A power grid resource-based charging pile intelligent management method and system

Abstract

The present application relates to the technical field of fault prediction, and is a charging pile intelligent management method and system based on power grid resources.The present application method obtains load data and charging pile density and charging pile position;according to the load data and the charging pile position, analysis is performed to obtain distribution characteristics; according to the charging pile density, load analysis is performed to obtain a load threshold; according to load data monitoring, a load change trend is obtained; the load value in the load change trend is compared with the load threshold; if the load value is greater than the load threshold, a warning signal is generated to remind; according to the warning signal, analysis is performed to obtain demand distribution; according to the demand distribution division, a high demand area and a low demand area are obtained; according to the high demand area and the low demand area, a matching model is constructed; based on the matching model, an allocation scheme is obtained, and the system is set to enable the system to adjust according to the allocation scheme.The present application can effectively balance the power grid load and charging demand, and improve the utilization efficiency of charging piles.

Description

Technical Field

[0001] This invention relates to the field of fault prediction technology, and in particular to a method and system for intelligent management of charging piles based on power grid resources. Background Technology

[0002] The rapid development of electric vehicles has led to a significant increase in the demand for charging stations. However, there is a potential conflict between the deployment of charging stations and the allocation of power grid resources. During peak charging periods, too many charging stations operating simultaneously can overload local power grids, affecting grid stability and security. Furthermore, the geographical distribution of charging stations does not always match the spatial distribution of power grid resources. This can result in some areas having abundant power grid resources but insufficient charging stations, while other areas have a high density of charging stations but strained power grid resources. In addition, the power allocation of charging stations is also a concern. Different electric vehicle models have varying charging power requirements, while the rated power of charging stations is constrained by power grid resources. How to dynamically adjust the power allocation of each charging station to meet the charging needs of different vehicle models while ensuring the safe and stable operation of the power grid, and simultaneously considering charging efficiency and fairness, is a pressing technical challenge that needs to be addressed.

[0003] In summary, there is a dynamic matching and optimization issue between charging piles and power grid resources. It is necessary to achieve real-time coordination between the operating status of charging piles and the power grid load in both time and space dimensions. Furthermore, the power allocation of charging piles should be dynamically adjusted and optimized based on the actual needs of electric vehicles and the availability of power grid resources to achieve coordinated operation between charging piles and the power grid. This is of great significance for promoting the application of electric vehicles and ensuring the safe and stable operation of the power grid.

[0004] In one existing technology, a specific implementation method includes a static planning method based on historical data. During implementation, the usage patterns and charging demands of electric vehicles are first analyzed. Then, within the grid load capacity range, the installation locations of charging piles are selected, and the number and power allocation of charging piles are determined. Subsequently, through a preset scheduling strategy, the real-time operating status of the charging piles is monitored to ensure reasonable allocation of power resources during peak hours and avoid local grid overload.

[0005] However, existing technologies fail to fully consider the dynamic changes in charging demand and the real-time fluctuations in grid load. As a result, in actual use, the layout and power allocation of charging piles cannot adapt to sudden peak demand in a timely manner, increasing the risk of grid instability and causing low charging efficiency and waste of resources. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides an intelligent management method and system for charging piles based on power grid resources. The invention primarily utilizes modules for data acquisition, feature analysis, area definition, data monitoring, judgment, demand analysis, area division, model building, scheme generation, and system settings. This solves the problem that existing technologies fail to adequately consider the dynamic changes in charging demand and the real-time fluctuations in power grid load. Consequently, in practical use, the layout and power allocation of charging piles cannot adapt to sudden peak demands in a timely manner, increasing the risk of power grid instability and causing low charging efficiency and resource waste.

[0007] The technical means employed in this invention are as follows:

[0008] A smart management method for charging piles based on power grid resources includes:

[0009] Acquire load data, charging pile density, and charging pile location;

[0010] Based on the load data and the location of the charging piles, feature analysis is performed to obtain the distribution characteristics;

[0011] Based on the charging pile density, a correlation analysis is performed on the load to obtain the load threshold.

[0012] Based on the load data, the load change trend can be obtained through monitoring.

[0013] The load value in the load change trend is compared with the load threshold; if the load value is greater than the load threshold, an early warning signal is generated to provide an alert.

[0014] Demand analysis is performed based on the aforementioned warning signals to obtain the demand distribution;

[0015] Based on the demand distribution, high-demand areas and low-demand areas are obtained;

[0016] A matching model is constructed based on the high-demand region and the low-demand region;

[0017] Based on the matching model, an allocation scheme is obtained;

[0018] The system is configured according to the allocation scheme so that the system adjusts the power grid in accordance with the allocation scheme.

