Electric Vehicle Charging / Discharging Spatiotemporal Graph Model Based on Graph Database and Its Applications

Abstract

The present invention discloses an electric vehicle charging / discharging spatio-temporal graph model based on a graph database and its application. It utilizes the charging / discharging power, charging / discharging time, and charging / discharging location of electric vehicles accessing the power grid collected by charging piles, and constructs an electric vehicle charging / discharging spatio-temporal graph model according to the requirements of the graph model. The charging / discharging spatio-temporal graph model is used to store the historical data and real-time data of the charging / discharging of electric vehicles, provide the spatio-temporal distribution of the schedulable capacity of electric vehicles at multiple time scales, realize the efficient storage, query, and analysis of large-scale electric vehicle cluster data, greatly improve the data management and analysis efficiency, and provide data support for large-scale electric vehicle clusters to participate in power auxiliary services in different scenarios of the power grid.

Description

Technical Field

[0001] The present invention relates to the technical field of power system load demand side response, and particularly relates to a method for constructing an electric vehicle charging / discharging spatio-temporal graph model based on a graph database and a method for dynamically analyzing schedulable capacity. Background Art

[0002] To solve energy and environmental problems, new energy electric vehicles that are environmentally friendly and pollution-free have developed rapidly. Electric vehicles have the characteristics of distributed energy storage. An electric vehicle cluster connected to the grid can be used as a networked energy storage system to provide various auxiliary services to the grid, such as peak shaving, frequency modulation, and economic dispatching. Predicting the schedulable capacity of electric vehicles is a prerequisite for participating in grid dispatching and providing auxiliary services. Therefore, it is necessary to dynamically predict the spatio-temporal distribution of the schedulable capacity of electric vehicles at multiple time scales using the charging / discharging data of electric vehicles.

[0003] Charging piles can be used to collect data such as the charging / discharging power and time of electric vehicles accessing the grid. By collecting and analyzing the charging / discharging data of electric vehicles in real time, the analysis of the change in the schedulable capacity of the electric vehicle cluster can be carried out. However, when a large number of electric vehicles access the grid for charging / discharging, a large amount of data will be generated. Using traditional relational databases to store and process the charging / discharging data of electric vehicles and dynamically analyze the spatio-temporal network of the schedulable capacity of electric vehicles has low efficiency and cannot meet the requirements of various power auxiliary services for dynamically analyzing the schedulable capacity of a large number of electric vehicles at multiple time scales. Summary of the Invention

[0004] To avoid the deficiencies of the above-mentioned prior art, the present invention provides an electric vehicle charging / discharging spatio-temporal graph model based on a graph database and its application, so as to achieve efficient storage and processing of the charging / discharging data of electric vehicles and meet the requirements for schedulable capacity data at multiple time scales when electric vehicles participate in grid dispatching and auxiliary services.

[0005] The present invention adopts the following technical solutions to achieve the invention purpose:

[0006] The characteristics of the electric vehicle charging / discharging spatio-temporal graph model based on the graph database of the present invention are:

[0007] Define each vertex of the charging / discharging spatio-temporal graph according to the requirements of the graph database and the electric vehicle charging / discharging spatio-temporal graph;

[0008] Define the edges between the vertices according to the requirements of the graph database and the electric vehicle charging / discharging spatio-temporal graph;

[0009] The attributes of each vertex and each edge are defined according to the requirements of the graph database and the electric vehicle charging / discharging spatiotemporal graph and the charging / discharging data of the electric vehicle; the charging / discharging data includes the charging / discharging power, charging / discharging time and charging / discharging position of the electric vehicle connected to the power grid.

[0010] The electric vehicle charging / discharging spatiotemporal graph model based on the graph database of the present invention is also characterized by:

[0011] The vertices of the charge / discharge time-space diagram include:

[0012] The time vertices at each level defined in the graph database, including the time vertices at the year, month, day, and hour levels, as well as the minute vertices, enable the electric vehicle charging / discharging spatiotemporal graph model to provide the charging / discharging time distribution information of electric vehicles;

[0013] The location vertices at each level defined in the graph database, including the location vertices at each level of the region and district, as well as the charging station vertices and charging pile vertices, enable the electric vehicle charging / discharging spatiotemporal graph model to provide the charging / discharging spatial distribution information of electric vehicles;

