Active power distribution network planning data processing method based on power system business data

Abstract

The invention discloses an active power distribution network planning data processing method based on power system business data. Relates to the technical field of active distribution networks. And efficient acquisition, processing, fusion, analysis and management of marketing, distribution and dispatching basic data can be realized. Comprising the following steps: 1) data acquisition and integration system construction; 2) data preprocessing and feature extraction; 3) constructing a data analysis model and method; and 4) performing data visualization and auxiliary support. According to the method, on the basis of a regional power supply department source attaching layer, data related to planning is collected and integrated, then preprocessing and feature extraction are carried out, a data analysis model is constructed, and finally a key data set for planning decision making is output. The power grid data resource can be efficiently aggregated, unified integration and unified management of city-county two-level planning data can be ensured, the basic data collection pressure can be effectively reduced, accurate and efficient data quality is ensured, labor is saved, and efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of active distribution network technology, and in particular to a method for extracting planning data from existing regional business data of power systems. Background Technology

[0002] Distribution network planning involves high precision requirements, a wide planning scope, and a large amount of planning information, resulting in a significant investment of time and effort and increasing the workload of relevant personnel. Currently, personnel engaged in distribution network planning must search, filter, analyze, and organize data related to planning from massive amounts of existing power system business data. This leads to an extremely large workload and low work efficiency.

[0003] Therefore, how local power supply system planning departments can leverage information technology to generate, clean, mine, analyze, share, and display multi-dimensional data by relying on a data platform and building customized data templates, and solve problems such as inconsistent data standards and data silos from both technical and business perspectives, has become an urgent technical issue to be addressed in this field. Summary of the Invention

[0004] To address the above-mentioned technical problems, this invention provides an active distribution network planning data processing method based on power system business data, which enables efficient collection, processing, fusion, analysis, and management of basic data from professional systems such as operation, distribution, and dispatch.

[0005] The technical solution of the present invention includes the following steps: 1) Construction of a data collection and integration system; 2) Data preprocessing and feature extraction; 3) Construction of data analysis models and methods; 4) Data visualization and auxiliary support.

[0006] Further, step 1) specifically involves: extracting grid model, real-time operation, equipment asset, and dispatch load data from existing power system regional business data; This includes offline data collection and processing, including power grid resource data, power source and user data, and historical planning project content, to construct a unified information model that conforms to the characteristics of power grid planning. By adopting multi-source data access, a governance dataset is obtained, covering data standards, quality, security, and lifecycle management.

[0007] Further, step 2) specifically includes: The governance dataset obtained in step 1) is preprocessed and its features are extracted. By addressing issues of missing, abnormal, and inconsistent data, interpolation, clustering, and / or association rule measures are used to perform automated data cleaning and repair before data preprocessing is performed, resulting in a power grid planning thematic dataset.

[0008] Further, step 3) specifically includes: The power grid planning business model is designed based on the power grid planning thematic dataset obtained in step 2), specifically as follows: 3.1) Scale-based resource search business model 3.1.1) Year-on-year growth rate of peak electricity load for the whole society = (current period value - same period last year value) / same period last year value * 100%; 3.1.2) Regional step-down load = Regional incoming load - Regional direct supply load; Regional capacity ratio = Regional step-down load / Sum of regional substation capacity; 3.1.3) Utilization hours of power supply connection = Annual power generation at different voltage levels / Annual installed capacity; 3.1.4) Power Grid Status (Substation / Main Transformer / Transformer Capacity / 10kV Bay): Public and Private = Calculated and summed according to asset attribute classification; Line Length = Overhead Length + Cable Length; Cableization Rate = Cable Length / Line Length * 100%; 3.2) Project resource search business model 3.2.1) Installed capacity of power supply projects of 110 kV and below = calculated and summed according to different access voltage levels; 3.2.2) Installed capacity of energy storage projects of 110 kV and below = calculated and summed according to different access voltage levels; 3.2.3) Installed capacity of 110 kV and below user projects = calculated and summed according to different access voltage levels; 3.3) Run the data search business model; 3.3.1) Calculate the maximum load rate of each main transformer at any point in the year. Specifically: the number of transformers with the maximum load rate distribution of 35 / 110 kV main transformers = summation according to different distribution intervals; Calculate the maximum load rate of each main transformer at any point in the year, specifically: the maximum load rate distribution of the 35 / 110 kV main transformer = sum of different distribution intervals / calculation of the total number of main transformers; 3.3.2) Calculate the average load rate of each main transformer in the previous year, specifically: the average load rate of the 35 / 110 kV main transformer distribution = summation according to different distribution intervals; The average load rate of the main transformer = the average annual load of the transformer / the transformer capacity; The average annual load of the transformer = the amount of electricity stepped down by the main transformer / the total number of hours of operation throughout the year; Total average load factor = Sum of annual average loads of all transformers in the region / Sum of capacities of all transformers in the region; 3.3.3) The load rate of each main transformer at the time of year in reverse is as follows: whether the 35 / 110 kV and heavily loaded main transformers are reverse-loaded = calculate the continuous heavy load for more than 2 hours throughout the year; 3.3.4) 35 / 110 kV and main transformer load imbalance = load imbalance according to different statistical methods; Load imbalance ; In the formula, Bs represents the load imbalance degree, and L... S_i For transformer load rate, is the average transformer load rate in the region, N is the number of substations, and s is the sample point.

