A method and apparatus for evaluating the load space flexibility of fast charging stations

Abstract

This invention proposes a method and apparatus for evaluating the spatial flexibility of fast charging station loads, belonging to the fields of electric vehicle load regulation and virtual power plants. The method includes: clustering electric vehicle users based on historical electric vehicle trajectory data and fast charging station data to obtain the selection probability of each type of electric vehicle user for a fast charging station and the transition probability between fast charging stations, thus obtaining the spatiotemporal probability distribution of the fast charging station load; based on the spatiotemporal probability distribution, obtaining the spatial flexibility characteristics of the fast charging station load to construct a fast charging station load spatial flexibility model; and evaluating the spatial flexibility of the fast charging station load based on the fast charging station load spatial flexibility model. This invention can obtain the spatial flexibility of fast charging stations that reflects the real situation, thereby effectively alleviating grid congestion and improving the safety and economy of grid operation by utilizing the spatial flexibility of fast charging stations.

Description

Technical Field

[0001] This invention belongs to the fields of electric vehicle load regulation and virtual power plant, and specifically relates to a method and device for evaluating the load space flexibility of a fast charging station. Background Technology

[0002] To achieve carbon neutrality, the penetration rate of renewable energy in the power system continues to rise, and the penetration rate of electric vehicles (EVs) in the transportation system is also increasing. Under this trend, there is a prospect for the integrated development of electric vehicles and renewable energy. The power system needs flexible resources to support the grid connection of renewable energy, and electric vehicles constitute a potential resource for spatiotemporal flexibility.

[0003] In exploring the spatial flexibility of electric vehicles (EVs), aggregators can act as a coordination layer between the power system and EVs. This scheduling architecture avoids the following problems: first, the spatial flexibility offered by a single EV is limited; second, when considering the charging scheduling of each EV individually, the power system needs to address a complex high-dimensional optimization problem. Common aggregators include EV fleet operators and charging station operators. For uncontrolled fast-charging EVs, charging station operators can indirectly guide EV users to change their charging station choices by adjusting charging station prices, thus exploring the spatial flexibility of charging station load.

[0004] Invention CN110378548A proposes a method for constructing a multi-timescale response capability assessment model for electric vehicle virtual power plants. This method, considering the temporal states (SOC state and charging / discharging state) during EV evaluation, proposes four typical scenarios under different response modes, analyzes the power constraints and available capacity of individual EVs in continuous time segments, and proposes single-time segment response capability assessment models for individual electric vehicles. This method approximates the flexibility provided by EV fleet operators aggregating EV groups, and the source of EV flexibility is the change in EV charging / discharging state, lacking exploration of the spatial flexibility of EVs.

[0005] As aggregators of spatial flexibility, charging station operators face the first challenge: spatial flexibility must account for the uncertainty of users' choices regarding charging stations. For mobile electric vehicles with fast charging needs, spatial flexibility implies uncertainty in charging station selection. This uncertainty must be considered when navigation methods that reduce this uncertainty are not employed. Some existing methods assume that drivers use a single factor, such as the shortest travel time, to select charging stations. However, in reality, different drivers have different preferences and may consider multiple factors with varying coefficients. When considering multiple factors, some methods derive relevant information from hypothetical data rather than real-world data. In contrast, results obtained using real-world data are closest to reality, which is a prerequisite for obtaining a more realistic spatial flexibility model.

[0006] As aggregators of spatial flexibility, charging station operators face a second challenge: the spatial flexibility of electric vehicles exhibits spatial correlation characteristics that vary with price. Spatial flexibility involves load shifting between fast charging stations (FCS), and the spatial correlation of charging load changes when price incentives are applied to alter the charging station selection of electric vehicles. Some methods assume spatial correlation parameters between charging stations, which leads to the following problems: First, the assumed spatial flexibility parameters are likely to be inconsistent with reality. Second, fixed correlation parameters ignore the changes in spatial correlation under different price incentives.

[0007] As aggregators of spatial flexibility, charging station operators face a third challenge: how to assess the spatial flexibility of fast charging station loads. This flexibility assessment needs to meet the following requirements: first, establishing matching aggregation characteristics after considering the two challenges mentioned above; and second, facilitating the subsequent formation of virtual power plants and participation in grid dispatch. Solutions to this challenge can be found in existing methods for addressing similar problems. Regarding the first point, considering the two challenges, the flexibility of a single charging station is uncertain, and the flexibility among multiple FCSs exhibits spatial correlation with price changes. Regarding the second point, existing research has a foundation for quantifying risk in power system dispatch under uncertain scenarios and for power system dispatch using Copula functions. Summary of the Invention

[0008] The purpose of this invention is to overcome the shortcomings of existing technologies and propose a method and apparatus for evaluating the spatial flexibility of fast charging station loads. This invention addresses the prospect of integrated development of electric vehicles and renewable energy, using charging station operators as aggregators and leveraging price incentives to explore the spatial flexibility of electric vehicles with fast charging needs. This invention obtains the spatial correlation between user preferences and fast charging station load changes with price from real-world data, and then obtains the spatial flexibility of fast charging stations reflecting the real situation. This allows for the effective alleviation of grid congestion and the improvement of grid operation safety and economy by utilizing the spatial flexibility of fast charging stations.

[0009] A first aspect of this invention provides a method for evaluating the load space flexibility of a fast charging station, comprising:

[0010] Based on historical electric vehicle trajectory data and fast charging station data, electric vehicle users are clustered to obtain the selection probability of each type of electric vehicle user for the fast charging station.

[0011] Based on the selection probability, the transition probability of the various types of electric vehicle users choosing among the fast charging stations is calculated to obtain the spatiotemporal probability distribution of the load of the fast charging stations;

[0012] Based on the spatiotemporal probability distribution, the spatial flexibility characteristics of the fast charging station load are obtained to construct a fast charging station load spatial flexibility model.

[0013] The load space flexibility of the fast charging station is evaluated based on the aforementioned fast charging station load space flexibility model.

[0014] In a specific embodiment of the present invention, the step of clustering electric vehicle users based on historical electric vehicle trajectory data and fast charging station data to obtain the selection probability of each type of electric vehicle user for the fast charging station includes:

[0015] 1) Collect historical electric vehicle trajectory data and fast charging station data, and statistically analyze the influencing factors and selection results of each electric vehicle user in choosing each fast charging station;

[0016] Among them, let Represent any user Time period Select the Fast charging station Historical selection results, if the time period user Choose in Charge =1 means otherwise ;

[0017] 2) Based on the statistical results of step 1), establish a logistic regression model for each user to reflect the relationship between the probability of selection and the influencing factors, solve the logistic regression model, and obtain the parameters of the logistic regression model that reflect the degree of importance users attach to the influencing factors.

