Power grid dispatching method, device, computer equipment and storage medium

Abstract

The present application relates to a power grid dispatching method, apparatus, computer equipment, and storage medium. The method comprises: obtaining electric vehicle charging demand data; obtaining a clustering object difference dataset of electric vehicles based on the electric vehicle charging demand data, obtaining a dynamically adjustable parameter based on the clustering object difference dataset, obtaining a fused object difference item for electric vehicles at each charging station based on the clustering object difference dataset and the dynamically adjustable parameter, and forming an electric vehicle data distance matrix from the fused object difference items; normalizing the distance matrix corresponding to each charging station and performing clustering processing to generate electric vehicle charging cluster groups corresponding to each charging station; obtaining aggregated information based on the electric vehicle charging cluster groups and an electric vehicle dispatching potential model constructed based on charging demand, and generating an electric vehicle charging dispatching plan based on the aggregated information. The present method can effectively reduce the peak-to-valley difference in load at power grid nodes and save power grid operating costs.

Description

Technical Field

[0001] The present application relates to the field of power system optimization and dispatching, and in particular to a method, device, computer equipment, and storage medium for electric vehicle clustering to participate in power grid dispatching. Background Art

[0002] With the development of computer technology, more and more objects have been widely used. The disorderly charging of a large number of electric vehicles has increased the pressure on power grid scheduling. Research programs that use electric vehicle resources to cluster and aggregate to participate in power grid scheduling optimization have attracted widespread attention from energy departments around the world.

[0003] The existing technology cannot fully utilize electric vehicle resources when clustering and aggregating electric vehicle resources, cannot effectively reduce the peak-to-valley difference of grid node loads and save grid operating costs. Summary of the Invention

[0004] Based on this, it is necessary to provide a power grid scheduling method, device, computer equipment, and computer-readable storage medium to address the above technical problems, which can effectively reduce the peak-to-valley difference in power grid node load and save power grid operating costs.

[0005] A power grid dispatching method, the method comprising:

[0006] Obtaining electric vehicle charging demand data uploaded by each distributed charging station. The charging demand data includes at least the following multi-dimensional data: the arrival time of the electric vehicle, the future departure time of the electric vehicle, the battery level of the electric vehicle upon arrival, and the user's desired charging level;

[0007] Perform cluster object difference calculation on the electric vehicle charging demand data corresponding to each distributed charging station, and obtain a cluster object difference data set corresponding to each electric vehicle in each distributed charging station. The cluster object difference data set includes cluster object difference items corresponding to each dimension data;

[0008] Based on the cluster object difference data set, the corresponding dynamic adjustable parameters of each dimension data are calculated;

[0009] The cluster object difference items in the cluster object difference data set are fused based on dynamically adjustable parameters to obtain the fused object difference items corresponding to each cluster object difference data set. The fused object difference items corresponding to the same distributed charging station form the electric vehicle data distance matrix corresponding to the distributed charging station.

[0010] Based on the electric vehicle data distance matrix corresponding to the distributed charging stations, the electric vehicles corresponding to each distributed charging station are clustered to obtain the electric vehicle charging cluster groups corresponding to each distributed charging station;

[0011] Based on each electric vehicle charging cluster group and the electric vehicle scheduling potential model, the aggregation information corresponding to each electric vehicle charging cluster group is calculated;

[0012] Generate charging scheduling plans for electric vehicles based on aggregated information.

[0013] A power grid dispatching device, comprising:

[0014] A charging demand data acquisition module is used to acquire the electric vehicle charging demand data uploaded by each distributed charging station. The charging demand data includes at least the following multi-dimensional data: the arrival time of the electric vehicle, the future departure time of the electric vehicle, the vehicle power level when the electric vehicle arrives, and the user's desired charging level;

[0015] The electric vehicle difference determination module is used to perform cluster object difference calculation on the electric vehicle charging demand data corresponding to each distributed charging station, and obtain a cluster object difference data set corresponding to each electric vehicle in each distributed charging station, wherein the cluster object difference data set includes cluster object difference items corresponding to each dimensional data;

[0016] A dynamic parameter determination module is used to calculate the corresponding dynamic adjustable parameters of each dimension data based on the cluster object difference data set;

[0017] A distance matrix determination module is used to fuse the cluster object difference items in the cluster object difference data set based on dynamically adjustable parameters to obtain the fused object difference items corresponding to each cluster object difference data set. The fused object difference items corresponding to the same distributed charging station constitute the electric vehicle data distance matrix corresponding to the distributed charging station;

[0018] An electric vehicle clustering module is used to cluster the electric vehicles corresponding to each distributed charging station based on the electric vehicle data distance matrix corresponding to the distributed charging station, and obtain the electric vehicle charging cluster group corresponding to each distributed charging station;

[0019] An aggregation information determination module is used to calculate the aggregation information corresponding to each electric vehicle charging cluster group based on each electric vehicle charging cluster group and the electric vehicle scheduling potential model;

[0020] The signal processing module is used to send the aggregated information to the city power grid and to send the electric vehicle charging scheduling plan to each electric vehicle charging station.

[0021] In one embodiment, the electric vehicle difference determination module is further used to calculate the difference distance corresponding to each dimension in the multi-dimensional data based on the charging demand data between electric vehicles in the same charging station, and obtain the cluster object difference items corresponding to each electric vehicle in each dimension; the cluster object difference items form a corresponding cluster object difference data set between each electric vehicle in the charging station.

[0022] In one embodiment, the dynamic parameter determination module is also used to fuse the difference items of each cluster object to obtain a difference fusion item; and respectively calculate the proportion of the corresponding cluster object difference item and the difference fusion item of each dimensional data to obtain the corresponding dynamic adjustable parameters of each dimensional data.

[0023] In one embodiment, the distance matrix determination module is also used to obtain clustering difference factors and parameter factors, where the clustering difference factors include the corresponding clustering object difference items of each dimensional data, and the parameter factors include the corresponding dynamic adjustable parameters of each dimensional data; the clustering difference factors and the parameter factors are weightedly fused to obtain the fused object difference items corresponding to each clustering object difference data set.

[0024] In one embodiment, the electric vehicle clustering module is further used to normalize the difference items of each fusion object in the electric vehicle data distance matrix to obtain adjacency difference items; form an adjacency matrix with the adjacency difference items; calculate the eigenvalues ​​of the adjacency matrix to obtain the eigenvectors corresponding to the eigenvalues; arrange the eigenvectors in descending order according to the corresponding eigenvalues ​​to form an eigenvector matrix; and cluster the row vectors of the eigenvector matrix using a clustering algorithm to generate electric vehicle charging cluster groups.

[0025] In one embodiment, the aggregate information determination module is further used to obtain the electric vehicle power constraints, which include the maximum power of the electric vehicle, the original power of the electric vehicle, the upper bound of the scheduling potential at the previous moment, the lower bound of the scheduling potential at the previous moment, the minimum power of the electric vehicle, the charging amount expected by the user, the power consumption of the electric vehicle, and the charging amount of the electric vehicle. The initial values ​​of the upper bound of the scheduling potential at the previous moment and the lower bound of the scheduling potential at the previous moment are both the original power of the electric vehicle; based on the electric vehicle power constraints, an electric vehicle scheduling potential model is obtained, which includes the upper bound of the scheduling potential and the lower bound of the scheduling potential. The upper bound of the scheduling potential takes the minimum value of the upper bound constraint items. The upper bound constraint items include the maximum power of the electric vehicle, the charging amount of the electric vehicle, and the upper bound of the scheduling potential. The upper bounds of the scheduling potential at a moment are integrated, and the lower bound of the scheduling potential takes the maximum value among the lower bound constraints. The lower bound constraints include the lower bound of the scheduling potential at the previous moment minus the power consumption of the electric vehicle, the minimum power of the electric vehicle, and the charging amount that the user expects to achieve minus the power consumption of the electric vehicle; the upper bounds of the scheduling potential of the electric vehicles in each cluster in the electric vehicle charging cluster group are integrated to obtain the upper bound of the scheduling potential of each cluster in the electric vehicle charging cluster group; the lower bounds of the scheduling potential of the electric vehicles in each cluster in the electric vehicle charging cluster group are integrated to obtain the lower bound of the scheduling potential of each cluster in the electric vehicle charging cluster group; the upper bound of the scheduling potential of each cluster in the electric vehicle charging cluster group and the upper bound of the scheduling potential are combined into the aggregated information corresponding to each electric vehicle charging cluster group.