[0009] Further, step 4) specifically includes: Based on the power grid planning business analysis model in step 3), and combined with the development tools and visualization platform, various power grid planning business analysis models are developed, demonstrated, and the accuracy of the data is verified.

[0010] Furthermore, the existing power system regional business data, including automatically collected PMS3.0 data, dispatch D5000 data, and marketing data, reflects the power grid model, real-time operation, equipment assets, and dispatch load.

[0011] This invention is based on the regional power supply department's source layer (including PMS data, dispatch data, marketing data, etc.). First, it collects and integrates planning-related data, then performs preprocessing and feature extraction, then builds a data analysis model, and finally outputs a key dataset for planning decisions.

[0012] This invention, through research and analysis, forms standardized data processing and management methods, which can efficiently aggregate power grid data resources, ensure the unified integration and management of planning data at the municipal and county levels, effectively reduce the data collection pressure at the grassroots level, ensure accurate and efficient data quality, save manpower, and improve efficiency. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a schematic diagram illustrating the working principle of the present invention. Figure 2 This is a flowchart of the process of the present invention. Detailed Implementation

[0015] The following is in conjunction with the appendix Figure 1 , 2 The technical solution of the present invention will be further illustrated through specific embodiments.

[0016] This invention relates to an active distribution network planning data processing method based on power system business data, comprising the following steps: 1) Construction of a data collection and integration system; 2) Data preprocessing and feature extraction; 3) Construction of data analysis models and methods; 4) Data visualization and auxiliary support.

[0017] Further, step 1) specifically involves: extracting grid model, real-time operation, equipment asset, and dispatch load data from existing power system regional business data; This includes offline data collection and processing, including power grid resource data, power source and user data, and historical planning project content, to construct a unified information model that conforms to the characteristics of power grid planning. By adopting multi-source data access, a governance dataset is obtained, covering data standards, quality, security, and lifecycle management.

[0018] By adopting the above technical measures, firstly, the data sources are covered in both directions, which not only clearly traces the source of the power grid business data sources, but also clearly analyzes the offline data collection and processing needs, thus ensuring the comprehensiveness of the data; Second, the data content focuses on the core of the plan, covering key data from multiple dimensions such as equipment assets, real-time operation, dispatch load, and power grid resources; Third, data processing is protected by two safeguards: data standardization is achieved by clarifying the data model of the data source, and data compliance and usability are ensured by multi-dimensional access combined with data governance rules.

[0019] Further, step 2) specifically includes: The governance dataset obtained in step 1) is preprocessed and its features are extracted. By addressing issues of missing, abnormal, and inconsistent data, interpolation, clustering, and / or association rule measures are used to perform automated data cleaning and repair before data preprocessing is performed, resulting in a power grid planning thematic dataset.

[0020] In this step, various characteristic data are extracted, including power grid resource data such as the ledgers of core equipment covering 220kV to 10kV substations, lines, main transformers, and switches; daily, weekly, and annual load curve characteristic values, load growth rate, simultaneity rate, and peak-valley difference are extracted from load data; spatiotemporal feature extraction is performed by extracting the spatial attribute characteristics of grids and cells from geographical data; equipment status characteristics are extracted by calculating equipment load rate, frequency and duration of heavy overload, and average service life based on operational data; power data such as the installed capacity and output of various power sources (especially distributed photovoltaics) are collected; and planning projects are extracted based on power sources / energy storage / user projects connected to different voltage levels. These are integrated to form a power grid planning thematic database, which serves as the basis for all analyses.

