiency, and improving the quality of external data services.

[0031] Additional aspects and advantages of the invention will become apparent from the description which follows, or may be learned by practice of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments with reference to the accompanying drawings, in which:

[0033] Figure 1 This is a flow chart of a method for power information management based on building user portraits on a data center platform according to the present invention;

[0034] Figure 2 This is a schematic diagram of the process of generating labels and building user portraits in the data center in steps S3 and S4 of the present invention;

[0035] Figure 3 yes Figure 2 Flow chart of label management in [1].

[0036] Figure 4 This is a data architecture diagram of a power information management method for building user portraits based on a data middle platform in the present invention;

[0037] Figure 5 This is the overall architecture diagram of data tags of an embodiment of the power information management method of the present invention that builds user portraits based on the data middle platform. DETAILED DESCRIPTION

[0038] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0039] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0040] Refer to the following Figures 1 to 5 To describe a power information management method for building user portraits based on a data middle platform provided in some embodiments of the present invention.

[0041] Some embodiments of the present application provide a power information management method for building user portraits based on a data middle platform.

[0042] like Figures 1 to 5 As shown, the first embodiment of the present invention proposes a power information management method based on building a user profile of a data center, and the method includes the following steps:

[0043] S1. Unified data interaction platform, which aggregates various power business system data collected by terminal devices to the power data center;

[0044] The original data interaction platforms, such as the full-service data center and ODS (operational data storage), are uniformly migrated to the data center, and all business data information of the power system is centrally statistically processed. The data center includes a source layer, a sharing layer, and an analysis layer. The source layer transmits information with each terminal information collection device to obtain data sources. The acquired data sources include various types of information such as customer data, operation and maintenance data, and financial data. The sharing layer is connected to the source layer. The sharing layer stores the data source in the corresponding data area and transmits it to the analysis layer through physical table mapping. The analysis layer then analyzes and processes the data. This realizes the unified management of power grid data.

[0045] S2. Create label application scenarios based on the company's core business of marketing, production and operation;

[0046] According to business needs, label application scenarios are constructed to provide an application background for subsequent label generation logic, establish the data application subject, and input the label application scenario application logic into the data middle platform for relevant configuration. Based on the label application scenario requirements, the data middle platform analysis capabilities can be used to realize rapid label query in the corresponding scenario, thereby improving label recognition and work efficiency. For example, by constructing a corporate energy consumption analysis label scenario, a comprehensive analysis of electricity consumption behavior data such as the number of electricity users in the target area, electricity consumption addresses, electricity consumption, electricity charges, and electricity consumption fluctuations is conducted to construct regional building vacancy rates, regional production capacity analysis, and corporate prosperity enterprise energy consumption analysis labels. This helps the government use corporate electricity data to build an enterprise energy consumption assessment model, analyze the matching degree between the enterprise's energy consumption level and its operating strength, understand the production and operation status of market enterprises, and conduct a macro analysis of the current market economic environment. At the same time, it helps the government identify shell companies, which can provide a reference for government financial regulatory departments to monitor the accuracy of loan issuance and credit fraud prevention.

[0047] Construct a labeling scenario for enterprise pollution monitoring, extract feature labels from enterprise electricity consumption information data such as industry classification, electricity consumption scale, electricity consumption area, 24-hour electricity data, etc., and combine it with real-time load data from the electricity consumption information collection system to provide real-time data for environmental protection departments to supervise the implementation of environmental protection measures of enterprises, conduct corresponding analysis, and construct labels such as enterprise pollution level, enterprise scale, enterprise electricity consumption range classification, enterprise energy conservation and emission reduction implementation index, and enterprise power outage sensitivity. Assist environmental protection departments in carrying out regional pollution source production electricity consumption monitoring, analyze and evaluate the production status of pollution sources, realize the monitoring of the implementation of emission reduction and emission limit measures of key enterprises under environmental pollution conditions, and enhance the business support capabilities of power data.