[0026] In one embodiment, the signal processing module is also used to obtain the charging power of the electric vehicle; send the aggregated information to the urban power grid to obtain a scheduling signal generated based on the aggregated information and returned by the urban power grid; fuse the scheduling signal with the charging power of the electric vehicle to obtain a signal decomposition objective function; calculate the minimum value of the signal decomposition objective function to obtain a charging scheduling plan for the electric vehicle, and the charging scheduling plan for the electric vehicle includes a charging curve of the electric vehicle.

[0027] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0028] Obtaining electric vehicle charging demand data uploaded by each distributed charging station. The charging demand data includes at least the following multi-dimensional data: the arrival time of the electric vehicle, the future departure time of the electric vehicle, the battery level of the electric vehicle upon arrival, and the user's desired charging level;

[0029] Perform cluster object difference calculation on the electric vehicle charging demand data corresponding to each distributed charging station, and obtain a cluster object difference data set corresponding to each electric vehicle in each distributed charging station. The cluster object difference data set includes cluster object difference items corresponding to each dimension data;

[0030] Based on the cluster object difference data set, the corresponding dynamic adjustable parameters of each dimension data are calculated;

[0031] The cluster object difference items in the cluster object difference data set are fused based on dynamically adjustable parameters to obtain the fused object difference items corresponding to each cluster object difference data set. The fused object difference items corresponding to the same distributed charging station form the electric vehicle data distance matrix corresponding to the distributed charging station.

[0032] Based on the electric vehicle data distance matrix corresponding to the distributed charging stations, the electric vehicles corresponding to each distributed charging station are clustered to obtain the electric vehicle charging cluster groups corresponding to each distributed charging station;

[0033] Based on each electric vehicle charging cluster group and the electric vehicle scheduling potential model, the aggregation information corresponding to each electric vehicle charging cluster group is calculated;

[0034] Generate charging scheduling plans for electric vehicles based on aggregated information.

[0035] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0036] Obtaining electric vehicle charging demand data uploaded by each distributed charging station. The charging demand data includes at least the following multi-dimensional data: the arrival time of the electric vehicle, the future departure time of the electric vehicle, the battery level of the electric vehicle upon arrival, and the user's desired charging level;

[0037] Perform cluster object difference calculation on the electric vehicle charging demand data corresponding to each distributed charging station, and obtain a cluster object difference data set corresponding to each electric vehicle in each distributed charging station. The cluster object difference data set includes cluster object difference items corresponding to each dimension data;

[0038] Based on the cluster object difference data set, the corresponding dynamic adjustable parameters of each dimension data are calculated;

[0039] The cluster object difference items in the cluster object difference data set are fused based on dynamically adjustable parameters to obtain the fused object difference items corresponding to each cluster object difference data set. The fused object difference items corresponding to the same distributed charging station form the electric vehicle data distance matrix corresponding to the distributed charging station.

[0040] Based on the electric vehicle data distance matrix corresponding to the distributed charging stations, the electric vehicles corresponding to each distributed charging station are clustered to obtain the electric vehicle charging cluster groups corresponding to each distributed charging station;

[0041] Based on each electric vehicle charging cluster group and the electric vehicle scheduling potential model, the aggregation information corresponding to each electric vehicle charging cluster group is calculated;

[0042] Generate charging scheduling plans for electric vehicles based on aggregated information.

[0043] The above-mentioned power grid dispatching method, apparatus, computer equipment and storage medium obtain electric vehicle charging demand data uploaded by each distributed charging station, the charging demand data including at least the following multi-dimensional data: the arrival time of the electric vehicle, the future departure time of the electric vehicle, the electric vehicle's battery level upon arrival, and the user's desired charging amount; perform cluster object difference calculation on the electric vehicle charging demand data corresponding to each distributed charging station to obtain a cluster object difference dataset corresponding to each electric vehicle in each distributed charging station, the cluster object difference dataset including cluster object difference items corresponding to each dimensional data; calculate the corresponding dynamic adjustable parameters for each corresponding dimensional data based on the cluster object difference dataset; fuse the cluster object difference items in the cluster object difference dataset based on the dynamic adjustable parameters to obtain fused object difference items corresponding to each cluster object difference dataset, and the fused object difference items corresponding to the same distributed charging station constitute the electric vehicle data distance matrix corresponding to the distributed charging station; cluster the electric vehicles corresponding to each distributed charging station based on the electric vehicle data distance matrix corresponding to the distributed charging station to obtain electric vehicle charging cluster groups corresponding to each distributed charging station; calculate the aggregation information corresponding to each electric vehicle charging cluster group based on each electric vehicle charging cluster group and the electric vehicle scheduling potential model; and generate the electric vehicle charging scheduling plan based on the aggregation information. In this way, by establishing multi-dimensional charging demand data and dynamically adjustable parameters, the time differences and power differences in electric vehicle charging behavior can be targetedly adjusted, so that electric vehicles that are closer in charging time and charging amount are aggregated into a cluster group. The electric vehicle aggregation information generated by the electric vehicle scheduling potential model constructed based on the electric vehicle charging demand can better reflect the time characteristics and power characteristics of electric vehicle charging behavior, thereby effectively reducing the peak-to-valley difference in grid node load and saving grid operating costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 A diagram showing an application environment of a power grid dispatching method in one embodiment;

[0045] Figure 2 1 is a flow chart of a power grid dispatching method according to an embodiment;

[0046] Figure 3 A schematic diagram of a process for determining a cluster object difference data set in one embodiment;

[0047] Figure 4 A schematic diagram of a process for determining dynamically adjustable parameters in one embodiment;

[0048] Figure 5 A schematic diagram of a process for determining difference items of fusion objects in one embodiment;

[0049] Figure 6 A schematic diagram of a process for determining charging clusters of electric vehicles in one embodiment;

[0050] Figure 7 A schematic diagram of a process for determining aggregate information in one embodiment;

[0051] Figure 8 A schematic diagram of a flow chart for determining a charging schedule for an electric vehicle in one embodiment;

[0052] Figure 9 A schematic diagram of changes in the upper and lower bound curves of a single electric vehicle scheduling potential model in one embodiment;

[0053] Figure 10 A schematic diagram of the change of the upper and lower bound curves of the electric vehicle charging cluster scheduling potential model in one embodiment;

[0054] Figure 11 A schematic diagram comparing power grid dispatching results of different schemes in one embodiment;

[0055] Figure 12 This is a structural block diagram of a power grid dispatching device in one embodiment;

[0056] Figure 13 is a diagram of the internal structure of a computer device in one embodiment;

[0057] Figure 14 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0059] The power grid dispatching method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Figure 1As shown, the application environment includes a terminal 102, a server 104, and a city grid 106. Terminal 102 communicates with server 104 via a network, and server 104 communicates with city grid 106 via a network. Terminal 102 can send electric vehicle charging demand data to server 104. Server 104 can cluster and aggregate the electric vehicle charging demand data to generate electric vehicle aggregate information, which is then sent to city grid 106. City grid 106 generates a scheduling signal based on the electric vehicle aggregate information and sends the scheduling signal back to server 104. Server 104 decomposes the scheduling signal to obtain an electric vehicle charging scheduling plan, which is then sent to terminal 102.

[0060] Terminal 102 may be, but is not limited to, various electric vehicles, charging stations, personal computers, laptops, smartphones, tablet computers, smart cameras, and portable wearable devices. Server 104 may be an independent physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The terminal and server may be connected directly or indirectly via wired or wireless communication, which is not limited in this application.

[0061] In one embodiment, Figure 2 As shown, a power grid dispatching method is provided, which is applied to Figure 1 The following steps are used as an example to illustrate the server in the example:

[0062] Step S202: Acquire the electric vehicle charging demand data uploaded by each distributed charging station. The charging demand data includes at least the following multi-dimensional data: the arrival time of the electric vehicle, the future departure time of the electric vehicle, the vehicle power level when the electric vehicle arrives, and the user's expected charging amount.

