Control method and system of intelligent photovoltaic charging pile

Abstract

The invention relates to the technical field of information, in particular to a control method and system for an intelligent photovoltaic charging pile in the technical field of automobile charging piles, and the method comprises the steps: obtaining the charging demands and residual electric quantity information of a plurality of electric automobiles, dynamically adjusting the charging power of each automobile through combining with an electric power distribution algorithm, and guaranteeing the reasonable distribution of electric power resources; acquiring real-time load data through a power grid load monitoring system, predicting a future power grid load change trend in combination with a historical load curve, and judging whether a charging strategy needs to be adjusted or not; according to historical data and real-time data, cooperative working parameters of photovoltaic power generation and storage battery power storage are optimized by adopting a machine learning algorithm, and the overall efficiency of the system is improved; a cooperative working mode of photovoltaic power generation, storage battery power storage and power grid power supply is dynamically adjusted through an intelligent optimization algorithm; if the illumination fluctuation is large, a standby power supply is started to supplement power supply, it is ensured that the charging process is not affected, meanwhile, the energy storage efficiency is optimized, and power resource waste is reduced.

Description

Technical Field

[0001] The present invention relates to the field of information technology, specifically to the technical field of automobile charging piles, and more particularly to a control method and system for an intelligent photovoltaic charging pile. Background Art

[0002] In existing technologies, the coordinated operation of photovoltaic power generation and battery storage primarily relies on preset fixed parameters and manual adjustments, lacking the ability to dynamically respond to real-time environmental changes. Existing systems are unable to accurately predict photovoltaic power generation when lighting conditions fluctuate significantly, resulting in overly conservative or aggressive charging strategies that fail to maximize photovoltaic energy utilization. For example, on cloudy or overcast days, photovoltaic power generation significantly decreases, yet the system may still overly rely on photovoltaic power generation, leading to low charging efficiency. Furthermore, when handling the simultaneous charging of multiple electric vehicles, existing technologies are unable to dynamically allocate power resources based on each vehicle's charging needs and remaining power, which can easily lead to some vehicles charging too slowly or wasting power. While existing systems can supplement grid power with batteries and photovoltaic power generation when grid power is insufficient, they lack real-time monitoring and forecasting of grid load, making it impossible to intelligently adjust charging strategies during peak periods, potentially further increasing grid load. Existing technologies lack intelligent algorithm support for the coordinated operation of photovoltaic power generation, battery storage, and grid power supply, making it impossible to optimize based on historical and real-time data. This results in overall low system efficiency and fails to meet the demands of a more efficient lifestyle. Summary of the Invention

[0003] The present invention provides a control method for an intelligent photovoltaic charging pile, comprising the following steps:

[0004] S101, acquiring real-time environmental data of light intensity, temperature, and humidity through a sensor network, establishing a photovoltaic power generation prediction model based on the physical characteristics of the photovoltaic modules, and outputting a power generation prediction value;

[0005] S102. Determine whether the charging strategy needs to be adjusted based on the predicted power generation value and the current capacity of the battery. If the predicted value is lower than a preset threshold, reduce the charging power to avoid over-reliance on photovoltaic power generation.

[0006] S103, obtaining charging demand and remaining power information of multiple electric vehicles, and dynamically adjusting the charging power of each vehicle in combination with a power distribution algorithm to ensure reasonable allocation of power resources;

[0007] S104. Obtain real-time load data through the power grid load monitoring system, combine it with historical load curves, predict future power grid load change trends, and determine whether the charging strategy needs to be adjusted;

[0008] S106. If the grid load forecast value exceeds a preset threshold, the intelligent optimization algorithm is activated to adjust the collaborative working mode of photovoltaic power generation, battery storage, and grid power supply to reduce the grid load;

[0009] S107. Based on historical data and real-time data, use machine learning algorithms to optimize the collaborative working parameters of photovoltaic power generation and battery storage to improve the overall efficiency of the system;

[0010] S108, through intelligent optimization algorithms, dynamically adjust the collaborative working mode of photovoltaic power generation, battery storage and grid power supply to ensure efficient operation of the system under different environmental conditions;

[0011] S109. If the light fluctuation is large, the backup power supply is activated to supplement the power supply to ensure that the charging process is not affected, while optimizing the energy storage efficiency and reducing the waste of electricity resources.

[0012] The present invention provides an intelligent photovoltaic charging system and an optimization system thereof, which mainly include:

[0013] The environmental data acquisition module is used to obtain real-time environmental data such as light intensity, temperature, and humidity through a sensor network. It then combines the physical characteristics of photovoltaic modules to establish a photovoltaic power generation prediction model and output the power generation prediction value.

[0014] The power generation prediction module is used to determine whether the charging strategy needs to be adjusted based on the power generation forecast value and the current capacity of the battery. If the forecast value is lower than the preset threshold, the charging power is reduced to avoid over-reliance on photovoltaic power generation;

[0015] The charging strategy adjustment module is used to obtain the charging demand and remaining power information of multiple electric vehicles, and dynamically adjust the charging power of each vehicle in combination with the power distribution algorithm to ensure the rational allocation of power resources;

[0016] The electric vehicle charging management module is used to obtain real-time load data from the power grid load monitoring system, combine it with historical load curves, predict future power grid load trends, and determine whether the charging strategy needs to be adjusted;

[0017] The grid load monitoring module is used to activate the intelligent optimization algorithm if the grid load forecast value exceeds the preset threshold, adjust the collaborative working mode of photovoltaic power generation, battery storage and grid power supply, and reduce the grid load;

[0018] An intelligent optimization module, which uses machine learning algorithms to optimize the collaborative working parameters of photovoltaic power generation and battery storage based on historical and real-time data to improve the overall efficiency of the system;

[0019] The collaborative work optimization module is used to dynamically adjust the collaborative work mode of photovoltaic power generation, battery storage and grid power supply through intelligent optimization algorithms to ensure efficient operation of the system under different environmental conditions;

[0020] The backup power management module is used to activate the backup power supply to supplement the power supply if the light fluctuation is large, ensuring that the charging process is not affected, while optimizing energy storage efficiency and reducing power resource waste.