[0021] This step begins by clarifying the data source support. Step 1) integrates existing business data (equipment assets, real-time operation, scheduling load, etc.) and offline processing data (power grid resources, power projects, etc.) to provide a complete preprocessing object for Step 2.

[0022] Secondly, to solidify the foundation of data quality, step 1) establishes standardized models and data governance rules (types, standards, quality, etc.), which reduces redundant work in data cleaning and standardization in step 2 preprocessing and improves processing efficiency.

[0023] Third, focusing on the core needs of planning, step 1 constructs data rules according to the characteristics of power grid planning, so that the feature extraction in step 2 can accurately plan the key data dimensions required and avoid invalid feature screening.

[0024] Further, step 3) specifically includes: The power grid planning thematic dataset obtained in step 2) is used to design a power grid planning business model. Based on the data preprocessing objects and feature extraction rules in step 2), and combined with the core requirements of power grid planning, various input-output analysis models are designed. Specifically: 3.1) Scale-based resource search business model 3.1.1) Year-on-year growth rate of peak electricity load for the whole society = (current period value - same period last year value) / same period last year value * 100%; 3.1.2) Regional step-down load = Regional incoming load - Regional direct supply load; Regional capacity ratio = Regional step-down load / Sum of regional substation capacity; 3.1.3) Utilization hours of power supply connection = Annual power generation at different voltage levels / Annual installed capacity; 3.1.4) Power Grid Status (Substation / Main Transformer / Transformer Capacity / 10kV Bay): Public and Private = Calculated and summed according to asset attribute classification; Line Length = Overhead Length + Cable Length; Cableization Rate = Cable Length / Line Length * 100%; 3.2) Project resource search business model 3.2.1) Installed capacity of power supply projects of 110 kV and below = calculated and summed according to different access voltage levels; 3.2.2) Installed capacity of energy storage projects of 110 kV and below = calculated and summed according to different access voltage levels; 2.2.3) Installed capacity of 110 kV and below user projects = calculated and summed according to different access voltage levels; 3.3) Run the data search business model; 3.3.1) Calculate the maximum load rate of each main transformer at any point in the year. Specifically: the number of transformers with the maximum load rate distribution of 35 / 110 kV main transformers = summation according to different distribution intervals; Calculate the maximum load rate of each main transformer at any point in the year, specifically: the maximum load rate distribution of the 35 / 110 kV main transformer = sum of different distribution intervals / calculation of the total number of main transformers; 3.3.2) Calculate the average load rate of each main transformer in the previous year, specifically: the average load rate of the 35 / 110 kV main transformer distribution = summation according to different distribution intervals; The average load rate of the main transformer = the average annual load of the transformer / the transformer capacity; The average annual load of the transformer = the amount of electricity stepped down by the main transformer / the total number of hours of operation throughout the year; Total average load factor = Sum of annual average loads of all transformers in the region / Sum of capacities of all transformers in the region; 3.3.3) The load rate of each main transformer at the time of year in reverse is as follows: whether the 35 / 110 kV and heavily loaded main transformers are reverse-loaded = calculate the continuous heavy load for more than 2 hours throughout the year; 3.3.4) 35 / 110 kV and main transformer load imbalance = load imbalance according to different statistical methods; Load imbalance ; In the formula, Bs represents the load imbalance degree, and L... S_i For transformer load rate, is the average transformer load rate in the region, N is the number of substations, and s is the sample point.

[0025] This step analyzes and solves the construction of data analysis models and methods. It is divided into two parts: load forecasting and power matching analysis models, which employ machine learning algorithms and combine multi-dimensional characteristics such as land use, economic population, and historical load to conduct refined load analysis at the city / county level. Distributed power absorption capacity analysis establishes a time-series production simulation model to analyze the output characteristics of distributed photovoltaic and wind power under different penetration rates and their impact on power flow, voltage, and network losses, assessing the maximum absorption capacity of the regional power grid. The power grid carrying capacity and adaptability assessment model uses power supply capacity assessment based on grid structure and equipment reliability parameters to calculate power supply reliability indicators for different regions, accurately identifying weak links. N-1 pass rate and transfer capacity analysis automatically performs power flow calculations and static security analysis to verify the compliance of the planned grid with safety criteria and quantitatively assess load transfer capacity, constructing a core business analysis and research model that conforms to power grid planning.

