Method and device for predicting adjustable capacity under grouping of charging and discharging resources of electric vehicle

Abstract

The invention relates to an adjustable capacity prediction method and device under electric vehicle charging and discharging resource grouping. The method comprises the following steps: acquiring historical charging and discharging behavior data of an electric vehicle; based on the historical charging and discharging behavior data, fuzzy reasoning of user charging and discharging behaviors is carried out, and users with preset regulation and control potential are obtained; carrying out clustering identification on the discharge behaviors of the users with the preset regulation and control potential; and based on the clustering identification result, quantitatively calculating the adjustable capacity of the charging and discharging resources of the given electric vehicle. According to the method, the charging and discharging resources are clustered by considering factors such as charging and discharging resource access characteristics and behavior characteristics of the electric vehicle, a charging and discharging behavior mode switching feasibility evaluation method is established on the basis, and an adjustable capacity prediction model of the charging and discharging resources of the electric vehicle is established; and technical support is provided for electric vehicle load aggregators to participate in transaction.

Description

Technical Field

[0001] The present invention relates to the technical field of power system management and electric vehicle, and in particular to a method and device for predicting adjustable capacity under grouping of electric vehicle charging and discharging resources. Background Art

[0002] With the rapid growth of electric vehicles, the safe and stable operation of the power system has brought challenges. It is urgent to promote the friendly interaction between large-scale electric vehicles and power grids through market transactions. Among them, how to measure the adjustable capacity of electric vehicles is the basis for carrying out vehicle-grid interactive transactions. The existing evaluation methods for the adjustable capacity of electric vehicles focus on the overall level, and predict the adjustable space of charging load by studying the relationship between the total charging load and the target load. Due to the lack of consideration of the internal characteristics of electric vehicle charging and discharging resources, it is difficult for existing methods to accurately evaluate the adjustable capacity and it is difficult to meet the requirements of high-frequency and deviation assessment of electric vehicles in power market transactions. Summary of the invention

[0003] The purpose of the present invention is to propose a method and device for predicting the adjustable capacity under the grouping of electric vehicle charging and discharging resources, so as to solve the problems existing in the above-mentioned prior art. By considering factors such as the access characteristics and behavioral characteristics of electric vehicle charging and discharging resources, the charging and discharging resources are clustered and grouped. On this basis, a feasibility evaluation method considering the switching of charging and discharging behavior modes is established, and a prediction model for the adjustable capacity of electric vehicle charging and discharging resources is established, so as to provide technical support for electric vehicle load aggregators to participate in transactions.

[0004] On the one hand, to achieve the above object, the present invention provides the following scheme:

[0005] A method for predicting adjustable capacity of electric vehicle charging and discharging resource groups, comprising:

[0006] Obtain historical charging and discharging behavior data of electric vehicles;

[0007] Based on the historical charging and discharging behavior data, fuzzy reasoning of user charging and discharging behavior is performed to obtain users with preset regulation potential;

[0008] Clustering and identifying the discharge behaviors of the users with preset regulation potential;

[0009] Based on the cluster identification results, the adjustable capacity of a given electric vehicle charging and discharging resource is quantitatively calculated.

[0010] Optionally, obtaining historical charging and discharging behavior data of the electric vehicle includes:

[0011] Collecting historical charging and discharging behavior data of electric vehicles; wherein the charging and discharging behavior data includes: vehicle parking time, charging time, initial state of charge, incentive electricity price level, battery power, and maximum charging and discharging power;

[0012] The historical charge and discharge behavior data is cleaned.

[0013] Optionally, performing fuzzy reasoning on user charging and discharging behavior includes:

[0014] Based on the historical charging and discharging behavior data, fuzzy modeling is performed on the constraint factors to obtain the membership functions of the constraint factors;

[0015] Fuzzy logic operation is adopted to comprehensively calculate the calculation results of the membership function of the constraint factors, set the membership threshold, and find users who have exceeded the membership threshold, that is, users with the preset regulation potential.

