New energy charging pile management and distribution method and system

Abstract

The invention relates to the technical field of electric vehicle charging pile management and distribution, and discloses a new energy charging pile management and distribution method and system, and the method comprises the steps: collecting charging user data, position data and environment data, and carrying out the preprocessing; and analyzing the preprocessed data, and constructing a prediction model of the charging demand of the user. And calculating the load condition of each charging pile according to a demand result predicted by the model, and adjusting a charging pile distribution strategy. According to the invention, through the steps of data preprocessing, charging demand prediction, load calculation and distribution strategy optimization, real-time monitoring and feedback and the like, intelligent management of charging pile resources is realized. The method plays a key role in improving the efficiency of the charging service, reducing overload and idle phenomena, optimizing the user experience and the like. A more intelligent, efficient and personalized charging pile management and distribution system is provided for electric vehicle users, the utilization rate of charging pile resources is effectively improved, the waiting time of the users is shortened, and the flexibility and stability of the system are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of electric vehicle charging pile management and distribution, and particularly to a management and distribution method and system for new energy charging piles. Background Art

[0002] With the popularization of new energy vehicles and the rapid development of the electric vehicle (EV) industry, the construction of charging infrastructure has gradually become one of the core issues in the development of electric vehicles. Especially in key areas such as cities and highways, the number and layout of charging piles directly affect the charging efficiency of electric vehicles and the user experience. In recent years, with the continuous progress of charging pile technology, intelligent charging pile systems have emerged. These systems have functions such as intelligent scheduling, remote monitoring, and load management. At the same time, in order to improve the utilization efficiency of charging piles, many studies and practices have begun to focus on the management and distribution methods of charging piles based on factors such as user demand prediction, geographical location, and traffic flow. The gradual maturity of these technologies has made the electric vehicle charging pile management system more intelligent and automated, which can alleviate the problem of uneven distribution of charging pile resources to a certain extent and provide more flexible and efficient charging services for electric vehicle users.

[0003] Although the existing charging pile management and distribution technologies have improved the resource utilization rate of charging piles to a certain extent, there are still many deficiencies, especially in intelligent management, load balancing, and dynamic scheduling. First of all, most of the existing technologies adopt charging pile allocation methods based on predetermined rules. These methods mainly perform static allocation according to the idle situation of charging piles, while ignoring the real-time changes in user demands. Since the charging demands of electric vehicles have strong time-periodicity and geographical dependence, the traditional static allocation methods cannot accurately reflect the dynamic changes in user demands, resulting in some charging piles being overloaded during peak hours and resource idleness during off-peak hours, thus affecting the optimal allocation of resources.

[0004] Secondly, the existing charging pile management systems lack real-time monitoring and feedback on environmental changes (such as traffic flow, weather changes, etc.) and user behaviors. This makes the existing systems lack flexible adjustment capabilities when facing emergencies or external environmental changes. For example, in case of bad weather or traffic jams, the charging demands of users may change significantly, but the existing systems usually cannot dynamically adjust the charging pile allocation strategy to cope with these changes. Furthermore, the existing technologies often do not combine the personalized needs and preferences of users. For example, users may prefer to choose charging piles at specific locations, but these preferences are not effectively utilized, resulting in a poor user experience.

[0005] Finally, the prior art lacks a high degree of optimization and refined control in the load management of charging piles. Although some systems attempt to use load balancing algorithms for scheduling, these algorithms are usually too simple to comprehensively consider multiple factors such as geographical location, environmental factors, charging demand, and the actual load capacity of charging piles. Therefore, the prior art has certain limitations in efficiently utilizing charging pile resources and improving the quality and efficiency of charging services.

[0006] In contrast, the present invention can achieve precise scheduling and optimized allocation of charging pile resources by introducing a dynamic allocation strategy based on user demand prediction and a load balancing algorithm that comprehensively considers geographical location and environmental factors, avoiding problems such as static allocation and uneven load in the prior art, thereby improving the efficiency of charging services and the user experience. Summary of the Invention

[0007] In view of the above problems, the present invention is proposed.

[0008] Therefore, the technical problems solved by the present invention are: the existing charging pile resource allocation methods have defects such as static allocation, uneven load, and failure to adjust in real time. The prior art cannot effectively combine user demand, environmental factors, and the load situation of charging piles for intelligent and dynamic scheduling, and how to optimize the resource allocation of charging piles to achieve real-time balance of charging pile load and efficient utilization of resources.

[0009] To solve the above technical problems, the present invention provides the following technical solutions: A new energy charging pile management and allocation method, including: Collect charging user data, location data, and environmental data, and perform preprocessing.

[0010] Analyze the preprocessed data and construct a prediction model for user charging demand.

[0011] According to the demand results predicted by the model, calculate the load situation of each charging pile and adjust the charging pile allocation strategy.

[0012] As a preferred solution of the new energy charging pile management and allocation method described in the present invention, wherein: The collection of charging user data, location data, and environmental data includes: The charging user data includes historical charging data and user behavior data.