[0026] The significance of calculating the load imbalance of main transformers goes far beyond a single number; it is a core manifestation of refined operation and management of the power grid. Obtaining the load imbalance from dynamic operational data has the following practical implications for the distribution network: The first priority is to ensure the safety of the power grid and equipment; Preventing overload of a single main transformer: When the load is severely unbalanced, the main transformer with the highest load rate may be the first to approach or exceed its stable overload capacity, which may lead to accelerated insulation aging and excessively high hot spot temperature, and in severe cases, may cause the transformer to burn out.

[0027] To avoid N-1 test failure: The important N-1 criterion in power grid planning requires that when one main transformer fails and is taken out of service, the remaining main transformers should be able to handle its load without overloading. If the imbalance is large, it is very likely that in the N-1 case, one of the remaining main transformers will be severely overloaded, leading to a cascading failure.

[0028] Secondly, optimizing the economic operation of the power grid can directly generate economic benefits; Reducing grid losses: Transformer losses (copper losses) are proportional to the square of the load current. When the load is unbalanced, the losses of the main transformer under high load will increase exponentially, while the losses of the main transformer under low load will decrease only slightly, resulting in the total active power loss of the entire system being greater than the losses under balanced load. Reducing the degree of imbalance is an effective means of reducing losses and increasing efficiency.

[0029] Improving equipment utilization efficiency: Imbalance means that some main transformer capacity is idle, while others are nearing their limits. Balancing the load can free up capacity from low-load main transformers, improving the overall power supply capacity of the region without additional investment.

[0030] Thirdly, it is crucial to improve the reliability and quality of power supply, which is also the key to the social significance of this case. Reduce the risk of power outages: When a heavily loaded main transformer needs to be shut down for maintenance, it may not be able to transfer all the load due to its excessive load, resulting in power outages for users.

[0031] Maintaining stable voltage: The voltage drop of the main transformer under heavy load is large, which may lead to low voltage on its low-voltage side bus, affecting power quality.

[0032] Fourth, it can guide power grid planning and renovation; Scientific Investment Basis: Load imbalance is a crucial reference for power grid planning. If the load imbalance remains consistently high in a region, it indicates problems with the grid structure or load distribution. Planners can then decide whether to prioritize upgrading heavily loaded main transformers or constructing new interconnection lines and adjusting operating modes to balance the load, rather than simply blindly adding new main transformer capacity.

[0033] Fifth, optimize the scheduling and operation mode; Dispatchers can proactively adjust the power grid operation mode based on real-time or predicted load imbalance. For example, by switching operations, some load can be transferred from heavily loaded main transformers to lightly loaded main transformers, so that the main transformer group can operate in the best condition.

[0034] Therefore, for main transformers in power systems, load imbalance is no longer a simple performance indicator, but a crucial safety, economic, and management indicator. Monitoring and optimizing the load imbalance of main transformers in power systems is an essential requirement for achieving "safe, economical, high-quality, and efficient" operation of the power grid, and a key link in the lean management of modern smart grids.

[0035] Further, step 4) specifically includes: Based on the power grid planning business analysis model in step 3), and combined with the development tools and visualization platform, various power grid planning business analysis models are developed, demonstrated, and the accuracy of the data is verified.

[0036] This step analyzes and addresses data visualization and support, automatically generating analytical reports on the scale, load levels, equipment lifespan distribution, grid structure rationality, and power supply safety standard verification of power grids at all levels through precise power grid status diagnosis and visualization. This changes the past practice of manual collection and statistics, enabling "one-click generation" of planning data reports, allowing planners to quickly, comprehensively, and accurately assess the current state of the power grid and identify weak points. Utilizing gridded load forecasting and spatial load distribution based on historical load data and regional development plans (such as land use and major projects), algorithmic models are used for medium- and long-term gridded and spatial load forecasting. This refines load forecasting from a single total to each power supply grid / unit, providing precise spatial basis for substation placement and line corridor planning, avoiding "over-investment" or "insufficient capacity." By analogy with the full lifecycle management and investment optimization of the project database, a dynamic correlation is established between the project database and power grid diagnostic results. Based on dimensions such as the urgency of the problem, the effectiveness of the project solution, and the investment cost, a project priority evaluation model can be constructed to assist in investment decisions and generate project ranking suggestions. This transforms investment decisions from "gut feeling" to "data-driven" ones, ensuring that limited investment is prioritized to solve the most urgent and beneficial power grid problems, thereby improving investment efficiency. A unified planning and visualization platform for the distribution network is also constructed to provide effective support for decision-making.

[0037] Furthermore, the existing power system regional business data, including automatically collected PMS3.0 data, dispatch D5000 data, and marketing data, reflects the power grid model, real-time operation, equipment assets, and dispatch load.

