Time-sharing and domain-based electricity price formulation method and device considering charging pressure-state-behavior

Abstract

The present application discloses a method and device for formulating time-of-use and region-of-use electricity prices considering charging pressure-state-behavior. The method introduces the PSR model. Based on historical data analysis, the charging demand is regarded as the pressure factor, the user's optimal charging pile and charging busyness are regarded as the state factors, and the charging behavior is regarded as the behavior factor. The parameter weights of the PSR-FM model are deduced reversely. Then, the PSR-FM model under a determined time scale is combined with the new charging demand, the user's optimal charging pile and charging busyness to predict the user's charging behavior. Finally, a dynamic correction mechanism combining long-time scale prediction and short-time scale correction is introduced to formulate electricity prices for different periods and different regions. The present invention incorporates the travel trajectories of electric vehicles into the charging pricing strategy, which can significantly improve the user experience and enhance user stickiness. It can be seen that the present invention has significant superiority, economy, feasibility and commercial transformation potential.

Description

Technical Field

[0001] The present invention relates to the field of power technology, and particularly to a method and device for formulating time-of-use and region-of-use electricity prices considering charging pressure-state-behavior. Background Art

[0002] In the context of the current booming electric vehicle market, the research on charging electricity price strategies is no longer limited to the traditional time-of-use pricing model, but is gradually developing towards a more refined and intelligent direction. Although the traditional time-of-use pricing strategy can effectively regulate the power grid load, it ignores the uneven spatial distribution of electric vehicle charging demands and the profound impact of electric vehicle travel trajectories on charging behaviors.

[0003] Currently, the research direction of methods for formulating electric vehicle charging electricity price strategies is mainly to find an optimal pricing scheme that takes into account both user costs and load peak shaving in the time dimension (Ma Pengbo, Huang Zhao, Guo Zhiwei, etc. Optimal Scheduling of Electric Vehicles Considering Load Peak Shaving and User Costs [J]. Electrical Automation, 2024, 46(05): 1-3.). The general research idea is to solve by taking the user's electricity consumption cost and the peak-valley difference of the power grid load as optimization objectives, and guide electric vehicle users to consume electricity during valley hours through prices. Although this mode can regulate the power grid load to a certain extent, on the one hand, it ignores the uneven spatial distribution of electric vehicle charging demands and the profound impact of electric vehicle travel trajectories on charging behaviors. On the other hand, the time-of-use electricity price has already formulated the electricity price for each period in advance based on prediction data, and cannot be dynamically adjusted immediately for sudden situations such as special weather and market changes. Summary of the Invention

[0004] The present application provides a method and device for formulating time-of-use and region-of-use electricity prices considering charging pressure-state-behavior, which is applicable to formulating time-of-use and region-of-use pricing strategies and provides technical support for the construction of the post-intelligent power grid.

[0005] The purpose of the present invention is achieved by at least one of the following technical solutions.

[0006] A method for formulating time-of-use and region-of-use electricity prices considering charging pressure-state-behavior includes the following steps:

[0007] S1. Collect data on the state of charge (SOC) of electric vehicle batteries, electric vehicle charging behavior data, occupancy data and queuing data of each charging pile at each moment within a set time period in the research area;

[0008] S2. Calculate the charging demand of electric vehicles based on the state-of-charge data of electric vehicle batteries, calculate the optimal charging pile for users according to the user's location in the electric vehicle charging behavior data and the time-of-use and region-of-use electricity prices, and calculate the busyness of the charging piles according to the occupancy data and queuing data of each charging pile;

[0009] S3. Take the user's charging demand as the pressure factor, the optimal charging pile for the user and the busyness degree of the charging pile as the state factor, and the historical charging behavior as the behavior factor to construct a PSR-FM model, and obtain the formula for calculating the weight parameters of the PSR-FM model.

[0010] S4. Set the long-term scale prediction - short-term scale correction mode, obtain the user's historical charging behavior, and obtain the weight parameters of the PSR-FM model under two time scales of long-term scale and short-term scale, so as to obtain the PSR-FM models under two time scales.