[0013] The historical charging data includes the charging time, charging duration, charging power, and charging frequency of the user. The user behavior data includes the user's reserved charging behavior, preference for selecting charging piles, and power demand during charging.

[0014] The location data includes the location of the charging pile and the user's historical charging location.

[0015] Environmental data includes weather data, traffic flow data, and power grid load data.

[0016] Weather data includes temperature, humidity, wind elements, air pressure, and precipitation. Traffic flow data includes the traffic volume on roads and in regions.

[0017] Perform different preprocessing on different data types.

[0018] As a preferred solution of the new energy charging pile management and allocation method described in the present invention, wherein: analyzing the preprocessed data and constructing a prediction model for user charging demand includes: By normalizing and processing historical charging data, obtain the charging demand characteristics of each user at time t, including the user charging timestamp , charging duration , charging power and charging frequency .

[0019] Conduct an analysis of the influence of historical data of user charging behavior, and construct an influence function of charging historical data on charging demand, expressed as: ; wherein, represents the influence of historical charging data on charging demand, represents the i-th charging data item, represents the time decay factor of each feature, represents the adjustment index for adjusting the influence of charging data.

[0020] Conduct an analysis of the influence of user behavior data, where user behavior data includes user reservation charging behavior and user's preference for charging pile selection , and construct an influence function of the user's behavior data on charging demand, expressed as: ; wherein, represents the influence of user behavior data on charging demand, represents the i-th behavior data item, represents the weight of each feature, represents the preferred location of the user, represents the location of the reference charging pile, represents the adjustment factor for user preference and location influence, represents the adjustment factor for controlling the contribution of behavior data to demand.

[0021] As a preferred solution of the new energy charging pile management and allocation method described in the present invention, wherein: the prediction model for user charging demand further includes: Analyze the impact of environmental factors, which include weather data and traffic flow. Then introduce the power grid load data and construct an impact function of environmental factor data on charging demand, expressed as: ; Among them, represents the impact of environmental data on charging demand, represents weather data, represents traffic flow data, represents power grid load data, represents the impact weight of environmental data on demand, represents the adjustment factor for influencing the environmental distance, represents the distance between the reference charging pile and the actual charging pile location.

[0022] Integrate each impact function and obtain the final charging demand prediction through weighted average, expressed as: ; Among them, represents the charging demand prediction at time t, represents the attenuation coefficient, which controls the impact of time on the prediction, represents the overall impact adjustment factor.

[0023] As a preferred solution of the new energy charging pile management and allocation method described in the present invention, among them: calculating the load conditions of each charging pile according to the demand result predicted by the model includes: According to the demand result output by the prediction model, combined with the actual available capacity of the charging pile, calculate the load of each charging pile at the current moment. The load calculation considers the spatial distribution of the charging piles, environmental conditions, and the spatial distribution of users.

[0024] Construct an exponential decay function to represent the impact of spatial distance on the load of the charging pile, expressed as: ; Among them, represents the geographical location of charging pile j at time t, represents the geographical location of user i at time t, represents the spatial distance between the charging pile and the user, represents the coefficient for adjusting the attenuation of the spatial impact, represents the coefficient for adjusting the attenuation of the spatial impact.

[0025] Construct a weighted Euclidean distance model to represent the impact of environmental factors on the load of the charging pile, expressed as: ;

[0026] Among them, represents the weather data at time t, represents the traffic flow data at time t, represents adjusting the influence of environmental factors on the load of the charging pile, represents adjusting the influence of environmental factors on the load of the charging pile.

[0027] As a preferred scheme of the new energy charging pile management and allocation method described in the present invention, among them: calculating the load conditions of each charging pile includes:

[0028] The load calculation formula is expressed as: ; Among them, represents the load condition of charging pile j at time t, represents the predicted charging demand of user i at time t, represents the indicator function. If user i uses charging pile j, then = 1, otherwise 0. represents the spatial influence function between user i and charging pile j, represents the environmental factor function, represents the maximum load capacity of charging pile j.

[0029] As a preferred scheme of the new energy charging pile management and allocation method described in the present invention, among them: the adjustment of the charging pile allocation strategy includes: Define the allocation strategy as the load allocation weight of charging pile j at time t, representing the load adjustment weight of the charging pile, which affects the load allocation ratio of the charging pile.

[0030] Adjust the allocation strategy by considering the load of the charging pile, the spatial distance between the charging pile and the reference position, and environmental factors. The allocation strategy optimization formula is expressed as: ; Among them, represents the allocation strategy of charging pile j at time t, represents the load of charging pile j, represents the spatial position of charging pile j at time t, represents the spatial position of reference charging pile r at time t. represents the rate of controlling the distance attenuation, represents the sensitivity of controlling the spatial influence, represents the total number of all charging piles, represents adjusting the influence of environmental factors on the allocation strategy, represents adjusting the influence of environmental factors on the allocation strategy.