[0014] The electric vehicle vertices defined in the graph database enable the electric vehicle charging / discharging spatiotemporal graph model to provide electric vehicle information;

[0015] The electric vehicle charging / discharging spatiotemporal graph model based on the graph database of the present invention is also characterized by:

[0016] The edges between the vertices include:

[0017] The edges between time vertices are determined based on the relationship between each time vertex;

[0018] The edges between position vertices are determined based on the relationship between the position vertices;

[0019] an edge between an electric vehicle vertex and a time vertex determined based on the relationship between the electric vehicle vertex and the time vertex;

[0020] an edge between an electric vehicle vertex and a position vertex determined based on the relationship between the electric vehicle vertex and the position vertex;

[0021] The edge between the time vertex and the location vertex is determined based on the relationship between the time vertex and the location vertex.

[0022] The electric vehicle charging / discharging spatiotemporal graph model based on the graph database of the present invention is also characterized by:

[0023] The vertex attributes and edge attributes include:

[0024] The electric vehicle vertex has electric vehicle information as its attribute;

[0025] The time vertex takes the charging / discharging time of the electric vehicle as an attribute;

[0026] The location vertex takes the charging / discharging location of the electric vehicle as an attribute;

[0027] Determine the edge attribute between the electric vehicle vertex and the time vertex according to the relationship between the electric vehicle information and the charging / discharging location;

[0028] Determine the edge attribute between the time vertex and the location vertex according to the relationship between the charging / discharging time and the charging / discharging location.

[0029] The characteristics of the method for dynamically analyzing the schedulable capacity of electric vehicles in the present invention are:

[0030] The schedulable capacity is the load capacity that can be scheduled, including the schedulable charging capacity and the schedulable discharging capacity. Using the above-mentioned spatio-temporal graph model of electric vehicle charging / discharging based on the graph database, dynamically analyze the schedulable capacity of electric vehicles according to the following steps:

[0031] Step 1: Data import

[0032] Import the charging / discharging data of the electric vehicle into the spatio-temporal graph model of electric vehicle charging / discharging;

[0033] Establish a mapping relationship between the charging / discharging data of the electric vehicle and the vertex attributes and edge attributes in the spatio-temporal graph model of electric vehicle charging / discharging, and run the data loading program to import the charging / discharging data into the spatio-temporal graph model of electric vehicle charging / discharging according to the mapping relationship to obtain the spatio-temporal distribution map of electric vehicle charging / discharging;

[0034] Step 2: Perform one-minute schedulable capacity analysis according to the following process

[0035] Step 2.1: Obtain the target minute vertex

[0036] Take the latest minute vertex obtained by traversing all minute vertices or the set minute value as the target minute vertex a

[0037] Step 2.2: Parallelly calculate the one-minute schedulable capacity of the electric vehicle

[0038] Query the spatio-temporal distribution map of electric vehicle charging / discharging according to the target minute vertex a, and construct all electric vehicle vertices connected to the target minute vertex a into an electric vehicle vertex set B;

[0039] Obtain the charging / discharging power P of each electric vehicle in the electric vehicle vertex set B by traversing the electric vehicle vertex set B, and according to the grid scheduling time t d Obtain the schedulable capacity SC of the electric vehicle EV , that is, SC EV =P×t d, then the one-minute dispatchable capacity of each electric vehicle in the electric vehicle vertex set B is obtained through parallel computing;

[0040] Step 2.3: Parallel calculation of the one-minute dispatchable capacity of the charging / discharging piles

[0041] Query the charge / discharge spatiotemporal distribution graph according to the electric vehicle vertex set B, and construct all charging pile vertices in the position vertex that are connected to each electric vehicle vertex in the electric vehicle vertex set B into a charging pile vertex set C;

[0042] Taking the schedulable capacity of the electric vehicle as the schedulable capacity of the corresponding charging pile, traversing the charging pile vertex set C and the electric vehicle vertex set B, and obtaining the one-minute schedulable capacity of each charging pile in the charging pile vertex set C in parallel, and storing the one-minute schedulable capacity of each charging pile on the edge connecting the corresponding charging pile vertex and the minute vertex;

[0043] Step 2.4: Parallel calculation of the one-minute dispatchable capacity of the charging station

[0044] Query the charging spatiotemporal distribution graph according to the charging pile vertex set C, and construct all charging station vertices in the position vertex that are connected to each charging / discharging pile vertex in the charging pile vertex set C into a charging station vertex set D;

