Charging pile group energy management method

Abstract

The invention relates to the technical field of power management, in particular to a charging pile group energy management method, which comprises the following steps: S1, predicting the power demand of each charging pile in a short period and the available amount of renewable energy; s2, adjusting the charging power of each charging pile through a dynamic allocation algorithm; s3, encouraging a user to charge in an off-peak period based on a preset pricing algorithm; s4, predicting a user charging demand mode; s5, monitoring electric energy quality in a charging process in real time, and automatically adjusting parameters of charging equipment by using a preset electric energy adjusting technology; and S6, automatically activating a preset standby energy supply scheme under the emergency condition of the power grid fault. According to the invention, energy distribution is optimized through real-time monitoring and intelligent adjustment of the charging strategy, the power grid efficiency is improved, the cost is reduced, and the continuity and safety of the charging service are ensured through the standby energy system, so that the operation efficiency of the charging infrastructure and the user satisfaction are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power management, and in particular to an energy management method for a charging pile group. Background Art

[0002] With the popularity of electric vehicles and the increasing demand of consumers for environmentally friendly travel modes, the construction of charging infrastructure has expanded rapidly. However, the existing charging infrastructure faces many challenges, especially in terms of grid load management, energy efficiency optimization and service reliability. The peaks and valleys of charging demand do not always match the power supply capacity of the grid, resulting in insufficient power supply during high-demand periods and energy waste during low-demand periods. In addition, grid failures or instability often lead to interruptions in the charging process, affecting user experience and the availability of electric vehicles.

[0003] Existing technologies mainly focus on improving the construction efficiency of charging stations and expanding network coverage, and pay less attention to how to intelligently manage the charging process to cope with grid fluctuations and the diversity of user behavior. The monitoring and management of power quality during the charging process are usually insufficient, and cannot effectively prevent or mitigate equipment damage and battery performance degradation caused by problems such as voltage fluctuations and harmonic pollution. In addition, there is a lack of effective demand response mechanisms to smooth grid loads and reduce pressure during peak hours.

[0004] Therefore, there is an urgent need to develop a comprehensive energy management method to monitor and adjust charging strategies in real time, optimize energy distribution, improve the intelligence level and operational efficiency of the charging network, and ensure the continuity and safety of charging services. Summary of the invention

[0005] Based on the above purpose, the present invention provides a charging pile group energy management method.

[0006] A charging pile group energy management method comprises the following steps: S1: Based on historical charging data and regional grid load information, a hybrid prediction model of SVR and ELM is used to predict the short-term power demand of each charging pile and the available amount of renewable energy; S2: Based on the prediction results in S1, combined with the current grid load and the predicted renewable energy output, the charging power of each charging pile is adjusted through a dynamic allocation algorithm to optimize energy distribution; S3: During peak demand periods, a preset pricing algorithm encourages users to charge during off-peak hours to smooth the electricity demand curve and reduce grid load; S4: Analyze users’ charging behaviors and preferences to predict users’ charging demand patterns and automatically adjust charging strategies and resource allocation based on the analysis results; S5: Real-time monitoring of power quality during charging, including voltage fluctuations and harmonic pollution, and automatic adjustment of charging equipment parameters using preset power regulation technology to optimize power quality, protect equipment and extend battery life; S6: In the event of an emergency with a power grid failure, the preset backup energy supply plan is automatically activated to ensure the continuity and safety of charging services.

[0007] Furthermore, the S1 specifically includes: S11: Collect historical charging data of each charging pile, including information on charging duration, charging start and end time, and charging power; collect grid load data and historical output data of renewable energy, and clean and pre-process the collected data to remove outliers and fill in missing values ​​to ensure data quality and integrity; S12: Based on the preprocessed data, develop a feature set that includes time characteristics, weather conditions, and historical electricity demand and energy supply data for the same period; S13: Use support vector regression and extreme learning machine to train the model, where the SVR model is used to analyze the nonlinear trends and complex dependencies of electricity demand, and the ELM model is used to quickly learn and adapt to the fluctuations of renewable energy output. The prediction results of the two models are then weighted averaged, with the weights determined based on the historical prediction accuracy, to form a hybrid prediction model; S14: Apply the hybrid prediction model formed in S13 to the real-time data stream to predict the power demand of each charging pile and the available amount of solar energy and wind energy in the next short term.

[0008] Furthermore, the S13 specifically includes: S131: First, the SVR model is used to process and predict power demand data. SVR is an effective nonlinear regression method based on statistical learning theory. Its goal is to find a function , so that the sum of squares of deviations between a given data point and the function is minimized while ensuring the smoothness of the function; S132: Then use ELM to process and predict renewable energy output, specifically the fluctuations of solar and wind energy. ELM is a single-layer feedforward neural network. The weights and biases of its hidden layer nodes are randomly initialized and fixed. The output weights are quickly calculated analytically. The optimization goal of ELM is to minimize the norm and training error of the output layer weights. The specific formula is: ,in, is the output matrix of the hidden layer, is the output layer weight, is the target output, is the regularization coefficient, which is used to control the complexity of the model and avoid overfitting; S133: Finally, the prediction results of the SVR and ELM models are combined to form a hybrid prediction model, which is achieved by applying a weighted average to the prediction results of each model. The weights are calculated based on their respective prediction performance on the validation set. The specific formula is: ,in, is the final prediction result. and are the prediction results of the SVR and ELM models, The weights are determined based on historical forecast accuracy.