[0031] As a preferred solution of the new energy charging pile management and allocation method described in the present invention, wherein: the adjustment of the charging pile allocation strategy includes: After the allocation strategy is determined, it is dynamically adjusted according to real-time demand prediction data. Assume that the real-time demand prediction data is , representing the latest charging demand prediction of user i at time t, and calculate the allocation weight of charging pile j: ; Wherein, represents the allocation weight of charging pile j, represents the charging demand of user i at time t predicted in real time, represents the relationship between controlling the charging pile load and the location, represents the relationship between controlling the charging pile load and the location, represents controlling the influence of environmental factors on the allocation strategy, represents controlling the influence of environmental factors on the allocation strategy, represents controlling the influence of real-time demand prediction on the allocation strategy.

[0032] As a preferred solution of the new energy charging pile management and allocation method described in the present invention, wherein: The automatic adjustment of the charging pile allocation strategy includes: The calculation result of the allocation weight of charging pile j The value range is [0, 1]. When it approaches 1, it means that the charging pile has a light load and a suitable location, and is allowed to be assigned higher charging requests.

[0033] When it approaches 0, it means that the charging pile has a high load and is far from the demand point, and the allocation of charging requests needs to be reduced.

[0034] The steps of adjusting the charging pile allocation strategy include calculating the allocation weight of each charging pile, calculating the number of users actually allocated to each charging pile based on the allocation weight, comparing the newly calculated number of users with the charging pile load. If it exceeds the maximum load of the charging pile, load redistribution is performed, which is expressed as: ; Wherein, represents the number of users after redistribution, represents the number of users actually allocated to each charging pile calculated based on the allocation weight, represents the maximum load of the charging pile.

[0035] Under the new allocation strategy, calculate the new load situation of the charging pile.

[0036] A new energy charging pile management and distribution system includes A preprocessing module that collects charging user data, location data, and environmental data for preprocessing.

[0037] A prediction module that analyzes the preprocessed data and constructs a prediction model for user charging demand.

[0038] An adjustment module that calculates the load conditions of each charging pile based on the demand results predicted by the model and adjusts the charging pile distribution strategy.

[0039] A computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the method described above.

[0040] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the steps of the method described above.

[0041] The beneficial effects of the present invention: Through steps such as data preprocessing, charging demand prediction, load calculation and distribution strategy optimization, real-time monitoring and feedback, etc., the present invention realizes the intelligent management of charging pile resources. Each step plays a key role in improving the efficiency of charging services, reducing overload and idle phenomena, and optimizing the user experience. Ultimately, the present invention provides a more intelligent, efficient, and personalized charging pile management and distribution system for electric vehicle users, effectively improving the utilization rate of charging pile resources, reducing the user waiting time, and enhancing the flexibility and stability of the system. Description of the Drawings

[0042] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. Among them: Figure 1 It is the overall flowchart of a new energy charging pile management and distribution method and system provided by the first embodiment of the present invention. Detailed Embodiments

[0043] To make the above-mentioned objects, features, and advantages of the present invention more obvious and understandable, the following will describe the detailed embodiments of the present invention in conjunction with the drawings of the specification. Obviously, the described embodiments are some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0044] Embodiment 1, referring toFigure 1 , as an embodiment of the present invention, provides a new energy charging pile management and allocation method, including: S1: Collect charging user data, location data, and environmental data, and perform preprocessing.

[0045] The charging user data includes historical charging data and user behavior data.

[0046] The historical charging data includes the charging time, charging duration, charging power, and charging frequency of the user. The user behavior data includes the user's reservation charging behavior, preference for selecting a charging pile, and power demand during charging.

[0047] The location data includes the location of the charging pile and the user's historical charging location.

[0048] The environmental data includes weather data, traffic flow data, and grid load data.

[0049] The weather data includes temperature, humidity, wind factor, air pressure, and precipitation. The traffic flow data includes the traffic flow of roads and regions.

[0050] Perform different preprocessing on different data types.

[0051] Furthermore, perform data cleaning, normalization, time series, and missing value processing on the historical charging data, perform data cleaning, feature extraction, preference analysis, and missing value filling on the user behavior data, perform data cleaning, spatial clustering, and time synchronization on the location data, perform data cleaning and time synchronization on the weather data, perform data cleaning, time-space alignment, and feature extraction on the traffic flow data, and perform data cleaning, normalization, and time synchronization on the grid load data.

[0052] It should be noted that when users select a charging pile, they often base on various factors such as personal habits, historical charging experience, charging pile location, and queuing time. In order to quantify the user's preference for a certain charging pile, a preference index is calculated to represent the user's charging tendency on a certain charging pile.

