Power distribution network cloud platform data fusion method

Abstract

The invention discloses a power distribution network cloud platform data fusion method. The method comprises the following steps: S1, analyzing an existing data source: identifying a required data type, format, quality and data updating frequency; s2, data acquisition and access: designing a data access interface to ensure that data of different service subsystems can be effectively accessed by the cloud platform; the data is collected through timing collection and event triggering collection, and the timeliness and integrity of the data are ensured; s3, data preprocessing: cleaning the data, and removing invalid, wrong or incomplete data records; s4, data classification and association; s5, data fusion: realizing a data fusion algorithm, performing comprehensive processing of multi-source data, and generating a fused data set; s6, spatio-temporal data mapping and fusion; and S7, data storage and management. According to the invention, the power distribution network cloud platform can effectively integrate multi-source heterogeneous data, and provides powerful data support for intelligent management and decision-making of the power distribution network.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network data fusion, and in particular to a distribution network cloud platform data fusion method. Background Art

[0002] The distribution network is directly facing users. It is a key link to ensure the quality of power supply, improve the efficiency of power grid operation, and provide users with high-quality services. Realizing a high degree of monitoring of the distribution network is the development trend of power companies. In order to facilitate the supervision of distribution network data, information integration technology and Internet of Things technology are adopted. With the help of big data analysis and mining, the data resources of the distribution and dispatching system are integrated to build a distribution network cloud platform. It provides information and decision-making support for engineering projects such as distribution network operation monitoring, power outage management, and fault repair command, and improves the service level of the power grid. The distribution network cloud platform is guided by proactive risk prevention, scientific production management, and deepening management status. With SOA as the core idea, through the service component architecture, service data objects and business process execution language, the three technologies solve the problems in the fields of service construction, data model and process orchestration of the distribution network cloud platform respectively. Through the integrated production and operation management support system, the full integration of relevant equipment information, power grid model and spatial information required for distribution production is realized, and the process of distribution network operation and maintenance data management is comprehensively promoted in terms of technology.

[0003] The existing publication number CN117272221A discloses a data fusion method and a distribution network cloud platform, which include the following steps: Step 1, input the distribution network data into the distribution network cloud platform through an input port, and the distribution network cloud platform is provided with multiple interactive ports. When the input data is large, multiple input ports simultaneously input data. The data fusion method and the distribution network cloud platform, the distribution network cloud platform shunts the cleaned distribution network data, so that the overall operation burden of the distribution network cloud platform is reduced, and the distribution network cloud platform is prevented from crashing due to the excessive amount of distribution network data processed. The distribution network cloud platform is set with a program that refuses to accept incomplete distribution network data, so as to reduce the workload of the distribution network cloud platform, so that the operation burden of the distribution network cloud platform is reduced. The distribution network cloud platform constructs the distribution network data into a statistical chart, which is convenient for intuitively viewing the distribution network data.

[0004] Although the existing technology has achieved data fusion, it cannot fuse multi-source heterogeneous data, so the fused data is relatively single and cannot provide good support for the subsequent management of the distribution network. For this reason, we propose a data fusion method for the distribution network cloud platform. Summary of the invention

[0005] The purpose of the present invention is to provide a distribution network cloud platform data fusion method to solve the problems in the prior art.

[0006] To achieve the above object, the present invention provides the following technical solution: a distribution network cloud platform data fusion method, comprising the following steps:

[0007] S1. Analyze existing data sources: identify the required data type, format, quality and data update frequency;

[0008] S2. Data collection and access: Design data access interfaces to ensure that data from different business subsystems can be effectively accessed by the cloud platform; collect data through scheduled collection and event-triggered collection to ensure the timeliness and integrity of the data;

[0009] S3. Data preprocessing: clean the data and remove invalid, erroneous or incomplete data records;

[0010] S4. Data classification and association: classify data and establish data association model;

[0011] S5, data fusion: implement data fusion algorithm, conduct comprehensive processing of multi-source data, and generate fused data sets;

[0012] S6, spatiotemporal data mapping and fusion: Process spatiotemporal data to ensure the consistency of data in spatiotemporal dimensions;

[0013] S7. Data storage and management: Use distributed databases or data lakes to store large amounts of fused data;

[0014] S8. Data analysis and application: Use the fused data for analysis and combine it with AI algorithms for prediction;

[0015] S9. Evaluation and optimization: Regularly evaluate the effectiveness of data fusion, adjust the fusion strategy and algorithm based on the evaluation results, and optimize the data fusion process.

