Distributed photovoltaic-containing rural power distribution network adaptability planning method and system

Abstract

The invention discloses an adaptive planning method and system for a rural power distribution network containing distributed photovoltaics, and relates to the technical field of adaptive planning, and the method comprises the steps: carrying out the fault simulation based on an optimal photovoltaic access scheme, collecting fault data corresponding to a fault event generated through simulation in a fault simulation process, and carrying out the fault simulation according to the fault data; the dynamic operation risk of the power distribution network is evaluated; and based on the characteristics of the rural power distribution network and the dynamic operation risk of the power distribution network, performing multi-objective optimization, and generating a distributed photovoltaic access planning scheme. According to the invention, multi-source data is collected and preprocessed to generate a high-quality data set, and photovoltaic output and rural power distribution network characteristics are obtained, so that full-process automatic processing from data collection to power grid characteristic extraction is realized.

Description

Technical Field

[0001] This invention relates to the field of adaptive planning technology, and in particular to an adaptive planning method and system for rural power distribution networks including distributed photovoltaic power. Background Technology

[0002] With the accelerated transformation of the global energy structure, distributed photovoltaic (PV) power generation technology has become an important form of renewable energy utilization. In recent years, the penetration rate of PV power generation in the distribution networks of rural areas in my country has continued to increase, and the large-scale integration of distributed PV has become a key path to promote the rural energy revolution. However, traditional distribution network planning methods are mainly designed for centralized power sources and unidirectional power flow characteristics, and still face many challenges in rural distribution network scenarios. For example, existing methods do not adequately assess the vulnerability of the distribution network after PV integration, making it difficult to accurately identify cascading failure risk points caused by topological defects. Furthermore, they do not fully consider the differentiated renewable energy integration needs of rural users (such as the continuous power supply requirements for agricultural irrigation loads and the tolerance for short-term power outages for residential users), leading to discrepancies between planning schemes and actual power supply reliability requirements. Further optimization and improvement are urgently needed.

[0003] The main shortcomings of existing technologies are: First, they lack a multi-objective optimization framework that integrates grid topology vulnerability assessment with differentiated user needs, resulting in insufficient adaptability and power supply reliability of planning schemes under fault scenarios. Second, they fail to fully consider the unique topology vulnerability characteristics of rural distribution networks, resulting in insufficient robustness of photovoltaic access schemes under fault scenarios. In addition, existing planning methods are mostly based on deterministic scenario analysis, making it difficult to comprehensively consider the operational risks under the coupled effects of multiple factors such as the uncertainty of meteorological conditions, the spatiotemporal heterogeneity of loads, and the dynamic aging of equipment. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides an adaptive planning method for rural power distribution networks with distributed photovoltaic power to address the problems of insufficient adaptability and power supply reliability in fault scenarios and the difficulty in accurately quantifying operational risks in existing technologies.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides an adaptive planning method for rural power distribution networks including distributed photovoltaic (PV) power. The method includes: collecting multi-source data and preprocessing it to generate a high-quality dataset, and obtaining PV output and rural power distribution network characteristics; generating a multi-objective balanced PV access scheme based on the PV output and the high-quality dataset; obtaining a comprehensive evaluation priority ranking result of the PV access scheme based on the multi-objective balanced PV access scheme, combined with the rural user's renewable energy access needs and rural user power supply reliability model, and performing a comprehensive evaluation to select the optimal PV access scheme; performing fault simulation based on the optimal PV access scheme, and collecting fault data corresponding to the simulated fault events during the fault simulation process, while simultaneously evaluating the dynamic operation risk of the power distribution network; and performing multi-objective optimization based on the rural power distribution network characteristics and the dynamic operation risk of the power distribution network to generate a distributed PV access planning scheme.

[0007] As a preferred embodiment of the adaptive planning method for rural power distribution networks with distributed photovoltaic power as described in this invention, the multi-source data includes meteorological data, photovoltaic parameters, power grid topology, equipment parameters, power grid operation data, and user load data. The preprocessing includes data cleaning, format conversion, deduplication, normalization, and outlier handling.

[0008] As a preferred embodiment of the adaptive planning method for rural power distribution networks including distributed photovoltaic power as described in this invention, the specific steps for obtaining photovoltaic power output and rural power distribution network characteristics are as follows: Calculate photovoltaic output based on high-quality datasets; Based on high-quality datasets, load distribution characteristics and photovoltaic absorption matching characteristics of rural power distribution networks are extracted; Based on the load distribution characteristics and photovoltaic power consumption matching characteristics, network connectivity and weak links are assessed to obtain the characteristics of rural power distribution networks.

[0009] As a preferred embodiment of the adaptive planning method for rural power distribution networks including distributed photovoltaic power as described in this invention, the specific steps for generating a multi-objective balanced photovoltaic grid connection scheme based on photovoltaic output and high-quality datasets are as follows. Based on high-quality datasets, obtain the physical constraints of the power grid; Based on photovoltaic output and high-quality datasets, the total cost of photovoltaic grid connection is minimized, and the combination of connected photovoltaic nodes and capacity is obtained to generate a preliminary grid connection scheme that meets the physical constraints of the power grid. Based on the initial grid connection scheme, a multi-objective balanced photovoltaic grid connection scheme is generated under the condition of satisfying the physical constraints of the power grid.