[0021] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0022] The present invention discloses a control method and system for an intelligent photovoltaic charging pile. Environmental data is collected in real time through a sensor network, and a power generation prediction model is established in combination with the characteristics of photovoltaic modules. Based on the prediction results and battery capacity, the charging strategy is dynamically adjusted to avoid over-reliance on photovoltaic power generation. At the same time, the system obtains the charging needs and remaining power information of multiple electric vehicles, and adopts a power distribution algorithm to reasonably distribute the charging power. In addition, the present invention also combines with the grid load monitoring system to predict future load change trends, and activates the intelligent optimization algorithm when the load is too high to adjust the collaborative working mode of photovoltaic power generation, battery storage and grid power supply. Through machine learning algorithms, the present invention continuously optimizes system parameters and improves overall efficiency. When the light fluctuations are large, the backup power supply can also be activated to ensure charging stability. This intelligent collaborative working mode significantly improves the efficiency and reliability of the photovoltaic charging system, and realizes the rational utilization and intelligent management of power resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a flow chart of a control method for an intelligent photovoltaic charging pile of the present invention.

[0024] Figure 2 Schematic diagram of a control method and system for an intelligent photovoltaic charging pile according to the present invention.

[0025] Figure 3 This is another schematic diagram of a control method and system for an intelligent photovoltaic charging pile according to the present invention.

[0026] Figure 4 This is a structural diagram of a control method and system for an intelligent photovoltaic charging pile according to the present invention. DETAILED DESCRIPTION

[0027] The following will describe the technical solutions in the embodiments of the present invention in detail with reference to the accompanying drawings. The described embodiments are only a part of the embodiments of the present invention.

[0028] like Figure 1-4 In this embodiment, a control method of a smart photovoltaic charging pile may specifically include:

[0029] S101. Acquire real-time environmental data of light intensity, temperature, and humidity through a sensor network, establish a photovoltaic power generation prediction model based on the physical characteristics of photovoltaic modules, and output a power generation prediction value.

[0030] Acquire real-time illuminance, temperature, and humidity data collected by a sensor network; perform data matching on the real-time illuminance, temperature, and humidity data with pre-stored physical parameters of photovoltaic modules to obtain matching data; use a regression algorithm to analyze the matching data and establish a photovoltaic power generation prediction model; determine a theoretical power generation prediction value under current environmental conditions based on the photovoltaic power generation prediction model; if the real-time illuminance, temperature, and humidity data change, update the real-time illuminance, temperature, and humidity data, and re-determine the theoretical power generation prediction value under current environmental conditions based on the updated real-time illuminance, temperature, and humidity data; compare the theoretical power generation prediction value with historical power generation data to determine the accuracy of the theoretical power generation prediction value.

[0031] For example, illuminance data is primarily collected using a light meter. For example, solar radiation intensity can reach approximately 1,000 watts per square meter at noon on a sunny day, but may drop below 200 watts in the morning and evening. Temperature and humidity sensors typically feature an integrated design, capable of measuring both parameters simultaneously. The temperature measurement range typically ranges from -40°C to 80°C, and the humidity measurement range from 0% to 100%. The physical parameters of photovoltaic modules include key indicators such as open-circuit voltage, short-circuit current, and maximum power point voltage and current. For example, under standard testing conditions, the open-circuit voltage of monocrystalline silicon modules is approximately 40 volts, the short-circuit current can reach 9 amperes, and the maximum power can reach 300 watts. When matching these parameters with real-time environmental data, the impact of the temperature coefficient must be considered. For example, for every degree increase in temperature, the output power decreases by approximately 0.4%. For regression analysis, a multivariate linear regression model can be used, with illuminance, temperature, and humidity as the independent variables and power generation as the dependent variable. The model training data should include historical data under various weather conditions. For example, on a cloudy day, illuminance may be around 200 watts per square meter, at which point power generation is approximately 20% of the rated power. When building a prediction model, it's necessary to account for module degradation, which typically ranges from 0.5% to 1% per year. System efficiency, including inverter efficiency, is typically above 95%, and line losses are approximately 2%, must also be considered when calculating theoretical power generation. Data updates can be set to five minutes to ensure real-time predictions. Prediction results can be verified by comparing them with historical data from the same period. For example, under similar weather conditions, if the predicted value deviates from the historical actual value within 5%, the prediction is considered accurate. For abnormal data, such as a sudden drop in illuminance without a corresponding change in temperature, the system should identify possible obstruction or sensor failure. During the data matching process, a mapping between environmental parameters and power generation efficiency must be established. For example, when the illuminance is 800 watts per square meter, the temperature is 25 degrees Celsius, and the humidity is 60%, the power generation efficiency of a PV module can reach 17%. At this time, if the module area is two square meters, the theoretical power generation should be approximately 270 watts. The accuracy of the prediction model is closely related to the data sampling time span. It is recommended to use at least one year of historical data for model training to fully account for seasonal variations. For example, power generation during midday hours in summer is typically more than twice that in winter, and this seasonal difference needs to be fully reflected in the model.

[0032] S102. Determine whether the charging strategy needs to be adjusted based on the predicted power generation value and the current capacity of the battery. If the predicted value is lower than a preset threshold, reduce the charging power to avoid over-reliance on photovoltaic power generation.

[0033] Obtain a predicted photovoltaic power generation value and a current battery capacity value; compare the predicted photovoltaic power generation value with a preset threshold value; if the predicted photovoltaic power generation value is less than the preset threshold value, calculate the charging power reduction amplitude, and generate a new charging power value according to the charging power reduction amplitude; generate an updated charging strategy value according to the new charging power value; use the updated charging strategy value to control the charging process of the battery and obtain an adjusted charging power; monitor the changes of the predicted photovoltaic power generation value and the current battery capacity value in real time to determine whether the charging strategy value needs to be further adjusted; if the predicted photovoltaic power generation value continues to be lower than the preset threshold value, gradually reduce the charging power value to avoid overcharging of the battery; maintain the battery within the current capacity range through cyclic monitoring and adjustment.