[0048] S3. In the tag application scenario, extract characteristic values of user-related data and generate corresponding user tags according to the characteristic values;

[0049] Using the data source of the data center in S1, in the label application scenario established in S2, the characteristic values of the data are extracted. The specific steps for generating labels are as follows:

[0050] S31. Data preparation: Through DAYU (an intelligent data lake operation platform that provides full-link monitoring of data flow from the source business system to the data middle platform's source layer, sharing layer, and analysis layer; supports link tracking and rapid fault location, can quickly locate fault information by calling link exception logs, and quickly solve problems based on specific fault logs; supports link visualization, showing the time consumption of each stage of the link, and performs performance analysis based on the time consumption information; supports the integration of visual link monitoring into the unified operation and management of the data middle platform), synchronize the data table from the source database to the shared layer DWS (an online data processing database based on public cloud infrastructure and platform) for use in label generation rules and model algorithms, and complete data processing, storage, calculation, and integration;

[0051] S32, Wide Table Creation: Through wide table configuration, mapping with business data tables is completed. The same intermediate result data used by different tags is uniformly processed into wide table data, effectively improving the tag generation speed.

[0052] S33. Tag management: Extract wide table data and create tags based on the basic information and business logic rules of tag configuration. Tag management includes the following processes:

[0053] S331. Configure basic information and business logic rules for tags, where the basic information includes tag name, directory, business meaning, and implementation logic.

[0054] S332, label update: Modify the basic information of the label and optimize the label configuration according to business needs. At the same time, you can deactivate the label and stop label publishing according to business status and business needs;

[0055] S333, Group Analysis: Complete preliminary analysis of data tags, store and classify according to tag attributes;

[0056] S334. Label publishing: The label value data obtained through analysis and calculation is published in the system for users to query and operate.

[0057] S4. Construct a report scenario based on the user tags, including constructing user portraits, tag group analysis, tag lifecycle monitoring, tag online and offline management information, and tag classification statistics. The method for constructing user portraits is as follows:

[0058] S41, Tag Analysis: Analyze and perform combined queries on tag values published in S3 through profiling and group filtering functions;

[0059] S42. Tag portrait: Generate tag application information such as tag portrait, group analysis, etc. required for the tag application scenario.

[0060] During and after the tag portrait construction process is completed, the BI (business intelligence platform) tools of the data center can be used to monitor and display reporting scenarios such as tag information and profile information, providing statistical data support for business use, facilitating tag display and monitoring, and further improving tag statistical analysis capabilities. Tag portraits can also be synchronized to the analysis layer DWS through tag sharing for the next step of data tag processing.

[0061] S5. Establish a data label library in the data middle platform, and the data label library provides label data services to the outside world through the data middle platform.

[0062] The data tag library includes tag value information, tag application information and tag management information. The tag value information includes attribute tags, fact tags, model tags and composite tags. The tag application information includes analysis subject data, portrait data and cluster data. The tag management information includes business scenario configuration data, portrait application configuration data, tag evaluation analysis data, tag directory data, tag rules, algorithm models, tag no data, tag service API information, tag service no data and tag sharing configuration data. The data middle platform is configured with an API (application programming interface). The data middle platform API provides tag data services to the outside world, realizes mutual communication with users, provides data sharing for different platforms, and realizes the data middle platform as a unified export for power grid data sharing services. The unified management of power information is realized. Based on this means, the tag service application review process can be further formulated to standardize data information application.

[0063] like Figure 4 As shown, Figure 4 This is a data architecture diagram, in which the State Grid headquarters data middle platform tag library as the upper layer and the provincial (municipal) company data middle platform tag library as the lower layer realize data interoperability through SG-UEP (data exchange platform). The analysis layers of the two are connected to each other to realize information transmission between the upper and lower databases.

