Supply chain demand prediction and dynamic optimization system based on artificial intelligence

Abstract

The invention discloses a supply chain demand prediction and dynamic optimization system based on artificial intelligence, and relates to the technical field of charging piles, and the system comprises a sensing module which is used for collecting the operation data of the charging piles in real time, and specifically comprises the daily average charging frequency of a single pile and the peak power duration time; the sensing module is also used for dynamically acquiring new energy vehicle growth data in the area and acquiring vehicle battery capacity distribution data; the early warning module is also used for periodically receiving extreme weather early warning information issued by a meteorological department; according to long-term demand prediction, the number of newly added charging piles is determined, and the corresponding number of charging piles are installed in the next long-term period; according to mid-term demand prediction, electric energy reserve and demand conditions of charging piles in each area in the area are monitored in real time, and distribution of the charging piles is dynamically adjusted; according to short-term demand prediction, dynamic time-sharing pricing of the charging pile is calculated, and a user is guided to reasonably use the charging pile in different time periods; the utilization rate of the charging pile can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of charging piles, and specifically refers to a supply chain demand prediction and dynamic optimization system based on artificial intelligence. Background Art

[0002] It should be noted that with the increasing market share of new energy vehicles, most people use new energy vehicles for short-distance trips such as going to work. Due to the low cruising range of new energy vehicles, charging stations are installed in large industrial areas (such as industrial areas and high-tech areas in cities), equipped with sufficient charging piles to meet the charging needs of employees or area visitors.

[0003] Since the electricity prices are different at night and during the day, in order to make a profit, charging stations use the low valley electricity price at night for charging and supply electricity to new energy vehicles in the area during the day. The charging piles equipped with energy storage devices store electricity at night and discharge during the day, reducing the regional electricity cost. However, since most employees will charge their vehicles after arriving at the area during the morning rush hour, in order to avoid the transformer overload caused by the superposition of regional electricity demand peaks, it is necessary to predict the charging demand in advance and formulate subsequent measures based on the prediction results.

[0004] However, in the actual use process, the prediction results are all short-term predictions. In seasons with frequent extreme weather, if the previous predictions are followed, the usage requirements cannot be met. Summary of the Invention

[0005] To solve the above problems, the present invention provides a supply chain demand prediction and dynamic optimization system based on artificial intelligence.

[0006] A supply chain demand prediction and dynamic optimization system based on artificial intelligence, characterized in that it includes a sensing module, and the sensing module is used to collect the operation data of the charging piles in real time, specifically including the average daily charging times of a single pile and the peak power duration.

[0007] The sensing module is also used to dynamically obtain the growth data of new energy vehicles in the area and obtain the vehicle battery capacity distribution data.

[0008] It is also used to periodically receive the extreme weather warning information issued by the meteorological department, and the sensing module transmits all the data.

[0009] It also includes a demand prediction module, and the demand prediction module is used to receive the data of the sensing module and obtain the demand prediction result after the prediction period according to the data of the sensing module. The demand prediction result specifically includes decomposing the prediction period into three time periods: long term, medium term and short term, and obtaining the demand prediction for the long term period, the demand prediction for the medium term period and the demand prediction for the short term period.

[0010] It further includes a dynamic optimization module, which is used to receive the data of the perception module and the demand prediction results of the demand prediction module, and perform corresponding dynamic optimizations according to the demand predictions in the long-term period, medium-term period, and short-term period in the above demand prediction results.

[0011] Preferably, the specific working steps of the perception module are as follows:

[0012] For the real-time collection of the operating data of the charging piles, specifically, through the sensors and data collection devices installed on the charging piles, the operating data of the charging piles are obtained in real time, including the average daily charging times Q of a single pile and the peak power duration T;

[0013] The specific acquisition steps include: First, through the data collection device of the charging pile, the start and end times of each charging are recorded in real time. Whenever a vehicle starts charging, the system will generate a charging record, including information such as the charging start time and end time. At the end of the statistical period, the system will count the number of charging times Q of this charging pile within this day;

[0014] A power sensor is installed in the charging pile to monitor the power output during the charging process in real time. The control system of the charging pile will record the data of the power change over time. When the power reaches the peak, the timing starts until the power drops below the peak and then stops timing, so as to obtain the peak power duration T.

[0015] Preferably, the specific working steps of the perception module further include:

[0016] By sharing data with the vehicle management system, the growth data of new energy vehicles within the range is dynamically obtained. At the same time, the vehicle battery capacity distribution data is obtained by using the V2X communication protocol;

[0017] The battery capacity distribution data specifically includes the average battery capacity V sales ; the average battery capacity V sales Obtained by dividing the sum of the battery capacities of all vehicles registered in the area by the total number of vehicles;

[0018] For the periodic reception of extreme weather warning information issued by the meteorological department, specifically, the 72-hour extreme weather warning information is obtained through the local meteorological department.