[0018] Among them, any user The logistic regression model is shown below:

[0019] (1)

[0020] (2)

[0021] in, Indicates influencing factors as In the case of user During the period right The probability of choosing;

[0022] and These are the parameters of the logistic regression model; To reflect user Influencing factors Weighting coefficients for the degree of importance; The coefficient of the constant term reflects the user's... To exclude The coefficient representing the degree of importance attached to other potential influencing factors;

[0023] 3) Use the results of step 2) to cluster electric vehicle users to obtain the selection probability of each type of user for each fast charging station;

[0024] Let the number of clusters be K, then the result after clustering reflects the first... User type influence factors Weighting coefficient of importance and reflect the first User class for excluding The constant term coefficient of the degree of importance attached to other potential influencing factors. , , and For the first The weight coefficients and constant coefficients corresponding to the cluster centers of each category;

[0025] Will Substituting into equation (1), the influencing factors are: In the case of the first User type during time period right Selection probability .

[0026] In one specific embodiment of the present invention, the influencing factors include: the distance between the user and the fast charging station, the average charging price of the fast charging station, the number of fast charging piles at the fast charging station, and the user's historical access probability to the fast charging station.

[0027] but , ;in, Representative time period user Location and the Fast charging station The distance between them; Representative time period user Go to The average charging price during the expected charging period; represent The number of fast charging stations; On behalf of users Select from all fast charging stations The probability of; for The weighting coefficient reflects the user's right The degree of importance attached to it; for The weighting coefficient reflects the user's right The degree of importance attached to it; for The weighting coefficient reflects the user's right The degree of importance attached to it; for The weighting coefficient reflects the user's right The degree of importance attached to it.

[0028] In a specific embodiment of the present invention, the maximum likelihood method is used to solve the logistic regression model, and the expression is as follows:

[0029] (3)

[0030] in, On behalf of users Time period choose The prediction result is expressed as follows:

[0031] (4)

[0032] in, For users In historical selection results The number, For users In historical selection results The number.

[0033] In a specific embodiment of the present invention, calculating the transition probability of the various types of electric vehicle users choosing among the fast charging stations based on the selection probability includes:

[0034] remember It is a collection of fast charging stations, the first User type during time period choose Other charging stations The probability is , Then the choice of history is No. User type during time period Still choose The probability is denoted as The choice of history is No. User type during time period The probability of choosing another charging station is , and For the first User type during time period The transition probability for selecting a fast charging station;

[0035] If the first User time period The choice is Any fast charging station And satisfy Then all those that satisfy the conditions Composition of the first User time period positive example set ,in For the first User time period choose The prediction results, if the prediction is within the time period No. User selection in Charging =1, otherwise ;

[0036] And thus obtain and The calculation expression is as follows:

[0037] (5)

[0038] (6)

[0039] in, .

[0040] In a specific embodiment of the present invention, obtaining the spatiotemporal probability distribution of the load of the fast charging station includes:

[0041] 1) Obtain spatial correlation parameters, including: the selection probability, transfer probability, and number of vehicles transferred between charging stations under different price combinations;

[0042] Recorded in the Choice of History The User time period Still choose The number of vehicles is Response transfer delay =0; historical selection The User time period from Transferred to The number of vehicles is , Indicates time period from Transferred to The time delay;

[0043] The charging station operator's price adjustment period is as follows: , This marks the start of the price adjustment period. This marks the end of the price adjustment phase. During the period The price adjustment range is ,in For time period Price adjustment cap For time period If the price adjustment floor is reached, then all fast charging stations will be subject to the following conditions: The price combination is denoted as Then the price combination index The range of values ​​is , Let be the total number of price combinations, and consider the price combinations as... middle hour, This is a combination of historical prices for charging stations;

[0044] The price combination is When, the corresponding selection probability is and The transition probability is and The number of vehicles transferred was and ;

[0045] but The calculation expression is as follows:

[0046] (7)

[0047] in, for arrive distance, For electric vehicles from time period Start from Transferred to The average speed; The set time interval;

[0048] Calculate historical price combinations for charging stations corresponding time period No. User arrival of the class Number of vehicles The expression is as follows:

[0049] (8)

[0050] (9)

[0051] (10)

[0052] in, Indicates the time period of The expected total charging load, For the time period of Total charging load, For time period The load that has finished charging and has left the facility; Indicates time period The expected charging load; For time period arrive The expected number of vehicles, for The charging station power;

[0053] according to probability distribution The three probability distributions obtained are as follows: , and ,Will Values The probability of the following is denoted as , Values The probability of the following is denoted as ;

[0054] Depend on and get All values and probability distribution , ;exist Values Down, Values The probability is obtained from equation (11). Values The probability is given by the formula get:

[0055] (11)

[0056] (12)

[0057] in, , , ;

[0058] Depend on , , get probability distribution as well as probability distribution ,but All values Values The probability of the final Values probability As shown in equation (13), the probability As shown in equation (14); obtain and After determining the probabilities for different values, we obtain... and The probability distributions are denoted as follows: and ;

[0059] (13)

[0060] (14)

[0061] 3) Obtain price combinations The following is the charging station load from the start of the period when the price adjustment is announced to the end of the period when the price adjustment is completed. The probability distribution;

[0062] Charging station operators during time periods Publicly announced time period The price combination is , , Previous period Maintain the probability distribution based on historical pricing;

[0063] The acquisition method is shown in equations (15)-(18); during the acquisition period of After that, the time period of The solution is obtained by rolling the solution of equation (8);

[0064] (15)

[0065] (16)

[0066] (17)

[0067] (18).

[0068] In a specific embodiment of the present invention, the step of obtaining the spatial flexibility characteristics of the fast charging station load based on the spatiotemporal probability distribution to construct a fast charging station load spatial flexibility model includes:

[0069] 1) Determine the characteristics of the flexibility of an individual charging station;

[0070] Among them, any price combination Down, During the period flexibility Values The probability is shown in equation (19), where the summation term satisfies constraint (20); during the time period Price combination The probability distribution of the lower flexibility is as follows The range of values ​​is denoted as , for The lower bound of the probability distribution. for The upper bound of the probability distribution;

[0071] (19)

[0072] (20)

[0073] The price combination for a single charging station Lower flexibility It is a probability distribution. The response probability is denoted as Response probability The methods for obtaining it are as follows:

[0074] (twenty one)

[0075] 2) Determine the characteristics of flexibility among multiple charging stations;

[0076] Will Record Instead, Record Then construct The joint probability distribution is used to reflect the spatial correlation of fast charging station flexibility with price variations, specifically including:

[0077] Smoothing discrete variables ;

[0078] because medium elements It is a bounded discrete variable, and its range of values ​​is... The lower bound of the value is The upper bound of the value is Gaussian kernel smoothing was used to obtain The corresponding probability density function is shown in equation (23). and the cumulative distribution function as shown in equation (24) ;in, for The number of samples, for Sample index, , For the first indivual sample, It is the bandwidth of the smoothing window, Gaussian kernel , , , ;

[0079] (twenty three)

[0080] (twenty four)

[0081] The interval between values ​​is ;

[0082] (25)

[0083] Solve for the Gaussian Copula function and its probability density function;

[0084] The dimension of the Gaussian Copula function is determined by corresponding time period and charging station The decision, its dimensions are ; Dimensional variables The cumulative distribution function is ;because ,therefore The Copula function is The joint distribution function, The estimated parameters of the Gaussian Copula are the covariance matrix between the variables. As shown in equation (26), where all elements belong to the range of -1 to 1:

[0085] (26)

[0086] After parameter estimation, the result is shown in equation (27). The expression for the Gaussian Copula function, where It is the standard normal distribution function;

[0087] (27)

[0088] The probability density function of the Gaussian Copula function is shown in equation (28), where ;

[0089] (28)

[0090] Obtain the joint probability distribution of discrete variables ;

[0091] The joint density function is shown in equation (29); Values The probability is shown in equation (30), then joint probability distribution according to The probabilities for different values ​​are obtained as follows:

[0092] (29)

[0093] (30)

[0094] 3) Based on the results of steps 1) and 2), construct a model of the load space flexibility of fast charging stations;

[0095] The fast charging station load space flexibility model includes: the flexibility characteristics of a single charging station and the flexibility characteristics of multiple charging stations; wherein, the flexibility characteristic of a single charging station is a price combination. ,flexibility Response probability The relationship between the two is given by equation (19) regarding the spatial flexibility of a single charging station. The probability distribution and the response probability of equation (21) and The relationship between the probability distributions constitutes the flexibility characteristics of the multiple charging stations, which are multi-time period charging station flexibility. The joint probability distribution is shown in equation (30).