[0016] Optionally, the constraint factors include: electricity price sensitivity, initial state of charge of the electric vehicle, power anxiety and compensation electricity price;

[0017] The compensation electricity price is:

[0018]

[0019]

[0020] Among them, f1, f2, and f3 are the membership functions of low electricity price compensation, medium electricity price compensation, and high electricity price compensation, respectively. r is the electricity price, A1, B1, C1, D1, and E1 are all electricity price parameters;

[0021] The initial state of charge of the electric vehicle is:

[0022]

[0023] Among them, f 4i is the membership function of the initial state of charge, i = 1, 2, 3, 4, corresponding to low charge, lower charge, higher charge and high charge respectively, S is the initial state of charge, σ is 2.5; μ i Take 0, 3.33, 6.67, and 10 respectively;

[0024] The power anxiety is:

[0025]

[0026] Among them, f 5i is the membership function of the initial state of charge, i=1, 2, 3, 4, corresponding to no anxiety, low anxiety, medium anxiety and high anxiety, respectively, and N is the state of charge anxiety;

[0027] The electricity price sensitivity is:

[0028]

[0029]

[0030] Among them, f6, f7, f8 are the membership functions of the sensitivity of electricity price, which are insensitive, relatively sensitive and highly sensitive respectively, SEN is the sensitivity of electricity price, and A2~E2 are sensitive parameters.

[0031] Optionally, the expression of the fuzzy logic operation is:

[0032]

[0033] Among them, RE is the possibility of regulation, 1 means that the user participates in grid regulation, and 0 means that the user does not participate in grid regulation. i is the behavioral decision factor, i = 1, 2, 3, 4, corresponding to the user's sensitivity to electricity price, the initial state of charge of electric vehicle charging, power anxiety and compensation electricity price, respectively.

[0034] Optionally, clustering and identifying the discharge behaviors of the users with preset regulation potential includes:

[0035] Based on a preset interval, collecting real-time charging data of the user with preset regulation potential;

[0036] Dividing the real-time charging data into k disjoint clusters by using a K-Means algorithm;

[0037] According to the division result, the power consumption mode to which the daily charge and discharge load baseline belongs is judged to obtain the adjustable capacity characteristics of electric vehicle charge and discharge; wherein, the adjustable capacity characteristics of electric vehicle charge and discharge include: charging behavior clustering characteristics and adjustable probability characteristics.

[0038] Optionally, based on the cluster identification result, quantitatively calculating the adjustable capacity of the user load for a given day includes:

[0039] Preset the flexible adjustable potential of the regulation day t period;

[0040] Based on quantitative indicators, the adjustable capacity of electric vehicle charging and discharging resources is calculated using the flexible adjustable potential of the control day t period; wherein the quantitative indicators include: load baseline and adjustable capacity confidence; the load baseline is determined based on the charging behavior clustering characteristics; the adjustable capacity confidence is determined based on the adjustable probability characteristics.

[0041] On the other hand, to achieve the above object, the present invention also provides the following solution:

[0042] A device for predicting adjustable capacity of electric vehicle charging and discharging resource groups, the device comprising: a data acquisition module, a fuzzy reasoning module, a clustering recognition module and a quantitative calculation module;

[0043] The data acquisition module is used to obtain historical charging and discharging behavior data of the electric vehicle;

[0044] The fuzzy reasoning module is used to perform fuzzy reasoning on user charging and discharging behavior based on the historical charging and discharging behavior data to obtain users with preset regulation potential;

[0045] The cluster identification module is used to perform cluster identification on the discharge behaviors of the users with preset regulation potential;

[0046] The quantitative calculation module is used to quantitatively calculate the adjustable capacity of the charging and discharging resources of a given electric vehicle based on the cluster recognition result.

[0047] Optionally, in the data acquisition module:

[0048] Collecting historical charging and discharging behavior data of electric vehicles; wherein the charging and discharging behavior data includes: vehicle parking time, charging time, initial state of charge, incentive electricity price level, battery power, and maximum charging and discharging power;

[0049] The historical charge and discharge behavior data is cleaned.