[0011] S5. Use the PSR-FM models under two time scales to predict the user's charging behavior respectively, and determine the time scale for prediction in the current time period according to the prediction error in the previous time period.

[0012] S6. Take the time-of-use and time-of-day electricity price as a variable, take the electric vehicle battery charging state data, the occupancy data and queuing data of each charging pile as the initial values, use the PSR-FM model under the time scale determined in step S5 to predict the user's charging behavior, set the sum of the charging benefit and the electricity selling benefit as the objective function, and use the particle swarm optimization algorithm to iteratively calculate the time-of-use and time-of-day electricity prices for each time period in the research area to ensure the user's charging benefit and the electricity selling benefit of the charging pile, complete the electricity price formulation for the current time period, and return to step S5.

[0013] Furthermore, in step S2, calculate the charging demand according to the electric vehicle battery charging state data, and perform quantitative evaluation and normalization processing on the charging demand. The specific calculation formula is as follows:

[0014] ;

[0015] where is the required calculation time, is the time division interval, is the charging demand of the th vehicle at time , the closer is to 1, the greater the charging demand of the vehicle, is the maximum value of all vehicles' SOC at time , is the minimum value of all vehicles' SOC at time , is the of the SOC th vehicle at time

[0016] ​​​Furthermore, in step S2, the optimal charging pile for the user is calculated according to the user's location and the time-sharing and domain-based electricity price in the electric vehicle charging behavior data, and the optimal charging pile for the user is quantitatively evaluated and normalized. The specific calculation formula is as follows:

[0017] ;

[0018] in, for Moment The charging cost performance of a vehicle at the optimal charging station is used to indicate the economic benefits of charging. The closer it is to 1, the more cost-effective the car charging is; for Moment Car SOC ; for Moment The price of charging the vehicle; for Moment The distance from the vehicle to its optimal charging station, Travel expenses per unit distance.

[0019] Furthermore, in step S2, the busyness of the charging piles is calculated according to the occupancy and queuing conditions of each charging pile, and a quantitative evaluation and normalization process is performed on the busyness of the charging piles. The specific calculation formula is as follows:

[0020] ;

[0021] in, Moment The optimal charging station selected by the vehicle is Charging stations; for Moment The busyness of the optimal charging pile selected by the vehicle is used to indicate the busyness of the charging pile. The closer it is to 1, the more idle the charging station is; for The total number of vehicles queuing at the charging station at any given moment; for Moment Car SOC ; for Moment The charging power of each charging station; For the The maximum charging power of each charging pile; For the The total number of vehicles in the queue when the charging power of a charging pile is maximum.

[0022] Further, in step S3, a PSR-FM model is constructed based on the PSR evaluation system and the FM algorithm model. The user's charging demand is used as the pressure factor, that is, the disaster-causing factor, and the vehicle-pile location distribution, time-of-use electricity price, and charging pile busyness are used as the state factors, that is, the disaster-bearing environment. Combining historical charging behaviors, a time-of-use electricity price pricing strategy is finally obtained;

[0023] To better explore the influence of the pressure factor and the state factor on the user's charging behavior, the state factor includes charging economic benefits and charging pile busyness. The pressure factor and the state factor expressed numerically are transformed into vectors existing in the first quadrant. The specific calculation formula is as follows:

[0024] ;

[0025] Among them, is the set of the pressure factor and the state factor of the th vehicle at time . When is 1, 2, or 3, it corresponds to the charging economic benefits in the pressure factor, the charging economic benefits in the state factor, and the charging pile busyness in the state factor respectively;

[0026] In the PSR-FM model, according to the pressure factor , the charging economic benefits in the state factor, and the charging pile busyness in the state factor, the predicted user charging behavior of the th vehicle at time is obtained, which is specifically as follows:

[0027] ;

[0028] According to the user's historical charging demand, the historical busyness of the charging pile, and the user's historical charging behavior, the prediction parameters and of the PSR-FM model can be deduced reversely; among them is the charging behavior error parameter of , and is

[0029] Further, in step S4, a long-time scale prediction - short-time scale correction mode is set;

[0030] The time period corresponding to the long-time scale is one day, that is, 96 15-minute intervals, and the predicted time period corresponding to the short-time scale is 1 15-minute interval. In the long-time scale prediction mode, The value is 96. In the short - time scale correction mode, the value is 1;

[0031] Since the prediction periods used by the two time scales are different, the weight parameters of the PSR - FM model under the two time scales and are also different;

[0032] Obtain the user's historical charging behavior:

[0033] ;

[0034] The weight parameters of the PSR - FM model under the two time scales; Obtain the PSR - FM models under the two time scales.

