Control system and method based on charging pile V2G

Abstract

The present invention discloses a control system and method based on charging pile V2G, which relates to the field of charging pile V2G control technology and includes the following steps: first, the sensor equipment obtains the grid operation status parameters in real time, and accurately monitors the actual working conditions of the grid by collecting the operation parameter data in real time, providing basic data support for subsequent analysis. Based on deep learning and intelligent feedback control, the present invention enables electric vehicles to adaptively adjust the output voltage and phase angle, accurately compensate for reactive power, reduce grid voltage fluctuations, and improve overall stability. This solution improves the response speed of EV inverters, achieves millisecond-level grid adaptation, enhances V2G collaborative regulation capabilities, reduces dependence on traditional reactive compensation equipment, and improves the efficiency of new energy consumption. By optimizing the power regulation mechanism, it promotes the deep integration of EV as distributed energy storage, promotes the development of smart energy management, and enhances the economic value and feasibility of V2G in the power market.

Description

Technical Field

[0001] The present invention relates to the field of charging pile V2G control technology, and in particular to a control system and method based on a charging pile V2G. Background Art

[0002] Charging pile-based V2G (Vehicle-to-Grid) control refers to an intelligent control method that uses electric vehicles (EVs) as energy storage units to exchange energy between the grid and vehicles through bidirectional charging and discharging technology. Specifically, EVs charge from the grid during periods of low electricity demand (such as at night). During periods of peak grid load or power shortages, they can discharge energy back to the grid through charging piles to provide power support, thereby achieving peak load shaving and valley filling for the grid, improving the ability to absorb new energy, and enhancing grid stability. V2G control typically involves technologies such as intelligent charging scheduling, load forecasting, dynamic electricity pricing strategies, and communication protocols (such as OCPP and ISO15118) to ensure efficient coordination between charging piles, EVs, and the grid, achieve optimized energy management, and provide economic benefits to vehicle owners (such as peak-valley electricity price arbitrage or compensation for ancillary services).

[0003] Existing technologies have the following shortcomings: When electric vehicles (EVs) discharge power into the grid, the voltage must remain synchronized with the grid to ensure grid stability. Otherwise, voltage fluctuations could impact the grid and affect other users' normal electricity use. EVs dynamically adjust voltage to maintain synchronization through feedback control systems, but voltage regulation alone cannot fully stabilize the grid. When the grid load changes, the phase angle must also be adjusted to change the output reactive power to compensate for the power factor and reduce voltage fluctuations. However, existing technologies are unable to achieve adaptive phase angle adjustment, resulting in EVs potentially being unable to provide sufficient reactive power compensation, causing large fluctuations in grid voltage. Such fluctuations directly impact all equipment connected to the grid, especially voltage-sensitive critical facilities such as hospital life-support equipment, precision industrial production lines, communication base stations, and data center servers. This can cause equipment failures, production interruptions, large-scale communication disruptions, and even endanger personal safety.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present invention is to provide a control system and method for charging pile V2G. Based on deep learning and intelligent feedback control, this system enables electric vehicles (EVs) to adaptively adjust their output voltage and phase angle, accurately compensate for reactive power, reduce grid voltage fluctuations, and improve overall stability. This solution improves the response speed of EV inverters, achieves millisecond-level grid adaptation, enhances V2G collaborative control capabilities, reduces reliance on traditional reactive power compensation equipment, and improves the efficiency of new energy consumption. By optimizing power control mechanisms, promoting the deep integration of EVs as distributed energy storage, advancing the development of smart energy management, and enhancing the economic value and feasibility of V2G in the power market, it provides innovative solutions for grid scheduling, energy trading, and renewable energy utilization, addressing the problems encountered in the aforementioned background technologies.

[0006] In order to achieve the above object, the present invention provides the following technical solution: a control method based on a charging pile V2G, comprising the following steps:

[0007] First, sensor equipment obtains grid operating status parameters in real time. By collecting operating parameter data in real time, it accurately monitors the actual working conditions of the grid and provides basic data support for subsequent analysis.

[0008] After obtaining the real-time operating parameters of the power grid, the acquired data is integrated to form an analysis set. To ensure data quality, the acquired data is pre-processed to improve the accuracy and consistency of the data, laying the foundation for subsequent analysis;

[0009] Extract key features reflecting power grid operation fluctuations from pre-processed data, and analyze and process the extracted key features within the detection window;

[0010] The analyzed key features are input into a pre-learned deep learning model, which is then used to intelligently evaluate power grid operation fluctuations.

[0011] When the model detects severe fluctuations in the power grid, the electric vehicle's bidirectional charging and discharging inverter will activate the feedback control system. Based on the deviation between the real-time grid voltage and the reference voltage, a closed-loop control algorithm is used to dynamically adjust the EV output voltage to keep it consistent with the grid, thereby reducing voltage shocks. At the same time, the phase angle of the EV output power is further adjusted according to the degree of grid fluctuation, thereby changing the reactive power and compensating for the grid power factor.

[0012] Preferably, the specific steps of obtaining the grid operation status parameters in real time through the sensor device are as follows:

[0013] First, smart sensors are deployed at key nodes of the power grid to collect grid operating parameters in real time;

[0014] The sensor then transmits the collected data to a data acquisition server or cloud platform via wireless communication, optical fiber network, or power carrier communication, where the data is stored and processed;

[0015] At the same time, the data will be time-synchronized to ensure the time consistency of all sampling points, thereby accurately reflecting the transient dynamic characteristics of the power grid.

[0016] Preferably, key features reflecting grid operation fluctuations are extracted from the preprocessed data, the extracted features including the degree of mutation of the grid line impedance in a short period of time and the degree of change of the grid negative-sequence voltage component. The degree of mutation of the grid line impedance in a short period of time and the degree of change of the grid negative-sequence voltage component are analyzed under the detection window, and a line impedance mutation reference value and a negative-sequence voltage fluctuation reference value are generated respectively. The stability and power quality status of the grid in a short period of time are quantified by the line impedance mutation reference value and the negative-sequence voltage fluctuation reference value.

