Electric vehicle charging load prediction method, system and device and storage medium

Abstract

The invention discloses an electric vehicle charging load prediction method, system and device and a storage medium, and the method comprises the steps: constructing a traffic network topological graph according to a graph theory and a trip chain theory, building a traffic characteristic model of an electric vehicle based on the traffic network topological graph, building an energy consumption model of the electric vehicle in consideration of traffic road conditions and vehicle driving data, a charging characteristic model is established by considering queuing time and charging time of a charging station and driving energy consumption of the electric vehicle, a trip chain utility function is established based on a traffic characteristic model, an energy consumption model and the charging characteristic model, and space-time distribution of charging load of the electric vehicle in a traffic balance state of the system is solved. According to the method, the real-time interaction among the electric vehicle users in the complex traffic network is considered, the spatial and temporal distribution of the charging load of the electric vehicle can be accurately predicted, the interaction among the electric vehicle users in the road network is considered, the user travel path selection is ensured to accord with the actual traffic condition, and the limitation of a traditional shortest path algorithm is avoided.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric vehicle charging load prediction, and in particular relates to an electric vehicle charging load prediction method, system, device and storage medium. Background Art

[0002] With the rapid growth of electric vehicles (EVs) worldwide, the impact of EV charging load on power and transportation systems is becoming increasingly significant. By the end of 2022, the global EV fleet was expected to reach 18 million. The widespread adoption of EVs has significantly altered the operational state of distribution and transportation networks. However, traditional EV charging load forecasting methods largely ignore the real-time interactions of EVs within road networks, resulting in inaccurate forecasts. This is particularly evident in several aspects: First, traditional methods are primarily based on static or semi-dynamic traffic models, which fail to accurately reflect the real-time interactions of EVs within road networks. Second, existing models often rely on a single factor (such as route planning or charging station selection) and fail to comprehensively consider the impact of multiple factors, including traffic conditions, energy consumption characteristics, and charging behavior. Third, existing methods struggle to cope with real-time changes in complex transportation networks, making them unable to provide accurate load forecasting support for power systems. Currently, research on EV charging load forecasting is increasingly focusing on the traffic attributes of EVs, but many challenges remain. For example, while existing methods consider the constraints imposed by the road network on electric vehicle travel, they fail to fully account for the interactive nature of route selection among electric vehicle users. Furthermore, existing traffic equilibrium models primarily apply to long-term and medium- to long-term timescales and cannot accurately describe the real-time interactions of electric vehicles within the road network. Therefore, a charging load forecasting method that comprehensively considers the real-time interactions among electric vehicle groups is urgently needed to improve forecast accuracy and provide a scientific basis for the coordinated optimization of power-transportation systems. Summary of the Invention

[0003] To solve the above problems, the present invention provides an electric vehicle charging load prediction method, system, device and storage medium to address the problems that existing electric vehicle charging load prediction methods do not fully consider the path selection interactivity between electric vehicle users, and existing traffic equilibrium models cannot accurately describe the real-time interactive behavior of electric vehicles in the road network.

[0004] A method for predicting charging load of an electric vehicle, comprising:

[0005] Construct a transportation network topology based on graph theory and trip chain theory;

[0006] Establish a traffic characteristic model of electric vehicles based on the traffic network topology;

[0007] Establish an energy consumption model for electric vehicles by considering traffic conditions and vehicle driving data;

[0008] The charging characteristic model of electric vehicles is established by considering the queuing time at the charging station, charging time and the driving energy consumption of electric vehicles;

[0009] Based on the traffic characteristic model, energy consumption model and charging characteristic model, the travel chain utility function is established, and the spatiotemporal distribution of electric vehicle charging load under the traffic equilibrium state is solved.

[0010] According to a specific embodiment of the present invention, constructing a transportation network topology graph based on graph theory and trip chain theory includes:

[0011] Obtain the daily travel data of electric vehicles in the forecasted area and construct an electric vehicle travel chain based on the travel data, where the travel data includes activity type, activity location, and departure time;

[0012] Graph theory is used to simplify the electric vehicle travel chain into a transportation network topology graph consisting of edges and vertices.

[0013] According to a specific embodiment of the present invention, the variables of the electric vehicle travel chain include the time of arrival at the activity location m Time of leaving the activity location m The time when the route c where the charging is decided and the time when leaving route c when going to the activity location m+1 Arrival time at fast charging station k The moment you leave the fast charging station The set of road segments that constitute the trip m path The set of road segments that constitute the travel path from road segment c to charging station k and the set of road segments that constitute the travel path from charging station k to activity location m+1

[0014] According to a specific embodiment of the present invention, establishing a traffic characteristic model of an electric vehicle based on a traffic network topology graph includes:

[0015] The mathematical model of upstream and downstream queues is defined based on the traffic network topology and motion wave theory as follows:

[0016] UQ(t)=N in (t)-N out (t+δ-l ij / w)

[0017] DQ(t)=N in (t+δ-l ij / v)-N out (t)

[0018] Among them, UQ(t) is the upstream queue, DQ(t) is the downstream queue, l ijis the length of the road section, δ is the simulation time step, w is the propagation speed of the congestion wave from downstream to upstream, v is the zero flow speed, N in (t) and N out (t) are the number of vehicles entering the queue and the number of vehicles leaving the queue as of time t; N out (t+δ-l ij / w) indicates the time point (t+δ-l ij / w) the number of vehicles flowing out of the queue, N in (t+δ-l ij / v) indicates the time point (t+δ-l ij / v) the number of vehicles entering the queue;

[0019] The motion wave model in the discrete time simulation state is defined based on the mathematical model of the upstream and downstream queues:

[0020]

[0021] Among them, R(t) and S(t) are the number of vehicles received and sent by the road section in the time interval t to t+δ, respectively. is the maximum density of the road section, is the maximum traffic flow of the road section, λ is the traffic scaling factor, and L is the length of the road section;

[0022] Based on the motion wave model, the entry and exit times of all vehicles passing through the road section within 15 minutes are calculated, and the travel time tr of all road sections in the road network is determined. ij :

[0023]

[0024] Where, and are the time when the i-th vehicle enters and leaves the road section, N v The number of vehicles passing the road section within 15 minutes;

[0025] According to the travel time of the road section ij and travel demand, and use Dijkstra's shortest path algorithm to plan the set of sections with the shortest travel time between activity locations

[0026] According to a specific embodiment of the present invention, establishing an energy consumption model for an electric vehicle by taking into account traffic conditions and vehicle driving data includes:

[0027] Considering real-time traffic conditions, an electric vehicle energy consumption model is established based on the electric vehicle's speed, air resistance, and rolling friction:

[0028]

[0029] in,

[0030] v ij =l ij / tr ij

[0031]

[0032] in, For an electric car with a constant speed v ij The energy consumed by passing the flat road section (i, j), v ij is the driving speed, is the power supplied to the motor, is the air resistance, is the rolling friction, η is the efficiency of the motor, ρ is the air density, C d is the air resistance coefficient, A is the area of the front area of the electric vehicle, m is the mass of the electric vehicle, C r is the rolling friction coefficient, and g is the acceleration due to gravity.