[0050] Optionally, in the fuzzy reasoning module:

[0051] Based on the historical charging and discharging behavior data, fuzzy modeling is performed on the constraint factors to obtain the membership functions of the constraint factors;

[0052] By using fuzzy logic operation and combining the calculation results of the membership function of the constraint factors, a membership threshold is set, and users who have passed the membership threshold are identified as users with preset control potential;

[0053] The constraint factors include: sensitivity to electricity prices, initial state of charge of electric vehicles, power anxiety and compensation electricity prices;

[0054] The compensation electricity price is:

[0055]

[0056] Among them, f1, f2, and f3 are the membership functions of low electricity price compensation, medium electricity price compensation, and high electricity price compensation, respectively. r is the electricity price, A1, B1, C1, D1, and E1 are all electricity price parameters;

[0057] The initial state of charge of the electric vehicle is:

[0058]

[0059] Among them, f 4i is the membership function of the initial state of charge, i = 1, 2, 3, 4, corresponding to low charge, lower charge, higher charge and high charge respectively, S is the initial state of charge, σ is 2.5; μ i Take 0, 3.33, 6.67, and 10 respectively;

[0060] The power anxiety is:

[0061]

[0062] Among them, f 5i is the membership function of the initial state of charge, i=1, 2, 3, 4, corresponding to no anxiety, low anxiety, medium anxiety and high anxiety, respectively, and N is the state of charge anxiety;

[0063] The electricity price sensitivity is:

[0064]

[0065] Among them, f6, f7, f8 are the membership functions of the sensitivity of electricity price, which are insensitive, relatively sensitive and highly sensitive respectively, SEN is the sensitivity of electricity price, and A2~E2 are sensitive parameters.

[0066] Optionally, the expression of the fuzzy logic operation is:

[0067]

[0068] Among them, RE is the possibility of regulation, 1 means that the user participates in grid regulation, and 0 means that the user does not participate in grid regulation. i is the behavioral decision factor, i = 1, 2, 3, 4, corresponding to the user's sensitivity to electricity price, the initial state of charge of electric vehicle charging, power anxiety and compensation electricity price, respectively.

[0069] Optionally, in the cluster identification module:

[0070] Based on a preset interval, collecting real-time charging data of the user with preset regulation potential;

[0071] Dividing the real-time charging data into k disjoint clusters by using a K-Means algorithm;

[0072] According to the division result, the power consumption mode to which the daily charge and discharge load baseline belongs is judged to obtain the adjustable capacity characteristics of electric vehicle charge and discharge; wherein, the adjustable capacity characteristics of electric vehicle charge and discharge include: charging behavior clustering characteristics and adjustable probability characteristics.

[0073] Optionally, in the quantitative calculation module:

[0074] Preset the flexible adjustable potential of the regulation day t period;

[0075] Based on quantitative indicators, the adjustable capacity of electric vehicle charging and discharging resources is calculated using the flexible adjustable potential of the control day t period; wherein the quantitative indicators include: load baseline and adjustable capacity confidence; the load baseline is determined based on the charging behavior clustering characteristics; the adjustable capacity confidence is determined based on the adjustable probability characteristics.

[0076] The beneficial effects of the present invention are:

[0077] The present invention first obtains the charging and discharging behavior data of electric vehicles; secondly, based on the charging and discharging behavior data, fuzzy reasoning of user charging and discharging behavior is performed to obtain users with preset regulation potential; then the discharge behavior of users with preset regulation potential is clustered and identified; finally, based on the cluster identification results, the adjustable capacity of the charging and discharging resources of a given electric vehicle is quantitatively calculated. The present invention clusters and groups the charging and discharging resources by considering factors such as the access characteristics and behavioral characteristics of the charging and discharging resources of electric vehicles, and on this basis, establishes a feasibility assessment method considering the switching of charging and discharging behavior modes, establishes an adjustable capacity prediction model for the charging and discharging resources of electric vehicles, and provides technical support for electric vehicle load aggregators to participate in transactions. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0079] Figure 1 The present invention is a flowchart of a method for predicting adjustable capacity of electric vehicles under grouping of charging and discharging resources according to an embodiment of the present invention. DETAILED DESCRIPTION

[0080] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0081] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0082] like Figure 1 As shown, this embodiment proposes a method for predicting adjustable capacity under electric vehicle charging and discharging resource grouping, including:

[0083] Obtain historical charging and discharging behavior data of electric vehicles;

[0084] Based on the historical charging and discharging behavior data, fuzzy reasoning of user charging and discharging behavior is performed to obtain users with preset regulation potential;

[0085] Cluster identification of discharge behaviors of users with preset regulation potential;

[0086] Based on the cluster identification results, the adjustable capacity of a given electric vehicle charging and discharging resource is quantitatively calculated.