[0045] The sum of the schedulable capacities of all charging pile vertices connected to the charging station vertex d in the charging station vertex set D is used as the schedulable capacity of the charging station vertex d; the one-minute schedulable capacity of each charging station vertex is obtained by traversing the charging station vertex set D and the charging pile vertex set C in parallel; the one-minute schedulable capacity of each charging station vertex is stored on the edge connecting each charging station vertex and the minute vertex a;

[0046] Step 2.5: Parallel calculation of the one-minute dispatchable capacity of the area

[0047] Query the charging spatiotemporal distribution graph according to the charging station vertex set D, and construct all the area vertices in the position vertex that are connected to each charging station vertex in the charging station vertex set D into an area vertex set E;

[0048] The sum of the schedulable capacities of all charging station vertices connected to the area vertex e in the area vertex set E is used as the schedulable capacity of the area vertex e; the one-minute schedulable capacity of each area vertex is obtained by traversing the area vertex set D and the charging station vertex set C in parallel; the one-minute schedulable capacity of each area vertex is stored on the edge connecting each area vertex and the minute vertex a;

[0049] Step 2.6: Parallel Compute the One-Minute Dispatched Capacity of the Region

[0050] Query the charging / discharging spatio-temporal distribution map according to the area vertex set E of the said area, and construct a regional vertex set F by taking all the regional vertices connected to each area vertex in the area vertex set E among the position vertices;

[0051] Take the sum of the schedulable capacities of all the charging station vertices connected to the regional vertex f in the regional vertex set F as the schedulable capacity of the regional vertex f; by traversing the regional vertex set F and the area vertex set E, calculate the one-minute schedulable capacity of each regional vertex in parallel; store the one-minute schedulable capacity of each regional vertex on the edge connecting each regional vertex and the minute vertex a; thus obtain the one-minute schedulable capacity of each hierarchical position vertex;

[0052] Step 3: Conduct the schedulable capacity analysis within 1 hour according to the following method

[0053] Calculate the one-hour schedulable capacity of each hierarchical position vertex based on the one-minute schedulable capacity of each hierarchical position vertex obtained in Step 2, and realize the schedulable capacity analysis within 1 hour.

[0054] The feature of the method for dynamically analyzing the schedulable capacity of electric vehicles in the present invention also lies in: in the said Step 3, according to the one-minute schedulable capacity of each hierarchical position vertex obtained in Step 2, calculate the average value of the schedulable capacity of each charging pile, each charging station, each area and each region every 15 minutes, and take the average value as the schedulable capacity of each minute within the corresponding 15 minutes. Then, within 1 hour, the schedulable capacity is divided into 4 average values of schedulable charging capacity and 4 average values of schedulable discharging capacity according to the 15-minute time period. Store the 8 average values of schedulable capacity on the edges connecting each position vertex and the corresponding hour vertex, and realize the schedulable capacity analysis within 1 hour.

[0055] The feature of the method for dynamically analyzing the schedulable capacity of electric vehicles in the present invention also lies in: by real-time loading the charging / discharging data of electric vehicles into the graph model, calculate the schedulable capacity of each hierarchical position vertex for one day, one week and one month according to Step 2 and Step 3.

[0056] Compared with the prior art, the beneficial effects of the present invention are reflected in:

[0057] 1. Based on the graph database and the charging / discharging data of electric vehicles, define each vertex, edge and attribute in the graph database, construct a spatio-temporal graph model of the charging / discharging of electric vehicles, and dynamically analyze the change of the schedulable capacity of electric vehicles based on the spatio-temporal graph, realizing the efficient storage and processing of the charging / discharging data of electric vehicles, and meeting the requirements of the schedulable capacity data of different time scales when electric vehicles participate in power grid dispatching and ancillary services;

[0058] 2. The present invention analyzes the one-minute dispatchable capacity at different locations based on the spatio-temporal distribution map of electric vehicle charging / discharging. The one-minute dispatchable capacity can provide data support for the dispatch control of electric vehicles participating in auxiliary services such as frequency modulation and voltage regulation. If the real-time charging / discharging data of electric vehicles is loaded, the dispatchable capacity of the latest minute can be used as the predicted value of the dispatchable capacity of the next minute.