[0033] According to a specific embodiment of the present invention, establishing a charging characteristic model of an electric vehicle by taking into account the queuing time at the charging station, the charging time, and the driving energy consumption of the electric vehicle includes:

[0034] Establish a charging station queuing model based on the queuing time and charging time of the charging station:

[0035]

[0036] in,

[0037]

[0038] in, is the queuing time required for charging station k, is the charging time required for charging station k, The remaining power of the electric vehicle when it arrives at the charging station. P is the remaining power of the electric vehicle when it decides to detour to charge. fc is the rated charging power of the charging station, is the set of road segments that constitute the travel path from road segment m to charging station c, E b is the rated capacity of the battery of the electric vehicle; For an electric car with a constant speed v ij The energy consumed by passing the flat road section (i, j);

[0039] Assume Q w ={t w1 ,t w2 L}, Q c={t c1 ,t c2 L} are the time sets of the vehicles queuing and charging in the charging station, respectively. The time set of all vehicles in the charging station that have completed charging is Q t =Q w UQ c , Q t Generate the set T={t1,t2L t n}, then the queuing time required for electric vehicles in charging station k is for:

[0040]

[0041] b k =N k -G k

[0042] Among them, b k is the actual queue number at charging station k, N k is the number of all vehicles at charging station k, G k is the number of chargers in charging station k;

[0043] When the remaining power of the electric vehicle meets When , the electric vehicle charging characteristic model is established with the minimum cost as the goal:

[0044] Select the objective function F of the charging station obj for:

[0045]

[0046] in, The real-time remaining power of electric vehicles during travel; E min is the charging threshold for electric vehicle battery protection, C k The cost of choosing different charging stations for electric vehicles, α q is the time cost of electric vehicle users corresponding to different average annual household incomes, and β is the unit price of charging.

[0047] According to a specific embodiment of the present invention, a travel chain utility function is established based on a traffic characteristic model, an energy consumption model, and a charging characteristic model, and the spatiotemporal distribution of the electric vehicle charging load of the system under a traffic equilibrium state is solved. The method includes:

[0048] Establish a daily travel chain utility function for a single electric vehicle user:

[0049]

[0050] Among them, U q,s is the daily travel utility of a single electric vehicle in the sth iteration, is the utility of the electric vehicle’s mth trip in the sth iteration, β tra is the direct marginal utility of activity travel time, is the travel time from the location of activity m to the location of activity m+1 in the sth iteration;

[0051] Based on the daily travel chain utility function of a single electric vehicle user, the average travel chain utility of all electric vehicles in the sth iteration is determined as:

[0052]

[0053] Where, is the average value of the travel chain utility of all electric vehicle users in the sth iteration, and NV is the total number of electric vehicle users participating in the simulation;

[0054] The successive averaging method is used to iteratively solve the spatiotemporal distribution of electric vehicle charging load in the system under traffic equilibrium state:

[0055]

[0056] Where, represents the expected travel time of link (i, j) at time t in the n+1th iteration, represents the expected travel time of link (i, j) at time t in the nth iteration, n represents the current iteration number, t∈[T h ,T h+1 ] represents the preset time interval of time t;

[0057] Based on the change in the average travel chain utility of all electric vehicles, it is determined whether the system meets the traffic equilibrium convergence condition. If so, the system reaches a traffic equilibrium state, the iteration is terminated, and the optimal electric vehicle travel and charging strategy is obtained;

[0058] The traffic equilibrium convergence condition is:

[0059]

[0060] Where NS is the number of simulations, ε DUE For simulation accuracy.

[0061] An electric vehicle charging load prediction system, comprising:

[0062] Topology creation module, used to construct transportation network topology based on graph theory and trip chain theory;

[0063] Traffic characteristic model building module, used to build a traffic characteristic model of electric vehicles based on the traffic network topology map;

[0064] Energy consumption model building module, used to build an energy consumption model for electric vehicles taking into account traffic conditions and vehicle driving data;

[0065] A charging characteristic model building module is used to build a charging characteristic model for electric vehicles by considering the queuing time at the charging station, the charging time, and the driving energy consumption of the electric vehicle;

[0066] The charging load prediction module is used to establish the travel chain utility function based on the traffic characteristic model, energy consumption model and charging characteristic model, and solve the spatiotemporal distribution of electric vehicle charging load under the traffic equilibrium state.

[0067] According to a specific embodiment of the present invention, the topology map creation module further includes:

[0068] The electric vehicle travel chain creation module is used to obtain the travel data of electric vehicles in the predicted area within a day and build an electric vehicle travel chain based on the travel data, where the travel data includes activity type, activity location and departure time;

[0069] The traffic network topology graph creation module is used to simplify the electric vehicle travel chain into a traffic network topology graph consisting of edges and vertices using graph theory.

[0070] According to a specific embodiment of the present invention, the variables of the electric vehicle travel chain include the time of arrival at the activity location m Time of leaving the activity location m The time when the route c where the charging is decided and the time when leaving route c when going to the activity location m+1 Arrival time at fast charging station k The moment you leave the fast charging station The set of road segments that constitute the trip m path The set of road segments that constitute the travel path from road segment c to charging station k and the set of road segments that constitute the travel path from charging station k to activity location m+1

[0071] According to a specific embodiment of the present invention, the traffic characteristics model building module further includes:

[0072] A motion wave model creation module, used to define mathematical models of the upstream queue and the downstream queue according to motion wave theory, and to define motion wave models in a discrete time simulation state based on the mathematical models of the upstream queue and the downstream queue;

[0073] The road section travel time calculation module is used to calculate the entry and exit times of all vehicles passing through the road section within 15 minutes based on the motion wave model, and determine the travel time of all road sections in the road network;

[0074] The shortest driving path calculation module is used to plan a set of paths with the shortest travel time between activity locations using the Dijkstra shortest path algorithm based on the road section travel time and travel demand.

[0075] According to a specific embodiment of the present invention, the charging characteristic model building module further includes:

[0076] A charging station queue model building module is used to build a charging station queue model based on the queue time and charging time of the charging station;

[0077] The electric vehicle charging characteristic model establishment module is used to establish the electric vehicle charging characteristic model according to the energy consumption of the electric vehicle.

[0078] According to a specific embodiment of the present invention, the charging characteristic model building module further includes:

[0079] The charging station determination module is used to determine the charging station based on the electric vehicle charging characteristic model.

[0080] According to a specific embodiment of the present invention, the charging load prediction module further includes:

[0081] The travel chain utility function creation module is used to establish the daily travel chain utility function of a single electric vehicle user;

[0082] An average trip chain utility function creation module is used to determine the average trip chain utility of all electric vehicles in the sth iteration based on the daily trip chain utility function of a single electric vehicle user;

[0083] The electric vehicle charging load spatiotemporal distribution calculation module is used to iteratively solve the spatiotemporal distribution of electric vehicle charging load when the system is in a traffic equilibrium state using the successive averaging method.

[0084] A computer device includes a processor and a memory. The memory stores a computer program, which is loaded and executed by the processor to implement the above-mentioned electric vehicle charging load prediction method.

[0085] A computer-readable storage medium stores a computer program, which is loaded and executed by a processor to implement the above-mentioned electric vehicle charging load prediction method.

[0086] Compared with the prior art, the electric vehicle charging load prediction method, system, device and storage medium provided by the present invention have the following advantages:

[0087] 1. By considering the real-time interactions between EV users in complex transportation networks, the spatiotemporal distribution of EV charging loads can be more accurately predicted, particularly in terms of real-time variations. This approach ensures that EV users' travel paths are tailored to actual traffic conditions, avoiding the limitations of traditional shortest path algorithms.

[0088] 2. The system can update traffic conditions, electric vehicle energy consumption and charging station selection in real time, provide real-time charging load forecast results, and provide real-time data support for the scheduling and optimization of the power system.

[0089] 3. The system is applicable to transportation networks of different sizes and complexities, and can provide a scientific basis for urban transportation planning and power system operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0091] Figure 1 The present invention provides a flow chart of a method for predicting charging load of an electric vehicle according to an embodiment of the present invention.

[0092] Figure 2 It is a flow chart of a method for constructing a traffic network topology map according to an embodiment of the present invention.

[0093] Figure 3 FIG. 4 is a flow chart of a method for establishing a traffic characteristic model according to an embodiment of the present invention.