[0053] The user's charging preference can be calculated through their historical selection records, considering factors including: the number of times N that the user has selected charging pile j in the past T days

[0054] ,

[0048] ,

[0049] ,

[0052] ,

[0053] , j ,

[0050] , u ,

[0051] , , , , j , , j , , u,j , , the total number of charging times N of the user in this area u , the charging availability A of this charging pile j (that is, whether there is an idle when the user goes to this pile). The average charging queuing time W of this charging pile j . The historical reliability R of the charging pile j (that is, the stability of this pile, for example, a pile with a lower failure rate is more favored).

[0054] The preference index calculation formula is expressed as: ; where, represents the preference index of user u for charging pile j, and the range is [0, 1]. represents the number of times user u has selected charging pile j in the past T days. represents the total number of charging times of user u within T days. represents the availability rate (unit: 0 - 1) of charging pile j when the user attempts to charge. represents the average waiting time of charging pile j (unit: minutes). represents the historical reliability of charging pile j (0 - 1), and the higher the value, the lower the failure rate.

[0055] Furthermore, based on the actual selection behavior of users, personal preferences are quantified. Factors such as availability, queuing time, and reliability are comprehensively considered. As the user's selection changes, this index can be updated over time.

[0056] S2: Analyze the pre - processed data and construct a prediction model for users' charging demands.

[0057] By normalizing and processing the historical charging data, the charging demand characteristics of each user at time t are obtained, including the user's charging timestamp , charging duration , charging power and charging frequency .

[0058] Conduct an analysis of the influence of historical data on users' charging behaviors. Historical charging data is an important basis for analyzing users' charging behaviors. Through historical data such as users' charging frequencies, charging durations, power, and charging times, users' charging habits and demand patterns can be revealed. First, it is necessary to extract the characteristics of users' historical charging data and convert it into digital data that can be analyzed. For example, the user's historical charging times (frequencies), average charging duration, and average charging power are used as input parameters for charging demand prediction. In addition, a time decay coefficient can be introduced. Considering that the influence of historical behaviors on future demands is non - linear, a gradually decaying function, such as exponential decay or logarithmic decay, is introduced to reflect the time sensitivity of users' habits. Construct an influence function of historical charging data on charging demand, expressed as: ; where, represents the influence of historical charging data on charging demand, represents the i - th charging data item, represents the time decay factor of each feature, represents the adjustment index used to adjust the influence of charging data.

[0059] Furthermore, by analyzing the historical data of users, the future demands of each user can be accurately predicted, improving the accuracy of charging pile allocation. After introducing the decay factor, over-reliance on past charging behaviors can be avoided, adapting to changes in user behaviors and preventing the impact of outdated data on predictions. By accurately predicting user demands, the load of charging piles can be more reasonably allocated, thus improving the utilization efficiency of resources and reducing overload and idle phenomena.

[0060] Conduct an impact analysis of user behavior data, which includes users' reservation behaviors, selection preferences, charging time period selections, etc. These behavior data reflect users' charging preferences and demand patterns. By analyzing these behavior data, the charging demand patterns of users can be extracted. For example, reserved charging can provide more information for prediction. Especially during high-demand periods, users' reservation behaviors can reflect in advance the increase in charging demands. Preference data (such as users' preference for charging piles at specific locations) can help the system make more personalized decisions during resource allocation. Design an impact function based on behavior data, such as adjusting the allocation strategy according to the reservation frequency or preference degree of users. User behavior data includes users' reserved charging behaviors and users' selection preferences for charging piles , construct an impact function of user behavior data on charging demand, expressed as: ; where represents the impact of user behavior data on charging demand, represents the i-th behavior data item, represents the weight of each feature, represents the preferred location of the user, represents the location of the reference charging pile, represents the adjustment factor for user preference and location impact, represents the adjustment factor, controlling the contribution of behavior data to demand.

[0061] Furthermore, this calculation can provide personalized charging pile allocation according to users' charging behaviors (such as reservation behaviors, preferences, etc.), enhancing the user experience. Combining users' behavior patterns makes the prediction of charging demands more accurate, reducing resource waste and overload problems. By analyzing reservation data, the system can predict demand fluctuations in advance and make corresponding preparations, reducing the pressure during peak periods.

[0062] Conduct an impact analysis of environmental factors. Environmental factors (such as weather, traffic flow, etc.) will affect users' charging demands to a certain extent. For example, extreme weather (such as rain and snow) may lead to a decrease in users going out, resulting in a decline in charging demands. While during peak traffic periods or hot weather, users' charging demands may surge. Therefore, when designing, factors such as weather data, traffic flow, time periods, etc. need to be considered. Combining these environmental change factors, establish a multi-dimensional impact function to better predict charging demands. Use a weighted average model to comprehensively consider the impacts of different environmental factors on demands. Environmental factors include weather data and traffic flow, and then introduce grid load data to construct an impact function of environmental factor data on charging demands, expressed as: ; Among them, represents the impact of environmental data on charging demands, represents weather data, represents traffic flow data, represents grid load data, represents the impact weight of environmental data on demands, represents the adjustment factor for influencing environmental distance, represents the distance between the reference charging pile and the actual charging pile location.