[0019] Preferably, the specific working steps of the demand prediction module are:

[0020] For the demand prediction of the long-term period, first, according to the new energy vehicle sales data in the past three long-term periods in the obtained region, the new energy vehicle sales data in the three long-term periods are normalized to the interval [0, 1]. Select 64 LSTM units, add a Dropout layer to reduce overfitting, connect a Dense layer, and output the new energy vehicle sales data in the next long-term period to complete the construction of the model. Input the normalized new energy vehicle sales data in the three long-term periods into the completed model to output the new energy vehicle sales data P in the next long-term period. new ;

[0021] Then, according to the formula calculate the basic demand prediction Qh for each long-term period. L , where Δ grid is the impact of grid expansion;

[0022] Then subtract the basic demand prediction Qh for each long-term period L from the actual demand for each long-term period to obtain the historical residual ∈1;

[0023] Select an MLP neural network model, divide all the obtained historical residual ∈1 data into a training set, a validation set, and a test set, and divide them in the ratio of 70%, 15%, and 15%. Train to obtain a residual model, and input the historical residual ∈1 into the residual model to obtain the residual ∈ in the next long-term period;

[0024] Then, according to the formula calculate the demand prediction Q for the long-term period. L .

[0025] Preferably, the specific working steps of the demand prediction module further include:

[0026] For the demand prediction of the medium-term period, according to the formula

[0027]

[0028] calculate the medium-term demand Q M , where σ is the regional economic vitality factor, S season is the seasonal fluctuation coefficient, A is the charging station radiation capacity value, μ is the distance attenuation coefficient, and R is the number of charging piles within the range of the charging station;

[0029] where e -μd is a distance attenuation factor used to describe the phenomenon that the charging demand gradually weakens as the distance increases, and d is the distance attenuation coefficient from the area where the charging station is located in the region.

[0030] Preferably, the specific working steps of the demand prediction module further include:

[0031] For the regional economic vitality factor ρ, obtained, where 1 is the annual GDP of the city where the region is located, 2 is the total annual GDP of the country, N1 is the number of newly registered vehicles in the city where the region is located during the medium-term, and N2 is the total number of vehicles in the city where the region is located;

[0032] For the seasonal fluctuation coefficient S season , obtained by fitting based on the charging volume data in the same period of the past three years. Specifically, obtained, where t represents the month;

[0033] For the charging station radiation capacity value A, according to the formula obtained, where P t is the average daily passenger flow within the range of the charging station, D p is the charging pile density, N b is the number of bus lines near the region.

[0034] Preferably, the specific working steps of the demand prediction module further include:

[0035] For the demand prediction in the short-term period, the specific method is;

[0036] Q S = Q b M *(1 + 0.3ΔT + 0.4R r + 0.3W S )*(1 + 0.15*ΔC)*X

[0037] where ΔT is the difference between the current temperature and the standard temperature, R r is the current rainfall intensity, W S is the current wind speed, ΔC is the battery capacity attenuation rate, obtained by subtracting the current battery capacity from the initial battery capacity and then dividing by the initial battery capacity, and X is the traffic congestion value, obtained by obtained, where TL is the traffic index, is the benchmark value of the charging demand, that is, the basic level of the charging demand without other influencing factors.

[0038] Preferably, the specific working steps of the dynamic optimization module include the following:

[0039] Every other cycle, calculate the prediction error E of the short-term demand prediction, specifically according to the formula: Calculate to obtain the prediction error E, where Q1 is the actual charging demand value. If E is greater than 15%, then adjust The weight coefficients of 0.3, 0.3, and 0.4. When E is greater than 0, adjust 0.3, 0.3, and 0.4 to increase by 2%. If E is less than 0, adjust 0.3, 0.3, and 0.4 to decrease by 2%.

[0040] Preferably, the specific working steps of the dynamic optimization module further include the following:

[0041] For the long-term period, according to the formula Calculate the newly added number of charging piles N, and according to the newly added number of charging piles N, in the next long-term period, install N charging piles in this area, where P x Is the power of a single charging pile, and T1 is the average working time of the charging pile;

[0042] In the medium-term period, obtain the real-time power reserve data E and the safe power threshold E1 of each charging pile in the area from the perception module, and obtain the demand forecasts Q L 、Q M And Q S ;

[0043] Calculate the inventory monitoring degree index H of each area. When the inventory monitoring degree index H is greater than the preset threshold, move the movable charging piles to this area;

[0044] In the short-term period, according to the formula N j =P L *(1 + 0.1*Q L ), calculate the dynamic time-of-use pricing N j Of the charging piles in the short-term period, where P L Is the basic electricity price.