[0096] A second aspect of the present invention provides an evaluation device for the load space flexibility of a fast charging station, comprising:

[0097] The probability acquisition module is used to cluster electric vehicle users based on historical electric vehicle trajectory data and fast charging station data, and to obtain the selection probability of each type of electric vehicle user for the fast charging station.

[0098] The spatiotemporal probability distribution acquisition module is used to calculate the transition probability of the various types of electric vehicle users choosing between the fast charging stations based on the selection probability, so as to obtain the spatiotemporal probability distribution of the load of the fast charging station.

[0099] The spatial flexibility model construction module is used to obtain the spatial flexibility characteristics of the fast charging station load based on the spatiotemporal probability distribution in order to construct a spatial flexibility model of the fast charging station load.

[0100] The evaluation module is used to evaluate the load space flexibility of the fast charging station based on the load space flexibility model of the fast charging station.

[0101] A third aspect of the present invention provides an electronic device comprising:

[0102] At least one processor; and a memory communicatively connected to said at least one processor;

[0103] The memory stores instructions that can be executed by the at least one processor, the instructions being configured to perform the aforementioned method for evaluating the load space flexibility of a fast charging station.

[0104] A fourth aspect of the present invention provides a computer-readable storage medium storing computer instructions for causing the computer to execute the above-described method for evaluating the load space flexibility of a fast charging station.

[0105] The features and beneficial effects of this invention are as follows:

[0106] This invention uses a logistic regression model to analyze the uncertainty of users' choices of charging stations, identifies model parameters using real data to obtain user preferences, and analyzes the transfer probability of fast charging station loads with price changes. It then uses probability theory to obtain the spatiotemporal distribution of fast charging station loads with price changes, and evaluates the spatial flexibility of fast charging station loads. This evaluation considers the variation of the flexibility range of a single fast charging station with response probability and the spatial correlation between multiple fast charging stations with price changes established using a Gaussian Copula function. This provides a foundation for electric vehicle fast charging loads to form virtual power plants as a flexibility resource, participate in quantitative risk power system dispatching under uncertain scenarios, and for power system dispatching after using the Copula function. This invention can fully utilize the spatial flexibility of fast charging station loads, providing flexibility resources for the power grid and contributing to the safe and economical operation of the power grid. Attached Figure Description

[0107] Figure 1 This is an overall flowchart of a method for evaluating the load space flexibility of a fast charging station according to an embodiment of the present invention.

[0108] Figure 2 This is a schematic diagram illustrating the principle of obtaining the transition probability from the selection probability in a specific embodiment of the present invention.

[0109] Figure 3 This is a price combination in a specific embodiment of the present invention. ,flexibility Response probability A diagram illustrating the relationship between them.

[0110] Figure 4 This is a specific embodiment of the flexibility of the present invention. With response probability A diagram illustrating the relationship between them.

[0111] Figure 5 This is a specific embodiment of the invention, showing charging stations under three price incentives during the time period of 12:00-13:30. A schematic diagram illustrating the range of load flexibility under different response probabilities.

[0112] Figure 6 This is a time period in a specific embodiment of the present invention. charging station and Load flexibility and A spatial relationship diagram. Detailed Implementation

[0113] This invention proposes a method and apparatus for evaluating the load space flexibility of fast charging stations, which will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0114] A first aspect of this invention provides a method for evaluating the load space flexibility of a fast charging station, comprising:

[0115] Based on historical electric vehicle trajectory data and fast charging station data, electric vehicle users are clustered to obtain the selection probability of each type of electric vehicle user for the fast charging station.

[0116] Based on the selection probability, the transition probability of the various types of electric vehicle users choosing among the fast charging stations is calculated to obtain the spatiotemporal probability distribution of the load of the fast charging stations;

[0117] Based on the spatiotemporal probability distribution, the spatial flexibility characteristics of the fast charging station load are obtained to construct a fast charging station load spatial flexibility model.

[0118] The load space flexibility of the fast charging station is evaluated based on the aforementioned fast charging station load space flexibility model.

[0119] The method described in this embodiment considers the uncertainty of user charging station selection and the resulting spatial correlation of charging station load. When using price incentives to obtain the flexibility of fast charging station load, the impact of price incentives on different users varies, thus affecting the spatial correlation of load to varying degrees. This embodiment uses charging station operators as aggregators. In the process of obtaining charging station load flexibility, it fully considers the different preferences of users in choosing charging stations, the transfer probability of charging station loads with price changes among users with different preferences, and the spatiotemporal probability distribution of fast charging station load flexibility. This provides a foundation for electric vehicle fast charging loads to form virtual power plants as a flexibility resource, participate in quantitative risk power system dispatching under uncertain scenarios, and perform power system dispatching after using Copula functions.

[0120] In a specific embodiment of the present invention, the overall process of the method for evaluating the load space flexibility of a fast charging station is as follows: Figure 1 As shown, it includes the following steps:

[0121] 1) Based on the collected historical electric vehicle trajectory data and fast charging station data, a logistic regression model is established to analyze the charging station selection preferences of electric vehicle users. Based on the solution results of the logistic regression model, electric vehicle users are clustered to obtain the selection probability of each type of user for each fast charging station.

[0122] 1-1) Collect electric vehicle trajectory data and fast charging station data within the electric vehicle driving area, and statistically analyze the influencing factors and selection results of each electric vehicle user in choosing each fast charging station.

[0123] In this embodiment, the electric vehicle trajectory data includes: time, the longitude and latitude of the electric vehicle at the corresponding time, vehicle status (parking, driving, charging), and battery level. The electric vehicle trajectory data needs to have a certain time span and appropriate time intervals. The time span is to ensure that multiple selection behaviors of a vehicle can be collected, and the time interval is to avoid failing to track vehicle status changes in a timely manner. The fast charging station data includes: the longitude and latitude of the charging station location, the time-of-use price of the charging station (charging fee + service fee), the number of fast charging piles, and the power of the fast charging piles. Only the time-of-use price is dynamic data.