[0019] Furthermore, obtaining the distribution characteristics specifically includes:

[0020] Spatiotemporal clustering analysis is performed on the load data to obtain the spatiotemporal distribution pattern;

[0021] Based on the location of the charging piles, spatial distribution data of the charging piles is obtained by mapping.

[0022] Based on the spatiotemporal distribution pattern and the spatial distribution data of the charging piles, a weighted interpolation is performed to obtain a charging pile density distribution map and a power grid load distribution map.

[0023] The distribution characteristics are obtained by overlaying the charging pile density distribution map and the power grid load distribution map.

[0024] Furthermore, obtaining the load threshold specifically includes:

[0025] Based on the charging pile density, a correlation coefficient is obtained by calculating the correlation with the load.

[0026] Based on the correlation coefficients, a model is constructed to obtain a linear regression model;

[0027] The linear regression model was used to fit the charging pile density and the power grid load distribution map to obtain the regression coefficients and intercepts.

[0028] A prediction model is constructed based on the regression coefficients and the intercept;

[0029] Based on the prediction model, different charging pile densities are predicted to obtain the predicted load value;

[0030] The load threshold is obtained by evaluating the predicted load value in conjunction with the power grid operation requirements.

[0031] Furthermore, the acquisition of the demand distribution specifically includes:

[0032] Demand data is obtained by reading the area indicated by the warning signal to obtain real-time demand data.

[0033] The required location data is obtained by reading the location within the area indicated by the warning signal.

[0034] The demand distribution is obtained by integrating the real-time demand data and the demand location data.

[0035] Furthermore, the demand distribution is further divided, specifically including:

[0036] Data is read based on the demand distribution to obtain an attribute dataset;

[0037] Cluster analysis was performed on the attribute dataset to obtain the clustering results;

[0038] Based on the clustering results, the regions are divided to obtain clustering regions;

[0039] The average power and spatial density of charging piles within the clustered area are calculated to obtain the power density index;

[0040] Based on the power density index, high-demand areas and low-demand areas are obtained.

[0041] Furthermore, the method for obtaining the allocation scheme includes:

[0042] Based on the matching model, the expected charging demand and the predicted number of vehicles are obtained;

[0043] A dynamic programming model is constructed based on the expected charging demand and the predicted number of vehicles.

[0044] Based on the dynamic programming model, the allocation scheme is obtained by solving the problem with the optimization objectives of maximizing the utilization rate of charging piles, minimizing vehicle waiting time, and ensuring that the total power does not exceed the pre-stored regional power supply capacity.

[0045] Furthermore, the allocation scheme compares and analyzes real-time demand data with pre-stored historical forecast data to determine whether there are abnormal fluctuations in the current area's charging demand. If abnormal fluctuations exist, peak demand areas are determined based on clustering results, and their matching degree with the charging pile layout is analyzed. For peak demand areas with low matching degree, the shortest path algorithm is used to plan the guidance path for vehicles to nearby charging piles. The power allocation of charging piles is dynamically adjusted.

[0046] This invention also provides a charging pile intelligent management system based on power grid resources, implemented using a method for intelligent management of charging piles based on power grid resources. The system includes: a data acquisition module, a feature analysis module, a range definition module, a data monitoring module, a judgment module, a demand analysis module, a region division module, a model building module, a scheme generation module, and a system setting module, wherein:

[0047] The data acquisition module is used to acquire load data, charging pile density, and charging pile location;

[0048] The feature analysis module is used to perform feature analysis based on load data and the location of the charging pile to obtain distribution features;

[0049] The defined range area is used to perform correlation analysis on the load based on the charging pile density to obtain the load threshold.

[0050] The data monitoring module is used to monitor load data and obtain load change trends;

[0051] The judgment module is used to compare the load value with the load threshold in the load change trend; if the load value is greater than the load threshold, an early warning signal is generated to remind the user.

[0052] The demand analysis module is used to perform demand analysis based on the early warning signal to obtain the demand distribution;

[0053] The region division module is used to divide regions according to demand distribution, resulting in high-demand regions and low-demand regions.

[0054] The model building module is used to build a matching model based on the high-demand region and the low-demand region.

[0055] The scheme generation module is used to perform allocation based on the matching model to obtain an allocation scheme;

[0056] The system setting module is used to configure the system according to the allocation scheme, so that the system can adjust the power grid according to the allocation scheme.