[0009] Furthermore, the S2 specifically includes: S21: First, the short-term power demand forecast results of each charging pile obtained in step S1 are integrated with the real-time grid load data and the output forecast results of renewable energy, ensuring that all data are up-to-date and complete before making energy allocation decisions; S22 defines a dynamic energy allocation strategy, which determines the charging power of each charging pile based on the current total grid load, demand forecast of each charging pile, and the availability of renewable energy. This strategy uses the following formula to calculate the charging power allocation of each charging pile: ,in, It is The distribution power of each charging pile, It is The predicted demand for charging piles, is the total number of charging piles, is the total power currently available on the grid, is the predicted renewable energy power; S23: Dynamically adjust the charging power setting of each charging pile according to the strategy defined in S22; S24: After implementing energy allocation, monitor the deviation between actual power consumption and predicted value, and optimize the allocation strategy according to the deviation result. Specifically, when there is a difference between actual consumption and prediction, the dynamic allocation algorithm will be automatically fine-tuned to reduce future prediction errors and optimize energy use.

[0010] Furthermore, the S24 specifically includes: S241: First, continuously monitor the deviation between the actual power consumption of each charging pile and the predicted power demand in S1, and specifically calculate the deviation using the following formula: ,in, Indicates The power consumption deviation of each charging pile, is the actual power consumption, and is the predicted power; S242: Analyze the patterns and causes of deviations, including whether there is persistent over- or under-consumption, and identify the need to adjust the forecast model or allocation strategy; S243: When the analysis results show deviations, the parameters of the dynamic allocation algorithm or prediction model will be automatically adjusted; S244: Based on the adjusted forecast model, the energy allocation strategy is redefined, specifically by modifying the coefficients in the energy allocation formula to respond to changes in the forecast model. The formula is: ,in, is a newly introduced adjustment factor used to adjust the real-time energy allocation according to historical deviations; S245: Implement the updated allocation strategy and continuously monitor its effectiveness. When the new strategy effectively reduces the prediction error and optimizes energy use, it will continue to be adopted; when the effect is not good, it will continue to be iterated and adjusted.

[0011] Furthermore, the S3 specifically includes: S31: First, determine the peak and non-peak power demand periods based on historical power grid data analysis; S32: Based on the identified peak and off-peak periods, a dynamic electricity price strategy is formulated, including increasing electricity prices during peak periods and reducing electricity prices during off-peak periods. The specific price adjustment formula is: ,in, It's in time The electricity price, is the basic electricity price, is the price adjustment factor, is the start time of the peak period, and the sign function is used to determine whether it is the peak period; S33: Implementing the mid-price strategy of S32 through the user’s charging management system or smart metering device, which will automatically adjust the electricity price and notify the user of current and expected electricity price changes to encourage the user to choose the charging time with the lowest cost; S34: Monitor changes in user charging behavior after the implementation of the dynamic pricing strategy, and analyze the impact of the adjustment of charging time periods on the grid load, including tracking the proportion of charging volume moving from peak hours to off-peak hours, and its contribution to the overall grid load smoothness; S35: Adjust pricing coefficient based on monitoring results Or modify the definition of peak and off-peak periods to encourage users to charge when the grid load is lower, optimize the electricity demand curve and reduce the grid load.

[0012] Furthermore, the S4 specifically includes: S41: First, the charging records of each charging pile user are collected regularly, including charging time, charging duration, charging frequency and charging amount data. In addition, relevant data affecting the user's charging behavior is collected, including vehicle model, battery capacity and historical charging behavior pattern; S42: Use data mining and machine learning technology to analyze the charging behavior patterns of users, and identify user groups with similar charging behaviors by grouping users, specifically dividing users into nighttime chargers, weekday chargers, and weekend chargers; S43: Based on the behavior pattern analysis results, a prediction model is established to predict the charging demand of different user groups to predict the future charging demand pattern; S44: Automatically adjust the charging strategy and resource allocation of charging piles based on the output of the demand forecasting model. Specifically, when the forecast shows that the demand for nighttime chargers will increase, the allocation of charging resources at night will be automatically increased to ensure that the demand is met. In addition, the electricity price is dynamically adjusted based on user preferences and charging needs to encourage users to charge during non-peak hours.

[0013] Furthermore, the S43 specifically includes: S431: extracting features from the user's charging history records, including charging time, duration, frequency, charging completion, and user type, and normalizing the extracted feature data to remove the impact of the dimension; S432: Construct a regression model based on Gaussian process to predict charging demand. Gaussian process is a random process defined on a continuous input space. Any set of finite observation points obeys the multivariate normal distribution. The model formula is expressed as: ; ;in, is the observed charging demand value; is the input feature vector; is the noise term, representing the random error in the measurement process; represents a Gaussian process; is the mean function, which is used to describe the average trend of the data; is the covariance function, which defines any two points in the input space and The covariance between S433: Train the model and optimize hyperparameters by maximizing the log-marginal likelihood function. The specific formula is: ,in, represents the log marginal likelihood function; represents the covariance matrix without noise; is the noise variance; represents the covariance matrix with noise added, ,in is the identity matrix; For the matrix The determinant of ; is the number of samples, indicating the total number of data points involved in training; S434: Use the trained Gaussian process regression model to predict future charging needs based on the latest user behavior data.

