Load balancing and scheduling method for new energy vehicle charging piles based on intelligent algorithm

Abstract

The present invention relates to the technical field of new energy vehicles, and specifically to a load balancing and scheduling method for charging piles of new energy vehicles based on intelligent algorithms, comprising the following steps: S1, data collection: real-time collection of operating data of charging piles; S2, charging demand prediction: prediction of the charging demand of each charging pile within a predetermined time in the future through a graph convolutional network model; S3, charging pile spatial scheduling optimization: spatial optimization of the scheduling of charging piles according to the geographical distribution of charging piles, the flow of surrounding vehicles and the results of charging demand prediction; S4, load scheduling and resource allocation: based on the results of charging demand prediction and charging pile spatial scheduling optimization, combined with the load of the power grid, resource allocation is performed. The present invention improves the utilization efficiency of charging pile resources, optimizes the layout of the overall charging network, and ensures the fairness of charging services and the balance of power grid load.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy vehicles, and in particular to a new energy vehicle charging pile load balancing and scheduling method based on an intelligent algorithm. Background Art

[0002] With the popularization of new energy vehicles, charging piles, as an important infrastructure, have become a key link in promoting the development of the new energy vehicle industry. However, the operating efficiency, resource utilization rate and coordination with the power grid of charging piles still face great challenges. The demand for charging piles is affected by multiple factors such as vehicle flow, power grid load and time period in the region. The difference in demand in different regions and time periods leads to uneven distribution of charging pile resources, too long idle time of charging piles in some areas, and overload in other areas. Therefore, how to achieve load balancing and scheduling optimization of the charging pile network, improve the resource utilization efficiency of charging piles, and ensure the stable operation of the power grid is the focus and difficulty of the industry's current attention.

[0003] Existing technologies mainly rely on traditional static scheduling methods or single prediction models to allocate charging pile resources, which have many deficiencies. On the one hand, static scheduling methods are difficult to respond to the dynamic changes of regional charging demand in real time, resulting in uneven resource allocation or scheduling delays. On the other hand, existing prediction models fail to fully consider the spatial correlation between charging piles and the dynamic changes of the surrounding environment (such as vehicle flow and grid load), resulting in low accuracy in charging demand prediction. In addition, existing technologies lack systematic optimization methods and cannot achieve full-process optimization from charging demand prediction, spatial scheduling to load distribution. Resource utilization efficiency and service capabilities are low, and it is difficult to ensure the utilization efficiency of charging piles and the balance of grid operation.

[0004] In order to address the deficiencies in the prior art, the present invention provides a new energy vehicle charging pile load balancing and scheduling method based on an intelligent algorithm, which realizes efficient collaboration between the charging pile network and the power grid, thereby optimizing the operating efficiency of the charging network, reducing the operating pressure of the power grid, and promoting the intelligent development of new energy vehicle charging infrastructure. Summary of the invention

[0005] The present invention provides a new energy vehicle charging pile load balancing and scheduling method based on an intelligent algorithm.

[0006] The load balancing and scheduling method of new energy vehicle charging piles based on intelligent algorithm includes the following steps: S1, data collection: real-time collection of charging pile operation data, including usage, surrounding vehicle flow, battery status and grid load; S2, charging demand prediction: Based on the collected operation data of charging piles, the charging demand of each charging pile within the future scheduled time (1 hour) is predicted through the graph convolutional network (GCN) model; S3, Optimization of charging pile spatial scheduling: Based on the geographical distribution of charging piles, the surrounding vehicle flow and the results of charging demand forecast, the scheduling of charging piles is spatially optimized, including: S31, regional demand assessment: based on the charging demand forecast results, assess the demand in the area where the charging piles are located; S32, scheduling priority sorting: according to the results of regional demand assessment, priority is set for charging pile resource allocation to optimize resource distribution; S33, charging pile location optimization and layout adjustment: based on the results of scheduling priority sorting, combined with the surrounding vehicle flow and charging demand prediction results, optimize the layout and location of the charging piles; S4, load scheduling and resource allocation: Based on the results of charging demand prediction and charging pile space scheduling optimization, combined with the grid load situation, resource allocation is carried out.