[0124] In one specific embodiment of the present invention, trajectory data of 170 electric vehicles with a time interval of 10 seconds over 100 days and data of all fast charging stations within the driving area of ​​the electric vehicles are obtained as raw data. The raw data is preprocessed to reproduce the user's selection scenario. Fast charging behavior is selected based on the rate of change of battery level over time when the vehicle is in a charging state, as shown in this embodiment. Fast charging behavior is defined as charging power of 25kW or higher and charging time of less than 2 hours. When there is a need for fast charging, the user decides which fast charging station to choose by comparing factors that influence the selection. Table 1 illustrates the time periods. user The user faces a choice scenario and historical choice results. The user's complete historical choice table includes all Fast Charging Station (FCS) selection behaviors extracted from trajectory data.

[0125]

[0126] in, It is an index of fast charging stations. , For the number of fast charging stations, It is an index for electric vehicle users. , It refers to the number of users (i.e., the number of electric vehicles). This is a time interval index. Since a user's fast charging session typically lasts for half an hour, this embodiment sets a time interval. It takes 5 minutes, so within a day The first row in Table 1 contains the serial numbers of the charging stations to be selected, and the second to fifth rows contain statistics on the four factors that influence the user's selection in this embodiment. Representative time period user Location relative to the i-th fast charging station The distance between them. Derived from user data in vehicle trajectory data. Time period Latitude and longitude information and charging station data The latitude and longitude information is obtained through processing. If the time period represents user Go to Charging, the average charging price during the expected charging period. (Based on time period) user Location and battery level, and The power output of the fast charging stations can predict the charging destinations users will visit. The charging period corresponding to charging, combined with During the period price From this, we can know the average price during the expected charging period. . represent The number of fast charging stations can be obtained directly from the charging station information. In actual operation, charging station operators can consider time periods. The impact of usage time correspond The number of available fast charging stations will be used as an influencing factor for a more precise analysis. On behalf of users Select from all fast charging stations The probability, based on the user Total number of fast charging actions and selections The number of times is obtained. (Selection) The number of times a user can be charged can be determined based on the correspondence between the vehicle's latitude and longitude and the fast charging station's latitude and longitude. The fast charging station selected each time. (Last line) On behalf of users Time period choose The result of historical selection, =1 indicates the time period user Choose in Charge it, otherwise .

[0127] 1-2) Using the results of step 1-1), establish a logistic regression model for each user that reflects the relationship between the probability of selection and influencing factors. Solve the logistic regression model to obtain the parameters of the logistic regression model that reflect the user's emphasis on the influencing factors. The specific steps are as follows:

[0128] 1-2-1) Using the results of step 1-1), establish a logistic regression model for each user;

[0129] In this embodiment, any user The logistic regression model is shown below:

[0130] (1)

[0131] (2)

[0132] Mode The logistic regression model shown reflects the probability of choice. Influencing factors The relationship between the probabilities of choice. As defined by equation (2), the influencing factors are: In the case of user During the period right The probability of selection. and These are the parameters to be determined for this model, where For corresponding The weighting coefficients. for The weighting coefficient reflects the user's Distance factors The degree of importance attached to it; for The weighting coefficient reflects the user's... Price factors The degree of importance attached to it; for The weighting coefficient reflects the user's Number of fast charging stations The degree of importance attached to it; for The weighting coefficient reflects the user's Historical preferences The degree of importance attached to it; The coefficient of the constant term reflects the user's... A coefficient representing the degree of importance attached to other potential influencing factors not listed.

[0133] Logistic regression models are simple to train, highly interpretable, and most importantly, they can be analyzed using the weight coefficients of each factor. Seeing the impact of different factors on the probability of choice The impact of this helps charging station operators clarify the influence of price factors on user choices. middle As the value increases, the probability of selection... As it changes, therefore Reflecting average price The impact on the probability of user selection.

[0134] 1-2-2) Solve the logistic regression model established in step 1-2-1) to obtain the logistic regression model parameters that reflect the degree of importance users attach to influencing factors;

[0135] The parameters to be determined in the model are those in equation (1). and In this embodiment, the maximum likelihood method shown in equation (3) is used to estimate the parameters. The key point is to let the prediction result... This is the actual result. The probability is highest for that. On behalf of users Time period choose The prediction results are as follows. The model solution expression is as follows:

[0136] (3)

[0137] Among them, the prediction results From the probability of choice The method for obtaining the result is shown in equation (4). If the predicted probability is not less than the observed probability, then the prediction result is... The value is 1 if the probability is positive and 0 otherwise. The predicted probability is the same as the selection probability. With the probability of not choosing The ratio, the observation probability is the user Complete history selection table number and number The ratio.

[0138] (4)

[0139] 1-3) Use the results of step 1-2) to cluster electric vehicle users to obtain the selection probability of each type of user for each fast charging station.

[0140] In this embodiment, based on the model in steps 1-2), the weighting coefficients reflecting the degree of importance all users attach to each influencing factor are denoted as follows: .

[0141] In this embodiment, due to the need to protect user privacy, it is not possible to obtain the trajectory data of all vehicles in the charging station's operating area, nor is it possible to obtain the selection preferences of all users in the operating area. To solve this problem, in obtaining... Then, the weight coefficients of the sample users (i.e., the users who collected the data) are clustered using the K-means clustering method. The clustering results and the proportion of each category are used as the corresponding weight coefficients for all users in the charging station's operating area. K-means uses minimizing the squared error between the sample and the cluster center as the objective function. The sum of the squared distance errors between the cluster center and the sample points within the cluster is called the distortion degree. For a cluster, the lower the distortion degree, the more compact the members within the cluster; the higher the distortion degree, the looser the cluster structure.

[0142] In this embodiment, the elbow method is used to determine the number of clusters K during clustering. For data with a certain degree of discriminative power, the distortion level will be greatly improved when a certain critical number of clusters is reached, and then slowly decrease. This critical number of clusters can be considered as the final selected number of clusters K. After clustering, the user's model parameters are divided into K classes, denoted as... , Indexing by user category, and For the first The weight coefficients and constant term coefficients for each user class are the corresponding coefficients for each cluster center. , . No. Number of samples within a cluster (i.e., the first) The proportion of users in category (i.e., total number of users) to the total sample is the [number of users in category 1]. Proportion of users of this type .Will Substitution Various user preferences are obtained from this. , Influencing factors are In the case of the first User type during time period right The probability of selection. For the ... User class, using Predict their charging station selection results. For the first User time period choose The prediction results, if the prediction is within the time period No. User selection in Charging =1, otherwise .

[0143] 2) Based on the selection probability obtained in step 1), calculate the transition probability of various types of users choosing between different fast charging stations to obtain the spatiotemporal probability distribution of the fast charging station load.

[0144] Uncertainty in user charging station selection and price incentives both lead to load shifting. The shifting probability between FCS loads can be obtained from the selection probability in step 1). Based on the shifting probability, probability theory is used to handle the uncertainty to obtain the spatiotemporal probability distribution of FCS loads. Then, charging station operators use price incentives to influence the shifting probability, changing the spatial correlation of FCSs, thereby obtaining the spatiotemporal probability distribution of charging station spatial flexibility. The specific steps are as follows:

[0145] 2-1) Based on the selection probabilities of each type of user in choosing each fast charging station and the historical selection results, calculate the transition probability of each type of user in choosing between different fast charging stations.

