Electric vehicle load aggregation method and system based on Markov chain

Abstract

The invention provides an electric vehicle load aggregation method and system based on a Markov chain. The method comprises the following steps: acquiring historical operation data of charging piles in a power distribution area; constructing a charging pile operation state distribution probability matrix based on a Markov chain and the operation data; based on the charging pile operation state distribution probability matrix, preliminarily predicting the up-regulation power and the down-regulation power of a station area charging pile load cluster, and determining charging piles capable of participating in demand response; and calculating real-time up-regulation power and down-regulation power of the charging piles participating in the demand response based on the operation data of the charging piles participating in the demand response. According to the method, the charging pile operation state distribution probability matrix is constructed based on the Markov chain, the state space of the charging pile is predicted, and the adjustable power prediction capability of the electric vehicle load cluster in the V2G mode is improved.

Description

Technical Field

[0001] The present invention relates to the field of power load analysis and management, and in particular to a method and system for electric vehicle load aggregation based on a Markov chain. Background Art

[0002] In recent years, with the rapid increase in the proportion of renewable energy grid-connected power generation, the power system regulation mode has changed from the "source follows load" mode of traditional centralized thermal power generation to the "source-load interaction" mode of high-proportion wind and solar power generation. By aggregating the flexible loads in the new power system, group regulation and control of the aggregated loads can be achieved, thereby adapting to the development trend of increasing uncertainty in power fluctuations on the power generation side. Load participation in power balance regulation is an important research direction in the current power system field. Electric vehicles are a rapidly developing type of flexible load in the new power system. Constructing an electric vehicle charging pile load aggregation model within the distribution station area and predicting the regulation potential of electric vehicle loads in the distribution station area are important decision-making bases for optimizing their regulation. Through "source-load interaction", the renewable energy absorption capacity is improved, the power grid loss in the distribution station area is reduced, and the green and low-carbon transformation and development of the power system is promoted.

[0003] Existing research on electric vehicle load aggregation methods still has shortcomings. There are two main types of electric vehicle load aggregation methods: one is to directly model the electric vehicle load cluster naturally distributed in a certain area as a whole; the other is to aggregate modeling based on the individual behavior of electric vehicle users. The first type of method usually uses clustering algorithms such as k-means, fuzzy C means, and self-organizing neural mapping networks to automatically divide electric vehicles into several clusters with the same driving characteristics, and uses the Monte Carlo method to assume that the starting charging time and charging time of electric vehicles in the same cluster all obey a certain probability distribution, thereby calculating the charging power of the charging station at different times. The models and parameters of this type of method are usually based on assumptions and experience, ignoring individual differences and having weak generalization ability. The second type of method usually constructs an adjustable potential model of a single electric vehicle based on parameters such as the state of charge and remaining charging time of the single electric vehicle, and obtains the aggregated adjustable power by summing the adjustable powers of all electric vehicles in the cluster. When the distribution capacity of the substation is large and the number of connected electric vehicles is large, the second type of method often has problems with data collection and incomplete data when constructing an adjustable model of a single electric vehicle. Therefore, this type of method is only suitable for the construction of electric vehicle load cluster aggregation models in smaller local distribution networks or microgrid systems. Summary of the invention

[0004] In order to solve the problems that the existing technology directly models the electric vehicle load cluster naturally distributed in a certain area as a whole, the model is based on assumptions and experience, ignores individual differences, and has weak generalization ability, and the aggregation modeling based on the individual behavior of electric vehicle users often has the problems of difficulty in data collection and incomplete data, the present invention proposes an electric vehicle load aggregation method based on Markov chain, including:

[0005] Obtain historical operation data of charging piles in distribution station areas;

[0006] Construct the probability matrix of charging pile operation status distribution based on Markov chain and historical operation data;

[0007] Preliminarily predict the adjustable power and adjustable power of the electric vehicle charging pile load cluster in the substation based on the charging pile operation status distribution probability matrix, and determine the charging piles that can participate in demand response;

[0008] Based on the historical operation data of the charging piles that can participate in the demand response, the real-time adjustable power and adjustable power of the charging piles that can participate in the demand response are calculated.

[0009] Optionally, constructing a charging pile operation state distribution probability matrix based on the Markov chain and historical operation data includes:

[0010] Setting the operating status of the electric vehicle charging pile in the station area, the operating status includes connection status and charging status;

[0011] Based on the historical operation data of charging piles in the distribution station area, the state distribution probability of charging piles in different time periods is analyzed to obtain the probability distribution of each state of the charging piles over time periods;

[0012] The operating state of the charging pile is represented by a Markov chain to obtain the state distribution probability of the charging pile in a certain period of time;

[0013] Constructing a charging pile operation state transfer matrix equation based on the state distribution probability of the charging pile in a certain period of time;

[0014] Performing time series recursion on the charging pile operation state transfer matrix equation to obtain a charging pile operation state distribution probability matrix;

[0015] The operating status of the electric vehicle charging pile in the station area includes: no electric vehicle connected, electric vehicle connected and the charge state value ∈ Connected electric vehicle with state of charge ∈ Connected electric vehicle with state of charge ∈ and They respectively represent the minimum and maximum state of charge of the battery of electric vehicles participating in demand response.

[0016] Optionally, the preliminarily predicting the adjustable power and adjustable power of the electric vehicle charging pile load cluster in the substation based on the charging pile operation status distribution probability matrix, and determining the charging piles that can participate in the demand response, includes: considering seasonal and holiday factors, and using the fuzzy C-means clustering algorithm to generate a typical energy usage scenario of the charging piles according to the set parameters based on the daily load curve of the substation throughout the year;

[0017] Based on the typical energy consumption scenario of the charging piles and the probability matrix of the charging pile operation status distribution, the adjustable power and adjustable power of the electric vehicle charging pile load cluster in the substation are predicted;

[0018] Identify the electric vehicle charging piles in the area with adjustable power and adjustable power as charging piles that can participate in demand response;

[0019] The setting parameters include: load peak / valley time, load value, and number of peaks and troughs.

[0020] Optionally, considering seasonal and holiday factors, the daily load curve of the area throughout the year is used according to the set parameters using the fuzzy C-means clustering algorithm to generate a typical energy usage scenario of the charging pile, including:

[0021] Step 1: Initialize the membership matrix U;

[0022] Step 2: Calculate the initial cluster center according to the numerical range and distribution of the physical parameters of the electric vehicle;

[0023] Step 3: Update the membership matrix U and cluster center V by calculating the distance between each data point and each cluster center and redistributing the membership of each data point according to the distance;

[0024] Step 4: Determine whether the iteration termination condition is met. If so, proceed to step 5, otherwise return to step 3;

[0025] Step 5: Output clustering results.