[0059] 3. The present invention analyzes the change trend of the dispatchable capacity at different time scales and different locations based on the spatio-temporal distribution map of electric vehicle charging / discharging. The change of the dispatchable capacity within a relatively long time such as one hour or one day can be used to analyze the charging / discharging rules of electric vehicles.

[0060] 4. The present invention uses a graph database to store the charging / discharging data of electric vehicles. According to the spatio-temporal distribution map after importing the data, the change trend of the spatio-temporal distribution of the dispatchable capacity of charging / discharging electric vehicles at different times and different locations can be intuitively displayed. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a flowchart of the method for analyzing the one-minute dispatchable capacity of electric vehicles in the present invention;

[0062] Figure 2 It is a spatio-temporal distribution map of the charging / discharging of some electric vehicles in the present invention;

[0063] Figure 3 It is a spatio-temporal distribution map of the dispatchable capacity at some hierarchical positions in a certain hour in the present invention; DETAILED DESCRIPTION OF THE EMBODIMENTS

[0064] In this embodiment, the spatio-temporal graph model of electric vehicle charging / discharging based on the graph database defines each vertex and the edges between vertices of the spatio-temporal graph of charging / discharging according to the requirements of the graph database and the spatio-temporal graph of electric vehicle charging / discharging; and defines the attributes of each vertex and each edge according to the charging / discharging data of electric vehicles and the requirements of the spatio-temporal graph. The charging / discharging data includes the charging / discharging power, charging / discharging time, and charging / discharging location of electric vehicles accessing the power grid.

[0065] In specific implementation, the corresponding measures also include:

[0066] Each vertex of the spatio-temporal graph of charging / discharging includes:

[0067] The EV charging / discharging spatiotemporal graph model needs to provide the temporal and spatial distribution of EV charging / discharging. Therefore, time and location vertices are defined in the graph database. To meet the requirements of grid ancillary services at different time scales, time vertices are defined for year, month, day, hour, and minute, while location vertices are defined for region and area, as well as charging station and charging pile vertices. Defining multiple levels of time and location vertices facilitates data queries at different scales. The electric vehicle vertices defined in the graph database enable the EV charging / discharging spatiotemporal graph model to provide EV information.

[0068] The edges between vertices include:

[0069] The edges between time vertices, determined by the relationships between each time vertex, are connected in sequence by year, month, day, hour, and minute vertices, representing the relationships between times at different granularities. The edges between location vertices, determined by the relationships between each location vertex, are connected in sequence by edges between region, district, charging station, and charging pile vertices, representing the relationships between locations at different levels. The edges between electric vehicle vertices and time and location vertices, determined by the relationships between electric vehicle vertices and time and location vertices, are connected by edges between electric vehicle vertices and time and location vertices, respectively. The electric vehicle vertex represents the location and time of electric vehicle charging and discharging by connecting to both the time and location vertices. The edges between time and location vertices, determined by the relationships between time and location vertices, are connected by edges between location vertices at each level and time vertices at each level, representing the changes in dispatchable capacity at each location at different times.

[0070] Vertex attributes and edge attributes include:

[0071] The electric vehicle vertex takes electric vehicle information as attributes, including charging / discharging power, battery capacity, charging / discharging status, and the operation order number of charging, and uses the operation order number as the unique identifier; the time vertex takes the charging / discharging time of the electric vehicle as attributes, and each level of time vertex takes the specific time value as attributes, where the year vertex takes the year value as the unique identifier, the month vertex takes the year-month value as the unique identifier, the day vertex takes the year-month-day value as the unique identifier, the hour vertex takes the year-month-day-hour value as the unique identifier, and the minute vertex takes the year-month-day-hour-minute value as the unique identifier; the location vertex takes the charging / discharging location of the electric vehicle as attributes, and each level of location vertex takes its number as the unique identifier; the edge attributes between the electric vehicle vertex and the time vertex are determined according to the relationship between the electric vehicle information and the charging / discharging time, including charging / discharging power, remaining time, and schedulable capacitance, indicating the above attributes of the electric vehicle change with time; the edge attributes between the time vertex and the location vertex are determined according to the relationship between the charging / discharging time and the charging / discharging location, and each edge is set with a schedulable capacitance attribute to store the schedulable capacitance at different times of each level of location.