[0087] Specifically, in this embodiment, the adjustable capacity prediction method includes: obtaining electric vehicle charging and discharging behavior data and performing data cleaning, including vehicle parking time, charging time, initial state of charge, incentive electricity price level, battery power, maximum charging and discharging power, etc.; conducting fuzzy reasoning on user charging and discharging behavior, simulating user behavior decisions, and realizing fuzzy reasoning prediction of user participation in power grid regulation, laying the foundation for further analysis of adjustable capacity; clustering and identifying user charging and discharging behaviors based on the KMeans algorithm; and using an electric vehicle user charging load flexible adjustable potential analysis method to quantitatively calculate the most likely adjustment amount and maximum adjustable amount of user load on a given day.

[0088] Specifically, in this embodiment, obtaining the electric vehicle charging and discharging behavior data and performing data cleaning includes:

[0089] The collected data includes vehicle stop time, charging time, initial state of charge, incentive electricity price level, battery power, maximum charging and discharging power, etc.

[0090] Use smoothing technology to detect abnormal load data. If the condition of formula (1) is met, the data is abnormal data. Abnormal data is eliminated.

[0091]

[0092] In the formula, x j,k is the user load at time k on day j, n is the total number of observation days, x- k is the average load of the user at time k, is the user load variance, ε is a constant factor, which is equal to 1.

[0093] Furthermore, the fuzzy reasoning of the user's charging and discharging behavior includes:

[0094] Based on the historical charging and discharging behavior data, the constraint factors are fuzzy modeled to obtain the membership functions of the constraint factors;

[0095] Fuzzy logic operation is adopted to calculate the membership function of the comprehensive constraint factors, and the membership threshold is set. Users who have exceeded the membership threshold are identified as users with the preset control potential.

[0096] Specifically, in this embodiment, fuzzy reasoning of user charging and discharging behavior is carried out to simulate user behavior decision-making, including:

[0097] Considering four factors, namely, the sensitivity of electric vehicle users to electricity prices, the initial state of charge of electric vehicles, power anxiety and compensation electricity prices, fuzzy modeling was performed on each factor to obtain their respective membership functions.

[0098] 1) Electricity price compensation

[0099]

[0100]

[0101] Where f1, f2, and f3 are the membership functions of low electricity price compensation, medium electricity price compensation, and high electricity price compensation, respectively, and the threshold intervals are [0, 3.33], [3.33, 6.67], and [6.67, 10]; P r is the electricity price; A1, B1, C1, D1, E1 are electricity price parameters.

[0102] 2) Electric vehicle charging initial state of charge

[0103]

[0104] In the formula, f 4i is the membership function of the initial state of charge, i = (1, 2, 3, 4), corresponding to low charge, lower charge, higher charge and high charge, respectively, and the threshold intervals are [0, 2.5], [2.5, 5], [5, 7.5], [7.5, 10], respectively; S is the initial state of charge; σ is 2.5; μ i Take 0, 3.33, 6.67, and 10 respectively.

[0105] 3) Battery anxiety

[0106]

[0107] In the formula, f 5i is the membership function of the initial state of charge, i = (1, 2, 3, 4), corresponding to no anxiety, low anxiety, medium anxiety and high anxiety, respectively, and the threshold intervals are [0, 2.5], [2.5, 5], [5, 7.5], [7.5, 10]; N is the state of charge anxiety; σ is 2.5; μ i Take 0, 3.33, 6.67, and 10 respectively.

[0108] 4) Sensitivity to electricity prices

[0109]

[0110]

[0111] In the formula: f6, f7, f8 are the membership functions of the sensitivity to electricity prices, which are insensitive, relatively sensitive and highly sensitive respectively, and the threshold intervals are [0, 3.33], [3.33, 6.67], [6.67, 10] respectively; SEN is the sensitivity to electricity prices; A2~E2 are sensitive parameters, A2=0, B2=2.5, C2=5, D2=7.5, E2=10.