[0017] Preferably, the specific steps of analyzing the degree of sudden change of the power grid line impedance in a short period of time within the detection window to generate a line impedance sudden change reference value are as follows:

[0018] Within the monitoring window, the mutation rate of line impedance in a short period of time is calculated to capture the drastic changes in line impedance. The transient change trend of line impedance is quantified by calculating the impedance change rate factor. The calculation expression is as follows:

[0019] , where and Respectively represent the maximum and minimum values of the line impedance within the detection window, and Respectively represent the maximum and minimum complex power of the power grid within the detection window, Is a very small number, to prevent the denominator from approaching zero and causing unstable calculations, is the impedance change rate factor;

[0020] Obtaining the line impedance change rate factor Finally, the dynamic impact factor of impedance is further introduced, and the line impedance mutation reference value is constructed through the high-order nonlinear relationship between grid voltage and impedance to quantitatively represent the fluctuation intensity of line impedance. The calculation expression is as follows:

[0021] , where It is the dynamic impact factor, which measures the direct impact of line impedance mutation on grid voltage fluctuation. and They are the maximum voltage and minimum voltage in the detection window, which are used to characterize the degree of voltage fluctuation. is the equivalent impedance ratio of the line, is the equivalent reactance, is the equivalent resistance, It is the sensitivity coefficient of reactive power to voltage, which indicates the degree of reactive power response caused by voltage changes. is the reactive power, is the voltage, is the regulating factor, It is the reference value of line impedance mutation.

[0022] Preferably, the specific steps of analyzing the degree of change of the negative sequence voltage component of the power grid within the detection window to generate the negative sequence voltage fluctuation reference value are as follows:

[0023] In the three-phase voltage system of the power grid, the symmetrical component method is used to calculate the negative sequence voltage component. The calculation expression is as follows:

[0024] , where 、 、 is the three-phase voltage, is the rotation factor of the three-phase symmetric transformation, is the negative sequence voltage component;

[0025] The rate of change of negative sequence voltage is an important indicator to measure the volatility. The discrete differential form is used to calculate the negative sequence voltage fluctuation rate. The calculation expression is as follows:

[0026] , where is the negative sequence voltage fluctuation rate, and It is at the moment and time The calculated negative sequence voltage component is is the equivalent impedance, is the moment and time The equivalent impedance of the power grid between Is a very small number, to prevent the denominator from approaching zero and causing unstable calculations, is the maximum value operator;

[0027] In order to fully quantify the dynamic instability of negative sequence voltage, it is necessary to combine its fluctuation rate, phase angle change and system impedance characteristics to calculate the reference value of negative sequence voltage fluctuation. The calculation expression is as follows:

[0028] , where is the negative sequence voltage fluctuation reference value, The power grid The phase angle change of the negative sequence voltage of each node, is the total number of nodes, It is the first strip , is a nonlinear adjustment factor, is the total number of lines, It is a small amount that prevents the denominator from being zero.

[0029] Preferably, the analyzed line impedance mutation reference value and negative sequence voltage fluctuation reference value are input into a pre-learned deep learning model, a power grid fluctuation risk coefficient is generated by the deep learning model, and the power grid fluctuation state is intelligently evaluated by the power grid fluctuation risk coefficient.

[0030] Preferably, the grid fluctuation risk coefficient generated when the grid fluctuation state is intelligently evaluated by the pre-learned deep learning model is compared with the pre-set grid fluctuation risk coefficient reference threshold to divide the grid fluctuation. The division steps are as follows:

[0031] If the grid fluctuation risk coefficient is greater than the grid fluctuation risk coefficient reference value threshold, it indicates that severe fluctuations occur during the current grid operation; if the grid fluctuation risk coefficient is less than or equal to the grid fluctuation risk coefficient reference value threshold, it indicates that no fluctuations occur during the current grid operation.

[0032] Preferably, when the model detects that there is a serious fluctuation in the power grid, the specific steps of dynamically adjusting the EV output voltage and further adjusting the phase angle of the EV output power according to the degree of power grid fluctuation are as follows:

[0033] When the power grid fluctuation risk coefficient When the set threshold is exceeded, the real-time voltage of the grid is firstly closed-loop feedback controlled, and the reference voltage is set to And the actual acquisition voltage is , calculate the voltage error, the calculation expression is as follows: , where is the voltage error;

[0034] Using closed-loop control algorithm, the control signal formula is as follows:

[0035] , where is the EV output voltage compensation value, is the proportional gain coefficient, is the integral gain coefficient, is the error integral term, the voltage error The integral over time, that is, the cumulative sum of historical errors, is the integration variable;

[0036] After completing voltage regulation, it is necessary to adjust the phase angle of the EV output power to change the reactive power output. First, the phase angle compensation formula is introduced: , where is the reference threshold of the power grid fluctuation risk coefficient, is the phase angle adjustment gain, is the phase angle increment required to adjust the EV inverter;

[0037] Then, the phase angle is updated. The updated phase angle calculation formula is as follows: , where is the updated EV inverter output phase angle, is the current phase angle of the EV inverter;

[0038] By adjusting the phase angle , the EV inverter dynamically changes the reactive power output to compensate for the power factor of the grid, thereby reducing the system risk caused by voltage fluctuations. The calculation expression is as follows: , where is the reactive power provided by the EV inverter to the grid, is the EV inverter output voltage, is the current output by the EV inverter, It is the key factor that determines the reactive power component;

[0039] Ultimately, the dual regulation of the EV’s output voltage and phase angle is integrated into the output control signal in complex form to achieve coordinated compensation of amplitude and phase. The calculation expression of the integrated control law is as follows: , where is the EV output control signal, is the phase angle compensation factor adjusted by the EV inverter, is the natural base, Is an imaginary unit.

[0040] Preferably, the control system based on the charging pile V2G includes a grid status monitoring module, a data processing and preprocessing module, a feature extraction and analysis module, a deep learning evaluation module, and an EV feedback control and phase angle optimization module:

[0041] The grid status monitoring module uses sensors to obtain grid operating status parameters in real time. By collecting operating parameter data in real time, it accurately monitors the actual working conditions of the grid and provides basic data support for subsequent analysis.

