Single-station dynamic pricing method and system based on market transaction and light storage power generation

Abstract

The invention discloses a single-station dynamic pricing method and system based on market transaction and light storage power generation, relates to the technical field of single-station dynamic pricing, and solves the technical problems that it is difficult to predict an optimal charging curve according to weather and a power market, and it is difficult to set dynamic pricing for balancing users and operators for a single station. The method comprises the following steps of: setting a self-adaptive time period and a user price sensitive factor of a single station; determining the output power of the photovoltaic panel in the next time period based on the surrounding environment data in the next time period; determining an ideal charging curve of the single station in the next time period based on the market electricity price, the storage electric quantity and the output power; setting a charging price curve of the next time period based on the ideal charging curve; optimizing the charging price curve through the user price sensitive factor to determine a final electricity price curve; according to the invention, the income capability of the single station in the electricity market can be improved.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent regulation of electricity prices and relates to a single-station dynamic pricing technology, specifically a single-station dynamic pricing method and system based on market transactions and photovoltaic storage power generation. Background Art

[0002] Renewable energy has become a key direction for current and future development. In the current power system, with the rapid development of new energy, particularly photovoltaic power generation, and the increasing popularity of electric vehicles, photovoltaic charging stations have become a crucial component. A single-station charging station integrates grid power supply, photovoltaic power generation, and energy storage systems. By leveraging multi-energy complementarity to achieve efficient energy utilization and grid interaction, it is a crucial infrastructure for new power systems and green mobility.

[0003] At present, most single-station dynamic pricing methods and systems based on market transactions and photovoltaic storage power generation find it difficult to predict the optimal charging curve based on weather and electricity market, resulting in insufficiently optimized charging strategies, inability to maximize the utilization of photovoltaic power generation, and reduced overall energy utilization efficiency; at the same time, most single-station dynamic pricing methods and systems based on market transactions and photovoltaic storage power generation find it difficult to set dynamic pricing for a single station that balances users and operators, which may cause users to bear higher electricity bills, increase dissatisfaction, and thus lose customers, and operators may miss potential profit opportunities due to unreasonable pricing.

[0004] Therefore, the present invention discloses a single-station dynamic pricing method and system based on market transactions and photovoltaic storage power generation, which are used to solve the above technical problems. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a single-station dynamic pricing method and system based on market transactions and photovoltaic storage power generation, which is used to solve the technical problems that it is difficult to predict the optimal charging curve based on weather and electricity market, and it is difficult to set dynamic pricing for a single station to balance users and operators. The present invention solves the above problems by setting an adaptive time period for a single station, obtaining the user price sensitivity factor of the single station and the output power of the photovoltaic panel in the next time period, obtaining the ideal charging curve of the single station in the next time period, setting the charging price curve for the next time period based on the ideal charging curve, and optimizing the charging price curve through the user price sensitivity factor to determine the final electricity price curve.

[0006] To achieve the above objectives, the first aspect of the present invention provides a single-station dynamic pricing method based on market transactions and photovoltaic power generation, comprising:

[0007] Based on the number of charging stations at multiple time points, at least one time period is set for the station, and the stored power of the station in the current time period, as well as the market electricity price and surrounding environmental data for the next time period, are obtained. The surrounding environmental data includes weather type, light intensity, and cloud thickness.

[0008] determining the output power of the photovoltaic panel in the next time period based on the ambient environment data in the next time period;

[0009] Determine the ideal charging curve for the next time period for the single station based on the market electricity price, stored electricity, and output power;

[0010] Set the charging price curve for the next time period based on the ideal charging curve;

[0011] The final electricity price curve is determined by optimizing the charging price curve using a user price sensitivity factor, where the user price sensitivity factor is determined by a user price sensitivity model constructed using historical data; wherein the historical data includes the number of people charging during low-price periods and the number of people charging during high-price periods.

[0012] Preferably, before setting at least one time period of a single station, the method further includes:

[0013] Obtain the number of charging stations at each time point in the history of several days, and mark the mode of the charging numbers at the time points as the analysis number at the corresponding time point;

[0014] The number of analyses at each time point is integrated into a quantity group, and it is determined whether the standard deviation of the quantity group is less than the defined threshold; if so, a day is divided into several time periods according to a fixed duration; if not, the time points corresponding to the number of analyses in the quantity group that is less than the percentile of the quantity group are marked as low-peak time points, and the non-low-peak time points in a day are marked as peak time points; wherein the defined threshold, fixed duration, and percentile of the percentile are all obtained through manual settings.

[0015] It should be noted that the charging quantity at each time point in several historical days has different charging quantities at the same time point, so there are several charging quantities at the same time point.

[0016] It should be noted that the charging quantity is the number of charging devices specified by a single station. For example, if the charging devices specified by a single station are electric vehicles, the charging quantity is the number of electric vehicles currently being charged in the single station.

[0017] Preferably, the setting of at least one time period for a single station includes:

[0018] Consolidate the off-peak time points into several off-peak periods, and determine the planned duration G1 of the off-peak period based on the average value P1 of the number of analysis points in the off-peak period; the planned duration G1 satisfies formula (1):

[0019]

[0020] Where DC is the fixed duration, BW is the percentile, α is the manually set amplitude adjustment coefficient, and the value range of α is (0,2];

[0021] Consecutive peak time points are integrated into several peak periods, and the planned duration G2 of the peak period is determined based on the average value P2 of the number of analysis points within the peak period; the planned duration G2 satisfies formula (2):

[0022]

[0023] The low-peak time periods that are shorter than the planned duration G1 and the peak time periods that are shorter than the planned duration G2 are merged into the smaller time periods adjacent on both sides. The ratio of each low-peak time period to the planned duration G1 is rounded up to obtain the planned number of the corresponding low-peak time period. The ratio of each peak time period to the planned duration G2 is rounded up to obtain the planned number of the corresponding peak time period. The corresponding low-peak time period or peak time period is evenly divided into several time periods according to the planned number.