[0072] Use the electric vehicle charging / discharging spatio-temporal graph model based on the graph database in this embodiment to dynamically analyze the schedulable capacitance of electric vehicles, where the schedulable capacitance is the load capacitance that can be scheduled, including schedulable charging capacitance and schedulable discharging capacitance. The specific analysis steps are as Figure 1 shown:

[0073] Step 1: Data import

[0074] Import the charging / discharging data of electric vehicles into the electric vehicle charging / discharging spatio-temporal graph model: Establish a mapping relationship between the charging / discharging data of electric vehicles and the vertex attributes and edge attributes in the charging / discharging spatio-temporal graph model, and run the data loading program to import the charging / discharging data into the charging / discharging spatio-temporal graph model according to the mapping relationship to obtain the charging / discharging spatio-temporal distribution map of electric vehicles. As Figure 2 shown is the partial charging / discharging spatio-temporal distribution map of electric vehicles for a certain minute. It can be seen that the vertex of type EV represents an electric vehicle, the vertex of type pi represents a charging pile, the vertex of type s represents a charging station, the vertex of type c represents a district, the vertex of type pr represents a region, and the vertex of type m represents a minute.

[0075] Step 2: Perform one-minute schedulable capacitance analysis according to the following process:

[0076] Step 2.1: Obtain the target minute vertex:

[0077] By traversing all minute vertices and reading their time values, the latest minute vertex obtained, or the set minute vertex obtained by traversing all minute vertices according to the set minute value, is used as the target minute vertex a.

[0078] Step 2.2: Parallel calculation of the one-minute dispatchable capacity of electric vehicles:

[0079] Query the minute-electric vehicle path of the charging / discharging spatiotemporal distribution graph according to the target minute vertex a, that is, query the electric vehicles that are charging / discharging in the target minute, obtain all electric vehicle vertices connected to the target minute vertex a, and construct them into the electric vehicle vertex set B; the dispatchable capacity of electric vehicles SC EV is the charging / discharging power P of the electric vehicle and the grid dispatching period t d The product of SC EV =P×t d When the electric vehicle is charging, the dispatchable charging capacity can be obtained, and the dispatchable discharging capacity is 0. When the electric vehicle is discharging, the dispatchable discharging capacity can be obtained, and the dispatchable charging capacity is 0. The charging / discharging power of each electric vehicle in the electric vehicle vertex set B is obtained by traversing the electric vehicle vertex set B through the minute-electric vehicle path, and the one-minute dispatchable capacity of each electric vehicle in the electric vehicle vertex set B can be obtained by parallel calculation according to the grid dispatch time, and the one-minute dispatchable capacity of each electric vehicle is stored in the dispatchable capacity attribute of the edge connecting the corresponding electric vehicle vertex and the target minute vertex a.

[0080] Step 2.3: Parallel calculation of the one-minute dispatchable capacity of the charging pile:

[0081] The electric vehicle-charging pile path in the charge / discharge spatiotemporal distribution graph is queried based on the electric vehicle vertex set B. Specifically, the charging / discharging locations of each electric vehicle are queried to obtain all charging pile vertices connected to each electric vehicle vertex in the electric vehicle vertex set B, and these are constructed into the charging pile vertex set C. For each charging pile connected to one electric vehicle, the schedulable capacity of the electric vehicle is the schedulable capacity of the charging pile. The charging pile vertex set C and the electric vehicle vertex set B are traversed through the electric vehicle-charging pile path, and the one-minute schedulable capacity of the electric vehicle connected to each charging pile is read in parallel to obtain the one-minute schedulable capacity of each charging pile in the charging pile vertex set C. The one-minute schedulable capacity of each charging pile is stored in the schedulable capacity attribute of the edge connecting the corresponding charging pile vertex and the target minute vertex a.

[0082] Step 2.4: Parallel calculation of the one-minute dispatchable capacity of the charging station:

[0083] Query the charging pile - charging pile path of the charging / discharging spatio - temporal distribution map according to the charging pile vertex set C, that is, query the charging stations to which each charging pile belongs, obtain all the charging station vertices connected to each charging pile vertex in the charging pile vertex set C, and construct them into the charging station vertex set D; the schedulable capacity of a charging station is the sum of the schedulable capacities of all the charging piles in the charging station, that is, the schedulable charging capacity of the charging station is the sum of the schedulable charging capacities of all the charging piles in the charging station, and the schedulable discharging capacity is the sum of the schedulable discharging capacities of all the charging piles in the charging station. Traverse the charging station vertex set D and the charging pile vertex set C through the charging pile - charging station path, read the schedulable capacity of each charging pile, and calculate the sum of the one - minute schedulable capacities of all the charging piles in each charging station in parallel to obtain the one - minute schedulable capacity of each charging station vertex; store the one - minute schedulable capacity of each charging station vertex in the schedulable capacity attribute of the edge connecting each charging station vertex to the target minute vertex a.