[0057] Compared with the prior art, the present invention has the following advantages:

[0058] The present invention provides a method and system for intelligent management of charging piles based on power grid resources, including: acquiring load data, charging pile density, and charging pile location; performing feature analysis based on load data and charging pile location to obtain distribution characteristics; performing correlation analysis on the load based on charging pile density to obtain a load threshold; monitoring the load data to obtain a load change trend; comparing the load value in the load change trend with the load threshold; generating an early warning signal if the load value exceeds the load threshold; performing demand analysis based on the early warning signal to obtain a demand distribution; dividing the demand distribution into high-demand areas and low-demand areas; constructing a matching model based on the high-demand areas and low-demand areas; allocating resources based on the matching model to obtain an allocation scheme; and setting the system according to the allocation scheme to adjust the power grid according to the allocation scheme.

[0059] This invention establishes a mapping model between grid load and charging pile density by real-time monitoring of grid load changes and combining the characteristics of charging pile density distribution. When the grid load exceeds a threshold, the invention analyzes the spatiotemporal distribution characteristics of charging demand and uses a clustering algorithm to re-divide charging pile areas, identifying high and low demand zones. Furthermore, a matching model between grid resources and charging demand is established, and a dynamic programming algorithm is used to dynamically optimize the allocation of charging pile power in different areas. By remotely distributing the optimized power scheme and monitoring it in real time, this invention can effectively balance grid load and charging demand, improve the utilization efficiency of charging piles, and ensure the safe and stable operation of the grid.

[0060] Based on the above reasons, this invention can be widely applied in fields such as fault prediction. Attached Figure Description

[0061] 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 some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0062] Figure 1 This is a schematic diagram of the intelligent management method for charging piles based on power grid resources in this invention.

[0063] Figure 2 This is a schematic diagram of the intelligent management system for charging piles based on power grid resources in this invention. Detailed Implementation

[0064] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0065] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the present invention or its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0066] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0067] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention. It should also be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters in the following figures denote similar items; therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0068] like Figure 1 As shown, this invention provides a smart management method for charging piles based on power grid resources, including:

[0069] The process involves acquiring load data, charging pile density, and charging pile locations. During implementation, real-time load data from each node of the power grid is obtained, and the data is preprocessed to remove outliers and missing values, resulting in a reliable load dataset. Based on the geographical coordinates of the charging piles, they are mapped to corresponding regional grids, and the number of charging piles within each grid is counted to obtain spatial distribution data of the charging piles.

[0070] By installing smart meters and data acquisition devices, parameters such as current, voltage, and power of various distribution transformers and lines can be monitored in real time. For example, a city's power grid has 500 monitoring points, collecting data every 5 minutes. During the data preprocessing stage, outliers need to be identified. For instance, data with instantaneous current fluctuations exceeding 300% are caused by equipment failure and should be discarded. For missing values, the average value of nearby time points can be used to fill in the gaps, ensuring data continuity and reliability.

[0071] Based on load data and charging pile locations, feature analysis is performed to obtain distribution characteristics. In a preferred embodiment of this invention, obtaining the distribution characteristics specifically includes:

[0072] Spatiotemporal clustering analysis is performed on the load data to obtain the spatiotemporal distribution pattern; the location of the charging piles is mapped to obtain the spatial distribution data of the charging piles; weighted interpolation is performed on the spatiotemporal distribution pattern and the spatial distribution data of the charging piles to obtain the charging pile density distribution map and the power grid load distribution map; the charging pile density distribution map and the power grid load distribution map are superimposed to obtain the distribution characteristics.

[0073] Spatial interpolation algorithms, such as Kriging interpolation or inverse distance weighted interpolation, are used to estimate the density of charging piles at unknown locations based on known charging pile density data, generating a continuous density distribution map. Spatiotemporal clustering analysis is performed on the power grid load data to identify the spatiotemporal distribution patterns of the load, determine peak load areas and time periods, and provide a reference for subsequent analysis. The charging pile density distribution map is overlaid with the power grid load distribution map to calculate their spatial correlation and identify the degree of matching between the charging pile density and load distribution.

[0074] The geographical location information of charging stations is crucial for analyzing their distribution characteristics. Suppose a city has 1000 charging stations, each with latitude and longitude coordinates. Divide the city into 10km × 10km grids and count the number of charging stations in each grid. For example, a grid in the city center might have 50 charging stations, while a grid in the suburbs might only have 5. This statistical method can intuitively reflect the uneven spatial distribution of charging infrastructure.

