Charging pile intelligent control management platform

Abstract

The invention discloses a charging pile intelligent control management platform, relates to the technical field of charging pile intelligent control, and can identify the peak and valley periods of the power grid load in advance by collecting and analyzing the power grid load data and the operation state of a charging pile in real time and combining a regression model of a load prediction and demand analysis module. The power output of each charging pile is dynamically adjusted through the charging pile scheduling and power adjusting module, the power grid resource distribution is optimized, the overload phenomenon is avoided, and the stability of power grid operation is improved. By monitoring and feeding back the charging behavior data of the user, the platform can reasonably plan the working state of the charging pile. Therefore, too fast loss of some charging piles due to continuous overload is avoided, resource waste of idle charging piles is reduced, the utilization efficiency of equipment is improved on the whole, and the service life of the charging piles is prolonged. By using the load prediction and demand analysis module, the platform can pre-judge the charging demand peak of the user and optimize the scheduling strategy in advance, thereby shortening the waiting time of the user.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control of charging piles, and in particular to an intelligent control management platform for charging piles. Background Art

[0002] With the rapid development of smart grid and new energy vehicle industries, the power system and energy management fields are gradually transforming towards intelligence, and many emerging intelligent control platforms have emerged. In this field, intelligent management of charging piles, as a key component of electric vehicle infrastructure construction, has received more and more attention. In particular, the intelligent control and management platform of charging piles not only needs to achieve efficient charging management, but also needs to optimize power resource allocation and grid load balancing. This intelligent platform combines data analysis, prediction and control technologies to provide users with convenient charging services and ensure the stable operation of the power grid by optimizing the power output and working status of charging piles.

[0003] At present, the management mode of most charging piles is relatively static, mainly relying on fixed power output and simple load distribution strategies, lacking the ability to dynamically predict and intelligently adjust the grid load and user demand. This mode easily leads to overload of charging piles during peak hours, which not only increases the risk of equipment failure, but also may increase user waiting time and reduce the charging experience. At the same time, grid load fluctuations are difficult to predict and adjust, and the utilization efficiency of power resources is low.

[0004] The shortcomings of traditional charging pile management methods are mainly due to their low dependence on real-time data, which makes it difficult to effectively respond to instantaneous changes in grid load and users' diverse charging needs. The lack of accurate load forecasting and intelligent adjustment mechanisms leads to excessive load pressure during peak hours, which easily leads to grid imbalance, local overload and even power outage risks. The above problems not only reduce the service life of charging piles, but also increase power loss, making it difficult to achieve energy saving goals. Therefore, by introducing load forecasting and power adjustment mechanisms based on data analysis and machine learning, intelligent distribution can be better achieved, and the stability of the power system and the utilization efficiency of resources can be improved. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a charging pile intelligent control management platform, which solves the problems mentioned in the background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a charging pile intelligent control management platform, including a data acquisition module, a data preprocessing module, a load forecasting and demand analysis module, a charging pile scheduling and power regulation module, a real-time monitoring and feedback module and a report generation and decision support module;

[0007] The data acquisition module is responsible for real-time collection of charging pile operation data, equipment load data and grid power demand data to form a data set W;

[0008] The data preprocessing module cleans and standardizes the data set W to obtain a data standard set WB;

[0009] The load forecasting and demand analysis module analyzes the data standard set WB by using a regression model to predict future grid load changes and charging pile requirements;

[0010] The charging pile scheduling and power regulation module dynamically adjusts the power output of each charging pile according to the predicted grid load changes and charging pile requirements, and balances the grid load by adjusting the working state of the charging piles;

[0011] The real-time monitoring and feedback module monitors the working status, power output and power consumption of each charging pile, and collects and feeds back the operating data;

[0012] The report generation and decision support module generates a decision support report on grid load, charging pile status and power adjustment based on the feedback operation data.

[0013] Preferably, the data acquisition module includes an operation data acquisition unit and a load and power grid data acquisition unit;

[0014] The operation data acquisition unit acquires the operation data of the charging pile, wherein the operation data of the charging pile includes the charging pile output power PC, the charging pile temperature TC and the charging pile cumulative power Eto, the charging pile output power PC is acquired by a power meter, the charging pile temperature TC is acquired by a temperature sensor, and the charging pile cumulative power Eto is acquired by a smart meter;

[0015] The power grid data acquisition unit acquires equipment load data and power grid power demand data, wherein the equipment load data includes equipment load fluctuation ΔPl, and the equipment load fluctuation is acquired through a current transformer;

[0016] The power demand data of the power grid is collected through power grid monitoring equipment and smart meters, including the current power grid load Lc, power grid voltage Vg and power grid load fluctuation rate ΔL. The operation data of charging piles, user charging behavior data, equipment load data and power grid power demand data are integrated to form a data set W.

[0017] Preferably, the data preprocessing module includes a data cleaning unit and a data standardization unit;

[0018] The data cleaning unit cleans the abnormal values ​​of the data set W and fills the missing values ​​by using statistical methods and interpolation algorithms to obtain a data cleaning set WC;

[0019] The data standardization unit converts the data cleaning set WC into a unified standard through the Z-score standard method, eliminates the dimensional differences between the data items, and obtains the data standard set WB.

[0020] Preferably, the load forecasting and demand analysis module includes a data feature extraction unit and a regression prediction unit;

[0021] The data feature extraction unit is responsible for extracting characteristic parameters related to load forecasting from the data standard set WB, including time characteristics Ttm and historical load characteristics LH, to form a feature set F;

[0022] Among them, the time feature Ttm is extracted through sine transform Tsin and cosine transform Tcos;

[0023] The historical load characteristics LH are extracted by using a sliding window based on the data standard set WB grid current load Lc, equipment load fluctuation ΔPl and grid load fluctuation rate ΔL;

[0024] The data feature extraction unit analyzes seasonal features, periodic features and trend features in historical data to provide input data for subsequent prediction models; specifically:

[0025] Seasonal characteristics include seasonal changes. In different seasons, the grid load is different. Summer and winter are peak load periods, while spring and autumn are low load periods.