[0010] As a preferred embodiment of the adaptive planning method for rural power distribution networks containing distributed photovoltaic power as described in this invention, the photovoltaic access scheme based on multi-objective balance combines the renewable energy access needs of rural users and the power supply reliability model of rural users to obtain a comprehensive evaluation of the photovoltaic access scheme priority ranking results, and performs a comprehensive evaluation to select the optimal photovoltaic access scheme. The specific steps are as follows: Through questionnaires and historical data analysis, we obtained the demand for new energy access for rural users and the impact of new energy on the reliability of power supply for rural users. At the same time, we obtained the differentiated needs of rural users for power supply reliability and formed a demand list. High-quality datasets are used as a framework for multi-dimensional evaluation metrics, and reliability quantification targets are set based on the requirements list. Based on a multi-dimensional evaluation index framework and reliability quantification objectives, an initial rural user power supply reliability model was constructed, and a BP neural network was used to train and finally construct the rural user power supply reliability model. The multi-objective balanced photovoltaic access scheme is input into the rural user power supply reliability model to predict the power supply reliability of each photovoltaic access scheme, and it is matched with the new energy access needs of rural users to evaluate the priority of photovoltaic access schemes and rank them. Based on the priority ranking of photovoltaic access schemes, the proximity of each photovoltaic access scheme to the optimal and worst values ​​is calculated, and the optimal photovoltaic access scheme is selected according to the proximity.

[0011] As a preferred embodiment of the adaptive planning method for rural power distribution networks including distributed photovoltaic power as described in this invention, the steps of performing fault simulation based on the optimal photovoltaic access scheme, collecting fault data corresponding to simulated fault events during the fault simulation process, and simultaneously assessing the dynamic operational risks of the power distribution network are as follows: Simulate random fault scenarios under the optimal photovoltaic access scheme, repeatedly randomly extract fault features, analyze the dynamic propagation process of each fault, obtain a comprehensive simulation dataset containing the dynamic characteristics of the entire fault process, and record the fault data of each simulation. By analyzing the timestamps of fault data, we can identify the frequency distribution of faults in different time periods, identify high-incidence periods and periodic patterns of faults, and generate the occurrence pattern of fault events in the time dimension. Obtain the spatial clustering patterns of hotspot areas distributed in the comprehensive simulation dataset, and generate the spatial distribution characteristics of fault events; Based on the occurrence patterns of fault data in the time dimension and the distribution characteristics in the spatial dimension, the spatiotemporal correlation between fault probability and impact range is calculated, thereby obtaining the dynamic operation risk of the distribution network. Based on the spatiotemporal correlation between fault probability and impact range, and combined with a comprehensive simulation dataset, the dynamic operation risk index of the distribution network is predicted, and the risk level of photovoltaic access schemes is classified.

[0012] As a preferred embodiment of the adaptive planning method for rural power distribution networks including distributed photovoltaic (PV) power according to the present invention, the specific steps for generating a distributed PV access planning scheme by performing multi-objective optimization based on the characteristics of the rural power distribution network and its dynamic operational risks are as follows. Based on the risk level of photovoltaic grid connection schemes, a set of constraints for multi-objective optimization is constructed; Based on the characteristics of rural power distribution networks, power grid topology vulnerability assessment results are generated; Based on the results of the power grid topology vulnerability assessment and the set of constraints for multi-objective optimization, a distributed photovoltaic access planning scheme is generated.

[0013] Secondly, this invention provides an adaptive planning system for rural power distribution networks including distributed photovoltaic (PV) power, comprising: a data acquisition and analysis module, an access scheme generation module, a scheme optimization and evaluation module, a dynamic risk analysis module, and a planning optimization module; the data acquisition and analysis module collects multi-source data and preprocesses it to generate a high-quality dataset, and obtains PV output and rural power distribution network characteristics; the access scheme generation module generates multi-objective balanced PV access schemes based on PV output and the high-quality dataset; the scheme optimization and evaluation module obtains a comprehensive evaluation of the priority ranking results of PV access schemes based on the multi-objective balanced PV access schemes combined with the new energy access needs of rural users and the power supply reliability model of rural users, and performs a comprehensive evaluation to select the optimal PV access scheme; the dynamic risk analysis module performs fault simulation based on the optimal PV access scheme, and collects fault data corresponding to the simulated fault events during the fault simulation process, while evaluating the dynamic operation risk of the power distribution network; the planning optimization module performs multi-objective optimization based on the characteristics of the rural power distribution network and the dynamic operation risk of the power distribution network to generate a distributed PV access planning scheme. Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein the computer program, when executed by the processor, implements any step of the adaptive planning method for rural power distribution networks containing distributed photovoltaic power as described in the first aspect of the present invention.

[0014] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the adaptive planning method for rural power distribution networks containing distributed photovoltaics as described in the first aspect of the present invention.

[0015] The beneficial effects of this invention are as follows: it collects multi-source data and preprocesses it to generate a high-quality dataset, and obtains photovoltaic power output and rural distribution network characteristics, realizing the fully automated processing from data acquisition to grid feature extraction; it performs fault simulation based on the optimal photovoltaic access scheme, and collects fault data corresponding to the simulated fault events during the fault simulation process, while assessing the dynamic operation risk of the distribution network, realizing a paradigm shift from static planning to dynamic risk prevention and control. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. 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.

[0017] Figure 1 This is a flowchart of an adaptive planning method for rural power distribution networks that include distributed photovoltaic power.

[0018] Figure 2 A flowchart generated for a multi-objective balanced photovoltaic grid connection scheme.

[0019] Figure 3 A flowchart for dynamic risk assessment and planning optimization.