[0063] It should be noted that this calculation fully considers the impact of the external environment on charging demands and avoids prediction deviations caused by a single data source. According to real-time weather and traffic flow data, the system can dynamically adjust the charging demand prediction to adapt to sudden demand fluctuations. The system can flexibly adjust the charging pile allocation strategy according to environmental changes, improving the flexibility and robustness of the system.

[0064] Integrate each impact function and obtain the final charging demand prediction through weighted average, expressed as: ; Among them, represents the charging demand prediction at time t, represents the attenuation coefficient, controlling the impact of time on the prediction, represents the overall impact adjustment factor.

[0065] S3: According to the demand results predicted by the model, calculate the load conditions of each charging pile and adjust the charging pile allocation strategy.

[0066] According to the demand results output by the prediction model and combined with the actual available capacity of the charging piles, calculate the load of each charging pile at the current moment. The load calculation considers the spatial distribution of the charging piles, environmental conditions, and the spatial distribution of users.

[0067] Construct an exponential decay function to represent the impact of spatial distance on the load of charging piles, expressed as: ; Wherein, represents the geographical location of charging pile j at time t, represents the geographical location of user i at time t, represents the spatial distance between the charging pile and the user, represents the coefficient for adjusting the attenuation of spatial influence, represents the coefficient for adjusting the attenuation of spatial influence.

[0068] Construct a weighted Euclidean distance model to represent the impact of environmental factors on the load of charging piles, expressed as: ; Wherein, represents the weather data at time t, represents the traffic flow data at time t, represents the adjustment of the impact of environmental factors on the load of charging piles, represents the adjustment of the impact of environmental factors on the load of charging piles.

[0069] It should be noted that the factor of spatial distance affecting the load of charging piles is obvious. Users tend to choose charging piles closer to themselves. Therefore, the geographical location of the charging pile determines its load situation. During design, we can simulate this impact through the attenuation function of spatial distance. In addition, environmental factors such as weather and traffic flow also affect the load of charging piles. For example, during peak traffic hours or in bad weather, the load of charging piles in certain areas will increase significantly. By combining these environmental factors with the load of charging piles, a more reasonable load prediction model can be obtained. The load calculation formula is expressed as: ; Wherein, represents the load situation of charging pile j at time t, represents the predicted charging demand of user i at time t, represents the indicator function. If user i uses charging pile j, then = 1, otherwise 0. represents the spatial influence function between user i and charging pile j, represents the environmental factor function, represents the maximum load capacity of charging pile j.

[0070] It should be noted that combining location and environmental factors can make the load prediction more accurate.

[0071] Define the allocation strategy $\omega_{j}(t)$ represents the load distribution weight of charging pile $j$ at time $t$, which is the load adjustment weight of the charging pile and affects the load distribution ratio of this charging pile.

[0072] It should be noted that the optimization of the allocation strategy dynamically adjusts the task allocation of charging piles according to the load conditions of charging piles and user needs to achieve load balancing and resource optimization. When designing, multiple factors need to be considered, such as the maximum load capacity, distance, environmental factors, and user preferences of charging piles. By introducing a load balancing algorithm, we can optimize the charging pile allocation based on real-time data. This optimization process not only focuses on the load of a single charging pile but also considers the mutual relationship between multiple charging piles to ensure the overall load balancing of the system. Considering the load of charging piles, the spatial distance between the charging pile and the reference position, and environmental factors to adjust the allocation strategy, the optimization formula of the allocation strategy is expressed as: ; where $\pi_{j}(t)$ represents the allocation strategy of charging pile $j$ at time $t$, $L_{j}$ represents the load of charging pile $j$, $S_{j}(t)$ represents the spatial position of charging pile $j$ at time $t$, $S_{r}(t)$ represents the spatial position of reference charging pile $r$ at time $t$. $\alpha$ represents the rate of controlling the distance attenuation, $\beta$ represents the sensitivity of controlling the spatial influence, $N$ represents the total number of all charging piles, $\gamma_{1}$ represents adjusting the influence of environmental factors on the allocation strategy, $\gamma_{2}$ represents adjusting the influence of environmental factors on the allocation strategy.

[0073] It should be noted that load balancing can avoid overloading of some charging piles. Spatial optimization can reduce the user waiting time.

[0074] It should be noted that the charging pile allocation weight is a key factor in determining the priority of each charging pile. When designing, we need to comprehensively consider multiple dimensions such as the actual load, geographical location, environmental factors, and user behavior of the charging pile. By introducing a dynamic adjustment factor, the allocation weight not only depends on the current load of the charging pile but also takes into account the changes in the external environment and the personalized needs of users. After the allocation strategy is determined, it is dynamically adjusted according to the real-time demand prediction data. Assuming the real-time demand prediction data is $d_{i}(t)$, which represents the latest charging demand prediction of user $i$ at time $t$, calculate the allocation weight of charging pile $j$: ; where $\omega_{j}$ represents the allocation weight of charging pile $j$, $d_{i}(t)$ represents the charging demand of user $i$ at time $t$ predicted in real time, Indicates the relationship between controlling the load and location of the charging pile, Indicates the relationship between controlling the load and location of the charging pile, Indicates the impact of controlling environmental factors on the allocation strategy, Indicates the impact of controlling environmental factors on the allocation strategy, Indicates the impact of controlling real-time demand forecasting on the allocation strategy.