[0007] Optionally, the data collection in S1 includes: S11, Charging pile usage: through the smart terminal installed on the charging pile, the usage status, charging power, charging time and number of users of the charging pile are monitored in real time; S12, surrounding vehicle flow: by installing vehicle flow monitoring equipment (traffic cameras) around the charging piles, the vehicle flow data of the roads around the charging piles is collected in real time, and combined with the real-time vehicle flow information provided by the traffic data platform, it is dynamically updated; S13, battery status: obtain the battery power, charging status and temperature of the vehicle battery in real time through the new energy vehicle battery management system (BMS) connected to the charging pile; S14, grid load: Power load monitoring sensors installed at key grid nodes are used to obtain grid load conditions, voltage fluctuations, and current data in real time.

[0008] Optionally, the charging demand prediction in S2 includes: S21, constructing a graph structure of charging piles: constructing a graph structure of charging piles according to the geographical locations and spatial relationships between charging piles, treating charging piles as nodes in the graph, and node features including historical charging pile usage, surrounding vehicle traffic, battery status, and grid load. Edges between charging piles represent spatial or traffic dependencies between charging piles, and edge weights are assigned by Euclidean distance and surrounding vehicle traffic. S22, graph convolution operation: In the graph convolution network (GCN) model, the graph convolution operation is used to process the charging pile graph structure. The graph convolution operation is performed through the adjacency matrix and the node feature matrix To spread; S23, Charging demand prediction: In the last layer of the graph convolutional network (GCN) model, the node features are mapped to the predicted values ​​through the fully connected layer, that is, the charging demand of each charging pile within the future scheduled time (1 hour).

[0009] Optionally, the regional needs assessment in S31 includes: S311, collecting charging demand prediction data: collecting the predicted charging demand of each charging pile; S312, calculating the weight of the charging pile: by considering the traffic flow around the charging pile and geographic location (via Euclidean distance ) to calculate the weight of the charging pile ; S313, evaluate the total charging demand in the region: based on the predicted charging demand and corresponding weight of each charging pile , evaluate the charging demand of the entire region, the total regional charging demand It is the weighted average of the demand for charging piles in the region.

[0010] Optionally, the scheduling priority sorting in S32 includes: S321, calculate charging pile priority: based on regional demand assessment results and the predicted charging demand of each charging pile, and calculate the priority score of each charging pile , prioritize the charging piles; S322, Priority sorting: based on the calculated charging pile priority score , sort all charging piles, and the charging piles with high priority are placed in front. The priority sorting of charging piles determines the order of resource allocation; S323, resource allocation optimization: During the scheduling process, charging resources are allocated in descending order according to the priority sorting results. When resources are limited, charging piles with high priority are given priority.

[0011] Optionally, the charging pile location optimization and layout adjustment in S33 includes: S331, determining the priority location distribution of charging piles: according to the priority sorting result of the charging piles, the charging piles are preferentially deployed in high priority areas; S332, adjust layout based on demand and flow: optimize the distribution of charging piles based on surrounding vehicle flow and charging demand forecast results.

[0012] Optionally, the demand- and traffic-based layout adjustment in S332 includes: S3321, Demand and flow weight calculation: Calculate the comprehensive demand flow weight of each charging pile area based on the charging demand forecast results of the area and the surrounding vehicle flow. ; S3322, Optimization of charging pile layout: based on comprehensive demand flow weight , optimize the layout of charging piles, when Exceeding high demand threshold When the area is defined as a high-demand high-flow area, the density of charging piles is increased. Below low demand threshold When defining areas with low demand and low traffic, the number of charging piles can be reduced or resource allocation can be optimized by sharing charging piles.