[0034] For example, the predicted photovoltaic power generation is closely related to battery charging management, requiring real-time adjustment of charging strategies to accommodate the fluctuating nature of photovoltaic power generation. Sensor-generated data such as light intensity can be used to predict photovoltaic power generation over a period of time. For example, in cloudy weather, the predicted power generation may drop to 60% of normal. The predicted value is then compared with a preset threshold. For example, the system's preset threshold is 70% of peak power generation. When the predicted value falls below this threshold, the charging strategy adjustment mechanism is triggered. The charging power adjustment needs to take into account the current battery capacity. For example, if the battery capacity is at 85%, the charging power can be appropriately reduced. The calculation of the charging power reduction range comprehensively considers factors such as the predicted power generation, battery capacity, and power demand. For example, if the predicted power generation drops to 80% of the threshold, the charging power can be reduced by 20% to ensure more stable system operation. A new charging strategy needs to be dynamically generated based on the adjusted charging power. For example, if the original charging strategy was constant power charging, the adjustment might switch to variable power charging. This strategy adjustment can reduce reliance on photovoltaic power generation. When photovoltaic power generation is insufficient, the system can automatically reduce charging power to avoid impacting the power grid. During real-time monitoring, the system continuously evaluates the effectiveness of the charging strategy. If the predicted value remains below a threshold for a certain period of time—for example, if it remains below 60% of the preset threshold for three consecutive hours—the system further reduces the charging power. This gradual adjustment prevents overcharging of the battery and prolongs its lifespan. Throughout the entire process, the charging strategy must be adjusted to maintain the battery capacity within a reasonable range. For example, when the capacity is above 90%, even if photovoltaic power generation is sufficient, the charging power should be appropriately reduced. The system continuously monitors the predicted value and capacity status to ensure that the charging strategy remains optimal. This dynamic adjustment mechanism improves system stability, extends battery life, and ensures that power demand is met. By establishing a correlation between the predicted value, threshold, charging power, and battery capacity, the system implements intelligent charging management. This management approach not only improves energy efficiency, but also extends equipment lifespan and reduces operation and maintenance costs. For example, on cloudy days with frequent changes in light intensity, the system can use a predictive model to proactively adjust the charging strategy to avoid damage to the battery caused by frequent charge and discharge cycles.

[0035] S103. Obtain charging demand and remaining power information of multiple electric vehicles, and dynamically adjust the charging power of each vehicle in combination with a power distribution algorithm to ensure reasonable distribution of power resources.

[0036] Obtain the remaining power information and charging demand information of multiple electric vehicles and establish a vehicle information database of the electric vehicles; calculate the charging priority of each electric vehicle based on the remaining power information and charging demand information of each electric vehicle in the vehicle information database; use a preset power distribution algorithm to determine an initial charging power distribution plan for electric vehicles with different charging priorities; obtain the real-time grid load situation and determine whether the real-time grid load exceeds a preset threshold; if so, based on the preset power distribution algorithm and combined with the real-time grid load situation, dynamically adjust the initial charging power distribution plan to obtain an updated charging power distribution plan; based on the updated charging power distribution plan, recalculate the charging power of each electric vehicle and generate an updated charging instruction; send the updated charging instruction to each electric vehicle through the charging pile controller to control the execution of the charging operation; continuously monitor the power changes of each electric vehicle and the grid load situation during the charging process, and cyclically execute the dynamic adjustment and power distribution process.

[0037] For example, the smart charging system first establishes a vehicle information database containing basic information about each electric vehicle. For example, consider a residential community with 50 electric vehicles. The database records each vehicle's model, battery capacity, remaining charge percentage, and expected time of use. For example, a certain vehicle has a 70 kWh battery capacity and a current remaining charge of 30%. The user sets the vehicle for use at 8:00 AM the following morning. Charging priority calculations take multiple factors into consideration. For example, if a vehicle has a remaining charge of 20% and is expected to be used in two hours, and another vehicle has a remaining charge of 15% and is expected to be used in six hours, the system will prioritize the vehicle expected to be used in two hours. Priority scoring uses a weighted approach, with remaining charge weighted at 0.4, usage time weighted at 0.4, and user-specific needs weighted at 0.2. The initial charging power allocation plan is based on the total power limit. Assuming the community's power distribution capacity is 500 kilowatts and the current grid base load is 200 kilowatts, the total power available for electric vehicle charging is 300 kilowatts. High-priority vehicles receive 11 kilowatts of charging power, medium-priority vehicles receive 7 kilowatts, and low-priority vehicles receive 3.5 kilowatts. Grid load monitoring utilizes real-time data collection. If the grid load exceeds the preset threshold of 450 kilowatts during a given period, the system immediately activates a dynamic adjustment mechanism. High-priority vehicles are reduced to 7 kilowatts, medium-priority vehicles to 3.5 kilowatts, and low-priority vehicles are suspended. This adjustment ensures stable grid operation and prevents overload trips. Charging power is redistributed in a decreasing manner. For example, if a high-priority vehicle was originally allocated 11 kilowatts of charging power, the system calculates that the charging time will be extended by approximately 57 percent after the power is reduced to 7 kilowatts. The system also estimates the time required to complete charging at this power level to ensure that the user's specified vehicle usage time is met. Charging commands are issued using a hierarchical control strategy. After receiving the new charging power command, the charging station controller uses a soft start method to gradually adjust to the target power within 30 seconds to prevent sudden large power fluctuations from impacting the grid. Each charging station is equipped with a power meter that provides real-time feedback on actual charging power. During charging monitoring, the system collects data every minute, recording information such as charging power, charged capacity, and battery temperature. If an anomaly is detected, such as a charging power exceeding the set value by 10%, the system automatically issues an alarm and, if necessary, initiates a protective power reduction. Through continuous monitoring and dynamic adjustment, safe and efficient management of the electric vehicle charging process can be achieved.