[0020] Figure 4 A flowchart for data preprocessing and forecasting for adaptive planning of photovoltaic access in rural power distribution networks. Detailed Implementation

[0021] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0022] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0023] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0024] Reference Figures 1-4As one embodiment of the present invention, this embodiment provides an adaptive planning method for rural power distribution networks including distributed photovoltaic power, comprising the following steps: S1. Collect multi-source data and preprocess it to generate a high-quality dataset, and obtain photovoltaic power output and rural power distribution network characteristics; Multi-source data includes meteorological data, photovoltaic parameters, power grid topology, equipment parameters, power grid operation data, and user load data; It should be noted that meteorological data is obtained by directly measuring parameters such as wind speed, temperature, and humidity through ground-based meteorological stations, or by acquiring large-scale irradiance data using satellite remote sensing technology. Photovoltaic parameters are obtained by reading DC-side voltage, current, and AC-side power data from the inverter, while portable testing equipment is used to periodically measure performance parameters such as surface temperature and irradiance of photovoltaic modules. The power grid topology is obtained by recording the location of poles and the connection relationships of lines through on-site surveys, and by compiling node and branch information in conjunction with power grid planning data to form complete power grid connection relationship data. Equipment parameters are obtained by manually inspecting and recording the nameplate parameters of equipment such as transformers and circuit breakers, while sensors installed on the equipment collect real-time status data such as temperature and oil pressure during operation. Power grid operation data is obtained by acquiring real-time voltage, current, and power data from the dispatch center, while historical operation records are collected for analyzing the long-term operating characteristics of the power grid. User load data is obtained by periodically collecting user electricity consumption and power data through smart meters, or by installing dedicated monitoring equipment on the side of large users to obtain the electricity consumption characteristics of subdivided loads.

[0025] Preprocessing includes data cleaning, format conversion, deduplication, normalization, and outlier handling; It should be noted that, firstly, the collected multi-source data, including meteorological data, photovoltaic parameters, power grid topology, equipment parameters, power grid operation data, and user load data, are cleaned to remove incomplete, erroneous, or irrelevant data records. Next, format conversion is performed to ensure that all data sources follow a unified format standard for subsequent processing. Then, deduplication is performed to eliminate duplicate data entries and ensure the uniqueness of the dataset. The normalization step is used to map data values ​​from different sources to the same scale to avoid certain features from having an undue impact on the results due to excessive differences in magnitude. Finally, outlier handling is used to identify and correct data points that deviate from the normal range to ensure the accuracy and reliability of the dataset.

[0026] Based on a high-quality dataset, the theoretical photovoltaic output is calculated using the complete workflow function of PVLib, and the theoretical photovoltaic output is compensated and corrected using a loss compensation algorithm to obtain the photovoltaic output. It should be noted that, based on high-quality meteorological and photovoltaic parameter data, the real-time solar altitude angle and azimuth angle are first calculated according to solar astronomical formulas. Combined with local latitude, longitude, and time zone information, the solar radiation incident angle is determined. Then, using the horizontal irradiance from meteorological data, it is converted into the total irradiance of the photovoltaic module's tilt surface through empirical formulas. Next, the DC-side current, voltage, and power output are calculated based on the module's electrical characteristic parameters. Finally, the theoretical AC-side output is calculated according to the standard conversion efficiency. Subsequently, various loss factors (such as decreased transmittance due to dust accumulation, efficiency degradation caused by increased module operating temperature, and DC line impedance voltage drop) are statistically analyzed based on historical operating data. The theoretical values ​​of each stage are corrected and compensated item by item using measured loss rates or piecewise functions, ultimately obtaining photovoltaic output time-series data that conforms to the actual on-site operating conditions.

[0027] It should also be noted that the PVLib toolkit contains a series of functions from solar position calculation and radiation conversion to power generation prediction, which can accurately reflect the power generation characteristics of photovoltaics under different environmental conditions. By calling the pre-defined function flow, photovoltaic output results that conform to actual operating conditions can be quickly obtained, providing reliable data support for photovoltaic project planning, performance evaluation, and operation and maintenance decisions.

[0028] Based on high-quality datasets, cluster analysis was used to extract the load distribution characteristics and photovoltaic absorption matching characteristics of rural power distribution networks. It should be noted that, based on a high-quality dataset and with user load data as the core foundation, this method integrates meteorological, photovoltaic, and other data, and constructs load curves for each node of the rural power distribution network through data correction (such as net load calculation after deducting photovoltaic output). Key characteristic parameters (including daily load peak, valley, peak-valley difference, peak occurrence time, and load fluctuation standard deviation) are calculated using time-series analysis. Subsequently, by analyzing the peak period distribution patterns, valley period concentration intervals, and overall fluctuation trends of various load curves, the spatial distribution patterns of load peaks and valleys in different geographical regions or node types are identified in the time dimension. Then, the photovoltaic output time series is matched with the time-series data of various load patterns on a time-by-time basis to calculate the time-series correlation coefficient between photovoltaic output and load demand (such as the Pearson correlation coefficient). The photovoltaic absorption rate index for different time periods and regions is calculated, and the photovoltaic absorption matching degree level is classified based on the clustering results. Finally, the matching characteristics of the rural power distribution network load characteristics and photovoltaic absorption capacity are obtained.

[0029] Based on load distribution characteristics and photovoltaic absorption matching characteristics, the network connectivity and weak links are dynamically evaluated through graph theory analysis to obtain the characteristics of rural power distribution networks. It should be noted that, firstly, a topological connection diagram of the rural power distribution network is constructed based on the load distribution characteristics and the matching degree characteristics of photovoltaic absorption. Nodes represent key locations in the power grid (such as substations and distribution transformers), and edges represent transmission lines and their electrical parameters. Then, connectivity indices in graph theory (such as node degree and network density) are used to evaluate the overall connectivity of the network. Key transmission paths and potential bottleneck nodes are identified through shortest path analysis and betweenness centrality. The power supply reliability of each region under load fluctuation and photovoltaic access conditions is quantified, and finally, the characteristics of the rural power distribution network are obtained.