[0014] Furthermore, the S5 specifically includes: S51: Install power quality monitoring equipment at the charging station to monitor and record the voltage fluctuation and harmonic pollution level during the charging process in real time. The monitoring equipment can continuously collect voltage and current data. Specifically, the voltage fluctuation and total harmonic distortion are calculated using the following formula : ; ,in, It is the scheduled time The voltage fluctuation rate, and are the maximum and minimum voltage values ​​during this time, respectively. is the nominal voltage, is the total harmonic distortion, It is The voltage amplitude of the subharmonics, is the fundamental voltage; S52: determining whether there is a power quality problem by setting a threshold, specifically when the voltage fluctuation exceeds a predetermined threshold or the THD exceeds a standard limit, identifying it as a power quality problem; S53: When a power quality problem is identified, the preset power conditioning technology will be automatically activated to adjust the charging equipment parameters, including dynamically adjusting the charging current or voltage to match the real-time conditions of the grid, reducing the impact of voltage fluctuations and harmonics, and applying harmonic filters or phase modulators to reduce harmonic content; S54: After the adjustment, the power quality continues to be monitored, and the strategy is adjusted according to the response of the equipment and the feedback of the grid status to ensure that the power quality is always maintained in the optimal state, thereby protecting the charging equipment and extending the battery life.

[0015] Furthermore, the S6 specifically includes: S61: First, the grid status is monitored in real time, including parameters of voltage, current and frequency, and grid monitoring equipment is used to continuously evaluate the stability and performance of the grid. Once it is detected that the grid parameters deviate from the normal range, it is determined to be a grid fault; S62: A preset backup energy supply plan, specifically including a battery storage system, a diesel generator and a renewable energy system. The battery storage system is used to provide immediate energy support when the power grid is out of power; the diesel generator is used to start in the event of a long-term power grid outage to maintain the operation of the charging station; the renewable energy system is used to cooperate with the energy storage system to provide a continuous energy supply for the charging station; S63: When the system detects a grid fault, it will select a predetermined backup energy supply solution based on the nature and expected duration of the fault; S64: Regularly test and maintain the backup energy supply system to ensure its reliability and effectiveness in emergency situations, including charging and discharging cycle testing of energy storage batteries, starting and operating testing of diesel generators, and performance inspection of renewable energy equipment.

[0016] Beneficial effects of the present invention: The present invention significantly improves the energy efficiency of the charging infrastructure by real-time monitoring and intelligent prediction of charging demand and the available amount of renewable energy. By accurately adjusting the charging power of each charging pile and optimizing energy distribution, it can not only ensure the load balance of the power grid, but also maximize the utilization of renewable energy. This method helps to reduce energy costs while reducing dependence on traditional energy, thereby supporting environmental protection and sustainable development goals.

[0017] The present invention effectively encourages users to charge during off-peak hours by dynamically adjusting electricity prices and automatically adjusting charging strategies, thereby smoothing the electricity demand curve and reducing the load on the power grid. This not only improves the operating efficiency and stability of the power grid, but also helps to reduce the pressure on the power grid during peak hours and avoid overload. In addition, by optimizing the distribution of charging time periods, charging stations can manage their operations more effectively and improve service quality and user satisfaction.

[0018] The present invention, through power quality monitoring and emergency response mechanism, can quickly switch to backup energy when the power grid is unstable or fails, ensuring the continuity and safety of charging services. This rapid response capability not only protects the charging equipment from damage caused by voltage fluctuations and harmonic pollution, but also extends the service life of the battery and reduces maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0020] Figure 1A schematic diagram of a charging pile group energy management method according to an embodiment of the present invention; Figure 2 Schematic diagram of a process for automatically adjusting charging device parameters according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0022] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. The words "include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects.

[0023] like Figure 1-Figure 2 As shown, a charging pile group energy management method includes the following steps: S1: Based on historical charging data and regional grid load information, a hybrid prediction model of SVR (support vector regression) and ELM (extreme learning machine) is used to predict the short-term power demand of each charging pile and the available amount of renewable energy (such as solar and wind energy); S2: Based on the prediction results in S1, combined with the current grid load and the predicted renewable energy output, the charging power of each charging pile is adjusted through a dynamic allocation algorithm to optimize energy distribution and ensure load balance of the grid and optimization of energy supply; S3: During peak demand periods, a preset pricing algorithm encourages users to charge during off-peak hours to smooth the electricity demand curve and reduce grid load; S4: Analyze users’ charging behaviors and preferences to predict users’ charging demand patterns, and automatically adjust charging strategies and resource allocation based on the analysis results to improve user experience and system efficiency; S5: Real-time monitoring of power quality during charging, including voltage fluctuations and harmonic pollution, and automatic adjustment of charging equipment parameters using preset power regulation technology to optimize power quality, protect equipment and extend battery life; S6: In the event of an emergency with a power grid failure, the system automatically activates a preset backup energy supply solution, such as a storage battery or a diesel generator, to ensure the continuity and safety of charging services.