[0035] Furthermore, in step S5, using the day - ahead data and real - time data, respectively use the PSR - FM models under the two time scales to predict the charging behavior of users in the previous time period and , and according to the prediction errors and of the PSR - FM models under the two time scales, determine which time scale to use for prediction in the current time period. The specific calculation formula is as follows:

[0036] ;

[0037] Among them, is the prediction error of the PSR - FM model under the long - time scale, is the current moment, is the real charging behavior of the th user at the moment, is the charging behavior of the user predicted by the PSR - FM model under the long - time scale, is the charging behavior of the user predicted by the PSR - FM model under the short - time scale.

[0038] Furthermore, in step S5, if the current time scale is the long - time scale, a time period is one day, that is, 96 15 - minute intervals. If the current time scale is the short - time scale, a time period is 15 minutes.

[0039] Furthermore, in step S6, taking the time - of - use and time - of - day electricity price as a variable, taking the electric vehicle battery charging state data, the occupancy data and queuing data of each charging pile as initial values, using the PSR - FM model under the time scale determined in step S5 for the The charging behavior of vehicle users is predicted;

[0040] Taking the sum of charging benefits and electricity selling benefits as the objective function, the time-of-use and sectional electricity prices for each period in the research area are iteratively calculated using the particle swarm optimization algorithm. The calculation formula of the objective function is as follows:

[0041] ;

[0042] ;

[0043] ;

[0044] Among them, and are the number of charging piles and the number of electric vehicle users respectively, is the pricing coefficient of the charging pile, is the electricity price of the th charging pile at time is the charging price of the th user at time is the total charging consumption of electric vehicle users, is the particle swarm optimization objective function.

[0045] Compared with the prior art, the advantages of the present invention are as follows:

[0046] The present invention proposes a method for formulating electricity prices considering charging pressure-state-behavior time-of-use and time-domain. By introducing the PSR model and based on historical data analysis, regarding the charging demand as the pressure factor, the user's optimal charging pile and charging busyness as the state factor, and the charging behavior as the behavior factor, the parameter weights of the PSR-FM model are inversely deduced. Then, the trained PSR-FM model is combined with new charging demands, the user's optimal charging pile and charging busyness to predict the user's charging behavior. Finally, a dynamic correction mechanism combining long-time scale prediction and short-time scale correction is introduced to formulate electricity prices for different periods and different regions.

[0047] It can be seen that, compared with the prior art, the present invention perfectly solves the deficiencies of the prior art. On the one hand, the present invention introduces the charging pile busyness degree and the user's location to formulate electricity prices for different regions, which can balance the supply and demand distribution of charging resources, relieve the peak pressure of charging stations in the city center, and guide users to reasonably choose the charging time and location, thereby improving the overall utilization rate of charging facilities. On the other hand, the present invention introduces a dynamic correction mechanism of day-ahead prediction - 15-minute correction, which helps to cope with various emergencies and instantaneously adjust the charging electricity price to ensure the user experience. In addition, the present invention incorporates the travel trajectory of electric vehicles into the charging pricing strategy, which can significantly improve the user experience and enhance user stickiness. It can be seen that the present invention has significant superiority, economy, feasibility and commercial transformation potential.