[0063] Among them, the electric vehicle charging demand data is a quantitative reflection of the charging demand of electric vehicles. The arrival time of the electric vehicle refers to the time when the electric vehicle arrives at the charging station. The future departure time of the electric vehicle is the time when the electric vehicle leaves the charging station after completing charging. The vehicle power at arrival is the remaining power of the vehicle when the electric vehicle arrives at the charging station. The user's expected charging amount is the amount of power that the electric vehicle needs to be charged at the charging station.

[0064] Specifically, when an electric vehicle arrives at a charging station, various sensors of various terminals such as electric vehicles or charging piles will collect data of various dimensions of the electric vehicle's charging demand data in real time, and the processing server will receive and store the charging demand data through the communication network in a wired or wireless manner.

[0065] Step S204 , performing cluster object difference calculation on the electric vehicle charging demand data corresponding to each distributed charging station, and obtaining a cluster object difference data set corresponding to each electric vehicle in each distributed charging station.

[0066] Among them, each distributed charging station refers to the charging station that each electric vehicle arrives at, and the cluster object difference calculation refers to the difference calculation of the charging demand data of electric vehicles in the same charging station in each dimension.

[0067] Specifically, the processing server calculates the difference between the charging demand data of two electric vehicles in the same charging station in each dimension, and then uses the calculation results as the clustering object difference items corresponding to the data of each dimension in the clustering object difference data set.

[0068] Step S206 : Calculate the corresponding dynamically adjustable parameters of the corresponding dimensional data based on the cluster object difference data set.

[0069] Among them, the dynamic adjustable parameters correspond to the various dimensional data of the cluster object difference dataset. Each cluster object difference dataset corresponds to a set of dynamic adjustable parameters. The number of each set of dynamic adjustable parameters is equal to the dimension of the corresponding cluster object difference dataset. Each dynamic adjustable parameter can determine its own weight relative to other adjustable parameters in the same group based on the numerical size of the cluster object difference item in the corresponding cluster object difference dataset.

[0070] Specifically, the processing server fuses the dimensional data in the cluster object difference dataset to obtain a corresponding set of dynamically adjustable parameters, and the number of dynamically adjustable parameters in each set is equal to the number of dimensions of the corresponding cluster object difference dataset.

[0071] Step S208 : The cluster object difference items in the cluster object difference data set are fused based on the dynamically adjustable parameters to obtain fused object difference items corresponding to each cluster object difference data set.

[0072] Among them, each fusion object difference item corresponds one-to-one to the cluster object difference dataset within the same charging station, that is, each cluster object difference dataset corresponds to a fusion object difference item. Similarly, each fusion object difference item also corresponds one-to-one to a set of dynamically adjustable parameters within the same charging station. Each fusion object difference item is used to characterize the degree of difference in charging requirements between every two electric vehicles.

[0073] Specifically, the processing server calculates each fused object difference item by fusing the corresponding dimension data of the cluster object difference data set with a corresponding set of dynamically adjustable parameters.

[0074] For example, the processing server will have n cluster object difference data sets (N1, N2, ..., N n ) and the corresponding n groups of dynamically adjustable parameters (V1, V2, ..., V n ), then the n clustering object difference data sets (N1, N2, ..., N n ) and the corresponding n groups of dynamically adjustable parameters (V1, V2, ..., V n ) to generate n fusion object difference items (D1, D2, ..., D n ), and D1 is obtained by the fusion of N1 and V1, D2 is obtained by the fusion of N2 and V2, and so on. n By N n and V n Fusion obtained.

[0075] Step S210 , forming an electric vehicle data distance matrix corresponding to the distributed charging station by using the difference items of the fusion objects corresponding to the same distributed charging station.

[0076] Among them, the electric vehicle data distance matrix corresponds to the distributed charging stations, that is, each charging station corresponds to its own electric vehicle data distance matrix. The electric vehicle data distance matrix is ​​used to characterize the degree of difference in charging requirements of electric vehicles in the same charging station. Each item in it characterizes the degree of difference in charging requirements between two cars in the same charging station.

[0077] Specifically, the processing server assigns an integer serial number from 1 to n to the n electric vehicles in the same charging station, and uses the difference items of the fusion objects between the i-th electric vehicle and the j-th electric vehicle as the matrix elements of the i-th row and j-th column in the electric vehicle data distance matrix. The diagonal elements of the electric vehicle data distance matrix (that is, the elements with the same number of rows and columns) are uniformly set to 0, and the matrix elements of the i-th row and j-th column are the same as the matrix elements of the j-th row and i-th column.

[0078] For example, the computer device uses the fusion object difference item D(i,j) between the i-th electric vehicle and the j-th electric vehicle as the matrix element of the i-th row and j-th column in the electric vehicle data distance matrix D, that is:

[0079] Where D(i,j) represents the fusion object difference term between the i-th electric vehicle and the j-th electric vehicle, and D(n,1)=D(1,n), both of which represent the degree of difference in charging requirements between the first electric vehicle and the n-th electric vehicle (i.e., the fusion object difference term).

[0080] Step S212 , clustering the electric vehicles corresponding to each distributed charging station based on the electric vehicle data distance matrix corresponding to the distributed charging station to obtain the electric vehicle charging cluster group corresponding to each distributed charging station.

[0081] Among them, clustering refers to the classification of many research individuals according to predetermined criteria and rules, so that individuals with similar attributes can be classified into one category under the predetermined criteria and rules. Here, clustering based on the electric vehicle data distance matrix to generate electric vehicle charging cluster groups means that electric vehicles with more similar charging time and charging amount demand attributes are classified into one statistical category. As a whole research object, each charging station corresponds to an electric vehicle data distance matrix, and based on the electric vehicle data distance matrix corresponding to each charging station, multiple electric vehicle charging cluster groups can be extracted.

[0082] Specifically, the processing server performs clustering calculations on the electric vehicle data distance matrix corresponding to each charging station, and generates electric vehicle charging cluster groups corresponding to each charging station.

[0083] Step S214 , obtaining aggregation information corresponding to each electric vehicle charging cluster group based on each electric vehicle charging cluster group and the electric vehicle scheduling potential model.

[0084] Among them, the electric vehicle scheduling potential model is constructed based on the charging demand of a single electric vehicle, and is used to characterize the power constraint range that the electric vehicle can allow from the time of arrival to the time of departure, such as Figure 9 As shown in ; the aggregation information reflects the power constraint range that can be allowed for each electric vehicle charging cluster from the arrival time to the departure time, as shown in Figure 10 shown.

[0085] Specifically, the processing server constructs an electric vehicle scheduling potential model based on the charging demand of a single electric vehicle, integrates the scheduling potential of each electric vehicle in the electric vehicle charging cluster group, and generates aggregate information corresponding to each electric vehicle charging cluster group.

[0086] Step S216: Generate a charging schedule for the electric vehicle based on the aggregated information.

[0087] Among them, the charging scheduling plan of electric vehicles includes the charging curve of each electric vehicle in each charging station.

[0088] Specifically, the processing server sends the aggregated information to the city grid 106 and obtains a dispatch signal returned by the city grid 106. The processing server then decomposes the dispatch signal according to the electric vehicle charging power constraint to generate a charging dispatch plan for a single electric vehicle.

[0089] In the above-mentioned power grid dispatching method, by acquiring electric vehicle charging demand data uploaded by each distributed charging station, the charging demand data includes at least the following multi-dimensional data: the arrival time of the electric vehicle, the future departure time of the electric vehicle, the electric vehicle's battery level upon arrival, and the user's desired charging amount; clustering object difference calculation is performed on the electric vehicle charging demand data corresponding to each distributed charging station to obtain a clustering object difference dataset corresponding to each electric vehicle in each distributed charging station, the clustering object difference dataset including clustering object difference items corresponding to each dimensional data; dynamically adjustable parameters corresponding to each dimensional data are calculated based on the clustering object difference dataset; the clustering object difference items in the clustering object difference dataset are fused based on the dynamic adjustable parameters to obtain fused object difference items corresponding to each clustering object difference dataset, and the fused object difference items corresponding to the same distributed charging station form an electric vehicle data distance matrix corresponding to the distributed charging station; the electric vehicles corresponding to each distributed charging station are clustered based on the electric vehicle data distance matrix corresponding to the distributed charging station to obtain electric vehicle charging cluster groups corresponding to each distributed charging station; aggregate information corresponding to each electric vehicle charging cluster group is calculated based on each electric vehicle charging cluster group and an electric vehicle scheduling potential model; and a charging dispatch plan for the electric vehicle is generated based on the aggregate information. In this way, by establishing multi-dimensional charging demand data and dynamically adjustable parameters, the time differences and power differences of electric vehicle charging behaviors can be adjusted in a targeted manner, so that electric vehicles that are closer in charging time and charging amount are aggregated into a cluster group. The electric vehicle aggregation information generated by the electric vehicle scheduling potential model constructed in combination with the electric vehicle charging demand can better reflect the time characteristics and power characteristics of electric vehicle charging behaviors, thereby effectively reducing the peak-to-valley difference of grid node loads and saving grid operating costs. Figure 11 shown.