[0024] S1 specifically includes: S11: Collect historical charging data of each charging pile, including information on charging duration, charging start and end time, and charging power; collect grid load data and historical output data of renewable energy (solar and wind energy), and clean and preprocess the collected data to remove outliers and fill in missing values ​​to ensure data quality and integrity; S12: Based on the preprocessed data, develop a feature set that includes time characteristics (such as hour, week, season), weather conditions (such as temperature, wind speed, sunshine duration), and historical electricity demand and energy supply data for the same period; these features help capture the key factors that affect charging demand and renewable energy output; S13: Use support vector regression (SVR) and extreme learning machine (ELM) to train the model, where the SVR model is used to interpret the nonlinear trends and complex dependencies of electricity demand, while the ELM model is used to quickly learn and adapt to the fluctuations of renewable energy output. The prediction results of the two models are then weighted averaged, with the weights determined based on the historical prediction accuracy, to form a hybrid prediction model; S14: Apply the hybrid prediction model formed in S13 to the real-time data stream to predict the power demand of each charging pile and the available amount of solar energy and wind energy in the next short term (for example, within the next 24 to 48 hours). The hybrid prediction model can also dynamically adjust the prediction and respond to the instant data changes to provide the most accurate prediction results; the hybrid prediction model can accurately predict the power demand of each charging pile and the available amount of solar energy and wind energy in the short term. Through optimized prediction, the efficiency and accuracy of energy distribution are improved, energy waste is reduced, and the stability and reliability of the power grid are enhanced. This prediction capability is particularly critical for managing large-scale charging pile networks, which helps to achieve more efficient energy utilization and better environmental protection effects.

[0025] S13 specifically includes: S131: First, the SVR model is used to process and predict power demand data. SVR is an effective nonlinear regression method based on statistical learning theory. Its goal is to find a function , so that the sum of the squares of the deviations between the given data points and the function is minimized while ensuring the smoothness of the function. Mathematically, SVR is implemented by solving the following optimization problem: ; ; ; ;in, is the weight vector, is the bias term, is a regularization parameter used to control model complexity and prevent overfitting; and is a slack variable used to handle data points that may not be completely within the error bound; The error tolerance defines the width of the no-penalty zone. The difference between the predicted value and the actual value of a data point must be smaller than this value, otherwise an error cost will be incurred. S132: Then use ELM to process and predict renewable energy output, specifically the fluctuations of solar and wind energy. ELM is a single-layer feedforward neural network. The weights and biases of its hidden layer nodes are randomly initialized and fixed. The output weights are quickly calculated analytically. The optimization goal of ELM is to minimize the norm and training error of the output layer weights. The specific formula is: ,in, is the output matrix of the hidden layer, is the output layer weight, is the target output, is the regularization coefficient, which is used to control the complexity of the model and avoid overfitting; S133: Finally, the prediction results of the SVR and ELM models are combined to form a hybrid prediction model, which is achieved by applying a weighted average to the prediction results of each model. The weights are calculated based on their respective prediction performance on the validation set. The specific formula is: ,in, is the final prediction result. and are the prediction results of the SVR and ELM models respectively. The weights are determined based on the accuracy of historical predictions. This hybrid prediction model combines the high accuracy of SVR and the fast learning ability of ELM to improve the accuracy and efficiency of the overall prediction. By accurately adjusting the weights of each model, it can ensure the best prediction results in different application environments, thereby effectively guiding the energy management and scheduling strategies of charging piles, optimizing the utilization of energy resources, and reducing waste.

[0026] S2 specifically includes: S21: First, the short-term power demand forecast results of each charging pile obtained in step S1 are integrated with the real-time grid load data and the output forecast results of renewable energy (such as solar and wind energy), ensuring that all data are up-to-date and complete before making energy allocation decisions; S22 defines a dynamic energy allocation strategy, which determines the charging power of each charging pile based on the current total grid load, demand forecast of each charging pile, and the availability of renewable energy. This strategy uses the following formula to calculate the charging power allocation of each charging pile: ,in, It is The distribution power of each charging pile, It is The predicted demand for charging piles, is the total number of charging piles, is the total power currently available on the grid, is the predicted renewable energy power; S23: According to the strategy defined in S22, the charging power setting of each charging pile is dynamically adjusted. The specific adjustment is performed by a preset central control system, which can receive and process data in real time and adjust the charging speed of each charging pile as needed; S24: After implementing energy allocation, monitor the deviation between actual power consumption and predicted value, and optimize the allocation strategy according to the deviation result. Specifically, when there is a difference between actual consumption and prediction, the dynamic allocation algorithm will be automatically fine-tuned to reduce future prediction errors and optimize energy use. Through the above dynamic allocation algorithm, it can flexibly respond to changes in power demand and supply, optimize the energy distribution of charging piles, improve energy utilization efficiency, and reduce energy waste. In addition, the system can adapt to various environmental changes through continuous feedback and optimization, and improve the operating efficiency and reliability of the entire charging network.