[0112] On the basis of fuzzy modeling, whether the user has the possibility of regulation is set as the output result, as shown in formula (11).

[0113]

[0114] Where RE is the possibility of regulation, 1 means the user participates in grid regulation, and 0 means the user does not participate in grid regulation. i is the behavioral decision factor, i = (1, 2, 3, 4), corresponding to the user's sensitivity to electricity price, the initial state of charge of electric vehicle charging, power anxiety and compensation electricity price respectively.

[0115] Then, the output rules of the membership function are formulated to obtain a fuzzy reasoning algorithm that takes into account the factors affecting user behavior. The algorithm simulates user behavior decisions and realizes fuzzy reasoning of users participating in power grid regulation.

[0116] Furthermore, clustering and identifying the discharge behaviors of users with preset regulation potential includes:

[0117] Based on preset intervals, collect real-time charging data of users with preset regulation potential;

[0118] The real-time charging data is divided into k disjoint clusters using the K-Means algorithm;

[0119] According to the division results, the power consumption mode to which the daily charge and discharge load baseline belongs is judged to obtain the adjustable capacity characteristics of electric vehicle charge and discharge. Among them, the adjustable capacity characteristics of electric vehicle charge and discharge include: charging behavior clustering characteristics and adjustable probability characteristics.

[0120] Specifically, in this embodiment, clustering and identifying user charging and discharging behaviors based on the K-Means algorithm includes:

[0121] For users with potential for regulation, data collection is performed on the park user's electricity meter every 15 minutes, with a total of 96 collection points per day, and 96 discrete points are used to approximate the user's power consumption changes within a day. The K-Means algorithm is used to divide the historical charging load sample set of electric vehicle users with potential for regulation into k non-overlapping clusters.

[0122] According to the classification results, the power consumption mode of the charge and discharge load baseline on the control day is judged, and the cluster h to which the control day belongs is:

[0123]

[0124] In the formula, c is the number of clusters, x t is the load baseline of the tth regulation day, which is the average load of n working days before the regulation day, and n is generally taken as 5; μ i is the center of the i-th cluster; ξ it For x t The degree of membership of the center of the i-th cluster.

[0125] Furthermore, based on the cluster identification results, the adjustable capacity of the given daily user load is quantitatively calculated, including:

[0126] Preset the flexible adjustable potential of the regulation day t period;

[0127] Based on quantitative indicators, the adjustable capacity of electric vehicle charging and discharging resources is calculated using the flexible adjustable potential of the control day t period. The quantitative indicators include: load baseline and adjustable capacity confidence; the load baseline is determined based on the charging behavior clustering characteristics; the adjustable capacity confidence is determined based on the adjustable probability characteristics.

[0128] Specifically, in this embodiment, the flexible adjustable potential analysis method of electric vehicle user charging load is adopted to quantitatively calculate the most likely adjustable capacity and the maximum adjustable capacity of the user load on a given day, including:

[0129] Let the cluster to which the load regulation day belongs be C m , the cluster center is μ m , the number of load curves in the cluster is θ, t = [t start , t end ]The load curve during the regulation period is x t , define the flexible adjustable potential of the regulation day t period as ΔP im (t) can be expressed as:

[0130]

[0131] In the formula, μ m is the center of the mth cluster that is not in the same cluster as the load baseline; x tis the load baseline during the adjustment period; x ψ is the load curve with the longest Euclidean distance within the same cluster as the load baseline; ΔP im (t) is the flexible adjustable potential of user load (kW). ΔP im (t)(m=i, i+1, i+2, ..., k) represents the adjustable potential of users (ki) under different power consumption modes.

[0132] Further calculations are performed to obtain the adjustable capacity of electric vehicle charging and discharging resources, using the most likely adjustable capacity ΔP ii (t), maximum adjustable capacity ΔP ik (t) Two quantitative indicators. Among them, the most likely control capacity ΔP ii (t) is the most likely adjustable capacity value for evaluating user load, which is calculated as the difference between the baseline load during the regulation period of the regulation day and the lowest load in the cluster to which it belongs; the maximum adjustable capacity ΔP ik (t) is used to evaluate the maximum adjustable capacity that can be achieved by the user. The specific calculation method is the difference between the baseline load during the adjustment period of the adjustment day and the cluster center of the cluster with the smallest load value.