[0090] In one embodiment, Figure 3 As shown, cluster object difference calculation is performed on the electric vehicle charging demand data corresponding to each distributed charging station to obtain a cluster object difference data set corresponding to each electric vehicle in each distributed charging station, including:

[0091] Step S302 : Based on the charging demand data of electric vehicles in the same charging station, the difference distance corresponding to each dimension in the multi-dimensional data is calculated to obtain the clustering object difference item corresponding to each electric vehicle in each dimension.

[0092] Specifically, the processing server calculates the bi-norm of the charging demand data of two electric vehicles in the same charging station in each dimension, where the bi-norm refers to the straight-line distance between two vector matrices in space, and then uses the calculation results as the clustering object difference items corresponding to the data of each dimension in the clustering object difference data set.

[0093] For example, there are electric cars i and j in a charging station: the charging demand data of electric car i is CR = {t a,i ,t d,i ,SOC a,i ,SOC desire,i}, where t a,i is the time when electric vehicle i arrives at the charging station, t d,i It is the time for electric vehicles to leave in the future, SOC a,i is the battery capacity of electric vehicle i when it arrives, SOC desire,i is the desired charging capacity of electric vehicle i; the charging demand data of electric vehicle j is CR = {t a,j ,t d,j ,SOC a,j ,SOC desire,j}, where t a,j is the time when electric vehicle j arrives at the charging station, t d,j Is the future departure time of electric vehicles, SOC a,j is the battery capacity of electric vehicle j when it arrives, SOC desire,j is the expected charging capacity of electric vehicle j; the clustering object difference calculation is performed on electric vehicle i and electric vehicle j, then N={||t a,i -t a,j ||2,||t d,i -t d,j ||2,||SOC a,i -SOC a,j ||2,||SOC desire,i -SOC desire,j ||2}, where N is the cluster object difference dataset, and ||t a,i -t a,j ||2 is the cluster object difference item in the cluster object difference dataset N corresponding to electric vehicle i and electric vehicle j in the electric vehicle arrival time dimension; ||t d,i -t d,j||2 is the cluster object difference item in the cluster object difference dataset N corresponding to electric vehicle i and electric vehicle j in the dimension of the future departure time of electric vehicles; ||SOC a,i -SOC a,j ||2 is the cluster object difference item in the vehicle power dimension when the electric vehicle arrives in the cluster object difference dataset N corresponding to electric vehicle i and electric vehicle j; ||SOC desire,i -SOC desire,j ||2 is the cluster object difference item in the cluster object difference dataset N corresponding to electric vehicle i and electric vehicle j in the dimension of user expected charging amount.

[0094] Step S304 : clustering object difference items form a clustering object difference data set corresponding to each electric vehicle in the corresponding charging station.

[0095] In this embodiment, the two norms corresponding to the data of each dimension of the charging demand data between electric vehicles are used as the difference distance of the charging demand data, and then these difference distances calculated based on the two norms are used as the clustering object difference items corresponding to each dimension in the clustering object difference data set, which can intuitively reflect the differences in the data of each dimension of the charging demand data of different electric vehicles.

[0096] In one embodiment, Figure 4 As shown, the corresponding dynamically adjustable parameters of each dimension data are calculated based on the cluster object difference data set, including:

[0097] Step S402: Fusion the difference items of the cluster objects to obtain difference fusion items.

[0098] Specifically, the cluster object difference items of each dimension corresponding to the cluster object difference data set are added together to obtain a difference fusion item.

[0099] For example, the clustering object difference dataset N = {||t a,i -t a,j ||2,||t d,i -t d,j ||2,||SOC a,i -SOC a,j ||2,||SOC desire,i -SOC desire,j ||2}, where N is the cluster object difference dataset, and ||t a,i -t a,j ||2 is the cluster object difference item in the cluster object difference dataset N corresponding to electric vehicle i and electric vehicle j in the electric vehicle arrival time dimension; ||t d,i -t d,j||2 is the cluster object difference item in the cluster object difference dataset N corresponding to electric vehicle i and electric vehicle j in the dimension of the future departure time of electric vehicles; ||SOC a,i -SOC a,j ||2 is the cluster object difference item in the vehicle power dimension when the electric vehicle arrives in the cluster object difference dataset N corresponding to electric vehicle i and electric vehicle j; ||SOC desire,i -SOC desire,j ||2 is the cluster object difference item in the cluster object difference dataset N corresponding to electric vehicle i and electric vehicle j in the dimension of user expected charging amount. The processing server converts ||t a,i -t a,j ||2 and ||t d,i -t d,j ||2 and ||SOC a,i -SOC a,j ||2 and ||SOC desire,i -SOC desire,j ||2 is summed to obtain the difference fusion term S, where the expression of S is S=||t a,i -t a,j ||2+||t d,i -t d,j ||2+||SOC a,i -SOC a,j ||2+||SOC desire,i -SOC desire,j ||2;

[0100] Step S404 : performing proportional calculations on the clustering object difference items and the difference fusion items corresponding to each dimension of data, respectively, to obtain the corresponding dynamically adjustable parameters of each dimension of data.

[0101] Specifically, the processing server uses the ratio of the clustering object difference item and the difference fusion item corresponding to each dimension data as the value of each corresponding dynamically adjustable parameter.

[0102] For example, the difference fusion term S = || t a,i -t a,j ||2+||t d,i -t d,j ||2+||SOC a,i -SOC a,j ||2+||SOC desire,i -SOC desire,j ||2, where ||t a,i -t a,j||2 is the cluster object difference item in the cluster object difference dataset N corresponding to electric vehicle i and electric vehicle j in the electric vehicle arrival time dimension, and its corresponding dynamically adjustable parameter is v1; ||t d,i -t d,j ||2 is the cluster object difference item in the cluster object difference dataset N corresponding to electric vehicle i and electric vehicle j in the dimension of the future departure time of electric vehicles, and its corresponding dynamically adjustable parameter is v2; ||SOC a,i -SOC a,j ||2 is the cluster object difference item in the vehicle power dimension when the electric vehicle arrives in the cluster object difference dataset N corresponding to electric vehicle i and electric vehicle j, and its corresponding dynamically adjustable parameter is v3; ||SOC desire,i -SOC desire,j ||2 is the cluster object difference item in the cluster object difference dataset N corresponding to electric vehicle i and electric vehicle j in the dimension of user expected charging amount. Its corresponding dynamic adjustable parameter is v4. Then the expression of v1, v2, v3, v4 is: v1=||t a,i -t a,j ||2 / S;v2=||t d,i -t d,j ||2 / S;v3=||SOC a,i -SOC a,j ||2 / S;v4=||SOC desire,i -SOC desire,j ||2 / S.

[0103] In this embodiment, the ratio of the clustering object difference item and the difference fusion item corresponding to each dimension data is used as the value of each corresponding dynamic adjustable parameter. This construction method makes it possible to construct the dynamic adjustable parameters so that when the electric vehicle time feature (t a ,t d ) is more different, according to the above formula of v1, v2, v3, v4, the weights of v1 and v2 are greater, so the points with large time differences in the cluster objects will repel each other and will not be clustered into one category. a ,SOC desire ) is larger, according to the above formula, the weights of v3 and v4 will be larger, so the points with large differences in power in the cluster object will repel each other and will not be clustered into one category. Therefore, compared with the fixed static parameter setting, this dynamic adjustable parameter setting can be used to dynamically adjust the weights according to the specific situation when clustering electric vehicles. a ,t d ,SOC a ,SOC desireThe samples with large differences in various items are further separated by setting weights so that they will not be clustered into one category. Finally, the points with similar time characteristics and charge characteristics are clustered into one category. The electric vehicle cluster obtained by this clustering method is closer to a stable power load. Therefore, the setting of this dynamic adjustable parameter makes the subsequent clustering algorithm with this dynamic adjustable parameter more flexible and targeted for the time characteristics and power characteristics of electric vehicle charging behavior.