[0042] The data processing and preprocessing module, after obtaining the real-time operating parameters of the power grid, integrates the acquired data to form an analysis set. To ensure data quality, the acquired data is preprocessed to improve the accuracy and consistency of the data, laying the foundation for subsequent analysis;

[0043] The feature extraction and analysis module extracts key features reflecting power grid operation fluctuations from the pre-processed data and analyzes and processes the extracted key features within the detection window;

[0044] The deep learning assessment module inputs the analyzed key features into a pre-learned deep learning model, and uses the deep learning model to perform intelligent assessment of power grid operation fluctuations;

[0045] EV feedback control and phase angle optimization module. When the model detects severe fluctuations in the power grid, the electric vehicle's bidirectional charging and discharging inverter will start the feedback control system. Based on the deviation between the real-time grid voltage and the reference voltage, a closed-loop control algorithm is used to dynamically adjust the EV output voltage to keep it consistent with the grid, thereby reducing voltage shocks. At the same time, the phase angle of the EV output power is further adjusted according to the degree of grid fluctuation, thereby changing the reactive power and compensating the grid power factor.

[0046] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0047] This invention, based on deep learning-driven dynamic grid fluctuation assessment and intelligent feedback control, enables electric vehicles (EVs) to adaptively adjust their output voltage and phase angle, thereby accurately compensating for reactive power, reducing the impact of grid voltage fluctuations on critical loads, and improving the overall stability of the grid. This solution optimizes the response speed of EV inverters, enabling them to quickly adapt to grid load changes within milliseconds. It also enhances the collaborative control capabilities of V2G devices, reduces reliance on traditional reactive compensation equipment, and improves the efficiency of new energy absorption. In addition, the present invention provides an efficient power control mechanism for smart grids, promotes the deep integration of EVs as distributed energy storage resources, promotes the development of smart energy management systems, and enhances the economic value and technical feasibility of V2G in future power markets, providing innovative solutions for grid dispatch optimization, energy trading, and renewable energy utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction to the drawings required for use in the embodiments will be given below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0049] Figure 1 This is a flow chart of the control method of the charging pile V2G based on the present invention.

[0050] Figure 2 This is a module diagram of the control system based on the charging pile V2G of the present invention. DETAILED DESCRIPTION

[0051] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art.

[0052] The present invention provides Figure 1 The control method based on charging pile V2G shown includes the following steps:

[0053] First, sensor equipment obtains grid operating status parameters in real time. By collecting operating parameter data in real time, it accurately monitors the actual working conditions of the grid and provides basic data support for subsequent analysis.

[0054] The specific steps for obtaining real-time grid operating status parameters through sensor devices are as follows: First, smart sensors (such as smart meters, synchronized phasor measurement units (PMUs), and voltage / current sensors) are deployed at key grid nodes (such as substations, distribution lines, and user-side). These sensors collect real-time grid operating parameters, including voltage, current, phase angle, frequency, power factor, reactive power, and load changes. The sensors then transmit the collected data via wireless communications (5G / LoRa), fiber optic networks, or power carrier communications (PLC) to a data acquisition server or cloud platform for storage and preliminary processing. Simultaneously, the data is time-synchronized (e.g., using GPS timing) to ensure time consistency across all sampling points, accurately reflecting the transient dynamic characteristics of the grid. This series of operations not only ensures the real-time and integrity of grid operating information but also provides reliable data support for subsequent data analysis, feature extraction, deep learning model evaluation, and grid control.

[0055] Keep abreast of grid operations and unexpected changes, providing realistic input for subsequent deep learning models and control strategies. PMUs (Phasor Measurement Units) enable microsecond-level data synchronization, while smart meters regularly transmit load information, ensuring data consistency and integrity, laying the foundation for big data analysis.

[0056] After obtaining the real-time operating parameters of the power grid, the acquired data will be integrated to form an analysis set. To ensure data quality, the acquired data will be preprocessed to improve the accuracy and consistency of the data, laying the foundation for subsequent analysis.

[0057] After acquiring real-time grid operating parameters, this data needs to be integrated and preprocessed to ensure accuracy, consistency, and usability. Because sensor devices may be affected by factors such as environmental noise, communication delays, and data loss, directly acquired raw data may contain noise, missing values, duplicate values, or outliers. Therefore, the data must first be denoised to filter out abnormal data caused by sensor errors or sudden interference. Second, time alignment and interpolation must be performed to ensure consistent timestamps across sensor data to prevent missing data from impacting analysis results. Finally, the data must be normalized or standardized to ensure comparability across different units and ranges, preventing significant numerical discrepancies from impacting subsequent modeling. These preprocessing steps transform acquired grid data into a high-quality analysis dataset, providing accurate and reliable data support for subsequent feature extraction, pattern recognition, deep learning analysis, and intelligent control.

[0058] Extract key features reflecting power grid operation fluctuations from pre-processed data, and analyze and process the extracted key features within the detection window;

[0059] Key features reflecting grid operation fluctuations are extracted from the preprocessed data. The extracted features include the degree of mutation of the grid line impedance in a short period of time and the degree of change of the grid negative-sequence voltage component. The degree of mutation of the grid line impedance and the degree of change of the grid negative-sequence voltage component in a short period of time are analyzed within a detection window, and a line impedance mutation reference value and a negative-sequence voltage fluctuation reference value are generated respectively. The line impedance mutation reference value and the negative-sequence voltage fluctuation reference value are used to quantify the stability and power quality status of the grid in a short period of time.

[0060] A sudden increase in grid line impedance over a short period of time may indicate severe grid fluctuations during the process of electric vehicles (EVs) supplying power to the grid. The root cause is that a sharp change in grid impedance typically indicates a sudden change in the grid's load, power supply capacity, or topology. From a grid fluctuation perspective, a rapid increase in line impedance can be caused by factors such as a grid short circuit, transformer switching, sudden disconnection of high-power loads, or remote line faults. These changes can lead to sudden voltage sags, sudden current fluctuations, and even grid frequency fluctuations. For V2G (Vehicle-to-Grid) systems, a sudden increase in grid impedance indicates a decrease in the impedance match between the EV inverter and the grid, potentially reducing the inverter's voltage regulation capability, resulting in an inability to provide stable power output and even causing inverter protection to trip the grid. Furthermore, a sudden change in line impedance can exacerbate reactive power imbalances in the grid, further impacting voltage stability, causing voltage oscillations and an increase in negative sequence components, which in turn affects the power supply quality of the entire region. Therefore, when electric vehicles are supplying power to the grid, if the line impedance increases abnormally in a short period of time, it usually indicates that the grid is experiencing severe dynamic fluctuations, which may affect the stability of the system and the quality of power. Rapid control measures need to be taken to prevent larger-scale grid instability.