[0133] This embodiment also proposes a device for predicting adjustable capacity under grouping of electric vehicle charging and discharging resources, including: a data acquisition module, a fuzzy reasoning module, a clustering recognition module and a quantitative calculation module;

[0134] Data acquisition module, used to obtain historical charging and discharging behavior data of electric vehicles;

[0135] The fuzzy reasoning module is used to perform fuzzy reasoning on user charging and discharging behavior based on historical charging and discharging behavior data, and obtain users with preset regulation potential;

[0136] A cluster identification module, used for cluster identification of discharge behaviors of users with preset regulation potential;

[0137] The quantitative calculation module is used to quantitatively calculate the adjustable capacity of a given electric vehicle charging and discharging resource based on the clustering recognition results.

[0138] Furthermore, in the data acquisition module:

[0139] Collect historical charging and discharging behavior data of electric vehicles; the charging and discharging behavior data includes: vehicle parking time, charging time, initial state of charge, incentive electricity price level, battery power, and maximum charging and discharging power;

[0140] Perform data cleaning on historical charging and discharging behavior data.

[0141] Furthermore, in the fuzzy reasoning module:

[0142] Based on the historical charging and discharging behavior data, the constraint factors are fuzzy modeled to obtain the membership functions of the constraint factors;

[0143] Fuzzy logic operation is used to calculate the membership function of the comprehensive constraint factors, and the membership threshold is set. Users who have exceeded the membership threshold are identified as users with the preset control potential.

[0144] Among them, the constraints include: sensitivity to electricity prices, initial state of charge of electric vehicles, power anxiety and compensation electricity prices;

[0145] The compensation electricity price is:

[0146]

[0147] Among them, f1, f2, and f3 are the membership functions of low electricity price compensation, medium electricity price compensation, and high electricity price compensation, respectively. r is the electricity price, A1, B1, C1, D1, and E1 are all electricity price parameters;

[0148] The initial state of charge of an electric vehicle is:

[0149]

[0150] Among them, f 4i is the membership function of the initial state of charge, i = 1, 2, 3, 4, corresponding to low charge, lower charge, higher charge and high charge respectively, S is the initial state of charge, σ is 2.5; μ i Take 0, 3.33, 6.67, and 10 respectively;

[0151] Power anxiety is:

[0152]

[0153] Among them, f 5i is the membership function of the initial state of charge, i=1, 2, 3, 4, corresponding to no anxiety, low anxiety, medium anxiety and high anxiety, respectively, and N is the state of charge anxiety;

[0154] The sensitivity of electricity prices is:

[0155]

[0156] Among them, f6, f7, f8 are the membership functions of the sensitivity of electricity price, which are insensitive, relatively sensitive and highly sensitive respectively, SEN is the sensitivity of electricity price, and A2~E2 are sensitive parameters.

[0157] Furthermore, the expression of fuzzy logic operation is:

[0158]

[0159] Among them, RE is the possibility of regulation, 1 means that the user participates in grid regulation, and 0 means that the user does not participate in grid regulation. i is the behavioral decision factor, i = 1, 2, 3, 4, corresponding to the user's sensitivity to electricity price, the initial state of charge of electric vehicle charging, power anxiety and compensation electricity price, respectively.

[0160] Furthermore, in the cluster identification module:

[0161] Based on preset intervals, collect real-time charging data of users with preset regulation potential;

[0162] The real-time charging data is divided into k disjoint clusters using the K-Means algorithm;

[0163] According to the division results, the power consumption mode to which the daily charge and discharge load baseline belongs is judged to obtain the adjustable capacity characteristics of electric vehicle charge and discharge; wherein, the adjustable capacity characteristics of electric vehicle charge and discharge include: charging behavior clustering characteristics and adjustable probability characteristics.

