An intelligent prediction method and system for the load of a vehicle charging station

Abstract

The present invention discloses an intelligent prediction method and system for the load of vehicle charging stations, which relates to the field of new energy technologies. First, the charging status information of each vehicle charging station within the target range is collected to construct the load intensity coefficient Xfh of a number of vehicle charging stations. After comparison, the number group of high-load vehicle charging stations is marked, and a load demand prediction instruction is issued. The charging piles within the number group of high-load vehicle charging stations are continuously monitored, and in combination with radio frequency identification technology, the load demand index Zyc is obtained by fitting. After comparing it with the threshold, it is judged whether the vehicle charging stations within the number group of high-load vehicle charging stations meet the vehicle charging demand of the day, so as to generate and execute corresponding level control instructions. This method effectively optimizes the resource allocation of vehicle charging stations and improves the charging user experience through real-time monitoring, dynamic analysis and precise control.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy, and particularly to an intelligent load prediction method and system for vehicle charging stations. Background Art

[0002] With the rapid development of global new energy technology, electric vehicles, as low-carbon and environmentally friendly means of transportation, have gradually occupied the mainstream market. Different from traditional fuel vehicles, the operation of electric vehicles highly depends on charging infrastructure. As the core support, vehicle charging stations directly affect the popularization of electric vehicles and user experience. In recent years, the number of vehicle charging stations has increased significantly, and various fast-charging stations and slow-charging stations have been gradually distributed in cities and rural areas. However, with the increasing regionalization and personalization of charging demands, the problem of unbalanced resource allocation among different vehicle charging stations has gradually emerged. In order to improve charging efficiency and user experience, and at the same time optimize the utilization rate of grid resources, carrying out intelligent load prediction for vehicle charging stations has become a key means to solve these problems.

[0003] Although the existing vehicle charging station load prediction methods have achieved some results in practical applications, there are still some deficiencies. First of all, the existing vehicle charging station load prediction methods usually focus on single and static indicators, failing to consider the real-time status and dynamic changes of vehicle charging stations, and relying on load monitoring records over a past period of time while ignoring the changes in real-time status data. This results in poor sensitivity of the prediction results to sudden load changes. In addition, the existing vehicle charging station load prediction usually lacks in-depth analysis of the load intensity differences and operating characteristics of each vehicle charging station, making it difficult to accurately evaluate the actual load intensity distribution of each vehicle charging station. More critically, these methods often focus on the current load status and lack the ability of future-oriented regulation and planning, especially the prediction of the next day's charging demand and the advance regulation of vehicle charging station resources. This situation seriously affects the practical application value of the prediction. Therefore, there is an urgent need to establish a set of intelligent, efficient and highly targeted load prediction and regulation methods to improve the utilization efficiency of vehicle charging station resources, optimize user experience, and provide support for the development and transformation of the new energy vehicle industry. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the present invention provides an intelligent load prediction method and system for vehicle charging stations, which solves the problems in the above background art.

[0005] To achieve the above objectives, the present invention is realized through the following technical solutions: An intelligent load prediction method for vehicle charging stations includes the following steps;

[0006] S1. Select several vehicle charging stations within the target range as monitoring objects, and combine with Internet of Things technology to collect the charging status information of each vehicle charging station to obtain relevant charging status data information;

[0007] S2. Based on the relevant charging status data information, construct the load intensity coefficients Xfh of several vehicle charging stations. After comparison, mark the group of high-load vehicle charging stations and issue a load demand prediction instruction;

[0008] S3. After receiving the issued load demand prediction instruction, continuously monitor the charging piles within the group of high-load vehicle charging stations, and combine with radio frequency identification technology to respectively obtain the real-time status data information and the charging pile quantity data information. After dimensionless processing, fit to obtain the load demand index Zyc;

[0009] S4. Preset a demand threshold Y, compare the load demand index Zyc with the preset demand threshold Y to judge whether the vehicle charging stations within the group of high-load vehicle charging stations meet the vehicle charging demand of the day, so as to generate corresponding level control instructions and execute them.

[0010] Preferably, the specific steps of S1 include:

[0011] S11. Select several vehicle charging stations within the target range as monitoring objects, combine with Internet of Things technology to connect several vehicle charging stations to the central data center, and perform data transmission through the OCPP protocol. Among them, the charging piles in the vehicle charging stations are all equipped with an electric energy metering unit, a fault diagnosis unit and an RFID chip for identity recognition;

[0012] S12. Connect to the data source interfaces of each vehicle charging station through the API interface of the central data center, and collect the charging status information of each vehicle charging station to obtain relevant charging status data information. The relevant charging status data information includes the total vehicle charging duration Tzs, the total vehicle charging times Ncs and the total fast charging times Nkc of each vehicle charging station. Among them, the collection duration is one day and night;

[0013] S13. Perform preprocessing on the relevant charging status data information. The preprocessing includes noise removal, missing value filling and data smoothing processing operations. Among them, the missing values are filled by mean filling, median filling and interpolation method filling.

[0014] Preferably, the specific steps of S2 include:

[0015] S21. Based on the relevant charging status data information, construct the load intensity coefficients Xfh of several vehicle charging stations. Taking the load intensity coefficient Xfh of the i-th vehicle charging station iFor example, it is obtained in the following specific manner:

[0016] ;

[0017] In the formula, Tzs i represents the total vehicle charging duration of the i-th vehicle charging station, Ncs i represents the total number of vehicle charging times of the i-th vehicle charging station, Nkc i represents the total number of fast charging times of the vehicles at the i-th vehicle charging station. Among them, 、 and all represent weight values, and A represents the first correction constant.

[0018] Preferably, the specific steps of S2 further include:

[0019] S22. According to the method of obtaining the load intensity coefficient Xfh of the i-th vehicle charging station i , obtain the load intensity coefficients Xfh of several vehicle charging stations respectively, and calculate and obtain the average value of the load intensity coefficient according to the statistical mean algorithm ;

[0020] S23. By comparing the load intensity coefficients Xfh of several vehicle charging stations with the average value respectively, obtain the vehicle charging stations corresponding to the load intensity coefficients Xfh that exceed the average value , and mark them all. According to the marked vehicle charging stations, construct a group of the number of high-load vehicle charging stations. When the number of high-load vehicle charging stations in the group exceeds 50% of the total number of vehicle charging stations, send out a load demand prediction instruction.

[0021] Preferably, the specific steps of S3 include:

[0022] S31. After receiving the sent load demand prediction instruction, continuously monitor the charging piles in the group of the number of high-load vehicle charging stations, and set a monitoring time period. Use the power metering unit and fault diagnosis unit equipped on the charging piles to continuously monitor the load power change state and fault shutdown state of the charging piles in the group of the number of high-load vehicle charging stations, and summarize and obtain the real-time status data information of the group of the number of high-load vehicle charging stations. The real-time status data information includes the total load power Pfh and the total number of faults Ncs at each monitoring time point during the monitoring time period, and through radio frequency identification technology, identify the identity and summarize the number of the charging piles in the group of the number of high-load vehicle charging stations to obtain the charging pile number data information. The charging pile number data information includes the total number of charging piles Ncd in the group of the number of high-load vehicle charging stations. Among them, the monitoring time period is one day and night.