[0146] In this embodiment, (the following is a record) It is a collection of fast charging stations within the electric vehicle driving area, the first User type during time period Historical Choice The probability is This type of user chooses Other charging stations The probability is , Index for charging stations, but This is used to distinguish different charging stations. So, the historical choice is... No. User type during time period Still choose The probability is denoted as The choice of history is No. User type during time period The probability of choosing another charging station is . and For the first User type during time period The transition probability for selecting a fast charging station.

[0147] Figure 2 This embodiment explains the selection probability. and Obtain the transition probability and The selection probability corresponds to the left rounded rectangle. and And there are , According to equation (4), if corresponding or corresponding ,So or Belongs to the set of positive examples ,therefore For time period All selections The For users, it satisfies A collection of charging stations. Users only consider [the following] when selecting an FCS. In FCS, when When the number of FCSs is greater than 1, the user's FCS selection becomes uncertain. Therefore, it is necessary to adjust the selection probability. and After normalization, the transition probabilities of the rounded rectangle on the right side of the figure are obtained. and See equations (5) and (6) for details, and there are During the time period The choice of history is The User class based on transition probability and choose The charging stations in the system, considering the uncertainty, show that the selection results conform to a multinomial distribution.

[0148] (5)

[0149] (6)

[0150] 2-2) Based on the transition probability and probability theory obtained in step 2-1), the spatiotemporal probability distribution of the FCS load is obtained. The specific steps are as follows:

[0151] 2-2-1) Obtain spatial correlation parameters, including: the selection probability, transfer probability, and number of vehicles transferred between charging stations under different price combinations.

[0152] In this embodiment, the number of EV transfers between the constructed FCSs is expressed as follows: and The corresponding transfer delay is expressed as and . Representatives in historical choices The User time period Still choose Number of vehicles, response transfer delay It is always 0; Representing historical choices The User time period from Transferred to Number of vehicles Indicates time period from Transferred to The time delay.

[0153] In this embodiment, if the vehicle transfer delay is less than a set time interval... (In this embodiment, the value is 5 minutes), so the transfer delay can be ignored. . The influencing factor is real-time traffic conditions. and The influencing factor is the price combination of all charging stations. The period for adjusting charging station operators' prices is set as follows: , This marks the start of the price adjustment period. This marks the end of the price adjustment phase. During the period The price adjustment range is With multiple operators competing, high prices can lose their appeal to users, thus leading to periods of low prices. Price adjustment cap The purpose of charging station operators providing flexibility is to generate revenue, therefore there are time periods. Price adjustment floor To simplify decision-making for charging station operators, the following measures are set: During the adjustment period Adjust internally to the same value. Set an appropriate price adjustment interval and record... The number of possible values ​​within the adjustment range is During the price adjustment period and within the adjustment range, all fast charging stations... The price combination is denoted as Then the price combination index The range of values ​​is , Let be the total number of price combinations, and consider the price combinations as... middle hour, This is a historical price combination for charging stations. Price Combination The average price over the selected period is determined by probability. and Affects the transition probability and This, in turn, affects and The price combination is When, the corresponding selection probability is and The transition probability is and The number of vehicles transferred was and .

[0154] It can be obtained from equation (7). Wherein, for arrive Distance, in km. For electric vehicles from time period Start from Transferred to The average speed, expressed in km / min. Affected by real-time traffic conditions, therefore It is a time variable. This represents rounding down the quantity.

[0155] (7)

[0156] and The methods for obtaining it are as follows. and It is not a fixed value, but a discrete probability distribution. Used to obtain... and The known information consists of two parts: first, information that charging station operators can obtain based on historical charging loads. The probability distribution, Historical price combinations Corresponding to time The expected total charging load; secondly, in the price mix The transition probability below and .

[0157] Depend on It can be obtained The probability distribution. Indicates the historical price combination of charging stations corresponding time period No. User arrival of the class The number of vehicles is obtained from equations (8) to (10).

[0158] (8)

[0159] (9)

[0160] (10)

[0161] in, Indicates the time period of The expected total charging load, For the time period of Total charging load, For time period The load that has finished charging and has left the facility; Indicates time period The expected charging load; For time period arrive The expected number of vehicles, for The charging station power. and During the period Given a known quantity. Substitute the known information. probability distribution The probability distributions of the three quantities can be obtained sequentially, including: , and . Specific value The probability of the following is denoted as . Specific value The probability of the following is denoted as .

[0162] Depend on and It can be obtained All values and probability distribution , Historical Choices The User time period choose middle and Therefore, according to the multinomial distribution formula and the conditional probability formula, Any specific value Down, Values The probability is obtained from equation (11). Values The probability is obtained from equation (12).

[0163] (11)

[0164] (12)

[0165] in, , , .

[0166] Depend on , , It can be obtained and probability distribution and According to the law of total probability, All values Values The probability is the final one. Values probability See equation (13). Similarly, probability See equation (14) for the method of acquisition. and After determining the probabilities for different values, we can obtain... and The probability distribution of is denoted as . and .

[0167] (13)

[0168] (14)

[0169] 2-2-3) Obtaining Price Combinations Next period charging station load The probability distribution.

[0170] Charging station operators during time periods Publicly announced time period The price combination is , The specific value depends on the time period of publication. Previous period Maintain the probability distribution based on historical pricing. Power forecast termination period. This requires considering the impact of the price adjustment period on the average price of subsequent periods. As shown in equation (9), can be Obtained. And The acquisition method is shown in equations (15) to (18). During the acquisition period... of after, Based on time period The charging status of all vehicles is estimated. Therefore, the time period... of It can be obtained by rolling the solution of equation (8).

[0171] (15)

[0172] (16)

[0173] (17)

[0174] (18)

[0175] Equation (15) shows that when considering time delay, the time period satisfies And the transfer delay satisfies all of equation (16). The sum of them is The probability distribution. The method of obtaining it is shown in equation (12). Equation (17) indicates that... It consists of two parts: the prediction obtained from equation (11) will still be selected vehicles and historical price levels The vehicle is expected to choose Number of vehicles Equation (18) shows the complete... From all categories The sum is obtained.

[0176] 3) Based on the results of step 2), obtain the spatial flexibility characteristics of the fast charging station load to construct a spatial flexibility model of the fast charging station load.

[0177] In this embodiment, to evaluate the flexibility resources of the aggregated FCS load, charging station operators need to establish the flexibility characteristics of the fast charging station load space. This needs to consider characteristics that account for user choice uncertainty and the spatial correlation of price changes, and should facilitate integration into power system dispatch. The flexibility characteristics of the fast charging station load space consist of two parts: the flexibility characteristics of an individual charging station and the flexibility characteristics among multiple charging stations. The flexibility characteristics of an individual charging station mainly focus on the relationship between price, flexibility, and response probability, employing a flexibility range that varies with response probability. The flexibility characteristics among multiple charging stations are the spatial correlation of the load flexibility of multiple fast charging stations with price changes, i.e., the joint probability distribution of the fast charging station load flexibility obtained by the Gaussian Copula function.