[0027] S24 specifically includes: S241: First, continuously monitor the deviation between the actual power consumption of each charging pile and the predicted power demand in S1, and specifically calculate the deviation using the following formula: ,in, Indicates The power consumption deviation of each charging pile, is the actual power consumption, and is the predicted power; S242: Analyze the patterns and causes of deviations, including whether there is persistent over- or under-consumption, and identify the need to adjust the forecast model or allocation strategy; S243: When the analysis results show deviations, the parameters of the dynamic allocation algorithm or prediction model will be automatically adjusted. The adjustment strategies include: Adjust weights In the forecasting model, in order to better balance the influence of SVR and ELM models; Adjusting the regularization parameter and In SVR and ELM models, to improve the model fit and generalization ability; S244: Based on the adjusted forecast model, the energy allocation strategy is redefined, specifically by modifying the coefficients in the energy allocation formula to respond to changes in the forecast model. The formula is: ,in, is a newly introduced adjustment factor used to adjust the real-time energy allocation according to historical deviations; S245: Implement the updated allocation strategy and continuously monitor its effect. When the new strategy effectively reduces the prediction error and optimizes energy use, it will continue to be adopted; when the effect is not good, it will continue to be iterated and adjusted. Through the dynamic monitoring and adjustment mechanism of the above steps, it can ensure that the energy allocation strategy is always based on the most accurate prediction data and responds to various changes in a timely manner. This not only improves the efficiency of energy use, but also enhances the system's adaptability to uncertainty, ultimately achieving higher economic benefits and user satisfaction.

[0028] S3 specifically includes: S31: First, determine the peak (such as morning and evening) and non-peak (such as night) electricity demand periods based on historical power grid data analysis. This step is based on the overall load level of the power grid and aims to identify the time periods with the most concentrated and sparse electricity use; S32: Based on the identified peak and off-peak periods, a dynamic electricity price strategy is formulated, including increasing electricity prices during peak periods and reducing electricity prices during off-peak periods. The specific price adjustment formula is: ,in, It's in time The electricity price, is the basic electricity price, is the price adjustment factor, is the start time of the peak period, and the sign function is used to determine whether it is the peak period; S33: Implementing the mid-price strategy of S32 through the user’s charging management system or smart metering device, which will automatically adjust the electricity price and notify the user of current and expected electricity price changes to encourage the user to choose the charging time with the lowest cost; S34: Monitor changes in user charging behavior after the implementation of the dynamic pricing strategy, and analyze the impact of the adjustment of charging time periods on the grid load, including tracking the proportion of charging volume moving from peak hours to off-peak hours, and its contribution to the overall grid load smoothness; S35: Adjust pricing coefficient based on monitoring results Or modify the definition of peak and off-peak time periods to encourage users to charge when the grid load is lower, optimize the electricity demand curve and reduce the grid load; through this time-based dynamic pricing strategy, users are effectively guided to charge during off-peak hours, which helps to smooth the load curve of the entire grid and reduce grid pressure during peak periods. This strategy can not only reduce grid operating costs, but also improve the grid's operating efficiency and reliability, while providing users with the possibility of saving energy costs.

[0029] S4 specifically includes: S41: First, the charging records of each charging pile user are collected regularly, including charging time, charging duration, charging frequency and charging amount data. In addition, relevant data affecting the user's charging behavior is collected, including vehicle model, battery capacity and historical charging behavior pattern; S42: Use data mining and machine learning techniques, such as cluster analysis (K-means), decision trees or neural networks, to analyze the charging behavior patterns of users, and identify groups of users with similar charging behaviors by grouping users, specifically dividing users into nighttime chargers, weekday chargers and weekend chargers; S43: Based on the behavior pattern analysis results, a prediction model is established to predict the charging demand of different user groups to predict the future charging demand pattern; S44: According to the output of the demand forecasting model, the charging strategy and resource allocation of the charging piles are automatically adjusted. Specifically, when the forecast shows that the demand of night chargers will increase, the allocation of charging resources at night will be automatically increased to ensure that the demand is met. In addition, the electricity price is dynamically adjusted according to user preferences and charging needs to encourage users to charge during non-peak hours. Through the above-mentioned refined user charging behavior analysis and demand forecasting, the present invention can effectively predict and meet the charging needs of different user groups and optimize the allocation of charging resources, which not only improves the operating efficiency of the charging station, but also enhances user satisfaction, reduces the overall load of the power grid, and ultimately achieves the optimization of energy use.