[0023] Preferably, the specific steps of S3 further include:

[0024] S32. Analyze the real-time status data information and the charging pile quantity data information. After dimensionless processing, construct the charging pile load density coefficient Xmd at several monitoring time points. Taking the charging pile load density coefficient Xmd at the t-th monitoring time point as an example, it is obtained in the following specific manner: t For example, it is obtained in the following manner:

[0025] ;

[0026] In the formula, represents the total load power at the t-th monitoring time point within the monitoring time period, represents the total number of charging piles within the group of high-load vehicle charging stations.

[0027] Preferably, the specific steps of S3 further include:

[0028] S33. According to the manner of obtaining the charging pile load density coefficient Xmd at the t-th monitoring time point, obtain the charging pile load density coefficients Xmd at several monitoring time points respectively. Associate the charging pile load density coefficients Xmd at several monitoring time points with the real-time status data information and the charging pile quantity data information. After dimensionless processing, construct the load satisfaction coefficient Xmz at several monitoring time points. Taking the load satisfaction coefficient Xmz at the t-th monitoring time point as an example, it is obtained in the following specific manner: t For example, it is obtained in the following manner: t For example, it is obtained in the following manner:

[0029] ;

[0030] In the formula, the charging pile load density coefficient at the t-th monitoring time point, represents the total number of charging piles within the group of high-load vehicle charging stations, represents the total number of faults at the t-th monitoring time point, represents the charging pile failure rate at the t-th monitoring time point.

[0031] Preferably, the specific steps of S3 further include:

[0032] S34. According to the manner of obtaining the load satisfaction coefficient Xmz at the t-th monitoring time point, obtain the load satisfaction coefficients Xmz at several monitoring time points respectively. According to the load level of the power supply line when the vehicle charging station is designed in the power grid planning and combined with the design capacity of the power supply transformer, obtain the total designed load capacity Cfh of the group of high-load vehicle charging stations; t For example, it is obtained in the following manner:

[0033] S35. By correlating the load satisfaction factor Xmz at several monitoring time points with the total designed load capacity Cfh, and combining with the real-time status data information, after dimensionless processing, the load demand index Zyc is obtained by fitting. The load demand index Zyc is obtained through the following formula:

[0034] ;

[0035] In the formula, Xmz t represents the load satisfaction factor at the t-th monitoring time point, represents the total load power at the t-th monitoring time point within the monitoring time period, t = 1, 2, 3,..., T, and T represents the number of monitoring time points, represents the ratio of the total load power at the t-th monitoring time point to the total designed load capacity, representing the operation pressure level of the vehicle charging stations within the group of high-load vehicle charging stations at the current monitoring time point, and both represent weight values, and B represents the second correction constant.

[0036] Preferably, the specific steps of S4 include:

[0037] S41. Preset a demand threshold Y, compare the load demand index Zyc with the preset demand threshold Y to determine whether the vehicle charging stations within the group of high-load vehicle charging stations meet the vehicle charging demand for the day, and generate corresponding level control instructions. The specific content is as follows:

[0038] If the load demand index Zyc > the demand threshold Y, it means that the vehicle charging stations within the group of high-load vehicle charging stations do not meet the vehicle charging demand for the day. Obtain a first-level control instruction. The control content is to formulate a user guidance strategy for the next day based on the distribution of the load intensity factor Xfh of the vehicle charging stations within the group of high-load vehicle charging stations; use the charging navigation system to recommend the surrounding low-load vehicle charging stations to electric vehicles as the preferred choice, deploy temporary energy storage devices and mobile charging vehicles in the surrounding area of the group of high-load vehicle charging stations as supplementary power sources during peak hours, combine the total number of faults Ncs index, arrange a maintenance team to repair the faulty charging piles of the vehicle charging stations within the group of high-load vehicle charging stations, and launch a peak-hour off-peak charging discount on the next day;

[0039] If the load demand index Zyc ≤ the demand threshold Y, it indicates that the charging stations within the high-load charging station quantity group for vehicles meet the vehicle charging demand for the current day. Obtain a secondary regulation instruction, and the regulation content is as follows: Continue to implement high-frequency dynamic monitoring on the charging stations within the high-load charging station quantity group for vehicles on the next day, guide the distribution of charging vehicles through the charging navigation system, promote the use of low-load charging stations through charging discounts and operation publicity on the next day, arrange an emergency response team to closely monitor, and prepare measures to deal with potential charging pile failure risks and sudden load increases, and use the load demand index Zyc on the next day as the prediction basis for subsequent vehicle charging demands.

[0040] An intelligent load prediction system for vehicle charging stations includes a data acquisition module, a marking module, a load demand prediction module, and a level regulation module;

[0041] The data acquisition module is used to select several vehicle charging stations within the target range as monitoring objects, and combine with the Internet of Things technology to collect the charging status information of each vehicle charging station to obtain relevant charging status data information;

[0042] The marking module is used to construct the load intensity coefficient Xfh of several vehicle charging stations based on the relevant charging status data information. After comparison, mark out the high-load charging station quantity group for vehicles and issue a load demand prediction instruction;

[0043] The load demand prediction module is used to continuously monitor the charging piles within the high-load charging station quantity group for vehicles after receiving the issued load demand prediction instruction, and combine with radio frequency identification technology to respectively obtain real-time status data information and charging pile quantity data information. After dimensionless processing, fit to obtain the load demand index Zyc;

[0044] The level regulation module is used to preset the demand threshold Y, compare the load demand index Zyc with the preset demand threshold Y, and judge whether the charging stations within the high-load charging station quantity group for vehicles meet the vehicle charging demand for the current day, so as to generate and execute corresponding level regulation instructions.