[0030] It should also be noted that graph theory analysis abstracts the power grid topology into a mathematical network of nodes and edges, and studies its connectivity characteristics and power flow transmission patterns through graph theory algorithms. Weak links refer to critical nodes (such as overloaded transformers) and bottleneck lines (such as high-impedance transmission lines) in the power grid that are prone to power outages or voltage quality degradation under fault or high load conditions.

[0031] S2. Based on photovoltaic output and high-quality datasets, generate a multi-objective balanced photovoltaic grid connection scheme; Based on high-quality datasets, power flow calculations are used to obtain the physical constraints of the power grid. It should be noted that, firstly, measured data such as the grid topology connections, load size and phase angle of each node, output of distributed generation sources and line impedance parameters are collected. Then, power flow calculation algorithms (such as the Newton-Raphson method) are used to process the measured data. By iteratively solving key electrical quantities such as node voltage amplitude and phase angle, branch power distribution, etc., the physical constraints of the grid, such as the allowable voltage deviation range, maximum transmission capacity of the line and stable operation boundary, which reflect the actual operating state of the grid, are finally obtained.

[0032] Based on photovoltaic power output and high-quality datasets, the total cost of photovoltaic grid connection is minimized through mixed-integer linear programming, and the combination of photovoltaic nodes and capacity is obtained to generate a preliminary grid connection scheme that meets the physical constraints of the power grid. The expression for minimizing the total cost of photovoltaic grid connection, calculated using mixed-integer linear programming, is as follows: ; in, It is a binary variable, representing a node. Whether to connect to photovoltaic power, Represents a node Photovoltaic grid connection capacity, Represents a node The investment cost per unit capacity of photovoltaic power. This represents the total number of nodes connected to the photovoltaic system. Represents a node Fixed costs, Represents a node Electricity revenue coefficient (value range: 0~1). This represents minimizing the total cost of photovoltaic (PV) grid connection. It is a symbolic function; Based on the preliminary grid connection scheme, the NSGA-II algorithm is used to generate a multi-objective balanced photovoltaic grid connection scheme under the condition of satisfying the physical constraints of the power grid, through non-dominated sorting and elite retention strategies. It should be noted that, firstly, the multi-objective optimization objectives of photovoltaic (PV) access are clearly defined (such as maximizing the absorption rate, minimizing voltage deviation, and minimizing line loss), and the physical constraints of the power grid are embedded into the optimization framework as boundary conditions. Then, the NSGA-II algorithm is used to initialize the population for different access locations and capacity combinations. The solution set is hierarchically sorted by non-dominated sorting, and the individual congestion degree is calculated to maintain population diversity. During the iteration process, the elite retention strategy is used to screen out high-quality solutions on the Pareto front, and finally, a multi-objective balanced PV access scheme is generated.

[0033] It should also be noted that the elite preservation strategy refers to merging the parent and offspring populations in each generation of the NSGA-II algorithm, and then selecting the best individual directly into the next generation based on the crowding comparison operator. This ensures that the high-quality solutions found are not lost during the evolution process, thereby accelerating convergence to the Pareto optimal front and maintaining population diversity.

[0034] S3. Based on the multi-objective balance of photovoltaic access schemes, the priority ranking results of photovoltaic access schemes are obtained by combining the new energy access needs of rural users and the power supply reliability model of rural users. The optimal photovoltaic access scheme is selected through comprehensive evaluation. Through questionnaires and historical data analysis, we obtained the demand for new energy access for rural users and the impact of new energy on the reliability of power supply for rural users. At the same time, we obtained the differentiated needs of rural users for power supply reliability and formed a demand list. It should be noted that, firstly, through questionnaires covering dimensions such as willingness to connect to new energy sources, electricity load characteristics, and sensitivity to power outages, the direct needs and feedback of rural users regarding new energy connection were collected through on-site visits or online channels. Simultaneously, historical power outage records, user repair reports, and electricity consumption fluctuation information were extracted to analyze the changing patterns of power supply reliability before and after new energy grid connection and its impact on users in different regions and industries. Subsequently, the two types of data were cross-validated to identify users' differentiated demands for power supply reliability (such as the high requirements of agricultural irrigation users for continuous power supply and the tolerance of residential users for short-term power outages). Finally, a demand list containing elements such as willingness to connect and priority of reliability requirements was compiled.

[0035] High-quality datasets are used as a framework for multi-dimensional evaluation metrics, and reliability quantification targets are set based on the requirements list. It should be noted that, firstly, the photovoltaic output fluctuation rate is calculated based on meteorological data and photovoltaic parameters. Then, stability indicators such as voltage qualification rate and power outage duration are statistically analyzed in conjunction with grid operation data. Fault recovery time and equipment health level are assessed through equipment parameters and grid topology. At the same time, user load data is analyzed to identify load peak and valley characteristics and electricity sensitivity. A multi-dimensional evaluation framework including power supply stability, response speed and equipment reliability is constructed. Subsequently, in combination with the differentiated demands of users for power supply reliability in the demand list, the weight of each indicator is determined by the analytic hierarchy process. Quantitative targets are set for different regions (such as urban areas and rural grid ends) and user types (such as residential, agricultural and industrial and commercial). For example, the voltage qualification rate in core urban areas is ≥99.99%, and the power outage duration in remote rural areas is ≤8 hours / year. Finally, a reliability quantitative target combining data-driven and demand-oriented approaches is formed.