[0104] In one embodiment, Figure 5 As shown, the cluster object difference items in the cluster object difference data set are fused based on the dynamically adjustable parameters to obtain fused object difference items corresponding to each cluster object difference data set, including:

[0105] Step S502: Obtain clustering difference factors and parameter factors.

[0106] The cluster difference factor includes the cluster object difference item corresponding to each dimension data, and the parameter factor includes the dynamic adjustable parameter corresponding to each dimension data.

[0107] Step S504 : performing weighted fusion on the cluster difference factors and the parameter factors to obtain fused object difference items corresponding to each cluster object difference data set.

[0108] Among them, each cluster object difference data set corresponds to a fusion object difference item. The fusion object difference item directly reflects the difference in the charging demand data of the two electric vehicles and is also an element that constitutes the subsequent electric vehicle data distance matrix.

[0109] Specifically, after obtaining the cluster difference factor and the parameter factor, the processing server weights the dimensional data of the cluster difference factor and the corresponding dynamically adjustable parameter and then sums them to generate the corresponding fusion object difference item.

[0110] For example, when calculating the fusion object difference item between electric vehicle i and electric vehicle j, the processing server clusters the object difference data set N = {||t a,i -t a,j ||2,||t d,i -t d,j ||2,||SOC a,i -SOC a,j ||2,||SOC desire,i -SOC desire,j The difference items of cluster objects in each dimension in ||2} are weighted and summed with the corresponding v1, v2, v3, and v4, that is: v1 and ||t a,i -t a,j ||2 corresponds to, where v1=||t a,i -t a,j||2 / S; v2 and ||t d,i -t d,j ||2 corresponds to, where v2=||t d,i -t d,j ||2 / S; v3 and ||SOC a,i -SOC a,j ||2 corresponds to, where v3=||SOC a,i -SOC a,j ||2 / S; v4 and ||SOC desire,i -SOC desire,j ||2 corresponds to, where v4=||SOC desire,i -SOC desire,j ||2 / S; where S is the difference fusion term, S=||t a,i -t a,j ||2+||t d,i -t d,j ||2+||SOC a,i -SOC a,j ||2+||SOC desire,i -SOC desire,j ||2; then the fusion object difference term D(i,j) between electric vehicle i and electric vehicle j = v1·||t a,i -t a,j ||2+v2·||t d,i -t d,j ||2+v3·||SOC a,i -SOC a,j ||2+v4·||SOC desire,i -SOC desire,j ||2.

[0111] In this embodiment, the difference factor and the parameter factor are weighted and summed to obtain the fused object difference item corresponding to each cluster object difference data set, and each difference factor is weighted and summed with its corresponding parameter factor. This fusion method enables the fused object difference item to be able to adjust the weights of each part of the cluster object difference item in a targeted manner according to the time characteristics and power characteristics of the electric vehicle charging demand data based on the dynamically adjustable parameters. The result obtained can better reflect the time characteristics and power characteristics of the electric vehicle charging behavior, thereby achieving the effect of fully utilizing the electric vehicle resources.

[0112] In one embodiment, Figure 6 As shown, based on the electric vehicle data distance matrix corresponding to the distributed charging stations, the electric vehicles corresponding to each distributed charging station are clustered to obtain the electric vehicle charging cluster groups corresponding to each distributed charging station, including:

[0113] Step S602 : normalizing the difference items of each fusion object in the electric vehicle data distance matrix to obtain adjacent difference items.

[0114] Among them, normalization is a way to simplify calculations, that is, to transform a dimensioned expression into a dimensionless expression and become a scalar. Here, the numerical values ​​of the difference items of each fusion object in the electric vehicle data distance matrix are mapped to the range of [0,1].

[0115] Specifically, the processing server normalizes the difference items of each fusion object in the electric vehicle data distance matrix according to the following formula to obtain the adjacent difference item S(i, j). The specific formula is: Where ξ is the Gaussian-Kernel function coefficient, exp is the exponential function, and D(i, j) is the fusion object difference term in the electric vehicle data distance matrix. S(i, j)∈[0,1], the larger the value of S(i, j), the closer the physical characteristics between the two electric vehicles are.

[0116] Step S604: forming an adjacency matrix using the adjacency difference items.

[0117] Step S606: Calculate the eigenvalues ​​of the adjacency matrix to obtain eigenvectors corresponding to the eigenvalues.

[0118] Specifically, the processing server calculates the eigenvalues ​​and corresponding eigenvectors of the adjacency matrix.

[0119] Step S608: Arrange the eigenvectors in descending order of their corresponding eigenvalues ​​to form an eigenvector matrix.

[0120] Specifically, the processing server sorts the feature values ​​in descending order: λ1≥λ2≥…≥λ k And arrange the eigenvectors in descending order of their corresponding eigenvalues ​​into a new matrix U=[u1,u2,…,u k ].

[0121] Step S610 , clustering the row vectors of the eigenvector matrix using a clustering algorithm to generate the electric vehicle charging cluster group.

[0122] Specifically, the processing server clusters the row vectors of the matrix U using the K-means algorithm to obtain k electric vehicle clusters: C1, C2, ..., C k .

[0123] In this embodiment, the electric vehicle data distance matrix determined based on the aforementioned steps is normalized, and then a new eigenvector matrix composed of eigenvectors arranged in descending order according to their corresponding eigenvalues ​​is clustered using the K-means algorithm. The resulting multiple electric vehicle charging clusters are more similar in time characteristics and power characteristics, and have more similar physical characteristics, thereby improving the similarity in charging behavior of electric vehicles in the electric vehicle charging clusters.

[0124] In one embodiment, Figure 7 As shown in FIG, based on each electric vehicle charging cluster group and the electric vehicle scheduling potential model, the aggregation information corresponding to each electric vehicle charging cluster group is calculated, including:

[0125] Step S702: Obtain the electric vehicle power constraint.

[0126] Among them, the electric vehicle power constraints include the maximum power of the electric vehicle B·SOC max 、Electric vehicle original power B·SOC a , the upper bound of the scheduling potential at the previous moment The lower bound of the scheduling potential at the previous moment e t-1 、Electric vehicle minimum power B·SOC min 、The charging capacity E that the user expects to achieve desire , the power consumption of electric vehicles Electric vehicle charging capacity Upper bound of scheduling potential at the previous moment and the lower bound of the scheduling potential at the previous moment e t-1 The initial value is the original power of the electric vehicle B·SOC a ; The electric vehicle power constraint expression is: SOC min ≤SOC t ≤SOC max , SOC desired ≤SOC d , where τ is a flag variable indicating the current charging and discharging status of the electric vehicle, with a value range of 0 or 1, SOC t is the electric vehicle's charge at time t, SOC max and SOC min Respectively represent the upper and lower limits of electric vehicle power, SOC desired User's expected charge capacity, SOC d is the charge of the electric car when leaving, and are the charging and discharging power of the electric vehicle at time t, η c and ηd A is the charge and discharge efficiency, and B is the battery capacity.

[0127] Step S704 : obtaining an electric vehicle dispatching potential model based on the electric vehicle power constraint. The electric vehicle dispatching potential model includes an upper bound of dispatching potential and a lower bound of dispatching potential.

[0128] The upper bound of the dispatching potential is the minimum value among the upper bound constraints, which include the maximum power of the electric vehicle, the charging capacity of the electric vehicle, and the upper bound of the dispatching potential at the previous moment. The lower bound of the dispatching potential is the maximum value among the lower bound constraints, which include the lower bound of the dispatching potential at the previous moment minus the power consumption of the electric vehicle, the minimum power of the electric vehicle, and the charging capacity that the user expects to achieve minus the power consumption of the electric vehicle.

[0129] Specifically, the upper bound of the scheduling potential in is the upper bound of the scheduling potential at the previous moment, when t=t a hour, B j ·SOC max The maximum power of an electric car. It is the sum of the charging capacity of the electric vehicle and the upper bound of the scheduling potential at the previous moment; the lower bound of the scheduling potential in e t-1 is the lower bound of the scheduling potential at the previous moment. When t=t a hour, e t-1 =B·SOC a ;SOC min B j is the minimum charge of an electric vehicle. It is the amount of charging that the user expects to achieve minus the power consumption of the electric vehicle; It is the lower bound of the dispatch potential at the previous moment minus the power consumption of electric vehicles.