[0038] S104. Obtain real-time load data through the power grid load monitoring system, combine it with historical load curves, predict future power grid load change trends, and determine whether the charging strategy needs to be adjusted.

[0039] Acquire real-time load data and historical load curve data of the power grid, and preprocess the real-time load data and the historical load data using a time series analysis method to obtain preprocessed load data; input the preprocessed load data into a pre-established load forecasting model, and use a long short-term memory network algorithm to perform load forecasting to obtain a load change trend forecast result for the next 24 hours; determine whether the forecast result exceeds a preset load threshold, and if so, trigger a charging strategy adjustment mechanism; use a decision tree algorithm to analyze the current charging strategy, and determine the optimal charging strategy adjustment plan based on the predicted load change trend; send the optimal charging strategy adjustment plan to the charging pile control system, and execute the adjusted charging strategy; monitor the load changes of the power grid in real time, and determine whether the load value continues to exceed the preset load threshold. If so, re-execute the load forecast and strategy adjustment process until the load value returns to a normal range.

[0040] For example, a power grid load monitoring system integrates data acquisition devices from multiple measurement points to acquire raw data such as voltage and current. This data is then fused and processed to generate real-time load data. For example, the load data for a regional power grid at 9:00 AM on a weekday is 8,000 kilowatts. This value is then compared with historical data for the same period. Historical load curve data typically includes electricity consumption data from the same time period and similar weather conditions over the past year. Data cleaning removes outliers, such as sudden high loads caused by equipment failures. Time series analysis uses a sliding window method to smooth the data, with a window size of up to four hours. Short-term fluctuations are eliminated by calculating a moving average. Long-short-term memory network algorithms offer significant advantages in load forecasting, as they can simultaneously consider short-term load fluctuations and long-term electricity consumption patterns. The forecasting model inputs include features such as temperature, date type, and historical load history, and outputs time-based load forecasts for the next 24 hours. In practice, a load forecast for a commercial area on a summer weekday indicates that the load between 2:00 PM and 4:00 PM is likely to exceed the 10,000 kilowatt warning threshold. This triggers the charging strategy adjustment mechanism, where a decision tree algorithm constructs judgment rules based on factors such as charging priority, remaining power, and user demand. For example, electric vehicles can be divided into three categories: emergency vehicles, public transportation, and private cars, each with a different charging weight. Charging strategy adjustments can include power limits and time shifting. When the forecasted load approaches a threshold, the charging power for private cars is limited, shifting some of the charging demand to the nighttime off-peak period. For example, the original 20-kilowatt charging power can be reduced to 10 kilowatts to extend the charging time. Emergency vehicles maintain their original charging strategy to ensure emergency service needs. The continuous load monitoring mechanism sets multiple early warning thresholds, initiating response measures step by step when the load reaches the warning value. For example, if the load in an industrial park exceeds the warning value for 30 minutes during peak electricity consumption, the system automatically re-runs the forecast analysis and dynamically adjusts the charging power allocation plan based on the updated load trend. This dynamic adjustment mechanism can smooth load fluctuations, improve grid operation stability, and meet the charging needs of electric vehicle users.

[0041] S105. If the predicted grid load value exceeds a preset threshold, the intelligent optimization algorithm is activated to adjust the collaborative working mode of photovoltaic power generation, battery storage and grid power supply to reduce the grid load.

[0042] Obtain a real-time prediction value of the grid load, and compare the real-time prediction value with a preset threshold; if the real-time prediction value exceeds the preset threshold, start the intelligent optimization algorithm, and use the intelligent optimization algorithm to calculate the optimal coordination mode of photovoltaic power generation, battery power storage and grid power supply; adjust the output power of photovoltaic power generation according to the optimal coordination mode to obtain an adjusted photovoltaic power generation output power; adjust the battery charging and discharging strategy according to the optimal coordination mode to obtain an optimized battery charging and discharging strategy; determine the optimized distribution strategy of grid power supply according to the optimal coordination mode; reduce the grid load according to the adjusted photovoltaic power generation output power, the optimized battery charging and discharging strategy and the optimized distribution strategy of grid power supply to obtain an adjusted grid load value; determine whether the adjusted grid load value meets the preset threshold requirement, and if so, output the adjusted photovoltaic power generation output power, the optimized battery charging and discharging strategy and the optimized distribution strategy of grid power supply.

[0043] For example, grid load forecasting is a key component of smart grid management. Data acquisition equipment collects real-time load data from key nodes, such as substations and distribution rooms. This data is combined with historical load curves and weather forecasts, and deep learning algorithms are used to generate load forecasts. The preset threshold is typically set at 85% of the substation's rated capacity, ensuring both power supply security and optimal utilization of power resources. When the predicted load reaches 1,200 kWh per hour, approaching the preset threshold, the system automatically initiates an intelligent optimization algorithm. The coordinated optimization of photovoltaic power generation, battery storage, and grid power supply involves multiple aspects. For example, an industrial park has a rooftop photovoltaic power generation system with a total installed capacity of 1,000 kilowatts and a 500 kWh energy storage system. During peak demand periods, increasing the output power of photovoltaic power generation can alleviate pressure on the grid. The photovoltaic power generation system dynamically adjusts inverter operating parameters based on sunlight intensity and temperature conditions to maintain optimal power generation. Battery energy storage systems play a vital role in grid peak load regulation. During daytime, when photovoltaic power generation is sufficient, excess power is stored in batteries. During peak electricity demand, battery discharge can supplement demand, significantly reducing grid load. The energy storage system employs an intelligent charging and discharging strategy, charging during periods of lower electricity prices and discharging during peak hours. This ensures power supply reliability and reduces electricity costs. The grid's optimized power distribution strategy prioritizes load balancing and economic efficiency. During the peak demand period of 7:00 AM to 9:00 AM on weekdays, the system prioritizes energy storage power while increasing the grid-connected power of photovoltaic power generation. During the midday peak of photovoltaic power generation, the grid's share of power is reduced, with excess power stored in the energy storage system. In the evening, the proportion of power supplied by the three power sources is appropriately allocated based on load forecasts. The load monitoring system collects real-time data from each power consumption link and uses data analysis to determine optimization results. For example, after implementing an intelligent optimization strategy at a commercial complex, peak grid load was reduced from 800 kilowatts to 600 kilowatts, reducing electricity costs by 20%. The system continuously monitors load changes and immediately initiates a new round of optimization calculations when a load value exceeds a threshold, ensuring the power supply system is always operating optimally. Through this dynamic optimization mechanism, the power supply quality is guaranteed and the efficient use of energy is achieved.