[0013] Optionally, the load scheduling and resource allocation in S4 includes: S41, load demand adjustment: based on the charging demand forecast results and the comprehensive demand flow weight of the charging pile space scheduling optimization , combined with the real-time load conditions of the power grid , dynamically adjust the load priority of each area; S42, resource allocation and execution: Calculate resource allocation amounts based on the adjusted regional load priorities.

[0014] Optionally, the load demand adjustment in S41 includes: S411, Assessment of regional demand and grid carrying capacity: Based on the comprehensive demand flow weight of the region and the real-time load situation of the power grid, and by calculating the remaining carrying capacity of the power grid, determine whether it is necessary to limit or adjust the regional load; S412, dynamically adjust regional priorities: dynamically adjust the load priority of each region based on the comprehensive demand flow weight and the remaining carrying capacity of the power grid.

[0015] Optionally, the resource allocation and execution in S42 includes: S421, calculating resource allocation: calculating the resource allocation of each region according to the adjusted regional priority and the total available resources; S422, executing resource allocation: allocating the calculated resource allocation amount to the charging piles in each area.

[0016] Beneficial effects of the present invention: The present invention realizes accurate prediction of charging demand through comprehensive collection and integrated analysis of new energy vehicle charging pile operation data. Based on the graph convolutional network model, it fully combines the geographical location of the charging piles, the surrounding vehicle flow, the battery status and the grid load, constructs a spatial relationship diagram between the charging piles, and predicts future charging demand through node feature propagation and the full connection layer, which significantly improves the accuracy and adaptability of the prediction. Accurate charging demand prediction provides a scientific basis for subsequent scheduling optimization, ensuring that the charging demand of different regions in different time periods is responded to in a timely manner, thereby effectively improving the load balance and operation efficiency of the charging pile network.

[0017] The present invention, through the optimization of charging pile spatial scheduling, combines regional demand assessment, priority sorting, and charging pile location optimization and layout adjustment, can dynamically optimize the spatial distribution of charging pile resources, clarify the characteristics of high-demand and high-flow areas and low-demand and low-flow areas, and formulate targeted resource allocation strategies, which not only reduces the idle time of charging piles, but also avoids the phenomenon of overloading, significantly improves the utilization efficiency of charging pile resources, optimizes the layout of the overall charging network, and ensures the fairness of charging services and the balance of power grid load.

[0018] The present invention ensures the efficiency and fairness of charging resources by dynamically adjusting regional priorities and power resource allocation strategies, calculates the remaining carrying capacity of the power grid in real time based on the real-time load conditions of the power grid and regional demands, dynamically adjusts the priorities of high-demand areas and low-demand areas, avoids power grid overload, and improves resource allocation flexibility. Resource allocation and execution are based on the adjusted priorities, and power resources are reasonably allocated, which not only meets the service requirements of high-demand areas, but also optimizes the resource utilization rate of low-demand areas, ensures the efficient coordination of the charging pile network and the power grid system, significantly reduces operating costs, and improves the responsiveness and service quality of the charging network as well as the operational stability and sustainability of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 A schematic diagram of a flow chart of a balancing and scheduling method according to an embodiment of the present invention; Figure 2 Schematic diagram of charging pile spatial scheduling optimization according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0022] like Figure 1-Figure 2 As shown, the load balancing and scheduling method for new energy vehicle charging piles based on intelligent algorithms includes the following steps: S1, data collection: real-time collection of charging pile operation data, including usage, surrounding vehicle flow, battery status and grid load; S2, charging demand prediction: Based on the collected operation data of charging piles, the charging demand of each charging pile within the future scheduled time (1 hour) is predicted through the graph convolutional network (GCN) model; S3, Optimization of charging pile spatial scheduling: Based on the geographical distribution of charging piles, the surrounding vehicle flow, and the results of charging demand forecast, the scheduling of charging piles is spatially optimized to reduce the idle time of charging piles and avoid overloading. Specifically, it includes: S31, regional demand assessment: based on the charging demand forecast results, assess the demand in the area where the charging piles are located; S32, scheduling priority sorting: according to the results of regional demand assessment, priority is set for charging pile resource allocation to optimize resource distribution; S33, charging pile location optimization and layout adjustment: Based on the results of scheduling priority sorting, combined with the surrounding vehicle flow and charging demand prediction results, the layout and location of charging piles are optimized to ensure efficient use of resources; S4, load scheduling and resource allocation: Based on the results of charging demand prediction and charging pile space scheduling optimization, combined with the grid load situation, resource allocation is carried out; Through the above content, charging demand can be predicted in real time, and through spatial scheduling optimization, load scheduling and resource allocation strategies, the utilization efficiency of charging pile resources can be effectively improved, the idle time of charging piles can be reduced, and excessive loading can be avoided, thereby achieving efficient configuration of charging pile resources and balance of grid load. This not only improves the service capacity of charging piles, but also helps to reduce the operating pressure of the grid and ensure the sustainable development of new energy vehicle charging infrastructure.