[0024] Preferably, the step of determining the output power of the photovoltaic panel in the next time period based on the ambient environment data in the next time period includes:

[0025] The weather type and cloud thickness YH of the next time period are extracted, and the weather impact factor β is determined based on the weather type and cloud thickness YH; the weather impact factor β satisfies formula (3):

[0026]

[0027] Wherein, when j = 1, it indicates sunny, when j = 2, it indicates cloudy, and when j = 3, it indicates overcast; BYH is the artificially set standard cloud thickness, and RH is the humidity in the area where the single station is located in the next time period;

[0028] Extract the illumination intensity I of the next time period, and determine the modified light intensity ZI based on the illumination intensity; the modified light intensity ZI satisfies formula (4):

[0029] ZI=I·e -γ·YH (4);

[0030] Wherein, γ is the artificially set cloud attenuation coefficient, and the value range of γ is (0,2];

[0031] Determine the output power Cp of the photovoltaic panel at each time point in the next time period based on the weather influence factor β and the corrected light intensity ZI i The output power Cp i Satisfying formula (5):

[0032] CP i =δ·β·θ·A·ZI·(1-0.0045(HT-25) (5);

[0033] Among them, δ is the manually set attenuation factor of the photovoltaic panel, θ is the photoelectric conversion efficiency, i is the number of the time point; A is the effective light-receiving area of ​​the photovoltaic panel, and HT is the battery temperature in the next time period.

[0034] Preferably, determining the ideal charging curve for the next time period of a single station includes:

[0035] Obtain the market electricity price for the next time period, and determine the grid power supply ratio BS1 at each time point based on the market electricity price i ; The proportion coefficient BS1 i Satisfying formula (6):

[0036]

[0037] Wherein, SD is the market electricity price in the next time period, BSD is the manually set standard market electricity price; μ1 is the manually set proportional adjustment coefficient, and the value range of μ1 is (0,2];

[0038] Get the level S of the stored power and the output power CP at each time point in the next time period i , based on the level S of the stored power and the output power CP i Determine the BS2 ratio of photovoltaic and energy storage power supply at each time point i ; The proportion coefficient BS2 i Satisfying formula (7):

[0039]

[0040] Wherein, BCP is the manually set standard output power, and the level S is 0, 1, and 2; μ2 is the manually set proportional adjustment coefficient, and the value range of μ2 is (0, 2];

[0041] Based on the proportion coefficient BS1 at each time point i , proportion coefficient BS2 i and output power CP i Determine the charging power CD at each time point i ; The charging power CD i Satisfy formula (8):

[0042]

[0043] Wherein, DP is the grid power supply;

[0044] The charging power CD at each time point i As the vertical axis, each time point as the horizontal axis, use the interpolation method to establish the ideal charging curve.

[0045] Preferably, setting the charging price curve for the next time period based on the ideal charging curve includes:

[0046] Extract the standard electricity price BE corresponding to the manually set standard charging power BD, and based on the standard charging power BD, the standard electricity price BE and the charging power CD at each time point i Get the reference electricity price CE at each time point i ; The reference electricity price CE i Satisfy formula (9):

[0047]

[0048] Among them, ρ is the price adjustment coefficient set manually and greater than 0;

[0049] Use each reference electricity price CE i The charging price curve is obtained by replacing the vertical coordinate value corresponding to each time point in the ideal charging curve.

[0050] Preferably, the step of optimizing the charging price curve by using the user price sensitivity factor to determine the final electricity price curve includes:

[0051] Extract the user price sensitivity factor HZ, and obtain the price adjustment coefficient DS based on the user price sensitivity factor HZ; the price adjustment coefficient DS satisfies formula (10):

[0052] DS=e -σ·HZ / BHZ (10);

[0053] Wherein, σ is the manually set amplitude adjustment coefficient, and the value range of σ is [0,1];

[0054] The final electricity price curve is obtained by multiplying the price adjustment coefficient DS by the vertical coordinate value of each time point in the charging price curve.

[0055] Preferably, the step of obtaining the stored electricity amount of a single station in the current time period, as well as the market electricity price and surrounding environment data for the next time period, includes:

[0056] The market electricity price of a single station in the next time period is obtained from the power company's website, the stored electricity of a single station is extracted from the database, and the weather type, light intensity and cloud thickness in the next time period in the area where the single station is located are obtained from the weather forecast website; among them, the weather types include sunny, cloudy and overcast.

[0057] It should be noted that if the market electricity price for the next time period is not unique, the market electricity price for the next time period is obtained by performing a proportional calculation based on the proportion of the time period of each market electricity price to the total time of the next time period.

[0058] Preferably, the user price sensitivity factor is determined by a user price sensitivity model constructed through historical data, including:

[0059] Extract the number of chargers during low-price periods, the number of chargers during high-price periods, and the user price sensitivity factor from the historical data of each single station. The historical data includes the number of chargers during low-price periods and the number of chargers during high-price periods, as well as the user price sensitivity factor set by experts based on the number of chargers during low-price periods and the number of chargers during high-price periods.

[0060] The number of people charging during low-price periods, the number of people charging during high-price periods, and the user price sensitivity factor are integrated into several sets of training data and test data; the artificial intelligence model is trained using the training data, and the trained artificial intelligence model is tested using the test data, and the artificial intelligence model is adjusted based on the test results; ultimately, a user price sensitivity model is obtained, whose input is the number of people charging during low-price periods and the number of people charging during high-price periods in the historical data of a single station, and the output is the user price sensitivity factor of the single station; the artificial intelligence model includes a BP neural network model and an RBF neural network model;

[0061] The number of charging users during low-price periods and high-price periods in the historical data of the current single station are input into the user price sensitivity model to obtain the user price sensitivity factor of the current single station.

[0062] It should be noted that the low-price period is the time period when the average price of a single station within the period is lower than the manually set low-price threshold, and the high-price period is the time period when the average price of a single station within the period is higher than the manually set high-price threshold.