[0044] S106. Based on historical data and real-time data, a machine learning algorithm is used to optimize the collaborative working parameters of photovoltaic power generation and battery storage to improve the overall efficiency of the system.

[0045] Acquire historical operating data and real-time operating data of the photovoltaic system and the battery system, establish a data integration model, and obtain integrated operating data; perform feature extraction on the integrated operating data to obtain a feature data set; use a random forest algorithm to optimize the parameters of the feature data set to obtain optimized feature parameters; determine the optimal collaborative working mode of the photovoltaic system and the battery system based on the optimized feature parameters; use a support vector machine algorithm to perform stability detection on the optimal collaborative working mode to obtain a detection result; if the detection result shows that the optimal collaborative working mode is unstable, correct the optimal collaborative working mode to obtain a corrected optimal collaborative working mode; obtain the corrected optimal collaborative working mode, and establish a dynamic balance model of photovoltaic power generation and battery storage capacity; use a gradient boosting decision tree algorithm to iteratively optimize the dynamic balance model to obtain an optimized dynamic balance model; generate final control strategy parameters of the photovoltaic system and the battery system based on the optimized dynamic balance model.

[0046] For example, data integration models are fundamental to understanding the operating patterns of photovoltaic and battery systems. By collecting historical data such as photovoltaic power generation, battery charge and discharge status, ambient temperature, and light intensity, combined with real-time monitoring data, a database of system operating characteristics is constructed. For example, the power generation patterns of a photovoltaic power station vary across seasons and weather conditions, as well as the battery performance at different temperatures and depths of charge and discharge. The random forest algorithm extracts system characteristics by constructing multiple decision trees. For photovoltaic systems, characteristics such as the relationship between light intensity and power generation efficiency and the impact of temperature on power generation efficiency can be extracted. For battery systems, characteristics such as the relationship between the number of charge and discharge cycles and capacity decay and the impact of ambient temperature on charging efficiency can be extracted. For example, analysis has found that when the ambient temperature exceeds 35 degrees Celsius, battery charging efficiency decreases by 15%. Determining the optimal coordinated operating mode requires consideration of multiple factors. During periods of sufficient sunlight, photovoltaic power generation is prioritized for power supply, while battery charging is also used. During peak hours, the ratio of photovoltaic power generation to battery discharge is appropriately allocated. For example, at noon, when sunlight is strongest, 70% of photovoltaic power generation can be used to power the load, and 30% can be used to charge the battery. The support vector machine algorithm is used to detect the stability of operating modes. By establishing a multidimensional feature space, it determines whether the current operating mode will cause system instability. For example, when rapid changes in light intensity are detected, the power allocation ratio between photovoltaic power generation and battery storage is adjusted promptly to avoid output power fluctuations. The dynamic balance model primarily addresses the intermittent nature of photovoltaic power generation. During cloudy days, battery discharge power is increased to supplement power shortages. During periods of abundant sunlight, excess power is stored in the battery. For example, if cloudy weather is predicted within the next four hours, battery charge can be increased in advance to prepare for subsequent power shortages. The gradient boosting decision tree algorithm improves system performance through continuous iterative optimization. Each iteration builds on the results of the previous iteration to make targeted improvements to the control strategy. For example, if analysis reveals a gradual increase in power consumption between 6:00 a.m. and 8:00 a.m., the algorithm will automatically adjust the battery discharge strategy during this period to ensure stable power supply. The final control strategy parameters include the photovoltaic power curve and the battery charge and discharge schedule. These parameters are dynamically adjusted with seasonal changes and changes in electricity demand. For example, when the air conditioning load increases in the summer, the power output of photovoltaic power generation will be increased before 9 am to prepare for the peak electricity consumption during the day.

[0047] S107. Through intelligent optimization algorithms, dynamically adjust the collaborative working mode of photovoltaic power generation, battery storage and grid power supply to ensure efficient operation of the system under different environmental conditions.

[0048] Acquire real-time data of photovoltaic power generation, the real-time data including the power generation power of the photovoltaic cells; acquire battery storage status data, the storage status data including the battery power level and charge / discharge power; acquire load demand data for grid power supply, the load demand data including the power demand of electrical equipment; determine whether the system operation mode is in an optimal state using a preset optimization algorithm based on the real-time data, the storage status data, and the load demand data; if the operation mode is not in the optimal state, dynamically adjust the coordinated operation mode of photovoltaic power generation, battery storage, and grid power supply based on current environmental conditions; when adjusting the coordinated operation mode, acquire new system operation parameters corresponding to the current environmental conditions, the new system operation parameters including photovoltaic power generation power, battery charge / discharge power, and grid power supply power; determine a coordinated operation strategy of photovoltaic power generation, battery storage, and grid power supply based on the new system operation parameters to obtain an adjusted coordinated operation mode; determine whether the adjusted coordinated operation mode is an efficient mode; if so, operate the control system according to the adjusted coordinated operation mode; if not, return to the step of dynamically adjusting the coordinated operation mode until an efficient coordinated operation mode is obtained.