[0023] Data collection in S1 includes: S11, Charging pile usage: through the smart terminal installed on the charging pile, the usage status, charging power, charging time and number of users of the charging pile are monitored in real time; S12, surrounding vehicle flow: by installing vehicle flow monitoring equipment (traffic cameras) around the charging piles, the vehicle flow data of the roads around the charging piles is collected in real time, and combined with the real-time vehicle flow information provided by the traffic data platform, it is dynamically updated; S13, battery status: obtain the battery power, charging status and temperature of the vehicle battery in real time through the new energy vehicle battery management system (BMS) connected to the charging pile; S14, grid load: through the power load monitoring sensors installed at the key nodes of the grid, the load conditions, voltage fluctuations and current data of the grid are obtained in real time; Through the above content, the operating conditions of the charging piles and the surrounding environment can be fully and accurately understood. By integrating the Internet of Things, sensors and communication technologies, the load distribution of the charging piles can be dynamically adjusted to ensure the efficiency and safety of the charging process while avoiding grid overload. This not only improves the utilization efficiency of charging pile resources, but also provides accurate data support for subsequent charging demand forecasting, load scheduling and resource optimization, thereby effectively reducing the operating costs of charging facilities and improving the overall coordination and sustainability of the power grid and transportation system.

[0024] The charging demand prediction in S2 includes: S21, constructing a graph structure of charging piles: constructing a graph structure of charging piles according to the geographical location and spatial relationship between charging piles, treating charging piles as nodes in the graph, and node features including the historical usage of charging piles, surrounding vehicle flow, battery status, and grid load. The edges between charging piles represent the spatial or traffic dependency between charging piles, and the edge weights are assigned by Euclidean distance and surrounding vehicle flow, expressed as: ; in, Indicates charging pile and charging piles The Euclidean distance between and These are the coordinates of the two charging piles; ; in, Indicates charging pile and charging piles The flow between them affects the edge weight. and Charging piles and Traffic flow in the vicinity; ; in, It is a charging station and charging piles The comprehensive edge weight between and is the weight coefficient; S22, graph convolution operation: In the graph convolution network (GCN) model, the graph convolution operation is used to process the charging pile graph structure. The graph convolution operation is performed through the adjacency matrix and the node feature matrix To spread, it is expressed as: ; in, It is The node feature representation of the layer is, initially, , is a normalized adjacency matrix used to represent the relationship between charging piles. It is The weight matrix of the layer, is the ReLU activation function, It is The node feature representation of the layer is used to extract the spatial features of the charging piles layer by layer; S23, Charging demand prediction: In the last layer of the graph convolutional network (GCN) model, the node features are mapped to the predicted values ​​through the fully connected layer, that is, the charging demand of each charging pile within the future scheduled time (1 hour), expressed as: ; in, It is The feature representation of a charging pile in the last layer is: is the weight matrix of the fully connected layer, is the bias term, It is The predicted charging demand of each charging station in the next hour; Through the above content, we can effectively combine multi-dimensional data such as the geographical location of the charging piles, the surrounding vehicle flow, and the power grid load to construct a spatial relationship diagram between the charging piles, so as to accurately predict the charging demand of the charging piles in the future. The spatial and traffic dependencies between the charging piles are extracted through graph convolution operations. Not only the historical usage of the charging piles, but also the influence of the surrounding environment are considered. It can more flexibly adapt to demand fluctuations in different regions and time periods, improve the accuracy of charging demand, optimize the load balancing and scheduling of charging piles, and provide strong support for the efficient utilization of power grid resources and the reasonable layout of new energy vehicle charging facilities, thereby promoting the coordinated development of smart grids and intelligent transportation systems.