[0045] The present invention provides an intelligent load prediction method and system for vehicle charging stations, having the following beneficial effects:

[0046] (1) By integrating Internet of Things technology, radio frequency identification technology, and real-time data analysis methods, it effectively overcomes many deficiencies existing in the prior art, improves the resource utilization efficiency of vehicle charging stations and optimizes the user experience, and provides support for the development and transformation of the new energy vehicle industry. First, based on real-time dynamic monitoring and multi-source data fusion, the load intensity coefficient Xfh of several vehicle charging stations is constructed. After comparison, the group of high-load vehicle charging stations is marked to reflect the dynamic operation status of the vehicle charging stations. This process effectively solves the limitation of relying on a single static indicator in traditional load forecasting methods and significantly improves the sensitivity and accuracy to load changes. Second, by accurately identifying high-load vehicle charging stations and dynamically fitting the load demand index Zyc based on real-time status data, the effective prediction and pre-judgment capabilities of the future demand of high-load vehicle charging stations are realized, providing a reliable basis for the optimal scheduling of vehicle charging station resources. Compared with traditional methods, it significantly improves the response ability to regional and personalized changes in charging demand. At the same time, on the basis of load forecasting, a hierarchical control mechanism is further introduced. According to the comparison result of the load demand index Zyc and the preset demand threshold Y, different levels of control instructions are generated, and through control strategies, including user guidance, energy storage deployment, mobile charging vehicle allocation, fault repair, and off-peak charging discounts the next day, the resource allocation is effectively optimized. This future-oriented control and planning ability not only improves the resource utilization efficiency of vehicle charging stations but also effectively alleviates the charging pressure during peak hours and ensures the stability of the user experience. In addition, through real-time dynamic monitoring and high-frequency data collection, while improving the accuracy of load forecasting, it also enhances the ability to handle sudden load changes, providing strong technical support for grid load optimization and the sustainable development of the new energy vehicle industry. In summary, a method and system for intelligent load forecasting of vehicle charging stations significantly improve the operation efficiency and service quality of vehicle charging stations through intelligent, efficient, and deeply optimized load forecasting and control methods, and play an important role in promoting the transformation and upgrading of the new energy transportation field.

[0047] (2) By integrating Internet of Things technology with the central data center, the intelligent level of load monitoring and analysis for vehicle charging stations has been improved; by connecting each charging station through the OCPP protocol and equipping with an electric energy metering unit, a fault diagnosis unit, and an RFID chip, all-day collection of key data such as the total vehicle charging duration Tzs, the total number of vehicle charging times Ncs, and the total number of fast vehicle charging times Nkc has been achieved; this multi-dimensional data collection method makes up for the defects of single data collection and insufficient real-time performance in traditional systems, and at the same time, through noise removal, missing value filling, and data smoothing processing, the quality and accuracy of the data are ensured; in the load intensity analysis, by constructing a load intensity coefficient Xfh, the operating load levels of different charging stations are quantified, and high-load stations are screened out through the mean algorithm, and a high-load station quantity group is accurately marked; when the proportion of high-load stations exceeds 50%, a load demand prediction instruction is automatically issued, providing a scientific basis for dynamic resource allocation; this method not only effectively solves the problem of insufficient analysis of the regional distribution of charging demand in traditional load prediction methods, but also significantly improves the accuracy and sensitivity of load prediction, providing strong technical support for the efficient management of charging station resources and the improvement of power grid resource utilization rate.

[0048] (3) Through continuous monitoring and multi-dimensional data analysis of the high-load vehicle charging station quantity group, the accuracy and dynamic response ability of load management have been significantly improved; after receiving the load demand prediction instruction, the total load power Pfh is monitored in real time through the electric energy metering unit, the total number of faults Ncs is obtained by combining the fault diagnosis unit, and the number of charging piles Ncd is summarized using radio frequency identification technology, realizing high-frequency data collection and accurate status perception; further, by constructing a dimensionless charging pile load density coefficient Xmd and a load satisfaction coefficient Xmz through dimensionless processing, the load intensity distribution and satisfaction ability at different monitoring time points are reflected, and combined with the total designed load capacity Cfh of the station, the load demand index Zyc of the high-load vehicle charging station quantity group is fitted to quantify the operating pressure and demand trend of high-load stations; this method effectively solves the problems of insufficient analysis of load dynamic changes and future demand prediction in traditional technologies, and provides a precise basis for resource regulation through the analysis of the load satisfaction coefficient Xmz and the load demand index Zyc; in addition, this method can identify problems such as charging peaks and uneven distribution, and combined with the designed capacity and real-time status data, formulate targeted regulation strategies, significantly improving the resource utilization efficiency and operation reliability of charging stations, providing important technical support for realizing intelligent load prediction of vehicle charging stations.

[0049] (4)Based on the comparison between the load demand index Zyc and the preset demand threshold Y, corresponding level control instructions are generated and executed. Through the execution of the control strategy, the future-oriented control planning ability is realized, especially the prediction of the next-day charging demand and the advance control of the resources of the vehicle charging stations; specific resource scheduling plans are formulated for different load situations, including off-peak charging discounts, user guidance strategies, deployment of temporary energy storage devices and arrangement of maintenance teams, which effectively relieve the peak charging pressure, optimize the utilization efficiency of grid resources, and improve the user experience; overall, the present invention takes intelligent, efficient and precise load prediction and control as the core, and can effectively solve the problems of uneven distribution of vehicle charging station resources and sudden load pressure, providing important support for the development of the new energy vehicle industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is a schematic flow chart of a method for intelligent prediction of the load of a vehicle charging station according to the present invention;

[0051] Figure 2 is a block diagram of a system for intelligent prediction of the load of a vehicle charging station according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0053] Embodiment 1

[0054] Please refer to Figure 1 , the present invention provides a method for intelligent prediction of the load of a vehicle charging station, including the following steps;

[0055] S1. Select a number of vehicle charging stations within the target range as monitoring objects, and collect the charging status information of each vehicle charging station in combination with the Internet of Things technology to obtain relevant charging status data information;

[0056] S2. Based on the relevant charging status data information, construct the load intensity coefficients Xfh of a number of vehicle charging stations. After comparison, mark the group of high-load vehicle charging stations and issue a load demand prediction instruction;

[0057] S3. After receiving the issued load demand prediction instruction, continuously monitor the charging piles in the group of high-load vehicle charging stations, and respectively obtain the real-time status data information and the charging pile quantity data information in combination with the radio frequency identification technology. After dimensionless processing, fit to obtain the load demand index Zyc;

[0058] S4. Preset a demand threshold Y, compare the load demand index Zyc with the preset demand threshold Y, and determine whether the number of vehicle charging stations in the high-load vehicle charging station group meets the vehicle charging demand for the day, so as to generate corresponding level control instructions and execute them.

[0059] In this embodiment, through the intelligent prediction of the load of vehicle charging stations, many deficiencies existing in the traditional technology are effectively overcome, and unique advantages and practical values are demonstrated; by combining the Internet of Things technology and radio frequency identification technology, the real-time dynamic monitoring of the charging status information of vehicle charging stations is realized, breaking through the limitations of relying on a single static index in the existing methods, and greatly improving the accuracy and sensitivity of load prediction; at the same time, by constructing a multi-dimensional analysis process of the load intensity coefficient Xfh, the charging pile load density coefficient Xmd, and the load satisfaction coefficient Xmz, and performing dimensionless processing, the load difference and dynamic change characteristics between vehicle charging stations are considered, effectively solving the problem of insufficient in-depth analysis of load distribution in traditional methods; in addition, with the load demand index Zyc as the core, preset the demand threshold Y and generate targeted level control instructions, realizing the resource optimization control and the ability to predict the next-day demand facing the future, thus effectively coping with the problem of lack of forward-looking planning in traditional methods; this method not only optimizes the resource allocation efficiency of vehicle charging stations, but also effectively alleviates the peak load pressure through off-peak charging, energy storage deployment and dynamic monitoring mechanisms, improving the stability and satisfaction of user experience; overall, it provides an intelligent and refined solution for the efficient operation and maintenance of new energy vehicle charging infrastructure, injecting new impetus into the sustainable development of the new energy vehicle industry.