[0164] Furthermore, in the quantitative calculation module:

[0165] Preset the flexible adjustable potential of the regulation day t period;

[0166] Based on quantitative indicators, the adjustable capacity of electric vehicle charging and discharging resources is calculated using the flexible adjustable potential during the t period of the control day. The quantitative indicators include: load baseline and adjustable capacity confidence. The load baseline is determined based on the clustering characteristics of charging behavior. The adjustable capacity confidence is determined based on the adjustable probability characteristics.

[0167] Specifically, this embodiment selects the real charging load data of a centralized charging station in a residential area of ​​a city for one month, and divides it into a 7-day cycle from Monday to Sunday. The selected load data uses 15 minutes as a sampling point, and there are 672 sampling points in one cycle. 224 users are screened out through fuzzy reasoning, and they are clustered to obtain 7 categories.

[0168] Table 1 Clustering results of electric vehicle charging and discharging resources

[0169]

[0170]

[0171] Furthermore, according to the analysis method of the flexible adjustable potential of electric vehicle user charging load, the most likely adjustable amount and the maximum adjustable amount of groups 1-7 in the time period from 19:00 to 23:00 on a certain operating day are calculated, and a comparative analysis is performed. The comparison results are shown in Table 2.

[0172] Table 2 Comparison of the most likely adjustment amount, maximum adjustable amount and actual adjustment load amount for different groups

[0173] Grouping Actual load adjustment value (kW) Most likely adjustment (kW) Maximum adjustable capacity (kW) Cluster 1 1656.5 1395.9 61.7 Cluster 2 472.9 258.3 173.8 Cluster 3 98.8 163.2 69.6 Cluster 4 129.4 288.1 197.1 Cluster 5 1524.7 985.1 224.3 Cluster 6 1891.8 1498.4 771.8 Group 7 90 188.1 41.6 total 5864.1 4777.1 1539.9

[0174] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for predicting the adjustable capacity of electric vehicle charging and discharging resources under grouping, characterized in that: include: Obtain historical charging and discharging behavior data of electric vehicles; Based on the historical charging and discharging behavior data, fuzzy reasoning of user charging and discharging behavior is performed to obtain users with preset regulation potential; Clustering and identifying the discharge behaviors of the users with preset regulation potential; Based on the cluster identification results, the adjustable capacity of a given electric vehicle charging and discharging resource is quantitatively calculated.

2. The method for predicting adjustable capacity of electric vehicle charging and discharging resource groups according to claim 1 is characterized in that: Obtaining historical charging and discharging behavior data of electric vehicles includes: Collecting historical charging and discharging behavior data of electric vehicles; wherein the charging and discharging behavior data includes: vehicle parking time, charging time, initial state of charge, incentive electricity price level, battery power, and maximum charging and discharging power; The historical charge and discharge behavior data is cleaned.

3. The method for predicting adjustable capacity of electric vehicle charging and discharging resource groups according to claim 1 is characterized in that: Fuzzy reasoning of user charging and discharging behavior includes: Based on the historical charging and discharging behavior data, fuzzy modeling is performed on the constraint factors to obtain the membership functions of the constraint factors; Fuzzy logic operation is adopted to comprehensively calculate the calculation results of the membership function of the constraint factors, set the membership threshold, and find users who have exceeded the membership threshold, that is, users with the preset regulation potential.

4. The method for predicting adjustable capacity of electric vehicle charging and discharging resource groups according to claim 3 is characterized in that: The constraint factors include: electricity price sensitivity, initial state of charge of electric vehicles, power anxiety and compensation electricity price; The compensation electricity price is: Among them, f1, f2, and f3 are the membership functions of low electricity price compensation, medium electricity price compensation, and high electricity price compensation, respectively. r is the electricity price, A1, B1, C1, D1, and E1 are all electricity price parameters; The initial state of charge of the electric vehicle is: Among them, f 4i is the membership function of the initial state of charge, i = 1, 2, 3, 4, corresponding to low charge, lower charge, higher charge and high charge respectively, S is the initial state of charge, σ is 2.5; μ i Take 0, 3.33, 6.67, and 10 respectively; The power anxiety is: Among them, f 5i is the membership function of the initial state of charge, i=1, 2, 3, 4, corresponding to no anxiety, low anxiety, medium anxiety and high anxiety, respectively, and N is the state of charge anxiety; The electricity price sensitivity is: Among them, f6, f7, f8 are the membership functions of the sensitivity of electricity price, which are insensitive, relatively sensitive and highly sensitive respectively, SEN is the sensitivity of electricity price, and A2~E2 are sensitive parameters.