[0045] Preferably, the specific steps for calculating the inventory monitoring degree index H of the charging piles are as follows:

[0046] Through H = (E - E1)+(Q L *0.2)+(Q M *0.1)+(Q S *0.02), calculate and obtain the inventory monitoring degree index H of each charging pile.

[0047] Beneficial effects: Determine the number of new charging piles according to long-term demand forecasts, and install the corresponding number of charging piles within the next long-term period; According to medium-term demand forecasts, monitor the power reserves and demands of charging piles in each area within the region in real time, and dynamically adjust the distribution of charging piles; According to short-term demand forecasts, calculate the dynamic time-of-use pricing of charging piles to guide users to reasonably use charging piles at different times; It can improve the utilization rate of charging piles, reduce investment costs, ensure sufficient charging resources at high-demand times and locations, and enhance the user experience. Brief Description of the Drawings

[0048] Figure 1 is a flowchart of the present invention. Detailed Embodiments

[0049] As Figure 1 shown: An artificial intelligence-based supply chain demand forecasting and dynamic optimization system includes a sensing module, and the sensing module is used to collect the operation data of charging piles in real time, specifically including the average daily charging times of a single pile and the peak power duration;

[0050] The sensing module is also used to dynamically obtain the growth data of new energy vehicles in the region and obtain the vehicle battery capacity distribution data;

[0051] It is also used to periodically receive extreme weather warning information issued by the meteorological department, and the sensing module transmits all data; It should be noted that collecting the operation data of charging piles in real time, including the average daily charging times of a single pile, the peak power duration, and the fault type code, can accurately master the usage frequency, performance, and fault conditions of charging piles, providing basic data support for subsequent operation and maintenance and management decisions;

[0052] Dynamically obtaining the growth data of new energy vehicles in the region and the vehicle battery capacity distribution data helps to understand the changing trends of market demands and the characteristics of user groups, so as to better plan charging resources and service strategies;

[0053] Periodically receiving extreme weather warning information issued by the meteorological department enables the system to predict in advance the severe weather conditions that may affect the use of charging facilities and new energy vehicles, so as to take countermeasures in advance;

[0054] By collecting the operation data of charging piles in real time, faults and performance problems of charging piles can be discovered in time, improving the reliability and availability of equipment, and reducing the user waiting time and operation losses caused by equipment failures;

[0055] Mastering the growth of new energy vehicles and the vehicle battery capacity distribution data helps to optimize the layout and configuration of charging facilities, improve the utilization efficiency of charging resources, meet the growing market demands, and enhance the user experience and satisfaction;

[0056] Receiving extreme weather warning information in advance enables timely adoption of protective measures, such as strengthening charging facilities, adjusting charging strategies, etc., reducing the damage risk of adverse weather to charging facilities and new energy vehicles, and ensuring the stable operation of the system and the safety of users;

[0057] It also includes a demand forecasting module, which is used to receive the data of the sensing module and obtain the demand forecasting result after the forecasting period based on the data of the sensing module. The demand forecasting result specifically includes decomposing the forecasting period into three time periods: long-term, medium-term, and short-term, and obtaining the demand forecasting for the long-term period, the demand forecasting for the medium-term period, and the demand forecasting for the short-term period. It should be noted that the time ranges of the three time periods are set during real-time use, and the initial values are three time periods: long-term (6 - 12 months), medium-term (1 - 6 months), and short-term (7 - 30 days);

[0058] It should be noted that receiving the data transmitted by the sensing module, including charging pile operation data, new energy vehicle growth data, battery capacity distribution data, and extreme weather warning information, etc., comprehensively analyzing these data, and obtaining the demand forecasting result after the forecasting period, providing a decision-making basis for the dynamic optimization module;

[0059] Accurate demand forecasting results help to reasonably plan the construction and upgrade of charging facilities, avoid over-investment or resource shortage situations, and improve investment efficiency and operation efficiency;

[0060] It can predict the change trend of market demand in advance, enabling the operation enterprise to adjust the operation strategy in advance, such as reasonably arranging the maintenance and repair plan of charging equipment, optimizing the charging price strategy, etc., and improving the competitiveness and profitability of the enterprise;

[0061] Provide accurate data support for the dynamic optimization module, enabling it to better make real-time adjustments and optimizations according to the changes in market demand, and improving the flexibility and adaptability of the entire system;

[0062] It also includes a dynamic optimization module, which is used to receive the data of the sensing module and the demand forecasting result of the demand forecasting module, and perform corresponding dynamic optimizations according to the demand forecasting for the long-term period, the demand forecasting for the medium-term period, and the demand forecasting for the short-term period in the above demand forecasting result.