[0026] Optionally, the adjustable power and adjustable power of the electric vehicle charging pile load cluster in the substation are calculated as follows:

[0027]

[0028] In the formula, The power can be adjusted downward or the expected power can be adjusted upward for the prediction of the electric vehicle charging pile cluster; The predicted downward adjustable power and upward adjustable power expected value of the electric vehicle charging pile in the kth state; The predicted adjustable power of the electric vehicle charging pile in the second state and the third state respectively; are the predicted adjustable power of the electric vehicle charging piles in the second and fourth states respectively; f(x) is the probability density function of the state of charge of the battery on the electric vehicle; N0 is the number of charging piles in the area; The maximum and minimum state of charge of the battery on board the electric vehicle participating in the demand response; is the state distribution probability of the i-th charging pile being in the second state in time period t; is the discharge power of the electric vehicle connected to the i-th charging pile; is the charging power of the electric vehicle connected to the i-th charging pile; The minimum battery state of charge that must be met when an electric vehicle connected to a charging pile in the second state participates in valley filling demand response; is the minimum battery state of charge that must be met when an electric vehicle connected to a charging pile in the second state participates in peak demand response; f(x) is the probability density function of the battery state of charge of the electric vehicle, and x is a random variable with a value in the interval [0,1], representing the battery state of charge of a certain electric vehicle.

[0029] Optionally, the real-time adjustable power and adjustable power of the charging pile participating in the demand response are calculated as follows:

[0030]

[0031] Where: They are respectively the adjustable power that can be reduced and the adjustable power that can be increased of the secondary prediction of the electric vehicle charging pile load cluster; They are respectively the secondary predicted power that can be adjusted downward and the power that can be adjusted upward for the electric vehicle charging pile in the kth state; is the total number of charging piles that are in the second state in the period before the start of demand response and can participate in a peak-shaving demand response. is the total number of charging piles in the second state that can participate in a valley filling demand response in the period before the demand response starts. is the total number of charging piles that are in the third state in the period before the start of demand response and can participate in a peak-shaving demand response. is the total number of charging piles in the fourth state before the start of demand response that can participate in a valley filling demand response, i is the serial number of the charging pile, are the discharge power and charging power of the electric vehicle connected to the i-th charging pile respectively.

[0032] On the other hand, the present invention also provides an electric vehicle load aggregation system based on a Markov chain, comprising:

[0033] A data acquisition module is used to obtain the historical operation data of charging piles in the distribution station area;

[0034] Probability matrix construction module, used to construct the probability matrix of charging pile operation status distribution based on Markov chain and historical operation data;

[0035] A preliminary judgment module, used to preliminarily predict the adjustable power and adjustable power of the electric vehicle charging pile load cluster in the substation based on the charging pile operation status distribution probability matrix, and determine the charging piles that can participate in the demand response;

[0036] The precise calculation module is used to calculate the real-time adjustable power and adjustable power of the charging pile participating in the demand response based on the historical operation data of the charging pile participating in the demand response.

[0037] Optionally, the probability matrix building module is specifically used for:

[0038] Setting the operating status of the electric vehicle charging pile in the station area, the operating status includes connection status and charging status;

[0039] Based on the historical operation data of charging piles in the distribution station area, the state distribution probability of charging piles in different time periods is analyzed to obtain the probability distribution of each state of the charging piles over time periods;

[0040] The operating state of the charging pile is represented by a Markov chain to obtain the state distribution probability of the charging pile in a certain period of time;

[0041] Constructing a charging pile operation state transfer matrix equation based on the state distribution probability of the charging pile in a certain period of time;

[0042] Performing time series recursion on the charging pile operation state transfer matrix equation to obtain a charging pile operation state distribution probability matrix;

[0043] The operating status of the electric vehicle charging pile in the station area includes: no electric vehicle connected, electric vehicle connected and the charge state value ∈ Connected electric vehicle with state of charge ∈ Connected electric vehicle with state of charge ∈ and They respectively represent the minimum and maximum state of charge of the battery of electric vehicles participating in demand response.

[0044] Optionally, the preliminary determination module includes:

[0045] The scenario determination submodule is used to consider seasonal and holiday factors, and use the fuzzy C-means clustering algorithm to generate typical energy usage scenarios for charging piles based on the daily load curve of the substation throughout the year according to the set parameters;

[0046] The adjustable power determination submodule is used to predict the adjustable power and adjustable power of the electric vehicle charging pile load cluster in the substation based on the typical energy consumption scenario of the charging pile and the probability matrix of the charging pile operation status distribution;

[0047] A judgment submodule, used to determine the electric vehicle charging piles in the area with adjustable power and adjustable power as charging piles that can participate in demand response;

[0048] The setting parameters include: load peak / valley time, load value, and number of peaks and troughs.

[0049] Optionally, the scene determination submodule is specifically used to:

[0050] Step 1: Initialize the membership matrix U;

[0051] Step 2: Calculate the initial cluster center according to the numerical range and distribution of the physical parameters of the electric vehicle;

[0052] Step 3: Update the membership matrix U and cluster center V by calculating the distance between each data point and each cluster center and redistributing the membership of each data point according to the distance;

[0053] Step 4: Determine whether the iteration termination condition is met. If so, proceed to step 5, otherwise return to step 3;

[0054] Step 5: Output clustering results.

[0055] Optionally, the adjustable power determination submodule calculates the adjustable power and the adjustable power of the electric vehicle charging pile load cluster in the substation area according to the following formula:

[0056]

[0057] In the formula, The power can be adjusted downward or the expected power can be adjusted upward for the prediction of the electric vehicle charging pile cluster; The predicted downward adjustable power and upward adjustable power expected value of the electric vehicle charging pile in the kth state; The predicted adjustable power of the electric vehicle charging pile in the second state and the third state respectively; are the predicted adjustable power of the electric vehicle charging piles in the second and fourth states respectively; f(x) is the probability density function of the state of charge of the battery on the electric vehicle; N0 is the number of charging piles in the area; The maximum and minimum state of charge of the battery on board the electric vehicle participating in the demand response; is the state distribution probability of the i-th charging pile being in the second state in time period t; is the discharge power of the electric vehicle connected to the i-th charging pile; is the charging power of the electric vehicle connected to the i-th charging pile; The minimum battery state of charge that must be met when an electric vehicle connected to a charging pile in the second state participates in valley filling demand response; is the minimum battery state of charge that must be met when an electric vehicle connected to a charging pile in the second state participates in peak demand response; f(x) is the probability density function of the battery state of charge of the electric vehicle, and x is a random variable with a value in the interval [0,1], representing the battery state of charge of a certain electric vehicle.