[0048] The present invention incorporates the travel trajectory of electric vehicles as a key factor into the charging pricing strategy, aiming to further improve and optimize the existing pricing system. Considering that the charging demand of electric vehicle users not only fluctuates with time but is also deeply affected by geographical location, by analyzing the driving routes and charging habits of vehicles, the charging demand at different locations can be accurately predicted. This innovative idea enables the charging price to be flexibly adjusted according to the actual demand. It can provide price discounts during periods with low grid load or at charging stations with low utilization rates, guiding users to reasonably choose the charging time and location, thereby improving the overall utilization rate of charging facilities and reducing resource waste. It can also balance the supply and demand distribution of charging resources through a differential pricing strategy, relieve the peak pressure of charging stations in the city center, and reduce traffic congestion problems caused by charging. In addition, incorporating the travel trajectory of electric vehicles into the charging pricing strategy can significantly improve the user experience and enhance user stickiness. Through personalized and intelligent charging services, it can meet the actual needs and preferences of different users, such as providing cross-regional charging discount packages for long-distance travel users and providing convenient and preferential charging services at their regular locations for short-distance commuting users, thereby enhancing the satisfaction and loyalty of users to electric vehicles and charging services. Through a reasonable charging price system, it can not only attract more consumers to purchase electric vehicles and promote the further expansion of the new energy vehicle market, but also promote the reasonable layout and efficient operation of charging infrastructure, laying a solid foundation for the long-term prosperity of the new energy vehicle industry. At the same time, it also helps to achieve the coordinated optimization of energy, transportation and the environment, contributing to the construction of a green and low-carbon sustainable development society.

[0049] In summary, incorporating the travel trajectory of electric vehicles into the charging pricing strategy is an important measure to address the challenges of the current electric vehicle market, improve the grid operation efficiency, optimize resource allocation, enhance the user experience and promote the sustainable development of the new energy vehicle industry. The implementation of this innovative idea will open a new chapter in electric vehicle charging services and inject new impetus into the popularization and green development of new energy vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 Schematic diagram of the time - of - use and domain - of - use electricity price formulation method considering charging pressure - state - behavior in the embodiment of the present invention;

[0051] Figure 2 Schematic diagram of the structure of the PSR - FM model in the embodiment of the present invention;

[0052] Figure 3 Schematic diagram of the distribution of charging power in time and space before optimization in the embodiment of the present invention;

[0053] Figure 4 Schematic diagram of the distribution of charging power in time and space after optimization in the embodiment of the present invention. Detailed implementation manners

[0054] To make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the following examples are given in conjunction with the accompanying drawings to elaborate on the specific implementation of the present invention.

[0055] Example:

[0056] The time - of - use and domain - of - use electricity price formulation method considering charging pressure - state - behavior, as Figure 1 shown, includes the following steps:

[0057] S1. Collect data on the charging state (State of Charge, SOC) of electric vehicle batteries, electric vehicle charging behavior data, occupancy data and queuing data of each charging pile at each moment within the set time period in the research area;

[0058] S2. Calculate the charging demand of electric vehicles according to the charging state data of electric vehicle batteries, calculate the optimal charging pile for users according to the user's location and time - of - use and domain - of - use electricity prices in the electric vehicle charging behavior data, and calculate the busyness degree of the charging piles according to the occupancy data and queuing data of each charging pile;

[0059] Calculate the charging demand according to the charging state data of electric vehicle batteries, and conduct quantitative evaluation and normalization processing on the charging demand. The specific calculation formula is as follows:

[0060] ;

[0061] Among them, is the required calculation moment, is the time division interval, is the charging demand of the th vehicle at the moment. The closer is to 1, the greater the charging demand of the vehicle. is SOC the maximum value in for all vehicles at the​ The minimum value among all vehicles at a certain moment SOC is the charging cost performance of the th vehicle at a certain moment SOC .