[0030] S43 specifically includes: S431: extracting features from the user's charging history records, including charging time, duration, frequency, charging completion, and user type (commercial user, residential user, etc.), and normalizing the extracted feature data to remove the impact of the dimension and ensure the generalization ability of the model; S432: Construct a regression model based on Gaussian process to predict charging demand. Gaussian process is a random process defined on a continuous input space. Any set of finite observation points obeys the multivariate normal distribution. The model formula is expressed as: ; ;in, is the observed charging demand value, which is expressed as the actual measured charging amount of each charging pile in a specific time period; The input feature vector includes relevant features of the user's charging behavior, such as charging start time, charging duration, calendar features (weekdays or weekends), weather conditions, etc. is the noise term, which represents the random errors in measurement and process, and is usually assumed to obey Gaussian distribution, expressing the deviation between model output and actual observation; represents a Gaussian process, which is a random process in which any finite set of observations is assumed to be multivariate normally distributed; is the mean function, which is used to describe the average trend of the data. In many practical applications, it can be simply set to zero to indicate that there is no prior trend; is the covariance function, which defines any two points in the input space and The covariance between them is the core of the model and is used to describe how the input features affect the correlation between the outputs. Commonly used covariance functions include radial basis function (RBF) and Matérn function. S433: Train the model and optimize hyperparameters by maximizing the log-marginal likelihood function. The specific formula is: ,in, Represents the logarithmic marginal likelihood function, which is used for model training and hyperparameter optimization to measure the given input Lower Output The probability of Represents the covariance matrix without noise, whose elements are Calculated to describe the correlation between different data points; is the noise variance, which indicates the level of random noise contained in the data and is a key hyperparameter of the GPR model; represents the covariance matrix with noise added, ,in is the identity matrix, which is used to consider the effect of noise in prediction; For the matrix The determinant of is used to calculate a component in the log-likelihood function, which affects parameter optimization; is the number of samples, indicating the total number of data points involved in training; S434: Use the trained Gaussian process regression model to predict future charging demand based on the latest user behavior data. Gaussian process regression can provide the mean and variance of each prediction point. The variance can be used as a measure of prediction uncertainty to assist in making more cautious strategic decisions. The Gaussian process regression model is particularly suitable for application scenarios such as demand forecasting due to its excellent prediction performance and ability to provide prediction uncertainty. Through this model, charging station operators can more accurately predict charging demand, optimize charging resource allocation, improve service quality and operational efficiency, and reduce energy waste.

[0031] S5 specifically includes: S51: Install power quality monitoring equipment at the charging station to monitor and record the voltage fluctuation and harmonic pollution level during the charging process in real time. The monitoring equipment can continuously collect voltage and current data. Specifically, the voltage fluctuation and total harmonic distortion are calculated using the following formula : ; ,in, It is the scheduled time The voltage fluctuation rate, and are the maximum and minimum voltage values ​​during this time, respectively. is the nominal voltage, is the total harmonic distortion, It is The voltage amplitude of the subharmonics, is the fundamental voltage; S52: determining whether there is a power quality problem by setting a threshold, specifically when the voltage fluctuation exceeds a predetermined threshold (such as 5%) or the THD exceeds a standard limit (such as 5% recommended by IEEE), it is identified as a power quality problem; S53: When a power quality problem is identified, the preset power conditioning technology will be automatically activated to adjust the charging equipment parameters. The specific technology includes dynamically adjusting the charging current or voltage to match the real-time conditions of the power grid, reducing the impact of voltage fluctuations and harmonics, and applying harmonic filters or phase modulators to reduce harmonic content. The specific calculation formula is: ; ,in, and are the adjusted charging current and phase angle, and is the setting value under normal working conditions. and is the adjustment factor; S54: Continue to monitor the power quality after the adjustment, and adjust the strategy according to the response of the equipment and the feedback of the grid status to ensure that the power quality is always maintained in the optimal state, thereby protecting the charging equipment and extending the battery life; by implementing the above steps, the charging facilities can effectively deal with power quality issues, improve the stability and safety of the charging process, extend the service life of batteries and charging equipment, and improve the overall charging efficiency. The intelligent response and automatic adjustment mechanism of this system significantly improves user satisfaction and operational efficiency.

[0032] S6 specifically includes: S61: First, the grid status is monitored in real time, including parameters of voltage, current and frequency, and the grid stability and performance are continuously evaluated using grid monitoring equipment. Once it is detected that the grid parameters deviate from the normal range (such as the voltage drops beyond a predetermined threshold or the frequency is abnormal), it is determined to be a grid fault; S62: A preset backup energy supply plan, specifically including a battery storage system, a diesel generator, and a renewable energy system (such as solar panels and wind power generation). The battery storage system is used to provide immediate energy support when the power grid is disconnected; the diesel generator is used to start in the event of a long-term power grid outage to maintain the operation of the charging station; the renewable energy system (such as solar panels and wind power generation) is used to cooperate with the energy storage system to provide a continuous energy supply for the charging station; S63: When the system detects a grid fault, it will select a predetermined backup energy supply plan based on the nature and expected duration of the fault. For example, a short-term grid fluctuation may only require the activation of the energy storage battery system, while a long power outage requires the start-up of a diesel generator; S64: Regularly test and maintain the backup energy supply system to ensure its reliability and effectiveness in emergency situations, including charging and discharging cycle testing of energy storage batteries, starting and operating testing of diesel generators, and performance inspection of renewable energy equipment. By implementing the above steps, the charging station can respond quickly when a power grid failure occurs and maintain the continuity and safety of the charging service through the preset backup energy supply plan. This not only ensures that the charging needs of electric vehicle users are met, but also improves the reliability of charging facilities and users' trust in charging services.

[0033] The present invention is intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A charging pile group energy management method, characterized in that: The following steps are involved: S1: Based on historical charging data and regional grid load information, a hybrid prediction model of SVR and ELM is used to predict the short-term power demand of each charging pile and the available amount of renewable energy; S2: Based on the prediction results in S1, combined with the current grid load and the predicted renewable energy output, the charging power of each charging pile is adjusted through a dynamic allocation algorithm to optimize energy distribution; S3: During peak demand periods, a preset pricing algorithm encourages users to charge during off-peak hours to smooth the electricity demand curve and reduce grid load; S4: Analyze users’ charging behaviors and preferences to predict users’ charging demand patterns and automatically adjust charging strategies and resource allocation based on the analysis results; S5: Real-time monitoring of power quality during charging, including voltage fluctuations and harmonic pollution, and automatic adjustment of charging equipment parameters using preset power regulation technology to optimize power quality, protect equipment and extend battery life; S6: In the event of an emergency with a power grid failure, the preset backup energy supply plan is automatically activated to ensure the continuity and safety of charging services.