[0060] Embodiment 2

[0061] Please refer to Figure 1 , specifically: The specific steps of S1 include:

[0062] S11. Select a number of vehicle charging stations within the target range as monitoring objects, connect the number of vehicle charging stations with the central data center in combination with the Internet of Things technology, and perform data transmission through the OCPP protocol. Among them, the charging piles in the vehicle charging stations are all equipped with electric energy metering units, fault diagnosis units and RFID chips for identity recognition;

[0063] It should be noted that the equipped electric energy metering unit, fault diagnosis unit, and RFID chip for identity recognition are respectively used to achieve core status monitoring and precise management of the charging pile; the electric energy metering unit is responsible for real-time monitoring of the electric energy consumption of the charging pile, providing accurate data support for subsequent load analysis and regulation; the fault diagnosis unit is used to detect the operating status of the charging pile, quickly identify and report fault conditions, ensuring the normal operation and timely maintenance of the charging pile; the RFID chip is used to identify the identity information of the charging pile, enhancing the identity verification security and data traceability ability of the charging pile;

[0064] S12. Connect to the data source interfaces of each vehicle charging station through the API interface of the central data center, collect the charging status information of each vehicle charging station to obtain relevant charging status data information, where the relevant charging status data information includes the total vehicle charging duration Tzs, the total number of vehicle charging times Ncs, and the total number of fast charging times Nkc of each vehicle charging station. Among them, the collection duration is one day and night;

[0065] S13. Preprocess the relevant charging status data information. The preprocessing includes operations such as noise removal, missing value filling, and data smoothing. Among them, mean filling, median filling, and interpolation method are used for missing value filling.

[0066] It should be noted that the data of the total vehicle charging duration Tzs, the total number of vehicle charging times Ncs, and the total number of fast charging times Nkc are collected in real time through the data source interfaces of each vehicle charging station connected to the central data center; the electric energy metering unit monitors the start and end times of each charging in real time, automatically calculates the total vehicle charging duration Tzs, and accumulatively counts the total number of vehicle charging times Ncs according to the number of charging events; the fast charging mode is judged by the charging power threshold of the charging pile to generate the total number of fast charging times Nkc of the vehicle. These data not only reflect the real-time operating load of the vehicle charging station, but also can provide key data support for load prediction, load intensity analysis, and subsequent regulation strategies by quantifying the charging demand differences of each vehicle charging station; this data acquisition method effectively overcomes the problems of single data collection dimension and lagging update in the traditional system, laying a solid foundation for improving the intelligence and real-time performance of vehicle charging station management, and at the same time providing an important basis for optimizing resource allocation and improving user experience.

[0067] In this embodiment, by introducing the Internet of Things technology and the OCPP protocol, multiple vehicle charging stations are efficiently connected to the central data center, achieving high-precision transmission of real-time data. Relying on the power metering unit, fault diagnosis unit, and RFID chip on the charging pile, the reliability of data collection is ensured. Through the API interface of the central data center, core charging status information covering multiple dimensions such as the total vehicle charging duration Tzs, the total number of vehicle charging times Ncs, and the total number of fast vehicle charging times Nkc can be obtained, thus significantly improving the dynamic control ability of the operating status of vehicle charging stations. In addition, through the preprocessing of the collected data, including removing noise, filling missing values, and data smoothing, the impact of data missing and noise interference on the prediction results is effectively solved, ensuring data quality and analysis accuracy. Compared with the deficiency of static single-index analysis in traditional load forecasting methods, the fusion analysis of multi-dimensional real-time data is realized, laying a solid data foundation for the subsequent construction of load intensity coefficients and demand forecasting. This data collection and preprocessing method not only improves the accuracy of vehicle charging station status monitoring but also significantly improves the dynamic response ability to load changes, providing innovative technical support for solving problems such as uneven resource allocation and difficulty in coping with sudden load changes, and meeting the actual needs of the new energy vehicle industry for an efficient and intelligent load forecasting system.

[0068] Embodiment 3

[0069] Please refer to Figure 1 , specifically: The specific steps of S2 include:

[0070] S21. Construct the load intensity coefficients Xfh of several vehicle charging stations based on relevant charging status data information. Taking the load intensity coefficient Xfh of the i-th vehicle charging station as an example, it is obtained in the following specific way: i For example, it is obtained in the following specific way:

[0071] ;

[0072] In the formula, Tzs i represents the total vehicle charging duration of the i-th vehicle charging station, Ncs i represents the total number of vehicle charging times of the i-th vehicle charging station, Nkc i represents the total number of fast vehicle charging times of the i-th vehicle charging station, where , and all represent weight values, and A represents the first correction constant.

[0073] Specifically, the specific steps of S2 also include:

[0074] S22. Based on the obtained load intensity coefficient Xfh of the i-th vehicle charging station iin a manner, obtain the load intensity coefficients Xfh of several vehicle charging stations respectively, and calculate the mean value of the load intensity coefficients according to the statistical mean algorithm ;

[0075] S23. By comparing the load intensity coefficients Xfh of several vehicle charging stations with the mean value respectively to obtain the vehicle charging stations corresponding to the load intensity coefficients Xfh that exceed the mean value , all are marked. Based on the marked vehicle charging stations, a high-load vehicle charging station quantity group is constructed. When the high-load vehicle charging station quantity group exceeds 50% of the total number of vehicle charging stations, a load demand prediction instruction is sent outwards.

[0076] In this embodiment, by constructing the load intensity coefficients Xfh of several vehicle charging stations, the load conditions of each vehicle charging station are accurately quantified from multiple dimensions, ensuring the flexibility and adaptability of the calculation model; compared with the static analysis of a single index in the traditional method, it can describe the operating characteristics of vehicle charging stations and effectively solve the problem of insufficient evaluation of load intensity differences and actual distribution in the prior art; in addition, through the statistical mean algorithm, the load intensity coefficients Xfh of several vehicle charging stations are overall evaluated, and the high-load vehicle charging stations that exceed the mean value are marked, and further a high-load vehicle charging station quantity group is constructed, realizing the accurate screening and identification of high-load vehicle charging stations; this method not only improves the scientificity and data processing efficiency of load prediction, but also automatically generates a load demand prediction instruction by setting a trigger condition of more than 50% of the high-load vehicle charging station ratio, providing a basis for subsequent demand prediction and regulation; this innovative step effectively makes up for the problem of insufficient response to the dynamic changes of high-load vehicle charging stations in the existing load prediction methods, significantly improving the identification and early warning ability of uneven distribution of vehicle charging station resources and providing reliable support for the optimal scheduling of vehicle charging stations.