[0130] Step S706 , integrating the upper bounds of the scheduling potentials of the electric vehicles in each cluster in the electric vehicle charging cluster group to obtain the upper bounds of the scheduling potentials of each cluster in the electric vehicle charging cluster group.

[0131] Specifically, the processing server adds up the scheduling potential upper bounds of the electric vehicles in each cluster in the electric vehicle charging cluster group to obtain the scheduling potential upper bounds of each cluster in the electric vehicle charging cluster group.

[0132] For example, the upper bound of the dispatch potential of each cluster in the electric vehicle charging cluster is in Represents cluster Ci The upper bound of the aggregation energy of a single electric vehicle j in , C i represents the i-th cluster of electric vehicle cluster sample data.

[0133] Step S708 : A lower bound of the scheduling potential of the electric vehicles in each cluster in the electric vehicle charging cluster group is merged to obtain a lower bound of the scheduling potential of each cluster in the electric vehicle charging cluster group.

[0134] Specifically, the processing server adds up the lower bounds of the scheduling potentials of the electric vehicles in each cluster in the electric vehicle charging cluster group to obtain the lower bounds of the scheduling potentials of each cluster in the electric vehicle charging cluster group.

[0135] For example, the lower bound of the dispatch potential of each cluster in the electric vehicle charging cluster is in e j Represents cluster C i The lower bound of the aggregate energy of a single electric vehicle j in , C i represents the i-th cluster of electric vehicle cluster sample data.

[0136] Step S710 : The scheduling potential upper bound of each cluster in the electric vehicle charging cluster group is combined with the scheduling potential upper bound to form aggregation information corresponding to each electric vehicle charging cluster group.

[0137] The aggregate information is the electric vehicle cluster scheduling potential constraint composed of the scheduling potential upper bound of each cluster in the electric vehicle charging cluster group and the scheduling potential upper bound.

[0138] In this embodiment, the scheduling potential model of electric vehicles is constructed based on the power constraints of individual electric vehicles and the charging demand of electric vehicles. This model can reflect the changes in the scheduling potential constraints and charging demand of electric vehicles from the time they arrive at the charging station to the time they leave the charging station. For example, Figure 9 What is shown is the change of the constraint range of the dispatch potential of a single electric vehicle, such as Figure 10 The figure shows the changes in the dispatching potential constraint range of electric vehicle charging clusters. The clarification of the dispatching potential constraint range of electric vehicle charging clusters improves the accuracy of electric vehicle aggregation information in reflecting the load characteristics of electric vehicle charging clusters as a single stable charging load, enabling subsequent urban power grids to generate dispatching signals based on the aggregated information to more accurately respond to the impact caused by large-scale electric vehicles connecting to the power grid.

[0139] In one embodiment, Figure 8 As shown in FIG, a charging schedule for electric vehicles is generated based on the aggregated information, including:

[0140] Step S802: Obtain the charging power of the electric vehicle.

[0141] Step S804: Send the aggregated information to the city power grid to obtain a dispatch signal generated based on the aggregated information returned by the city power grid.

[0142] The dispatching signal includes the charging curves of all electric vehicles at each charging station.

[0143] Specifically, the urban power grid constructs the grid dispatch objective function based on the aggregated information. The grid dispatch objective function formula is: Where: and EA i,t are the active power output of the generator at node i at time t and the active power output of the charging station at time t, P Ai,t is the active power exchange between the local power grid and the wholesale market, α i and β i is the fuel cost coefficient of generator i, λ t is the predicted day-ahead market electricity price at time t, ρ t is the electricity price in the wholesale market at time t, V D is the node set of the urban power grid, T is the total number of scheduling plan moments; under the basic constraints of the power grid physical model, the minimum value of the power grid scheduling objective function is solved to solve the EA i,t The optimal solution is used as the scheduling signal, and the processing server solves the decision variables This problem is a convex problem and is solved by the commercial software Cplex. The basic constraints of the power grid physical model are:

[0144] Power balance constraints of the basic constraints of the power grid physical model:

[0145]

[0146]

[0147] Where: P i and Q i is the injected active power and injected reactive power of node i, is the renewable energy power generation output at node i, and is the active and reactive traditional loads of node i, Q Ai is the amount of reactive power exchanged from the wholesale market;

[0148] The linear power flow equations with basic constraints of the power grid physical model are:

[0149]

[0150]

[0151] Where: V i and θi is the voltage amplitude and phase angle at node i, L ij is the line connecting node i and node j, L is the set of power lines, P ij and Q ij Line L ij The active power flow and reactive power flow on the ij and x ij Line L ij Resistance and reactance; basic constraints of the physical model of the power grid and system security constraints:

[0152]

[0153]

[0154]

[0155]

[0156] Where: S l Indicates the line capacity, V i and are the lower and upper bounds of the voltage amplitude at node i, respectively, P ij and Line L ij The lower and upper bounds of the meritorious current, Q ij and Line L ij The lower and upper bounds of the reactive current, and are the lower and upper bounds of the generator’s active output, and are the lower and upper bounds of the generator reactive output, It is the upper bound of the renewable energy power generation power of node i.

[0157] Step S806: The dispatching signal is integrated with the charging power of the electric vehicle to obtain a signal decomposition objective function.

[0158] Specifically, the processing server constructs the decomposition process objective function based on the integration of the dispatch signal sent back from the city power grid and the charging power of the electric vehicle. The specific construction method is shown in the following formula:

[0159]

[0160] in, is the numerical difference between the actual charging power and the dispatch curve. is the power change between two moments of all vehicles, that is, reducing the charging power fluctuation between consecutive moments, M is the coefficient used to adjust the accuracy of signal decomposition, is the charging power of electric vehicle j at time t, is the dispatch signal, is the number of aggregated vehicles.

[0161] Step S808 , calculating the minimum value of the signal decomposition objective function to obtain a charging scheduling plan for the electric vehicle, where the charging scheduling plan for the electric vehicle includes a charging curve of the electric vehicle.

[0162] Specifically, the processing server solves the minimum value of the decomposed objective function, which is a quadratic convex optimization problem with decision variables. Solved by the commercial software Cplex.

[0163] In this embodiment, a power grid dispatch objective function is constructed based on the aggregated information, and the minimum value of the power grid dispatch objective function is solved under the basic constraints of various power grid physical models to obtain a dispatch signal. Then, a decomposition process objective function is constructed based on the dispatch signal and the electric vehicle charging power, and the minimum value of the decomposition process objective function is solved again to obtain an electric vehicle charging dispatch plan. The charging dispatch plan is an electric vehicle charging curve, which reduces the difference between the actual charging power of the electric vehicle and the dispatch signal, and at the same time reduces the volatility of the electric vehicle charging power change between consecutive moments, thereby improving the stability of the electric vehicle charging power.

[0164] This application also provides an application scenario, which applies the above-mentioned power grid dispatching method, and the method is applied to the urban power grid dispatching scenario. Specifically, the application of the power grid dispatching method in this application scenario is as follows:

[0165] The computer equipment obtains the charging demand data of each electric vehicle in the electric vehicle charging station CR={t a ,t d ,SOC a ,SOC desire}, according to the charging demand data, the cluster object difference data set N = {||t a,i -t a,j ||2,||t d,i -t d,j ||2,||SOC a,i -SOC a,j ||2,||SOC desire,i -SOC desire,j||2}, where the cluster object difference dataset in each charging station is equal to the number of pairwise comparisons of electric vehicles in the corresponding charging station. Then, based on the cluster object difference dataset in each charging station, the corresponding number of dynamic adjustable parameter groups is obtained, and the dimension of the parameters in each group of dynamic adjustable parameters is equal to the dimension of the corresponding cluster object difference dataset. Based on the weighted fusion of the dynamic adjustable parameters and the cluster object difference dataset, the similarity between every two electric vehicles in the charging station is obtained, that is, the fusion object difference item. Then, the electric vehicle data distance matrix D of each charging station composed of the fusion object difference item is subjected to K-means clustering to generate electric vehicle charging cluster groups C1, C2,…, C k , according to the charging cluster group and electric vehicle scheduling potential model, the aggregate information is obtained. At this time, the aggregate information includes the upper and lower bounds of the scheduling potential of the electric vehicle charging cluster group of each charging station (this scheduling potential range stipulates the power constraint range in the subsequent electric vehicle charging scheduling plan). The computer device transmits the aggregate information to the urban power grid scheduling system through the network. The urban power grid scheduling system then incorporates the aggregate information into the basic constraint framework of the power grid system to generate a scheduling signal, which is then sent back to the computer device through the network. The computer device then decomposes and processes the scheduling signal to generate an electric vehicle charging scheduling plan, which is then sent back to each corresponding charging station. Each charging station then configures the power grid according to the electric vehicle charging scheduling plan, adjusts the generator output power, etc., to realize the charging of each electric vehicle.