[0063] The second aspect of the present invention provides a single-station dynamic pricing method and system based on market transactions and photovoltaic storage power generation, characterized by comprising: an intelligent analysis module, and a data collection module and a dynamic pricing module connected thereto;

[0064] The data collection module is used to set at least one time period for a single station based on the number of charging operations at multiple time points, and obtain the stored electricity volume of the single station in the current time period, as well as the market electricity price and surrounding environmental data for the next time period; wherein the surrounding environmental data includes weather type, light intensity, and cloud thickness;

[0065] The intelligent analysis module is used to determine the output power of the photovoltaic panel in the next time period based on the ambient environment data in the next time period; and to determine the ideal charging curve for the next time period of the single station based on the market electricity price, the amount of stored electricity and the output power;

[0066] The dynamic pricing module is used to set the charging price curve for the next time period based on the ideal charging curve; the final electricity price curve is determined by optimizing the charging price curve using a user price sensitivity factor, where the user price sensitivity factor is determined by a user price sensitivity model constructed using historical data; the historical data includes the number of users charging during low-price periods and the number of users charging during high-price periods;

[0067] The database is used to store data.

[0068] Compared with the prior art, the present invention has the following beneficial effects:

[0069] 1. The present invention sets an adaptive time period and a user price sensitivity factor for a single station to obtain the stored electricity in the current time period of the single station, as well as the market electricity price and surrounding environmental data for the next time period; determines the output power of the photovoltaic panel for the next time period based on the surrounding environmental data for the next time period; determines the ideal charging curve for the next time period of the single station based on the market electricity price, the stored electricity, and the output power; sets the charging price curve for the next time period based on the ideal charging curve; and optimizes the charging price curve using the user price sensitivity factor to determine the final electricity price curve, thereby solving the technical problems of difficulty in predicting the optimal charging curve based on weather and the electricity market, and difficulty in setting dynamic pricing for a single station that balances users and operators; the present invention can improve the profitability of a single station in the electricity market.

[0070] 2. This invention achieves the coordinated optimization of the power grid and renewable energy by integrating the power supply ratio of the grid, the power supply ratio of photovoltaic and energy storage, and the real-time output power, constructs a multi-dimensional dynamic model, and ultimately generates an optimal charging power curve that reflects the balance between economy and power supply.

[0071] 3. This invention achieves an ideal charging curve through economic optimization of the photovoltaic and energy storage output in conjunction with electricity prices. This not only considers electricity prices but also simultaneously integrates the photovoltaic and energy storage output power and energy storage status. For example, during the midday hours, when sunlight is abundant and electricity prices are high, the system prioritizes photovoltaic power generation while reducing grid charging power to match user demand flexibility. During the early morning hours, when electricity prices are low and energy storage is sufficient, the system can simultaneously utilize both the grid and energy storage for charging, shortening charging time and reducing marginal costs. This multi-objective optimization strategy enables a single station to serve as both an "electricity consumer" and an "electricity seller," enhancing its profitability in the electricity market. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0073] Figure 1 Schematic diagram of the operating steps of the present invention;

[0074] Figure 2 Schematic diagram of the system module of the present invention;

[0075] Figure 3 Schematic diagram of the operating steps for setting a single station time period in the present invention. DETAILED DESCRIPTION

[0076] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0077] See also Figure 1 The first embodiment of the present invention provides a single-station dynamic pricing method based on market transactions and photovoltaic power generation, including:

[0078] A single-station dynamic pricing method based on market transactions and photovoltaic storage power generation is characterized by including:

[0079] Based on the number of charging stations at multiple time points, at least one time period is set for the station, and the stored power of the station in the current time period, as well as the market electricity price and surrounding environmental data for the next time period, are obtained. The surrounding environmental data includes weather type, light intensity, and cloud thickness.

[0080] determining the output power of the photovoltaic panel in the next time period based on the ambient environment data in the next time period;

[0081] Determine the ideal charging curve for the next time period for the single station based on the market electricity price, stored electricity, and output power;

[0082] Set the charging price curve for the next time period based on the ideal charging curve;

[0083] The final electricity price curve is determined by optimizing the charging price curve using a user price sensitivity factor, where the user price sensitivity factor is determined by a user price sensitivity model constructed using historical data; wherein the historical data includes the number of people charging during low-price periods and the number of people charging during high-price periods.

[0084] See also Figure 3 , before setting at least one time period of a single station in this application, it also includes:

[0085] Obtain the number of charging stations at each time point in the history of several days, and mark the mode of the charging numbers at the time points as the analysis number at the corresponding time point;

[0086] The number of analyses at each time point is integrated into a quantity group, and it is determined whether the standard deviation of the quantity group is less than the defined threshold; if so, a day is divided into several time periods according to a fixed duration; if not, the time points corresponding to the number of analyses in the quantity group that is less than the percentile of the quantity group are marked as low-peak time points, and the non-low-peak time points in a day are marked as peak time points; among them, the defined threshold, fixed duration, and percentile of the percentile are all obtained through manual settings.

[0087] It is worth noting that the present invention uses percentiles when calculating characteristic values. This value is added because the system can set different percentiles according to needs, so that the obtained low-peak time points and peak time points can adapt to different situations, increasing the flexibility of the system.

[0088] It should be noted that the charging quantity at each time point in several historical days has different charging quantities at the same time point, so there are several charging quantities at the same time point.

[0089] It should be noted that the charging quantity is the number of charging devices specified by a single station. For example, if the charging devices specified by a single station are electric vehicles, the charging quantity is the number of electric vehicles currently being charged in the single station.

[0090] It should be noted that the time length occupied by the time point in the present invention is manually set. For example, the time length occupied by the time point can be selected as: 1s, 1min or 5min.