[0075] It should be noted that, combining the load and demand, dynamically adjust the usage amount of the charging pile. Avoid overloading some charging piles and improve the charging efficiency.

[0076] The calculation result of the allocation weight of charging pile j The value range is [0,1], When approaching 1, it indicates that the charging pile has a light load and a suitable location, and is allowed to be assigned higher charging requests.

[0077] When approaching 0, it indicates that the charging pile has a high load and is far from the demand point, and the allocation of charging requests needs to be reduced.

[0078] The steps to adjust the charging pile allocation strategy include calculating the allocation weight of each charging pile, and calculating the actual number of users assigned to each charging pile based on the allocation weight, which is expressed as: ; Among them, Indicates the expected number of users assigned to charging pile j (unit: number). [[ID=3A]]Indicates the allocation weight of the charging pile. Indicates the total number of users applying for charging at time t.

[0079] If the calculated of a certain charging pile j exceeds its maximum load , then load reallocation is required, which is expressed as: ; Among them, Indicates the number of users after reallocation, Indicates the actual number of users assigned to each charging pile calculated based on the allocation weight, Indicates the maximum load of the charging pile.

[0080] The excess user demand will be assigned to other charging piles with lower loads.

[0081] Under the new allocation strategy, calculate the new load situation of the charging pile: ; Among them, Indicates the new load of charging pile j after adjusting the allocation. Represents the adjusted number of users. Represents the estimated charging demand of user j.

[0082] It should be noted that according to the calculated and , the charging management system will take the following automatic adjustment measures: Dynamic charging pile recommendation: If a certain charging pile is overloaded (i.e., >0.9 ), the system will recommend to users to go to the charging piles with lighter loads on the APP or the display screen of the charging station.

[0083] Intelligent queuing adjustment: For users who cannot charge immediately, the system will arrange the optimal charging queue based on the sorting to reduce the waiting time.

[0084] Load balancing and shunting: If some charging piles are idle ( <0.3 ), the system will offer price discounts to attract users to go to the charging piles with lower loads.

[0085] It should be noted that all the influencing parameters in the present invention are set by those skilled in the art according to experience and historical data, and can be adjusted according to the situation in specific implementations.

[0086] In the above embodiments, there is also a new energy charging pile management and distribution system, specifically: A preprocessing module, which collects charging user data, location data, and environmental data and performs preprocessing.

[0087] A prediction module, which analyzes the preprocessed data and constructs a prediction model for users' charging demands.

[0088] An adjustment module, which calculates the load conditions of each charging pile according to the demand results predicted by the model and adjusts the charging pile allocation strategy.

[0089] The computer device can be a server. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the data cluster data of the power monitoring system. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals through a network connection. When the computer program is executed by the processor, it implements a new energy charging pile management and allocation method.

[0090] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., and are not limited thereto. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logics, data processing logics based on quantum computing, etc., and are not limited thereto.

[0091] Embodiment 2, which is an embodiment of the present invention, provides a new energy charging pile management and allocation method and system. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.

[0092] In this embodiment, the management and allocation process of a group of electric vehicle charging piles is simulated. Through steps such as data collection, demand prediction, load calculation, and allocation strategy optimization, the superiority and innovation of the present invention are demonstrated. Assuming the experimental scenario is the charging pile system of a certain city, the system involves multiple charging piles, and the following aspects are mainly investigated: the accuracy of charging demand prediction, the dynamic allocation of charging pile loads, and the optimized charging pile allocation strategy based on the environment and user behavior.

[0093] Experimental environment: Site: An area with 10 charging poles in a city was selected for the experiment. The charging poles were distributed at different locations to simulate the diversity of actual charging demands in the city.

[0094] Data collection: Historical usage data of the charging poles in this area was collected, including users' charging duration, charging frequency, reservation records, etc.; real-time weather and traffic flow data, as well as the load situation of each charging pole, were also considered.

[0095] Experimental process: Data preprocessing: First, the collected historical charging data was cleaned and normalized. The charging historical data includes each user's charging time, charging duration, charging power, and charging frequency. After standardization, the scale of the input data is unified, facilitating subsequent model training and prediction.

[0096] Charging demand prediction: Through a prediction model constructed based on user behavior and environmental factors (such as temperature, humidity, traffic flow, etc.), the demand for each charging pole in the next few time periods was predicted. Machine learning algorithms (such as regression analysis, decision trees, etc.) were used to model the historical data to obtain the predicted values of the charging demand for each charging pole.