[0025] The regional needs assessment in S31 includes: S311, collecting charging demand prediction data: collecting the predicted charging demand of each charging pile; S312, calculating the weight of the charging pile: by considering the traffic flow around the charging pile and geographic location (via Euclidean distance ) to calculate the weight of the charging pile , expressed as: ; in, For charging pile The weight of For charging pile The surrounding traffic volume, For charging pile The Euclidean distance to the nearest traffic node (such as an intersection), is the total number of all charging piles; S313, evaluate the total charging demand in the region: based on the predicted charging demand and corresponding weight of each charging pile , evaluate the charging demand of the entire region, the total regional charging demand It is the weighted average of the demand for charging piles in the region, expressed as: ; in, For the region at time The total charging demand, For charging pile In time The predicted charging demand value; Through the above content, combined with the predicted charging demand of charging piles, surrounding traffic flow and geographical location, an accurate decision-making basis is provided for the allocation of charging pile resources. By weighted consideration of traffic flow and geographical distribution, high-demand areas and key charging piles can be better identified to avoid excessive concentration or excessive dispersion of resources, which helps to improve the utilization efficiency of charging piles, optimize the layout of charging piles, and ensure that charging resources can be reasonably allocated during periods of high charging demand, thereby improving the load balance and stability of the overall charging network, reducing the idle time of charging piles, avoiding overload, and maximizing the utilization of power grid resources.

[0026] Scheduling priorities in S32 include: S321, calculate charging pile priority: based on regional demand assessment results and the predicted charging demand of each charging pile, and calculate the priority score of each charging pile , the priority of charging piles is sorted as follows: ; S322, Priority sorting: based on the calculated charging pile priority score , sort all charging piles, and the charging piles with high priority are placed in front. The priority sorting of charging piles determines the order of resource allocation. The sorting formula is expressed as: ; in, This is the priority list of charging piles arranged by priority. Represents the priority score of the charging pile; S323, resource allocation optimization: During the scheduling process, charging resources are allocated in descending order according to the priority sorting results. When resources are limited, charging piles with high priority are given priority to ensure that charging piles in high-demand areas can obtain the required power support in a timely manner to avoid resource waste or overload. It is expressed as: ; in, It is a charging station The resources allocated, It is a charging station The priority score of is the total available resources (total power capacity or number of charging ports available for charging piles), is the total number of charging piles; Through the above content, resources can be reasonably allocated to charging piles. Through priority scoring, it is ensured that charging piles with higher demand obtain charging resources first, thereby avoiding waste of resources and overload of some charging piles, effectively improving the resource utilization efficiency of charging piles, optimizing the overall operation of the charging network, and ensuring that charging needs are met in a timely manner. In addition, the priority-based resource allocation method can flexibly respond to dynamic demand changes in different regions, help balance the grid load, reduce the impact of power fluctuations on charging piles, and improve the stability and sustainability of the system.