[0049] For example, the real-time photovoltaic power generation data collected by intelligent algorithms includes environmental parameters such as solar irradiance, temperature, and humidity, as well as operating parameters such as PV module output voltage, current, and power. For example, on sunny summer days, the PV system will adjust its maximum power point tracking control based on changes in sunlight intensity. When the irradiance reaches 1,000 watt-hours per square meter, the system output power can reach over 90% of its design rating. Battery storage status monitoring focuses on key indicators such as charge and discharge depth, state of charge, and operating temperature. For example, lithium iron phosphate batteries maximize their service life when operating within the 20% to 80% state of charge range. If the energy storage system detects a battery temperature exceeding 40°C, it activates the cooling system to regulate the temperature. Grid load demand analysis is based on the characteristics of the power load curve. For example, industrial areas experience higher loads during the day and lower loads at night. In this case, energy storage scheduling can be optimized based on load forecasts. When peak power demand is predicted, the system will schedule battery charging in advance to prepare for peak power supply. System operational status is assessed using multiple metrics, including photovoltaic power generation efficiency, energy storage conversion efficiency, and load response speed. When PV power generation efficiency falls below the target, the system automatically checks whether the module surface requires cleaning and whether the tracking system is functioning properly. If the energy storage system's charge and discharge efficiency drops below 85%, power adjustment or equipment maintenance is required. Dynamic adjustments to the collaborative operation mode are based on real-time data and predictive models. For example, during rainy weather, the system adjusts the PV power generation plan based on the weather forecast and increases the energy storage system's discharge power to ensure a stable supply of electricity to the load. If a sudden drop in PV power is detected, the energy storage system responds within milliseconds to promptly fill the power gap. System parameter adjustments due to changing environmental conditions involve multiple levels. For example, during spring and autumn, when there is a large temperature difference between day and night, the system adjusts the operating voltage of the PV modules to optimize power generation efficiency. The energy storage system also adjusts the charge and discharge current based on ambient temperature fluctuations to protect battery life. The ultimate collaborative operation strategy balances the needs of the power generation side, the energy storage side, and the power consumption side. When PV power generation is sufficient, the power consumption is prioritized, and the excess power is used to charge the energy storage. When the power consumption exceeds the PV generation capacity, the energy storage system cooperates with the grid to share the load, achieving the most economical power supply solution.

[0050] S108. If the light fluctuation is large, the backup power supply is activated to supplement the power supply to ensure that the charging process is not affected, while optimizing the energy storage efficiency and reducing the waste of electricity resources.

[0051] Obtain the current illumination value, calculate the fluctuation rate of the current illumination value, compare the fluctuation rate with a preset threshold value, and if the fluctuation rate exceeds the threshold value, calculate the amount of electricity that needs to be replenished, determine the startup value of the backup power supply based on the amount of electricity that needs to be replenished, and obtain the current storage energy of the backup power supply. Based on the amount of electricity that needs to be replenished and the current storage energy, determine whether the backup power supply meets the power supply requirements. If so, control the backup power supply to start and supply power. If not, calculate the optimal energy storage efficiency value, adjust the operating parameters of the energy storage device based on the optimal energy storage efficiency value, and obtain the storage energy of the adjusted energy storage device. Based on the storage energy of the adjusted energy storage device, re-judge the power supply capacity of the backup power supply, determine the final amount of electricity that needs to be replenished, and control the backup power supply to start and supply power. Obtain the power consumption of the backup power supply during the power supply process, calculate the amount of electricity resource waste, and determine whether the waste exceeds the preset range. If so, adjust the power supply strategy of the backup power supply to optimize energy storage efficiency and reduce electricity resource waste.

[0052] For example, detecting sunlight fluctuations is a key indicator of photovoltaic power generation stability. Light sensors in the photovoltaic array collect real-time light intensity data, comparing changes in light values ​​between adjacent time periods to calculate the fluctuation rate. For example, on a sunny day with heavy white clouds, light intensity can drop from 1,000 watts per square meter to 300 watts per square meter in a short period of time, resulting in a fluctuation rate of 70%, far exceeding the normal operating threshold of 20%. The system needs to calculate the required supplemental power to maintain a stable power supply. When light fluctuations cause insufficient photovoltaic output power, the actual power gap must be calculated and the startup parameters of the backup power supply must be determined. For example, if the power load is 50 kilowatts and the actual photovoltaic output is only 30 kilowatts, a backup power supply of 20 kilowatts is required. The backup power supply typically uses a battery energy storage system, and its energy storage status must be monitored in real time. Assume that the battery pack has a rated capacity of 100 kilowatt-hours and currently has 60 kilowatt-hours of remaining capacity, it can meet the 20 kilowatt supplemental capacity requirement. However, if the remaining capacity falls below the required supplemental capacity, storage efficiency must be optimized. This optimization includes adjusting charging and discharging strategies and temperature management. By controlling the charge and discharge currents and regulating battery temperature, the battery can be kept operating at its optimal state. For example, reducing the charging current from 0.5x to 0.3x can reduce heat loss and improve charging efficiency by approximately 5%. Monitoring power consumption during the power supply process is crucial for energy conservation and consumption reduction. The system records the power profile of each consumer and analyzes whether there is unnecessary energy waste. If it detects that certain devices are still consuming high power even in standby mode, or if the supply voltage is too high, resulting in additional losses, the power supply strategy needs to be adjusted promptly. Dynamically adjusting the backup power supply strategy can significantly improve system efficiency. For example, when there is sufficient sunlight, photovoltaic power generation is prioritized while simultaneously charging the battery. When rainy weather is predicted, the battery storage capacity is increased in advance to prepare for subsequent power supply fluctuations. This proactive adjustment method can improve overall system efficiency by over 15%, effectively reducing power waste.

[0053] The present invention provides an intelligent photovoltaic charging system and an optimization system thereof, which mainly include:

[0054] The environmental data acquisition module is used to obtain real-time environmental data such as light intensity, temperature, and humidity through a sensor network. It then combines the physical characteristics of photovoltaic modules to establish a photovoltaic power generation prediction model and output the power generation prediction value.