[0038] It should be stated that the above-described specific embodiments are merely preferred embodiments of the present invention and the technical principles employed. Those skilled in the art should understand that various modifications, equivalent substitutions, and variations can be made to the present invention based on the technical content disclosed in this application. However, such variations, as long as they do not depart from the spirit of the present invention, should be within the protection scope of the present invention. Furthermore, some terminology used in this specification and claims is not limiting but merely for ease of description.

Claims

1. A proactive power distribution network planning data processing method based on power system business data, characterized in that, It comprises the following steps: 1) data acquisition and integration system construction; 2) data preprocessing and feature extraction; 3) data analysis model and method construction; 4) data visualization and auxiliary support.

2. The active power system service data based distribution network planning data processing method according to claim 1, wherein, Step 1) is specifically extracting power grid model, real-time operation, equipment asset and dispatching load data from existing power system regional business data; and offline data collection and processing, including power grid resource data, power supply and user data and historical planning project content, to construct a unified information model conforming to the characteristics of power grid planning, adopting multi-source data access to obtain governance data sets, covering data standards, quality, security and life cycle management.

3. The active power system service data based distribution network planning data processing method of claim 1, wherein, Step 2) is specifically: data preprocessing and feature extraction are performed on the governance data sets obtained in step 1), and through interpolation, clustering and / or association rule measures, automatic data cleaning and repair are performed for data preprocessing to solve the problems of data missing, abnormality and inconsistency; obtaining power grid planning special data sets.

4. The proactive power system service data based distribution network planning data processing method of claim 1, wherein, Step 3) is specifically: power grid planning business model design is performed on the power grid planning special data sets obtained in step 2), specifically: 3.1) scale search business model, 3.1.1) year-on-year growth rate of highest load of whole society = (current value - last year's same period value) / last year's same period value * 100%; 3.1.2) regional voltage reduction load = regional incoming load - regional direct supply load; regional load capacity ratio = regional voltage reduction load / sum of regional transformer capacity; 3.1.3) utilization hours of connected power supply = annual power generation according to different voltage levels / annual installed capacity; 3.1.4) power grid situation (substation / main transformer / transformer capacity / 10 kV interval) public and private = calculated and summed according to asset attribute classification; line length = overhead length + cable length; cable rate = cable length / line length * 100%; 3.2) project search business model 3.2.1) 110 kV and below power supply project installed capacity = calculated and summed according to connection to different voltage levels; 3.2.2) 110 kV and below energy storage project installed capacity = calculated and summed according to connection to different voltage levels; 3.2.3) 110 kV and below user project installed capacity = calculated and summed according to connection to different voltage levels; 3.3) operation data search business model; 3.3.1) calculate the maximum load rate of each main transformer at all times, specifically: 35 / 110 kV main transformer load rate distribution maximum load rate number = sum according to different distribution intervals; calculate the maximum load rate of each main transformer at all times, specifically: 35 / 110 kV main transformer load rate distribution maximum load rate proportion = sum according to different distribution intervals / calculate the total number of main transformers; 3.3.2) calculate the average load rate of each main transformer in the last year, specifically: 35 / 110 kV main transformer load rate distribution average load rate number = sum according to different distribution intervals; main transformer average load rate = transformer annual average load / transformer capacity; transformer annual average load = main transformer voltage reduction power / annual utilization hours; Full caliber average load rate = the sum of the annual average load of all transformers in the region / the sum of the capacity of all transformers in the region; 3.3.3), the load rate of each main transformer in reverse at all times of the year, specifically: 35 / 110 kV and heavy load main transformer whether to send heavy load = calculate the full year continuous heavy load for more than 2 hours; 3.3.4), 35 / 110 kV and main transformer load imbalance = load imbalance according to different statistical caliber; Load imbalance ； In the formula, Bs is the load imbalance degree, L S_i is the transformer load rate, is the average value of transformer load rate in the region, N is the number of substations, and s is the sample point.

5. The proactive power system service data based distribution network planning data processing method of claim 1, wherein, Said step 4) is specifically: Based on the power grid planning business analysis model of step 3), combined with the development tool, the visual platform is used to develop various types of power grid planning business analysis model demonstration verification data real accuracy.

6. The proactive power system service data based distribution network planning data processing method of claim 1, wherein, Among the existing power system regional business data, including automatic collection PMS3.0 data, dispatching D5000 data and marketing data, the data reflecting the power grid model, real-time operation, equipment assets and dispatching load.

Images