[0075] A load threshold is obtained by performing a correlation analysis on the load based on the charging pile density. Specifically, in a preferred embodiment of this invention, obtaining the load threshold includes: calculating the correlation coefficient between the charging pile density and the load; constructing a model based on the correlation coefficient to obtain a linear regression model; fitting the charging pile density and power grid load distribution map based on the linear regression model to obtain regression coefficients and intercepts; constructing a prediction model based on the regression coefficients and intercepts; predicting the load value for different charging pile densities based on the prediction model; and evaluating the predicted load value in conjunction with power grid operation requirements to obtain the load threshold.

[0076] Historical power grid load data and charging pile density data within the target area are acquired. Both types of data are preprocessed to remove outliers and missing values, resulting in a standardized dataset. The Pearson correlation coefficient method is used to calculate the correlation between power grid load and charging pile density, determining whether a significant linear correlation exists. If the absolute value of the correlation coefficient is greater than a preset threshold, the correlation is considered strong. If the correlation between power grid load and charging pile density is strong, a linear regression model is constructed. Using charging pile density as the independent variable and power grid load as the dependent variable, a linear function is fitted using the least squares method to obtain the regression coefficients and intercept. Based on the linear regression model, the power grid load value under different charging pile densities is predicted. Simultaneously, the mean squared error and coefficient of determination of the model are calculated to evaluate the model's goodness of fit and prediction accuracy. For the predicted power grid load value, combined with the requirements for safe power grid operation, upper and lower limit threshold ranges for the load are determined. Load values ​​exceeding these ranges are considered to pose a risk to power grid stability. Based on the power grid load threshold range, a reasonable range for charging pile density is derived through the inverse linear regression model. Substituting the load thresholds into the regression model yields the corresponding upper and lower limits for charging pile density.

[0077] In power grid planning and management, analyzing the relationship between power grid load and charging pile density is crucial. First, historical data for the target area needs to be obtained. For example, the daily peak power grid load and the number of charging piles in a city over a year. During data preprocessing, abnormally high load data may be observed on some days, such as a sudden increase of three times the normal load, caused by temporary large-scale events or equipment failures. For these outliers, the median substitution method can be used to ensure data reliability. Next, the Pearson correlation coefficient between power grid load and charging pile density is calculated. Assuming a correlation coefficient of 0.85, this indicates a strong positive correlation between the two. This means that as the number of charging piles increases, the power grid load also tends to increase. This relationship provides a basis for subsequent prediction and planning. Based on the significant correlation, a linear regression model can be constructed. Using charging pile density as the independent variable and power grid load as the dependent variable, the regression equation is obtained by fitting the data using the least squares method. For example, the equation could be: Y = 500 + 0.2X, where Y represents the power grid load (unit: MW) and X represents the number of charging piles per square kilometer. This equation indicates that without charging stations, the basic grid load is 500MW, and each additional charging station increases the load by an average of 0.2MW. To assess the reliability of the model, the coefficient of determination is calculated. Assuming A value of 0.75 indicates that the model can explain 75% of the load variation, demonstrating a good fit. Simultaneously, the mean squared error (MSE) is calculated to measure the average deviation between the predicted and actual values. A lower MSE value indicates higher model prediction accuracy. Based on the requirements for safe grid operation, a safe load threshold needs to be determined. Assuming the safe operating load range for the grid is 600-800 MW, substituting this range into the regression equation allows for the calculation of a reasonable range for charging pile density. For example, the calculated number of charging piles per square kilometer should be controlled between 500-1500. This provides clear guidance for urban planning departments, ensuring both the charging needs of electric vehicles and the safe and stable operation of the power grid.

[0078] Load data is monitored to obtain load change trends. By deploying sensors and smart meters in the power grid, load data such as voltage and current at each node are collected in real time. This data is then transmitted to a cloud data center for aggregation and processing. Within the cloud data center, big data analytics are used to clean, organize, and analyze the aggregated power grid load data, yielding information such as load change curves and trends.

[0079] By deploying smart sensors at key nodes such as substations and distribution rooms, parameters such as voltage, current, and power can be collected. For example, in a city's smart grid, each substation is equipped with 10 high-precision current transformers and 10 voltage transformers, achieving millisecond-level data sampling. These devices transmit data to the nearest edge computing node via industrial Ethernet, where it is initially processed before being uploaded to the cloud data center. The cloud data center is the core of power grid data analysis. Employing a distributed storage and computing architecture, it can efficiently process massive amounts of data. Taking a provincial power grid as an example, its data center processes over 100 million data records daily. The data cleaning process uses machine learning algorithms to identify outliers, such as sudden voltage drops caused by equipment failure. The cleaned data is then used to plot power grid load curves using methods such as time series analysis, revealing load variation patterns.