[0091] It should be noted that in the present invention, the percentile is the value at a specific percentile in the quantity group; for example, if the quantity group data is sorted from small to large, if the manually set percentile is 60%, then the data at the 60% position of the quantity group is the percentile of the group. If there is no data at the 60% position of the group, the data closest to the 60% position is used as the percentile of the group.

[0092] In this application, at least one time period for a single station is set, including:

[0093] Consolidate the off-peak time points into several off-peak periods, and determine the planned duration G1 of the off-peak period based on the average value P1 of the number of analysis points in the off-peak period; the planned duration G1 satisfies formula (1):

[0094]

[0095] Where DC is the fixed duration, BW is the percentile, α is the manually set amplitude adjustment coefficient, and the value range of α is (0,2];

[0096] Consecutive peak time points are integrated into several peak periods, and the planned duration G2 of the peak period is determined based on the average value P2 of the number of analysis points within the peak period; the planned duration G2 satisfies formula (2):

[0097]

[0098] The low-peak time periods that are shorter than the planned duration G1 and the peak time periods that are shorter than the planned duration G2 are merged into the smaller time periods adjacent on both sides. The ratio of each low-peak time period to the planned duration G1 is rounded up to obtain the planned number of the corresponding low-peak time period. The ratio of each peak time period to the planned duration G2 is rounded up to obtain the planned number of the corresponding peak time period. The corresponding low-peak time period or peak time period is evenly divided into several time periods according to the planned number.

[0099] It should be noted that the low-peak time period that is less than the planned duration G1 and the peak time period that is less than the planned duration G2 are merged into the smaller time periods adjacent on both sides. The merging operation is carried out on the low-peak time period and the peak time period that is less than the corresponding planned duration in sequence according to the time order, and the name of the merged time period is the name of the time period to be merged; when a low-peak time period or peak time period that is less than the corresponding planned duration is merged by the previous time period, if the two merged time periods are still less than the corresponding planned duration, the merging operation will continue to be carried out on this merged time period; if the two merged time periods are not less than the corresponding planned duration, there is no need to continue to merge this merged time period.

[0100] It should be noted that the name of the merged time period is the name of the time period of the merged party. For example, if the off-peak time period 6 is merged into the peak time period 9, the off-peak time period 6 is the merging party, the peak time period 9 is the merged party, and the name of the merged time period is peak time period 9.

[0101] It should be noted that the off-peak period shorter than the planned duration G1 and the peak period shorter than the planned duration G2 are merged into the adjacent smaller periods on both sides. For example:

[0102] Example 1: If the consecutive time periods from front to back are off-peak period 1, peak period 1, and off-peak period 2; among them, peak period 1 is less than the planned duration G2, off-peak period 1 and off-peak period 2 are not less than the planned duration G1, and off-peak period 1 is less than off-peak period 2; then peak period 1 will be merged into off-peak period 1, and the name of the merged time period will be off-peak period 1.

[0103] Example 2: If the consecutive time periods from front to back are peak period 2, off-peak period 3, peak period 3, and off-peak period 4; among them, peak period 2 is not less than the planned duration G2, peak period 3 is less than the planned duration G2, off-peak period 3 is less than the planned duration G1, and off-peak period 4 is not less than the planned duration G1; then off-peak period 3 is merged into peak period 3, and the name of the merged period is peak period 3.

[0104] It should be noted that α is an manually set amplitude adjustment coefficient, which is used to adjust the influence of the average value P1 on the planned duration G1 during the off-peak period and the influence of the average value P2 on the planned duration G2 during the peak period; when other conditions remain unchanged, the larger α is, the greater the impact on the planned duration G1 and the planned duration G2, and the smaller α is, the smaller the impact on the planned duration G1 and the planned duration G2.

[0105] It should be noted that the ratio of each off-peak period to the planned duration G1 is rounded up to obtain the planned number of the corresponding off-peak period, and the ratio of each peak period to the planned duration G2 is rounded up to obtain the planned number of the corresponding peak period. This can be explained by the formula: Among them, DT is the duration of the off-peak period, H1 is the planned number corresponding to the off-peak period; GT is the duration of the peak period, H2 is the planned number corresponding to the peak period.

[0106] This application obtains the storage capacity of a single station in the current time period, as well as the market electricity price and surrounding environmental data for the next time period, including:

[0107] The market electricity price of a single station in the next time period is obtained from the power company's website, the stored electricity of a single station is extracted from the database, and the weather type, light intensity and cloud thickness in the next time period in the area where the single station is located are obtained from the weather forecast website; among them, the weather types include sunny, cloudy and overcast.

[0108] It should be noted that if the market electricity price for the next time period is not unique, the market electricity price for the next time period will be obtained by proportional calculation based on the proportion of the time period of each market electricity price to the total time of the next time period. For example, the next time period is 17:00-19:00 on March 20, 2025, and the market electricity price for the next time period is 0.6 yuan / kWh from 17:00 to 18:00 and 0.65 yuan / kWh from 18:00 to 19:00. The market electricity price from 17:00 to 19:00 on March 20, 2025 is: 0.6×0.5+0.65×0.5=0.625 yuan / kWh.

[0109] In this application, the output power of the photovoltaic panel in the next time period is determined based on the ambient environment data in the next time period, including:

[0110] The weather type and cloud thickness YH of the next time period are extracted, and the weather impact factor β is determined based on the weather type and cloud thickness YH; the weather impact factor β satisfies formula (3):

[0111]

[0112] Wherein, when j = 1, it indicates sunny, when j = 2, it indicates cloudy, and when j = 3, it indicates overcast; BYH is the artificially set standard cloud thickness, and RH is the humidity in the area where the single station is located in the next time period;

[0113] Extract the illumination intensity I of the next time period, and determine the modified light intensity ZI based on the illumination intensity; the modified light intensity ZI satisfies formula (4):

[0114] ZI=I·e -γ·YH (4);

[0115] Wherein, γ is the artificially set cloud attenuation coefficient, and the value range of γ is (0,2];

[0116] Determine the output power CP of the photovoltaic panel at each time point in the next time period based on the weather influence factor β and the corrected light intensity ZI i ; The output power CP i Satisfying formula (5):

[0117] CP i =δ·β·θ·AZI·(1-0.0045(HT-25)(5);

[0118] Among them, δ is the manually set attenuation factor of the photovoltaic panel, θ is the photoelectric conversion efficiency, i is the number of the time point; A is the effective light-receiving area of ​​the photovoltaic panel, and HT is the battery temperature in the next time period.