[0094] Figure 4 The figure is a flow chart of a method for establishing an energy consumption model of an electric vehicle according to an embodiment of the present invention.

[0095] Figure 5 FIG. 4 is a flow chart of a method for establishing a charging characteristic model of an electric vehicle according to an embodiment of the present invention.

[0096] Figure 6 4 is a flow chart of a method for establishing a travel chain utility function according to an embodiment of the present invention.

[0097] Figure 7 1 is a schematic structural diagram of a daily travel chain of an electric vehicle according to an embodiment of the present invention.

[0098] Figure 8 2 is a schematic diagram of a queue-based motion wave model provided according to an embodiment of the present invention.

[0099] Figure 9 The figure is a schematic diagram of the structure of an electric vehicle charging load prediction system provided according to one embodiment of the present invention.

[0100] Figure 10 It is a structural diagram of a topology map creation module provided according to an embodiment of the present invention.

[0101] Figure 11 FIG. 4 is a structural diagram of a traffic characteristic model establishment module provided according to an embodiment of the present invention.

[0102] Figure 12 FIG. 4 is a structural diagram of a charging characteristic model establishment module according to an embodiment of the present invention.

[0103] Figure 13 4 is a structural diagram of a charging load prediction module provided according to an embodiment of the present invention.

[0104] Figure 14 The figure is a schematic diagram of the structure of a computer device according to an embodiment of the present invention.

[0105] Figure 15 Schematic diagram of road network topology and fast charging station distribution according to an embodiment of the present invention.

[0106] Reference numerals:

[0107] 01-Topology map creation module; 02-Traffic characteristics model establishment module; 03-Energy consumption model establishment module; 04-Charging characteristics model establishment module; 05-Charging load prediction module;

[0108] 011-Electric vehicle travel chain creation module; 012-Traffic network topology creation module;

[0109] 021-Motion wave model creation module; 022-Road section travel time calculation module; 023-Shortest driving path calculation module;

[0110] 041-Charging station queue model construction module; 042-Electric vehicle charging characteristic model establishment module; 043-Charging station determination module;

[0111] 051-Travel chain utility function creation module; 052-Average travel chain effect function creation module; 053-Electric vehicle charging load spatiotemporal distribution calculation module. DETAILED DESCRIPTION

[0112] In order to make those skilled in the art understand the concept and thought of the present invention more clearly, the present invention is described in detail below in conjunction with specific embodiment.It should be understood that the embodiment provided herein is only a part of all possible embodiments of the present invention.After reading the specification of the application, those skilled in the art have the ability to make improvements, transformations, or replacements to part or all of the following embodiments, and these improvements, transformations, or replacements are also included in the scope of protection claimed in the present invention.

[0113] In this document, the terms "advance", "entry" and other similar words are not intended to imply any order, quantity and importance, but are merely used to distinguish different elements. In this document, the terms "one", "an" and other similar words are not intended to indicate that there is only one thing, but rather that the relevant description is only for one of the things, and the thing may have one or more. In this document, the terms "comprise", "include" and other similar words are intended to indicate logical relationships, and cannot be regarded as indicating relationships in spatial structure. For example, "A includes B" is intended to indicate that B logically belongs to A, and does not mean that B is spatially located inside A. In addition, the meanings of the terms "comprise", "include" and other similar words should be regarded as open, not closed. For example, "A includes B" is intended to indicate that B belongs to A, but B does not necessarily constitute the whole of A, and A may also include other elements such as C, D, and E.

[0114] In this document, the terms "embodiment," "this embodiment," "one embodiment," and "an embodiment" do not indicate that the description applies only to a specific embodiment, but rather indicate that the description may also apply to one or more other embodiments. Those skilled in the art should understand that any description of a particular embodiment herein may be substituted, combined, or otherwise combined with the description of one or more other embodiments. New embodiments resulting from such substitution, combination, or other combination are readily conceivable by those skilled in the art and fall within the scope of protection of this invention.

[0115] Example 1

[0116] Additional aspects and advantages of embodiments of the present invention will be given in part in the following description and will become apparent from the following description or learned through practice of embodiments of the present invention. Figures 1-8 , an embodiment of the present invention provides a method for predicting charging load of an electric vehicle, comprising:

[0117] S1: Construct a transportation network topology diagram based on graph theory and trip chain theory.

[0118] S2: Establish a traffic characteristic model of electric vehicles based on the traffic network topology diagram.

[0119] S3: Establish an energy consumption model for electric vehicles considering traffic conditions and vehicle driving data.

[0120] S4: Establish a charging characteristic model of electric vehicles by considering the queuing time at the charging station, the charging time, and the driving energy consumption of electric vehicles.

[0121] S5: Establish a travel chain utility function based on the traffic characteristic model, energy consumption model, and charging characteristic model, and solve the spatiotemporal distribution of electric vehicle charging load under the traffic equilibrium state.

[0122] The present invention first uses graph theory to construct a traffic network topology map, and establishes an electric vehicle traffic characteristics model based on trip chain theory and a motion wave queuing model to study the spatiotemporal distribution of electric vehicles. Then, considering real-time traffic conditions, an electric vehicle energy consumption model is established based on factors such as the electric vehicle's driving speed, air resistance, and rolling friction to calculate the energy consumption of the electric vehicle during driving. Furthermore, a charging characteristics model is established by considering the charging station's queuing time, charging time, and the electric vehicle's driving energy consumption to determine the choice of charging station for the electric vehicle based on the electric vehicle's remaining power. Finally, considering the interaction between electric vehicle users, a trip chain utility function is established to determine the spatiotemporal distribution of the electric vehicle's charging load when the system is in a traffic equilibrium state. By combining traffic equilibrium theory, the electric vehicle trip chain model, the energy consumption model, and the charging characteristics model, the present invention can predict the spatiotemporal distribution of the real-time changing electric vehicle charging load in a complex traffic network, providing a scientific basis for the coordinated optimization operation of the power-transportation system.

[0123] Specifically, step S1 constructs a transportation network topology diagram based on graph theory and trip chain theory, including:

[0124] S11: Obtain the travel data of electric vehicles in the area to be predicted within one day, and build an electric vehicle travel chain based on the travel data, where the travel data includes activity type, activity location and departure time.

[0125] The trip chain theory is a theoretical framework for analyzing and understanding people's travel behavior. It is mainly used in traffic planning and urban planning. The trip chain is used to describe all the activities of travelers from leaving home to returning home, which are connected by trip chains to form a closed chain. The electric vehicle trip chain established in the embodiment of the present invention is as follows: Figure 7 As shown in the figure, the outer circle of the trip chain represents the time chain, which is used to describe the time when the electric vehicle arrives at or leaves the activity location. The inner circle of the trip chain represents the space chain, which is used to describe the spatial position relationship of the various activities that the electric vehicle goes to. Specifically, the variables of the electric vehicle trip chain include the time when the electric vehicle arrives at the activity location m Time of leaving the activity location m The time when the route c where the charging is decided and the time when leaving route c when going to the activity location m+1 Arrival time at fast charging station k The moment you leave the fast charging station The set of road segments that constitute the trip m path The set of road segments that constitute the travel path from road segment c to charging station k and the set of road segments that constitute the travel path from charging station k to activity location m+1 By leveraging these variables, the system can accurately capture the behavioral changes of electric vehicle users during their travels, providing fundamental data support for subsequent traffic network modeling and charging load forecasting. This paper describes the daily travel needs of electric vehicles by constructing an electric vehicle travel chain. Specific travel demand information includes activity type, location, and departure time, comprehensively reflecting users' daily travel behaviors.

[0126] S12: Use graph theory to simplify the electric vehicle travel chain into a transportation network topology graph consisting of edges and vertices.