[0016] Preferably, the data processing in S3 also includes data standardization processing: unifying the data format to ensure comparability between data.

[0017] Preferably, the data processing in S3 also includes data quality assessment: checking the accuracy, completeness and consistency of the data, and performing missing value filling or outlier processing.

[0018] Preferably, the classification of data in S4 includes the source, type and geographical location of the data; data association is to establish a data association model to identify the logical and physical connections between different data sources.

[0019] Preferably, the data fusion in S5 selects one or more of weighted averaging, Kalman filtering or deep learning models according to data characteristics and application requirements.

[0020] Preferably, the spatiotemporal data processing in S6 includes spatiotemporal data interpolation, spatiotemporal data matching, and application of a spatiotemporal grid method to improve the spatial and temporal resolution of the data.

[0021] Preferably, the management of data in S7 includes access permission control, backup and recovery, and version management of data.

[0022] Preferably, the data analysis in S8 includes trend analysis and fault pattern recognition; the prediction performed by the AI ​​algorithm includes load prediction, equipment health assessment, and intelligent scheduling.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] 1. The distribution network cloud platform can effectively integrate multi-source heterogeneous data and provide strong data support for intelligent management and decision-making of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0026] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention.

[0028] See also Figure 1 In an embodiment of the present invention, a distribution network cloud platform data fusion method includes the following steps:

[0029] S1. Analyze existing data sources: identify the required data type, format, quality and data update frequency;

[0030] S2. Data collection and access: Design data access interfaces to ensure that data from different business subsystems can be effectively accessed by the cloud platform; collect data through scheduled collection and event-triggered collection to ensure the timeliness and integrity of the data;

[0031] S3. Data preprocessing: clean the data and remove invalid, erroneous or incomplete data records;

[0032] S4. Data classification and association: classify data and establish data association model;

[0033] S5, data fusion: implement data fusion algorithm, conduct comprehensive processing of multi-source data, and generate fused data sets;

[0034] S6, spatiotemporal data mapping and fusion: Process spatiotemporal data to ensure the consistency of data in spatiotemporal dimensions;

[0035] S7. Data storage and management: Use distributed databases or data lakes to store large amounts of fused data;

[0036] S8. Data analysis and application: Use the fused data for analysis and combine it with AI algorithms for prediction;

[0037] S9. Evaluation and optimization: Regularly evaluate the effectiveness of data fusion, adjust the fusion strategy and algorithm based on the evaluation results, and optimize the data fusion process.

[0038] Preferably, the data processing in S3 also includes data standardization processing: unifying the data format to ensure comparability between data.

[0039] Preferably, the data processing in S3 also includes data quality assessment: checking the accuracy, completeness and consistency of the data, and performing missing value filling or outlier processing.

[0040] Preferably, the classification of data in S4 includes the source, type and geographical location of the data; data association is to establish a data association model to identify the logical and physical connections between different data sources.

[0041] Preferably, the data fusion in S5 selects one or more of weighted averaging, Kalman filtering or deep learning models according to data characteristics and application requirements.

[0042] Preferably, the spatiotemporal data processing in S6 includes spatiotemporal data interpolation, spatiotemporal data matching, and application of a spatiotemporal grid method to improve the spatial and temporal resolution of the data.

[0043] Preferably, the management of data in S7 includes access permission control, backup and recovery, and version management of data.

[0044] Preferably, the data analysis in S8 includes trend analysis and fault pattern recognition; the prediction performed by the AI ​​algorithm includes load prediction, equipment health assessment, and intelligent scheduling.

[0045] Clarify the goals and application scenarios of data fusion, such as condition monitoring, fault prediction, load management, etc. Analyze existing data sources and identify the required data type, format, quality, and data update frequency.

[0046] Data collection and access: Design data access interfaces to ensure that data from different business subsystems (such as SCADA, DMS, GIS, AMR, etc.) can be effectively accessed by the cloud platform. Implement data collection strategies, including timed collection, event-triggered collection, etc., to ensure the timeliness and integrity of data.

[0047] 3. Data preprocessing: Data cleaning: Eliminate invalid, erroneous or incomplete data records. Data standardization: Unify data formats, such as timestamp conversion, dimension conversion, etc., to ensure comparability between data. Data quality assessment: Check the accuracy, completeness and consistency of data, and fill in missing values ​​or handle outliers when necessary.