[0036] Based on a multi-dimensional evaluation index framework and reliability quantification objectives, an initial rural user power supply reliability model was constructed, and a BP neural network was used to train and finally construct the rural user power supply reliability model. It should be noted that, firstly, historical data is collected to construct a training sample set, and the mapping relationship between input features (meteorological conditions, photovoltaic output, load characteristics, etc.) and output targets (actual power supply reliability indicators) is determined. Then, a BP neural network is used to design the network structure (such as input layer nodes corresponding to key indicators, the number of hidden layers and nodes determined by trial and error, and the output layer corresponding to reliability scores). The output error is calculated through forward propagation and the weight parameters are adjusted through backpropagation using the training samples. After multiple iterations and optimizations until convergence, a neural network model that can accurately predict the power supply reliability of rural users is constructed.

[0037] The multi-objective balanced photovoltaic access scheme is input into the rural user power supply reliability model to predict the power supply reliability of each photovoltaic access scheme, and it is matched with the new energy access needs of rural users to evaluate the priority of photovoltaic access schemes and rank them. It should be noted that, firstly, the multi-objective balanced photovoltaic access scheme is input into the pre-trained rural user power supply reliability model. The rural user power supply reliability model will, based on parameters such as photovoltaic access capacity, location, load, and output in the scheme, quickly calculate indicators such as voltage qualification rate, outage duration, and power supply reliability for each scheme using data-driven methods (e.g., machine learning prediction) or physical-data hybrid methods (e.g., power flow calculation combined with reliability assessment). Finally, it generates reliability prediction results. Subsequently, the reliability prediction results (including specific numerical indicators such as voltage qualification rate and outage duration for each scheme) are compared with the user renewable energy access requirements in the demand list (such as those stipulated by national / local standards). The system compares and matches the lower limit of voltage qualification rate and the upper limit of acceptable power outage duration for users. A multi-index comprehensive evaluation method is adopted, which selects key indicators such as voltage qualification rate, power outage duration, and power supply reliability. First, each indicator is standardized to eliminate dimensional differences. Then, the weight of each indicator is determined by the Analytic Hierarchy Process (AHP) and the entropy weight method to reflect the demand preferences of rural users for power supply reliability. Finally, the standardized indicator values ​​are combined with the weights to calculate the comprehensive score using the weighted summation method, the TOPSIS method, or the fuzzy comprehensive evaluation method. This calculates the fit between each photovoltaic access scheme and the power supply demand. Finally, the photovoltaic access schemes are prioritized according to the fit.

[0038] Based on the priority ranking of photovoltaic access schemes, the TOPSIS method is used to calculate the closeness of each photovoltaic access scheme to the optimal and worst values, and the optimal photovoltaic access scheme is selected according to the closeness. It should be noted that, firstly, based on the priority ranking of photovoltaic access schemes, a decision matrix is ​​constructed that includes multi-dimensional evaluation indicators such as power supply reliability, economic efficiency, and user demand matching degree. The indicator data is normalized and the entropy weight method is used to determine the weight of each indicator. Then, the TOPSIS method is used to calculate the Euclidean distance from each scheme to the positive ideal solution (optimal value) and the negative ideal solution (worst value), and then the relative closeness of each scheme to the ideal solution is obtained. The higher the closeness, the better the overall performance of the scheme. Finally, the optimal photovoltaic access scheme is selected based on the magnitude of the closeness value.

[0039] S4. Based on the optimal photovoltaic access scheme, conduct fault simulation and collect fault data corresponding to the simulated fault events during the fault simulation process, while assessing the dynamic operation risk of the distribution network. The Monte Carlo method was used to simulate random fault scenarios under the optimal photovoltaic access scheme. Fault features were randomly extracted multiple times, and the dynamic propagation process of each fault was analyzed using power time-domain simulation tools. A comprehensive simulation dataset containing the dynamic characteristics of the entire fault process was obtained, and the fault data of each simulation was recorded. It should be noted that, firstly, the grid topology connection relationship, equipment parameters and operating data under the optimal photovoltaic access scheme are organized, and initial conditions such as fault type (such as short circuit, open circuit), occurrence probability and location range are set. Then, the Monte Carlo method is used to randomly extract fault characteristics (such as fault location, type and duration), and the fault environment is simulated in combination with meteorological data. The voltage and current changes, protection action sequence and sag amplitude under each fault are calculated by power time domain simulation tools. The data of the entire fault propagation process are recorded, and the results of multiple simulations are summarized to form a comprehensive dataset containing dynamic characteristics of multiple scenarios.

[0040] It should also be noted that random fault scenarios refer to various random fault situations that may occur in the power grid after photovoltaic grid connection, including fault type (such as single-phase grounding, three-phase short circuit), location of occurrence (such as any node of the line), duration and superimposed meteorological conditions, etc. The specific characteristics of the fault are dynamic parameters such as the magnitude of the fault current, the voltage sag, the timing of the protection device's operation, and the fault propagation range, which are used to analyze the power grid's anti-interference capability and weak links.