[0127] Graph theory is a branch of mathematics that mainly studies the properties and applications of graphs. Graphs are mathematical structures composed of vertices and edges, used to describe the relationships between things. Vertices represent things, and edges represent the relationships between things. This paper uses graph theory to simplify the electric vehicle travel chain into a traffic network topology graph composed of edges (road sections) and vertices (intersections or nodes), such as Figure 8 As shown, Figure 8 Given a long and uniform road section, the sections are connected by nodes. Each section includes an upstream queue and a downstream queue. By defining the upstream queue and the downstream queue, the inflow and outflow of vehicles in the section can be described.

[0128] Specifically, step S2 of establishing a traffic characteristic model of electric vehicles based on the traffic network topology diagram includes:

[0129] S21: Based on the traffic network topology and motion wave theory, the mathematical model for upstream and downstream queues is defined as follows:

[0130] UQ(t)=N in (t)-N out (t+δ-l ij / w) (1)

[0131] w=l ij / δ (2)

[0132] DQ(t)=N in (t+δ-l ij / v)-N out (t) (3)

[0133] Among them, UQ(t) is the upstream queue, DQ(t) is the downstream queue, and li j is the length of the road section, δ is the simulation time step, w is the propagation speed of the congestion wave from downstream to upstream, v is the zero flow speed, N in (t) and N out (t) are the number of vehicles entering the queue and the number of vehicles leaving the queue as of time t, N out (t+δ-l ij / w) indicates the time point (t+δ-l ij / w), taking into account the time it takes for congestion information to propagate from downstream to upstream. ij / w reflects the propagation delay effect of queue changes, N in (t+δ-l ij / v) indicates the time point (t+δ-l ij / v) takes into account the time it takes for a vehicle to travel from the upstream to the downstream position l ij / v reflects the actual spatial propagation process of the vehicle.

[0134] Kinematic wave theory treats traffic flow as a continuous medium similar to a fluid, using wave propagation to describe vehicle movement and queuing. Based on kinematic wave theory and traffic network topology, this paper establishes mathematical models of upstream and downstream queues to simulate the propagation of vehicles between road sections. By defining upstream and downstream queues, this model describes the inflow and outflow of vehicles within a road section, as well as the changes in the number of vehicles in the upstream and downstream queues within the road section. Within each time step, the number of vehicles in the upstream queue is adjusted based on the changes in the number of incoming and outgoing vehicles. The number of incoming vehicles is related to the vehicle speed and road section length, while the number of outgoing vehicles is related to the inter-pass propagation speed. At time t, the upstream and downstream queues in the kinematic wave model are derived by inputting the number of vehicles in the upstream queue at the previous moment, the number of vehicles in the downstream queue at the previous moment, the number of vehicles in the current queue, and the number of vehicles out of the current queue. These models simulate the inflow and outflow of vehicles within the road section, thereby describing the propagation of traffic flow and the formation of congestion. Its upstream queue UQ(t) and downstream queue DQ(t) are determined by formula (1) and formula (3) respectively.

[0135] S22: Based on the mathematical models of the upstream and downstream queues, the motion wave model in the discrete time simulation state is defined as:

[0136]

[0137] Among them, R(t) and S(t) are the number of vehicles received and sent by the road section in the time interval t to t+δ, respectively. is the maximum density of the road section, is the maximum traffic flow of the road section, λ is the traffic scaling factor, which is determined by the scaling ratio of the samples in the study area, and L is the length of the road section.

[0138] Formula (4) and Formula (5) are traffic motion wave models under discrete time simulation state, which are used to describe the number of vehicles received and sent by the road section in each time step. In the time interval t, the number of vehicles that the road section can receive is limited by two factors: the number of vehicles attempting to flow in and the remaining capacity of the road section. Ultimately, the number of vehicles received by the road section is the smaller of the two values, ensuring that the road section does not exceed its maximum capacity. In the time interval t, the number of vehicles that the road section can send is limited by two factors: the number of vehicles in the downstream queue and the maximum traffic capacity of the road section. The number of vehicles sent by the road section is the smaller of the two values, ensuring that the road section does not exceed its maximum traffic capacity. In the embodiment of the present invention, the smaller value of the number of vehicles attempting to flow in and the remaining capacity of the road section, and the smaller value of the number of vehicles in the downstream queue and the maximum traffic capacity of the road section are selected to obtain the number of vehicles received and sent by the road section in a simulation step. The motion wave queuing model of Formulas (1)-(5) can be used to obtain the formation, propagation and dissipation process of traffic congestion. For example, when the traffic density exceeds a critical value, the compression wave will propagate backward, forming a traffic jam.

[0139] S23: Calculate the entry and exit times of all vehicles passing through the road section within 15 minutes based on the motion wave model, and determine the travel time tr of all road sections in the road network ij :

[0140]

[0141] Where, and are the time when the i-th vehicle enters and leaves the road section, N v The number of vehicles passing the road section within 15 minutes.

[0142] According to the motion wave model obtained by the embodiment of the present invention based on formulas (1)-(5), the entry and exit times of all vehicles passing through the road section within 15 minutes can be obtained, thereby determining the road section travel time. The travel time of all road sections in the road network is determined by formula (6).

[0143] S24: According to the road section travel time tr ij and travel demand, and use Dijkstra's shortest path algorithm to plan the set of sections with the shortest travel time between activity locations

[0144] The Dijkstra algorithm is an algorithm for calculating the shortest path of a unit in a weighted directed graph. The Dijkstra algorithm adopts a greedy strategy, each time selecting the unresolved node closest to the starting node, ensuring that the path selected each time is the current optimal one. The present invention uses the Dijkstra algorithm to obtain the set of sections with the shortest travel time between activity locations.

[0145] Specifically, step S3 considers traffic conditions and vehicle driving data to establish an energy consumption model for electric vehicles, including:

[0146] S31: Considering real-time traffic conditions, establish an electric vehicle energy consumption model based on the electric vehicle's speed, air resistance, and rolling friction:

[0147]

[0148] in,

[0149] v ij =l ij / tr ij (8)

[0150]

[0151] in, For an electric car with a constant speed v ij The energy consumed by passing the flat road section (i, j), v ij is the driving speed, It is the power supplied to the motor to overcome air resistance and rolling friction. is the air resistance, is the rolling friction, η is the efficiency of the motor, ρ is the air density, C d is the air resistance coefficient, A is the area of the front area of the electric vehicle, m is the mass of the electric vehicle, C r is the rolling friction coefficient, and g is the acceleration due to gravity.

[0152] The embodiment of the present invention establishes an electric vehicle energy consumption model based on the traffic characteristic model established in step 2 and taking into account the real-time traffic conditions. Since the driving energy consumption of an electric vehicle is proportional to the motor power and the length of the road section, and inversely proportional to the driving speed, the greater the motor power or the longer the road section, the greater the driving energy consumption. The faster the driving speed, the lower the driving energy consumption. Based on the above characteristics, the embodiment of the present invention takes into account the real-time traffic conditions and establishes an electric vehicle energy consumption model based on the driving speed, air resistance and rolling friction of the electric vehicle, as shown in formula (7). Formula (7) can be used to determine the energy consumption of an electric vehicle at a constant speed v ij The energy consumed by passing the flat road section (i, j).

[0153] S32: Calculating the energy consumption of the electric vehicle during driving based on the electric vehicle energy consumption model.