[0058] Optionally, the precise calculation module calculates the real-time adjustable power and adjustable power of the charging pile participating in the demand response according to the following formula:

[0059]

[0060] Where: They are respectively the adjustable power that can be reduced and the adjustable power that can be increased of the secondary prediction of the electric vehicle charging pile load cluster; They are respectively the secondary predicted power that can be adjusted downward and the power that can be adjusted upward for the electric vehicle charging pile in the kth state; is the total number of charging piles that are in the second state in the period before the start of demand response and can participate in a peak-shaving demand response. is the total number of charging piles in the second state that can participate in a valley filling demand response in the period before the demand response starts. is the total number of charging piles that are in the third state in the period before the start of demand response and can participate in a peak-shaving demand response. is the total number of charging piles in the fourth state before the start of demand response that can participate in a valley filling demand response, i is the serial number of the charging pile, are the discharge power and charging power of the electric vehicle connected to the i-th charging pile respectively.

[0061] In another aspect, the present application further provides an electronic device, comprising: at least one processor and a memory; the memory and the processor are connected via a bus;

[0062] The memory is used to store one or more programs;

[0063] When the one or more programs are executed by the at least one processor, the electric vehicle load aggregation method based on Markov chain as described above is implemented.

[0064] On the other hand, the present application also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed, the electric vehicle load aggregation method based on Markov chain as described above is implemented.

[0065] Compared with the prior art, the present invention has the following beneficial effects:

[0066] The present invention provides a method and system for aggregating electric vehicle loads based on a Markov chain, including: obtaining historical operating data of charging piles in a distribution station area; constructing a probability matrix for the distribution of operating states of charging piles based on the Markov chain and the historical operating data; preliminarily predicting the adjustable power and adjustable power of the charging pile load cluster of electric vehicles in the distribution station area based on the probability matrix for the distribution of operating states of charging piles, and determining the charging piles that can participate in the demand response; and calculating the real-time adjustable power and adjustable power of the charging piles that can participate in the demand response based on the historical operating data of the charging piles that can participate in the demand response. The present invention constructs a probability matrix for the distribution of operating states of charging piles during demand response based on the historical operating data of the charging piles in the distribution station area and the Markov chain, considers the differences between the individual charging piles in the area, determines the charging piles that can participate in the demand response, realizes the prediction of the state space of the charging piles, and improves the accuracy of the prediction of the adjustable power of the electric vehicle load cluster under the V2G mode. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 This is a flow chart of a method for aggregating electric vehicle loads based on a Markov chain according to the present invention;

[0068] FIG2( a ) is a schematic diagram of a controllable charging area of ​​a single electric vehicle in a V2G mode of the present invention;

[0069] FIG2( b ) is a schematic diagram of a controllable discharge area of ​​a single electric vehicle in a V2G mode of the present invention;

[0070] Figure 3 It is a state transition relationship diagram of the electric vehicle charging pile of the present invention;

[0071] Figure 4 This is a typical scene clustering flow chart of an electric vehicle charging station based on the fuzzy C-means clustering method of the present invention;

[0072] FIG5(a) is a diagram showing the state of charge range of the on-board battery of an electric vehicle connected to a charging pile in the k=2 state when the demand response period is relatively long;

[0073] FIG5( b ) is a diagram showing the state of charge range of the on-board battery of an electric vehicle connected to a charging pile in the k=2 state when the demand response period is relatively short;

[0074] Figure 6 The figure is a schematic diagram of the structure of a computer device of the present invention. DETAILED DESCRIPTION

[0075] The present invention proposes a method for electric vehicle load aggregation based on Markov chain. The method is based on the single electric vehicle charging and discharging model in V2G mode, and the electric vehicle and its charging pile are equivalent to a virtual energy storage device. The Markov chain is used to establish an electric vehicle charging pile load aggregation model in the substation area, and the adjustable power of the charging pile cluster is accurately predicted through the two-stage regulation potential prediction model of the electric vehicle charging pile cluster in the substation area. V2G here refers to the coordinated interaction between electric vehicles and power grids.

[0076] In order to better understand the present invention, the content of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0077] Embodiment 1:

[0078] A Markov chain-based electric vehicle load aggregation method, such as Figure 1 As shown, including:

[0079] Step I: Obtain historical operation data of charging piles in the distribution station area;

[0080] Step II: Construct the charging pile operation status distribution probability matrix based on the Markov chain and historical operation data;

[0081] Step III: Preliminarily predict the adjustable power and adjustable power of the electric vehicle charging pile load cluster in the substation based on the charging pile operation status distribution probability matrix, and determine the charging piles that can participate in demand response;

[0082] Step IV: Based on the historical operation data of the charging piles that can participate in the demand response, the real-time adjustable power and adjustable power of the charging piles that can participate in the demand response are calculated.

[0083] This invention studies the maximum adjustable power of electric vehicle loads within a certain distribution station area. In view of the fact that the existing load aggregation model cannot effectively support the prediction of load cluster regulation potential and its group regulation and control, a Markov chain-based electric vehicle load aggregation model is proposed to improve the accuracy of the prediction of the adjustable power of electric vehicle load clusters in the V2G mode, and provide a decision-making basis for formulating more reasonable load dispatching strategies. By optimizing the power configuration of the substation area, electric vehicles are helped to participate in the grid demand response, the renewable energy absorption capacity is improved, the distribution loss of the substation area power grid is reduced, and the green and low-carbon transformation and development of the power system are promoted.

[0084] The present invention proposes a method for aggregating electric vehicle loads based on a Markov chain, the method comprising the following contents:

[0085] Step I: Obtain the historical operation data of the charging piles in the distribution area, including:

[0086] Obtain the charging and discharging time of a single electric vehicle in V2G mode, the rated charging and discharging power of a single electric vehicle, the state of charge of the electric vehicle's onboard battery at the start and end of charging, and the rated capacity of the electric vehicle's onboard battery.