[0084] Step 2.5. Calculate the one - minute schedulable capacity of the area in parallel:

[0085] Query the charging station - area path of the charging / discharging spatio - temporal distribution map according to the charging station vertex set D, that is, query the areas to which each charging station belongs, obtain all the area vertices connected to each charging station vertex in the charging station vertex set D, and construct them into the area vertex set E; the schedulable capacity of an area is the sum of the schedulable capacities of all the charging stations in the area, that is, the schedulable charging capacity of the area is the sum of the schedulable charging capacities of all the charging stations in the area, and the schedulable discharging capacity is the sum of the schedulable discharging capacities of all the charging stations in the area. Traverse the area vertex set D and the charging station vertex set C, read the schedulable capacity of each charging station, and calculate the sum of the one - minute schedulable capacities of all the charging stations in each area in parallel to obtain the one - minute schedulable capacity of each area vertex; store the one - minute schedulable capacity of each area vertex in the schedulable capacity attribute of the edge connecting each area vertex to the target minute vertex a.

[0086] Step 2.6. Calculate the one - minute schedulable capacity of the region in parallel:

[0087] Query the area-region path of the charge / discharge spatiotemporal distribution graph based on the area vertex set E. Specifically, query the region to which each area belongs, obtain all regional vertices connected to each area vertex in the area vertex set E, and construct a regional vertex set F. The dispatchable capacity of a region is the sum of the dispatchable capacities of all areas within the region. Specifically, the dispatchable charging capacity of a region is the sum of the dispatchable charging capacities of all areas within the region, and the dispatchable discharge capacity is the sum of the dispatchable discharge capacities of all areas within the region. Traverse the regional vertex set F and the area vertex set E through the area-region path, read the dispatchable capacity of each area, and calculate the one-minute dispatchable capacity of all areas within each region in parallel to obtain the one-minute dispatchable capacity of each regional vertex. Store the one-minute dispatchable capacity of each regional vertex in the dispatchable capacity attribute of the edge connecting each regional vertex to the minute vertex a. This yields the one-minute dispatchable capacity of vertices at each level.

[0088] Step 3: Perform dispatchable capacity analysis within 1 hour as follows:

[0089] The analysis of the dispatchable capacity of each level location within one hour can be converted into the analysis of the dispatchable capacity every minute within one hour. According to step 2, the one-minute dispatchable capacity of each level location vertex every minute within one hour is obtained. However, each level location has 60 one-minute dispatchable capacities, and the dispatchable capacity changes every minute is small. Therefore, the average dispatchable capacity of each charging pile, each charging station, each area and each region every 15 minutes can be calculated based on the dispatchable capacity every minute, and the average value is used as the dispatchable capacity of each minute within the corresponding 15 minutes. Then, each level location can obtain 4 average dispatchable charging capacities and 4 average dispatchable discharging capacities within one hour, so as to analyze the changing trend of the dispatchable capacity of each level location within one hour, realize the dispatchable capacity analysis within one hour, and store the 8 average dispatchable capacities in the dispatchable capacity attribute of the edge connecting each location vertex with the corresponding hour vertex.

[0090] In the specific implementation, by loading the charging / discharging data of electric vehicles into the graph model in real time, the dispatchable capacity of the position vertices at each level for one day, one week, and one month is calculated according to steps 2 and 3.

[0091] Figure 3 To calculate some results of the dispatchable capacity within a certain hour in a certain region, the figure shows the dispatchable capacity of each charging station, each area, and each region within one hour. In the figure, SCC represents the dispatchable charging capacity. As described in step 3, the connecting edge between each location vertex and the hour vertex stores four dispatchable charging capacity values within one hour. The results can be used to analyze the changing trend of the dispatchable charging capacity of each location within one hour.

[0092] Table 1 shows the time required for the dynamic analysis of the schedulable capacity of electric vehicles using the TigeGraph graph database and MySQL respectively under different data scales.