[0026] The periodic characteristics include daily cycle, weekly cycle and monthly cycle; the daily cycle includes the power grid load fluctuations in the morning, noon, evening and midnight, and the comparison is carried out to obtain the time period of high load time and mark it as the load fluctuation of the day;

[0027] The weekly cycle includes the grid load fluctuations on working days and the grid load fluctuations on weekends, which are obtained by accumulating the characteristic load fluctuations of the daily cycle and marked as weekly load fluctuations;

[0028] The monthly cycle includes the grid load fluctuation at the beginning of the month, the grid load fluctuation in the middle of the month and the grid load fluctuation at the end of the month, which is obtained by accumulating the weekly load fluctuations with weekly cycle characteristics;

[0029] In a one-month cycle, the grid load fluctuation is at its peak at the end of the month, the last week of the month, and the morning and evening of the working day.

[0030] Trend characteristics include long-term load growth trends, which are captured through historical load data.

[0031] Preferably, the regression prediction unit trains a regression model through the feature set F and the data standard set WB, and outputs a load prediction value Ye of future load and charging demand; obtains the load prediction value Ye through the regression model;

[0032] Through the intercept β0 of the regression model and the regression coefficient β of the jth feature of the feature set F j Perform weighted combination to obtain the load forecast value Ye, identify peak and valley periods, and use it to regulate the power output of the charging pile;

[0033] Through the load prediction value Ye, the preset peak threshold TGY and the valley threshold TDY, the power output of the charging pile is dynamically regulated to obtain the charging demand, which is obtained by matching in the following ways:

[0034] When the load forecast value Ye ≥ the peak threshold TGY, it means that the charging pile is in the peak period, and the load of the charging pile in use is adjusted;

[0035] When the load forecast value Ye is less than the valley threshold TDY, it means that the charging pile is in the valley period, and the load adjustment of the charging pile in use is not performed;

[0036] When the peak threshold TGY>load prediction value Ye≥valley threshold TDY, it means that the charging pile is in a stable period and the charging pile load is not adjusted.

[0037] Through the above judgment conditions, the system can identify the peak and trough periods of demand, which is convenient for rationally regulating resource allocation and grid load scheduling. The regression prediction unit outputs the predicted future load changes and demand distribution. The generated results help the grid to make scheduling plans in advance and ensure grid load balance. Charging piles can automatically adjust power output during different demand periods.

[0038] Preferably, the charging pile scheduling and power regulation module includes a power distribution unit and a working state regulation unit;

[0039] The power distribution unit determines the power output of the charging pile in different time periods according to the load prediction value Ye, the preset peak power threshold Phi and the valley power threshold Plo; adjusts the power during the peak period and keeps the power unchanged during the valley period; and achieves the best allocation of resources;

[0040] The power allocation decision is obtained by matching:

[0041] When the load prediction value Ye ≥ the peak power threshold Phi, the power of the charging pile is adjusted;

[0042] When the load prediction value Ye≤the valley power threshold Plo, the power of the charging pile is maintained.

[0043] Preferably, the working state adjustment unit is responsible for managing the start, pause and low power mode switching of the charging pile;

[0044] According to the power distribution decision result output by the power distribution unit, the working state adjustment unit adjusts the number of charging piles in use Nac during peak hours and maintains the working frequency during off-peak hours, thereby ensuring the dynamic balance of the grid load;

[0045] When the load forecast value Ye ≥ the peak power threshold Phi, the number of charging piles in use Nac is equal to the maximum number of charging piles Nmax, and the output power PC of the charging pile is adjusted to 120% of the original power;

[0046] When the load forecast value Ye≤the valley power threshold Plo, the number of charging piles in use Nac is obtained by the difference between the maximum number of charging piles Nmax, the load forecast value Ye and the peak power threshold Phi, and the output power PC of the charging pile is adjusted to 90% of the original power.

[0047] Preferably, the real-time monitoring and feedback module includes a status monitoring unit and an alarm feedback unit;

[0048] The state monitoring unit is responsible for real-time monitoring of the charging pile output power PC and the current load Lc of the power grid, and judges the state of the charging pile by comparing the charging pile output power PC with the preset power threshold Pth:

[0049] The status of the charging pile is obtained by matching in the following way:

[0050] When the charging pile output power PC>power threshold Pth, the charging pile overload alarm signal is triggered;

[0051] When the charging pile output power PC ≤ power threshold Pth, the charging pile overload alarm signal will not be triggered;

[0052] By comparing the current grid load Lc with the preset grid load threshold Lth, the grid load state is determined;

[0053] When the current grid load Lc>grid load threshold Lth, the grid load overload alarm signal is triggered;

[0054] When the current grid load Lc ≤ the grid load threshold Lth, the grid load overload alarm signal is not triggered.

[0055] Preferably, the alarm feedback unit provides feedback on overload and fault conditions according to the alarm signal output by the status monitoring unit, sends a real-time alarm to the operator, and automatically triggers a strategy adjustment signal to adjust the charging pile output power PC and the number of active charging piles Nac;

[0056] The charging pile power overload alarm signal and the power grid overload alarm signal are integrated to obtain an alarm feedback signal AF;

[0057] When the alarm feedback signal AF=1, the alarm information is fed back, triggering the strategy adjustment to reduce the power of the charging pile.

[0058] Preferably, the report generation and decision support module includes a data analysis and report generation unit and a decision support suggestion generation unit;

[0059] The data analysis and report generation unit receives the feedback of grid load, charging pile status, and power adjustment data, aggregates, counts, and performs trend analysis on these data; based on the analysis results, generates report content, including future grid load forecast, charging pile power dispatch strategy, and equipment operation status;

[0060] The decision support suggestion generating unit generates adjustment decision suggestions according to the data results of the analysis report;

[0061] Specifically: during peak load periods, reduce the output power of the charging piles and adjust the charging piles to start during low load periods;

[0062] During off-load periods, users are encouraged to charge to reduce overall grid fluctuations and increase the output power of charging piles.