[0027] The optimization and layout adjustment of charging pile locations in S33 include: S331, determining the priority location distribution of charging piles: according to the priority sorting result of the charging piles, the charging piles are preferentially deployed in high priority areas; S332, adjust layout based on demand and flow: optimize the distribution of charging piles based on the surrounding vehicle flow and charging demand forecast results; Through the above content, efficient allocation of charging pile resources is achieved, ensuring the concentrated layout of charging piles in areas with higher demand, improving the coverage and response speed of charging services, while avoiding resource waste in low-demand areas, improving the utilization efficiency of charging piles, balancing the load, reducing the impact of power grid fluctuations, and enhancing the service capabilities of charging facilities and the stability of the power grid.

[0028] The demand-based and flow-adjusted layout in S332 includes: S3321, Demand and flow weight calculation: Calculate the comprehensive demand flow weight of each charging pile area based on the charging demand forecast results of the area and the surrounding vehicle flow. , expressed as: ; in, , are the corresponding weight coefficients, is the charging demand forecast result of the region, is the surrounding vehicle flow, and are the maximum values ​​of charging demand and flow rate in all areas respectively; S3322, Optimization of charging pile layout: based on comprehensive demand flow weight , optimize the layout of charging piles, when Exceeding high demand threshold When the area is defined as a high-demand and high-flow area, the density of charging piles is increased to meet higher charging and flow demands. Below low demand threshold When it is defined as a low-demand and low-flow area, the number of charging piles can be reduced or resource allocation can be optimized by sharing charging piles to avoid resource waste; High demand threshold , low demand threshold The settings include: Calculate the average and standard deviation of the comprehensive demand flow weight: According to the comprehensive demand flow weight of each area , calculate the mean of the comprehensive weights of all regions and standard deviation , expressed as: ; ; in, is the total number of regions, is the comprehensive demand flow weight; Setting a high demand threshold: High demand threshold The setting is based on the mean and standard deviation of the comprehensive weights, expressed as: ; in, is the adjustment coefficient, which ranges from 1 to 2; Setting a low demand threshold: Low demand threshold The setting is based on the mean and standard deviation of the comprehensive weights, expressed as: ; in, is the adjustment coefficient, which ranges from 1 to 2; Through the above content, the layout of charging piles can be accurately optimized, and high-demand and high-traffic areas and low-demand and low-traffic areas can be effectively identified, so as to achieve reasonable allocation of resources. For high-demand areas, the density of charging piles is increased to meet actual needs. For low-demand areas, resource input is reduced or utilization is increased through shared charging piles. This significantly improves the efficiency of charging pile resource utilization, optimizes the overall layout, reduces resource waste, and ensures the fairness of charging services and the balance of power grid load.

[0029] Load scheduling and resource allocation in S4 include: S41, load demand adjustment: based on the charging demand forecast results and the comprehensive demand flow weight of the charging pile space scheduling optimization , combined with the real-time load conditions of the power grid , dynamically adjust the load priority of each region to adapt to the load conditions of the power grid and regional needs; S42, resource allocation and execution: based on the adjusted regional load priority, calculate the resource allocation amount to ensure the efficiency and fairness of resource allocation; Through the above content, the charging needs of high-demand and high-weight areas can be met first, and the power distribution can be dynamically adjusted to avoid grid overload, thereby achieving efficient and fair resource allocation. Through real-time monitoring and feedback optimization, it is ensured that load scheduling can adapt to grid load fluctuations and regional demand changes, thereby improving the responsiveness of the charging network, resource utilization and overall service quality, while ensuring the stability and reliability of the grid.

[0030] Load demand adjustments in S41 include: S411, Assessment of regional demand and grid carrying capacity: Based on the comprehensive demand flow weight of the region And the real-time load of the power grid, by calculating the remaining carrying capacity of the power grid, it is determined whether the regional load needs to be restricted or adjusted, which is expressed as: ; in, Indicates the current remaining carrying capacity of the power grid. is the maximum load carrying capacity of the power grid, is the real-time load of the current power grid; S412, dynamically adjust regional priorities: dynamically adjust the load priority of each region based on the comprehensive demand flow weight and the remaining carrying capacity of the power grid; ; in, Prioritize load demands for the region; Through the above content, the remaining carrying capacity of the power grid is evaluated in real time to ensure that the power grid will not be overloaded when meeting the charging needs of high-demand areas. The dynamic priority adjustment mechanism enables high-demand areas to be allocated resources first, while resource usage in low-demand areas is optimized, thereby achieving efficient coordination between the charging pile network and the power grid, improving the flexibility of resource allocation and the stability of system operation.