[0178] 3-1) Determine the characteristics of the flexibility of a single charging station;

[0179] In this embodiment, based on the results of step 2), the spatial flexibility is first obtained. The spatiotemporal correlation probability distribution. Price combination. Down, During the period flexibility Values The probability is shown in equation (19), where the summation term needs to satisfy constraint (20). During the time period... Price combination The probability distribution of the lower flexibility is as follows The range of values ​​is denoted as , for The lower bound of the probability distribution. for The upper bound of the probability distribution. Similarly, the probability distribution of flexibility under different price combinations can be obtained. .

[0180] (19)

[0181] (20)

[0182] Single charging station price combination Lower flexibility It is a probability distribution. Response probability Defined to facilitate grid access and dispatch, its meaning is within a time period. The price combination is If flexibility is Then the response probability is denoted as Because the single-point probability of flexibility is not very practical, the response requirements for charging station flexibility are relaxed here. If the actual flexibility is greater than the required flexibility, then the adjustment requirement is considered met. Under this setting, according to The relationship with zero results in three possible response intervals. Response probability. The method of obtaining it is shown in equation (21):

[0183] (twenty one)

[0184] The transformation process of equation (21) will change the probability distribution The response probability is obtained by summing the probabilities within the response interval. This reveals the required flexibility. The smaller the absolute value, the wider the corresponding interval, and the higher the probability of a response satisfying the interval. The higher.

[0185] In this embodiment, the flexibility of a single charging station is as follows: Figure 3The price combination shown ,flexibility Response probability The relationship between them. Price combinations. As the dependent variable, flexibility and response probability The independent variable is . For all price combinations The range of values ​​is ,in, This represents the minimum value for flexibility across all price combinations. This represents the maximum flexibility value across all price combinations. Taking point A as an example, the price combination is... At that time, if the flexibility is Then the response probability Approaching zero.

[0186] A load flexibility range that varies with response probability is adopted to facilitate grid connection and dispatch. Figure 4 This is a specific embodiment of the flexibility of the present invention. With response probability A diagram illustrating the relationship between them. Figure 4 In the middle, the left side illustrates the transformation process embodied in formula (22), while the right side illustrates flexibility. With response probability Changes. If the response probability Therefore, the range of flexibility corresponds to the span of the dashed line segment in the diagram. Response probability The larger it is, the less flexible it becomes.

[0187] (twenty two)

[0188] 3-2) Determine the characteristics of flexibility among multiple charging stations;

[0189] In this embodiment, the definition is... The joint probability distribution is used to reflect the spatial correlation of fast charging station flexibility with price changes, where for Each charging station during the period flexibility Obtained in 3-1) The samples were used to construct a Gaussian Copula function. The subsequent steps included smoothing discrete variables. Solve for the Gaussian Copula function and obtain the joint probability distribution of the discrete variables. For the sake of simplicity, the price combination will no longer be emphasized in these three steps. ,use replace ,use replace .

[0190] 3-2-1) Smoothing discrete variables ;

[0191] because medium elements It is a bounded discrete variable, and its range of values ​​is... The lower bound of the value is The upper bound of the value is The Coupla function is for continuous variables, therefore Gaussian kernel smoothing is used to obtain the result. The corresponding probability density function (23) and the cumulative distribution function of equation (24) .in, for The number of samples, for Sample index, , For the first indivual sample, It is the bandwidth of the smoothing window, Gaussian kernel , , , .

[0192] (twenty three)

[0193] (twenty four)

[0194] The interval between values ​​is According to observations, the approximate relationship of equation (25) is a reasonable assumption.

[0195] (25)

[0196] 3-2-2) Solve for the Gaussian Copula function and its probability density function;

[0197] The dimension of the Gaussian Copula function is determined by corresponding time period and charging station The decision, its dimensions are . Dimensional variables Their cumulative distribution function is .because ,therefore The Copula function is what establishes... The joint distribution function. The estimated parameters of the Gaussian Copula are the covariance matrix between the variables. For details, see equation (26), where all elements belong to the range of -1 to 1.

[0198] (26)

[0199] After parameter estimation, the result is shown in equation (27). The functional expression of the Gaussian Copula, where It is the standard normal distribution function.

[0200] (27)

[0201] The above The probability density function of the Gaussian Copula function is shown in equation (28), where .

[0202] (28)

[0203] 3-2-3) Obtain the joint probability distribution of discrete variables ;

[0204] The joint density function is given by equation (29). Then, based on equation (25), it is assumed that... Values The probability is shown in equation (30). Therefore, joint probability distribution According to The probabilities are obtained under different values.

[0205] (29)

[0206] (30)

[0207] 3-3) Based on the results of steps 3-1) and 3-2), construct a model of the load space flexibility of fast charging stations;

[0208] In this embodiment, the final fast charging station load space flexibility model includes: the flexibility characteristic of a single charging station is the price combination. ,flexibility Response probability The relationship between them is specifically determined by equation (19) regarding the spatial flexibility of a single charging station. The probability distribution and response probability of equation (21) and The relationship between probability distributions constitutes the flexibility characteristics of multiple charging stations, which are the flexibility of multiple charging stations across multiple time periods. The joint probability distribution is shown in equation (30).

[0209] 4) Based on the fast charging station load space flexibility model, evaluate the load space flexibility of each fast charging station under different price combinations.

[0210] In this embodiment, it is assumed that the power grid congestion occurs at The affiliated substation, through continuous improvement The charging station's pricing from 12:30 to 13:00 leverages the spatial flexibility of the fast charging station's load to alleviate grid congestion. Figure 5 The exhibition showcased charging stations under different price incentives during the 12:00-13:30 time period. Results for the range of load flexibility under different response probabilities. The first subplot corresponds to... After the charging price increases by 0.1 yuan / kWh, the response probability, taking values ​​uniformly in intervals of 0.1 between 0.1 and 1.0, The load flexibility changes over time. Similarly, the second subplot corresponds to... The charging price will increase by 0.5 yuan per kilowatt-hour. The third sub-image corresponds to... The charging price increases by 1.0 yuan per kilowatt-hour. The graph shows that, at the same price, a higher probability of demand response guarantees better service. The smaller the load flexibility range, the less effective the price incentives become, and for the same response probability. Load flexibility will increase. In scenarios involving risk-based dispatching under uncertainty, charging station operators can better assess the risks and benefits of participating in grid dispatching based on the proposed response probabilities. Simultaneously, when grid congestion occurs... When it belongs to a substation, it can be increased The price of charging stations indirectly guides users to change their choice of charging stations, and the spatial flexibility of the load can alleviate the congestion of substations.

[0211] To showcase Spatial correlation results of charging station load flexibility Figure 6 Using two-dimensional data as an example, the time period is shown. and flexibility and The correlation between them, i.e. and The discrete joint probability distribution, and the corresponding off-diagonal elements of the Gaussian Copula covariance matrix are: It can be observed that the probability values ​​are higher on the negative diagonal of the horizontal plane formed by the two variables, indicating that during the time period... and The flexibility of [the product / service] has a negative correlation. and The probability of both taking smaller or larger values ​​is relatively low. Therefore, when actually utilizing the flexibility of electric vehicles, the spatiotemporal correlation between charging station load flexibility and price changes cannot be ignored, and different flexibility ranges should be assessed based on different pricing. When utilizing the spatial flexibility of load to alleviate substation congestion, the correlation changes of associated charging station loads should be considered to assess the risks and benefits of participating in grid dispatch.