[0087] Before step II, also include:

[0088] S1: Construct a charging and discharging model of electric vehicles in a certain area under V2G mode;

[0089] The S1 includes the following steps.

[0090] S101: Electric vehicle charging and discharging model under V2G mode. The energy exchange between the power grid and electric vehicle load under V2G mode is bidirectional, that is, during peak power consumption periods, the peak load power consumption pressure can be alleviated by discharging the on-board battery of the electric vehicle, and during low power consumption periods, the surplus new energy electric energy can be consumed by charging the on-board battery of the electric vehicle, so as to realize peak load shaving and valley filling of the distribution network load. The electric vehicle charging and discharging model under V2G mode is:

[0091]

[0092] Where: They are the charging and discharging time of a single electric vehicle in V2G mode, in h; They are the rated charging and discharging power of a single electric vehicle, in kW; E is the state of charge of the battery in the electric vehicle at the start and end of charging. N Rated capacity of the battery on board an electric vehicle, in kWh.

[0093] S102: Charging and discharging characteristics of the onboard battery of electric vehicles. In the process of optimizing and controlling the load cluster of electric vehicle charging piles in a certain area, in order to prevent the influence of excessive charging and discharging on the battery life, the charging and discharging feasible domain of the onboard battery of a single electric vehicle in the V2G mode is established, as shown in Figure 2(a) and Figure 2(b), where θ is the angle in the figure during the drawing process, and its physical meaning is that tanθ is the rate at which the state of charge value of the electric vehicle battery changes with time. Based on the actual onboard battery charging and discharging curve, the present invention converts the charging and discharging S SOC The -t curve is approximated as a piecewise line segment with one inflection point.

[0094] Step II: Construct the charging pile operation status distribution probability matrix based on the Markov chain and historical operation data, including:

[0095] Setting the operating status of the electric vehicle charging pile in the station area, the operating status includes connection status and charging status;

[0096] Based on the historical operation data of charging piles in the distribution station area, the state distribution probability of charging piles in different time periods is analyzed to obtain the probability distribution of each state of the charging piles over time periods;

[0097] The operating state of the charging pile is represented by a Markov chain to obtain the state distribution probability of the charging pile in a certain period of time;

[0098] Constructing a charging pile operation state transfer matrix equation based on the state distribution probability of the charging pile in a certain period of time;

[0099] Performing time series recursion on the charging pile operation state transfer matrix equation to obtain a charging pile operation state distribution probability matrix;

[0100] The operating status of the electric vehicle charging pile in the station area includes: no electric vehicle connected, electric vehicle connected and the charge state value ∈ Connected electric vehicle with state of charge ∈ Connected electric vehicle with state of charge ∈ and They respectively represent the minimum and maximum state of charge of the battery of electric vehicles participating in demand response.

[0101] Step II specifically includes:

[0102] S2: Constructing an aggregation model of electric vehicle charging piles in the substation based on Markov chain;

[0103] The S2 includes the following steps.

[0104] S201: Construct a charging pile state transfer matrix based on the Markov chain. Set the electric vehicle charging piles in the area to have four operating states: no electric vehicle connected, electric vehicle connected and the state of charge value ∈ Connected electric vehicle with state of charge ∈ Connected electric vehicle with state of charge ∈ The state variable k is set accordingly to represent the above four states, and k takes an integer value in the interval [1,4]. ij Represents the transition probability between states, such as a 11 represents the probability that a charging pile is not connected to an electric vehicle in the current period and will still be not connected to an electric vehicle in the next period, a 12 Indicates that the current period is in the state of no electric vehicle connected and the next period is transferred to the corresponding state of k = 2 - the electric vehicle is connected and the vehicle battery state of charge value ∈ , and so on. Figure 3 This is a state transfer relationship diagram of an electric vehicle charging pile. From the charging pile operation state transfer relationship in the figure, we can see that:

[0105]

[0106] Where: P k (1) is the distribution probability that the charging pile is in the kth state in the next period; k is the state variable, indicating the operating state of the electric vehicle; P k (0) is the distribution probability of the charging pile being in the kth state in the current period; a ij is the state transition probability of the charging pile transferring from the current period i state to the next period j state.

[0107] Based on the big data of vehicle-pile operation monitoring, the state distribution probability of charging piles in different time periods is analyzed to obtain the probability distribution of each state of the charging pile at any time period, and the probability distribution curves of the four operating states are drawn accordingly.

[0108] Formula (2) can be expressed in matrix form:

[0109]

[0110] S202: Solve the state transfer matrix equation. The charging pile state transfer matrix equation of formula (3) is recursively deduced in time series, that is,

[0111] P k (t) = A T ·P k (t-1) (4)

[0112] Where: P k (t) is the probability matrix of the operating state of the charging pile at time t; A T is the state transfer matrix of the charging pile from the operating state at time t-1 to the operating state at time t; P k (t-1) is the probability matrix of the operating status distribution of the charging pile at time t-1.

[0113] Based on the big data of charging pile operation status monitoring in adjacent time periods, the state transition probability distribution of the charging pile in two adjacent time periods is analyzed, and 16 state transition probability distribution curves corresponding to the 16 state transitions are drawn over time periods.

[0114] According to the limit theory in calculus, when the period is small enough, it can be considered that P k(t) =P k(t-1) , that is,

[0115] P k(t) =A T ·P k(t) (5)

[0116] Due to the existence of:

[0117]

[0118] So for the determinant |A T -E|, there exists an equation:

[0119]

[0120] Because the determinant |A T -E| is equal to 0, so P k(t) Yes A T The corresponding eigenvector when the eigenvalue is 1. Therefore, by representing the operating state of the charging pile with a Markov chain, the state distribution probability P of the charging pile in a certain period of time can be obtained. k(t) .

[0121] Considering the charging time, the operation time of connecting and disconnecting the vehicle and the charging pile, etc., when performing the iterative calculation of the state space of formula (4), it is recommended to select the time range of the iteration step to be 5 to 30 minutes, which is sufficient to support the accurate prediction of the distribution probability of the charging pile state. When the time period is small enough, the operating state of the charging pile in the current time period can be used to approximate the operating state of the next time period, and it can be used to more accurately predict the adjustable power of the charging pile after aggregation.

[0122] S3: Construct a model to calculate the load regulation potential of electric vehicle charging pile clusters in the substation.

[0123] The S3 includes the following steps.