[0080] The system compares the load value in the load change trend with the load threshold; if the load value exceeds the load threshold, an early warning signal is generated. Based on the actual carrying capacity of the power grid and the distribution of charging piles, a pre-set early warning threshold for the power grid load is established, serving as the standard for determining whether the load limit has been exceeded. A power grid load prediction model is built using machine learning algorithms. By analyzing historical load data and change patterns, it predicts the power grid load changes over a future period, providing a reference for threshold determination. If the power grid load change trend analysis or load prediction results indicate that the load at a future moment exceeds the preset threshold, a charging pile density early warning signal is triggered.

[0081] For example, setting a power grid load warning threshold is an important means of ensuring power grid security. The determination of the threshold needs to consider multiple factors, such as transformer capacity and line carrying capacity. A power grid operator in a certain region sets 90% of the rated capacity of the main transformer as the warning threshold. When the load reaches this value, a yellow warning is triggered; when it reaches 95%, a red warning is triggered. This tiered warning mechanism helps to timely dispatch power grid resources. Machine learning plays an important role in power grid load forecasting. Commonly used algorithms include Support Vector Machine (SVM) and Long Short-Term Memory (LSTM) networks. A typical LSTM model can use historical data from the past 7 days, combined with external factors such as weather and holidays, to predict load changes in the next 24 hours. A city using this model for forecasting has kept the average error rate within 3%, providing a reliable basis for power grid dispatching. Charging pile density warning is a new issue in the context of the widespread adoption of electric vehicles. When the predicted load approaches the threshold, the system will issue a warning signal, prompting relevant departments to suspend the addition of new charging piles or take load shifting measures. For example, load forecasting in a commercial area shows that the weekend peak period exceeds the warning threshold. The system automatically sends information to nearby charging stations, suggesting that users charge during off-peak hours or go to areas with lower loads to charge, thereby balancing the pressure on the power grid.

[0082] Demand analysis is performed based on the early warning signals to obtain the demand distribution; in a preferred embodiment of the present invention, obtaining the demand distribution specifically includes:

[0083] Demand data is obtained by reading the area of ​​the warning signal; location data is obtained by reading the location of the warning signal; and demand distribution is obtained by integrating the real-time demand data and the demand location data.

[0084] Taking a city as an example, by installing smart meters at charging stations, data such as current and voltage can be collected in real time. When setting warning thresholds, factors such as the rated power of the charging stations and the grid's carrying capacity must be considered. Assuming the rated power of charging stations in a certain area is 100kW, the warning threshold can be set at 90kW. When the load exceeds 90kW, the system automatically triggers a warning signal. After the warning is triggered, obtaining the real-time location distribution of electric vehicles is crucial. Using onboard GPS or mobile app positioning functions, the location information of electric vehicles within the area can be accurately determined. For example, if there are 500 electric vehicles in a commercial area, the system can update their specific coordinates in real time. Analysis of historical charging demand data helps predict future demand trends.

[0085] Based on the demand distribution, high-demand areas and low-demand areas are obtained. In a preferred embodiment of the present invention, the demand distribution is divided into the following steps: reading data based on the demand distribution to obtain an attribute dataset; performing cluster analysis on the attribute dataset to obtain clustering results; dividing the clusters based on the clustering results to obtain clustered regions; calculating the average power and spatial density of charging piles within the clustered regions to obtain a power density index; and dividing the high-demand areas and low-demand areas based on the power density index.

[0086] The system acquires the geographical location information and rated power attributes of charging piles to construct a charging pile attribute dataset. Based on the spatiotemporal distribution characteristics of charging demand, the dataset is preprocessed to extract key features related to charging demand. A K-means clustering algorithm is used to perform cluster analysis on the preprocessed dataset, grouping charging piles with similar power and concentrated spatial distribution into the same region. For each cluster, the average power and spatial density of charging piles within that region are calculated to obtain the region's power density index. Based on the region's power density index and the spatiotemporal distribution characteristics of charging demand, the charging demand level of each region is determined. If a region has high power density and frequent charging demand, it is classified as a high-demand region; otherwise, it is classified as a low-demand region.

[0087] A matching model is constructed based on high-demand and low-demand areas; an allocation scheme is obtained based on the matching model; in a preferred embodiment of the present invention, the method for obtaining the allocation scheme includes: predicting the expected charging demand and the predicted number of vehicles based on the matching model; constructing a dynamic programming model based on the expected charging demand and the predicted number of vehicles; and solving the dynamic programming model with the optimization objectives of maximizing the utilization rate of charging piles, minimizing vehicle waiting time, and ensuring that the total power does not exceed the pre-stored regional power supply capacity to obtain the allocation scheme.