[0063] It should be noted that receiving the data of the sensing module and the demand forecasting result of the demand forecasting module, combining the actual inventory situation, comprehensively analyzing and evaluating, and triggering corresponding inventory management operations, such as cross-regional transfer, emergency replenishment, etc., to ensure the normal operation of charging facilities and the satisfaction of market demand;

[0064] It can adjust the inventory management strategy in a timely manner according to real-time data and prediction results, ensure an adequate supply of spare parts for charging equipment, reduce equipment downtime and repair delays caused by insufficient inventory, and improve the reliability and availability of the equipment;

[0065] Through reasonable inventory management operations, inventory costs can be reduced, inventory backlogs and waste can be avoided, and the real-time adjustment and optimization functions of the dynamic optimization module can enable the entire system to better cope with fluctuations and uncertainties in market demand, improve the stability and reliability of the system, and provide users with more stable and efficient charging services;

[0066] As an optional embodiment, the specific working steps of the sensing module are as follows:

[0067] Regarding the real-time collection of the operating data of charging piles, specifically, through sensors and data collection devices installed on the charging piles, the operating data of the charging piles are obtained in real time, including the average daily charging times Q of a single pile and the peak power duration T;

[0068] The specific acquisition steps include, first, through the data collection device of the charging pile, the start and end times of each charge are recorded in real time. Whenever a vehicle starts charging, the system generates a charging record, including information such as the charging start time and end time. At the end of the statistical period, the system will count the number of charging times Q of this charging pile within this day;

[0069] Install a power sensor in the charging pile to monitor the power output during the charging process in real time. The control system of the charging pile will record the data of the power change over time. When the power reaches the peak, start timing until the power drops below the peak, so as to obtain the peak power duration T. It should be noted that by counting the charging times Q and the peak power duration T, the usage frequency and load conditions of the charging pile can be understood. These data can provide a basis for the configuration and management of charging resources, such as reasonably arranging the number and layout of charging piles, optimizing the charging price strategy, etc.; According to the change trends of the charging times Q and the peak power duration T, the future charging demand can be predicted, and the planning and construction of charging facilities can be carried out in advance, which helps to improve the utilization efficiency of charging resources, meet the growing charging needs of new energy vehicles, and promote the development of the new energy vehicle industry.

[0070] As an optional embodiment, the specific working steps of the sensing module further include:

[0071] By sharing data with the vehicle management system, the growth data of new energy vehicles within the range is dynamically obtained. At the same time, the vehicle battery capacity distribution data is obtained by using the V2X communication protocol. It should be noted that the specific range within this range is set according to needs, and the initial value is 5KM;

[0072] The battery capacity distribution data specifically includes the average battery capacity V sales ; the average battery capacity V sales is obtained by the sum of the battery capacities of all vehicles registered in the area / the total number of vehicles;

[0073] For periodically receiving extreme weather warning information issued by the meteorological department, specifically obtain 72-hour extreme weather warning information through the local meteorological department; it should be noted that it specifically includes temperature, humidity, wind speed, and precipitation data.

[0074] As an optional embodiment, the specific working steps of the demand prediction module are as follows:

[0075] For the demand prediction in the long term, first, according to the new energy vehicle sales data in the past three long periods in the area obtained, normalize the new energy vehicle sales data in the three long periods to the interval [0, 1], select 64 LSTM units, add a Dropout layer to reduce overfitting, connect a Dense layer, output the new energy vehicle sales data in the next long period, complete the construction of the model, input the normalized new energy vehicle sales data in the three long periods into the completed model, and output the new energy vehicle sales data P in the next long period new ;

[0076] Then, according to the formula calculate the basic demand prediction Qh for each long period L ; it should be noted that it refers to the total demand for charging piles predicted in the long term based on factors such as the sales growth of new energy vehicles, vehicle battery capacity, and grid expansion, with the purpose of predicting future charging demand, evaluating market development trends, and optimizing resource allocation. The long-term demand Q L is an indicator that comprehensively considers factors such as the sales growth of new energy vehicles, vehicle battery capacity, and grid expansion, and is used to guide the long-term planning and construction of charging piles;

[0077] where Δ grid is the impact of grid expansion; for ΔG grid , according to the grid expansion planning data, determine the scale of grid expansion in the future long term, including information such as the expansion capacity and expansion area, and calculate the impact ΔG of grid expansion on charging demand according to the grid expansion scale and market conditions grid , for example, if the power supply capacity of the charging pile can be increased by 20% after grid expansion, then ΔG grid = 0.2;

[0078] Grid expansion can improve the power supply capacity of charging piles, thereby increasing charging demand. If the charging power of the charging piles can be increased and the charging time can be shortened after grid expansion, this will attract more new energy vehicle users to choose charging, thus increasing charging demand;

[0079] Then subtract the basic demand forecast Qh for each long-term period L from the actual demand for each long-term period to obtain the historical residual ∈1;