[0124] S301: Generate typical energy usage scenarios for electric vehicle charging stations. Considering seasonal and holiday differences, typical days are selected and the fuzzy C-means clustering algorithm is used to generate typical energy usage scenarios for charging stations. The fuzzy C-means clustering algorithm obtains the optimal clustering result of the typical scenario set of charging stations through iterative optimization of cluster centers and membership matrices. The flowchart is as follows: Figure 4 shown.

[0125] S302: Preliminarily predict the regulation potential of the charging pile cluster in the substation area to provide a decision basis for whether to participate in demand response.

[0126] From the demand side, the power that can be flexibly increased during the regulation period is its adjustable power, and the power that can be flexibly reduced during the regulation period is its adjustable power. For the electric vehicle charging pile load cluster in the substation area, the electric vehicles connected to it can discharge to the grid, which means it has the ability to adjust downward (peak shaving capability), and the electric vehicles connected to it can charge, which means it has the ability to adjust upward (valley filling capability). It is generally believed that the charging pile in the state of not connecting to the electric vehicle has no adjustment capability, and the charging pile in the state of connecting the electric vehicle and the charge state value ∈ The charging pile has the ability to adjust downward and upward, and is connected to an electric vehicle and has a charge state ∈ The charging pile has only the ability to adjust downward, and is connected to an electric vehicle and the charge state ∈ The charging piles only have upward adjustment capabilities.

[0127] Since the loads participating in the invitation are required to meet the power stability requirement during the demand response period, it is not difficult to know that all charging piles in the k=3 state can participate in the peak shaving demand response, and all charging piles in the k=4 state can participate in the valley filling demand response, but not all charging piles in the k=2 state can participate in a certain "peak shaving" or "valley filling" demand response.

[0128] The following is a further introduction to the electric vehicle connected to the charging pile in the k=2 state in combination with Figure 5(a) and Figure 5(b), where f(x) is the probability density function of the battery state of charge of the electric vehicle, and x is a random variable with a value in the interval [0,1], representing the battery state of charge of a certain electric vehicle. The state of charge of its on-board battery must also be within the specified range in order to provide stable adjustable power during the specified invitation duration. The allowable range of the on-board battery state of charge is related to the invitation duration and can be expressed as:

[0129]

[0130] Where: It is the maximum and minimum rated value of the state of charge of the battery on board the electric vehicle;

[0131] The battery state of charge range that needs to be met when an electric vehicle connected to a charging pile in the second state participates in valley filling and peak shaving demand response;

[0132] The maximum battery state of charge required for an electric vehicle connected to a charging pile in the second state to participate in valley filling demand response;

[0133] The minimum battery state of charge that must be met when an electric vehicle connected to a charging pile in the second state participates in peak demand response;

[0134] is the maximum charging power of electric vehicle batteries, It is the maximum discharge power of electric vehicle battery, in kW.

[0135] t 邀约 Invitation duration, unit: h.

[0136] The expected adjustable power of the charging pile cluster participating in demand response is calculated based on the distribution probability of the operating status of the charging pile at the start time of the invitation. The specific calculation formula is:

[0137]

[0138] In the formula, It is the predicted adjustable power and expected adjustable power of the electric vehicle charging pile cluster, in kW;

[0139] is the expected value of the predicted adjustable power and adjustable power of the electric vehicle charging pile in the kth state, in kW;

[0140] N0 is the number of charging piles in the area;

[0141] f(x) is the probability density function of the state of charge of the battery on board the electric vehicle;

[0142] is the discharge and charging power of the electric vehicle battery connected to the i-th charging pile, in kW;

[0143] is the state distribution probability that the i-th charging pile is in the k-th state in time period t.

[0144] Step IV: Based on the historical operation data of the charging piles that can participate in the demand response, the real-time adjustable power and adjustable power of the charging piles that can participate in the demand response are calculated, including:

[0145] S303: Accurate prediction of the regulation potential of the cluster of electric vehicle charging piles in the substation.

[0146] As can be seen from S202, during the iterative calculation of the Markov chain state space, when the iteration step is small enough (5 to 30 minutes), the operating state of the charging pile in the previous period can be used to approximate the operating state of the current period, and used to more accurately predict the adjustable power after the charging piles are aggregated. The secondary accurate prediction of the regulation potential of the electric vehicle charging pile cluster in the substation area needs to be supported by the measured vehicle and pile data in the period before the start of demand response. The specific calculation formula is:

[0147]

[0148] Where: It is the secondary predicted adjustable power and adjustable power of the electric vehicle charging pile cluster, in kW;

[0149] The secondary predicted power reduction and power increase of the electric vehicle charging pile in the kth state, in kW;

[0150] The total number of charging piles in the second state that can participate in a peak-shaving demand response in the period before the demand response starts;

[0151] The total number of charging piles in the second state that can participate in a valley filling demand response in the period before the demand response starts;

[0152] The total number of charging piles in the third state that can participate in a peak-shaving demand response in the period before the demand response starts;

[0153] It is the total number of charging piles in the fourth state that can participate in a valley filling demand response in the period before the demand response starts.

[0154] The Markov chain-based electric vehicle load aggregation model proposed in this paper divides the operating status of the electric vehicle charging piles in the substation into two types: no electric vehicle connected, electric vehicle connected and the charge state value ∈ Connected electric vehicle with state of charge ∈ Connected electric vehicle with state of charge ∈ Four categories are identified, and matrix equations are constructed through the state transfer matrix and the state distribution matrix to solve the state distribution probability of the cluster of electric vehicle charging piles in the substation. Compared with the traditional load aggregation method that only considers the needs of electric vehicle owners, the operability of electric vehicle load data information collection is enhanced, thereby improving the soundness of information related to load aggregation, and improving the prediction accuracy of electric vehicle load considering the multi-dimensional distribution of time and space, so that the substation can cope with the high proportion of new energy output fluctuations more elastic and provide power system peak load regulation services.

[0155] The two-stage regulation potential prediction model for the cluster of electric vehicle charging piles in the substation proposed in the present invention is based on the charging pile aggregation model. In the first stage, the SOC allowable range of the on-board batteries of the electric vehicles participating in the response is calculated according to the duration of the demand response invitation. Then, based on the predicted battery SOC probability density distribution function and the distribution of the charging pile operation status, the adjustable power and adjustable power of the cluster of electric vehicle charging piles in the substation are preliminarily predicted, providing theoretical and data support for whether the charging station participates in a certain power grid demand response. In the second stage, based on the Markov chain theory, the operation status of the charging pile in the previous period is used to approximate the operation status of the current period, that is, the vehicle and pile data counted by the charging station in the period before the start of the demand response is used to accurately predict the real-time adjustable power and adjustable power of the charging piles participating in this demand response. Constructing a two-stage regulation potential prediction model for the cluster of electric vehicle charging piles in the substation is helpful for the formulation of resource optimization allocation strategies and promotes the improvement of the overall operation efficiency of the distribution substation.