[0064] like Figure 5 As shown, Figure 5For the overall data tag architecture of Chongqing Power Grid Company, data access includes data replication (DRS: distributed resource scheduler), ETL (the process of extracting, converting, and loading data from the source to the destination, which requires the use of DAYU), and data exchange (SG-UEP) to obtain the data source; in the tag construction, the acquired data source is used to build the tag. First, the tag configuration is performed, including tag attribute configuration, tag wide table configuration, tag update configuration, etc. After the configuration is completed, a data tag library is established through analysis and calculation, and statistical analysis is performed on the data; in the tag service, based on the intelligent data lake operation platform, RestFul (a design style and development method for network applications that enables third parties to call mobile network resources) services are provided, while supporting rapid packaging of tag services and pre-packaging of hot tag services to achieve flexible calling of the data middle platform and unified export of power grid data; finally, the tag application provides report analysis scenarios, digital products, etc.

[0065] Throughout this specification, terms such as "one embodiment," "some embodiments," and "specific embodiments" mean that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

Claims

1. A power information management method based on building user portraits in a data center, characterized in that: The following steps are involved: S1. Unified data interaction platform, which aggregates various power business system data collected by terminal devices to the power data center; S2. Create label application scenarios based on the company's core business of marketing, production and operation; S3. In the tag application scenario, extract characteristic values of user-related data and generate corresponding user tags according to the characteristic values; S4. Construct a report scenario based on the user tags, including constructing user portraits, tag group analysis, tag lifecycle monitoring, tag online and offline management information, and tag classification statistics; S5. Establish a data label library in the data center, and the data label library provides label data services to the outside world through the data center; The specific process of generating user tags in S3 is as follows: S31. Data preparation: Synchronize data tables from the source database to DWS through DAYU for use in label generation rules and model algorithms, and complete data processing, storage, calculation, and integration. S32, wide table creation: Complete the mapping with the business data table through wide table configuration, and uniformly process the same intermediate result data used by different tags into wide table data; S33, Tag Management: Extract wide table data and create tags based on the basic information and business logic rules of tag configuration; The tag management includes the following steps: S331. Configure basic information and business logic rules for tags, where the basic information includes tag name, directory, business meaning, and implementation logic. S332, label update: Modify the basic information of the label and optimize the label configuration according to business needs. At the same time, you can deactivate the label and stop label publishing according to business status and business needs; S333, Group Analysis: Complete preliminary analysis of data tags, store and classify according to tag attributes; S334. Label publishing: The label value data obtained through analysis and calculation is published in the system for users to query and operate.

2. The power information management method based on building user portraits based on a data middle platform according to claim 1 is characterized in that: The method for constructing user portraits in S4 is as follows: S41. Tag analysis: Analyze and perform combined queries on tag values through profiling and group filtering functions; S42. Tag portrait: Generate tag application information such as tag portrait, group analysis, etc. required for the tag application scenario.

3. The power information management method based on building user portraits based on a data middle platform according to claim 1 is characterized in that: The data tag library in S5 includes tag value information, tag application information and tag management information. The tag value information includes attribute tags, fact tags, model tags and composite tags. The tag application information includes analysis subject data, portrait data and clustering data. The tag management information includes business scenario configuration data, portrait application configuration data, tag evaluation analysis data, tag directory data, tag rules, algorithm models, tag no data, tag service no data and tag sharing configuration data.

4. The power information management method based on building user portraits based on a data middle platform according to claim 1 is characterized in that: The data middle platform is configured with an API interface, and the data middle platform API provides tag data services to the outside world. The tag management information also includes tag service API information.

5. The power information management method based on building user portraits based on a data middle platform according to claim 1 is characterized in that: The data middle platform includes a source layer, a sharing layer and an analysis layer. The source layer obtains the data source, which includes customer data, operation and maintenance data and financial data. The sharing layer stores the data source in the corresponding data area and passes it to the analysis layer. The analysis layer analyzes and processes the data and establishes a data label library.

6. The power information management method based on building user portraits based on a data middle platform according to claim 5 is characterized in that: The upper-layer data middle-station analysis layer and the lower-layer data middle-station analysis layer are connected to each other through SG-UEP.

7. The power information management method based on building user portraits based on a data middle platform according to claim 1 is characterized in that: The data center is equipped with BI tools, and the report scenario is displayed and monitored using BI tools to achieve visual operations.

Images