[0077] Embodiment 4

[0078] Please refer to Figure 1 , specifically: The specific steps of S3 include:

[0079] S31. After receiving the issued load demand forecasting instruction, continuously monitor the charging piles within the high-load vehicle charging station quantity group, and set a monitoring time period. Using the power metering unit and fault diagnosis unit equipped on the charging piles, continuously monitor the load power change status and fault shutdown status of the charging piles within the high-load vehicle charging station quantity group, and summarize to obtain the real-time status data information of the high-load vehicle charging station quantity group. The real-time status data information includes the total load power Pfh and the total number of faults Ncs at each monitoring time point within the monitoring time period, and through radio frequency identification technology, identify the identities and summarize the quantities of the charging piles within the high-load vehicle charging station quantity group to obtain the charging pile quantity data information. The charging pile quantity data information includes the total number of charging piles Ncd within the high-load vehicle charging station quantity group. Among them, the monitoring time period is one day and night.

[0080] It should be noted that the data of the total load power Pfh, the total number of faults Ncs at each monitoring time point within the monitoring time period, and the total number of charging piles Ncd within the high-load vehicle charging station quantity group are obtained through continuous monitoring of the high-load vehicle charging station, and are specifically completed jointly by the power metering unit, the fault diagnosis unit, and radio frequency identification technology within the charging pile; the power metering unit records the power output of each charging pile in real time and summarizes the total load power Pfh at each monitoring time point; the fault diagnosis unit identifies fault events by detecting the operating status of the equipment and accumulatively counts the total number of faults Ncs at each time point; radio frequency identification technology conducts identity verification and data collection through the RFID chip within the charging pile to ensure the accurate summarization of the number of charging piles Ncd; the role of these data is to accurately reflect the operating conditions of the high-load vehicle charging station; the total load power Pfh describes the real-time load intensity of the vehicle charging station, the total number of faults Ncs provides a key indicator of equipment reliability and operating risks, and the total number of charging piles Ncd within the high-load vehicle charging station quantity group provides a basis for subsequent load density analysis and resource scheduling; this multi-dimensional data acquisition method significantly improves the perception ability of the status of the high-load vehicle charging station, provides scientific support for formulating accurate load forecasting and regulation strategies, and effectively responds to the challenges of charging demand fluctuations and sudden equipment failures.

[0081] In this embodiment, by continuously monitoring the charging piles within the high-load vehicle charging station group, the dynamic control ability of the operation status of the vehicle charging station is significantly enhanced; by using the power metering unit and fault diagnosis unit equipped on the charging piles, the load power change and fault shutdown status of the charging piles can be captured in real time, and combined with radio frequency identification technology, the identity of the charging piles can be identified and the quantity summarized, accurately obtaining the total number Ncd of the charging piles and the real-time status data information; this process not only improves the perception ability of the real-time operation status of the vehicle charging station, but also provides key data support for subsequent load prediction and regulation strategies through the total load power Pfh and the total number of faults Ncs at the monitoring time points; especially by setting the monitoring time period of one day and night, high-frequency data collection and analysis can be realized, quickly responding to load changes and sudden fault problems, and compared with the static prediction method relying on historical data in the traditional method, its dynamic and real-time performance is significantly enhanced; effectively solving the problem of the lack of real-time monitoring of the fault status and load dynamic changes of the charging piles in the existing technology, and at the same time optimizing the timeliness of data collection, providing a solid foundation for the accurate fitting of subsequent load demands and the formulation of regulation plans, thereby effectively alleviating the operation pressure of the high-load vehicle charging station and improving the overall resource allocation efficiency and operation reliability of the vehicle charging station.

[0082] Embodiment 5

[0083] Please refer to Figure 1 , specifically: The specific steps of S3 further include:

[0084] S32. Analyze the real-time status data information and the charging pile quantity data information. After dimensionless processing, construct the charging pile load density coefficient Xmd at several monitoring time points. Taking the charging pile load density coefficient Xmd at the t-th monitoring time point as an example, it is obtained in the following specific manner: t For example, it is obtained in the following specific manner:

[0085] ;

[0086] In the formula, represents the total load power at the t-th monitoring time point within the monitoring time period, represents the total number of charging piles within the high-load vehicle charging station group.

[0087] It should be noted that the charging pile load density coefficient Xmd is an important indicator used to quantify the charging pile load distribution in a vehicle charging station at a specific monitoring time point. Its specific definition is the ratio of the total load power Pfh to the total number of charging piles Ncd at each monitoring time point; it reflects the average load intensity borne by each charging pile per unit time and is a key parameter for evaluating the operation efficiency and load distribution balance of a vehicle charging station; by constructing the charging pile load density coefficient Xmd, it is possible to effectively identify high-load and low-load time periods and the load distribution of vehicle charging stations, providing data support for optimizing resource allocation and improving operation efficiency; in addition, the dynamic change trend of the charging pile load density coefficient Xmd can reveal the load fluctuation law, help predict future load changes, and thus provide a scientific basis for formulating precise control strategies, improving the user charging experience, and enhancing the overall management level of vehicle charging stations; this indicator has good universality after dimensionless processing, facilitating comparison and analysis at different time points.

[0088] Specifically, the specific steps of S3 also include:

[0089] S33. According to the method of obtaining the charging pile load density coefficient Xmd at the t-th monitoring time point t Obtain the charging pile load density coefficients Xmd at several monitoring time points respectively, associate the charging pile load density coefficients Xmd at several monitoring time points with the real-time status data information and the charging pile quantity data information, and after dimensionless processing, construct the load satisfaction coefficients Xmz at several monitoring time points. Taking the load satisfaction coefficient Xmz at the t-th monitoring time point t as an example, it is obtained specifically in the following way:

[0090] ;

[0091] In the formula, The charging pile load density coefficient at the t-th monitoring time point, Represents the total number of charging piles in the high-load vehicle charging station quantity group, Represents the total number of faults at the t-th monitoring time point, Represents the charging pile failure rate at the t-th monitoring time point.