[0062] Calculate the optimal charging pile for users according to the user's location and time-of-use and domain-of-use electricity prices in the electric vehicle charging behavior data, and conduct quantitative evaluation and normalization processing on the optimal charging pile for users. The specific calculation formula is as follows:

[0063] ;

[0064] Among them, is the charging cost performance of the th vehicle at a certain moment at the optimal charging pile, which is used to represent the charging economic benefit, the closer it is to 1, the higher the charging cost performance of the vehicle; is the th vehicle at a certain moment SOC ; is the charging price of the th vehicle at a certain moment; is the distance from the th vehicle to its optimal charging pile at a certain moment, and

[0065] Calculate the busyness of the charging pile according to the occupancy and queuing situation of each charging pile, and conduct quantitative evaluation and normalization processing on it. The specific calculation formula is as follows:

[0066] ;

[0067] Among them, the optimal charging pile selected by the th vehicle at a certain moment is the th charging pile; is the busyness of the optimal charging pile selected by the th vehicle at a certain moment, which is used to represent the busyness of the charging pile, the closer it is to 1, the more idle the charging pile is; is the total number of queuing vehicles at the charging pile at a certain moment; is the th vehicle at a certain moment SOC ; is the The charging power of a charging pile; is the maximum charging power of the th charging pile; is the total number of queuing vehicles when the charging power of the

[0068] S3. Take the user's charging demand as the stress factor, the optimal charging pile for the user and the busyness degree of the charging pile as the state factor, and the historical charging behavior as the behavior factor, construct the PSR-FM model, and obtain the formula for calculating the weight parameters of the PSR-FM model;

[0069] In one embodiment, as Figure 2 shown, based on the PSR evaluation system (see: Xu Bailu, Wang Kuang. Comprehensive evaluation of soil environmental quality in railway vehicle depots based on the PSR model [J / OL]. Railway Standard Design, 1-8 [2024-12-04]. https: / / doi.org / 10.13238 / j.issn.1004-2954.202405280004.) and the FM algorithm model (see: Li Xingbing, Xie Jun, Xu Xinying, et al. Deep click-through rate prediction model based on improved FM algorithm and attention mechanism [J]. Journal of Nanjing University of Science and Technology, 2021, 45(04): 429-438. DOI: 10.14177 / j.cnki.32-1397n.2021.45.04.006.), construct the PSR-FM model. Compared with the traditional charging price strategy, take the user's charging demand as the stress factor, that is, the disaster-causing factor, and the vehicle-pile location distribution, time-of-use electricity price, and the busyness degree of the charging pile as the state factor, that is, the disaster-bearing environment, and comprehensively consider the historical charging behavior. Finally, obtain the time-of-use electricity price pricing strategy;

[0070] To better explore the influence of the stress factor and the state factor on the user's charging behavior, the state factor includes charging economic benefits and the busyness degree of the charging pile. Convert the stress factor and the state factor expressed numerically into vectors existing in the first quadrant. The specific calculation formula is as follows:

[0071] ;

[0072] Among them, is the stress factor and state factor set of the th vehicle at time . When

[0073] is 1, 2, or 3, it corresponds to the stress factor, charging economic benefits in the state factor, and the busyness degree of the charging pile in the state factor respectively; In the PSR-FM model, according to the stress factor , the busyness level of the charging pile in the state factor , to obtain the predicted At the user charging behavior of the vehicle, specifically as follows:

[0074] ;

[0075] According to the user's historical charging demand, the historical busyness level of the charging pile, and the user's historical charging behavior, the prediction parameters of the PSR-FM model can be deduced reversely and ; where is the charging behavior error parameter of is the mutual influence parameter of the charging behavior pressure state of

[0076] S4. Set the long-time scale prediction - short-time scale correction mode, obtain the user's historical charging behavior, obtain the weight parameters of the PSR-FM model under two time scales of long-time scale and short-time scale, and obtain the PSR-FM models under two time scales;

[0077] Set the long-time scale prediction - short-time scale correction mode;

[0078] The time period corresponding to the long-time scale is one day, that is, 96 15-minute intervals, and the predicted time period corresponding to the short-time scale is 1 15-minute interval. In the long-time scale prediction mode, takes the value of 96, and in the short-time scale correction mode, takes the value of 1;

[0079] Since the prediction periods used for the two time scales are different, the weight parameters and of the PSR-FM model under the two time scales are also different;

[0080] Obtain the user's historical charging behavior:

[0081] ;

[0082] The weight parameters of the PSR-FM model under the two time scales; obtain the PSR-FM models under the two time scales.