[0088] Based on the regional division results, the distribution information of areas with surplus and shortage of power grid resources is obtained; the charging demand of electric vehicles is divided into regions to obtain the charging demand distribution in different regions; the distribution of power grid resources and the distribution of charging demand are clustered using a clustering algorithm to determine whether the cluster centers of the two match; if the matching degree of the cluster centers is lower than a preset threshold, the regional division scheme is adjusted according to the clustering results until the matching degree meets the requirements; based on the matched distribution of power grid resources and the distribution of charging demand, an optimization model of resource supply and demand response is constructed.

[0089] The system is configured according to the allocation scheme to adjust the power grid accordingly. In a preferred embodiment of this invention, the allocation scheme compares and analyzes real-time demand data with pre-stored historical forecast data to determine if there are abnormal fluctuations in current regional charging demand. If abnormal fluctuations exist, peak demand areas are identified based on clustering results, and their matching degree with the charging pile layout is analyzed. For peak demand areas with low matching degree, a shortest path algorithm is used to plan the guidance path for vehicles to nearby charging piles; and the power allocation of charging piles is dynamically adjusted.

[0090] Acquire real-time operating status data of charging piles in each area, including current power, occupancy status, and fault information; predict the number of vehicles arriving and expected charging demand in each area in the future based on the vehicle matching model; take the charging pile and vehicle demand in each area as input, establish a dynamic programming model, take maximizing the utilization rate of charging piles and minimizing vehicle waiting time as optimization objectives, and take the constraint that the total power does not exceed the regional power supply capacity, and solve for the optimal power allocation scheme.

[0091] This invention also provides a smart management system for charging piles based on power grid resources, implemented using a smart management method for charging piles based on power grid resources. The system includes: a data acquisition module, a feature analysis module, a range definition module, a data monitoring module, a judgment module, a demand analysis module, a region division module, a model building module, a scheme generation module, and a system setting module, wherein:

[0092] The data acquisition module is used to acquire load data, charging pile density, and charging pile location.

[0093] The feature analysis module is used to perform feature analysis based on load data and charging pile locations to obtain distribution characteristics;

[0094] The range is defined to perform correlation analysis on the load based on the charging pile density, and to obtain the load threshold.

[0095] The data monitoring module is used to monitor load data and obtain load change trends;

[0096] The judgment module is used to compare the load value with the load threshold in the load change trend; if the load value is greater than the load threshold, an early warning signal is generated to remind the user.

[0097] The demand analysis module is used to perform demand analysis based on early warning signals and obtain the demand distribution.

[0098] The region division module is used to divide regions based on demand distribution, resulting in high-demand regions and low-demand regions.

[0099] The model building module is used to build matching models based on high-demand and low-demand regions.

[0100] The allocation scheme generation module is used to generate allocation schemes based on the matching model.

[0101] The system settings module is used to configure the system according to the allocation scheme, so that the system can adjust the power grid according to the allocation scheme.

[0102] Example

[0103] like Figure 1As shown, this invention provides a smart charging pile management method and system based on power grid resources. By analyzing historical data and considering special events (such as holidays, large-scale events, etc.), the model predicts that approximately 50 electric vehicles will need charging in the area within the next 4 hours, with an average charging requirement of 40 kWh per vehicle. This prediction helps managers increase the supply of charging piles or adjust charging strategies in advance. The dynamic programming model aims to maximize charging pile utilization and minimize vehicle waiting time. Consider a specific scenario: there are 5 charging piles in an area, each with a maximum power of 100 kW, and the area's power supply capacity is limited to 400 kW. When 8 vehicles arrive simultaneously, the model needs to decide how to allocate charging resources. Through dynamic programming, an optimal solution can be derived: 3 charging piles charge 3 vehicles at 100 kW each, 2 charging piles charge 2 vehicles at 50 kW each, and the remaining 3 vehicles enter a waiting queue. This allocation scheme ensures high utilization of charging piles while minimizing vehicle waiting time. The optimization objective reflects the core requirements of charging network management. Maximizing charging pile utilization means improving equipment efficiency and reducing idle costs. Minimizing vehicle waiting time directly impacts user experience and improves the convenience of using electric vehicles. There is a certain conflict between these two goals, requiring a trade-off based on specific circumstances in practical applications. For example, during peak charging demand periods, some waiting time needs to be sacrificed to ensure high utilization of charging stations. The constraints reflect considerations for grid security. The constraint that the total power output does not exceed the area's power supply capacity ensures that the charging network does not place excessive pressure on the grid. For instance, if an area has a power supply capacity of 1MW, even if there are 20 100kW charging stations in that area, they cannot all operate at maximum power simultaneously; instead, an optimization algorithm is needed to rationally allocate power.