[0156] In order to solve the problems of weak generalization ability of current electric vehicle load aggregation models and difficulty in acquiring on-board data, the present invention regards electric vehicles connected to charging piles as virtual energy storage devices, and constructs an electric vehicle charging pile aggregation model with the substation as the minimum unit of V2G demand response based on the Markov chain. By constructing a state distribution matrix and a state transfer matrix, the Markov chain random process equation of the electric vehicle charging piles in the substation is established to realize the prediction of the state space of the charging piles.

[0157] The present invention can predict the probability density function of the state of charge of the on-board battery of the electric vehicle and the probability distribution of the state of charge of the charging pile by collecting, preprocessing and analyzing the data of the historical state of charge of the electric vehicle connected to the charging pile in the substation. On this basis, considering the difference in the adjustment capacity of the charging pile under different operating conditions, a two-stage adjustable capacity prediction model of the cluster of electric vehicle charging piles in the substation is constructed. In the first stage, the adjustable and adjustable power of the electric vehicle load cluster in the response period is predicted based on the probability of the distribution of the state of charge of the electric vehicle charging pile and the probability density function of the state of charge of the on-board battery, providing a decision basis for whether the charging pile participates in this demand response. In the second stage, the vehicle and pile data collected by the electric vehicle charging station in the previous period of demand response are used to accurately predict the adjustable and adjustable power of the electric vehicle load cluster when participating in the demand response, providing technical support for the precise optimization and scheduling of flexible loads in new power systems.

[0158] Example 2

[0159] The present invention based on the same inventive concept also provides an electric vehicle load aggregation system based on a Markov chain, comprising:

[0160] A data acquisition module is used to obtain the historical operation data of charging piles in the distribution station area;

[0161] Probability matrix construction module, used to construct the probability matrix of charging pile operation status distribution based on Markov chain and historical operation data;

[0162] A preliminary judgment module, used to preliminarily predict the adjustable power and adjustable power of the electric vehicle charging pile load cluster in the substation based on the charging pile operation status distribution probability matrix, and determine the charging piles that can participate in the demand response;

[0163] The precise calculation module is used to calculate the real-time adjustable power and adjustable power of the charging pile participating in the demand response based on the historical operation data of the charging pile participating in the demand response.

[0164] Optionally, the probability matrix building module is specifically used for:

[0165] Setting the operating status of the electric vehicle charging pile in the station area, the operating status includes connection status and charging status;

[0166] Based on the historical operation data of charging piles in the distribution station area, the state distribution probability of charging piles in different time periods is analyzed to obtain the probability distribution of each state of the charging piles over time periods;

[0167] The operating state of the charging pile is represented by a Markov chain to obtain the state distribution probability of the charging pile in a certain period of time;

[0168] Constructing a charging pile operation state transfer matrix equation based on the state distribution probability of the charging pile in a certain period of time;

[0169] Performing time series recursion on the charging pile operation state transfer matrix equation to obtain a charging pile operation state distribution probability matrix;

[0170] The operating status of the electric vehicle charging pile in the station area includes: no electric vehicle connected, electric vehicle connected and the charge state value ∈ Connected electric vehicle with state of charge ∈ Connected electric vehicle with state of charge ∈ and They respectively represent the minimum and maximum state of charge of the battery of electric vehicles participating in demand response.

[0171] Optionally, the preliminary determination module includes:

[0172] The scenario determination submodule is used to consider seasonal and holiday factors, and use the fuzzy C-means clustering algorithm to generate typical energy usage scenarios for charging piles based on the daily load curve of the substation throughout the year according to the set parameters;

[0173] The adjustable power determination submodule is used to predict the adjustable power and adjustable power of the electric vehicle charging pile load cluster in the substation based on the typical energy consumption scenario of the charging pile and the probability matrix of the charging pile operation status distribution;

[0174] A judgment submodule, used to determine the electric vehicle charging piles in the area with adjustable power and adjustable power as charging piles that can participate in demand response;

[0175] The setting parameters include: load peak / valley time, load value, and number of peaks and troughs.

[0176] Optionally, the scene determination submodule is specifically used to:

[0177] Step 1: Initialize the membership matrix U;

[0178] Step 2: Calculate the initial cluster center according to the numerical range and distribution of the physical parameters of the electric vehicle;

[0179] Step 3: Update the membership matrix U and cluster center V by calculating the distance between each data point and each cluster center and redistributing the membership of each data point according to the distance;

[0180] Step 4: Determine whether the iteration termination condition is met. If so, proceed to step 5, otherwise return to step 3;

[0181] Step 5: Output clustering results.

[0182] Optionally, the adjustable power determination submodule calculates the adjustable power and the adjustable power of the electric vehicle charging pile load cluster in the substation area according to the following formula:

[0183]

[0184] In the formula, The power can be adjusted downward or the expected power can be adjusted upward for the prediction of the electric vehicle charging pile cluster; The predicted downward adjustable power and upward adjustable power expected value of the electric vehicle charging pile in the kth state; The predicted adjustable power of the electric vehicle charging pile in the second state and the third state respectively; are the predicted adjustable power of the electric vehicle charging piles in the second and fourth states respectively; f(x) is the probability density function of the state of charge of the battery on the electric vehicle; N0 is the number of charging piles in the area; The maximum and minimum state of charge of the battery on board the electric vehicle participating in the demand response; is the state distribution probability of the i-th charging pile being in the second state in time period t; is the discharge power of the electric vehicle connected to the i-th charging pile; is the charging power of the electric vehicle connected to the i-th charging pile; The minimum battery state of charge that must be met when an electric vehicle connected to a charging pile in the second state participates in valley filling demand response; The minimum battery state of charge that must be met when an electric vehicle connected to a charging pile in the second state participates in peak demand response.