[0119] It is worth noting that the present invention establishes a multi-parameter coupled prediction model by constructing a multi-dimensional environmental parameter system of weather type, cloud thickness, ambient humidity, light intensity, and battery temperature. Compared with the traditional prediction method that relies solely on light intensity, this solution takes into account the differentiated impact mechanism of environmental parameters on photovoltaic output under different weather types. For example, in cloudy weather, a cloud thickness dynamic compensation model is used. When j = 2 in formula (3), the standard cloud thickness BYH is introduced as the reference value for linear correction; on cloudy days, a humidity compensation factor is introduced. The RH / 100 term with j = 3 in formula (3) fully reflects the light transmission loss of the photovoltaic panel due to water vapor. This parameter classification process enables the prediction model to have environmental adaptability and improves the practicality of the prediction model.

[0120] It is worth noting that in the calculation of the corrected light intensity, the present invention uses the exponential function e -γ·YH A cloud attenuation model is constructed that better reflects the physical properties of the atmosphere than traditional linear attenuation models. As cloud thickness YH increases, the attenuation of light intensity exhibits a nonlinear acceleration, accurately simulating the multiple scattering effect of water vapor particles in the cloud layer on light. By setting an adjustable cloud attenuation coefficient γ, the model is made regionally adaptable: lower γ values ​​can be used in maritime climates to reflect the high light transmittance of clouds, while higher γ values ​​can be used in inland arid regions to account for the combined effects of low-altitude dust. This parameterized design enables the forecast system to rapidly adapt to different regions.

[0121] It should be noted that the temperature compensation term in formula (5), 1-0.0045 (HT-25), accurately quantifies the temperature effect of photovoltaic cells. The 0.45% / °C temperature coefficient setting conforms to the physical characteristics of crystalline silicon cells. The measured temperature coefficient range is 0.4%-0.5% / °C, and the 25°C reference temperature corresponds to the standard test conditions. This compensation term effectively eliminates power degradation in high-temperature environments. For example, when the cell temperature reaches 45°C, the system can automatically identify a 9% power loss, avoiding the prediction bias caused by traditional models that ignore temperature effects.

[0122] It should be noted that γ is an artificially set cloud attenuation coefficient, which is obtained according to the thickness and penetration of the cloud.

[0123] It should be noted that δ is an artificially set attenuation factor of the photovoltaic panel, which is obtained according to the usage time and dust shielding.

[0124] It should be noted that HT is the battery temperature in the next time period, which is obtained by comparing the current ambient temperature, the current battery temperature, the predicted number of charging people in the next time period, and the ambient temperature in the next time period. For example: DW2 = SW × DW1 × HW2 / HW1; where SW is the power generation temperature rise coefficient, which is obtained based on the predicted number of charging people in the next time period for a single station; HW1 is the current ambient temperature, DW1 is the current battery temperature, and HW2 is the predicted number of charging people in the next time period and the ambient temperature in the next time period.

[0125] In this application, the ideal charging curve for a single station in the next time period is determined, including:

[0126] Obtain the market electricity price for the next time period, and determine the grid power supply ratio BS1 at each time point based on the market electricity price i ; The proportion coefficient BS1 i Satisfying formula (6):

[0127]

[0128] Wherein, SD is the market electricity price in the next time period, BSD is the manually set standard market electricity price; μ1 is the manually set proportional adjustment coefficient, and the value range of μ1 is (0,2];

[0129] Get the level S of the stored power and the output power CP at each time point in the next time period i , based on the level S of the stored power and the output power CP i Determine the BS2 ratio of photovoltaic and energy storage power supply at each time point i ; The proportion coefficient BS2 i Satisfying formula (7):

[0130]

[0131] Wherein, BCP is the manually set standard output power, and the level S is 0, 1, and 2; μ2 is the manually set proportional adjustment coefficient, and the value range of μ2 is (0, 2];

[0132] Based on the proportion coefficient BS1 at each time point i , proportion coefficient BS2 i and output power CP i Determine the charging power CD at each time point i ; The charging power CD i Satisfy formula (8):

[0133]

[0134] Wherein, DP is the grid power supply;

[0135] The charging power CD at each time point i As the vertical axis, each time point as the horizontal axis, use the interpolation method to establish the ideal charging curve.

[0136] It is worth noting that the present invention achieves the coordinated optimization of the power grid and renewable energy by integrating the power supply ratio of the power grid, the power supply ratio of photovoltaic and energy storage, and the real-time output power, constructs a multi-dimensional dynamic model, and ultimately generates the optimal charging power curve that reflects the balance between economy and power supply.

[0137] It is worth noting that the design of the exponential functions in formulas (6) and (7) gives the proportion coefficient a nonlinear response characteristic: when the market electricity price is lower than the standard electricity price, the proportion coefficient of grid power supply increases exponentially, giving priority to low-priced grid power; when the photovoltaic output power exceeds the standard power, the proportion coefficient of photovoltaic and energy storage power supply increases rapidly to maximize the absorption of renewable energy. This dynamic balance mechanism enables the system to absorb grid power during periods of low electricity prices and give priority to absorbing clean energy during periods of peak photovoltaic output, thereby improving the utilization rate of comprehensive energy.