[0080] Select the MLP neural network model, divide all the obtained historical residual ∈1 data into a training set, a validation set and a test set, and divide them according to the ratio of 70%, 15%, 15%. Train to obtain a residual model, and bring the historical residual ∈1 into the residual model to obtain the residual ∈ for the next long-term period;

[0081] Then according to the formula calculate the demand forecast Q for the long-term period L ; It should be noted that by calculating the basic demand forecast, the future demand for charging piles can be initially estimated. Then, calculate the historical residual, that is, the difference between the actual charging demand and the basic forecast. This step helps to identify the deficiencies in the basic model prediction. Using the intermediate output, external dynamic features and time series derivative features of the basic model, train a residual prediction model. By inputting the relevant features of future time points into the trained residual model, predict the possible residuals in the future, so as to dynamically correct the basic demand forecast. Finally, add the predicted residual back to the basic demand forecast to obtain a more accurate and comprehensive total demand forecast for charging piles. This series of steps not only improves the accuracy of the prediction, but also optimizes the resource allocation;

[0082] It should be noted that since the proportion of new energy vehicles in the region has been changing during the long-term period, along with local policies or the price of new energy vehicles, the proportion of new energy vehicles in the region may change greatly during the long-term period. If the long-term demand cannot be predicted, because the installation of charging piles in the charging station takes time, there will be a shortage of charging piles.

[0083] As an optional embodiment, the specific working steps of the demand forecasting module further include:

[0084] For the demand forecast of the medium-term period, according to the formula

[0085]

[0086] calculate the medium-term demand Q M , where ρ is the regional economic vitality factor, S seasonis the seasonal fluctuation coefficient, A is the radiation capacity value of the charging station, μ is the distance attenuation coefficient, and R is the number of charging piles within the range of the charging station; it should be noted that since the sensing module is also used to dynamically obtain the growth data of new energy vehicles in the area, the above calculation is based on the charging pile where the sensing module is initially located, and the data within a surrounding range of 5 km;

[0087] It should also be noted that through multi-dimensional data fusion and dynamic modeling, the prediction accuracy and practicality have been significantly improved, providing strong support for the planning, operation, and decision-making of the charging pile supply chain. By integrating multi-dimensional data such as economy, transportation, and policies, the limitations of traditional single data sources are broken through. Through parameter dynamic adjustment and policy sensitivity modeling, the self-optimization of the prediction model is realized, quantifying the radiation effect and distance attenuation law of the charging pile, and supporting refined resource layout;

[0088] where e -μd is a distance attenuation factor, used to describe the phenomenon that the charging demand gradually weakens as the distance increases. d is the distance attenuation coefficient of the area where the charging station is located within the distance area. The value is obtained by calculating the ratio of the total charging volume of each area in the area to the total charging volume of the area where the charging station is located in the previous quarter. This ratio represents how many times the charging demand in the area where the charging station is located in the area is that of other areas. Take the natural logarithm of the ratio. The natural logarithm reflects the exponential attenuation relationship of the charging demand with distance. Divide the natural logarithm result by the marginal distance. μ represents the change rate of the charging demand per unit distance, that is, for every additional kilometer, by what proportion the charging demand will decrease; it should be noted that the larger μ is, which describes the speed at which the charging demand decreases as the distance increases, the faster the charging demand decreases with distance; conversely, it decreases slowly;

[0089] It should be noted that for large areas, the demands in each area within the area are different. Although charging stations will be installed in each area within the area, in some areas, such as parking lots, the demand is greater and more charging piles are required. If fixed charging piles are installed in each area, the cost will increase. Therefore, a small number of movable energy storage charging piles need to be configured. In special cases, according to the predicted demand, they can be moved to the appropriate area to meet the usage demand. By calculating the demand prediction for the medium-term time period within the area, it can better provide a reference for the subsequent allocation of movable energy storage charging piles.

[0090] As an optional embodiment, the specific working steps of the demand prediction module further include:

[0091] For the regional economic vitality factor ρ, Obtained, where 1 is the annual GDP of the city where the region is located, 2 is the total annual GDP of the country, N1 is the number of newly registered vehicles in the city where the region is located during the medium-term, and N2 is the total number of vehicles in the city where the region is located; it should be noted that it reflects the regional economic vitality and the popularity of new energy vehicles, so as to predict the potential of the charging market;

[0092] For the seasonal fluctuation coefficient S season , it is obtained by fitting based on the charging volume data of the same period in the past three years. Specifically, Obtained, where t represents the month; it should be noted that it captures the seasonal changes in charging demand and improves the accuracy of prediction;

[0093] For the charging station radiation capacity value A, according to the formula Obtained, where P t is the average daily passenger flow within the range of the charging station, D p is the charging pile density, N b is the number of bus lines near the region. It should be noted that P t is the average daily passenger flow within the range of the charging station, this range is 5 km around the charging pile, and the number of bus lines near the region is specifically the number of bus lines within 1 km near the region.