[0185] Optionally, the precise calculation module calculates the real-time adjustable power and adjustable power of the charging pile participating in the demand response according to the following formula:

[0186]

[0187] Where: They are respectively the adjustable power that can be reduced and the adjustable power that can be increased of the secondary prediction of the electric vehicle charging pile load cluster; They are respectively the secondary predicted power that can be adjusted downward and the power that can be adjusted upward for the electric vehicle charging pile in the kth state; is the total number of charging piles that are in the second state in the period before the start of demand response and can participate in a peak-shaving demand response. is the total number of charging piles in the second state that can participate in a valley filling demand response in the period before the demand response starts. is the total number of charging piles that are in the third state in the period before the start of demand response and can participate in a peak-shaving demand response. is the total number of charging piles in the fourth state before the start of demand response that can participate in a valley filling demand response, i is the serial number of the charging pile, are the discharge power and charging power of the electric vehicle connected to the i-th charging pile respectively.

[0188] Example 3

[0189] like Figure 6 As shown, the present invention also provides an electronic device, which may be a computer device, a single-chip device, an intelligent mobile device, etc. The electronic device in this embodiment may include a processor, a memory, a transceiver component, etc. The memory, the processor, and the transceiver component are connected via a bus; the memory may be used to store an execution program, and an exemplary execution program may include instructions; the processor is used to execute the instructions stored in the memory. The memory may also be used to store data, which may be called and / or modified when the instructions are executed.

[0190] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in a storage medium to implement a corresponding method flow or a corresponding function, so as to implement the steps of a Markov chain-based electric vehicle load aggregation method in the above-mentioned embodiment.

[0191] Example 4

[0192] Based on the same inventive concept, the present invention also provides a readable storage medium, specifically an electronic device readable storage medium (Memory), which is a memory device in an electronic device for storing programs and data. It can be understood that the storage medium here can include both built-in storage media in electronic devices and, of course, extended storage media supported by electronic devices. The storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by a processor are also stored in the storage space, and these instructions can be one or more execution programs (including program codes). It should be noted that the storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor loads and executes one or more instructions stored in the storage medium, which can implement the steps of a Markov chain-based electric vehicle load aggregation method in the above embodiment.

[0193] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0194] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0195] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1A function specified in one or more boxes.

[0196] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0197] The above are merely embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are included in the scope of the claims of the present invention to be approved.

Claims

1. A method for electric vehicle load aggregation based on Markov chain, characterized in that: include: Obtain historical operation data of charging piles in distribution station areas; Construct the probability matrix of charging pile operation status distribution based on Markov chain and historical operation data; Preliminarily predict the adjustable power and adjustable power of the electric vehicle charging pile load cluster in the substation based on the charging pile operation status distribution probability matrix, and determine the charging piles that can participate in demand response; Based on the historical operation data of the charging piles that can participate in the demand response, the real-time adjustable power and adjustable power of the charging piles that can participate in the demand response are calculated.

2. The method according to claim 1, characterized in that The method of constructing a charging pile operation state distribution probability matrix based on the Markov chain and historical operation data includes: Setting the operating status of the electric vehicle charging pile in the station area, the operating status includes connection status and charging status; Based on the historical operation data of charging piles in the distribution station area, the state distribution probability of charging piles in different time periods is analyzed to obtain the probability distribution of each state of the charging piles over time periods; The operating state of the charging pile is represented by a Markov chain to obtain the state distribution probability of the charging pile in a certain period of time; Constructing a charging pile operation state transfer matrix equation based on the state distribution probability of the charging pile in a certain period of time; Performing time series recursion on the charging pile operation state transfer matrix equation to obtain a charging pile operation state distribution probability matrix; The operating status of the electric vehicle charging pile in the station area includes: no electric vehicle connected, and They respectively represent the minimum and maximum state of charge of the battery of electric vehicles participating in demand response.

3. The method according to claim 2, characterized in that The preliminarily predicting the adjustable power and adjustable power of the electric vehicle charging pile load cluster in the substation based on the charging pile operation state distribution probability matrix, and determining the charging piles that can participate in the demand response, includes: Taking into account seasonal and holiday factors, the daily load curve of the substation throughout the year is used according to the set parameters using the fuzzy C-means clustering algorithm to generate typical energy usage scenarios for charging piles; Based on the typical energy consumption scenario of the charging piles and the probability matrix of the charging pile operation status distribution, the adjustable power and adjustable power of the electric vehicle charging pile load cluster in the substation are predicted; Identify the electric vehicle charging piles in the area with adjustable power and adjustable power as charging piles that can participate in demand response; The setting parameters include: load peak / valley time, load value, and number of peaks and troughs.

4. The method according to claim 3, characterized in that Taking into account seasonal and holiday factors, the daily load curve of the substation throughout the year is used according to the set parameters using the fuzzy C-means clustering algorithm to generate typical energy usage scenarios for charging piles, including: Step 1: Initialize the membership matrix U; Step 2: Calculate the initial cluster center according to the numerical range and distribution of the physical parameters of the electric vehicle; Step 3: Update the membership matrix U and cluster center V by calculating the distance between each data point and each cluster center and redistributing the membership of each data point according to the distance; Step 4: Determine whether the iteration termination condition is met. If so, proceed to step 5, otherwise return to step 3; Step 5: Output clustering results.

5. The method according to claim 3, characterized in that The adjustable power and adjustable power of the electric vehicle charging pile load cluster in the substation area are calculated as follows: In the formula, The power can be adjusted downward or the expected power can be adjusted upward for the prediction of the electric vehicle charging pile cluster; The predicted downward adjustable power and upward adjustable power expected value of the electric vehicle charging pile in the kth state; The predicted adjustable power of the electric vehicle charging pile in the second state and the third state respectively; are the predicted adjustable power of the electric vehicle charging piles in the second and fourth states respectively; f(x) is the probability density function of the state of charge of the battery on the electric vehicle; N0 is the number of charging piles in the area; The maximum and minimum state of charge of the battery on board the electric vehicle participating in the demand response; is the state distribution probability of the i-th charging pile being in the second state in time period t; is the discharge power of the electric vehicle connected to the i-th charging pile; is the charging power of the electric vehicle connected to the i-th charging pile; The minimum battery state of charge that must be met when an electric vehicle connected to a charging pile in the second state participates in valley filling demand response; is the minimum battery state of charge that must be met when an electric vehicle connected to a charging pile in the second state participates in peak demand response; f(x) is the probability density function of the battery state of charge of the electric vehicle, and x is a random variable with a value in the interval [0,1], representing the battery state of charge of a certain electric vehicle.