[0166] In the above-mentioned grid dispatching method, a charging dispatching plan for each electric vehicle in each charging station can be automatically generated according to the charging demand of the electric vehicles at each charging station, which can fully utilize the time characteristics and power characteristics of electric vehicle resources. The charging dispatching plan is generated quickly and reasonably. Figure 11 As shown in Table 1, the above-mentioned grid dispatching method can effectively reduce the peak-valley difference of grid node load and save grid operation costs.

[0167] Table 1

[0168] Discharge method Peak-to-valley difference (MW) Grid operating costs ($) Orderly discharge (method proposed in this patent) 7.93 5645.4 Disorderly discharge (no charging planning method involved) 14.36 15325.65

[0169] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0170] In one embodiment, Figure 12 As shown, a power grid dispatching device is provided. The device can adopt a software module or a hardware module, or a combination of the two to form a part of a computer device. The device specifically includes: a charging demand data acquisition module 1202, an electric vehicle difference determination module 1204, a dynamic parameter determination module 1206, a distance matrix determination module 1208, an electric vehicle clustering module 1210, an aggregate information determination module 1212, and a signal processing module 1214, wherein:

[0171] The charging demand data acquisition module 1202 is used to acquire the electric vehicle charging demand data uploaded by each distributed charging station. The charging demand data includes at least the following multi-dimensional data: the arrival time of the electric vehicle, the future departure time of the electric vehicle, the battery level of the electric vehicle when it arrives, and the user's desired charging level;

[0172] The electric vehicle difference determination module 1204 is configured to perform cluster object difference calculation on the electric vehicle charging demand data corresponding to each distributed charging station, and obtain a cluster object difference data set corresponding to each electric vehicle in each distributed charging station, wherein the cluster object difference data set includes cluster object difference items corresponding to each dimensional data;

[0173] A dynamic parameter determination module 1206 is configured to calculate corresponding dynamic adjustable parameters of each dimension data based on the cluster object difference data set;

[0174] The distance matrix determination module 1208 is configured to fuse the cluster object difference items in the cluster object difference data set based on dynamically adjustable parameters to obtain fused object difference items corresponding to each cluster object difference data set. The fused object difference items corresponding to the same distributed charging station constitute the electric vehicle data distance matrix corresponding to the distributed charging station.

[0175] The electric vehicle clustering module 1210 is configured to cluster the electric vehicles corresponding to each distributed charging station based on the electric vehicle data distance matrix corresponding to the distributed charging station, and obtain the electric vehicle charging cluster groups corresponding to each distributed charging station;

[0176] Aggregation information determination module 1212, configured to calculate aggregation information corresponding to each electric vehicle charging cluster based on each electric vehicle charging cluster and the electric vehicle scheduling potential model;

[0177] The signal processing module 1214 is used to send the aggregated information to the city power grid and to send the electric vehicle charging scheduling plan to each electric vehicle charging station.

[0178] In one embodiment, the electric vehicle difference determination module 1204 is further used to calculate the difference distance corresponding to each dimension in the multi-dimensional data based on the charging demand data between electric vehicles in the same charging station, and obtain the cluster object difference items corresponding to each electric vehicle in each dimension; the cluster object difference items form a corresponding cluster object difference data set between each electric vehicle in the charging station.

[0179] In one embodiment, the dynamic parameter determination module 1206 is also used to fuse the difference items of each clustering object to obtain a difference fusion item; and respectively calculate the proportion of the corresponding clustering object difference item and the difference fusion item of each dimensional data to obtain the corresponding dynamic adjustable parameters of each dimensional data.

[0180] In one embodiment, the distance matrix determination module 1208 is also used to obtain clustering difference factors and parameter factors, where the clustering difference factors include the corresponding clustering object difference items of each dimensional data, and the parameter factors include the corresponding dynamic adjustable parameters of each dimensional data; the clustering difference factors and the parameter factors are weightedly fused to obtain the fused object difference items corresponding to each clustering object difference data set.

[0181] In one embodiment, the electric vehicle clustering module 1210 is further used to normalize the difference items of each fusion object in the electric vehicle data distance matrix to obtain adjacency difference items; form an adjacency matrix with the adjacency difference items; calculate the eigenvalues ​​of the adjacency matrix to obtain the eigenvectors corresponding to the eigenvalues; arrange the eigenvectors in descending order according to the corresponding eigenvalues ​​to form an eigenvector matrix; and cluster the row vectors of the eigenvector matrix using a clustering algorithm to generate electric vehicle charging cluster groups.

[0182] In one embodiment, the aggregate information determination module 1212 is further used to obtain the electric vehicle power constraints, which include the maximum power of the electric vehicle, the original power of the electric vehicle, the upper bound of the scheduling potential at the previous moment, the lower bound of the scheduling potential at the previous moment, the minimum power of the electric vehicle, the charging amount expected by the user, the power consumption of the electric vehicle, and the charging amount of the electric vehicle. The initial values ​​of the upper bound of the scheduling potential at the previous moment and the lower bound of the scheduling potential at the previous moment are both the original power of the electric vehicle; based on the electric vehicle power constraints, an electric vehicle scheduling potential model is obtained, which includes the upper bound of the scheduling potential and the lower bound of the scheduling potential. The upper bound of the scheduling potential takes the minimum value of the upper bound constraint items. The upper bound constraint items include the maximum power of the electric vehicle, the charging amount of the electric vehicle and the The upper bound of the scheduling potential at the previous moment is fused, and the lower bound of the scheduling potential takes the maximum value among the lower bound constraints. The lower bound constraints include the lower bound of the scheduling potential at the previous moment minus the power consumption of the electric vehicle, the minimum power of the electric vehicle, and the charging amount expected by the user minus the power consumption of the electric vehicle; the upper bounds of the scheduling potential of the electric vehicles of each cluster in the electric vehicle charging cluster group are fused to obtain the upper bound of the scheduling potential of each cluster in the electric vehicle charging cluster group; the lower bounds of the scheduling potential of the electric vehicles of each cluster in the electric vehicle charging cluster group are fused to obtain the lower bound of the scheduling potential of each cluster in the electric vehicle charging cluster group; the upper bound of the scheduling potential of each cluster in the electric vehicle charging cluster group is combined with the upper bound of the scheduling potential to form aggregated information corresponding to each electric vehicle charging cluster group.

[0183] In one embodiment, the signal processing module 1214 is also used to obtain the charging power of the electric vehicle; send the aggregated information to the urban power grid to obtain a scheduling signal generated based on the aggregated information and returned by the urban power grid; fuse the scheduling signal with the charging power of the electric vehicle to obtain a signal decomposition objective function; calculate the minimum value of the signal decomposition objective function to obtain a charging scheduling plan for the electric vehicle, and the charging scheduling plan for the electric vehicle includes a charging curve of the electric vehicle.

[0184] For the specific definition of the power grid dispatching device, please refer to the definition of the power grid dispatching method above, which will not be repeated here. The various modules in the above-mentioned power grid dispatching device can be implemented in whole or in part by software, hardware, or a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0185] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 13As shown. The computer device includes a processor, a memory, and a network interface connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store electric vehicle charging demand data, etc. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a power grid scheduling method is implemented.

[0186] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 14 As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a power grid scheduling method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad provided on the computer device housing, or an external keyboard, touchpad or mouse.

[0187] Those skilled in the art will understand that Figure 13 、 14 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0188] In one embodiment, a computer device is further provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0189] In one embodiment, a computer-readable storage medium is provided, storing a computer program, which implements the steps in the above-mentioned method embodiments when executed by a processor.