[0061] The specific steps for analyzing the sudden change degree of the power grid line impedance in a short period of time within the detection window to generate the line impedance sudden change reference value are as follows:

[0062] Within the monitoring window, the mutation rate of line impedance in a short period of time is calculated to capture the drastic changes in line impedance. The transient change trend of line impedance is quantified by calculating the impedance change rate factor. The calculation expression is as follows:

[0063] , where and They represent the maximum and minimum values of the line impedance within the detection window, respectively, and are used to reflect the fluctuation range of the line impedance. and Respectively represent the maximum and minimum complex power of the power grid within the detection window, Is a very small number, to prevent the denominator from approaching zero and causing unstable calculations, is the impedance change rate factor;

[0064] This step measures the maximum disturbance in line impedance within the detection window and incorporates a power factor to account for dynamic changes in grid load. If power fluctuates significantly, this factor will increase even if the impedance change is minimal, thus more accurately reflecting the impact of impedance changes on grid stability.

[0065] Obtaining the line impedance change rate factor Finally, the dynamic impact factor of impedance is further introduced, and the line impedance mutation reference value is constructed through the high-order nonlinear relationship between grid voltage and impedance to quantitatively represent the fluctuation intensity of line impedance. The calculation expression is as follows:

[0066] , where It is the dynamic impact factor, which measures the direct impact of line impedance mutation on grid voltage fluctuation. and They are the maximum voltage and minimum voltage in the detection window, which are used to characterize the degree of voltage fluctuation. is the equivalent impedance ratio of the line, is the equivalent reactance, is the equivalent resistance. This ratio reflects the inductive or resistive characteristics of the line and affects the degree of influence of impedance mutation on the grid voltage. It is the sensitivity coefficient of reactive power to voltage, which indicates the degree of reactive power response caused by voltage changes. is the reactive power, is the voltage, Is an adjustment factor to prevent the denominator from approaching zero in extreme cases, It is the reference value of line impedance mutation.

[0067] This step combines impedance change, voltage fluctuation, line impedance characteristics and reactive power response capability, and can accurately quantify the impact of sudden changes in line impedance on grid fluctuations. When the value is large, it means that the power grid has experienced more severe fluctuations, and the EV inverter may need to intervene to regulate reactive power or adjust the power output strategy; when A smaller value indicates that the grid is operating relatively stably and does not require additional adjustments. This reference value can be used not only to monitor grid status during V2G operations but also in applications such as renewable energy grid integration and intelligent dispatch, improving the sensitivity and accuracy of grid fluctuation detection.

[0068] The larger the line impedance mutation reference value, generated by analyzing the magnitude of sudden changes in grid line impedance within a detection window, the more severe the grid fluctuations experienced during the electric vehicle (EV) power supply. From a grid stability perspective, rapid changes in line impedance over a short period of time typically indicate internal grid load changes, short-term faults, reactive power fluctuations, or power supply topology adjustments, all of which can affect grid voltage stability. An increase in the line impedance mutation reference value indicates a dramatic change in grid impedance, potentially causing power output fluctuations in the EV inverter, which in turn affects overall grid power quality. For example, if the grid load suddenly decreases and the line impedance increases, the EV's power output may face the risk of voltage increase or even inverter disconnection. Conversely, if the load suddenly increases or a short-circuit fault occurs, the line impedance sharply decreases, and the EV may require additional reactive power compensation to maintain grid stability. Therefore, a higher line impedance mutation reference value indicates more severe grid fluctuations and poorer voltage stability, requiring a more sophisticated regulation strategy to prevent grid instability during EV power supply.

[0069] A sudden increase in the negative-sequence voltage component of the power grid may indicate severe fluctuations during the process of electric vehicles (EVs) supplying power to the grid. However, this assessment requires consideration of other indicators. The negative-sequence voltage component primarily reflects the degree of three-phase imbalance in the power grid. A sudden increase in this component may indicate load imbalance, inverter output asymmetry, nonlinear load input, or a localized grid fault. A sudden increase in the negative-sequence voltage component during EV power supply may indicate a three-phase mismatch in the EV inverter's output voltage, leading to power imbalance in the grid and exacerbating voltage fluctuations. Furthermore, if the grid is already heavily loaded or has a high background negative-sequence voltage, EV power supply may further exacerbate voltage imbalances, causing more severe fluctuations and impacting the stable operation of other grid-connected equipment. For example, motors and power conversion equipment require high three-phase balance. Excessive negative-sequence voltage can cause these devices to malfunction, experience increased vibration, or even damage. Therefore, when a sudden rise in the negative sequence voltage component is detected, it is necessary to further confirm the severity of the grid fluctuation by combining parameters such as line impedance change, voltage transient rate, and phase angle fluctuation, and optimize the control strategy of the EV inverter to ensure the stability and power quality of the grid-connected power supply.

[0070] The specific steps for analyzing the change degree of the negative sequence voltage component of the power grid within the detection window to generate the negative sequence voltage fluctuation reference value are as follows:

[0071] In the three-phase voltage system of the power grid, the negative sequence voltage component is used to characterize the three-phase imbalance. The symmetrical component method is used to calculate the negative sequence voltage component. The calculation expression is as follows:

[0072] , where 、 、 is the three-phase voltage, is the rotation factor of the three-phase symmetric transformation, is the negative sequence voltage component;

[0073] This step converts the three-phase voltage into positive sequence, negative sequence and zero sequence components, where the negative sequence voltage component It mainly reflects the asymmetry of the three-phase voltage. The larger the value, the more serious the three-phase imbalance of the power grid, which may lead to power grid fluctuations when electric vehicles are connected to the grid.