[0093] Table 1

[0094]

[0095] As can be seen from Table 1, whether it is the data scale of one minute or one hour, the time required for graph database calculation is much less than that required for the relational database MySQL. Therefore, the present invention stores charge / discharge data of electric vehicles and analyzes schedulable capacity based on graph data technology, which can better meet the needs of schedulable capacity analysis of electric vehicles at multiple time scales.

Claims

1. An electric vehicle charging / discharging spatio-temporal graph model based on a graph database, characterized in that: According to the requirements of the graph database and the spatio-temporal graph of electric vehicle charging / discharging, define each vertex of the spatio-temporal graph of charging / discharging in the graph database; according to the requirements of the graph database and the spatio-temporal graph of electric vehicle charging / discharging, define the edges between the above-mentioned vertices in the graph database; according to the requirements of the graph database and the spatio-temporal graph of electric vehicle charging / discharging, as well as the charging / discharging data of electric vehicles, define each vertex attribute and each edge attribute; the charging / discharging data includes the charging / discharging power, charging / discharging time, and charging / discharging location of the electric vehicle accessing the power grid. Each vertex of the spatio-temporal graph of charging / discharging includes: each hierarchical time vertex defined in the graph database, including each hierarchical time vertex of year, month, day, and hour, as well as the minute vertex, so that the spatio-temporal graph model of electric vehicle charging / discharging can provide the charging / discharging time distribution information of electric vehicles; each hierarchical location vertex defined in the graph database, including each hierarchical location vertex of region and area, as well as the charging station vertex and charging pile vertex, so that the spatio-temporal graph model of electric vehicle charging / discharging can provide the charging / discharging space distribution information of electric vehicles; the electric vehicle vertex defined in the graph database, so that the spatio-temporal graph model of electric vehicle charging / discharging can provide electric vehicle information; the edges between the above-mentioned vertices include: the edges between time vertices determined according to the relationship between each time vertex; the edges between location vertices determined according to the relationship between each location vertex; the edges between the electric vehicle vertex and the time vertex determined according to the relationship between the electric vehicle vertex and the time vertex; the edges between the electric vehicle vertex and the location vertex determined according to the relationship between the electric vehicle vertex and the location vertex; the edges between the time vertex and the location vertex determined according to the relationship between the time vertex and the location vertex.

2. The spatio-temporal graph model of electric vehicle charging / discharging based on the graph database according to claim 1, wherein: Each vertex attribute and each edge attribute include: the electric vehicle vertex takes the electric vehicle information as the attribute; the time vertex takes the charging / discharging time of the electric vehicle as the attribute; the location vertex takes the charging / discharging location of the electric vehicle as the attribute; the edge attribute of the edge between the electric vehicle vertex and the time vertex is determined according to the connection between the electric vehicle information and the charging / discharging location; the edge attribute of the edge between the time vertex and the location vertex is determined according to the relationship between the charging / discharging time and the charging / discharging location.

3. A method for dynamically analyzing the dispatchable capacity of an electric vehicle, characterized in that: The schedulable capacity is the load capacity that can be scheduled, including the schedulable charging capacity and the schedulable discharging capacity; using the spatio-temporal graph model of electric vehicle charging / discharging based on the graph database described in claim 2, dynamically analyze the schedulable capacity of electric vehicles according to the following steps: Step 1: Data import: Import the charging / discharging data of the electric vehicle into the spatio-temporal graph model of electric vehicle charging / discharging; establish a mapping relationship between the charging / discharging data of the electric vehicle and the vertex attributes and edge attributes in the spatio-temporal graph model, and run the data loading program to import the charging / discharging data into the spatio-temporal graph model according to the mapping relationship to obtain the spatio-temporal distribution map of electric vehicle charging / discharging. Step 2: Conduct one-minute schedulable capacity analysis according to the following process Step 2.1: Obtain the target minute vertex: Take the latest minute vertex obtained by traversing all minute vertices or the set minute value as the target minute vertex a. Step 2.

2. Parallelly calculate the one-minute dispatchable capacity of electric vehicles: Query the charge / discharge spatio-temporal distribution map according to the target minute vertex a, and construct all electric vehicle vertices connected to the target minute vertex a into an electric vehicle vertex set B; Obtain the charge / discharge power of each electric vehicle in the electric vehicle vertex set B by traversing the electric vehicle vertex set B, and according to the grid dispatch time t d obtain the dispatchable capacity of the electric vehicle SC EV , that is SC EV =P× t d , then parallelly calculate the one-minute dispatchable capacity of each electric vehicle in the electric vehicle vertex set B; Step 2.