[0031] Resource allocation and execution in S42 include: S421, calculating resource allocation: according to the adjusted regional priority and the total available resources, calculating the resource allocation of each region, expressed as: ; in, is the amount of resources allocated to the region, is the total resources currently available in the power grid, is the total number of regions; S422, executing resource allocation: allocating the calculated resource allocation amount to charging piles in each area; Through the above content, we ensure that resource allocation matches the actual needs of the region, so that high-priority and high-demand areas can obtain more adequate resource support, while ensuring that the basic resource allocation of low-demand areas is not ignored, and timely adjust the allocation strategy according to changes in grid load and demand, effectively avoiding resource waste and grid overload, and improving the overall resource utilization efficiency, charging service capabilities, and grid operation stability and flexibility.

[0032] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A new energy vehicle charging pile load balancing and scheduling method based on an intelligent algorithm, characterized in that: The following steps are involved: S1, data collection: real-time collection of charging pile operation data, including usage, surrounding vehicle flow, battery status and grid load; S2, charging demand prediction: Based on the collected operation data of charging piles, the charging demand of each charging pile within the future scheduled time is predicted through the graph convolutional network model; S3, Optimization of charging pile spatial scheduling: Based on the geographical distribution of charging piles, the surrounding vehicle flow and the results of charging demand forecast, the scheduling of charging piles is spatially optimized, including: S31, regional demand assessment: based on the charging demand forecast results, assess the demand in the area where the charging piles are located; S32, scheduling priority sorting: according to the results of regional demand assessment, priority is set for charging pile resource allocation to optimize resource distribution; S33, charging pile location optimization and layout adjustment: based on the results of scheduling priority sorting, combined with the surrounding vehicle flow and charging demand prediction results, optimize the layout and location of the charging piles; S4, load scheduling and resource allocation: Based on the results of charging demand prediction and charging pile space scheduling optimization, combined with the grid load situation, resource allocation is carried out; The charging demand prediction in S2 includes: S21, constructing a graph structure of charging piles: constructing a graph structure of charging piles according to the geographical locations and spatial relationships between charging piles, treating charging piles as nodes in the graph, and node features including historical charging pile usage, surrounding vehicle traffic, battery status, and grid load. Edges between charging piles represent spatial or traffic dependencies between charging piles, and edge weights are assigned by Euclidean distance and surrounding vehicle traffic. S22, graph convolution operation: In the graph convolution network model, the graph convolution operation is used to process the charging pile graph structure. The graph convolution operation is performed through the adjacency matrix and the node feature matrix To spread; S23, Charging demand prediction: In the last layer of the graph convolutional network model, the node features are mapped to the predicted values ​​through the fully connected layer, that is, the charging demand of each charging pile within the scheduled time in the future.

2. The method for load balancing and scheduling of new energy vehicle charging piles based on intelligent algorithm according to claim 1 is characterized in that: The data collection in S1 includes: S11, Charging pile usage: through the smart terminal installed on the charging pile, the usage status, charging power, charging time and number of users of the charging pile are monitored in real time; S12, surrounding vehicle flow: by installing vehicle flow monitoring equipment around the charging pile, the vehicle flow data of the roads around the charging pile is collected in real time, and combined with the real-time vehicle flow information provided by the traffic data platform, it is dynamically updated; S13, battery status: obtain the battery power, charging status and temperature of the vehicle battery in real time through the new energy vehicle battery management system connected to the charging pile; S14, grid load: Power load monitoring sensors installed at key grid nodes are used to obtain grid load conditions, voltage fluctuations, and current data in real time.