[0074] The rate of change of negative sequence voltage is an important indicator to measure the volatility. The discrete differential form is used to calculate the negative sequence voltage fluctuation rate. The calculation expression is as follows:

[0075] , where is the negative sequence voltage fluctuation rate, and It is at the moment and time The calculated negative sequence voltage component is is the equivalent impedance, is the moment and time The equivalent impedance of the power grid between It is a very small number, which prevents the denominator from approaching zero, causing unstable calculations, and avoids singular situations in mathematical calculations. is the maximum operator, which selects the maximum value among the volatility calculated between all moments;

[0076] This step quantifies the degree of dynamic fluctuation of the power grid by measuring the maximum ratio of negative sequence voltage change. If the standard deviation is too large, it indicates that the negative sequence voltage fluctuates violently during the EV power supply process, which may cause a decrease in grid power quality and malfunction of relay protection. Compared with traditional standard deviation calculation, this method incorporates the influence of equivalent impedance and can more accurately reflect the fluctuation transmission characteristics within the grid.

[0077] In order to fully quantify the dynamic instability of negative sequence voltage, it is necessary to combine its fluctuation rate, phase angle change and system impedance characteristics to calculate the reference value of negative sequence voltage fluctuation. The calculation expression is as follows:

[0078] , where is the negative sequence voltage fluctuation reference value, The power grid The phase angle change of the negative sequence voltage of each node, is the total number of nodes, It is the first strip , It is a nonlinear adjustment factor used to amplify the effect of high impedance path on voltage fluctuation. is the total number of lines, It is to avoid small quantities with denominators equal to zero to prevent unstable calculations.

[0079] This formula comprehensively considers the negative sequence voltage fluctuation rate ( ), phase angle change ( ) and grid impedance distribution ( ), thus providing a comprehensive quantitative indicator of negative sequence voltage fluctuation. When the value is high, it indicates that during the current process of electric vehicles supplying power to the grid, the negative sequence voltage of the grid fluctuates severely and the three-phase imbalance is significant, which may affect the power factor of the grid and even cause relay protection action or power equipment failure.

[0080] The negative-sequence voltage fluctuation reference value, generated by analyzing the variation in the grid's negative-sequence voltage component within the detection window, indicates greater grid fluctuations during the current electric vehicle (EV) supply to the grid. The negative-sequence voltage component reflects the degree of imbalance in the grid's three-phase voltage. Instability in the grid's negative-sequence voltage is often caused by factors such as asymmetric loads, uneven access to distributed power sources, unbalanced reactive power distribution, or short-circuit faults. A higher negative-sequence voltage fluctuation reference value indicates increased asymmetry in the grid's three-phase voltage, potentially causing output power fluctuations in EV inverters and impacting grid stability. For example, a sudden increase in the grid's negative-sequence voltage could result in increased reactive power demand from EV inverters, potentially triggering inverter protection mechanisms and causing grid disconnection. Furthermore, high negative-sequence voltage can affect motor equipment in the grid, reducing operating efficiency and increasing losses. Therefore, a higher negative-sequence voltage fluctuation reference value indicates greater grid fluctuations and poorer grid stability for EV power supply, necessitating measures such as reactive power compensation and phase angle adjustment to improve power quality. On the contrary, when the reference value is low or remains stable, it indicates that the power grid operates smoothly during the EV power supply process, the three-phase voltage is well symmetrical, and no significant fluctuations are caused.

[0081] The analyzed key features are input into a pre-learned deep learning model, which is then used to intelligently evaluate power grid operation fluctuations.

[0082] The analyzed line impedance mutation reference value and negative sequence voltage fluctuation reference value are input into the pre-learned deep learning model, and the grid fluctuation risk coefficient is generated by the deep learning model. The grid fluctuation state is intelligently evaluated by the grid fluctuation risk coefficient.

[0083] A pre-learned deep learning model refers to a neural network model trained with historical power grid data and possessing a certain degree of generalization capability. This model can automatically identify power grid operating modes and assess power fluctuation risks based on input features (such as reference values for line impedance mutations and negative-sequence voltage fluctuations). This model typically employs supervised or semi-supervised learning methods, utilizing extensive historical power grid data (including normal operation data, fault data, and fluctuation event data). This allows it to learn the complex nonlinear relationships between characteristics and fluctuation risks under different power grid conditions. To improve the model's adaptability, training employs techniques such as data augmentation, time series modeling, and feature selection, ensuring that the model not only accurately identifies common patterns of power grid fluctuations but also accurately assesses sudden abnormalities. Compared to traditional rule-based methods, deep learning models can extract key features from high-dimensional data and automatically adapt to different power grid operating conditions, significantly improving the accuracy of power grid fluctuation prediction and assessment. For example, LSTM (Long Short-Term Memory) or Transformer timing models can be used to learn the temporal dependencies of grid parameters and extract long-term trends in grid fluctuations, while CNN (Convolutional Neural Network) or GNN (Graph Neural Network) can be used to analyze local disturbance patterns in the grid topology and improve anomaly detection capabilities.

[0084] In practical applications, a pre-trained deep learning model receives real-time grid status characteristics, such as line impedance mutation reference values and negative-sequence voltage fluctuation reference values. This input data is then processed through a neural network layer to generate a grid fluctuation risk factor. This risk factor is a quantitative indicator, typically expressed as a 0-1 or 0-100 scale, that measures the current level of grid fluctuation. For example, a risk factor close to 0 indicates a stable grid with no significant fluctuations; a value close to 100 indicates a high-risk grid state, potentially experiencing a short-term voltage collapse, reactive power imbalance, or equipment failure. This model not only provides real-time fluctuation assessment but also predicts fluctuation trends over short periods of time, providing early warning of potential grid instability events, enabling the dispatch center or V2G devices to take timely compensatory measures. By introducing a deep learning assessment mechanism, the grid can achieve an intelligent transition from passive response (intervention only after a fault occurs) to proactive prediction (pre-emptive perception of fluctuations and adjustment), thereby improving overall power supply security and system stability.