[0063] If the system detects that the load demand is too high for a long time in certain periods, the unit will recommend managers to expand infrastructure or add charging piles. At the same time, for areas with low charging pile utilization during off-peak periods, the unit will recommend reasonable equipment deployment to better allocate resources. If it is detected that the power load of some charging piles exceeds the preset safety threshold, the unit will generate emergency suggestions, such as immediately reducing the power output of the charging pile, or guiding users to complete charging at other charging piles. At the same time, when the grid load fluctuates beyond the warning range, it will recommend power limiting for all charging piles to reduce the impact on the grid. This strategy ensures that the system can maintain stable operation even under extreme conditions.

[0064] The present invention provides a charging pile intelligent control management platform, which has the following beneficial effects:

[0065] (1) The platform can identify peak and valley periods of grid load in advance by collecting and analyzing grid load data and charging pile operating status in real time, combined with the regression model of the load forecasting and demand analysis module. The charging pile scheduling and power regulation module dynamically adjusts the power output of each charging pile, optimizes grid resource allocation, avoids overload, and improves the stability of grid operation. By monitoring and feeding back user charging behavior data, the platform can reasonably plan the working status of charging piles. It can prevent some charging piles from being worn out too quickly due to continuous overload, and reduce the waste of resources of idle charging piles, thereby improving the overall utilization efficiency of equipment and extending the service life of charging piles.

[0066] Using the load forecasting and demand analysis module, the platform can predict the peak of user charging demand and optimize the dispatching strategy in advance to shorten the user's waiting time. During the peak load period of the power grid, the platform provides stable charging services by intelligently allocating power.

[0067] (2) The data cleaning unit improves the integrity and accuracy of the data set by cleaning outliers and filling missing values, eliminates analysis bias caused by data anomalies, and provides a high-quality data foundation for subsequent model training and prediction. The data standardization unit eliminates the differences in different parameter dimensions through the Z-score standard method, so that charging pile operation data, power grid load data, and user charging behavior data can be input into the prediction model with a unified standard, improving the algorithm's computational efficiency and the interpretability of the results.

[0068] The data acquisition module integrates charging pile operation data, user charging behavior data, equipment load data and power grid power demand data to form a wide-coverage, rich-dimensional data set W, which provides comprehensive data support for power grid load forecasting and charging pile scheduling. Through real-time monitoring of parameters such as power grid load fluctuation rate and equipment load fluctuation, the platform can refine its perception of the power grid operation status, provide a more accurate basis for the intelligent scheduling of charging piles, and effectively balance the power grid load.

[0069] (3) The data feature extraction unit extracts the key time feature Ttm and historical load feature LH from the data standard set WB, encodes the time feature through sine and cosine transforms, and averages the historical load features using the sliding window method, which can fully capture seasonal, cyclical, and trend features. This method improves the accuracy of feature expression and helps the prediction model more effectively identify load fluctuation patterns in different time periods.

[0070] The system integrates seasonal, cyclical and trend characteristics and combines regression models to dynamically predict future load change trends and charging needs. In particular, the in-depth analysis of multi-dimensional characteristics such as daily cycles, weekly cycles, and monthly cycles significantly improves the accuracy of load forecasting and provides a reliable basis for dynamic regulation.

[0071] The load forecast value Ye output by the regression prediction unit is combined with the preset peak threshold TGY and valley threshold TDY to build a dynamic control mechanism. During peak periods, the system can actively adjust the power output of the charging pile to reduce the pressure on the power grid; maintain stable output during valley periods to avoid waste of resources; and balance the load during stable periods to ensure that power output matches demand. This control method effectively reduces energy waste and improves system efficiency.

[0072] (4) The power allocation unit combines the load prediction value Ye, the peak power threshold Phi, and the valley power threshold Plo to achieve dynamic adjustment of the charging pile power, avoiding grid overload or resource waste caused by excessive or low load. Power allocation formulas for peak, valley, and medium loads are provided, so that the system can accurately adjust the output power of the charging pile in different load periods to achieve the optimal configuration of power resources.

[0073] The working state adjustment unit is used to dynamically switch the charging pile between start, pause and low power mode, and adjust the number of active charging piles Nac according to the real-time load conditions. The maximum number of charging piles Nmax is enabled during peak load periods, and the number of active charging piles is dynamically reduced during off-peak periods to improve resource utilization efficiency and reduce energy consumption. The state monitoring unit monitors the output power Pc of the charging pile and the grid load status Lc in real time, and identifies abnormal conditions in a timely manner in combination with the power threshold Pth and the grid load threshold Lth. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 This is a schematic diagram of the flow chart of a charging pile intelligent control management platform of the present invention. DETAILED DESCRIPTION

[0075] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0076] Example 1

[0077] The present invention provides a charging pile intelligent control management platform, please refer to Figure 1 , including data acquisition module, data preprocessing module, load forecasting and demand analysis module, charging pile scheduling and power regulation module, real-time monitoring and feedback module and report generation and decision support module;

[0078] The data acquisition module is responsible for real-time collection of charging pile operation data, equipment load data and grid power demand data to form a data set W;

[0079] The data preprocessing module cleans and standardizes the data set W to obtain a data standard set WB;

[0080] The load forecasting and demand analysis module analyzes the data standard set WB by using a regression model to predict future grid load changes and charging pile requirements;

[0081] The charging pile scheduling and power regulation module dynamically adjusts the power output of each charging pile according to the predicted grid load changes and charging pile requirements, and balances the grid load by adjusting the working state of the charging piles;

[0082] The real-time monitoring and feedback module monitors the working status, power output and power consumption of each charging pile, and collects and feeds back the operating data;

[0083] The report generation and decision support module generates a decision support report on grid load, charging pile status and power adjustment according to the fed-back operation data.