3. The method for load balancing and scheduling of new energy vehicle charging piles based on intelligent algorithm according to claim 1 is characterized in that: The regional needs assessment in S31 includes: S311, collecting charging demand prediction data: collecting the predicted charging demand of each charging pile; S312, calculating the weight of the charging pile: by considering the traffic flow around the charging pile and geographic location to calculate the weight of the charging pile ; S313, evaluate the total charging demand in the region: based on the predicted charging demand and corresponding weight of each charging pile , evaluate the charging demand of the entire region, the total regional charging demand It is the weighted average of the demand for charging piles in the region.

4. The method for load balancing and scheduling of new energy vehicle charging piles based on intelligent algorithm according to claim 3 is characterized in that: The scheduling priority sorting in S32 includes: S321, calculate charging pile priority: based on regional demand assessment results and the predicted charging demand of each charging pile, and calculate the priority score of each charging pile , prioritize the charging piles; S322, Priority sorting: based on the calculated charging pile priority score , sort all charging piles, and the charging piles with high priority are placed in front. The priority sorting of charging piles determines the order of resource allocation; S323, resource allocation optimization: During the scheduling process, charging resources are allocated in descending order according to the priority sorting results. When resources are limited, charging piles with high priority are given priority.

5. The method for load balancing and scheduling of new energy vehicle charging piles based on intelligent algorithm according to claim 4 is characterized in that: The charging pile location optimization and layout adjustment in S33 include: S331, determining the priority location distribution of charging piles: according to the priority sorting result of the charging piles, the charging piles are preferentially deployed in high priority areas; S332, adjust layout based on demand and flow: optimize the distribution of charging piles based on surrounding vehicle flow and charging demand forecast results.

6. The method for load balancing and scheduling of new energy vehicle charging piles based on intelligent algorithm according to claim 5 is characterized in that: The demand- and traffic-based layout adjustment in S332 includes: S3321, Demand and flow weight calculation: Calculate the comprehensive demand flow weight of each charging pile area based on the charging demand forecast results of the area and the surrounding vehicle flow. ; S3322, Optimization of charging pile layout: based on comprehensive demand flow weight , optimize the layout of charging piles, when Exceeding high demand threshold When the area is defined as a high-demand high-flow area, the density of charging piles is increased. Below low demand threshold When defining areas with low demand and low traffic, the number of charging piles can be reduced or resource allocation can be optimized by sharing charging piles.

7. The method for load balancing and scheduling of new energy vehicle charging piles based on intelligent algorithm according to claim 6 is characterized in that: The load scheduling and resource allocation in S4 include: S41, load demand adjustment: based on the charging demand forecast results and the comprehensive demand flow weight of the charging pile space scheduling optimization , combined with the real-time load conditions of the power grid , dynamically adjust the load priority of each area; S42, resource allocation and execution: Calculate resource allocation amounts based on the adjusted regional load priorities.

8. The method for load balancing and scheduling of new energy vehicle charging piles based on intelligent algorithm according to claim 7 is characterized in that: The load demand adjustment in S41 includes: S411, Assessment of regional demand and grid carrying capacity: Based on the comprehensive demand flow weight of the region and the real-time load situation of the power grid, and by calculating the remaining carrying capacity of the power grid, determine whether it is necessary to limit or adjust the regional load; S412, dynamically adjust regional priorities: dynamically adjust the load priority of each region based on the comprehensive demand flow weight and the remaining carrying capacity of the power grid.

9. The method for load balancing and scheduling of new energy vehicle charging piles based on intelligent algorithm according to claim 8 is characterized in that: The resource allocation and execution in S42 include: S421, calculating resource allocation: calculating the resource allocation of each region according to the adjusted regional priority and the total available resources; S422, executing resource allocation: allocating the calculated resource allocation amount to the charging piles in each area.

Images