[0083] S5. Use the PSR-FM models under the two time scales to predict the user's charging behavior respectively, and determine the time scale for prediction in the current time period according to the prediction error in the previous time period;

[0084] Using the pre-use data and real-time data, the PSR-FM model under two time scales is used to predict the charging behavior of users in the previous time period and , and based on the prediction errors and of the PSR-FM model under two time scales, determine which time scale to use for prediction in the current time period. The specific calculation formula is as follows:

[0085] ;

[0086] where is the prediction error of the PSR-FM model under the long time scale, is the current moment, is the true charging behavior of the th user at the moment, is the charging behavior of the user predicted by the PSR-FM model under the long time scale, is the charging behavior of the user predicted by the PSR-FM model under the short time scale.

[0087] If the current time scale is the long time scale, a time period is one day, that is, 96 15-minute intervals. If the current time scale is the short time scale, a time period is 15 minutes.

[0088] S6. Taking the time-of-use and time-domain electricity price as a variable, using the electric vehicle battery charging status data, the occupancy data and queuing data of each charging pile as initial values, predicting the user charging behavior using the PSR-FM model under the time scale determined in step S5, setting the sum of the charging benefit and the electricity selling benefit as the objective function, and using the particle swarm optimization algorithm to iteratively calculate the time-of-use and time-segmented electricity prices for each time period in the research area to ensure the user charging benefit and the electricity selling benefit of the charging pile, completing the electricity price formulation for the current time period, and returning to step S5;

[0089] Taking the time-of-use and time-domain electricity price as a variable, using the electric vehicle battery charging status data, the occupancy data and queuing data of each charging pile as initial values, and using the PSR-FM model under the time scale determined in step S5 to predict the charging behavior of the user of the th vehicle at the moment;

[0090] Taking the sum of the charging benefit and the electricity selling benefit as the objective function, using the particle swarm optimization algorithm to iteratively calculate the time-of-use and time-segmented electricity prices for each time period in the research area. The calculation formula of the objective function is as follows:

[0091] ;

[0092] ;

[0093] ;

[0094] Among them, and are the number of charging piles and the number of electric vehicle users respectively, is the pricing coefficient of the charging pile, is the electricity price of the th charging pile at time is the charging price of the th user at time is the total charging consumption of electric vehicle users, is the particle swarm optimization objective function.

[0095] In one embodiment, considering that the battery industry is developing rapidly and its capacity is continuously increasing, the capacity of the considered EV battery is 80 kW·h, the power consumption per 100 km of driving is 20 kW·h, and the purchase price of the battery is 80,000 yuan. The initial SOC of the EV follows a normal distribution with a mean of 0.5 and a standard deviation of 0.1, the travel time of the user follows a normal distribution with a mean of 480 and a standard deviation of 60, and the return time of the user follows a normal distribution with a mean of 1080 and a standard deviation of 60. Based on the fitted probability function, the EV travel situation is simulated, and the charging and discharging behavior of the EV is obtained by using the Monte Carlo sampling method.

[0096] The intelligent charging device adopts the conventional charging method and the constant power charging and discharging mode, the charging and discharging power is 8 kW, and the charging and discharging efficiency is 90%. Considering safety, the upper and lower limits of SOC are set to 90% and 10% respectively.

[0097] The charging and discharging (discharging) efficiency of the EV battery is set to 90%.

[0098] The simulation scenario is a residential community - company park with 200 EVs, and the park is powered by 6 transformers with a capacity of 1500 kV·A.

[0099] In one embodiment, the IEEE33 model is used for simulation. The distribution of the charging power in space and time before optimization is as Figure 3 shown; the distribution of the charging power in space and time after optimization is as Figure 4As shown. It can be seen that the charging power before optimization is significantly unreasonable in spatial distribution, with a large amplitude of curve fluctuation. The charging load is concentrated on a small number of nodes, and the peak-valley difference is significant. The highest charging power is 49 kW. At this time, the variance of the charging power in space is 45.77. After optimization, the spatial distribution of the charging power is reasonable, the charging load is relatively evenly distributed on each node, the amplitude of curve fluctuation is small, the highest charging power drops to 31 kW, and the variance of the charging power in space is 22.91 at this time. It can be seen the effectiveness of the present invention in adjusting the spatial distribution of the charging power.