[0154] Specifically, step S4 considers the queuing time at the charging station, the charging time, and the driving energy consumption of the electric vehicle to establish a charging characteristic model of the electric vehicle, including:

[0155] S41: Establish a charging station queue model based on the charging station queue time and charging time:

[0156]

[0157] in,

[0158]

[0159] in, is the queuing time required for charging station k, is the charging time required for charging station k, The remaining power of the electric vehicle when it arrives at the charging station. P is the remaining power of the electric vehicle when it decides to detour to charge. fc is the rated charging power of the charging station, is the set of road segments that constitute the travel path from road segment m to charging station c, E b is the rated capacity of the battery of the electric vehicle, For an electric car with a constant speed v ij The energy consumed by passing the flat road section (i, j).

[0160] Since the temporal and spatial distribution of electric vehicle charging load is affected by the choice of electric vehicle charging station, the present invention determines the selection of electric vehicle charging stations by establishing a charging characteristic model. First, a charging station queuing model is established based on the queuing time and charging time of the charging station. Assuming that all electric vehicles in the charging station leave the charging station after being fully charged, the time when the electric vehicle leaves the charging station is equal to the time when it arrives at the charging station plus the queuing time and charging time. If the queuing time or charging time of the charging station is longer, the time when the electric vehicle leaves the charging station will be later. Therefore, the time when the electric vehicle leaves the charging station is It can be determined by formula (10).

[0161] Assume Q w ={t w1 ,t w2 L}, Q c ={t c1 ,t c2 L} are the time sets of the vehicles queuing and charging in the charging station to complete charging, respectively. The elements can be determined by formula (10). The time set of all vehicles in the charging station to complete charging is Qt =Q w UQ c , Q t Generate the set T={t1,t2L t n}, calculate the difference between the time when all vehicles in the charging station complete charging and the arrival time of the current vehicle, and take the maximum value of these differences to represent the time the current vehicle needs to wait for all vehicles in front of it to complete charging. If the calculated maximum value is a negative number (that is, no vehicle is charging when the current vehicle arrives), the queue time is 0, indicating that it can charge directly. Therefore, the queue time required for electric vehicles in charging station k is for:

[0162]

[0163] b k =N k -G k (13)

[0164] Among them, b k is the actual queue number at charging station k, N k is the number of all vehicles at charging station k, G k is the number of chargers in charging station k.

[0165] S42: When the remaining power of the electric vehicle meets When , the electric vehicle charging characteristic model is established with the minimum cost as the goal:

[0166] Select the objective function F of the charging station obj for:

[0167]

[0168] in, The real-time remaining power of the electric vehicle during travel, E min is the charging threshold for electric vehicle battery protection, C k The cost of choosing different charging stations for electric vehicles, α q is the time cost of electric vehicle users for different average annual household incomes, and β is the charging price. The time cost is typically related to the user's waiting time and charging time. For example, if the user has to wait a long time before charging begins, or if the charging time is long, the time cost will be higher. The charging fee is the amount the user pays to charge at the charging station. A higher charging price or a larger amount of electricity required will also result in a higher charging fee.

[0169] When the electric vehicle is low on power during travel (i.e. ), the electric vehicle will abandon the planned route to the activity location and choose the detour charging station in the graph with the minimum cost to quickly charge to full power, and then re-plan the route to the activity location. The choice of charging station can be determined by formula (14).

[0170] Specifically, step S5 establishes a travel chain utility function based on the traffic characteristic model, energy consumption model, and charging characteristic model, and solves the spatiotemporal distribution of the electric vehicle charging load under the traffic equilibrium state, including:

[0171] When each electric vehicle user is unable to unilaterally improve their plan under constraints, the system has reached a traffic equilibrium state. The traffic equilibrium criterion is as follows:

[0172] S51: Establish a daily travel chain utility function for a single electric vehicle user:

[0173]

[0174] Where U q,s is the daily travel utility of a single electric vehicle in the sth iteration, is the utility of the mth electric vehicle trip in the sth iteration, reflecting the utility or cost incurred by the user during travel from one activity location to another. For example, factors such as travel time and travel comfort can affect trip utility. This can be further broken down into the direct marginal utility of travel time and the marginal utility of activity duration.

[0175]

[0176] Where, β tra is the direct marginal utility of activity travel time, is the travel time from the location of activity m to the location of activity m+1 in the sth iteration.

[0177] S52: Based on the daily travel chain utility function of a single electric vehicle user, determine the average travel chain utility of all electric vehicles in the sth iteration as:

[0178]

[0179] Where, is the average value of the travel chain utility of all electric vehicle users in the sth iteration, and NV is the total number of electric vehicle users participating in the simulation.

[0180] S53: The time-space distribution of the electric vehicle charging load of the system in the traffic equilibrium state is solved by the successive averaging method. The expected link travel time used in the next iteration of the path selection is shown in formula (19):

[0181]

[0182] Where, represents the expected travel time of link (i, j) at time t in the n+1th iteration, which can be used as the input of the model for path selection in the next iteration. It represents the expected travel time of link (i, j) at time t in the nth iteration, which can be used as the input of the model for path selection in the current iteration. n represents the number of current iterations, and t∈[T h ,T h+1 ] represents the preset time interval of time t.

[0183] S54: Determine whether the system meets the traffic equilibrium convergence condition based on the change in the average trip chain utility of all electric vehicles. If so, the system reaches traffic equilibrium, the iteration is terminated, and the optimal electric vehicle travel and charging strategy is obtained. If the change in the total utility of all users between two iterations is small, it indicates that the user's travel plan has hardly changed between the two iterations, and the system is stable.

[0184] The traffic equilibrium convergence condition is:

[0185]

[0186] Where NS is the number of simulations, ε DUE For simulation accuracy.

[0187] Example 2

[0188] Based on the above method, the embodiment of the present invention also provides an electric vehicle charging load prediction system, such as Figures 9-13 Shown, including:

[0189] Topology map creation module 01 is used to construct a transportation network topology map based on graph theory and trip chain theory.

[0190] The traffic characteristic model establishment module 02 is used to establish a traffic characteristic model of electric vehicles based on the traffic network topology diagram.

[0191] The energy consumption model establishment module 03 is used to establish an energy consumption model of the electric vehicle taking into account traffic conditions and vehicle driving data.

[0192] The charging characteristic model building module 04 is used to build a charging characteristic model of the electric vehicle by considering the queuing time at the charging station, the charging time, and the driving energy consumption of the electric vehicle.

[0193] The charging load prediction module 05 is used to establish a travel chain utility function based on the traffic characteristic model, energy consumption model and charging characteristic model, and solve the spatiotemporal distribution of the electric vehicle charging load under the traffic equilibrium state.

[0194] The present invention first constructs a traffic network topology using a topology map creation module 01, and establishes a traffic characteristic model of electric vehicles based on a traffic characteristic model establishment module 02 to study the spatiotemporal distribution of electric vehicles. Then, an energy consumption model establishment module 03 is used to establish an energy consumption model of electric vehicles to calculate the energy consumption of electric vehicles during driving. A charging characteristic model establishment module 04 is used to establish a charging characteristic model of electric vehicles to determine the selection of charging stations for electric vehicles based on the remaining power of the electric vehicles. Finally, a charging load prediction module 05 is used to establish a travel chain utility function to determine the spatiotemporal distribution of the charging load of electric vehicles in a traffic equilibrium state. By combining the electric vehicle travel chain model, energy consumption model, and charging characteristic model with traffic equilibrium theory, the present invention can predict the spatiotemporal distribution of the real-time changing charging load of electric vehicles in complex traffic networks, providing a scientific basis for the coordinated optimization operation of the power-transportation system.

[0195] Specifically, the topology map creation module 01 further includes:

[0196] The electric vehicle travel chain creation module 011 is used to obtain the travel data of electric vehicles in the predicted area within one day and build an electric vehicle travel chain based on the travel data, where the travel data includes activity type, activity location and departure time.