[0092] Specifically, the specific steps of S3 also include:

[0093] S34. According to the method of obtaining the load satisfaction coefficient Xmz at the t-th monitoring time point tIn this way, the load satisfaction coefficient Xmz at several monitoring time points is obtained respectively. According to the load level of the power supply line during the grid planning and design of the vehicle charging station, and combined with the design capacity of the power supply transformer, the total design load capacity Cfh of the high-load vehicle charging station quantity group is obtained;

[0094] It should be noted that the total design load capacity Cfh of the high-load vehicle charging station quantity group is calculated by comprehensively analyzing the load level of the power supply line and the design capacity of the power supply transformer during the grid planning and design. Specifically, the load level of the power supply line provides the theoretical upper limit of the station power supply system under full-load operation, while the design capacity of the power supply transformer defines the actual upper limit of the station's power supply capacity. By accumulating the design capacities of each charging station in the high-load station quantity group, the total design load capacity Cfh is obtained. Its role is to provide a benchmark for the analysis of load demand. By comparing the actual load satisfaction with the total design load capacity Cfh, the operation pressure and resource utilization efficiency of the high-load vehicle charging station quantity group can be effectively evaluated. This index can not only identify potential resource bottlenecks, but also provide key data support for formulating load regulation strategies, optimizing grid resource allocation, and ensuring the operation stability of the station.

[0095] S35. By associating the load satisfaction coefficient Xmz at several monitoring time points with the total design load capacity Cfh, and combining with the real-time status data information, after dimensionless processing, the load demand index Zyc is obtained by fitting. The load demand index Zyc is obtained through the following formula:

[0096] ;

[0097] In the formula, Xmz t represents the load satisfaction coefficient at the t-th monitoring time point, represents the total load power at the t-th monitoring time point within the monitoring time period, t = 1, 2, 3,..., T, and T represents the number of monitoring time points, represents the ratio of the total load power at the t-th monitoring time point to the total design load capacity, representing the operation pressure level of the vehicle charging stations within the high-load vehicle charging station quantity group at the current monitoring time point, and both represent weight values, and B represents the second correction constant.

[0098] It should be noted that the load demand index Zyc is a comprehensive index used to quantify the matching degree between the charging demand and the resource carrying capacity within the group of high-load vehicle charging stations; by combining the load satisfaction coefficient Xmz and the total designed load capacity Cfh at several monitoring time points, as well as the real-time status data information, the constructed load demand index Zyc reflects the operating pressure level and load demand trend of high-load vehicle charging stations during the monitoring period; specifically, the calculation of the load demand index Zyc takes into account the ratio of the actual load power to the designed capacity of the vehicle charging station at each time point, and by introducing the weight value 、 and the second correction constant B, it is ensured that the load demand index Zyc can flexibly adapt to the operating conditions of different vehicle charging stations. The role of the load demand index Zyc is to provide a scientific basis for load forecasting and resource regulation, as well as the ability to dynamically analyze load changes.

[0099] In this embodiment, by constructing the charging pile load density coefficient Xmd and the load satisfaction coefficient Xmz, the load distribution and operation characteristics of high-load vehicle charging stations are dynamically analyzed from multiple monitoring time points, effectively solving the limitation of static analysis of load status in traditional technologies. First, based on the real-time status data information and the charging pile quantity data information, using the dimensionless processing technology, the charging pile load density coefficient Xmd per monitoring time point is constructed, accurately quantifying the load intensity distribution of each charging pile at each moment. This method can clearly reflect the dynamic changes of the load in the time dimension, making up for the problem that the description of the load distribution characteristics in traditional load forecasting methods is not comprehensive enough. In addition, by associating the load density coefficient Xmd with the real-time status data information and the charging pile quantity data information, the load satisfaction coefficient Xmz is further constructed, which can more comprehensively evaluate the load bearing capacity and operation stability of vehicle charging stations at different time points. Further combining the load level of the power supply line and the design capacity of the power supply transformer when the vehicle charging station is planned and designed in the power grid, the total design load capacity Cfh of the high-load vehicle charging station is obtained, and based on this, the load demand index Zyc within the high-load vehicle charging station quantity group is dynamically fitted. This process realizes the accurate quantification of the operation pressure level of vehicle charging stations, effectively solving the problem of the lack of in-depth analysis of load dynamic changes and future demand trends in the existing technology. Especially through the comprehensive evaluation of several monitoring time points, the key periods of load fluctuations and peak pressures can be captured, providing a scientific basis for formulating more efficient and accurate resource regulation strategies. This method not only improves the accuracy of load forecasting, but also provides technical support for the optimal allocation of power grid resources and the forward-looking regulation of high-load vehicle charging stations, thus significantly improving the user experience and enhancing the intelligent management level of new energy vehicle charging infrastructure. Generally speaking, this process has unique advantages in dynamic load analysis and future demand forecasting, providing an innovative solution for effectively solving the uneven distribution of resources and sudden load pressures.

[0100] Embodiment 6

[0101] Please refer to Figure 1 , specifically: The specific steps of S4 include:

[0102] S41. Preset a demand threshold Y, compare the load demand index Zyc with the preset demand threshold Y to determine whether the vehicle charging stations in the high-load vehicle charging station quantity group meet the vehicle charging demand for the current day, and generate corresponding level regulation instructions, the specific content is as follows:

[0103] If the load demand index Zyc > the demand threshold Y, it indicates that the vehicle charging stations within the high-load vehicle charging station quantity group do not meet the vehicle charging demand for the current day. Obtain a first-level regulation instruction, and the regulation content is to formulate a user guidance strategy for the next day based on the distribution of the load intensity coefficient Xfh of the vehicle charging stations within the high-load vehicle charging station quantity group; use the charging navigation system to recommend the surrounding low-load vehicle charging stations to electric vehicles as the preferred choice, deploy temporary energy storage devices and mobile charging vehicles in the surrounding area of the high-load vehicle charging station quantity group as supplementary power sources during peak hours, combine with the total number of faults Ncs index, arrange a maintenance team to repair the faulty charging piles of the vehicle charging stations within the high-load vehicle charging station quantity group, and launch peak-hour off-peak charging discounts on the next day;

[0104] If the load demand index Zyc ≤ the demand threshold Y, it indicates that the vehicle charging stations within the high-load vehicle charging station quantity group meet the vehicle charging demand for the current day. Obtain a second-level regulation instruction, and the regulation content is: continue to implement high-frequency dynamic monitoring of the vehicle charging stations within the high-load vehicle charging station quantity group on the next day, guide the distribution of charging vehicles through the charging navigation system, promote the use of low-load vehicle charging stations through charging discounts and operation publicity on the next day, arrange an emergency response team to pay close attention, and prepare measures to deal with potential charging pile faults and sudden load increases, and use the load demand index Zyc on the next day as the prediction basis for subsequent vehicle charging demands.