[0041] By analyzing the timestamps of fault data, we can identify the frequency distribution of faults in different time periods, identify high-incidence periods and periodic patterns of faults, and generate the occurrence pattern of fault events in the time dimension. It should be noted that, firstly, timestamp information is extracted from the fault data, and the data is classified and statistically analyzed according to time granularities such as hour, day, week and month. The number of faults in each time period is calculated and frequency distribution curves are plotted. Then, by comparing the fault frequencies in different time periods, high-incidence periods of faults (such as peak load periods and extreme weather periods) are identified, and periodic characteristics (such as seasonal fluctuations and diurnal differences) are analyzed. Finally, the occurrence pattern of fault events in the time dimension is formed.

[0042] By performing cluster analysis and density calculation on the geographical location of fault data, the spatial clustering pattern of hotspot areas distributed in the comprehensive simulation dataset is obtained, and the spatial distribution characteristics of fault events are generated. It should be noted that, firstly, the geographical location information (such as latitude and longitude coordinates) of the fault data is extracted. The K-means clustering algorithm is used to spatially group the fault points and identify the clusters of densely distributed faults. Then, the fault density value of each region is calculated by the kernel density estimation algorithm. The fault density distribution map is constructed by the spatial interpolation algorithm (such as the inverse distance weighting method or the Kriging method) to intuitively display the fault hotspot areas. The significant high-value clusters (hotspots) are identified by hotspot analysis (such as Getis-Ord Gi* statistics). The spatial autocorrelation and local clustering patterns (such as high-high clustering) are analyzed by combining global and local Moran's I index. Then, the distribution pattern of hotspots along the line corridor or around key equipment is observed based on GIS visualization, thereby analyzing the spatial clustering pattern of fault hotspot areas and finally generating the spatial distribution characteristics of fault events.

[0043] Based on the occurrence patterns of fault data in the time dimension and the distribution characteristics in the spatial dimension, the spatiotemporal correlation between fault probability and impact range is calculated through spatiotemporal statistical analysis, thereby obtaining the dynamic operation risk of the distribution network. It should be noted that, firstly, based on the fault frequency distribution in the time dimension and the hotspot area distribution in the spatial dimension, a spatiotemporal joint statistical framework is constructed, and the number of fault occurrences in each fixed spatiotemporal unit is counted (by counting the number of fault records in each spatiotemporal unit and dividing by the total number of observations to calculate the fault probability) and the corresponding impact range indicators (such as the power outage area and the number of affected users) to establish a complete spatiotemporal joint dataset. Subsequently, by calculating the spatiotemporal correlation coefficient (such as the Moran index or spatiotemporal mutual information), the coupling pattern between high-incidence periods and high-risk areas of faults is identified. Combined with the power grid topology and load characteristics, the fault propagation range and the severity of the consequences are assessed, and finally, a spatiotemporal distribution map of the dynamic operation risk of the distribution network is generated.

[0044] Based on the spatiotemporal correlation between fault probability and impact range, and combined with a comprehensive simulation dataset, the dynamic operation risk index of the distribution network is predicted by geographically weighted regression, and the risk level of photovoltaic access schemes is classified. It should be noted that, firstly, based on the spatiotemporal correlation results of fault probability and impact range, dynamic fault characteristic parameters (such as voltage sag amplitude and protection action time) for corresponding time periods and regions are extracted from the comprehensive simulation dataset. Then, a geographically weighted calculation method is adopted, with the spatiotemporal correlation strength as the weight, combined with the fault propagation range and power grid operation parameters, to calculate the dynamic operation risk index of the distribution network for each region. Finally, based on the magnitude of the risk index and the spatial distribution location of the photovoltaic access scheme, the risk level is classified.

[0045] S5. Based on the characteristics of rural power distribution networks and the dynamic operation risks of power distribution networks, multi-objective optimization is carried out to generate distributed photovoltaic access planning schemes; Based on the risk level of photovoltaic grid connection schemes, a set of constraints for multi-objective optimization is constructed using the analytic hierarchy process (AHP). It should be noted that, firstly, based on the risk level of the photovoltaic access scheme, the risk level classification standards for each risk indicator (such as voltage fluctuation, fault cascading, equipment overload, etc.) are clarified. Then, the analytic hierarchy process (AHP) is used to transform the risk level classification standards into a hierarchical constraint structure. High-level risk objectives (such as overall reliability) are decomposed into mid-level key indicators (voltage stability, fault recovery capability, etc.), and further refined into bottom-level quantifiable constraint parameters (voltage fluctuation amplitude ≤ ±5%, fault recovery time < 2 hours, etc.). Consistency checks are used to ensure the logical consistency between constraints at all levels, ultimately forming a multi-objective optimization constraint set with clear hierarchy and well-defined objectives.

[0046] Based on the characteristics of rural power distribution networks, the N-1 static security analysis method is used to generate power grid topology vulnerability assessment results; It should be noted that, firstly, based on the power grid topology data and equipment parameters combined with the characteristics of rural distribution networks, the vulnerability of rural power grids under single component failures is analyzed using the N-1 static security analysis criterion. Then, through topology connectivity analysis (such as node degree and network density calculation), the number of connecting lines for each node (node ​​degree) is calculated, and nodes with low node degree are marked as potential isolated points. Next, the network density of the entire power grid (the ratio of actual connections to the theoretical maximum number of connections) is calculated, and areas with low network density are identified as weak interconnection areas. Finally, the distribution of interconnection lines is analyzed to identify nodes with only one interconnection line (single interconnection point) and key points connecting different subnets. Key interconnection nodes are identified, and finally, by combining node degree, network density, and interconnection line distribution, areas prone to isolation due to insufficient interconnection (such as low node degree, low density areas) and key interconnection nodes with a large impact range during faults (such as hub nodes or single interconnection points) are located. The risk areas of voltage drop after faults are assessed by combining equipment capacity and line impedance parameters, with particular attention paid to the difficulty of power restoration due to faults in intermediate nodes of long lines. Finally, the results of topology analysis and equipment parameter assessment are integrated to obtain vulnerable links in the rural power grid topology (such as single interconnection points, high impedance lines), forming a power grid topology vulnerability assessment result that includes the risk location, impact degree, and risk level unique to the rural power grid.