[0055] The power generation prediction module is used to determine whether the charging strategy needs to be adjusted based on the power generation forecast value and the current capacity of the battery. If the forecast value is lower than the preset threshold, the charging power is reduced to avoid over-reliance on photovoltaic power generation;

[0056] The charging strategy adjustment module is used to obtain the charging demand and remaining power information of multiple electric vehicles, and dynamically adjust the charging power of each vehicle in combination with the power distribution algorithm to ensure the rational allocation of power resources;

[0057] The electric vehicle charging management module is used to obtain real-time load data from the power grid load monitoring system, combine it with historical load curves, predict future power grid load trends, and determine whether the charging strategy needs to be adjusted;

[0058] The grid load monitoring module is used to activate the intelligent optimization algorithm if the grid load forecast value exceeds the preset threshold, adjust the collaborative working mode of photovoltaic power generation, battery storage and grid power supply, and reduce the grid load;

[0059] An intelligent optimization module, which uses machine learning algorithms to optimize the collaborative working parameters of photovoltaic power generation and battery storage based on historical and real-time data to improve the overall efficiency of the system;

[0060] The collaborative work optimization module is used to dynamically adjust the collaborative work mode of photovoltaic power generation, battery storage and grid power supply through intelligent optimization algorithms to ensure efficient operation of the system under different environmental conditions;

[0061] The backup power management module is used to activate the backup power supply to supplement the power supply if the light fluctuation is large, ensuring that the charging process is not affected, while optimizing energy storage efficiency and reducing power resource waste.

[0062] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to specific embodiments. Obviously, many modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A control method for an intelligent photovoltaic charging pile, characterized in that: The method comprises the following steps: S101, acquiring real-time environmental data of light intensity, temperature, and humidity through a sensor network, establishing a photovoltaic power generation prediction model based on the physical characteristics of the photovoltaic modules, and outputting a power generation prediction value; S102. Determine whether the charging strategy needs to be adjusted based on the predicted power generation value and the current capacity of the battery. If the predicted value is lower than a preset threshold, reduce the charging power to avoid over-reliance on photovoltaic power generation. S103, obtaining charging demand and remaining power information of multiple electric vehicles, and dynamically adjusting the charging power of each vehicle in combination with a power distribution algorithm to ensure reasonable allocation of power resources; S104. Obtain real-time load data through the power grid load monitoring system, combine it with historical load curves, predict future power grid load change trends, and determine whether the charging strategy needs to be adjusted; S105. If the grid load forecast value exceeds a preset threshold, the intelligent optimization algorithm is activated to adjust the collaborative working mode of photovoltaic power generation, battery storage, and grid power supply to reduce the grid load; S106. Using machine learning algorithms to optimize the collaborative working parameters of photovoltaic power generation and battery storage based on historical data and real-time data to improve the overall efficiency of the system; S107, through intelligent optimization algorithms, dynamically adjust the collaborative working mode of photovoltaic power generation, battery storage and grid power supply to ensure efficient operation of the system under different environmental conditions; S108. If the light fluctuation is large, the backup power supply is activated to supplement the power supply to ensure that the charging process is not affected, while optimizing the energy storage efficiency and reducing the waste of electricity resources.

2. The control method of a smart photovoltaic charging pile according to claim 1, characterized in that: The S101 includes: Obtain real-time illumination, temperature and humidity data collected by the sensor network; Matching the real-time illuminance, temperature, and humidity data with pre-stored physical parameters of the photovoltaic module to obtain matching data; Using a regression algorithm to analyze the matching data and establish a photovoltaic power generation prediction model; Determining a theoretical power generation prediction value under current environmental conditions based on the photovoltaic power generation prediction model; If the real-time illuminance, temperature and humidity data change, the real-time illuminance, temperature and humidity data are updated, and the theoretical power generation prediction value under the current environmental conditions is re-determined based on the updated real-time illuminance, temperature and humidity data; The theoretical power generation prediction value is compared with the historical power generation data to determine the accuracy of the theoretical power generation prediction value.

3. The control method of a smart photovoltaic charging pile according to claim 1, characterized in that: The S102 includes: Obtain the predicted value of photovoltaic power generation and the current capacity of the battery; Comparing the photovoltaic power generation prediction value with a preset threshold value, and if the photovoltaic power generation prediction value is less than the preset threshold value, calculating a charging power reduction range, and generating a new charging power value according to the charging power reduction range; Generating an updated charging strategy value according to the new charging power value; Using the updated charging strategy value to control the charging process of the battery to obtain an adjusted charging power; Monitor the changes of the photovoltaic power generation forecast value and the current capacity value of the battery in real time to determine whether the charging strategy value needs to be further adjusted; If the photovoltaic power generation forecast value continues to be lower than the preset threshold, gradually reducing the charging power value to avoid overcharging of the battery; The battery is maintained within the current capacity range through cyclic monitoring and adjustment.

4. The control method of a smart photovoltaic charging pile according to claim 1, characterized in that: The S103 includes: Acquiring remaining power information and charging demand information of a plurality of electric vehicles, and establishing a vehicle information database of the electric vehicles; Calculating the charging priority of each electric vehicle based on the remaining power information and charging demand information of each electric vehicle in the vehicle information database; Using a preset power allocation algorithm, the initial charging power allocation plan is determined for electric vehicles with different charging priorities; Obtaining real-time grid load conditions, and determining whether the real-time grid load exceeds a preset threshold; If so, dynamically adjusting the initial charging power allocation plan based on the preset power allocation algorithm and the real-time grid load to obtain an updated charging power allocation plan; Recalculating the charging power of each electric vehicle according to the updated charging power allocation scheme and generating an updated charging instruction; The updated charging instruction is sent to each of the electric vehicles through the charging pile controller to control the execution of the charging operation; The power changes of each electric vehicle and the load condition of the power grid are continuously monitored during the charging process, and the dynamic adjustment and power distribution process is cyclically executed.