[0197] The electric vehicle travel chain established by the electric vehicle travel chain creation module 011 in the embodiment of the present invention is as follows: Figure 7 As shown in the figure, the outer circle of the trip chain represents the time chain, which is used to describe the time when the electric vehicle arrives at or leaves the activity location. The inner circle of the trip chain represents the space chain, which is used to describe the spatial position relationship of the various activities that the electric vehicle goes to. Specifically, the variables of the electric vehicle trip chain include the time when the electric vehicle arrives at the activity location m Time of leaving the activity location m The time when the route c where the charging is decided and the time when leaving route c when going to the activity location m+1 Arrival time at fast charging station k The moment you leave the fast charging station The set of road segments that constitute the trip m path The set of road segments that constitute the travel path from road segment c to charging station k and the set of road segments that constitute the travel path from charging station k to activity location m+1 By leveraging these variables, the system can accurately capture the behavioral changes of electric vehicle users during their travels, providing fundamental data support for subsequent traffic network modeling and charging load forecasting. This paper describes the daily travel needs of electric vehicles by constructing an electric vehicle travel chain. Specific travel demand information includes activity type, location, and departure time, comprehensively reflecting users' daily travel behaviors.

[0198] The traffic network topology graph creation module 012 is used to simplify the electric vehicle travel chain into a traffic network topology graph consisting of edges and vertices using graph theory.

[0199] The embodiment of the present invention simplifies the electric vehicle travel chain into a traffic network topology diagram consisting of edges (road sections) and vertices (intersections or nodes) through the traffic network topology diagram creation module 012, such as Figure 8 As shown, Figure 8 Given a long and uniform road section, the sections are connected by nodes. Each section includes an upstream queue and a downstream queue. By defining the upstream queue and the downstream queue, the inflow and outflow of vehicles in the section can be described.

[0200] Specifically, the traffic characteristics model building module 02 further includes:

[0201] The motion wave model creation module 021 is used to define the mathematical models of the upstream queue and the downstream queue according to the motion wave theory, and to define the motion wave model in the discrete time simulation state based on the mathematical models of the upstream queue and the downstream queue.

[0202] The embodiment of the present invention first utilizes motion wave model to create module 021 to set up the mathematical model of upstream queue and downstream queue, in order to simulate the propagation process of vehicle between road section.This model has described the inflow and outflow situation of vehicle in road section, and the variation of number of vehicles in the upstream and downstream queues of road section by defining upstream queue and downstream queue.In each time step, the number of vehicles in upstream queue can be adjusted according to the variation of number of vehicles flowing in and out.The number of vehicles flowing in is relevant to the driving speed of vehicle and road section length, and the number of vehicles flowing out is relevant to the gap propagation speed. At the t moment, by inputting the number of vehicles of upstream queue at the previous moment, the number of vehicles of downstream queue at the previous moment, the number of vehicles flowing into queue at the current moment, the number of vehicles flowing out queue at the current moment, obtain upstream queue and downstream queue in the motion wave model, simulate the inflow and outflow process of vehicle in road section respectively, and then describe the propagation and congestion formation of traffic flow.

[0203] Next, a traffic motion wave model is established in a discrete-time simulation state using the motion wave model creation module 021. This model describes the number of vehicles received and sent by a road section within each time step. Within a time interval t, the number of vehicles a road section can receive is limited by two factors: the number of vehicles attempting to enter and the remaining capacity of the road section. Ultimately, the number of vehicles received by the road section is the smaller of these two values, ensuring that the road section does not exceed its maximum capacity. Conversely, within a time interval t, the number of vehicles a road section can send is limited by two factors: the number of vehicles in the downstream queue and the maximum capacity of the road section. The number of vehicles sent by the road section is the smaller of these two values, ensuring that the road section does not exceed its maximum capacity. In this embodiment of the present invention, the smaller value of the number of vehicles attempting to enter or the remaining capacity of the road section, and the smaller value of the number of vehicles in the downstream queue or the maximum capacity of the road section, is selected to determine the number of vehicles received and sent by the road section within a simulation step. This model can be used to illustrate the formation, propagation, and dissipation of traffic congestion. For example, when the traffic density exceeds a critical value, a compression wave propagates backward, forming a traffic jam.

[0204] The road section travel time calculation module 022 is used to calculate the entry and exit times of all vehicles passing through the road section within 15 minutes based on the motion wave model, and determine the travel time of all road sections in the road network.

[0205] In the embodiment of the present invention, the road section travel time calculation module 022 can obtain the entry and exit times of all vehicles passing through the road section within 15 minutes, thereby determining the travel time of all road sections in the road network.

[0206] The shortest driving route calculation module 023 is used to plan the driving route with the shortest travel time between activity locations using the Dijkstra shortest path algorithm based on the road section travel time and travel demand.

[0207] The embodiment of the present invention can obtain the set of road sections with the shortest travel time between activity locations through the shortest travel path calculation module 023.

[0208] Specifically, the charging characteristic model building module 04 also includes:

[0209] The charging station queue model building module 041 is used to build a charging station queue model according to the queue time and charging time of the charging station.

[0210] The electric vehicle charging characteristic model establishing module 042 is used to establish an electric vehicle charging characteristic model according to the energy consumption of the electric vehicle.

[0211] The charging station determination module 043 is used to determine a charging station based on the electric vehicle charging characteristic model.

[0212] The embodiment of the present invention first uses the charging station queuing model construction module 041 to establish a charging station queuing model to determine the queuing time and charging time required for electric vehicles at the charging station, and then uses the electric vehicle charging characteristic model construction module 042 to determine the remaining power of the electric vehicle. When the electric vehicle is low on power during the trip, the electric vehicle will abandon the planned route to the activity location and use the charging station determination module 043 to determine the charging station to be selected. It will then select a detour charging station in the diagram with the minimum cost to quickly charge the vehicle to full power, and then re-plan the route to the activity location.

[0213] Specifically, the charging load prediction module 05 also includes:

[0214] The travel chain utility function creation module 051 is used to establish the daily travel chain utility function of a single electric vehicle user.

[0215] The average trip chain utility function creation module 052 is used to determine the average trip chain utility of all electric vehicles in the sth iteration based on the daily trip chain utility function of a single electric vehicle user.

[0216] The electric vehicle charging load spatiotemporal distribution calculation module 053 is used to iteratively solve the spatiotemporal distribution of the electric vehicle charging load when the system is in a traffic equilibrium state using a successive averaging method.

[0217] This embodiment of the present invention first uses the trip chain utility function creation module 051 to establish a daily trip chain utility function for a single electric vehicle user, reflecting the utility or cost incurred by the user during travel from one activity location to another. Then, the average trip chain utility function creation module 052 is used to establish an average trip chain utility function, which is used to determine the average trip chain utility of all electric vehicles in the sth iteration. Finally, the electric vehicle charging load spatiotemporal distribution calculation module 053 is used to iteratively solve the spatiotemporal distribution of the electric vehicle charging load in the system under traffic equilibrium. This module first determines whether the system meets the traffic equilibrium convergence condition based on the change in the average trip chain utility of all electric vehicles. If so, the system reaches traffic equilibrium, the iteration is terminated, and the optimal electric vehicle travel and charging strategy is obtained.

[0218] Example 3

[0219] like Figure 14 As shown, an embodiment of the present invention further provides a computer device, comprising a processor and a memory. The memory stores a computer program, which is loaded and executed by the processor to implement the above-mentioned electric vehicle charging load prediction method. The device in the present invention can be a server, a PC, a PAD, a mobile phone, etc.

[0220] Furthermore, an embodiment of the present invention also provides a computer-readable storage medium, in which a computer program is stored. The computer program is loaded and executed by a processor to implement the above-mentioned electric vehicle charging load prediction method.