[0105] In this embodiment, by presetting the demand threshold Y, the load demand index Zyc is compared with the preset demand threshold Y to generate and execute the first-level control instruction and the second-level control instruction, significantly improving the intelligent and precise level of the load management of the vehicle charging station; when the vehicle charging stations in the high-load vehicle charging station quantity group do not meet the vehicle charging demand of the day, a first-level control instruction is generated, and multi-level control measures are taken. Through the charging navigation system, users are guided to preferentially select low-load vehicle charging stations, temporary energy storage devices and mobile charging vehicles are deployed to relieve the power consumption pressure during peak hours, and the maintenance team is arranged to repair the faulty charging piles in combination with the total number of faults Ncs index. At the same time, a peak-shifting charging discount strategy for the next day is launched; these measures not only effectively relieve the operating pressure of high-load vehicle charging stations, but also optimize the resource scheduling efficiency and improve the user experience; when the vehicle charging stations in the high-load vehicle charging station quantity group meet the vehicle charging demand of the day, a second-level control instruction is generated, high-frequency dynamic monitoring is implemented, and the use of low-load vehicle charging stations is promoted through charging discounts and operation publicity. At the same time, an emergency response team is arranged to closely monitor potential risks; this control method can ensure the reasonable allocation of resources and effectively prevent the adverse effects of sudden load growth on the power grid and user experience. Compared with the lack of forward-looking control planning and the inability to predict future demands in traditional methods, this control mechanism effectively solves the problems of uneven resource allocation and failure to respond to sudden demands in a timely manner through a targeted hierarchical strategy; moreover, the control measures construct an intelligent management system for efficient resource utilization and user-friendly experience by dynamically adjusting the vehicle charging station resources and optimizing user behavior, providing strong technical support for the efficient operation and sustainable development of electric vehicle charging infrastructure.

[0106] Embodiment 7

[0107] Please refer to Figure 1 and Figure 2 , specifically: An intelligent load prediction system for vehicle charging stations includes a data acquisition module, a marking module, a load demand prediction module, and a hierarchical control module;

[0108] The data acquisition module is used to select several vehicle charging stations within the target range as monitoring objects, and combined with the Internet of Things technology, collect the charging status information of each vehicle charging station to obtain relevant charging status data information;

[0109] The marking module is used to construct the load intensity coefficient Xfh of several vehicle charging stations based on the relevant charging status data information. After comparison, the high-load vehicle charging station quantity group is marked, and a load demand prediction instruction is issued;

[0110] The load demand prediction module is used to continuously monitor the charging piles in the high-load vehicle charging station quantity group after receiving the issued load demand prediction instruction, and combine with radio frequency identification technology to respectively obtain the real-time status data information and the charging pile quantity data information. After dimensionless processing, the load demand index Zyc is obtained by fitting.

[0111] The level regulation module is used to preset the demand threshold Y, compare the load demand index Zyc with the preset demand threshold Y, and judge whether the vehicle charging stations in the high-load vehicle charging station quantity group meet the vehicle charging demand of the day, so as to generate and execute the corresponding level regulation instruction.

[0112] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principle and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent prediction method for the load of an automobile charging station, characterized in that: It includes the following steps; S1. Select several vehicle charging stations within the target range as monitoring objects, and combine Internet of Things technology to collect the charging status information of each vehicle charging station to obtain relevant charging status data information; S2. Based on the relevant charging status data information, construct the load intensity coefficient Xfh of several vehicle charging stations. After comparison, mark the group of high-load vehicle charging stations and issue a load demand prediction instruction; Based on the relevant charging status data information, construct the load intensity coefficient Xfh of several vehicle charging stations, and the specific construction process is as follows; Based on the total vehicle charging duration, total vehicle charging times, and total fast charging times of each charging station in the relevant charging status data information, construct the load intensity coefficient Xfh of the corresponding vehicle charging station by weighted superposition and deviation correction; S3. After receiving the issued load demand prediction instruction, continuously monitor the charging piles within the group of high-load vehicle charging stations, and combine radio frequency identification technology to respectively obtain real-time status data information and charging pile quantity data information. After dimensionless processing, fit to obtain the load demand index Zyc, and the specific acquisition process is as follows; Extract the characteristics of the real-time status data information and the charging pile quantity data information, calculate the ratio of the total load power of each monitored time point after extraction to the total number of charging piles in the group of high-load vehicle charging stations to obtain the charging pile load density coefficient Xmd of several monitored time points, and correlate the charging pile load density coefficient Xmd of several monitored time points with the total number of fault times of the corresponding monitored time points and the total number of charging piles in the group of high-load vehicle charging stations to obtain the load satisfaction coefficient Xmz of several monitored time points; S34. Obtain the load satisfaction coefficient Xmz at the t-th monitoring time point, and respectively obtain the load satisfaction coefficients Xmz at a number of monitoring time points. According to the load level of the power supply line during the grid planning and design of the vehicle charging station, and in combination with the designed capacity of the power supply transformer, obtain the total designed load capacity Cfh of the high-load vehicle charging station quantity group; t ​ S35. By correlating the load satisfaction coefficient Xmz of several monitored time points with the total designed load capacity Cfh and combining the real-time status data information, after dimensionless processing, fit to obtain the load demand index Zyc, and the load demand index Zyc is obtained through the following formula: ； Wherein, Xmz t represents the load satisfaction coefficient at the t-th monitoring time point, represents the total load power at the t-th monitoring time point within the monitoring time period, t = 1, 2, 3,..., T, and T represents the number of monitoring time points, represents the ratio of the total load power at the t-th monitoring time point to the total designed load capacity, and represents the operating pressure level of the electric vehicle charging stations within the group of high-load electric vehicle charging stations at the current monitoring time point, and both represent weight values, and B represents the second correction constant; S4. Preset a demand threshold Y, compare the load demand index Zyc with the preset demand threshold Y, and judge whether the vehicle charging stations within the group of high-load vehicle charging stations meet the vehicle charging demand for the current day, so as to generate a corresponding level control instruction and execute it.

2. The intelligent prediction method for the load of an automotive charging station according to claim 1, characterized in that: The specific steps of S1 include: S11. Select several vehicle charging stations within the target range as monitoring objects, combine Internet of Things technology to connect several vehicle charging stations to the central data center, and perform data transmission through the OCPP protocol. Among them, the charging piles in the vehicle charging stations are all equipped with an electric energy metering unit, a fault diagnosis unit, and an RFID chip for identity recognition; S12. Connect to the data source interfaces of each vehicle charging station through the API interface of the central data center, collect the charging status information of each vehicle charging station to obtain relevant charging status data information. The relevant charging status data information includes the total vehicle charging duration Tzs, the total number of vehicle charging times Ncs, and the total number of fast charging times Nkc of each vehicle charging station. Among them, the collection duration is one day and night. S13. Preprocess the relevant charging status data information. The preprocessing includes noise removal, missing value filling, and data smoothing operations. Among them, the missing value filling adopts mean filling, median filling, and interpolation method filling.