[0104] The system is configured according to the power allocation scheme to adjust the power grid accordingly. Specifically, this includes: distributing an optimized power allocation scheme via remote communication, adjusting the operating power of each charging pile, and simultaneously collecting real-time charging pile operating parameters and grid load data to monitor changes in charging pile operating status and grid load levels. The optimized power allocation scheme, distributed from the cloud, is obtained via a wireless communication module, and the target operating power of each charging pile is analyzed. Based on the target operating power, the actual operating power of each charging pile is adjusted by the control module to match the target operating power. Sensors collect parameters such as operating voltage and current of each charging pile in real time and upload them to the cloud via the communication module. The total grid load power at the current moment is obtained, and it is determined whether the grid load level exceeds a preset threshold. If the grid load level exceeds the threshold, the power allocation scheme is re-optimized. An updated power allocation scheme is generated by combining historical operating data and currently collected data using a machine learning algorithm. Finally, a reinforcement learning algorithm is used to dynamically adjust the power allocation ratio of each charging pile based on changes in charging pile operating status and fluctuations in grid load levels, achieving overall power optimization for the charging pile group.

[0105] For power allocation schemes, comparing real-time demand with historical forecast data helps identify abnormal fluctuations. If actual charging demand suddenly exceeds forecasts by 30% on a given day, it may indicate a large-scale event or special circumstance, requiring timely adjustments to resource allocation. Matching analysis between charging demand and charging pile layout can optimize resource allocation. For example, suppose there are 10 charging piles in a commercial area, with 8 located around the shopping mall and only 2 near office buildings. If there is significant vehicle congestion in the office building area, it indicates a mismatch between the charging pile layout and actual demand. Shortest path algorithms can plan optimal charging routes for drivers. Using Dijkstra's algorithm, considering factors such as road conditions and distance, the optimal route to nearby charging piles is planned for vehicles near office buildings, alleviating localized pressure. Dynamically adjusting charging power is an effective means of balancing grid load. During peak demand periods, the power of charging piles around the shopping mall can be increased from 100kW to 120kW, while the power of charging piles in other low-load areas can be reduced to 80kW, achieving optimized allocation of power resources. This intelligent management approach not only improves charging efficiency but also reduces grid fluctuation risks, providing strong infrastructure support for the widespread application of electric vehicles.

[0106] In summary, this invention discloses an intelligent management method for charging piles based on power grid resources, including: acquiring load data, charging pile density, and charging pile location; performing feature analysis based on load data and charging pile location to obtain distribution characteristics; performing correlation analysis on the load based on charging pile density to obtain a load threshold; monitoring the load data to obtain a load change trend; comparing the load value in the load change trend with the load threshold; generating an early warning signal if the load value exceeds the load threshold; performing demand analysis based on the early warning signal to obtain a demand distribution; dividing the demand distribution into high-demand areas and low-demand areas; constructing a matching model based on the high-demand areas and low-demand areas; allocating resources based on the matching model to obtain an allocation scheme; and setting the system according to the allocation scheme to adjust the power grid according to the allocation scheme. This invention establishes a mapping model between grid load and charging pile density by real-time monitoring of grid load changes and combining the characteristics of charging pile density distribution. When the grid load exceeds a threshold, the invention analyzes the spatiotemporal distribution characteristics of charging demand and uses a clustering algorithm to re-divide charging pile areas, identifying high and low demand zones. Furthermore, a matching model between grid resources and charging demand is established, and a dynamic programming algorithm is used to dynamically optimize the allocation of charging pile power in different areas. By remotely distributing the optimized power scheme and monitoring it in real time, this invention can effectively balance grid load and charging demand, improve the utilization efficiency of charging piles, and ensure the safe and stable operation of the grid.