[0084] In this embodiment, the platform can identify the peak and valley periods of the power grid load in advance by collecting and analyzing the power grid load data and the operating status of the charging piles in real time, combined with the regression model of the load forecasting and demand analysis module. Through the charging pile scheduling and power regulation module, the power output of each charging pile is dynamically adjusted to optimize the allocation of power grid resources, avoid overload, and improve the stability of power grid operation. By monitoring and feeding back the user's charging behavior data, the platform can reasonably plan the working status of the charging piles. Avoid some charging piles from being worn out too quickly due to continuous overload, and reduce the waste of resources of idle charging piles, thereby improving the overall utilization efficiency of the equipment and extending the service life of the charging piles.

[0085] Using the load forecasting and demand analysis module, the platform can predict the peak of user charging demand and optimize the dispatching strategy in advance to shorten the user's waiting time. During the peak load period of the power grid, the platform provides stable charging services by intelligently allocating power to avoid service interruptions or user dissatisfaction caused by overload. Through the report generation and decision support module, the platform can generate a comprehensive report containing future power grid load forecasts, charging pile power dispatching strategies and system optimization suggestions, providing operators and managers with a scientific basis for decision-making. This comprehensive data support improves the management efficiency and responsiveness of the platform.

[0086] The platform introduces load forecasting models and user behavior analysis, which makes up for the lack of forecasting and control capabilities in traditional charging pile management methods, enabling the platform to respond to real-time changing charging needs and optimize resource allocation. The system is designed in a modular way, with each module focusing on specific tasks such as data preprocessing, forecasting analysis, scheduling and feedback, ensuring that each function works together, facilitating function expansion and updating, and improving the system's ability to adapt to complex scenarios.

[0087] Through the real-time monitoring and feedback module, the platform can quickly identify equipment failures, load anomalies and other problems, and take emergency measures. Users and managers can grasp the system operation status in real time, significantly improving the reliability of the system. Dynamic power regulation and intelligent scheduling strategies reduce energy waste and maintenance costs caused by equipment overload or idleness, while improving user satisfaction and bringing higher economic benefits.

[0088] Example 2

[0089] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the data acquisition module includes an operation data acquisition unit and a load and power grid data acquisition unit;

[0090] The operation data acquisition unit acquires the operation data of the charging pile, wherein the operation data of the charging pile includes the charging pile output power PC, the charging pile temperature TC and the charging pile cumulative power Eto, the charging pile output power PC is acquired by a power meter, the charging pile temperature TC is acquired by a temperature sensor, and the charging pile cumulative power Eto is acquired by a smart meter;

[0091] The power grid data acquisition unit acquires equipment load data and power grid power demand data, wherein the equipment load data includes equipment load fluctuation ΔPl, and the equipment load fluctuation is acquired through a current transformer;

[0092] The power demand data of the power grid is collected through power grid monitoring equipment and smart meters, including the current power grid load Lc, power grid voltage Vg and power grid load fluctuation rate ΔL. The operation data of charging piles, user charging behavior data, equipment load data and power grid power demand data are integrated to form a data set W.

[0093] The data preprocessing module includes a data cleaning unit and a data standardization unit;

[0094] The data cleaning unit cleans the abnormal values ​​of the data set W and fills the missing values ​​by using statistical methods and interpolation algorithms to obtain a data cleaning set WC;

[0095] The data standardization unit converts the data cleaning set WC into a unified standard through the Z-score standard method, eliminates the dimensional differences between the data, and obtains the data standard set WB

[0096] In this embodiment, the data cleaning unit improves the integrity and accuracy of the data set by cleaning outliers and filling missing values, eliminates analysis biases caused by data anomalies, and provides a high-quality data foundation for subsequent model training and prediction. The data standardization unit eliminates the differences in different parameter dimensions through the Z-score standard method, so that charging pile operation data, power grid load data, and user charging behavior data can be input into the prediction model with a unified standard, improving the algorithm's computational efficiency and the interpretability of the results.

[0097] The data acquisition module integrates charging pile operation data, user charging behavior data, equipment load data and power grid power demand data to form a wide-coverage, rich-dimensional data set W, which provides comprehensive data support for power grid load forecasting and charging pile scheduling. Through real-time monitoring of parameters such as power grid load fluctuation rate and equipment load fluctuation, the platform can refine its perception of the power grid operation status, provide a more accurate basis for the intelligent scheduling of charging piles, and effectively balance the power grid load.

[0098] Compared with the traditional system that only monitors some parameters, this system provides more comprehensive information on the status of power grids and charging piles through multi-source data collection. It improves the accuracy of load forecasting and scheduling strategies, allowing the platform to more effectively deal with complex power grid load scenarios. The data cleaning unit can effectively solve data anomalies and missing problems, eliminating prediction deviations caused by data noise; the data standardization unit further optimizes the expression of each parameter to make the data features consistent.

[0099] The system collects user charging behavior data such as charging duration, frequency, and power consumption, and combines it with operation data and power grid data to establish a direct connection between user needs and power grid resources, making the dispatch plan more in line with actual needs. While meeting user charging needs, it effectively balances the power grid load, improves user experience and energy efficiency. The system can monitor parameters such as power grid load fluctuation rate and equipment load fluctuation, perceive the power grid status in real time and respond quickly to fluctuations, and optimize power grid resource dispatch through load forecasting models. Through high-quality data preprocessing and multi-source data integration, the platform realizes the combination of intelligent prediction and real-time regulation, avoiding the lack of real-time performance or inaccurate analysis in existing systems, and making the power grid load more balanced.

[0100] By combining user behavior data with grid demand data, the platform can accurately match the working mode and power distribution plan of the charging pile, avoiding the conflict between user demand and grid load, and improving the utilization efficiency of the charging pile and user satisfaction.