[0047] Based on the results of the power grid topology vulnerability assessment and the constraint set of multi-objective optimization, a distributed photovoltaic access planning scheme is generated by searching for the Pareto optimal solution set and co-evolution of the population through the Pareto multi-objective evolution algorithm.

[0048] It should be noted that, firstly, based on the results of the power grid topology vulnerability assessment and the constraint set of multi-objective optimization, a multi-objective optimization framework including photovoltaic access location, capacity configuration, and risk control objectives is constructed. Then, the Pareto multi-objective evolutionary algorithm is used to initialize the population, and the Pareto front solution set is identified through non-dominated sorting. During the iteration process, an elite retention strategy is used to maintain high-quality solutions. At the same time, the topology vulnerability assessment results are used as the key constraint condition of the fitness function to guide the population to evolve towards photovoltaic access schemes that reduce the risk of voltage drop. Finally, the Pareto optimal solution set of distributed photovoltaic access planning that takes into account both power supply reliability improvement and risk control is generated.

[0049] It should also be noted that Pareto optimality, in multi-objective optimization, refers to the set of all solutions that cannot be surpassed by any other solution in a multi-objective optimization problem, representing the best balance state among different objectives.

[0050] This embodiment also provides an adaptive planning system for rural power distribution networks with distributed photovoltaic power, including: a data acquisition and analysis module, an access scheme generation module, a scheme optimization and evaluation module, a dynamic risk analysis module, and a planning optimization module; The data acquisition and analysis module collects multi-source data, preprocesses it to generate a high-quality dataset, and obtains photovoltaic power output and rural power distribution network characteristics. The grid connection scheme generation module generates a multi-objective balanced grid connection scheme based on photovoltaic power output and a high-quality dataset. The optimal photovoltaic access scheme evaluation module, based on a multi-objective balanced photovoltaic access scheme, combines the renewable energy access needs of rural users with the power supply reliability model of rural users to obtain the priority ranking results of the photovoltaic access scheme after comprehensive evaluation, and performs a comprehensive evaluation to select the optimal photovoltaic access scheme. The dynamic risk analysis module performs fault simulation based on the optimal photovoltaic access scheme, collects fault data corresponding to the simulated fault events during the fault simulation process, and assesses the dynamic operation risk of the distribution network. The planning and optimization module performs multi-objective optimization based on the characteristics of rural power distribution networks and the dynamic operation risks of the power distribution networks, and generates distributed photovoltaic access planning schemes.

[0051] This embodiment also provides a computer device applicable to the adaptive planning method for rural power distribution networks with distributed photovoltaic power, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the adaptive planning method for rural power distribution networks with distributed photovoltaic power as proposed in the above embodiment.

[0052] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0053] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the adaptive planning method for rural power distribution networks with distributed photovoltaic power as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0054] In summary, this invention achieves fully automated processing from data acquisition to grid feature extraction by: collecting multi-source data and preprocessing it to generate a high-quality dataset, and obtaining photovoltaic power output and rural distribution network characteristics; and by performing fault simulation based on the optimal photovoltaic access scheme and collecting fault data corresponding to the simulated fault events during the fault simulation process, while assessing the dynamic operation risk of the distribution network, thus realizing a paradigm shift from static planning to dynamic risk prevention and control.

[0055] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An adaptive planning method for rural power distribution networks incorporating distributed photovoltaic power, characterized in that: include, Collect multi-source data and preprocess it to generate a high-quality dataset, and obtain photovoltaic power output and rural power distribution network characteristics; Based on photovoltaic power output and high-quality datasets, a multi-objective balanced photovoltaic grid connection scheme is generated. The photovoltaic access scheme based on multi-objective balance combines the new energy access needs of rural users and the power supply reliability model of rural users to obtain the priority ranking results of the photovoltaic access scheme after comprehensive evaluation, and conducts a comprehensive evaluation to select the optimal photovoltaic access scheme. Fault simulation is performed based on the optimal photovoltaic access scheme, and fault data corresponding to the simulated fault events are collected during the fault simulation process. At the same time, the dynamic operation risk of the distribution network is assessed. Based on the characteristics of rural power distribution networks and their dynamic operational risks, multi-objective optimization is performed to generate distributed photovoltaic access planning schemes.

2. The adaptive planning method for rural power distribution networks including distributed photovoltaic power as described in claim 1, characterized in that: The multi-source data includes meteorological data, photovoltaic parameters, power grid topology, equipment parameters, power grid operation data, and user load data. The preprocessing includes data cleaning, format conversion, deduplication, normalization, and outlier handling.

3. The adaptive planning method for rural power distribution networks including distributed photovoltaic power as described in claim 2, characterized in that: The specific steps for obtaining photovoltaic power output and rural power distribution network characteristics are as follows. Calculate photovoltaic output based on high-quality datasets; Based on high-quality datasets, load distribution characteristics and photovoltaic absorption matching characteristics of rural power distribution networks are extracted; Based on the load distribution characteristics and photovoltaic power consumption matching characteristics, network connectivity and weak links are assessed to obtain the characteristics of rural power distribution networks.