3. The intelligent prediction method for the load of an automotive charging station according to claim 2, wherein: The specific steps of S2 include: S21. Construct a load intensity coefficient Xfh for several vehicle charging stations based on relevant charging status data information. Taking the load intensity coefficient Xfh of the i-th vehicle charging station as an example, it is obtained in the following specific manner: i For example, it is obtained in the following specific manner: ； Where, Tzs i represents the total vehicle charging duration of the i-th vehicle charging station, Ncs i represents the total number of vehicle charging times of the i-th vehicle charging station, Nkc i represents the total number of fast charging times of the vehicles at the i-th vehicle charging station, where , and all represent weight values, and A represents the first correction constant.

4. The intelligent prediction method for the load of an automotive charging station according to claim 3, characterized in that: The specific steps of S2 also include: S22. Obtain the load intensity coefficient Xfh of the i-th vehicle charging station, i and obtain the load intensity coefficients Xfh of several vehicle charging stations respectively. Then, calculate the mean value of the load intensity coefficient according to the statistical mean algorithm. ; S23. By comparing the load intensity coefficients Xfh of several vehicle charging stations with the mean value respectively to obtain the vehicle charging stations corresponding to the load intensity coefficients Xfh that exceed the mean value , all of them are marked. Based on the marked vehicle charging stations, a high-load vehicle charging station quantity group is constructed. When the high-load vehicle charging station quantity group exceeds 50% of the total number of vehicle charging stations, a load demand prediction instruction is sent outwards.

5. The intelligent prediction method for the load of an automotive charging station according to claim 4, wherein: The specific steps of S3 include: S31. After receiving the issued load demand prediction instruction, continuously monitor the charging piles in the high-load vehicle charging station quantity group, set the monitoring time period, and use the power metering unit and fault diagnosis unit equipped on the charging piles to continuously monitor the load power change status and fault shutdown status of the charging piles in the high-load vehicle charging station quantity group, and summarize to obtain the real-time status data information of the high-load vehicle charging station quantity group. The real-time status data information includes the total load power Pfh and the total number of faults Ncs at each monitoring time point during the monitoring time period. And through radio frequency identification technology, identify the identities and summarize the quantities of the charging piles in the high-load vehicle charging station quantity group to obtain the charging pile quantity data information. The charging pile quantity data information includes the total number of charging piles Ncd in the high-load vehicle charging station quantity group. Among them, the monitoring time period is one day and night.

6. The intelligent prediction method for the load of an automotive charging station according to claim 5, wherein: The specific steps of S3 also include: S32. Analyze the real-time status data information and the charging pile quantity data information. After dimensionless processing, construct the charging pile load density coefficient Xmd at several monitoring time points. Taking the charging pile load density coefficient Xmd at the t-th monitoring time point as an example, it is obtained in the following specific manner: t For example, it is obtained in the following specific manner: ； In the formula, represents the total load power at the t-th monitoring time point within the monitoring time period, represents the total number of charging piles within the group of high-load vehicle charging stations.

7. The intelligent prediction method for the load of an automotive charging station according to claim 6, characterized in that: The specific steps of S3 also include: S33. According to the method of obtaining the charging pile load density coefficient Xmd at the t-th monitoring time point t , the charging pile load density coefficients Xmd at a number of monitoring time points are obtained respectively. The charging pile load density coefficients Xmd at a number of monitoring time points are associated with the real-time status data information and the charging pile quantity data information. After dimensionless processing, the load satisfaction coefficients Xmz at a number of monitoring time points are constructed. Taking the load satisfaction coefficient Xmz at the t-th monitoring time point t as an example, it is obtained specifically according to the following method: ； In the formula, represents the charging pile load density coefficient at the t-th monitoring time point, represents the total number of charging piles within the group of high-load vehicle charging stations, represents the total number of faults at the t-th monitoring time point, represents the failure rate of charging piles at the t-th monitoring time point.

8. The intelligent prediction method for the load of an automotive charging station according to claim 1, wherein: The specific steps of S4 include: S41. Preset the demand threshold Y, compare the load demand index Zyc with the preset demand threshold Y to judge whether the vehicle charging stations in the high-load vehicle charging station quantity group meet the vehicle charging demand of the day, and generate corresponding level regulation instructions. The specific content is as follows: If the load demand index Zyc > the demand threshold Y, it means that the vehicle charging stations in the high-load vehicle charging station quantity group do not meet the vehicle charging demand of the day. Obtain the first-level regulation instruction. The regulation content is to formulate the user guidance strategy for the next day according to the distribution of the load intensity coefficient Xfh of the vehicle charging stations in the high-load vehicle charging station quantity group; use the charging navigation system to recommend the surrounding low-load vehicle charging stations to electric vehicles as the first choice, deploy temporary energy storage devices and mobile charging vehicles in the surrounding area of the high-load vehicle charging station quantity group as supplementary power sources during peak hours, combine with the total number of faults Ncs index, arrange the maintenance team to repair the faulty charging piles of the vehicle charging stations in the high-load vehicle charging station quantity group, and launch off-peak charging discounts during peak hours on the next day. If the load demand index Zyc ≤ the demand threshold Y, it indicates that the vehicle charging stations within the high-load vehicle charging station quantity group meet the vehicle charging demand for the current day. Obtain a secondary regulation instruction, and the regulation content is as follows: Continue to implement high-frequency dynamic monitoring on the vehicle charging stations within the high-load vehicle charging station quantity group for the next day, guide the distribution of charging vehicles through the charging navigation system, promote the use of low-load vehicle charging stations the next day, arrange the emergency response team to pay close attention, and prepare measures to deal with potential charging pile failure risks and sudden load increases. Use the load demand index Zyc of the next day as the prediction basis for subsequent vehicle charging demands.

9. An intelligent load prediction system for a vehicle charging station yard, which is used to implement the intelligent load prediction method for a vehicle charging station yard described in any one of the above claims 1 to 8, and is characterized in that: It includes a data acquisition module, a marking module, a load demand prediction module, and a level regulation module; The data acquisition module is used to select a number of vehicle charging stations within the target range as monitoring objects, and combine with the Internet of Things technology to collect the charging status information of each vehicle charging station to obtain relevant charging status data information; The marking module is used to construct the load intensity coefficients Xfh of a number of vehicle charging stations based on the relevant charging status data information. After comparison, mark out the high-load vehicle charging station quantity group and issue a load demand prediction instruction; The load demand prediction module is used to continuously monitor the charging piles within the high-load vehicle charging station quantity group after receiving the issued load demand prediction instruction, and combine with radio frequency identification technology to respectively obtain the real-time status data information and the charging pile quantity data information. After dimensionless processing, fit to obtain the load demand index Zyc; The level regulation module is used to preset the demand threshold Y, compare the load demand index Zyc with the preset demand threshold Y, and judge whether the vehicle charging stations within the high-load vehicle charging station quantity group meet the vehicle charging demand for the current day, so as to generate and execute corresponding level regulation instructions.

Images