5. A control method for a smart photovoltaic charging pile according to any one of claims 1 to 4, characterized in that: The S104 includes: Acquire real-time load data and historical load curve data of the power grid, and preprocess the real-time load data and the historical load data using a time series analysis method to obtain preprocessed load data; The pre-processed load data is input into a pre-established load forecasting model, and a long short-term memory network algorithm is used to perform load forecasting to obtain a load change trend forecast result for the next 24 hours; Determining whether the prediction result exceeds a preset load threshold, and if so, triggering a charging strategy adjustment mechanism; Analyze the current charging strategy using a decision tree algorithm, and determine the optimal charging strategy adjustment plan based on the predicted load change trend; Sending the optimal charging strategy adjustment plan to the charging pile control system to execute the adjusted charging strategy; Monitor the load changes of the power grid in real time to determine whether the load value continues to exceed the preset load threshold. If so, re-execute the load forecasting and strategy adjustment process until the load value returns to the normal range.

6. A control method for a smart photovoltaic charging pile according to any one of claims 1 to 4, characterized in that: The S105 includes: Obtaining a real-time predicted value of the grid load, and comparing the real-time predicted value with a preset threshold; If the real-time prediction value exceeds the preset threshold, the intelligent optimization algorithm is activated to calculate the optimal coordination mode of photovoltaic power generation, battery storage and grid power supply; adjusting the output power of photovoltaic power generation according to the optimal coordination mode to obtain an adjusted output power of photovoltaic power generation; Adjusting the battery charging and discharging strategy according to the optimal coordinated mode to obtain an optimized battery charging and discharging strategy; Determining an optimal distribution strategy for power supply from the power grid according to the optimal coordination mode; According to the adjusted photovoltaic power generation output power, the optimized battery charging and discharging strategy and the optimized distribution strategy of the power grid, the grid load is reduced to obtain the adjusted grid load value; Determine whether the adjusted grid load value meets the preset threshold requirement, and if so, output the adjusted photovoltaic power generation output power, the optimized battery charging and discharging strategy, and the optimized distribution strategy for grid power supply.

7. A control method for a smart photovoltaic charging pile according to any one of claims 1 to 4, characterized in that: The S106 includes: Obtain historical and real-time operating data of the photovoltaic system and battery system, establish a data integration model, and obtain integrated operating data; performing feature extraction on the integrated operation data to obtain a feature data set; Using a random forest algorithm to perform parameter optimization on the feature data set to obtain optimized feature parameters; Determining the optimal collaborative working mode of the photovoltaic system and the battery system based on the optimized characteristic parameters; A support vector machine algorithm is used to perform stability detection on the optimal collaborative working mode to obtain a detection result; If the detection result indicates that the optimal collaborative working mode is unstable, the optimal collaborative working mode is corrected to obtain a corrected optimal collaborative working mode; Obtaining the modified optimal collaborative working mode and establishing a dynamic balance model of photovoltaic power generation and battery storage capacity; Iteratively optimizing the dynamic equilibrium model using a gradient boosting decision tree algorithm to obtain an optimized dynamic equilibrium model; According to the optimized dynamic balance model, final control strategy parameters of the photovoltaic system and the battery system are generated.

8. A control method for a smart photovoltaic charging pile according to any one of claims 1 to 4, characterized in that: The S107 includes: Acquiring real-time data of photovoltaic power generation, wherein the real-time data includes power generation of photovoltaic cells; Acquiring battery storage status data, wherein the battery storage status data includes battery power and charge and discharge power; Obtaining load demand data for power supply from the power grid, wherein the load demand data includes power demand of electrical equipment; Based on the real-time data, the power storage state data and the load demand data, a preset optimization algorithm is used to determine whether the system operation mode is in an optimal state; If the operating mode does not reach the optimal state, the coordinated working mode of photovoltaic power generation, battery storage and grid power supply is dynamically adjusted according to the current environmental conditions; When adjusting the collaborative working mode, new system operating parameters corresponding to the current environmental conditions are obtained, wherein the new system operating parameters include photovoltaic power generation, battery charge and discharge power, and grid power supply power; Determining a coordinated working strategy of photovoltaic power generation, battery power storage, and grid power supply based on the new system operating parameters to obtain an adjusted coordinated working mode; Determining whether the adjusted collaborative working mode is an efficient mode; If so, the control system operates according to the adjusted collaborative working mode; If not, return to the step of dynamically adjusting the collaborative working mode until an efficient collaborative working mode is obtained.

9. A control system for an intelligent photovoltaic charging pile, characterized in that: The system is used to implement the control method of a smart photovoltaic charging pile according to any one of claims 1 to 8, and the system includes: The environmental data acquisition module is used to obtain real-time environmental data such as light intensity, temperature, and humidity through a sensor network. It then combines the physical characteristics of photovoltaic modules to establish a photovoltaic power generation prediction model and output the power generation prediction value. The power generation prediction module is used to determine whether the charging strategy needs to be adjusted based on the power generation forecast value and the current capacity of the battery. If the forecast value is lower than the preset threshold, the charging power is reduced to avoid over-reliance on photovoltaic power generation; The charging strategy adjustment module is used to obtain the charging demand and remaining power information of multiple electric vehicles, and dynamically adjust the charging power of each vehicle in combination with the power distribution algorithm to ensure the rational allocation of power resources; The electric vehicle charging management module is used to obtain real-time load data from the power grid load monitoring system, combine it with historical load curves, predict future power grid load trends, and determine whether the charging strategy needs to be adjusted; The grid load monitoring module is used to activate the intelligent optimization algorithm if the grid load forecast value exceeds the preset threshold, adjust the collaborative working mode of photovoltaic power generation, battery storage and grid power supply, and reduce the grid load; An intelligent optimization module, which uses machine learning algorithms to optimize the collaborative working parameters of photovoltaic power generation and battery storage based on historical and real-time data to improve the overall efficiency of the system; The collaborative work optimization module is used to dynamically adjust the collaborative work mode of photovoltaic power generation, battery storage and grid power supply through intelligent optimization algorithms to ensure efficient operation of the system under different environmental conditions; The backup power management module is used to activate the backup power supply to supplement the power supply if the light fluctuation is large, ensuring that the charging process is not affected, while optimizing energy storage efficiency and reducing power resource waste.

Images