5. The method for predicting adjustable capacity of electric vehicle charging and discharging resource groups according to claim 3 is characterized in that: The expression of the fuzzy logic operation is: Among them, RE is the possibility of regulation, 1 means that the user participates in the grid regulation, 0 means that the user does not participate in the grid regulation, f i is the behavioral decision factor, i = 1, 2, 3, 4, corresponding to the user's sensitivity to electricity price, the initial state of charge of electric vehicle charging, power anxiety and compensation electricity price, respectively.

6. The method for predicting adjustable capacity of electric vehicle charging and discharging resource groups according to claim 1 is characterized in that: Clustering and identifying the discharge behavior of the user with preset regulation potential includes: Based on a preset interval, collecting real-time charging data of the user with preset regulation potential; Dividing the real-time charging data into k disjoint clusters by using a K-Means algorithm; According to the division result, the power consumption mode to which the daily charge and discharge load baseline belongs is judged to obtain the adjustable capacity characteristics of electric vehicle charge and discharge; wherein, the adjustable capacity characteristics of electric vehicle charge and discharge include: charging behavior clustering characteristics and adjustable probability characteristics.

7. The method for predicting adjustable capacity of electric vehicle charging and discharging resource groups according to claim 6 is characterized in that: Based on the cluster identification results, the adjustable capacity of the given daily user load is quantitatively calculated, including: Preset the flexible adjustable potential of the regulation day t period; Based on quantitative indicators, the adjustable capacity of electric vehicle charging and discharging resources is calculated using the flexible adjustable potential of the control day t period; wherein the quantitative indicators include: load baseline and adjustable capacity confidence; the load baseline is determined based on the charging behavior clustering characteristics; the adjustable capacity confidence is determined based on the adjustable probability characteristics.

8. A device for predicting adjustable capacity of electric vehicle charging and discharging resource groups, characterized in that: Used to implement the method for predicting the adjustable capacity of electric vehicle charging and discharging resource groups as described in any one of claims 1 to 7, the device comprises: a data acquisition module, a fuzzy reasoning module, a clustering recognition module and a quantitative calculation module; The data acquisition module is used to obtain historical charging and discharging behavior data of the electric vehicle; The fuzzy reasoning module is used to perform fuzzy reasoning on user charging and discharging behavior based on the historical charging and discharging behavior data to obtain users with preset regulation potential; The cluster identification module is used to perform cluster identification on the discharge behaviors of the users with preset regulation potential; The quantitative calculation module is used to quantitatively calculate the adjustable capacity of the charging and discharging resources of a given electric vehicle based on the cluster recognition result.

9. The device for predicting adjustable capacity of electric vehicle charging and discharging resource groups according to claim 8, characterized in that: In the cluster identification module: Based on a preset interval, collecting real-time charging data of the user with preset regulation potential; Dividing the real-time charging data into k disjoint clusters by using a K-Means algorithm; According to the division result, the power consumption mode to which the daily charge and discharge load baseline belongs is judged to obtain the adjustable capacity characteristics of electric vehicle charge and discharge; wherein, the adjustable capacity characteristics of electric vehicle charge and discharge include: charging behavior clustering characteristics and adjustable probability characteristics.

10. The device for predicting adjustable capacity of electric vehicle charging and discharging resource groups according to claim 9, characterized in that: In the quantitative calculation module: Preset the flexible adjustable potential of the regulation day t period; Based on quantitative indicators, the adjustable capacity of electric vehicle charging and discharging resources is calculated using the flexible adjustable potential of the control day t period; wherein the quantitative indicators include: load baseline and adjustable capacity confidence; the load baseline is determined based on the charging behavior clustering characteristics; the adjustable capacity confidence is determined based on the adjustable probability characteristics.

Images