[0100] The preferred embodiments of the present application disclosed above are only used to help understand the present invention and its core idea. For those of ordinary skill in the art, according to the idea of the present invention, there will be changes in specific application scenarios and implementation operations. This specification should not be construed as a limitation on the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A time-based and domain-based electricity price formulation method considering charging pressure, state, and behavior, characterized in that: The following steps are involved: S1. Collect the battery charging status data of electric vehicles, the charging behavior data of electric vehicles, the occupancy data of each charging pile and the queuing data at each time in the set time period of the research area; S2. Calculate the charging demand of electric vehicles based on the battery charging status data of electric vehicles, calculate the optimal charging pile for users based on the user's location and the time-based and domain-based electricity price in the electric vehicle charging behavior data, and calculate the busyness of the charging piles based on the occupancy data and queuing data of each charging pile; S3. Taking the user's charging demand as the pressure factor, the user's optimal charging pile and the busyness of the charging pile as the state factor, and the historical charging behavior as the behavior factor, the PSR-FM model is constructed, and the formula for obtaining the weight parameters of the PSR-FM model is obtained; the PSR-FM model is constructed based on the PSR evaluation system and the FM algorithm model, and the user's charging demand is taken as the pressure factor, that is, the disaster-causing factor; the vehicle and pile location distribution, time-sharing and domain-sharing electricity prices and the busyness of the charging piles are taken as the state factor, that is, the disaster-prone environment, and the historical charging behavior is combined to finally obtain the time-sharing and domain-sharing electricity price pricing strategy; The state factor includes the economic benefits of charging and the busyness of the charging pile. The pressure factor and state factor expressed in numerical values ​​are converted into vectors existing in the first quadrant. The specific calculation formula is as follows: ； in, for Moment The pressure factor and state factor set of the vehicle, when When it is 1, 2, and 3, it corresponds to the pressure factor, the charging economic benefit in the state factor, and the busyness of the charging pile in the state factor respectively; In the PSR-FM model, according to the pressure factor , Charging economic benefits in state factors , the busyness of the charging pile in the state factor , get the predicted Moment Vehicle user charging behavior , as follows: ， According to the user's historical charging demand, the historical busyness of the charging pile, and the user's historical charging behavior, the prediction parameters of the PSR-FM model can be reversely derived. and ;in for The charging behavior error parameter, for The charging behavior and pressure state interaction parameters; S4. Setting a long time scale prediction-short time scale correction mode, obtaining the user's historical charging behavior, obtaining weight parameters of the PSR-FM model at two time scales, the long time scale and the short time scale, and obtaining the PSR-FM model at two time scales; S5. Use the PSR-FM model under two time scales to predict the user's charging behavior, and determine the time scale for prediction in the current time period based on the prediction error of the previous time period; S6. Take the time-sharing and domain-based electricity prices as variables, and the electric vehicle battery charging status data, charging pile occupancy data and queuing data as initial values. Use the PSR-FM model under the time scale determined in step S5 to predict the user's charging behavior. Set the sum of the charging benefit and the electricity selling benefit as the objective function. Use the particle swarm algorithm to iteratively calculate the time-sharing and segmented electricity prices for each time period in the study area to ensure the user's charging benefit and the charging pile's electricity selling benefit. Complete the electricity price setting for the current time period and return to step S5.

2. The method for formulating time-based and domain-based electricity prices considering charging pressure-state-behavior according to claim 1 is characterized in that: In step S2, the charging demand is calculated according to the charging status data of the electric vehicle battery, and the charging demand is quantitatively evaluated and normalized. The specific calculation formula is as follows: ， in, is the required calculation time, Divide the time into intervals, for Moment The charging needs of the vehicle, The closer it is to 1, the greater the car's charging demand. for All vehicles at the moment SOC The maximum value in for All vehicles at the moment SOC The minimum value in for Moment Car SOC .