[0101] Example 3

[0102] This embodiment is explained in Example 2. Please refer to Figure 1 ,Specifically: the load forecasting and demand analysis module includes a data feature extraction unit and a regression prediction unit;

[0103] The data feature extraction unit is responsible for extracting characteristic parameters related to load forecasting from the data standard set WB, including time characteristics Ttm and historical load characteristics LH, to form a feature set F = {Ttm, LH};

[0104] Among them, the time feature Ttm is extracted by sine transform Tsin and cosine transform Tcos;

[0105] The data feature extraction unit analyzes seasonal features, periodic features and trend features in historical data to provide input data for subsequent prediction models; specifically:

[0106] Seasonal characteristics include seasonal changes. In different seasons, the grid load is different. Summer and winter are peak load periods, while spring and autumn are low load periods.

[0107] The periodic characteristics include daily cycle, weekly cycle and monthly cycle; the daily cycle includes the power grid load fluctuations in the morning, noon, evening and midnight, and the comparison is carried out to obtain the time period of high load time and mark it as the load fluctuation of the day;

[0108] The weekly cycle includes the grid load fluctuations on working days and the grid load fluctuations on weekends, which are obtained by accumulating the characteristic load fluctuations of the daily cycle and marked as weekly load fluctuations;

[0109] The monthly cycle includes the grid load fluctuation at the beginning of the month, the grid load fluctuation in the middle of the month and the grid load fluctuation at the end of the month, which is obtained by accumulating the weekly load fluctuations with weekly cycle characteristics;

[0110] In a one-month cycle, the grid load fluctuation is at its peak at the end of the month, the last week of the month, and the morning and evening of the working day.

[0111] Trend characteristics include long-term load growth trends, which are captured through historical load data;

[0112] The period of the time feature Ttm is one day: sine and cosine are used to encode the time feature, and the formula is as follows:

[0113]

[0114] In the formula, Tsin represents the sine transform of the time feature, and Tcos represents the cosine transform of the time feature. This processing method helps the model capture the fluctuation of charging demand within 24 hours a day. The sine transform Tsin of the time feature and the cosine transform Tcos of the time feature are fitted to form the time feature Ttm.

[0115] The historical load characteristics LH are extracted by using a sliding window based on the data standard set WB grid current load Lc, equipment load fluctuation ΔPl and grid load fluctuation rate ΔL;

[0116]

[0117] Where N represents the total number of time periods.

[0118] The regression prediction unit trains a regression model through the feature set F and the data standard set WB, and outputs a load prediction value Ye of future load and charging demand; obtains the load prediction value Ye through the regression model;

[0119] Through the intercept β0 of the regression model and the regression coefficient β of the jth feature of the feature set F j Perform weighted combination to obtain the load forecast value Ye, identify peak and valley periods, and use it to regulate the power output of the charging pile;

[0120] The load forecast value Ye is obtained by the following formula:

[0121]

[0122] Where Ye represents the load forecast value, specifically the predicted future grid load and charging pile demand, β0 represents the intercept of the model, and F j represents the jth feature of feature set F, β j represents the regression coefficient of the jth feature, and m represents the number of features;

[0123] Through the load prediction value Ye, the preset peak threshold TGY and the valley threshold TDY, the power output of the charging pile is dynamically regulated to obtain the charging demand, which is obtained by matching in the following ways:

[0124] When the load forecast value Ye ≥ the peak threshold TGY, it means that the charging pile is in the peak period, and the load of the charging pile in use is adjusted;

[0125] When the load forecast value Ye is less than the valley threshold TDY, it means that the charging pile is in the valley period, and the load adjustment of the charging pile in use is not performed;

[0126] When the peak threshold TGY>load prediction value Ye≥valley threshold TDY, it means that the charging pile is in a stable period and the charging pile load is not adjusted.

[0127] The charging demand is obtained by the following formula:

[0128] Through the above judgment conditions, the system can identify the peak and trough periods of demand, which is convenient for rationally regulating resource allocation and grid load scheduling. The regression prediction unit outputs the predicted future load changes and demand distribution. The generated results help the grid to make scheduling plans in advance and ensure grid load balance. Charging piles can automatically adjust power output during different demand periods.

[0129] In this embodiment, the data feature extraction unit extracts the key time feature Ttm and historical load feature LH from the data standard set WB, encodes the time feature through sine and cosine transforms, and averages the historical load features using the sliding window method, which can fully capture seasonal, periodic and trend features. This method improves the accuracy of feature expression and helps the prediction model to more effectively identify load fluctuation patterns in different time periods.

[0130] The system integrates seasonal, cyclical and trend characteristics and combines regression models to dynamically predict future load change trends and charging needs. In particular, the in-depth analysis of multi-dimensional characteristics such as daily cycles, weekly cycles, and monthly cycles significantly improves the accuracy of load forecasting and provides a reliable basis for dynamic regulation.

[0131] The load forecast value Ye output by the regression prediction unit is combined with the preset peak threshold TGY and valley threshold TDY to build a dynamic control mechanism. During peak periods, the system can actively adjust the power output of the charging pile to reduce the pressure on the power grid; maintain stable output during valley periods to avoid waste of resources; and balance the load during stable periods to ensure that power output matches demand. This control method effectively reduces energy waste and improves system efficiency.

[0132] The system uses the load forecast value Ye to identify the peak and valley periods of the power grid in advance, realize reasonable resource allocation, and optimize the load dispatch of the power grid. The forecast results provide data support for the grid to formulate dispatch plans, avoiding grid instability caused by sudden load changes. During peak periods, by regulating the power output of charging piles, the risk of grid overload is alleviated, while ensuring the safe operation of charging piles; during valley periods, maintaining stable power output improves charging efficiency and realizes the rational use of power resources.