[0048] Data classification and association: Classify data, for example, by data source, type (such as electrical parameters, environmental parameters), geographic location, etc. Establish a data association model to identify the logical and physical connections between different data sources, laying the foundation for subsequent integration.

[0049] Data fusion method selection and implementation: Select appropriate fusion methods according to data characteristics and application requirements, such as weighted average, Kalman filter, deep learning model, etc. Implement data fusion algorithm, conduct comprehensive processing of multi-source data, and generate fused data sets.

[0050] Spatiotemporal data mapping and fusion: Processing of spatiotemporal data, such as spatiotemporal data interpolation and spatiotemporal data matching, ensures the consistency of data in spatiotemporal dimensions. Apply spatiotemporal grid methods or other spatiotemporal fusion technologies to improve the spatial and temporal resolution of data.

[0051] Data storage and management: Design data storage architecture and use distributed databases or data lakes to store large amounts of integrated data. Implement data management strategies to ensure data access control, backup and recovery, version management, etc.

[0052] Data analysis and application: Use the fused data for advanced analysis, such as trend analysis, fault pattern recognition, and optimized decision making. Combine AI algorithms for prediction and decision support, such as load forecasting, equipment health assessment, and intelligent scheduling.

[0053] Performance evaluation and optimization: Regularly evaluate the effect of data fusion, including data quality, fusion efficiency, application effect, etc. Adjust the fusion strategy and algorithm based on the evaluation results and continuously optimize the data fusion process.

[0054] The distribution network cloud platform can effectively integrate multi-source heterogeneous data and provide strong data support for intelligent management and decision-making of the distribution network.

[0055] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A data fusion method for a distribution network cloud platform, characterized in that: The following steps are involved: S1. Analyze existing data sources: identify the required data type, format, quality and data update frequency; S2. Data collection and access: Design data access interfaces to ensure that data from different business subsystems can be effectively accessed by the cloud platform; collect data through scheduled collection and event-triggered collection to ensure the timeliness and integrity of the data; S3. Data preprocessing: clean the data and remove invalid, erroneous or incomplete data records; S4. Data classification and association: classify data and establish data association model; S5, data fusion: implement data fusion algorithm, conduct comprehensive processing of multi-source data, and generate fused data sets; S6, spatiotemporal data mapping and fusion: Process spatiotemporal data to ensure the consistency of data in spatiotemporal dimensions; S7. Data storage and management: Use distributed databases or data lakes to store large amounts of fused data; S8. Data analysis and application: Use the fused data for analysis and combine it with AI algorithms for prediction; S9. Evaluation and optimization: Regularly evaluate the effectiveness of data fusion, adjust the fusion strategy and algorithm based on the evaluation results, and optimize the data fusion process.

2. A distribution network cloud platform data fusion method according to claim 1, characterized in that: The data processing in S3 also includes data standardization processing: unifying the data format to ensure comparability between data.

3. A distribution network cloud platform data fusion method according to claim 1 or 2, characterized in that: The data processing in S3 also includes data quality assessment: checking the accuracy, completeness and consistency of the data, and filling in missing values ​​or handling outliers.

4. A distribution network cloud platform data fusion method according to claim 1, characterized in that: The classification of data in S4 includes the source, type and geographical location of the data; Data association is to establish a data association model and identify the logical and physical connections between different data sources.

5. A distribution network cloud platform data fusion method according to claim 1, characterized in that: The data fusion in S5 selects one or more of weighted averaging, Kalman filtering or deep learning models according to data characteristics and application requirements.

6. A distribution network cloud platform data fusion method according to claim 1, characterized in that: The spatiotemporal data processing in S6 includes spatiotemporal data interpolation, spatiotemporal data matching, and application of spatiotemporal grid method to improve the spatial and temporal resolution of the data.

7. A distribution network cloud platform data fusion method according to claim 1, characterized in that: The management of data in S7 includes data access permission control, backup and recovery, and version management.

8. A distribution network cloud platform data fusion method according to claim 1, characterized in that: The data analysis in S8 includes trend analysis and fault pattern recognition; the prediction performed by the AI ​​algorithm includes load prediction, equipment health assessment, and intelligent scheduling.

Images