[0107] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A smart management method for charging piles based on power grid resources, characterized in that, include: Acquire load data, charging pile density, and charging pile location; Based on the load data and the location of the charging piles, feature analysis is performed to obtain the distribution characteristics; Based on the charging pile density, a correlation analysis is performed on the load to obtain the load threshold. Based on the load data, the load change trend can be obtained through monitoring. The load value in the load change trend is compared with the load threshold; if the load value is greater than the load threshold, an early warning signal is generated to provide an alert. Demand analysis is performed based on the aforementioned warning signals to obtain the demand distribution; Based on the demand distribution, high-demand areas and low-demand areas are obtained; The demand distribution is divided, specifically including: Data is read based on the demand distribution to obtain an attribute dataset; Cluster analysis was performed on the attribute dataset to obtain the clustering results; Based on the clustering results, the regions are divided to obtain clustering regions; The average power and spatial density of charging piles within the clustered area are calculated to obtain the regional power density index. Based on the regional power density index and the spatiotemporal distribution characteristics of charging demand, high-demand regions and low-demand regions are obtained. A matching model is constructed based on the high-demand region and the low-demand region; Based on the matching model, an allocation scheme is obtained. The allocation scheme compares and analyzes real-time demand data with pre-stored historical prediction data to determine whether there are abnormal fluctuations in the current area's charging demand. If there are abnormal fluctuations, the peak demand areas are determined based on the clustering results, and their matching degree with the charging pile layout is analyzed. For peak demand areas with low matching degree, the shortest path algorithm is used to plan the guidance path of vehicles to nearby charging piles. The power allocation of charging piles is dynamically adjusted. The system is configured according to the allocation scheme so that the system adjusts the power grid in accordance with the allocation scheme.

2. The intelligent management method for charging piles based on power grid resources according to claim 1, characterized in that, The acquisition of the distribution characteristics specifically includes: Spatiotemporal clustering analysis is performed on the load data to obtain the spatiotemporal distribution pattern; Based on the location of the charging piles, spatial distribution data of the charging piles is obtained by mapping. Based on the spatiotemporal distribution pattern and the spatial distribution data of the charging piles, a weighted interpolation is performed to obtain a charging pile density distribution map and a power grid load distribution map. The distribution characteristics are obtained by overlaying the charging pile density distribution map and the power grid load distribution map.

3. The intelligent management method for charging piles based on power grid resources according to claim 2, characterized in that, The acquisition of the load threshold specifically includes: Based on the charging pile density, a correlation coefficient is obtained by calculating the correlation with the load. Based on the correlation coefficients, a model is constructed to obtain a linear regression model; The linear regression model was used to fit the charging pile density and the power grid load distribution map to obtain the regression coefficients and intercepts. A prediction model is constructed based on the regression coefficients and the intercept; Based on the prediction model, different charging pile densities are predicted to obtain the predicted load value; The load threshold is obtained by evaluating the predicted load value in conjunction with the power grid operation requirements.

4. The intelligent management method for charging piles based on power grid resources according to claim 1, characterized in that, The acquisition of the demand distribution specifically includes: Demand data is obtained by reading the area indicated by the warning signal to obtain real-time demand data. The required location data is obtained by reading the location within the area indicated by the warning signal. The demand distribution is obtained by integrating the real-time demand data and the demand location data.

5. The intelligent management method for charging piles based on power grid resources according to claim 1, characterized in that, The method for obtaining the allocation scheme includes: Based on the matching model, the expected charging demand and the predicted number of vehicles are obtained; A dynamic programming model is constructed based on the expected charging demand and the predicted number of vehicles. Based on the dynamic programming model, the allocation scheme is obtained by solving the problem with the optimization objectives of maximizing the utilization rate of charging piles, minimizing vehicle waiting time, and ensuring that the total power does not exceed the pre-stored regional power supply capacity.

6. A charging pile intelligent management system based on power grid resources, implemented according to the intelligent management method for charging piles based on power grid resources as described in any one of claims 1-5, characterized in that, include: The system comprises the following modules: data acquisition, feature analysis, scope definition, data monitoring, judgment, requirements analysis, region division, model building, solution generation, and system settings. The data acquisition module is used to acquire load data, charging pile density, and charging pile location; The feature analysis module is used to perform feature analysis based on load data and the location of the charging pile to obtain distribution features; The defined range area is used to perform correlation analysis on the load based on the charging pile density to obtain the load threshold. The data monitoring module is used to monitor load data and obtain load change trends; The judgment module is used to compare the load value with the load threshold in the load change trend; if the load value is greater than the load threshold, an early warning signal is generated to remind the user. The demand analysis module is used to perform demand analysis based on the early warning signal to obtain the demand distribution; The region division module is used to divide regions according to demand distribution, resulting in high-demand regions and low-demand regions. The model building module is used to build a matching model based on the high-demand region and the low-demand region. The scheme generation module is used to perform allocation based on the matching model to obtain an allocation scheme; The system setting module is used to configure the system according to the allocation scheme, so that the system can adjust the power grid according to the allocation scheme.

Images