[0190] In one embodiment, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps of each of the above-described method embodiments.

[0191] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0192] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0193] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A power grid dispatching method, characterized in that: The method comprises: Obtaining electric vehicle charging demand data uploaded by each distributed charging station, wherein the charging demand data includes at least the following multi-dimensional data: the arrival time of the electric vehicle, the future departure time of the electric vehicle, the vehicle power level of the electric vehicle upon arrival, and the user's desired charging level; Performing cluster object difference calculation on the electric vehicle charging demand data corresponding to each distributed charging station to obtain a cluster object difference data set corresponding to each electric vehicle in each distributed charging station, wherein the cluster object difference data set includes cluster object difference items corresponding to each dimensional data; The difference items of each cluster object are fused to obtain a difference fusion item; the difference items of the cluster objects corresponding to each dimension data are proportionally calculated with the difference fusion item to obtain the corresponding dynamic adjustable parameters of each dimension data; Obtaining clustering difference factors and parameter factors; performing weighted fusion on the clustering difference factors and the parameter factors to obtain fused object difference items corresponding to each clustering object difference data set; forming an electric vehicle data distance matrix corresponding to the distributed charging station with the fused object difference items corresponding to the same distributed charging station; clustering the electric vehicles corresponding to each distributed charging station based on the electric vehicle data distance matrix corresponding to the distributed charging station to obtain electric vehicle charging cluster groups corresponding to each distributed charging station; Based on each electric vehicle charging cluster group and the electric vehicle scheduling potential model, the aggregation information corresponding to each electric vehicle charging cluster group is calculated; Generating a charging scheduling plan for an electric vehicle based on the aggregated information includes: sending the aggregated information to a city power grid, obtaining a scheduling signal returned by the city power grid, and decomposing the scheduling signal according to the electric vehicle charging power constraint to generate a charging scheduling plan for a single electric vehicle.

2. The method according to claim 1, characterized in that The clustering object difference calculation is performed on the electric vehicle charging demand data corresponding to each distributed charging station to obtain a clustering object difference data set corresponding to each electric vehicle in each distributed charging station, including: Based on the charging demand data of electric vehicles in the same charging station, respectively calculating the difference distance corresponding to each dimension in the multi-dimensional data, and obtaining the cluster object difference item corresponding to each electric vehicle in each dimension; The cluster object difference items form a cluster object difference data set corresponding to each electric vehicle in the charging station.

3. The method according to claim 2, characterized in that The method of calculating the difference distance corresponding to each dimension in the multi-dimensional data based on the charging demand data between electric vehicles in the same charging station to obtain the cluster object difference items corresponding to each dimension between the electric vehicles includes: The bi-norm calculation is performed on the data of each dimension of the charging demand data of two electric vehicles in the same charging station, and the calculation results are used as the clustering object difference items corresponding to the data of each dimension in the clustering object difference dataset.

4. The method according to claim 1, wherein The cluster difference factor includes the cluster object difference item corresponding to each dimensional data; the parameter factor includes the dynamic adjustable parameter corresponding to each dimensional data.

5. The method according to claim 1, wherein The electric vehicles corresponding to each distributed charging station are clustered based on the electric vehicle data distance matrix corresponding to the distributed charging station to obtain the electric vehicle charging cluster group corresponding to each distributed charging station, including: Normalize the difference items of each fusion object in the electric vehicle data distance matrix to obtain the adjacent difference items; An adjacency matrix is ​​formed by the adjacency difference items; Calculating the eigenvalues ​​of the adjacency matrix to obtain eigenvectors corresponding to the eigenvalues; Arrange the eigenvectors in descending order of their corresponding eigenvalues ​​to form an eigenvector matrix; The row vectors of the eigenvector matrix are clustered using a clustering algorithm to generate the electric vehicle charging cluster group.

6. The method according to claim 1, characterized in that The aggregation information corresponding to each electric vehicle charging cluster group is obtained by calculating the aggregation information corresponding to each electric vehicle charging cluster group based on the electric vehicle scheduling potential model, including: Get the electric vehicle power constraint; The electric vehicle power constraints include the maximum power of the electric vehicle, the original power of the electric vehicle, the upper bound of the scheduling potential at the previous moment, the lower bound of the scheduling potential at the previous moment, the minimum power of the electric vehicle, the charging capacity expected by the user, the power consumption of the electric vehicle, and the charging capacity of the electric vehicle. The initial values ​​of the upper bound of the scheduling potential at the previous moment and the lower bound of the scheduling potential at the previous moment are both the original power of the electric vehicle. Obtaining an electric vehicle scheduling potential model based on the electric vehicle power constraint, wherein the electric vehicle scheduling potential model includes an upper bound of scheduling potential and a lower bound of scheduling potential; The upper bound of the scheduling potential is the minimum value of the upper bound constraint items, and the upper bound constraint items include the fusion of the maximum power of the electric vehicle, the charging capacity of the electric vehicle and the upper bound of the scheduling potential at the previous moment; The lower bound of the scheduling potential is the maximum value among the lower bound constraints, and the lower bound constraints include the lower bound of the scheduling potential at the previous moment minus the power consumption of the electric vehicle, the minimum power of the electric vehicle, and the charging amount expected by the user minus the power consumption of the electric vehicle; fusing the upper bounds of the scheduling potentials of the electric vehicles of each cluster in the electric vehicle charging cluster group to obtain the upper bounds of the scheduling potentials of each cluster in the electric vehicle charging cluster group; fusing the lower bounds of the scheduling potentials of the electric vehicles of each cluster in the electric vehicle charging cluster group to obtain the lower bounds of the scheduling potentials of each cluster in the electric vehicle charging cluster group; The scheduling potential upper bound of each cluster in the electric vehicle charging cluster group is combined with the scheduling potential upper bound to form aggregation information corresponding to each electric vehicle charging cluster group.

7. The method according to claim 1, characterized in that Generating a charging schedule for electric vehicles based on the aggregated information includes: Get the charging power of electric vehicles; Sending the aggregated information to the city power grid to obtain a dispatch signal generated based on the aggregated information returned by the city power grid; Fusing the dispatch signal with the charging power of the electric vehicle to obtain a signal decomposition objective function; The minimum value of the signal decomposition objective function is calculated to obtain a charging scheduling plan for the electric vehicle, wherein the charging scheduling plan for the electric vehicle includes a charging curve of the electric vehicle.

8. A power grid dispatching device, characterized in that: The device comprises: A charging demand data acquisition module is used to acquire the electric vehicle charging demand data uploaded by each distributed charging station. The charging demand data includes at least the following multi-dimensional data: the arrival time of the electric vehicle, the future departure time of the electric vehicle, the vehicle power level of the electric vehicle when it arrives, and the user's desired charging level; An electric vehicle difference determination module is used to perform cluster object difference calculation on the electric vehicle charging demand data corresponding to each distributed charging station, and obtain a cluster object difference data set corresponding to each electric vehicle in each distributed charging station, wherein the cluster object difference data set includes cluster object difference items corresponding to each dimensional data; A dynamic parameter determination module is used to fuse the difference items of each cluster object to obtain a difference fusion item; respectively calculate the ratio of the cluster object difference item corresponding to each dimension data and the difference fusion item to obtain the corresponding dynamic adjustable parameter of each dimension data; The distance matrix determination module is used to obtain cluster difference factors and parameter factors; the cluster difference factors and parameter factors are weightedly fused to obtain the fused object difference items corresponding to each cluster object difference data set; the fused object difference items corresponding to the same distributed charging station form the electric vehicle data distance matrix corresponding to the distributed charging station; An electric vehicle clustering module is used to cluster the electric vehicles corresponding to each distributed charging station based on the electric vehicle data distance matrix corresponding to the distributed charging station to obtain the electric vehicle charging cluster group corresponding to each distributed charging station; An aggregation information determination module, configured to calculate the aggregation information corresponding to each electric vehicle charging cluster based on each electric vehicle charging cluster and an electric vehicle scheduling potential model; A signal processing module is used to send the aggregated information to the urban power grid and to send the electric vehicle charging scheduling plan to each electric vehicle charging station, including: sending the aggregated information to the urban power grid, obtaining a scheduling signal returned by the urban power grid, and decomposing the scheduling signal according to the electric vehicle charging power constraint to generate a charging scheduling plan for a single electric vehicle.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images