[0138] It is worth noting that the present invention achieves an ideal charging curve through economic optimization of the photovoltaic and energy storage output in conjunction with electricity prices. This not only takes into account electricity prices, but also simultaneously integrates the photovoltaic and energy storage output power and energy storage status. For example, during the midday hours when sunlight is abundant and electricity prices are high, the system prioritizes photovoltaic power supply while reducing grid charging power to match user demand elasticity. During the early morning hours when electricity prices are low, if there is sufficient energy storage capacity, the system can simultaneously utilize the grid and energy storage for charging, shortening charging time and reducing marginal costs. This multi-objective optimization strategy enables a single station to play the dual roles of "electricity user" and "electricity seller," improving the profitability of the single station in the electricity market.

[0139] It's important to note that by regulating the ratio of solar-to-storage power generation, the present invention's system can combine the intermittent output of photovoltaics with the energy storage characteristics of energy storage. For example, during periods of rapidly fluctuating light intensity, the system dynamically adjusts to smooth out fluctuations in photovoltaic output, reducing impact on the grid. In situations where energy storage discharge efficiency is low, the system reduces its reliance on energy storage by reducing the μ2 value through Level S, thereby extending equipment life.

[0140] It should be noted that the current level S of stored power is 0, 1, and 2; the level S of stored power is determined based on the threshold, for example: if the stored power is less than power threshold 1, the level S of stored power is 0; if the stored power is between power threshold 1 and power threshold 2, the level S of stored power is 1; if the stored power is greater than power threshold 2, the level S of stored power is 2; wherein, power threshold 1 and power threshold 2 are both obtained through manual setting, and power threshold 1 is less than power threshold 2.

[0141] It should be noted that μ1 and μ2 are manually set proportional adjustment coefficients, which are set based on the total storage battery capacity and photovoltaic panel area of ​​a single station.

[0142] In this application, the charging price curve for the next time period is set based on the ideal charging curve, including:

[0143] Extract the standard electricity price BE corresponding to the manually set standard charging power BD, based on the standard charging power BD, standard electricity price BE and charging power CD at each time point i Get the reference electricity price CE at each time point i ; The reference electricity price CE i Satisfying formula (9):

[0144]

[0145] Among them, ρ is a price adjustment coefficient that is set manually and is greater than 0;

[0146] Use each reference electricity price CE i The charging price curve is obtained by replacing the vertical coordinate value corresponding to each time point in the ideal charging curve.

[0147] It should be noted that ρ is a price adjustment coefficient that is set manually and is greater than 0. The level, as well as the manually set photovoltaic and energy storage power supply prices and standard electricity prices are analyzed.

[0148] In this application, the charging price curve is optimized by the user price sensitivity factor to determine the final electricity price curve, including:

[0149] The user price sensitivity factor HZ is extracted, and the price adjustment coefficient DS is obtained based on the user price sensitivity factor HZ; the price adjustment coefficient DS satisfies formula (10):

[0150] DS=e -σ·HZ / BHZ (10);

[0151] Wherein, σ is the manually set amplitude adjustment coefficient, and the value range of σ is [0,1];

[0152] The final electricity price curve is obtained by multiplying the price adjustment coefficient DS by the vertical coordinate value of each time point in the charging price curve.

[0153] It should be noted that after obtaining the final electricity price curve, the electricity prices for each time period published by the user can be the final electricity price curve, or the extreme value comparison of the final electricity price curve can be performed to obtain a charging price range, or the mode of the vertical coordinate in the final electricity price curve can be used as the published charging price.

[0154] The user price sensitivity factor in this application is determined by a user price sensitivity model constructed through historical data, including:

[0155] Extract the number of chargers during low-price periods, the number of chargers during high-price periods, and the user price sensitivity factor from the historical data of each single station. The historical data includes the number of chargers during low-price periods and the number of chargers during high-price periods, as well as the user price sensitivity factor set by experts based on the number of chargers during low-price periods and the number of chargers during high-price periods.

[0156] The number of people charging during low-price periods, the number of people charging during high-price periods, and the user price sensitivity factor are integrated into several sets of training data and test data; the artificial intelligence model is trained using the training data, and the trained artificial intelligence model is tested using the test data, and the artificial intelligence model is adjusted based on the test results; ultimately, a user price sensitivity model is obtained, whose input is the number of people charging during low-price periods and the number of people charging during high-price periods in the historical data of a single station, and the output is the user price sensitivity factor of the single station; the artificial intelligence model includes a BP neural network model and an RBF neural network model;

[0157] The number of charging users during low-price periods and high-price periods in the historical data of the current single station are input into the user price sensitivity model to obtain the user price sensitivity factor of the current single station.

[0158] It should be noted that the low-price period is the time period when the average price of a single station within the period is lower than the manually set low-price threshold, and the high-price period is the time period when the average price of a single station within the period is higher than the manually set high-price threshold.

[0159] It should be noted that the user price sensitivity factor is a factor set by experts based on the number of people charging during low-price periods and the number of people charging during high-price periods, which can reflect the price sensitivity of users at a single station. The number of people charging during low-price periods is directly proportional to the user price sensitivity factor, and the number of people charging during high-price periods is inversely proportional to the user price sensitivity factor. The larger the user price sensitivity factor, the more sensitive the user is to price, and the greater the impact on subsequent high price setting. When setting prices subsequently, it is more important to avoid excessively high prices. The smaller the user price sensitivity factor, the less sensitive the user is to price, and the smaller the impact on subsequent high price setting.

[0160] Specifically, the trained artificial intelligence model is tested using test data, and the specific steps for adjusting the artificial intelligence model based on the test results are as follows:

[0161] The number of people charging during low-price periods and the number of people charging during high-price periods in the test data are input into the trained artificial intelligence model to obtain the corresponding user price sensitivity factors. The corresponding user price sensitivity factors are compared with the corresponding user price sensitivity factors in the test data. When the difference between the two is within the threshold, which is obtained based on experience, there is no need to adjust the parameters, and the next set of test data is tested; if it is not within the threshold, the corresponding parameters are adjusted until the difference between the two is within the threshold, and then the next set of test data is tested. When the number of test data with user price sensitivity factors within the threshold obtained from all test data accounts for 90% or more of the total test data, a user price sensitivity model is obtained, whose input is the number of people charging during low-price periods and the number of people charging during high-price periods in the historical data of a single station, and whose output is the user price sensitivity factor of a single station.