2. A charging pile group energy management method according to claim 1, characterized in that: The S1 specifically includes: S11: Collect historical charging data of each charging pile, including information on charging duration, charging start and end time, and charging power; collect grid load data and historical output data of renewable energy, and clean and pre-process the collected data to remove outliers and fill in missing values ​​to ensure data quality and integrity; S12: Based on the preprocessed data, develop a feature set that includes time characteristics, weather conditions, and historical electricity demand and energy supply data for the same period; S13: Use support vector regression and extreme learning machine to train the model, where the SVR model is used to analyze the nonlinear trends and complex dependencies of electricity demand, and the ELM model is used to quickly learn and adapt to the fluctuations of renewable energy output. The prediction results of the two models are then weighted averaged, with the weights determined based on the historical prediction accuracy, to form a hybrid prediction model; S14: Apply the hybrid prediction model formed in S13 to the real-time data stream to predict the power demand of each charging pile and the available amount of solar energy and wind energy in the next short term.

3. A charging pile group energy management method according to claim 2, characterized in that: The S13 specifically includes: S131: First, the SVR model is used to process and predict power demand data. SVR is an effective nonlinear regression method based on statistical learning theory. Its goal is to find a function , so that the sum of squares of deviations between a given data point and the function is minimized while ensuring the smoothness of the function; S132: Then use ELM to process and predict renewable energy output, specifically the fluctuations of solar and wind energy. ELM is a single-layer feedforward neural network. The weights and biases of its hidden layer nodes are randomly initialized and fixed. The output weights are quickly calculated analytically. The optimization goal of ELM is to minimize the norm and training error of the output layer weights. The specific formula is: ,in, is the output matrix of the hidden layer, is the output layer weight, is the target output, is the regularization coefficient, which is used to control the complexity of the model and avoid overfitting; S133: Finally, the prediction results of the SVR and ELM models are combined to form a hybrid prediction model, which is achieved by applying a weighted average to the prediction results of each model. The weights are calculated based on their respective prediction performance on the validation set. The specific formula is: ,in, is the final prediction result. and are the prediction results of the SVR and ELM models respectively. The weights are determined based on historical forecast accuracy.

4. A charging pile group energy management method according to claim 1, characterized in that: The S2 specifically includes: S21: First, the short-term power demand forecast results of each charging pile obtained in step S1 are integrated with the real-time grid load data and the output forecast results of renewable energy, ensuring that all data are up-to-date and complete before making energy allocation decisions; S22 defines a dynamic energy allocation strategy, which determines the charging power of each charging pile based on the current total grid load, demand forecast of each charging pile, and the availability of renewable energy. This strategy uses the following formula to calculate the charging power allocation of each charging pile: ,in, It is The distribution power of each charging pile, It is The predicted demand for charging piles, is the total number of charging piles, is the total power currently available on the grid, is the predicted renewable energy power; S23: Dynamically adjust the charging power setting of each charging pile according to the strategy defined in S22; S24: After implementing energy allocation, monitor the deviation between actual power consumption and predicted value, and optimize the allocation strategy according to the deviation result. Specifically, when there is a difference between actual consumption and prediction, the dynamic allocation algorithm will be automatically fine-tuned to reduce future prediction errors and optimize energy use.

5. A charging pile group energy management method according to claim 4, characterized in that: The S24 specifically includes: S241: First, continuously monitor the deviation between the actual power consumption of each charging pile and the predicted power demand in S1, and specifically calculate the deviation using the following formula: ,in, Indicates The power consumption deviation of each charging pile, is the actual power consumption, and is the predicted power; S242: Analyze the patterns and causes of deviations, including whether there is persistent over- or under-consumption, and identify the need to adjust the forecast model or allocation strategy; S243: When the analysis results show deviations, the parameters of the dynamic allocation algorithm or prediction model will be automatically adjusted; S244: Based on the adjusted forecast model, the energy allocation strategy is redefined, specifically by modifying the coefficients in the energy allocation formula to respond to changes in the forecast model. The formula is: ,in, is a newly introduced adjustment factor used to adjust the real-time energy allocation according to historical deviations; S245: Implement the updated allocation strategy and continuously monitor its effectiveness. When the new strategy effectively reduces the prediction error and optimizes energy use, it will continue to be adopted; when the effect is not good, it will continue to be iterated and adjusted.