3. The method for formulating time-based and domain-based electricity prices considering charging pressure-state-behavior according to claim 1, characterized in that: In step S2, the optimal charging pile for the user is calculated according to the user's location and the time-sharing and domain-sharing electricity price in the electric vehicle charging behavior data, and the optimal charging pile for the user is quantitatively evaluated and normalized. The specific calculation formula is as follows: ， in, for Moment The charging cost performance of a vehicle at the optimal charging station is used to indicate the economic benefits of charging. The closer it is to 1, the more cost-effective the car charging is; for Moment Car SOC ; for Moment The price of charging the vehicle; for Moment The distance from the vehicle to its optimal charging station, Travel expenses per unit distance.

4. The method for formulating time-based and domain-based electricity prices considering charging pressure-state-behavior according to claim 1, characterized in that: In step S2, the busyness of the charging piles is calculated according to the occupancy and queuing conditions of each charging pile, and a quantitative evaluation and normalization process is performed on the busyness of the charging piles. The specific calculation formula is as follows: ； in, Moment The optimal charging station selected by the vehicle is Charging stations; for Moment The busyness of the optimal charging pile selected by the vehicle is used to indicate the busyness of the charging pile. The closer it is to 1, the more idle the charging station is; for The total number of vehicles queuing at the charging station at any given moment; for Moment Car SOC ; for Moment The charging power of each charging station; For the The maximum charging power of each charging pile; For the The total number of vehicles in the queue when the charging power of a charging pile is maximum.

5. The method for formulating time-based and domain-based electricity prices considering charging pressure-state-behavior according to claim 1, characterized in that: In step S4, a long time scale prediction-short time scale correction mode is set; The time period corresponding to the long time scale is one day, that is, 96 15-minute periods. The time period corresponding to the short time scale is 1 15-minute period. In the long time scale prediction model, The value is 96, in the short time scale correction mode, The value is 1; Since the prediction periods used in the two time scales are different, the weight parameters of the PSR-FM model in the two time scales are and They are also different; Get the user's historical charging behavior: ； Weight parameters of the PSR-FM model at two time scales; and obtaining the PSR-FM model at two time scales.

6. The method for formulating time-based and domain-based electricity prices considering charging pressure-state-behavior according to claim 5 is characterized in that: In step S5, the day-ahead data and real-time data are used to predict the user's charging behavior in the previous time period using the PSR-FM model at two time scales. and , and according to the prediction errors of the PSR-FM model at two time scales and , determine which time scale to use for prediction in the current time period. The specific calculation formula is as follows: ； ； in, is the prediction error of the PSR-FM model on a long time scale, For the current moment, for Moment The actual charging behavior of the car, is the prediction error of the PSR-FM model at short time scale; The charging behavior of users predicted by the PSR-FM model in a long time scale. The user's charging behavior predicted by the PSR-FM model in a short time scale.

7. The method for formulating time-based and domain-based electricity prices considering charging pressure-state-behavior according to claim 1, characterized in that: In step S5, if the current time scale is a long time scale, one time period is one day, that is, 96 15-minute periods; if the current time scale is a short time scale, one time period is 15 minutes.

8. The method for formulating time-based and domain-based electricity prices considering charging pressure-state-behavior according to claim 6, characterized in that: In step S6, the time-based and domain-based electricity prices are used as variables, the battery charging status data of electric vehicles, the occupancy data of each charging pile and the queue data are used as initial values, and the PSR-FM model under the time scale determined in step S5 is used to Moment Vehicle user charging behavior Make predictions; The sum of charging benefit and electricity selling benefit is taken as the objective function, and the particle swarm algorithm is used to iteratively calculate the time-sharing and segmented electricity prices in each period of the study area. The objective function calculation formula is as follows: ； ； ； in, and are the number of charging piles and the number of electric vehicle users, is the charging pile pricing coefficient, for Moment The electricity price of a charging station, for Moment The charging price of the car, The total amount of charging consumption for electric vehicle users, Optimize the objective function for the particle swarm.

9. A computer device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 8.

Images