[0212] To achieve the above embodiments, a second aspect of the present invention provides an evaluation device for the load space flexibility of a fast charging station, comprising:

[0213] The probability acquisition module is used to cluster electric vehicle users based on historical electric vehicle trajectory data and fast charging station data, and to obtain the selection probability of each type of electric vehicle user for the fast charging station.

[0214] The spatiotemporal probability distribution acquisition module is used to calculate the transition probability of the various types of electric vehicle users choosing between the fast charging stations based on the selection probability, so as to obtain the spatiotemporal probability distribution of the load of the fast charging station.

[0215] The spatial flexibility model construction module is used to obtain the spatial flexibility characteristics of the fast charging station load based on the spatiotemporal probability distribution in order to construct a spatial flexibility model of the fast charging station load.

[0216] The evaluation module is used to evaluate the load space flexibility of the fast charging station based on the load space flexibility model of the fast charging station.

[0217] It should be noted that the aforementioned explanation of an embodiment of a method for evaluating the spatial flexibility of a fast charging station load also applies to an evaluation device for the spatial flexibility of a fast charging station load in this embodiment, and will not be repeated here. According to an embodiment of the present invention, an evaluation device for the spatial flexibility of a fast charging station load clusters electric vehicle users based on historical electric vehicle trajectory data and fast charging station data to obtain the selection probability of each type of electric vehicle user for the fast charging station; based on the selection probability, it calculates the transition probability of each type of electric vehicle user choosing between the fast charging stations to obtain the spatiotemporal probability distribution of the fast charging station load; based on the spatiotemporal probability distribution, it obtains the spatial flexibility characteristics of the fast charging station load to construct a fast charging station load spatial flexibility model; and based on the fast charging station load spatial flexibility model, it evaluates the spatial flexibility of the fast charging station load. This allows charging station operators to act as aggregators, using price incentives to tap into the spatial flexibility of electric vehicles with fast charging needs, effectively alleviating grid congestion and improving the safety and economy of grid operation.

[0218] To implement the above embodiments, a third aspect of the present invention provides an electronic device, comprising:

[0219] At least one processor; and a memory communicatively connected to said at least one processor;

[0220] The memory stores instructions that can be executed by the at least one processor, the instructions being configured to perform the aforementioned method for evaluating the load space flexibility of a fast charging station.

[0221] To implement the above embodiments, a fourth aspect of the present invention provides a computer-readable storage medium storing computer instructions for causing the computer to execute the above-described method for evaluating the load space flexibility of a fast charging station.

[0222] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0223] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform a method for evaluating the load space flexibility of a fast charging station according to the above embodiments.

[0224] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0225] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0226] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0227] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the function involved, as will be understood by those skilled in the art to which embodiments of this application pertain.

[0228] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0229] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0230] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0231] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0232] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A method for evaluating the load space flexibility of a fast charging station, characterized in that, include: Based on historical electric vehicle trajectory data and fast charging station data, electric vehicle users are clustered to obtain the selection probability of each type of electric vehicle user for the fast charging station, including: 1) Collect historical electric vehicle trajectory data and fast charging station data, and statistically analyze the influencing factors and selection results of each electric vehicle user in choosing each fast charging station; 2) Based on the statistical results of step 1), establish a logistic regression model for each user to reflect the relationship between the probability of selection and the influencing factors, solve the logistic regression model, and obtain the parameters of the logistic regression model that reflect the degree of importance users attach to the influencing factors. 3) Use the results of step 2) to cluster electric vehicle users to obtain the selection probability of each type of user for each fast charging station; Based on the selection probability, the transition probability of the various types of electric vehicle users choosing between the fast charging stations is calculated to obtain the spatiotemporal probability distribution of the fast charging station load, including: 1) Obtain spatial correlation parameters, including: the selection probability, transfer probability, and number of vehicles transferred between charging stations under different price combinations; 2) Obtain the probability distribution of charging station load from the start of the period when the price adjustment is announced to the end of the period when the price adjustment is completed, under the price combination; Based on the spatiotemporal probability distribution, the spatial flexibility characteristics of the fast charging station load are obtained to construct a fast charging station load spatial flexibility model; wherein: the fast charging station load spatial flexibility model includes: the flexibility characteristics of a single charging station and the flexibility characteristics of multiple charging stations; wherein, the flexibility characteristics of a single charging station are the relationship between price combination, flexibility, and response probability; the flexibility characteristics of multiple charging stations are the joint probability distribution of the flexibility of multiple charging stations in multiple time periods; the flexibility of a single charging station under price combination is a probability distribution, and the response probability is obtained by accumulating the probability distribution within the response interval; The load space flexibility of the fast charging station is evaluated based on the aforementioned fast charging station load space flexibility model.

2. The method according to claim 1, characterized in that, The process of clustering electric vehicle users based on historical electric vehicle trajectory data and fast charging station data to obtain the selection probability of each type of electric vehicle user for the fast charging station includes: 1) Collect historical electric vehicle trajectory data and fast charging station data, and statistically analyze the influencing factors and selection results of each electric vehicle user in choosing each fast charging station; Among them, let Represent any user Time period Select the Fast charging station Historical selection results, if the time period user Choose in Charge =1 means otherwise ; 2) Based on the statistical results of step 1), establish a logistic regression model for each user to reflect the relationship between the probability of selection and the influencing factors, solve the logistic regression model, and obtain the parameters of the logistic regression model that reflect the degree of importance users attach to the influencing factors. Among them, any user The logistic regression model is shown below: （1） （2） in, Indicates influencing factors as In the case of user During the period right The probability of choosing; and These are the parameters of the logistic regression model; To reflect user Influencing factors Weighting coefficients for the degree of importance; The coefficient of the constant term reflects the user's... To exclude The coefficient representing the degree of importance attached to other potential influencing factors; 3) Use the results of step 2) to cluster electric vehicle users to obtain the selection probability of each type of user for each fast charging station; Let the number of clusters be K, then the result after clustering reflects the first... User type influence factors Weighting coefficient of importance and reflect the first User class for excluding The constant term coefficient of the degree of importance attached to other potential influencing factors. , , and For the first The weight coefficients and constant coefficients corresponding to the cluster centers of each category; Will Substituting into equation (1), the influencing factors are: In the case of the first User type during time period right Selection probability .

3. The method according to claim 2, characterized in that, The influencing factors include: the distance between the user and the fast charging station, the average charging price of the fast charging station, the number of fast charging piles at the fast charging station, and the user's historical access probability to the fast charging station. but , ;in, Representative time period user Location and the Fast charging station The distance between them; Representative time period user Go to The average charging price during the expected charging period; represent The number of fast charging stations; On behalf of users Select from all fast charging stations The probability of; for The weighting coefficient reflects the user's right The degree of importance attached to it; for The weighting coefficient reflects the user's right The degree of importance attached to it; for The weighting coefficient reflects the user's right The degree of importance attached to it; for The weighting coefficient reflects the user's right The degree of importance attached to it.

4. The method according to claim 2, characterized in that, The maximum likelihood method is used to solve the logistic regression model, and the expression is as follows: （3） in, On behalf of users Time period choose The prediction result is expressed as follows: （4） in, For users In historical selection results The number, For users In historical selection results The number.