6. A charging pile group energy management method according to claim 1, characterized in that: The S3 specifically includes: S31: First, determine the peak and non-peak power demand periods based on historical power grid data analysis; S32: Based on the identified peak and off-peak periods, a dynamic electricity price strategy is formulated, including increasing electricity prices during peak periods and reducing electricity prices during off-peak periods. The specific price adjustment formula is: ,in, It's in time The electricity price, is the basic electricity price, is the price adjustment factor, is the start time of the peak period, and the sign function is used to determine whether it is the peak period; S33: Implementing the mid-price strategy of S32 through the user’s charging management system or smart metering device, which will automatically adjust the electricity price and notify the user of current and expected electricity price changes to encourage the user to choose the charging time with the lowest cost; S34: Monitor changes in user charging behavior after the implementation of the dynamic pricing strategy, and analyze the impact of the adjustment of charging time periods on the grid load, including tracking the proportion of charging volume moving from peak hours to off-peak hours, and its contribution to the overall grid load smoothness; S35: Adjust pricing coefficient based on monitoring results Or modify the definition of peak and off-peak periods to encourage users to charge when the grid load is lower, optimize the electricity demand curve and reduce the grid load.

7. A charging pile group energy management method according to claim 1, characterized in that: The S4 specifically includes: S41: First, the charging records of each charging pile user are collected regularly, including charging time, charging duration, charging frequency and charging amount data. In addition, relevant data affecting the user's charging behavior is collected, including vehicle model, battery capacity and historical charging behavior pattern; S42: Use data mining and machine learning technology to analyze the charging behavior patterns of users, and identify user groups with similar charging behaviors by grouping users, specifically dividing users into nighttime chargers, weekday chargers, and weekend chargers; S43: Based on the behavior pattern analysis results, a prediction model is established to predict the charging demand of different user groups to predict the future charging demand pattern; S44: Automatically adjust the charging strategy and resource allocation of charging piles based on the output of the demand forecasting model. Specifically, when the forecast shows that the demand for nighttime chargers will increase, the allocation of charging resources at night will be automatically increased to ensure that the demand is met. In addition, the electricity price is dynamically adjusted based on user preferences and charging needs to encourage users to charge during non-peak hours.

8. A charging pile group energy management method according to claim 7, characterized in that: The S43 specifically includes: S431: extracting features from the user's charging history records, including charging time, duration, frequency, charging completion, and user type, and normalizing the extracted feature data to remove the impact of the dimension; S432: Construct a regression model based on Gaussian process to predict charging demand. Gaussian process is a random process defined on a continuous input space. Any set of finite observation points obeys the multivariate normal distribution. The model formula is expressed as: ； ;in, is the observed charging demand value; is the input feature vector; is the noise term, representing the random error in the measurement process; represents a Gaussian process; is the mean function, which is used to describe the average trend of the data; is the covariance function, which defines any two points in the input space and The covariance between S433: Train the model and optimize hyperparameters by maximizing the log-marginal likelihood function. The specific formula is: ,in, represents the log marginal likelihood function; represents the covariance matrix without noise; is the noise variance; represents the covariance matrix with noise added, ,in is the identity matrix; For the matrix The determinant of ; is the number of samples, indicating the total number of data points involved in training; S434: Use the trained Gaussian process regression model to predict future charging needs based on the latest user behavior data.

9. A charging pile group energy management method according to claim 1, characterized in that: The S5 specifically includes: S51: Install power quality monitoring equipment at the charging station to monitor and record the voltage fluctuation and harmonic pollution level during the charging process in real time. The monitoring equipment can continuously collect voltage and current data. Specifically, the voltage fluctuation and total harmonic distortion are calculated using the following formula : ； ,in, It is the scheduled time The voltage fluctuation rate, and are the maximum and minimum voltage values ​​during this time, respectively. is the nominal voltage, is the total harmonic distortion, It is The voltage amplitude of the subharmonics, is the fundamental voltage; S52: determining whether there is a power quality problem by setting a threshold, specifically when the voltage fluctuation exceeds a predetermined threshold or the THD exceeds a standard limit, identifying it as a power quality problem; S53: When a power quality problem is identified, the preset power conditioning technology will be automatically activated to adjust the charging equipment parameters, including dynamically adjusting the charging current or voltage to match the real-time conditions of the grid, reducing the impact of voltage fluctuations and harmonics, and applying harmonic filters or phase modulators to reduce harmonic content; S54: After the adjustment, the power quality continues to be monitored, and the strategy is adjusted according to the response of the equipment and the feedback of the grid status to ensure that the power quality is always maintained in the optimal state, thereby protecting the charging equipment and extending the battery life.

10. A charging pile group energy management method according to claim 1, characterized in that: The S6 specifically includes: S61: First, the grid status is monitored in real time, including parameters of voltage, current and frequency, and grid monitoring equipment is used to continuously evaluate the stability and performance of the grid. Once it is detected that the grid parameters deviate from the normal range, it is determined to be a grid fault; S62: A preset backup energy supply plan, specifically including a battery storage system, a diesel generator and a renewable energy system. The battery storage system is used to provide immediate energy support when the power grid is out of power; the diesel generator is used to start in the event of a long-term power grid outage to maintain the operation of the charging station; the renewable energy system is used to cooperate with the energy storage system to provide a continuous energy supply for the charging station; S63: When the system detects a grid fault, it will select a predetermined backup energy supply solution based on the nature and expected duration of the fault; S64: Regularly test and maintain the backup energy supply system to ensure its reliability and effectiveness in emergency situations, including charging and discharging cycle testing of energy storage batteries, starting and operating testing of diesel generators, and performance inspection of renewable energy equipment.

Images