[0133] Example 4

[0134] This embodiment is explained in Example 3, please refer to Figure 1 ,Specifically: the charging pile scheduling and power regulation module includes a power distribution unit and a working state regulation unit;

[0135] The power distribution unit determines the power output of the charging pile in different time periods according to the load prediction value Ye, the preset peak power threshold Phi and the valley power threshold Plo; adjusts the power during the peak period and keeps the power unchanged during the valley period;

[0136] The power allocation decision is obtained by matching:

[0137] When the load prediction value Ye ≥ the peak power threshold Phi, the power of the charging pile is adjusted;

[0138] When the load prediction value Ye≤the valley power threshold Plo, the power of the charging pile is maintained.

[0139] The specific formula is:

[0140]

[0141] In the formula, Pou represents the final output power of the charging pile, Plim represents the power limit during peak hours, Pmax represents the maximum power during off-peak hours, and Padj represents the charging pile power after demand adjustment during medium load. Through the power allocation formula, the system can automatically adjust the output power during different load periods to avoid grid overload.

[0142] The working state adjustment unit is responsible for managing the start, pause and low power mode switching of the charging pile;

[0143] According to the power allocation decision result output by the power allocation unit, the working state adjustment unit adjusts the number of charging piles in use Nac during the peak period and maintains the working frequency during the valley period;

[0144] When the load forecast value Ye ≥ the peak power threshold Phi, the number of charging piles in use Nac is equal to the maximum number of charging piles Nmax, and the output power PC of the charging pile is adjusted to 120% of the original power;

[0145] When the load forecast value Ye≤the valley power threshold Plo, the number of charging piles in use Nac is obtained by the difference between the maximum number of charging piles Nmax, the load forecast value Ye and the peak power threshold Phi, and the output power PC of the charging pile is adjusted to 90% of the original power.

[0146] The number of active charging piles Nac is obtained by the following formula:

[0147]

[0148] Where Nmax represents the maximum number of charging piles, Nadj represents the number of active piles at medium load, which is dynamically adjusted by demand balance, and k represents the adjustment coefficient, which determines the proportion of reducing the number of active charging piles under high load conditions.

[0149] The real-time monitoring and feedback module includes a status monitoring unit and an alarm feedback unit;

[0150] The state monitoring unit is responsible for real-time monitoring of the charging pile output power PC and the current load Lc of the power grid, and judges the state of the charging pile by comparing the charging pile output power PC with the preset power threshold Pth:

[0151] The status of the charging pile is obtained by matching in the following way:

[0152] When the charging pile output power PC>power threshold Pth, the charging pile overload alarm signal is triggered;

[0153] When the charging pile output power PC ≤ power threshold Pth, the charging pile overload alarm signal will not be triggered;

[0154] By comparing the current grid load Lc with the preset grid load threshold Lth, the grid load state is determined;

[0155] When the current grid load Lc>grid load threshold Lth, the grid load overload alarm signal is triggered;

[0156] When the current grid load Lc ≤ the grid load threshold Lth, the grid load overload alarm signal will not be triggered;

[0157] The grid load fluctuation rate ΔL can also be used to determine the grid load status, and through continuous data collection and analysis, potential overload and abnormal problems can be identified in a timely manner.

[0158] In this embodiment, the power distribution unit combines the load prediction value Ye, the peak power threshold Phi and the valley power threshold Plo to achieve dynamic adjustment of the charging pile power, avoiding grid overload or resource waste caused by excessively high or low loads. Power distribution formulas for peak, valley and medium loads are provided, so that the system can accurately adjust the output power of the charging pile in different load periods to achieve the optimal configuration of power resources.

[0159] The working state adjustment unit is used to dynamically switch the charging pile between start, pause and low power mode, and adjust the number of active charging piles Nac according to the real-time load conditions. The maximum number of charging piles Nmax is enabled during peak load periods, and the number of active charging piles is dynamically reduced during off-peak periods to improve resource utilization efficiency and reduce energy consumption. The state monitoring unit monitors the output power Pc of the charging pile and the grid load status Lc in real time, and identifies abnormal conditions in a timely manner in combination with the power threshold Pth and the grid load threshold Lth.

[0160] Dynamically adjust power output and the number of active charging piles to meet user needs while reducing energy waste and equipment idle rate, thereby improving the overall operating efficiency of the system. Realize the deep integration of charging pile power regulation and dynamic monitoring of power grid load, reduce the impact of load fluctuations on power grid stability, and ensure that the power grid can operate smoothly during both peak and off-peak periods. Through dynamic power regulation and flexible state management, power resources are reasonably allocated to avoid overload problems during peak periods and idle resources during off-peak periods. Real-time monitoring and alarm feedback mechanisms enable the system to promptly detect and respond to potential overload risks, avoiding the risk of equipment damage or downtime caused by excessive load or abnormal fluctuations.

[0161] Example 5

[0162] This embodiment is explained in Example 4. Please refer to Figure 1 Specifically: the alarm feedback unit feeds back the overload condition and the fault condition according to the alarm signal output by the status monitoring unit, sends a real-time alarm to the operator, and automatically triggers a strategy adjustment signal to adjust the output power PC of the charging pile and the number of active charging piles Nac;

[0163] The charging pile power overload alarm signal and the power grid overload alarm signal are integrated to obtain an alarm feedback signal AF;

[0164] When the alarm feedback signal AF=1, the alarm information is fed back, triggering the strategy adjustment to reduce the power of the charging pile.

[0165] The strategy adjustment formula is:

[0166]

[0167] Where Padj represents the charging pile power after demand adjustment when the load is medium, and Psa represents the power output in safety mode to avoid overload.