[0162] See also Figure 2 The second embodiment of the present invention provides a single-station dynamic pricing method and system based on market transactions and photovoltaic storage power generation, characterized by comprising: an intelligent analysis module, and a data collection module and a dynamic pricing module connected thereto;

[0163] Data collection module: used to set at least one time period for a single station based on the number of charging stations at multiple time points, obtain the stored power of the single station in the current time period, as well as the market electricity price and surrounding environmental data for the next time period; the surrounding environmental data includes weather type, light intensity, and cloud thickness;

[0164] Intelligent analysis module: used to determine the output power of the photovoltaic panels in the next time period based on the surrounding environmental data of the next time period; based on the market electricity price, storage capacity and output power, determine the ideal charging curve for the next time period of a single station;

[0165] Dynamic Pricing Module: This module sets the charging price curve for the next time period based on the ideal charging curve. The final price curve is determined by optimizing the charging price curve using a user price sensitivity factor, which is determined by a user price sensitivity model constructed using historical data. This historical data includes the number of users charging during low-price periods and the number of users charging during high-price periods.

[0166] Database: used to store data.

[0167] Some of the data in the above formula are calculated by removing the dimensions and taking their numerical values. The formula is a formula that is closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0168] Working principle of the present invention:

[0169] Used to set at least one time period for a single station, obtain the stored power of the single station in the current time period, as well as the market electricity price and surrounding environmental data for the next time period; determine the output power of the photovoltaic panel in the next time period based on the surrounding environmental data of the next time period. This step improves the accuracy of the prediction by constructing a multi-dimensional environmental parameter system of weather type, cloud thickness, ambient humidity, light intensity, and battery temperature; determine the ideal charging curve for the next time period of the single station based on the market electricity price, stored power and output power. This step generates an optimal charging power curve reflecting the balance of economy and power supply by constructing a multi-dimensional dynamic model; set the charging price curve for the next time period based on the ideal charging curve; determine the user price sensitivity factor of the single station through a user price sensitivity model constructed through historical data. This step analyzes the user's sensitivity to price changes and adjusts the charging price accordingly; optimize the charging price curve through the user price sensitivity factor to determine the final electricity price curve.

[0170] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A single-station dynamic pricing method based on market transactions and photovoltaic storage power generation is characterized by: include: Based on the number of charging stations at multiple time points, at least one time period is set for the station, and the stored power of the station in the current time period, as well as the market electricity price and surrounding environmental data for the next time period, are obtained. The surrounding environmental data includes weather type, light intensity, and cloud thickness. determining the output power of the photovoltaic panel in the next time period based on the ambient environment data in the next time period; Determine the ideal charging curve for the next time period for the single station based on the market electricity price, stored electricity, and output power; Set the charging price curve for the next time period based on the ideal charging curve; The final electricity price curve is determined by optimizing the charging price curve using a user price sensitivity factor, where the user price sensitivity factor is determined by a user price sensitivity model constructed using historical data; wherein the historical data includes the number of people charging during low-price periods and the number of people charging during high-price periods.

2. The single-station dynamic pricing method based on market transactions and photovoltaic power generation according to claim 1 is characterized in that: Before setting at least one time period of a single station, the method further includes: Obtain the number of charging stations at each time point in the history of several days, and mark the mode of the charging numbers at the time points as the analysis number at the corresponding time point; The number of analyses at each time point is integrated into a quantity group, and it is determined whether the standard deviation of the quantity group is less than the defined threshold; if so, a day is divided into several time periods according to a fixed length; if not, the time points corresponding to the number of analyses in the quantity group that is less than the percentile of the quantity group are marked as low-peak time points, and the non-low-peak time points in a day are marked as peak time points.

3. The single-station dynamic pricing method based on market transactions and photovoltaic power generation according to claim 2 is characterized in that: The step of setting at least one time period for a single station includes: Consolidate the off-peak time points into several off-peak periods, and determine the planned duration G1 of the off-peak period based on the average value P1 of the number of analysis points in the off-peak period; the planned duration G1 satisfies formula (1): Where DC is the fixed duration, BW is the percentile, α is the amplitude adjustment coefficient, and the value range of α is (0,2]; Consecutive peak time points are integrated into several peak periods, and the planned duration G2 of the peak period is determined based on the average value P2 of the number of analysis points within the peak period; the planned duration G2 satisfies formula (2): The low-peak time periods that are shorter than the planned duration G1 and the peak time periods that are shorter than the planned duration G2 are merged into the smaller time periods adjacent on both sides. The ratio of each low-peak time period to the planned duration G1 is rounded up to obtain the planned number of the corresponding low-peak time period. The ratio of each peak time period to the planned duration G2 is rounded up to obtain the planned number of the corresponding peak time period. The corresponding low-peak time period or peak time period is evenly divided into several time periods according to the planned number.

4. The single-station dynamic pricing method based on market transactions and photovoltaic power generation according to claim 1 is characterized in that: The acquisition of the stored electricity of a single station in the current time period, as well as the market electricity price and surrounding environment data for the next time period, includes: The market electricity price of a single station in the next time period is obtained from the power company's website, the stored electricity of a single station is extracted from the database, and the weather type, light intensity and cloud thickness in the next time period in the area where the single station is located are obtained from the weather forecast website; among them, the weather types include sunny, cloudy and overcast.