[0085] The deep learning model is not limited here and can achieve the line impedance sudden change reference value and negative sequence voltage fluctuation reference value Conduct comprehensive analysis to generate power grid fluctuation risk coefficient In order to realize the technical solution of the present invention, the present invention provides a specific implementation method;

[0086] Grid fluctuation risk coefficient The generation formula is as follows:

[0087] , where 、 They are line impedance mutation reference values and negative sequence voltage fluctuation reference value The preset scaling factor of 、 Both are greater than 0.

[0088] Preset scale factor and Refers to the reference value used to balance the line impedance mutation and negative sequence voltage fluctuation reference value Calculating the risk factor of power grid fluctuation Since the numerical range, change trend and impact degree of different features on grid stability may vary significantly, direct linear combination may cause a feature to have too much or too little impact on the final result. Therefore, the introduction of and As an adjustment parameter, it is used to adjust the line impedance sudden change reference value and negative sequence voltage fluctuation reference value Grid fluctuation risk factor The contribution ratio of the calculation is calculated to ensure that the calculation results can reflect the combined impact of the two and avoid a certain indicator dominating the entire risk assessment result. The values of these coefficients can usually be adjusted through statistical analysis, expert experience setting or deep learning optimization to ensure that the power grid fluctuation risk coefficient It can accurately quantify the risk of power grid fluctuations and is applicable to different power grid operating environments.

[0089] It can be seen from the grid fluctuation risk coefficient that the larger the line impedance mutation reference value generated after analyzing the mutation degree of the grid line impedance in a short period of time within the detection window, and the larger the negative-sequence voltage fluctuation reference value generated after analyzing the change degree of the grid negative-sequence voltage component within the detection window, the larger the grid fluctuation risk coefficient generated when the grid fluctuation state is intelligently evaluated by the pre-learned deep learning model, indicating that the probability of severe fluctuations in the grid is greater, and vice versa, the probability of severe fluctuations in the grid is smaller.

[0090] The grid fluctuation risk coefficient generated by the pre-learned deep learning model during the intelligent assessment of the grid fluctuation state is compared with the pre-set grid fluctuation risk coefficient reference threshold to classify the grid fluctuation. The classification steps are as follows:

[0091] If the grid fluctuation risk coefficient is greater than the grid fluctuation risk coefficient reference value threshold, it indicates that severe fluctuations occur during the current grid operation; if the grid fluctuation risk coefficient is less than or equal to the grid fluctuation risk coefficient reference value threshold, it indicates that no fluctuations occur during the current grid operation.

[0092] When the model detects severe fluctuations in the power grid, the electric vehicle's bidirectional charging and discharging inverter activates a feedback control system. Based on the deviation between the real-time grid voltage and the reference voltage, a closed-loop control algorithm (such as PI control or adaptive control) is used to dynamically adjust the EV output voltage to keep it consistent with the grid, thereby reducing voltage shocks. At the same time, the phase angle of the EV output power is further adjusted according to the degree of grid fluctuation, thereby changing the reactive power and compensating for the grid power factor.

[0093] When the model detects severe grid fluctuations, it dynamically adjusts the EV output voltage and further adjusts the phase angle of the EV output power based on the degree of grid fluctuation. The specific steps are as follows:

[0094] When the power grid fluctuation risk coefficient When the set threshold is exceeded, the real-time voltage of the grid is firstly closed-loop feedback controlled, and the reference voltage is set to And the actual acquisition voltage is , calculate the voltage error, the calculation expression is as follows: , where is the voltage error;

[0095] Using a closed-loop control algorithm (such as adaptive PI control), the control signal formula is as follows:

[0096] , where is the EV output voltage compensation value, which indicates the voltage increment that the electric vehicle bidirectional inverter needs to adjust to make the EV output voltage consistent with the grid target voltage. Is the proportional gain coefficient, which represents the proportional gain in the PI controller and determines the influence of the voltage error on the output voltage adjustment. Is the integral gain coefficient, which represents the integral gain in the PI controller and determines the degree of influence of the long-term voltage error on the output voltage adjustment. is the error integral term, the voltage error The integral over time, that is, the cumulative sum of historical errors, is the integration variable;

[0097] Through this step, the EV's bidirectional charge and discharge inverter adjusts the output voltage according to the voltage deviation, so that it can quickly Closer together, thereby reducing the impact of voltage shocks on the power grid.

[0098] After completing voltage regulation, it is necessary to adjust the phase angle of the EV output power to change the reactive power output. First, the phase angle compensation formula is introduced: , where is the reference threshold of the power grid fluctuation risk coefficient, is the phase angle adjustment gain, It is the phase angle increment that the EV inverter needs to adjust, which determines the change in reactive power;

[0099] Then, the phase angle is updated. The updated phase angle calculation formula is as follows: , where is the updated EV inverter output phase angle, is the current phase angle of the EV inverter;

[0100] By adjusting the phase angle , the EV inverter dynamically changes the reactive power output to compensate for the power factor of the grid, thereby reducing the system risk caused by voltage fluctuations. The calculation expression is as follows: , where is the reactive power provided by the EV inverter to the grid, is the EV inverter output voltage, is the current output by the EV inverter, It is the key factor that determines the reactive power component. increases (i.e. the phase angle increases), becomes larger, EV provides more reactive power to the grid to make up for the reactive power gap of the grid and improve voltage stability; when Reduce, EV inverter reduces reactive power output to avoid excess reactive power in the grid;

[0101] Ultimately, the dual regulation of the EV’s output voltage and phase angle is integrated into the output control signal in complex form to achieve coordinated compensation of amplitude and phase. The calculation expression of the integrated control law is as follows: , where is the EV output control signal, is the phase angle compensation factor adjusted by the EV inverter, is the natural base, Is an imaginary unit.

[0102] This step means that the EV inverter will adjust the voltage change according to the closed loop feedback and phase angle compensation Synchronously adjusting the amplitude and phase of its output power ensures that the output voltage remains synchronized with the grid, effectively compensating for reactive power shortfalls and mitigating the risk of grid fluctuations. This coordinated control strategy, by simultaneously adjusting amplitude and phase, forms an advanced, dynamic, and complex control mechanism, ensuring stable grid operation even in the face of severe fluctuations.