3. Parallelly calculate the one-minute dispatchable capacity of charging / discharging piles: Query the charging / discharging spatio-temporal distribution map according to the set B of electric vehicle vertices. Construct the set C of charging pile vertices by taking all the charging pile vertices connected to each electric vehicle vertex in the set B of electric vehicle vertices among the location vertices. Take the dispatchable capacity of the electric vehicle as the dispatchable capacity of the corresponding charging pile. By traversing the set C of charging pile vertices and the set B of electric vehicle vertices, parallelly obtain the one-minute dispatchable capacity of each charging pile in the set C of charging pile vertices, and store the one-minute dispatchable capacity of each charging pile on the edge connecting the corresponding charging pile vertex and the minute vertex. Step 2.

4. Parallelly calculate the one-minute dispatchable capacity of the charging station: Query the charging spatio-temporal distribution map according to the set C of charging pile vertices. Construct the set D of charging station vertices by taking all the charging station vertices connected to each charging / discharging pile vertex in the set C of charging pile vertices among the location vertices. Take the sum of the dispatchable capacities of all the charging pile vertices connected to the charging station vertex d in the set D of charging station vertices as the dispatchable capacity of the charging station vertex d. By traversing the set D of charging station vertices and the set C of charging pile vertices, parallelly calculate and obtain the one-minute dispatchable capacity of each charging station vertex. Store the one-minute dispatchable capacity of each charging station vertex on the edge connecting each charging station vertex and the minute vertex a. Step 2.

5. Parallelly calculate the one-minute dispatchable capacity of the area: Query the charging spatio-temporal distribution map according to the set D of charging station vertices. Construct the set E of area vertices by taking all the area vertices connected to each charging station vertex in the set D of charging station vertices among the location vertices. Take the sum of the dispatchable capacities of all the charging station vertices connected to the area vertex e in the set E of area vertices as the dispatchable capacity of the area vertex e. By traversing the set D of charging station vertices and the set C of charging pile vertices, parallelly calculate and obtain the one-minute dispatchable capacity of each area vertex. Store the one-minute dispatchable capacity of each area vertex on the edge connecting each area vertex and the minute vertex a. Step 2.

6. Parallelly calculate the one-minute dispatchable capacity of the region: Query the charging / discharging spatio-temporal distribution map according to the set E of area vertices. Construct the set F of region vertices by taking all the region vertices connected to each area vertex in the set E of area vertices among the location vertices. Take the sum of the dispatchable capacities of all the charging station vertices connected to the region vertex f in the set F of region vertices as the dispatchable capacity of the region vertex f. By traversing the set F of region vertices and the set E of area vertices, parallelly calculate and obtain the one-minute dispatchable capacity of each region vertex. Store the one-minute dispatchable capacity of each region vertex on the edge connecting each region vertex and the minute vertex a. Thus, obtain the one-minute dispatchable capacity of each hierarchical location vertex. Step 3: Perform the dispatchable capacity analysis within 1 hour according to the following method: Calculate the one-hour dispatchable capacity of each hierarchical location vertex based on the one-minute dispatchable capacity of each hierarchical location vertex obtained in Step 2, and realize the dispatchable capacity analysis within 1 hour.

4. The dynamic analysis method for the dispatchable capacity of an electric vehicle according to claim 3, characterized in that: at In step 3, according to the one-minute dispatchable capacity of the position vertices at each level obtained in step 2, calculate the average value of the dispatchable capacity of each charging pile, each charging station, each area, and each region every 15 minutes. Take the average value as the dispatchable capacity of each minute within the corresponding 15 minutes. Then, the dispatchable capacity within 1 hour is divided into 4 dispatchable charging capacity average values and 4 dispatchable discharging capacity average values according to 15-minute time periods. Store the 8 dispatchable capacity average values on the edges connecting each position vertex to the corresponding hourly vertex to realize the analysis of the dispatchable capacity within 1 hour.

5. The method for dynamically analyzing the dispatchable capacity of an electric vehicle according to claim 3, characterized in that: By real-time loading the charging / discharging data of electric vehicles into the graph model, calculate the dispatchable capacity of each level of position vertices for one day, one week, and one month according to steps 2 and 3.

Images