5. The single-station dynamic pricing method based on market transactions and photovoltaic power generation according to claim 1 is characterized in that: The determining the output power of the photovoltaic panel in the next time period based on the ambient environment data in the next time period includes: The weather type and cloud thickness YH of the next time period are extracted, and the weather impact factor β is determined based on the weather type and cloud thickness YH; the weather impact factor β satisfies formula (3): Wherein, when j = 1, it indicates sunny, when j = 2, it indicates cloudy, and when j = 3, it indicates overcast; BYH is the standard cloud thickness, and RH is the humidity in the area where the single station is located in the next time period; Extract the illumination intensity I of the next time period, and determine the modified light intensity ZI based on the illumination intensity; the modified light intensity ZI satisfies formula (4): ZI = I·e -γ·YH (4); Where γ is the cloud attenuation coefficient, and the value range of γ is (0,2]; Determine the output power CP of the photovoltaic panel at each time point in the next time period based on the weather influence factor β and the corrected light intensity ZI i ; The output power CP i Satisfying formula (5): CP i =δ·β·θ·A·ZI·(1-0.0045(HT-25) (5); Wherein, δ is the attenuation factor of the photovoltaic panel, θ is the photoelectric conversion efficiency, i is the number of the time point; A is the effective light-receiving area of ​​the photovoltaic panel, and HT is the battery temperature in the next time period.

6. The single-station dynamic pricing method based on market transactions and photovoltaic power generation according to claim 1 is characterized in that: Determining the ideal charging curve for the next time period of a single station includes: Obtain the market electricity price for the next time period, and determine the grid power supply ratio BS1 at each time point based on the market electricity price i ; The proportion coefficient BS1 i Satisfy formula (6): Among them, SD is the market electricity price in the next time period, BSD is the standard market electricity price; μ1 is the proportional adjustment coefficient, and the value range of μ1 is (0,2]; Get the level S of the stored power and the output power CP at each time point in the next time period i , based on the level S of the stored power and the output power CP i Determine the BS2 ratio of photovoltaic and energy storage power supply at each time point i ; The proportion coefficient BS2 i Satisfy formula (7): Wherein, BCP is the standard output power, the level S is 0, 1, 2; μ2 is the proportional adjustment coefficient, and the value range of μ2 is (0,2]; Based on the proportion coefficient BS1 at each time point i , proportion coefficient BS2 i and output power CP i Determine the charging power CD at each time point i ; The charging power CD i Satisfy formula (8): Wherein, DP is the grid power supply; The charging power CD at each time point i As the vertical axis, each time point as the horizontal axis, use the interpolation method to establish the ideal charging curve.

7. The single-station dynamic pricing method based on market transactions and photovoltaic power generation according to claim 6 is characterized in that: The step of setting a charging price curve for the next time period based on the ideal charging curve includes: Extract the standard electricity price BE corresponding to the manually set standard charging power BD, and based on the standard charging power BD, the standard electricity price BE and the charging power CD at each time point i Get the reference electricity price CE at each time point i ; The reference electricity price CE i Satisfy formula (9): Where ρ is the price adjustment coefficient greater than 0; Use each reference electricity price CE i The charging price curve is obtained by replacing the vertical coordinate value corresponding to each time point in the ideal charging curve.

8. The single-station dynamic pricing method based on market transactions and photovoltaic power generation according to claim 1 is characterized in that: The step of optimizing the charging price curve by using the user price sensitivity factor to determine the final electricity price curve includes: Extract the user price sensitivity factor HZ, and obtain the price adjustment coefficient DS based on the user price sensitivity factor HZ; the price adjustment coefficient DS satisfies formula (10): DS=e -σ·HZ / BHZ (10); Among them, σ is the amplitude adjustment coefficient, and the value range of σ is [0,1]; The final electricity price curve is obtained by multiplying the price adjustment coefficient DS by the vertical coordinate value of each time point in the charging price curve.

9. The single-station dynamic pricing method based on market transactions and photovoltaic power generation according to claim 1 is characterized in that: The user price sensitivity factor is determined by a user price sensitivity model constructed through historical data, including: Extract the number of chargers during low-price periods, the number of chargers during high-price periods, and the user price sensitivity factor from the historical data of each single station. The historical data includes the number of chargers during low-price periods and the number of chargers during high-price periods, as well as the user price sensitivity factor set by experts based on the number of chargers during low-price periods and the number of chargers during high-price periods. The number of people charging during low-price periods, the number of people charging during high-price periods, and the user price sensitivity factor are integrated into several sets of training data and test data; the artificial intelligence model is trained using the training data, and the trained artificial intelligence model is tested using the test data, and the artificial intelligence model is adjusted based on the test results; ultimately, a user price sensitivity model is obtained, whose input is the number of people charging during low-price periods and the number of people charging during high-price periods in the historical data of a single station, and the output is the user price sensitivity factor of the single station; the artificial intelligence model includes a BP neural network model and an RBF neural network model; The number of charging users during low-price periods and high-price periods in the historical data of the current single station are input into the user price sensitivity model to obtain the user price sensitivity factor of the current single station.

10. A method and system for dynamic pricing of a single station based on market transactions and photovoltaic power generation with energy storage, which is based on the method for dynamic pricing of a single station based on market transactions and photovoltaic power generation with energy storage as claimed in any one of claims 1 to 9, and is characterized in that: include: Intelligent analysis module, and its connected data collection module and dynamic pricing module; The data collection module is used to set at least one time period for a single station based on the number of charging operations at multiple time points, and obtain the stored electricity volume of the single station in the current time period, as well as the market electricity price and surrounding environmental data for the next time period; wherein the surrounding environmental data includes weather type, light intensity, and cloud thickness; The intelligent analysis module is used to determine the output power of the photovoltaic panel in the next time period based on the ambient environment data in the next time period; and to determine the ideal charging curve for the next time period of the single station based on the market electricity price, the amount of stored electricity and the output power; The dynamic pricing module is used to set the charging price curve for the next time period based on the ideal charging curve; the final electricity price curve is determined by optimizing the charging price curve using a user price sensitivity factor, where the user price sensitivity factor is determined by a user price sensitivity model constructed using historical data; wherein the historical data includes the number of people charging during low-price periods and the number of people charging during high-price periods.

Images