[0097] Load calculation and allocation optimization: According to the charging demand prediction results, the load situation of each charging pole was calculated, and the allocation strategy of the charging poles was adjusted through optimization algorithms (such as genetic algorithms, simulated annealing algorithms, etc.). Considering the geographical location, load capacity of the charging poles, and changes in the external environment, the allocation weights of the charging poles were adjusted to avoid overloading or idling.

[0098] Real-time monitoring and feedback adjustment: The system monitored the usage situation of each charging pole in real time, and dynamically adjusted the allocation strategy of the charging poles according to the actual charging demand and environmental factors (such as sudden weather changes or traffic congestion).

[0099] The experimental results are shown in Table 1.

[0100] Table 1 Experimental results

[0101] By analyzing the data in the experimental table, it is possible to intuitively see the load and allocation strategy of each charging pole under different environmental conditions.

[0102] The "predicted charging demand" column in the table shows the predicted charging demand (unit: kWh) of each charging pole in a specific time period, and the data reflects the change of demand. It can be seen that the charging poles with larger charging demands (such as charging pole 6 and charging pole 10) are usually located in areas with larger traffic flows. By comparing with the "actual load" column, the effectiveness of the prediction model was verified. The actual load of most charging poles is close to the predicted value, indicating that the prediction model has a high accuracy.

[0103] Changes in temperature and traffic flow affect the usage of charging piles. Taking Charging Pile 3 as an example, its temperature is relatively high (33°C) and the traffic flow is relatively low (80 vehicles per hour). Although the predicted demand is 45 kWh, its actual load is 40 kW, indicating that high temperature may lead to a reduction in users going out for charging, and the load is slightly lower than the predicted value. For Charging Piles 6 and 9, the temperature is low and the traffic flow is high, so the actual load is closer to the predicted demand.

[0104] Through the load adjustment factor (the "charging pile load adjustment factor" column), the system makes dynamic adjustments according to the actual load of each charging pile and changes in environmental factors. The load adjustment factor of Charging Pile 6 is 1.4, indicating that the load capacity of its charging pile is preferentially increased, and the distribution strategy weight is 0.93. The relatively high distribution strategy weight shows that the system preferentially allocates more resources to charging piles with higher demand.

[0105] It can be seen from the "distribution strategy weight" column that for charging piles with relatively high demand (such as Charging Pile 10), the weight of the distribution strategy is relatively high (0.96), while for charging piles with relatively low demand (such as Charging Pile 7), the weight is relatively low (0.85). This optimization strategy effectively avoids the phenomena of charging pile overload or idleness and improves resource utilization.

[0106] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not restrictive. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A management and allocation method for new energy charging piles, characterized in that, Including: Collect charging user data, location data, and environmental data, and perform preprocessing; Analyze the preprocessed data and construct a prediction model for user charging demand; According to the demand results predicted by the model, calculate the load conditions of each charging pile and adjust the charging pile allocation strategy.

2. The new energy charging pile management and allocation method according to claim 1, characterized in that: The charging user data includes historical charging data and user behavior data; The historical charging data includes the charging time, charging duration, charging power, and charging frequency of the user; the user behavior data includes the user's reserved charging behavior, preference for selecting a charging pile, and power demand during charging; The location data includes the location of the charging pile and the user's historical charging location; The environmental data includes weather data, traffic flow data, and grid load data; The weather data includes temperature, humidity, wind factor, air pressure, and precipitation; the traffic flow data includes the traffic flow of roads and regions; Perform different preprocessing on different data types.

3. The new energy charging pile management and allocation method according to claim 2, characterized in that, The analysis of the preprocessed data and the construction of the prediction model for user charging demand include: By normalizing and processing historical charging data, the charging demand characteristics of each user at time t are obtained, including the user's charging timestamp , charging duration , charging power and charging frequency ; Conduct an analysis of the impact of historical data on user charging behavior, and construct an impact function of charging historical data on charging demand, expressed as: ； Among them, represents the impact of historical charging data on the charging demand, represents the i-th charging data item, represents the time decay factor of each feature, represents the adjustment index for adjusting the influence of charging data; Conduct an impact analysis of user behavior data, where the user behavior data includes users' reservation charging behavior and users' preference for charging pile selection , and construct an impact function of the user's behavior data on the charging demand, expressed as: ； Among them, represents the impact of user behavior data on the charging demand, represents the i-th behavior data item, represents the weight of each feature, represents the preferred location of the user, represents the location of the reference charging pile, represents the adjustment factor of user preference and location influence, represents the adjustment factor that controls the contribution of behavior data to the demand.