[0094] As an optional embodiment, the specific working steps of the demand prediction module further include:

[0095] For the demand prediction of the short-term period, the specific method is;

[0096]

[0097] where ΔT is the difference between the current temperature and the standard temperature, R r is the current rainfall intensity, W S is the current wind speed, ΔC is the battery capacity attenuation rate, obtained by subtracting the current battery capacity from the initial battery capacity and then dividing by the initial battery capacity, and X is the traffic congestion value, obtained by Obtained, where TL is the traffic index, is the baseline value of the charging demand, that is, the basic level of the charging demand in the absence of other influencing factors; it should be noted that it improves the prediction accuracy:

[0098] By considering actual influencing factors such as weather, battery degradation, and traffic conditions, the formula can more accurately reflect the changes in charging demand. Each factor is adjusted according to the current conditions, enabling the prediction to respond to environmental changes in real time. More accurate demand prediction helps optimize the use of charging infrastructure, ensuring sufficient charging resources at high-demand times and locations. By avoiding overcharging at night during low-demand periods and ensuring sufficient capacity during high-demand periods, the return on investment can be increased, and sufficient charging capacity can be ensured in case of a surge in demand such as in bad weather or traffic congestion;

[0099] It should be noted that the traffic index can be obtained through map software.

[0100] As an optional embodiment, the specific working steps of the dynamic optimization module are as follows:

[0101] Every other cycle, calculate the prediction error E of the short-term demand prediction, specifically according to the formula: Calculate and obtain the prediction error E, where Q1 is the actual charging demand value. If E is greater than 15%, then adjust the weight coefficients 0.3, 0.3, and 0.4 in. When E is greater than 0, adjust 0.3, 0.3, and 0.4 to increase by 2%. If E is less than 0, adjust 0.3, 0.3, and 0.4 to decrease by 2%. It should be noted that through the error feedback mechanism, the model can monitor the prediction error in real time and dynamically adjust the weight coefficients according to the error situation. When the prediction error continuously exceeds a certain threshold, the model will trigger an automatic update to ensure the accuracy and reliability of the prediction. This method helps improve the accuracy of charging demand prediction and optimize the operation and management of charging stations.

[0102] As an optional embodiment, the specific working steps of the dynamic optimization module further include the following:

[0103] For the long term, according to the formula calculate the number of new charging piles N to be added, and according to the number of new charging piles N, in the next long-term period, install N charging piles in this area, where P x is the power of a single charging pile, and T1 is the average working time of the charging pile; it should be noted that the average working time of the charging pile is generally the working time of the area, that is, 8 hours;

[0104] In the medium term, obtain the real-time power reserve data E and the safe power threshold E1 of each charging pile in the area from the perception module, and obtain the demand predictions Q L 、Q M and Q S for each charging pile in the short term, medium term, and long term from the demand prediction module;

[0105] Calculate the inventory monitoring index H for each area. When the inventory monitoring index H is greater than the preset threshold, move the movable charging pile to that area;

[0106] In the short - term period, according to the formula N j =P L *(1 + 0.1*Q L ), calculate the dynamic time - of - use pricing N j of the charging pile in the short - term period, where P L is the basic electricity price.

[0107] It should be noted that through long - term, medium - term and short - term optimization measures, it is possible to comprehensively respond to the changes in the number of new energy vehicles in the area, the demand differences in different areas and the actual influencing factors, ensuring the reasonable allocation and efficient use of charging piles;

[0108] According to the long - term demand forecast, determine the number of new charging piles to be added and install the corresponding number of charging piles in the next long - term period; according to the medium - term demand forecast, monitor the power reserve and demand of each charging pile in the area in real time and dynamically adjust the distribution of charging piles;

[0109] According to the short - term demand forecast, calculate the dynamic time - of - use pricing of the charging pile to guide users to use the charging pile reasonably at different times;

[0110] It can improve the utilization rate of charging piles, reduce the investment cost, ensure sufficient charging resources at high - demand times and locations, and enhance the user experience.

[0111] As an optional embodiment, the specific steps for calculating the inventory monitoring index H of the charging pile are as follows:

[0112] Through H=(E - E1)+(Q L *0.2)+(Q M *0.1)+(Q S *0.02), calculate and obtain the inventory monitoring index H of each charging pile. It should be noted that by considering the predicted charging demands in the short - term, medium - term and long - term, the electric energy resources can be reasonably allocated. During the peak demand period, the power reserve can be increased in advance to avoid charging interruptions caused by insufficient electric energy. During the low - demand period, the power reserve can be reduced to lower the operating cost. By optimizing the power reserve, the resource utilization efficiency of the charging pile can be improved, and electric energy waste can be reduced. By replenishing the power reserve in a timely manner, the continuity of the charging service of the charging pile can be ensured, the user satisfaction can be improved. By reasonably arranging the power reserve, the waiting time of users due to insufficient electric energy of the charging pile can be reduced, and the user experience can be improved.