[0103] This invention, based on deep learning-driven dynamic grid fluctuation assessment and intelligent feedback control, enables electric vehicles (EVs) to adaptively adjust their output voltage and phase angle, thereby accurately compensating for reactive power, reducing the impact of grid voltage fluctuations on critical loads, and improving the overall stability of the grid. This solution optimizes the response speed of EV inverters, enabling them to quickly adapt to grid load changes within milliseconds. It also enhances the collaborative control capabilities of V2G devices, reduces reliance on traditional reactive compensation equipment, and improves the efficiency of new energy absorption. In addition, the present invention provides an efficient power control mechanism for smart grids, promotes the deep integration of EVs as distributed energy storage resources, promotes the development of smart energy management systems, and enhances the economic value and technical feasibility of V2G in future power markets, providing innovative solutions for grid dispatch optimization, energy trading, and renewable energy utilization.

[0104] The present invention provides Figure 2 The charging pile V2G-based control system shown in the figure includes a grid status monitoring module, a data processing and preprocessing module, a feature extraction and analysis module, a deep learning evaluation module, and an EV feedback control and phase angle optimization module:

[0105] The grid status monitoring module uses sensors to obtain grid operating status parameters in real time. By collecting operating parameter data in real time, it accurately monitors the actual working conditions of the grid and provides basic data support for subsequent analysis.

[0106] The data processing and preprocessing module, after obtaining the real-time operating parameters of the power grid, integrates the acquired data to form an analysis set. To ensure data quality, the acquired data is preprocessed to improve the accuracy and consistency of the data, laying the foundation for subsequent analysis;

[0107] The feature extraction and analysis module extracts key features reflecting power grid operation fluctuations from the pre-processed data and analyzes and processes the extracted key features within the detection window;

[0108] The deep learning assessment module inputs the analyzed key features into a pre-learned deep learning model, and uses the deep learning model to perform intelligent assessment of power grid operation fluctuations;

[0109] EV feedback control and phase angle optimization module. When the model detects severe fluctuations in the power grid, the electric vehicle's bidirectional charging and discharging inverter will start the feedback control system. Based on the deviation between the real-time grid voltage and the reference voltage, a closed-loop control algorithm is used to dynamically adjust the EV output voltage to keep it consistent with the grid, thereby reducing voltage shocks. At the same time, the phase angle of the EV output power is further adjusted according to the degree of grid fluctuation, thereby changing the reactive power and compensating the grid power factor.

[0110] The control method based on charging pile V2G provided in an embodiment of the present invention is implemented by the above-mentioned control system based on charging pile V2G. The specific method and process of the control system based on charging pile V2G are detailed in the embodiment of the control method based on charging pile V2G, which will not be repeated here.

[0111] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0112] The above description is merely illustrative of certain exemplary embodiments of the present invention. It goes without saying that those skilled in the art will be able to modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims.

[0113] It should be noted that, in this document, if there are relational terms such as first and second, etc., they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device that includes the element.

[0114] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0115] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0116] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0117] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0118] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0119] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0120] The above description is merely illustrative of certain exemplary embodiments of the present invention. It goes without saying that those skilled in the art will be able to modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims.

Claims

1. A control method based on charging pile V2G, characterized in that: The following steps are involved: First, sensor equipment obtains grid operating status parameters in real time. By collecting operating parameter data in real time, it accurately monitors the actual working conditions of the grid and provides basic data support for subsequent analysis. After obtaining the real-time operating parameters of the power grid, the acquired data is integrated to form an analysis set. To ensure data quality, the acquired data is pre-processed to improve the accuracy and consistency of the data, laying the foundation for subsequent analysis; Extract key features reflecting power grid operation fluctuations from pre-processed data, and analyze and process the extracted key features within the detection window; The analyzed key features are input into a pre-learned deep learning model, which is then used to intelligently evaluate power grid operation fluctuations. When the model detects severe fluctuations in the power grid, the electric vehicle's bidirectional charging and discharging inverter activates a feedback control system. Based on the deviation between the real-time grid voltage and the reference voltage, a closed-loop control algorithm is used to dynamically adjust the EV output voltage to keep it consistent with the grid, thereby reducing voltage shocks. At the same time, the phase angle of the EV output power is further adjusted according to the degree of grid fluctuation, thereby changing the reactive power and compensating the grid power factor. Extract key features reflecting grid operation fluctuations from the preprocessed data. The extracted features include the degree of sudden change in grid line impedance over a short period of time and the degree of change in the grid negative-sequence voltage component. These are analyzed within a detection window to generate line impedance sudden change reference values and negative-sequence voltage fluctuation reference values, respectively. These values are used to quantify the grid's stability and power quality over a short period of time. The specific steps for analyzing the sudden change degree of the power grid line impedance in a short period of time within the detection window to generate the line impedance sudden change reference value are as follows: Within the monitoring window, the mutation rate of line impedance in a short period of time is calculated to capture the drastic changes in line impedance. The transient change trend of line impedance is quantified by calculating the impedance change rate factor. The calculation expression is as follows: , where and Respectively represent the maximum and minimum values of the line impedance within the detection window, and Respectively represent the maximum and minimum complex power of the power grid within the detection window, Is a very small number, to prevent the denominator from approaching zero and causing unstable calculations, is the impedance change rate factor; Obtaining the line impedance change rate factor Finally, the dynamic impact factor of impedance is further introduced, and the line impedance mutation reference value is constructed through the high-order nonlinear relationship between grid voltage and impedance to quantitatively represent the fluctuation intensity of line impedance. The calculation expression is as follows: , where It is the dynamic impact factor, which measures the direct impact of line impedance mutation on grid voltage fluctuation. and They are the maximum voltage and minimum voltage in the detection window, which are used to characterize the degree of voltage fluctuation. is the equivalent impedance ratio of the line, is the equivalent reactance, is the equivalent resistance, It is the sensitivity coefficient of reactive power to voltage, which indicates the degree of reactive power response caused by voltage changes. is the reactive power, is the voltage, is the regulating factor, It is the reference value of line impedance mutation.