[0168] The report generation and decision support module includes a data analysis and report generation unit and a decision support suggestion generation unit;

[0169] The data analysis and report generation unit receives the feedback of grid load, charging pile status, and power adjustment data, aggregates, counts, and performs trend analysis on these data; based on the analysis results, generates report content, including future grid load forecast, charging pile power dispatch strategy, and equipment operation status;

[0170] The decision support suggestion generating unit generates adjustment decision suggestions according to the data results of the analysis report;

[0171] Specifically: during peak load periods, reduce the output power of the charging piles and adjust the charging piles to start during low load periods;

[0172] During off-load periods, users are encouraged to charge to reduce overall grid fluctuations and increase the output power of charging piles.

[0173] The future grid load forecasting formula is:

[0174] L t+1 =α*Lt+β*Ut+γ*Et+δ;

[0175] Where, L t+1 represents the predicted grid load at time t+1, Ut represents the user charging behavior data at time t, Et represents the current load status of the charging pile equipment, α, β, γ and δ represent regression coefficients, which are obtained based on historical data and model training;

[0176] This formula is used to predict the load of the power grid in the next period of time, providing a basis for decision support to generate load adjustment suggestions. This value is used to determine whether to adjust the power output of the charging pile in a specific time period;

[0177] Adjust the power output of the charging pile according to the set load forecast results;

[0178] The formula for charging pile power scheduling strategy is:

[0179]

[0180] Where, Padj represents the adjusted charging pile power, Pmax represents the maximum power output of the charging pile, and Lth represents the grid load threshold;

[0181] By predicting that the load exceeds the grid load threshold Lth, the power output is automatically reduced to prevent grid overload;

[0182] If the load is lower than the grid load threshold Lth, the charging pile will maintain maximum power;

[0183] The equipment operation status is judged by the load balance index B

[0184]

[0185] Where P i adj represents the power after adjustment of the i-th charging pile, P i base represents the power of the i-th charging pile before load balancing, and n represents the number of charging piles;

[0186] The load balance index B shows that the more reasonable the power distribution of the charging pile is, the better the equipment operation status is, which helps to reduce the load of a single charging pile and balance the pressure on the power grid.

[0187] In this embodiment, an alarm feedback unit is introduced to trigger a strategy adjustment signal through comprehensive analysis of the charging pile power overload alarm signal and the power grid overload alarm signal, so as to realize automatic adjustment of the charging pile output power PC and the number of active charging piles Nac. The response speed of charging pile scheduling is significantly improved, the impact of manual intervention delay on power grid overload is avoided, and the safety and stability of the system are enhanced. Combining historical data with real-time data, a trend analysis report is generated through the data analysis and report generation unit, including future power grid load forecasts and charging pile power scheduling strategies. The decision support unit generates load adjustment recommendations based on the report data.

[0188] Design a charging pile power dispatch strategy formula to dynamically adjust the power output according to the predicted load and actual load, automatically reduce the power during peak hours, and encourage users to charge during off-peak hours. Effectively balance the grid load pressure during peak hours, while making full use of power resources during off-peak hours, improve the overall system utilization and improve user experience. Continuously monitor and analyze the charging pile power PC, grid load Lc, and load fluctuation rate ΔL through a real-time monitoring unit to ensure that potential abnormal situations are identified in a timely manner. Establish an efficient closed-loop management mechanism, so that the charging pile dispatch system can achieve flexible and intelligent power regulation under various load scenarios, and improve the safety and reliability of overall operation.

[0189] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A charging pile intelligent control management platform, characterized by: It includes data acquisition module, data preprocessing module, load forecasting and demand analysis module, charging pile scheduling and power regulation module, real-time monitoring and feedback module, and report generation and decision support module; The data acquisition module is responsible for real-time collection of charging pile operation data, equipment load data and grid power demand data to form a data set W; The data preprocessing module cleans and standardizes the data set W to obtain a data standard set WB; The load forecasting and demand analysis module analyzes the data standard set WB by using a regression model to predict future grid load changes and charging pile requirements; The charging pile scheduling and power regulation module dynamically adjusts the power output of each charging pile according to the predicted grid load changes and charging pile requirements, and balances the grid load by adjusting the working state of the charging piles; The real-time monitoring and feedback module monitors the working status, power output and power consumption of each charging pile, and collects and feeds back the operating data; The report generation and decision support module generates a decision support report on grid load, charging pile status and power adjustment based on the feedback operation data.

2. The charging pile intelligent control management platform according to claim 1 is characterized by: The data acquisition module includes an operation data acquisition unit and a load and power grid data acquisition unit; The operation data acquisition unit acquires the operation data of the charging pile, wherein the operation data of the charging pile includes the charging pile output power PC, the charging pile temperature TC and the charging pile cumulative power Eto, the charging pile output power PC is acquired by a power meter, the charging pile temperature TC is acquired by a temperature sensor, and the charging pile cumulative power Eto is acquired by a smart meter; The power grid data acquisition unit acquires equipment load data and power grid power demand data, wherein the equipment load data includes equipment load fluctuation ΔPl, and the equipment load fluctuation is acquired through a current transformer; The power demand data of the power grid is collected through power grid monitoring equipment and smart meters, including the current power grid load Lc, power grid voltage Vg and power grid load fluctuation rate ΔL. The operation data of charging piles, user charging behavior data, equipment load data and power grid power demand data are integrated to form a data set W.

3. The charging pile intelligent control management platform according to claim 1 is characterized by: The data preprocessing module includes a data cleaning unit and a data standardization unit; The data cleaning unit cleans the abnormal values ​​of the data set W and fills the missing values ​​by using statistical methods and interpolation algorithms to obtain a data cleaning set WC; The data standardization unit converts the data cleaning set WC into a unified standard through the Z-score standard method, eliminates the dimensional differences between the data items, and obtains the data standard set WB.