[0113] Working principle

[0114] It includes a perception module, which is used to collect the operation data of the charging pile in real time, specifically including the average daily charging times of a single pile and the peak power duration;

[0115] The perception module is also used to dynamically obtain the growth data of new energy vehicles in the area and obtain the vehicle battery capacity distribution data;

[0116] It is also used to periodically receive the extreme weather warning information released by the meteorological department, and the perception module transmits all the data;

[0117] It further includes a demand prediction module, which is used to receive the data of the perception module and obtain the demand prediction result after the prediction period according to the data of the perception module. The demand prediction result specifically includes decomposing the prediction period into three time periods: long term, medium term and short term, and obtaining the demand prediction for the long term period, the demand prediction for the medium term period and the demand prediction for the short term period;

[0118] It also includes a dynamic optimization module, which is used to receive the data of the perception module and the demand prediction result of the demand prediction module, and perform corresponding dynamic optimization according to the demand prediction for the long term period, the demand prediction for the medium term period and the demand prediction for the short term period in the above demand prediction result.

[0119] The above is only the preferred implementation mode of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the idea of the present invention belong to the protection scope of the present invention. It should be pointed out that for those ordinary technical staff in the technical field, several improvements and refinements made without departing from the principle of the present invention should also be regarded as the protection scope of this template.

Claims

1. A supply chain demand forecasting and dynamic optimization system based on artificial intelligence, characterized in that: It includes a sensing module, which is used to collect the operation data of the charging pile in real time, specifically including the average daily charging times of a single pile and the peak power duration; The sensing module is also used to dynamically obtain the growth data of new energy vehicles in the region and obtain the vehicle battery capacity distribution data; It is also used to periodically receive extreme weather warning information issued by the meteorological department, and the sensing module transmits all data; It also includes a demand forecasting module, which is used to receive the data of the sensing module and obtain the demand forecasting result after the forecasting period according to the data of the sensing module. The demand forecasting result specifically includes decomposing the forecasting period into three periods: long-term, medium-term and short-term, and obtaining the demand forecast for the long-term period, the demand forecast for the medium-term period and the demand forecast for the short-term period; It also includes a dynamic optimization module, which is used to receive the data of the perception module and the demand forecast results of the demand forecast module, and perform corresponding dynamic optimization based on the long-term demand forecast, medium-term demand forecast and short-term demand forecast in the above demand forecast results.

2. The supply chain demand forecasting and dynamic optimization system based on artificial intelligence according to claim 1 is characterized in that: The specific working steps of the perception module are as follows: For real-time collection of charging pile operation data, specifically through sensors and data acquisition equipment installed on the charging pile, real-time acquisition of charging pile operation data, including the average daily charging times Q of a single pile and the peak power duration T; The specific acquisition steps include, first, using the data acquisition device of the charging pile to record the start and end time of each charging in real time. Whenever a vehicle starts charging, the system will generate a charging record, including the charging start time, end time and other information. At the end of the statistical period, the system will count the number of charging times Q of the charging pile in that day; A power sensor is installed in the charging pile to monitor the power output during the charging process in real time. The control system of the charging pile will record the data of power changes over time. When the power reaches the peak, the timing will start and the timing will stop when the power drops below the peak, thereby obtaining the duration T of the peak power.

3. The supply chain demand forecasting and dynamic optimization system based on artificial intelligence according to claim 1 is characterized in that: The specific working steps of the perception module also include: By sharing data with the vehicle management system, the growth data of new energy vehicles in the range can be dynamically obtained. At the same time, the vehicle battery capacity distribution data can be obtained using the V2X communication protocol; The battery capacity distribution data specifically includes the average battery capacity V sales ; Average battery capacity V sales Obtained by the sum of the battery capacities of all vehicles registered in the region / total number of vehicles; For periodic receipt of extreme weather warning information issued by the meteorological department, specifically obtain 72-hour extreme weather warning information through the local meteorological department.