4. The adaptive planning method for rural power distribution networks including distributed photovoltaic power as described in claim 3, characterized in that: The specific steps for generating a multi-objective balanced photovoltaic grid connection scheme based on photovoltaic output and high-quality datasets are as follows: Based on high-quality datasets, obtain the physical constraints of the power grid; Based on photovoltaic output and high-quality datasets, the total cost of photovoltaic grid connection is minimized, and the combination of connected photovoltaic nodes and capacity is obtained to generate a preliminary grid connection scheme that meets the physical constraints of the power grid. Based on the initial grid connection scheme, a multi-objective balanced photovoltaic grid connection scheme is generated under the condition of satisfying the physical constraints of the power grid.

5. The adaptive planning method for rural power distribution networks including distributed photovoltaic power as described in claim 4, characterized in that: The photovoltaic (PV) grid connection scheme based on multi-objective balance combines the renewable energy access needs of rural users and the power supply reliability model of rural users to obtain a comprehensive evaluation of the PV grid connection scheme priority ranking results, and then performs a comprehensive evaluation to select the optimal PV grid connection scheme. The specific steps are as follows: Through questionnaires and historical data analysis, we obtained the demand for new energy access for rural users and the impact of new energy on the reliability of power supply for rural users. At the same time, we obtained the differentiated needs of rural users for power supply reliability and formed a demand list. High-quality datasets are used as a framework for multi-dimensional evaluation metrics, and reliability quantification targets are set based on the requirements list. Based on a multi-dimensional evaluation index framework and reliability quantification objectives, an initial rural user power supply reliability model was constructed, and a BP neural network was used to train and finally construct the rural user power supply reliability model. The multi-objective balanced photovoltaic access scheme is input into the rural user power supply reliability model to predict the power supply reliability of each photovoltaic access scheme, and it is matched with the new energy access needs of rural users to evaluate the priority of photovoltaic access schemes and rank them. Based on the priority ranking of photovoltaic access schemes, the proximity of each photovoltaic access scheme to the optimal and worst values ​​is calculated, and the optimal photovoltaic access scheme is selected according to the proximity.

6. The adaptive planning method for rural power distribution networks including distributed photovoltaic power as described in claim 5, characterized in that: The method involves performing fault simulation based on the optimal photovoltaic access scheme, collecting fault data corresponding to simulated fault events during the simulation process, and assessing the dynamic operational risks of the distribution network. The specific steps are as follows: Simulate random fault scenarios under the optimal photovoltaic access scheme, repeatedly randomly extract fault features, analyze the dynamic propagation process of each fault, obtain a comprehensive simulation dataset containing the dynamic characteristics of the entire fault process, and record the fault data of each simulation. By analyzing the timestamps of fault data, we can identify the frequency distribution of faults in different time periods, identify high-incidence periods and periodic patterns of faults, and generate the occurrence pattern of fault events in the time dimension. Obtain the spatial clustering patterns of hotspot areas distributed in the comprehensive simulation dataset, and generate the spatial distribution characteristics of fault events; Based on the occurrence patterns of fault data in the time dimension and the distribution characteristics in the spatial dimension, the spatiotemporal correlation between fault probability and impact range is calculated, thereby obtaining the dynamic operation risk of the distribution network. Based on the spatiotemporal correlation between fault probability and impact range, and combined with a comprehensive simulation dataset, the dynamic operation risk index of the distribution network is predicted, and the risk level of photovoltaic access schemes is classified.

7. The adaptive planning method for rural power distribution networks including distributed photovoltaic power as described in claim 6, characterized in that: Based on the characteristics of rural power distribution networks and their dynamic operational risks, multi-objective optimization is performed to generate a distributed photovoltaic (PV) grid connection planning scheme. The specific steps are as follows: Based on the risk level of photovoltaic grid connection schemes, a set of constraints for multi-objective optimization is constructed; Based on the characteristics of rural power distribution networks, power grid topology vulnerability assessment results are generated; Based on the results of the power grid topology vulnerability assessment and the set of constraints for multi-objective optimization, a distributed photovoltaic access planning scheme is generated.

8. An adaptive planning system for rural power distribution networks incorporating distributed photovoltaic power, based on the adaptive planning method for rural power distribution networks incorporating distributed photovoltaic power as described in any one of claims 1 to 7, characterized in that: include, Data acquisition and analysis module, access scheme generation module, scheme optimization and evaluation module, dynamic risk analysis module, planning optimization module; The data acquisition and analysis module collects multi-source data, preprocesses it to generate a high-quality dataset, and obtains photovoltaic power output and rural power distribution network characteristics. The grid connection scheme generation module generates a multi-objective balanced grid connection scheme based on photovoltaic power output and a high-quality dataset. The optimal photovoltaic access scheme evaluation module, based on a multi-objective balanced photovoltaic access scheme, combines the renewable energy access needs of rural users with the power supply reliability model of rural users to obtain the priority ranking results of the photovoltaic access scheme after comprehensive evaluation, and performs a comprehensive evaluation to select the optimal photovoltaic access scheme. The dynamic risk analysis module performs fault simulation based on the optimal photovoltaic access scheme, collects fault data corresponding to the simulated fault events during the fault simulation process, and assesses the dynamic operation risk of the distribution network. The planning and optimization module performs multi-objective optimization based on the characteristics of rural power distribution networks and the dynamic operation risks of the power distribution networks, and generates distributed photovoltaic access planning schemes.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the adaptive planning method for rural power distribution networks with distributed photovoltaic power as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the adaptive planning method for rural power distribution networks with distributed photovoltaic power as described in any one of claims 1 to 7.

Images