5. The method according to claim 2, characterized in that, The step of calculating the transition probability of various types of electric vehicle users choosing between the fast charging stations based on the selection probability includes: remember It is a collection of fast charging stations, the first User type during time period choose Other charging stations The probability is , Then the choice of history is No. User type during time period Still choose The probability is denoted as The choice of history is No. User type during time period The probability of choosing another charging station is , and For the first User type during time period The transition probability for selecting a fast charging station; If the first User time period The choice is Any fast charging station And satisfy Then all those that satisfy the conditions Composition of the first User time period positive example set ,in For the first User time period choose The prediction results, if the prediction is within the time period No. User selection in Charging =1, otherwise ; And thus obtain and The calculation expression is as follows: （5） （6） in, .

6. The method according to claim 5, characterized in that, The process of obtaining the spatiotemporal probability distribution of the load of the fast charging station includes: 1) Obtain spatial correlation parameters, including: the selection probability, transfer probability, and number of vehicles transferred between charging stations under different price combinations; Recorded in the Choice of History The User time period Still choose The number of vehicles is Response transfer delay =0; historical selection The User time period from Transferred to The number of vehicles is , Indicates time period from Transferred to The time delay; The charging station operator's price adjustment period is as follows: , This marks the start of the price adjustment period. This marks the end of the price adjustment phase. During the period The price adjustment range is ,in For time period Price adjustment cap For time period If the price adjustment floor is reached, then all fast charging stations will be subject to the following conditions: The price combination is denoted as Then the price combination index The range of values ​​is , Let be the total number of price combinations, and consider the price combinations as... middle hour, This is a combination of historical prices for charging stations; The price combination is When, the corresponding selection probability is and The transition probability is and The number of vehicles transferred was and ; but The calculation expression is as follows: （7） in, for arrive distance, For electric vehicles from time period Start from Transferred to The average speed; The set time interval; Calculate historical price combinations for charging stations corresponding time period No. User arrival of the class Number of vehicles The expression is as follows: （8） （9） （10） in, Indicates the time period of The expected total charging load, For the time period of Total charging load, For time period The load that has finished charging and has left the facility; Indicates time period The expected charging load; For time period arrive The expected number of vehicles, for The charging station power; according to probability distribution The three probability distributions obtained are as follows: , and ,Will Values The probability of the following is denoted as , Values The probability of the following is denoted as ; Depend on and get All values and probability distribution , ;exist Values Down, Values The probability is obtained from equation (11). Values The probability is obtained from equation (12): （11） （12） in, , , ; Depend on , , get probability distribution as well as probability distribution ,but All values Values The probability of the final Values probability As shown in equation (13), the probability As shown in equation (14); obtain and After determining the probabilities for different values, we obtain... and The probability distributions are denoted as follows: and ; （13） （14） 2) Obtain price combinations The following is the charging station load from the start of the period when the price adjustment is announced to the end of the period when the price adjustment is completed. The probability distribution; Charging station operators during time periods Publicly announced time period The price combination is , , Previous period Maintain the probability distribution based on historical pricing; This marks the start of the price adjustment period. This marks the end of the price adjustment phase. The acquisition method is shown in equations (15)-(18); during the acquisition period of After that, time period of The solution is obtained by rolling the solution of equation (8); This is the end period for power prediction; （15） （16） （17） （18）。 7. The method according to claim 6, characterized in that, The step of obtaining the spatial flexibility characteristics of the fast charging station load based on the spatiotemporal probability distribution to construct a fast charging station load spatial flexibility model includes: 1) Determine the characteristics of the flexibility of an individual charging station; Among them, any price combination Down, During the period flexibility Values The probability is shown in equation (19), where the summation term satisfies constraint (20); during the time period Price combination The probability distribution of the lower flexibility is as follows The range of values ​​is denoted as , for The lower bound of the probability distribution. for The upper bound of the probability distribution; （19） （20） The price combination for a single charging station Lower flexibility It is a probability distribution. The response probability is denoted as Response probability The methods for obtaining it are as follows: （21） 2) Determine the characteristics of flexibility among multiple charging stations; Will Record Instead, Record Then construct The joint probability distribution is used to reflect the spatial correlation of fast charging station flexibility with price variations, specifically including: Smoothing discrete variables ; because medium elements It is a bounded discrete variable, and its range of values ​​is... The lower bound of the value is The upper bound of the value is Gaussian kernel smoothing was used to obtain The corresponding probability density function is shown in equation (23). and the cumulative distribution function as shown in equation (24) ;in, for The number of samples, for Sample index, , For the first indivual sample, It is the bandwidth of the smoothing window, Gaussian kernel , , , ; （23） （24） The interval between values ​​is ; （25） Solve for the Gaussian Copula function and its probability density function; The dimension of the Gaussian Copula function is determined by corresponding time period and charging station The decision, its dimensions are ; Dimensional variables The cumulative distribution function is ;because ,therefore The Copula function is The joint distribution function, The estimated parameters of the Gaussian Copula are the covariance matrix between the variables. As shown in equation (26), where all elements belong to the range of -1 to 1: （26） After parameter estimation, the result is shown in equation (27). The expression for the Gaussian Copula function, where It is the standard normal distribution function; （27） The probability density function of the Gaussian Copula function is shown in equation (28), where ; （28） Obtain the joint probability distribution of discrete variables ; The joint density function is shown in equation (29); Values The probability is shown in equation (30), then joint probability distribution according to The probabilities for different values ​​are obtained as follows: （29） （30） 3) Based on the results of steps 1) and 2), construct a model of the load space flexibility of fast charging stations; The fast charging station load space flexibility model includes: the flexibility characteristics of a single charging station and the flexibility characteristics of multiple charging stations; wherein, the flexibility characteristic of a single charging station is a price combination. ,flexibility Response probability The relationship between the two is given by equation (19) regarding the spatial flexibility of a single charging station. The probability distribution and the response probability of equation (21) and The relationship between the probability distributions constitutes the flexibility characteristics of the multiple charging stations, which are multi-time period charging station flexibility. The joint probability distribution is shown in equation (30).

8. An evaluation apparatus for the load space flexibility of a fast charging station performing the method as described in any one of claims 1-7, characterized in that, include: The selection probability acquisition module is used to cluster electric vehicle users based on historical electric vehicle trajectory data and fast charging station data, and obtain the selection probability of each type of electric vehicle user for the fast charging station. The spatiotemporal probability distribution acquisition module is used to calculate the transition probability of the various types of electric vehicle users choosing between the fast charging stations based on the selection probability, so as to obtain the spatiotemporal probability distribution of the load of the fast charging station. The spatial flexibility model construction module is used to obtain the spatial flexibility characteristics of the fast charging station load based on the spatiotemporal probability distribution in order to construct a spatial flexibility model of the fast charging station load. The evaluation module is used to evaluate the load space flexibility of the fast charging station based on the load space flexibility model of the fast charging station.

9. An electronic device, characterized in that, include: At least one processor; And, a memory communicatively connected to the at least one processor; The memory stores instructions executable by the at least one processor, the instructions being configured to perform the method described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the method according to any one of claims 1-7.

Images