4. The new energy charging pile management and allocation method according to claim 3, wherein: The prediction model for user charging demand further includes: Conduct an analysis of the impact of environmental factors. The environmental factors include weather data and traffic flow, and then introduce grid load data to construct an impact function of environmental factor data on charging demand, expressed as: ； Among them, represents the impact of environmental data on the charging demand, represents weather data, represents traffic flow data, represents grid load data, represents the impact weight of environmental data on the demand, represents the adjustment factor affecting the environmental distance, represents the distance between the reference charging pile and the actual charging pile location; Integrate each impact function and obtain the final charging demand prediction through weighted averaging, expressed as: ； Among them, represents the predicted charging demand at time t, represents the attenuation coefficient, which controls the influence of time on the prediction, represents the overall influence adjustment factor.

5. The new energy charging pile management and allocation method according to claim 4, characterized in that: The calculation of the load conditions of each charging pile according to the demand results predicted by the model includes: According to the demand results output by the prediction model, combined with the actual available capacity of the charging pile, calculate the load of each charging pile at the current moment; the load calculation considers the spatial distribution of the charging piles, environmental conditions, and the spatial distribution of users; Construct an exponential decay function to represent the impact of spatial distance on the load of the charging pile, expressed as: ； Among them, represents the geographical location of charging pile j at time t, represents the geographical location of user i at time t, represents the spatial distance between the charging pile and the user, represents the coefficient for adjusting the attenuation of spatial influence, represents the coefficient for adjusting the attenuation of spatial influence; Construct a weighted Euclidean distance model to represent the impact of environmental factors on the load of the charging pile, expressed as: ； Among them, represents the weather data at time t, represents the traffic flow data at time t, represents adjusting the influence of environmental factors on the load of the charging pile, represents adjusting the influence of environmental factors on the load of the charging pile.

6. The new energy charging pile management and distribution method according to claim 5, characterized in that: The calculation of the load conditions of each charging pile includes: The load calculation formula is expressed as: ； Among them, represents the load condition of charging pile j at time t, represents the predicted charging demand of user i at time t, represents the indicator function. If user i uses charging pile j, then = 1, otherwise it is 0; represents the spatial influence function between user i and charging pile j, represents the environmental factor function, represents the maximum load capacity of charging pile j.

7. The new energy charging pile management and allocation method according to claim 6, characterized in that: The adjustment of the charging pile allocation strategy includes: Define the allocation strategy It represents the load allocation weight of charging pile j at time t, indicating the load adjustment weight of the charging pile and affecting the load allocation ratio of this charging pile; Consider the load of the charging pile, the spatial distance between the charging pile and the reference location, and environmental factors to adjust the allocation strategy. The allocation strategy optimization formula is expressed as: ； Among them, represents the allocation strategy of charging pile j at time t, represents the load of charging pile j, Denote the spatial position of charging pile j at time t, Denote the spatial position of reference charging pile r at time t; Denote the rate of controlling distance attenuation, Denote the sensitivity of controlling spatial influence, Denote the total number of all charging piles, Denote the influence of adjusting environmental factors on the allocation strategy, Denote the influence of adjusting environmental factors on the allocation strategy.

8. The new energy charging pile management and allocation method according to claim 7, wherein: The adjustment of the charging pile allocation strategy includes: After the allocation strategy is determined, it is dynamically adjusted according to the real-time demand prediction data. Assume that the real-time demand prediction data is , which represents the latest charging demand prediction of user i at time t, and calculate the allocation weight of charging pile j: ； Among them, represents the allocation weight of charging pile j, represents the charging demand of user i at time t predicted in real time, represents the relationship between controlling the load and location of the charging pile, represents the relationship between controlling the load and location of the charging pile, represents controlling the influence of environmental factors on the allocation strategy, represents controlling the influence of environmental factors on the allocation strategy, represents controlling the influence of real-time demand prediction on the allocation strategy.

9. The new energy charging pile management and allocation method according to claim 8, characterized in that: The automatic adjustment of the charging pile allocation strategy includes: Calculation result of the allocation weight of charging pile j The value range is [0, 1], When approaching 1, it indicates that the charging pile has a light load and a suitable location, and is allowed to be assigned higher charging requests; When it approaches 0, it means that the charging pile load is high and the distance from the demand point is far, and the allocation of charging requests needs to be reduced; The steps of adjusting the charging pile allocation strategy include calculating the allocation weight of each charging pile, calculating the actual number of users allocated to each charging pile based on the allocation weight, comparing the newly calculated number of users with the load of the charging pile. If it exceeds the maximum load of the charging pile, perform load redistribution, expressed as: ； Among them, represents the number of users after reallocation, represents the number of users actually allocated to each charging pile calculated based on the allocation weight, represents the maximum load of the charging pile; Under the new allocation strategy, calculate the new load conditions of the charging pile.

10. A new energy charging pile management and allocation system adopting the method according to any one of claims 1-9, characterized in that: A preprocessing module that collects charging user data, location data, and environmental data and performs preprocessing; A prediction module that analyzes the preprocessed data and constructs a prediction model for user charging demand; Adjustment module, according to the demand results predicted by the model, calculates the load conditions of each charging pile and adjusts the charging pile allocation strategy.

Images