2. The control method based on charging pile V2G according to claim 1, characterized in that: The specific steps for obtaining grid operation status parameters in real time through sensor equipment are as follows: First, smart sensors are deployed at key nodes of the power grid to collect grid operating parameters in real time; The sensor then transmits the collected data to a data acquisition server or cloud platform via wireless communication, optical fiber network, or power carrier communication, where the data is stored and processed; At the same time, the data will be time-synchronized to ensure the time consistency of all sampling points, thereby accurately reflecting the transient dynamic characteristics of the power grid.

3. The control method based on charging pile V2G according to claim 1, characterized in that: The specific steps for analyzing the change degree of the negative sequence voltage component of the power grid within the detection window to generate the negative sequence voltage fluctuation reference value are as follows: In the three-phase voltage system of the power grid, the symmetrical component method is used to calculate the negative sequence voltage component. The calculation expression is as follows: , where 、 、 is the three-phase voltage, is the rotation factor of the three-phase symmetric transformation, is the negative sequence voltage component; The rate of change of negative sequence voltage is an important indicator to measure volatility. The discrete differential form is used to calculate the negative sequence voltage fluctuation rate. The calculation expression is as follows: , where is the negative sequence voltage fluctuation rate, and It is at the moment and time The calculated negative sequence voltage component is is the equivalent impedance, is the moment and time The equivalent impedance of the power grid between Is a very small number, to prevent the denominator from approaching zero and causing unstable calculations, is the maximum value operator; In order to fully quantify the dynamic instability of negative sequence voltage, it is necessary to combine its fluctuation rate, phase angle change and system impedance characteristics to calculate the reference value of negative sequence voltage fluctuation. The calculation expression is as follows: , where is the negative sequence voltage fluctuation reference value, The power grid The phase angle change of the negative sequence voltage of each node, is the total number of nodes, It is the first strip , is a nonlinear adjustment factor, is the total number of lines, It is a small amount that prevents the denominator from being zero.

4. The control method based on charging pile V2G according to claim 1, characterized in that: The analyzed line impedance mutation reference value and negative sequence voltage fluctuation reference value are input into the pre-learned deep learning model, and the grid fluctuation risk coefficient is generated by the deep learning model. The grid fluctuation state is intelligently evaluated by the grid fluctuation risk coefficient.

5. The control method based on charging pile V2G according to claim 4, characterized in that: The grid fluctuation risk coefficient generated by the pre-learned deep learning model during the intelligent assessment of the grid fluctuation state is compared with the pre-set grid fluctuation risk coefficient reference threshold to classify the grid fluctuation. The classification steps are as follows: If the grid fluctuation risk coefficient is greater than the grid fluctuation risk coefficient reference value threshold, it indicates that severe fluctuations occur during the current grid operation; if the grid fluctuation risk coefficient is less than or equal to the grid fluctuation risk coefficient reference value threshold, it indicates that no fluctuations occur during the current grid operation.

6. The control method based on charging pile V2G according to claim 5, characterized in that: When the model detects severe grid fluctuations, it dynamically adjusts the EV output voltage and further adjusts the phase angle of the EV output power based on the degree of grid fluctuation. The specific steps are as follows: When the power grid fluctuation risk coefficient When the set threshold is exceeded, the real-time voltage of the grid is firstly closed-loop feedback controlled, and the reference voltage is set to And the actual acquisition voltage is , calculate the voltage error, the calculation expression is as follows: , where is the voltage error; Using closed-loop control algorithm, the control signal formula is as follows: , where is the EV output voltage compensation value, is the proportional gain coefficient, is the integral gain coefficient, is the error integral term, the voltage error The integral over time, that is, the cumulative sum of historical errors, is the integration variable; After completing voltage regulation, it is necessary to adjust the phase angle of the EV output power to change the reactive power output. First, the phase angle compensation formula is introduced: , where is the reference threshold of the power grid fluctuation risk coefficient, is the phase angle adjustment gain, is the phase angle increment required to adjust the EV inverter; Then, the phase angle is updated. The updated phase angle calculation formula is as follows: , where is the updated EV inverter output phase angle, is the current phase angle of the EV inverter; By adjusting the phase angle , the EV inverter dynamically changes the reactive power output to compensate for the power factor of the grid, thereby reducing the system risk caused by voltage fluctuations. The calculation expression is as follows: , where is the reactive power provided by the EV inverter to the grid, is the EV inverter output voltage, is the current output by the EV inverter, It is the key factor that determines the reactive power component; Ultimately, the dual regulation of the EV’s output voltage and phase angle is integrated into the output control signal in complex form to achieve coordinated compensation of amplitude and phase. The calculation expression of the integrated control law is as follows: , where is the EV output control signal, is the phase angle compensation factor adjusted by the EV inverter, is the natural base, Is an imaginary unit.

7. A control system based on a charging pile V2G, used to implement the control method based on a charging pile V2G according to any one of claims 1 to 6, characterized in that: It includes a grid status monitoring module, a data processing and preprocessing module, a feature extraction and analysis module, a deep learning evaluation module, and an EV feedback control and phase angle optimization module: The grid status monitoring module uses sensors to obtain grid operating status parameters in real time. By collecting operating parameter data in real time, it accurately monitors the actual working conditions of the grid and provides basic data support for subsequent analysis. The data processing and preprocessing module, after obtaining the real-time operating parameters of the power grid, integrates the acquired data to form an analysis set. To ensure data quality, the acquired data is preprocessed to improve the accuracy and consistency of the data, laying the foundation for subsequent analysis; The feature extraction and analysis module extracts key features reflecting power grid operation fluctuations from the pre-processed data and analyzes and processes the extracted key features within the detection window; The deep learning assessment module inputs the analyzed key features into a pre-learned deep learning model, and uses the deep learning model to perform intelligent assessment of power grid operation fluctuations; EV feedback control and phase angle optimization module. When the model detects severe fluctuations in the power grid, the electric vehicle's bidirectional charging and discharging inverter will start the feedback control system. Based on the deviation between the real-time grid voltage and the reference voltage, a closed-loop control algorithm is used to dynamically adjust the EV output voltage to keep it consistent with the grid, thereby reducing voltage shocks. At the same time, the phase angle of the EV output power is further adjusted according to the degree of grid fluctuation, thereby changing the reactive power and compensating the grid power factor.

Images