6. The method according to claim 1, characterized in that The real-time adjustable power and adjustable power of the charging pile participating in the demand response are calculated as follows: Where: They are respectively the adjustable power that can be reduced and the adjustable power that can be increased of the secondary prediction of the electric vehicle charging pile load cluster; They are respectively the secondary predicted power that can be adjusted downward and the power that can be adjusted upward for the electric vehicle charging pile in the kth state; is the total number of charging piles that are in the second state in the period before the start of demand response and can participate in a peak-shaving demand response. is the total number of charging piles in the second state that can participate in a valley filling demand response in the period before the demand response starts. is the total number of charging piles that are in the third state in the period before the start of demand response and can participate in a peak-shaving demand response. is the total number of charging piles in the fourth state before the start of demand response that can participate in a valley filling demand response, i is the serial number of the charging pile, are the discharge power and charging power of the electric vehicle connected to the i-th charging pile respectively.

7. An electric vehicle load aggregation system based on Markov chain, characterized in that: include: A data acquisition module is used to obtain the historical operation data of charging piles in the distribution station area; Probability matrix construction module, used to construct the probability matrix of charging pile operation status distribution based on Markov chain and historical operation data; A preliminary judgment module, used to preliminarily predict the adjustable power and adjustable power of the electric vehicle charging pile load cluster in the substation based on the charging pile operation status distribution probability matrix, and determine the charging piles that can participate in the demand response; The precise calculation module is used to calculate the real-time adjustable power and adjustable power of the charging pile participating in the demand response based on the historical operation data of the charging pile participating in the demand response.

8. The system according to claim 7, characterized in that The probability matrix building module is specifically used for: Setting the operating status of the electric vehicle charging pile in the station area, the operating status includes connection status and charging status; Based on the historical operation data of charging piles in the distribution station area, the state distribution probability of charging piles in different time periods is analyzed to obtain the probability distribution of each state of the charging piles over time periods; The operating state of the charging pile is represented by a Markov chain to obtain the state distribution probability of the charging pile in a certain period of time; Constructing a charging pile operation state transfer matrix equation based on the state distribution probability of the charging pile in a certain period of time; Performing time series recursion on the charging pile operation state transfer matrix equation to obtain a charging pile operation state distribution probability matrix; The operating status of the electric vehicle charging pile in the station area includes: no electric vehicle connected, and They respectively represent the minimum and maximum state of charge of the battery of electric vehicles participating in demand response.

9. The system according to claim 7, characterized in that The preliminary judgment module includes: The scenario determination submodule is used to consider seasonal and holiday factors, and use the fuzzy C-means clustering algorithm to generate typical energy usage scenarios for charging piles based on the daily load curve of the substation throughout the year according to the set parameters; The adjustable power determination submodule is used to predict the adjustable power and adjustable power of the electric vehicle charging pile load cluster in the substation based on the typical energy consumption scenario of the charging pile and the probability matrix of the charging pile operation status distribution; A judgment submodule, used to determine the electric vehicle charging piles in the area with adjustable power and adjustable power as charging piles that can participate in demand response; The setting parameters include: load peak / valley time, load value, and number of peaks and troughs.

10. The system according to claim 9, characterized in that The scene determination submodule is specifically used for: Step 1: Initialize the membership matrix U; Step 2: Calculate the initial cluster center according to the numerical range and distribution of the physical parameters of the electric vehicle; Step 3: Update the membership matrix U and cluster center V by calculating the distance between each data point and each cluster center and redistributing the membership of each data point according to the distance; Step 4: Determine whether the iteration termination condition is met. If so, proceed to step 5, otherwise return to step 3; Step 5: Output clustering results.

11. The system according to claim 9, characterized in that The adjustable power determination submodule calculates the adjustable power and adjustable power of the electric vehicle charging pile load cluster in the substation area according to the following formula: In the formula, The power can be adjusted downward or the expected power can be adjusted upward for the prediction of the electric vehicle charging pile cluster; The predicted downward adjustable power and upward adjustable power expected value of the electric vehicle charging pile in the kth state; The predicted adjustable power of the electric vehicle charging pile in the second state and the third state respectively; are the predicted adjustable power of the electric vehicle charging piles in the second and fourth states respectively; f(x) is the probability density function of the state of charge of the battery on the electric vehicle; N0 is the number of charging piles in the area; The maximum and minimum state of charge of the battery on board the electric vehicle participating in the demand response; is the state distribution probability of the i-th charging pile being in the second state in time period t; is the discharge power of the electric vehicle connected to the i-th charging pile; is the charging power of the electric vehicle connected to the i-th charging pile; The minimum battery state of charge that must be met when an electric vehicle connected to a charging pile in the second state participates in valley filling demand response; is the minimum battery state of charge that must be met when an electric vehicle connected to a charging pile in the second state participates in peak demand response; f(x) is the probability density function of the battery state of charge of the electric vehicle, and x is a random variable with a value in the interval [0,1], representing the battery state of charge of a certain electric vehicle.

12. The system of claim 7, wherein: The precise calculation module calculates the real-time adjustable power and adjustable power of the charging pile participating in the demand response according to the following formula: Where: They are respectively the adjustable power that can be reduced and the adjustable power that can be increased of the secondary prediction of the electric vehicle charging pile load cluster; They are respectively the secondary predicted power that can be adjusted downward and the power that can be adjusted upward for the electric vehicle charging pile in the kth state; is the total number of charging piles that are in the second state in the period before the start of demand response and can participate in a peak-shaving demand response. is the total number of charging piles in the second state that can participate in a valley filling demand response in the period before the demand response starts. is the total number of charging piles that are in the third state in the period before the start of demand response and can participate in a peak-shaving demand response. is the total number of charging piles in the fourth state before the start of demand response that can participate in a valley filling demand response, i is the serial number of the charging pile, are the discharge power and charging power of the electric vehicle connected to the i-th charging pile respectively.

13. An electronic device, characterized in that: include: at least one processor and memory; The memory and the processor are connected via a bus; The memory is used to store one or more programs; When the one or more programs are executed by the at least one processor, an electric vehicle load aggregation method based on a Markov chain as described in any one of claims 1 to 6 is implemented.

14. A readable storage medium, characterized in that: An execution program is stored thereon, and when the execution program is executed, an electric vehicle load aggregation method based on a Markov chain as described in any one of claims 1 to 6 is implemented.

Images