[0221] Example 4

[0222] like Figure 15 As shown in the figure, the embodiment of the present invention uses a certain region as the research area for simulation analysis. According to OpenStreetMap (open source map), the road network of level 4 or above in the research area is selected as the traffic network. After simplification, it consists of 579 nodes and 781 one-way roads. A total of 18 fast charging stations were established with reference to the charging facility planning. The road network topology and the distribution of fast charging stations are shown in the attached figure. Figure 15 As shown in the figure, based on the 2020 census results, a sample of 2% of the sample population is selected, considering private car travel. Five activity types are considered: home (H), work (W), shopping (S), leisure (L), and other (O). A travel plan is generated for each individual in the sample population.

[0223] Two scenarios were considered for analysis: Scenario S1 did not consider the interaction of electric vehicle travel within the road network, while Scenario S2 did. This paper primarily analyzes the impact of real-time interactions among electric vehicle groups on the spatiotemporal distribution of electric vehicle loads. Therefore, the influence of random electric vehicle type was not considered. Other simulation parameters are shown in Table 1.

[0224] Table 1 Other simulation parameters

[0225]

[0226] Simulation results show that as the number of iterations increases, user travel in the network gradually meets the dynamic traffic equilibrium criterion. At this point, all electric vehicle users cannot improve their daily travel utility by changing their travel routes. Simultaneously, the average active travel time for all users reaches its minimum, decreasing by 28.3% compared to scenario S1. Compared to scenario S1, in scenario S2, all electric vehicles tend to choose the shortest route rather than the shortest distance. Therefore, users traveling during congested periods increase their travel distance to reduce their travel time. Because the shortest route is chosen in scenario S1, peak network traffic loads during peak periods are higher and last longer, resulting in more congestion. In scenario S2, users choose the shortest route, allowing them to leave the network earlier, resulting in better traffic conditions. During off-peak periods, network traffic is almost the same. Regarding charging load, the total energy consumed by electric vehicles in scenario S2 is higher than in scenario S1. In terms of spatial distribution, scenario S2 has more charging stations experiencing peak loads than scenario S1. In terms of temporal distribution, the peak charging load for electric vehicles lasts longer. As a result, the spatiotemporal distribution characteristics of the charging load of electric vehicles in the entire study area have changed.

[0227] In summary, the electric vehicle charging load prediction method, system, device, and storage medium provided by the present invention have the following advantages:

[0228] 1. By introducing traffic equilibrium theory and combining the trip chain model, motion wave model and Dijkstra algorithm, a model of electric vehicle traffic characteristics, energy consumption characteristics and charging characteristics is constructed. This model can simulate the interactive behavior of electric vehicles in the road network in real time and is used to predict the spatiotemporal distribution of charging load.

[0229] 2. By establishing a travel chain utility function and equilibrium criteria, the interactive problem of real-time route selection for electric vehicle groups is solved, ensuring that the system reaches the optimal traffic equilibrium state. At the same time, considering real-time traffic conditions and charging station selection, the energy consumption and charging demand of electric vehicles are calculated, and the charging station selection strategy is optimized. Through the successive averaging method, intelligent prediction of electric vehicle charging load is achieved, providing a scientific basis for load management of the power system and route planning of the transportation system.

[0230] 3. The present invention can predict the real-time changing charging load of electric vehicles in complex transportation networks, reduce the operating pressure of the power system, optimize the smoothness of the transportation network, and provide strong technical support for the coordinated operation of the power-transportation system.

[0231] 4. By considering the real-time interactions between EV users in complex transportation networks, the spatiotemporal distribution of EV charging loads can be more accurately predicted, particularly in terms of real-time variations. This approach ensures that EV users' travel paths are consistent with actual traffic conditions, avoiding the limitations of traditional shortest path algorithms.

[0232] 5. The system can update traffic conditions, electric vehicle energy consumption and charging station selection in real time, provide real-time charging load forecast results, and provide real-time data support for the scheduling and optimization of the power system.

[0233] 6. The system is applicable to transportation networks of different sizes and complexities, and can provide a scientific basis for urban transportation planning and power system operation.

[0234] The concepts, principles, and ideas of the present invention are described in detail above in conjunction with specific implementation methods (including embodiments and examples). Those skilled in the art should understand that the implementation methods of the present invention are not limited to the forms given above. After reading this application document, those skilled in the art can make any possible improvements, replacements, and equivalent forms to the steps, methods, systems, and components in the above-mentioned implementation methods. These improvements, replacements, and equivalent forms should be deemed to fall within the scope of the present invention, and the scope of protection of the present invention shall be subject only to the claims.

Claims

1. A method for predicting electric vehicle charging load, characterized in that: include: Construct a transportation network topology based on graph theory and trip chain theory; Establishing a traffic characteristic model of electric vehicles based on the traffic network topology diagram; Establish an energy consumption model for electric vehicles by considering traffic conditions and vehicle driving data; The charging characteristic model of electric vehicles is established by considering the queuing time at the charging station, charging time and the driving energy consumption of electric vehicles; A travel chain utility function is established based on the traffic characteristic model, energy consumption model and charging characteristic model, and the spatiotemporal distribution of the electric vehicle charging load under the traffic equilibrium state of the system is solved.

2. The electric vehicle charging load prediction method according to claim 1, characterized in that: The construction of a transportation network topology graph based on graph theory and trip chain theory includes: Acquire travel data of electric vehicles in a forecasted area within one day, and construct an electric vehicle travel chain based on the travel data, wherein the travel data includes activity type, activity location, and departure time; Graph theory is used to simplify the electric vehicle travel chain into a traffic network topology graph consisting of edges and vertices.

3. The electric vehicle charging load prediction method according to claim 2, characterized in that: The variables of the electric vehicle travel chain include the time of arrival at the activity location m Time of leaving the activity location m The time when the route c where the charging is decided and the time when leaving route c when going to the activity location m+1 Arrival time at fast charging station k The moment you leave the fast charging station The set of road segments that constitute the trip m path The set of road segments that constitute the travel path from road segment c to charging station k and the set of road segments that constitute the travel path from charging station k to activity location m+1 4. The electric vehicle charging load prediction method according to claim 3, characterized in that: The establishing of the traffic characteristic model of the electric vehicle based on the traffic network topology diagram includes: Based on the traffic network topology and motion wave theory, the mathematical model of upstream and downstream queues is defined as follows: UQ(t)=N in (t)-N out (t+δ-l ij / w) DQ(t)=N in (t+δ-l ij / v)-N out (t) Among them, UQ(t) is the upstream queue, DQ(t) is the downstream queue, and li j is the length of the road section, δ is the simulation time step, w is the propagation speed of the congestion wave from downstream to upstream, v is the zero flow speed, N in (t) and N out (t) are the number of vehicles entering the queue and the number of vehicles leaving the queue as of time t; N out (t+δ-l ij / w) indicates the time point (t+δ-l ij / w) the number of vehicles flowing out of the queue, N in (t+δ-l ij / v) indicates the time point (t+δ-l ij / v) the number of vehicles entering the queue; The motion wave model under the discrete time simulation state is defined based on the mathematical model of the upstream queue and the downstream queue: Among them, R(t) and S(t) are the number of vehicles received and sent by the road section in the time interval t to t+δ, respectively. is the maximum density of the road section, is the maximum traffic flow of the road section, λ is the traffic scaling factor, and L is the length of the road section; Based on the motion wave model, the entry and exit times of all vehicles passing through the road section within 15 minutes are calculated, and the travel time tr of all road sections in the road network is determined. ij : Where, and are the time when the i-th vehicle enters and leaves the road section, N v The number of vehicles passing the road section within 15 minutes; According to the road section travel time tr ij and travel demand, and use Dijkstra's shortest path algorithm to plan the set of sections with the shortest travel time between activity locations 5. The electric vehicle charging load prediction method according to claim 4, characterized in that: The energy consumption model of the electric vehicle established by considering traffic conditions and vehicle driving data includes: Considering real-time traffic conditions, an electric vehicle energy consumption model is established based on the electric vehicle's speed, air resistance, and rolling friction: in, v ij =l ij / tr ij in, For an electric car with a constant speed v ij The energy consumed by passing the flat road section (i, j), v ij is the driving speed, is the power supplied to the motor, is the air resistance, is the rolling friction, η is the efficiency of the motor, ρ is the air density, C d is the air resistance coefficient, A is the area of the front area of the electric vehicle, m is the mass of the electric vehicle, C r is the rolling friction coefficient, and g is the acceleration due to gravity.