4. The charging pile intelligent control management platform according to claim 1, characterized in that: The load forecasting and demand analysis module includes a data feature extraction unit and a regression prediction unit; The data feature extraction unit is responsible for extracting characteristic parameters related to load forecasting from the data standard set WB, including time characteristics Ttm and historical load characteristics LH, to form a feature set F; Among them, the time feature Ttm is extracted through sine transform Tsin and cosine transform Tcos; The historical load characteristics LH are extracted by using a sliding window based on the data standard set WB grid current load Lc, equipment load fluctuation ΔPl and grid load fluctuation rate ΔL; The data feature extraction unit analyzes seasonal features, periodic features and trend features in historical data to provide input data for subsequent prediction models; specifically: Seasonal characteristics include seasonal changes. In different seasons, the grid load is different. Summer and winter are peak load periods, while spring and autumn are low load periods. The periodic characteristics include daily cycle, weekly cycle and monthly cycle; the daily cycle includes the power grid load fluctuations in the morning, noon, evening and midnight, and the comparison is carried out to obtain the time period of high load time and mark it as the load fluctuation of the day; The weekly cycle includes the grid load fluctuations on working days and the grid load fluctuations on weekends, which are obtained by accumulating the characteristic load fluctuations of the daily cycle and marked as weekly load fluctuations; The monthly cycle includes the grid load fluctuation at the beginning of the month, the grid load fluctuation in the middle of the month and the grid load fluctuation at the end of the month, which is obtained by accumulating the weekly load fluctuations with weekly cycle characteristics; Trend characteristics include long-term load growth trends, which are captured through historical load data.

5. The charging pile intelligent control management platform according to claim 4 is characterized by: The regression prediction unit trains a regression model through the feature set F and the data standard set WB, and outputs a load prediction value Ye of future load and charging demand; obtains the load prediction value Ye through the regression model; Through the intercept β0 of the regression model and the regression coefficient β of the jth feature of the feature set F j Perform weighted combination to obtain the load forecast value Ye, identify peak and valley periods, and use it to regulate the power output of the charging pile; Through the load prediction value Ye, the preset peak threshold TGY and the valley threshold TDY, the power output of the charging pile is dynamically regulated to obtain the charging demand, which is obtained by matching in the following ways: When the load prediction value Ye ≥ the peak threshold TGY, it means that the charging pile is in the peak period, and the load of the charging pile in use is adjusted; When the load forecast value Ye is less than the valley threshold TDY, it means that the charging pile is in the valley period, and the load adjustment of the charging pile in use is not performed; When the peak threshold TGY>load prediction value Ye≥valley threshold TDY, it means that the charging pile is in a stable period and the charging pile load is not adjusted.

6. The charging pile intelligent control management platform according to claim 5 is characterized by: The charging pile scheduling and power regulation module includes a power distribution unit and a working state regulation unit; The power distribution unit determines the power output of the charging pile in different time periods according to the load prediction value Ye, the preset peak power threshold Phi and the valley power threshold Plo; adjusts the power during the peak period and keeps the power unchanged during the valley period to obtain the power distribution decision; The power allocation decision is obtained by matching: When the load forecast value Ye ≥ the peak power threshold Phi, the power of the charging pile is adjusted; When the load prediction value Ye≤the valley power threshold Plo, the power of the charging pile is maintained.

7. The charging pile intelligent control management platform according to claim 6 is characterized by: The working state adjustment unit is responsible for managing the start, pause and low power mode switching of the charging pile; According to the power allocation decision result output by the power allocation unit, the working state adjustment unit adjusts the number of charging piles in use Nac during the peak period and maintains the working frequency during the off-peak period; When the load forecast value Ye ≥ the peak power threshold Phi, the number of charging piles in use Nac is equal to the maximum number of charging piles Nmax, and the output power PC of the charging pile is adjusted to 120% of the original power; When the load prediction value Ye≤the valley power threshold Plo, the number of charging piles in use Nac is obtained by the difference between the maximum number of charging piles Nmax and the load prediction value Ye and the peak power threshold Phi, and the output power PC of the charging pile is adjusted to 90% of the original power.

8. The charging pile intelligent control management platform according to claim 2 is characterized by: The real-time monitoring and feedback module includes a status monitoring unit and an alarm feedback unit; The state monitoring unit is responsible for real-time monitoring of the charging pile output power PC and the current load Lc of the power grid, and judges the state of the charging pile by comparing the charging pile output power PC with the preset power threshold Pth: The status of the charging pile is obtained by matching in the following way: When the charging pile output power PC>power threshold Pth, the charging pile overload alarm signal is triggered; When the charging pile output power PC ≤ power threshold Pth, the charging pile overload alarm signal will not be triggered; By comparing the current grid load Lc with the preset grid load threshold Lth, the grid load state is determined; When the current grid load Lc>grid load threshold Lth, the grid load overload alarm signal is triggered; When the current grid load Lc ≤ the grid load threshold Lth, the grid load overload alarm signal is not triggered.

9. The charging pile intelligent control management platform according to claim 8, characterized in that: The alarm feedback unit provides feedback on overload and fault conditions according to the alarm signal output by the status monitoring unit, sends a real-time alarm to the operator, and automatically triggers a strategy adjustment signal to adjust the charging pile output power PC and the number of active charging piles Nac; The charging pile power overload alarm signal and the power grid overload alarm signal are integrated to obtain an alarm feedback signal AF; When the alarm feedback signal AF=1, the alarm information is fed back, triggering the strategy adjustment to adjust the charging pile power.

10. The charging pile intelligent control management platform according to claim 1, characterized in that: The report generation and decision support module includes a data analysis and report generation unit and a decision support suggestion generation unit; The data analysis and report generation unit receives the feedback of grid load, charging pile status, and power adjustment data, aggregates, counts, and performs trend analysis on these data; and generates report content based on the analysis results, including future grid load forecasts, charging pile power scheduling strategies, and equipment operating status; The decision support suggestion generating unit generates adjustment decision suggestions according to the data results of the analysis report; Specifically: during peak load periods, adjust the output power of the charging pile; Users are encouraged to charge during off-peak load periods.

Images