4. The supply chain demand forecasting and dynamic optimization system based on artificial intelligence according to claim 1 is characterized in that: The specific working steps of the demand forecasting module are: For long-term demand forecasting, firstly, based on the sales data of new energy vehicles in the past three long-term periods in the region, the sales data of new energy vehicles in the three long-term periods are normalized to the interval [0, 1], 64 LSTM units are selected, a Dropout layer is added to reduce overfitting, a Dense layer is connected, and the sales data of new energy vehicles in the next long-term period is output to complete the construction of the model. The normalized sales data of new energy vehicles in the three long-term periods are input into the completed model, and the sales data of new energy vehicles in the next long-term period P is output. new ; Then according to the formula Calculate the basic demand forecast Qh for each long-term period L , where Δ grid Impact of power grid expansion; Then the basic demand forecast Qh for each long-term period is L Subtract the actual demand in each long-term period to obtain the historical residual ∈ 1; Select the MLP neural network model, divide all the historical residual ∈1 data into training set, validation set and test set, and divide them in the proportion of 70%, 15%, and 15%. Train to get the residual model, bring the historical residual ∈1 into the residual model, and get the residual ∈ of the next long-term period; Then according to the formula Calculate the long-term demand forecast Q L .

5. The supply chain demand forecasting and dynamic optimization system based on artificial intelligence according to claim 4 is characterized in that: The specific working steps of the demand forecasting module also include: For the demand forecast in the medium term, according to the formula Calculate the medium-term demand Q M , where ρ is the regional economic vitality factor, S season is the seasonal fluctuation coefficient, A is the radiation capacity value of the charging station, μ is the distance attenuation coefficient, and R is the number of charging piles within the range of the charging station; where e -μd It is a distance attenuation factor, which is used to describe the phenomenon that the charging demand gradually weakens with the increase of distance. d is the distance attenuation coefficient of the area where the charging station is located in the distance area.

6. The supply chain demand forecasting and dynamic optimization system based on artificial intelligence according to claim 5 is characterized in that: The specific working steps of the demand forecasting module also include: For the regional economic vitality factor ρ, Get, where 1 is the annual GDP of the city where the region is located, 2 is the annual total GDP of the country, N1 is the number of newly registered cars in the city where the region is located in the medium term, and N2 is the total number of cars in the city where the region is located; For the seasonal fluctuation coefficient S season , based on the charging data of the past three years, specifically, Get, where t represents the month; For the charging station radiation capacity value A, according to the formula Get, where P t is the average daily passenger flow within the charging station, D p is the density of charging piles, N b is the number of bus routes near the area.

7. The supply chain demand forecasting and dynamic optimization system based on artificial intelligence according to claim 4 is characterized in that: The specific working steps of the demand forecasting module also include: For short-term demand forecasting, the specific method is as follows; Where ΔT is the difference between the current temperature and the standard temperature, R r is the current rainfall intensity, W S is the current wind speed, ΔC is the battery capacity attenuation rate, which is obtained by subtracting the current battery capacity from the initial battery capacity and dividing it by the initial battery capacity, and X is the traffic congestion value, which is obtained by Get, where TL is the traffic index, It is the benchmark value of charging demand, that is, the basic level of charging demand in the absence of other influencing factors.

8. The supply chain demand forecasting and dynamic optimization system based on artificial intelligence according to claim 7 is characterized in that: The specific working steps of the dynamic optimization module include the following: Every other cycle, the forecast error E of the short-term demand forecast is calculated according to the formula: The prediction error E is calculated, where Q1 is the actual charging demand value. If E is greater than 15%, adjust The weight coefficients are 0.3, 0.3 and 0.

4. When E is greater than 0, adjust 0.3, 0.3 and 0.4 to increase by 2%. If E is less than 0, adjust 0.3, 0.3 and 0.4 to decrease by 2%.

9. The supply chain demand forecasting and dynamic optimization system based on artificial intelligence according to claim 8 is characterized in that: The specific working steps of the dynamic optimization module also include the following: For long-term periods, according to the formula Calculate the number of newly added charging piles N, and according to the number of newly added charging piles N, in the next long-term period, the number of charging piles installed in this area is N, where P x is the power of a single charging pile, T1 is the average working time of the charging pile; In the medium term, the real-time power reserve data E and the safety power threshold E1 of each charging pile in the region are obtained from the perception module, and the demand forecast Q of each charging pile in the short term, medium term and long term is obtained from the demand forecast module. L , Q M and Q S ; The inventory monitoring index H of each area is calculated, and when the inventory monitoring index H is greater than a preset threshold, the movable charging pile is moved to the area; In the short term, according to the formula N j =P L *(1+0.1*Q L ), calculate the dynamic time-sharing pricing N of the charging pile in the short term j , where P L The basic electricity price.

10. The supply chain demand forecasting and dynamic optimization system based on artificial intelligence according to claim 9, characterized in that: The specific steps of calculating the inventory monitoring index H of the charging pile are: By H = (E-E1) + (Q L *0.2)+(Q M *0.1)+(Q S *0.02), and calculate and obtain the inventory monitoring index H of each charging pile.

Images