6. The electric vehicle charging load prediction method according to claim 5, characterized in that: The charging characteristic model of the electric vehicle is established by considering the queuing time at the charging station, the charging time, and the driving energy consumption of the electric vehicle, including: Establish a charging station queuing model based on the queuing time and charging time of the charging station: in, in, is the queuing time required for charging station k, is the charging time required for charging station k, The remaining power of the electric vehicle when it arrives at the charging station. P is the remaining power of the electric vehicle when it decides to detour to charge. fc is the rated charging power of the charging station, is the set of road segments that constitute the travel path from road segment m to charging station c, E b is the rated capacity of the battery of the electric vehicle; For an electric car with a constant speed v ij The energy consumed by passing the flat road section (i, j); Assume Q w ={t w1 ,t w2 L}, Q c ={t c1 ,t c2 L} are the time sets of the vehicles queuing and charging in the charging station, respectively. The time set of all vehicles in the charging station that have completed charging is Q t =Q w UQ c , Q t Generate a set T from small to large T = {t1, t2Lt n }, then the queuing time required for electric vehicles in charging station k is for: b k =N k -G k Among them, b k is the actual queue number at charging station k, N k is the number of all vehicles at charging station k, G k is the number of chargers in charging station k; When the remaining power of the electric vehicle meets When , the electric vehicle charging characteristic model is established with the minimum cost as the goal: Select the objective function F of the charging station obj for: in, The real-time remaining power of electric vehicles during travel; E min is the charging threshold for electric vehicle battery protection, C k The cost of choosing different charging stations for electric vehicles, α q is the time cost of electric vehicle users corresponding to different average annual household incomes, and β is the unit price of charging.

7. The electric vehicle charging load prediction method according to claim 1, characterized in that: The method of establishing a travel chain utility function based on the traffic characteristic model, energy consumption model, and charging characteristic model, and solving the spatiotemporal distribution of the electric vehicle charging load under the traffic equilibrium state includes: Establish a daily travel chain utility function for a single electric vehicle user: Among them, U q,s is the daily travel utility of a single electric vehicle in the sth iteration, is the utility of the electric vehicle’s mth trip in the sth iteration, β tra is the direct marginal utility of activity travel time, is the travel time from the location of activity m to the location of activity m+1 in the sth iteration; Based on the daily trip chain utility function of a single electric vehicle user, the average trip chain utility of all electric vehicles in the sth iteration is determined as: Where, is the average value of the travel chain utility of all electric vehicle users in the sth iteration, and NV is the total number of electric vehicle users participating in the simulation; The successive averaging method is used to iteratively solve the spatiotemporal distribution of electric vehicle charging load in the system under traffic equilibrium state: Where, represents the expected travel time of link (i, j) at time t in the n+1th iteration, represents the expected travel time of link (i, j) at time t in the nth iteration, n represents the current iteration number, t∈[T h ,T h+1 ] represents the preset time interval of time t; Based on the change in the average travel chain utility of all electric vehicles, it is determined whether the system meets the traffic equilibrium convergence condition. If so, the system reaches a traffic equilibrium state, the iteration is terminated, and the optimal electric vehicle travel and charging strategy is obtained. The traffic equilibrium convergence condition is: Where NS is the number of simulations, ε DUE For simulation accuracy.

8. An electric vehicle charging load prediction system, characterized in that: include: Topology creation module, used to construct transportation network topology based on graph theory and trip chain theory; A traffic characteristic model building module, used to build a traffic characteristic model of electric vehicles based on the traffic network topology map; Energy consumption model building module, used to build an energy consumption model for electric vehicles taking into account traffic conditions and vehicle driving data; A charging characteristic model building module is used to build a charging characteristic model for electric vehicles by considering the queuing time at the charging station, the charging time, and the driving energy consumption of the electric vehicle; The charging load prediction module is used to establish a travel chain utility function based on the traffic characteristic model, energy consumption model and charging characteristic model, and solve the spatiotemporal distribution of the electric vehicle charging load under the traffic equilibrium state.

9. The electric vehicle charging load prediction system according to claim 8, characterized in that: The topology map creation module also includes: An electric vehicle travel chain creation module is used to obtain travel data of electric vehicles in a forecasted area within a day and construct an electric vehicle travel chain based on the travel data, wherein the travel data includes activity type, activity location and departure time; The traffic network topology diagram creation module is used to simplify the electric vehicle travel chain into a traffic network topology diagram consisting of edges and vertices by using graph theory.

10. The electric vehicle charging load prediction system according to claim 9, characterized in that: The variables of the electric vehicle travel chain include the time of arrival at the activity location m Time of leaving the activity location m The time when the route c where the charging is decided and the time when leaving route c when going to the activity location m+1 Arrival time at fast charging station k The moment you leave the fast charging station The set of road segments that constitute the trip m path The set of road segments that constitute the travel path from road segment c to charging station k and the set of road segments that constitute the travel path from charging station k to activity location m+1 11. The electric vehicle charging load prediction system according to claim 10, characterized in that: The traffic characteristics model building module also includes: A motion wave model creation module, configured to define mathematical models of an upstream queue and a downstream queue according to motion wave theory, and to define a motion wave model in a discrete time simulation state based on the mathematical models of the upstream queue and the downstream queue; A road section travel time calculation module is used to calculate the entry and exit times of all vehicles passing through the road section within 15 minutes based on the motion wave model, and determine the travel time of all road sections in the road network; The shortest driving path calculation module is used to plan a set of paths with the shortest travel time between activity locations using the Dijkstra shortest path algorithm based on the travel time of the road section and travel demand.

12. The electric vehicle charging load prediction system according to claim 11, characterized in that: The charging characteristic model building module also includes: A charging station queue model building module is used to build a charging station queue model based on the queue time and charging time of the charging station; The electric vehicle charging characteristic model establishment module is used to establish the electric vehicle charging characteristic model according to the energy consumption of the electric vehicle.

13. The electric vehicle charging load prediction system according to claim 12, characterized in that: The charging characteristic model building module also includes: The charging station determination module is used to determine the charging station based on the electric vehicle charging characteristic model.

14. The electric vehicle charging load prediction system according to claim 13, characterized in that: The charging load prediction module also includes: The travel chain utility function creation module is used to establish the daily travel chain utility function of a single electric vehicle user; an average trip chain utility function creation module, configured to determine the average trip chain utility of all electric vehicles in the sth iteration based on the daily trip chain utility function of the single electric vehicle user; The electric vehicle charging load spatiotemporal distribution calculation module is used to iteratively solve the spatiotemporal distribution of electric vehicle charging load when the system is in a traffic equilibrium state using the successive averaging method.

15. A computer device, characterized in that: The computer device includes a processor and a memory, wherein a computer program is stored in the memory, and the computer program is loaded and executed by the processor to implement the electric vehicle charging load prediction method according to any one of claims 1 to 7.

16. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which is loaded and executed by a processor to implement the electric vehicle charging load prediction method according to any one of claims 1 to 7.

Images