Electric vehicle schedulable capacity prediction method, system, equipment and medium

Abstract

The invention relates to the field of electric vehicles, and discloses an electric vehicle schedulable capacity prediction method, system and device and a medium, and the method comprises the steps: constructing a travel chain model of an electric vehicle in a target area, and constructing a travel space-time model of each electric vehicle in combination with a Markov chain; a traffic network model of the target area is constructed to be combined with the travel space-time models, and a multi-target optimization algorithm is adopted to predict travel paths of the electric vehicles; determining the time cost of introducing a time emergency factor based on the travel path prediction result of each electric vehicle, and combining the time cost with the battery degradation cost, the discharge cost and the mood loss cost to establish a scheduling willingness function, so that each electric vehicle responds to the received real-time scheduling request based on the respective scheduling willingness function; the electric vehicle dispatching capacity in the target area is obtained; accurate prediction of the dispatchable capacity of the electric vehicle in the target area is realized.

Description

Technical Field

[0001] The present invention relates to the field of electric vehicles, and in particular to a method, system, device and medium for predicting the dispatchable capacity of electric vehicles. Background Art

[0002] With the rapid growth of electric vehicles in recent years, their charging and discharging behavior has significantly impacted grid operations. Therefore, rationally predicting the dispatchable capacity of electric vehicles has become a key issue in improving grid stability and optimizing energy dispatch. Current predictions of EV dispatchable capacity rely heavily on historical dispatch data, but these methods don't fully consider the dispatchability preferences of EV owners, significantly reducing the accuracy of these predictions.

[0003] It can be seen that how to improve the prediction results of the dispatchable capacity of electric vehicles has become a technical problem that needs to be solved urgently by those skilled in the art. Summary of the Invention

[0004] The present invention provides a method, system, device and medium for predicting the dispatchable capacity of electric vehicles, which solve the problem of how to improve the prediction result of the dispatchable capacity of electric vehicles.

[0005] To solve the above technical problems, the present invention provides a method for predicting the dispatchable capacity of electric vehicles, comprising:

[0006] Constructing a travel chain model for electric vehicles in a target area, and using a Markov chain to model the transfer probability of each electric vehicle in the target area based on the travel chain model to obtain a travel spatiotemporal model of each electric vehicle;

[0007] Constructing a transportation network model of the target area, and based on each of the travel spatiotemporal models and the transportation network model, using a multi-objective optimization algorithm to search for travel paths of each of the electric vehicles to obtain predicted travel paths of each of the electric vehicles;

[0008] Determining a time cost of introducing a time urgency factor based on each predicted travel path, and combining this with a battery degradation cost, a discharge cost, and a mood loss cost of each electric vehicle to establish a dispatch willingness function for each electric vehicle;

[0009] A real-time dispatch request is sent to each of the electric vehicles, so that each of the electric vehicles performs a dispatch willingness judgment based on the real-time dispatch request using each of the dispatch willingness functions to obtain the electric vehicle dispatch capacity within the target area.

[0010] A second aspect of the present invention provides a system for predicting the dispatchable capacity of electric vehicles, comprising:

[0011] a model building module, configured to build a travel chain model for electric vehicles in a target area, and to model the transfer probability of each electric vehicle in the target area using a Markov chain based on the travel chain model, thereby obtaining a travel spatiotemporal model for each electric vehicle;

[0012] a path prediction module, configured to construct a traffic network model of the target area, and based on each of the travel spatiotemporal models and the traffic network model, use a multi-objective optimization algorithm to search for a travel path of each of the electric vehicles to obtain a predicted travel path of each of the electric vehicles;

[0013] a function establishment module for determining a time cost of introducing a time urgency factor based on each predicted travel path, and combining this with a battery degradation cost, a discharge cost, and a mood loss cost of each electric vehicle to establish a dispatch willingness function for each electric vehicle;

[0014] The capacity determination module is used to send a real-time scheduling request to each of the electric vehicles, so that each of the electric vehicles performs a scheduling willingness judgment based on the real-time scheduling request through each of the scheduling willingness functions to obtain the electric vehicle scheduling capacity in the target area.

[0015] A third aspect of the present invention provides an electronic device comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the above-described method for predicting the dispatchable capacity of electric vehicles when executing the computer program.

[0016] A fourth aspect of the present invention provides a computer-readable storage medium, which includes a stored computer program. When the device where the computer-readable storage medium is located executes the computer program, the electric vehicle dispatchable capacity prediction method as described above is implemented.

[0017] Compared with the prior art, the embodiments of the present invention have the following advantages:

[0018] (1) By introducing the traffic model, the all-weather spatiotemporal distribution of electric vehicles can be more accurately simulated, thereby more accurately predicting the dispatchable capacity of electric vehicles in the target area, providing reliable data support for the grid's dispatch decision-making;

[0019] (2) Based on the constructed travel chain model and traffic network model, a multi-objective optimization algorithm is used to search for travel paths, which can more accurately plan the travel routes of electric vehicles, comprehensively consider multiple factors such as time and energy consumption, and improve travel efficiency and resource utilization;

[0020] (3) By sending dispatch requests in real time and making judgments based on the dispatch willingness function that considers conventional costs, time costs with time urgency factors, and mood loss costs, the dispatch status of electric vehicles can be flexibly adjusted according to actual conditions, and different dispatch needs can be responded to in a timely manner, thereby improving the flexibility and adaptability of the entire electric vehicle dispatch system. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for use in the implementation. Obviously, the drawings described below are only some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0022] Figure 1 This is a flow chart of a method for predicting the dispatchable capacity of electric vehicles provided by one embodiment of the present invention;

[0023] Figure 2 This is a structural diagram of the fusion of the distribution network and transportation network model provided by an embodiment of the present invention;

[0024] Figure 3 This is a structural diagram of an electric vehicle dispatchable capacity prediction system provided by an embodiment of the present invention;

[0025] Figure 4 This is a structural diagram of an electronic device provided by one embodiment of the present invention;

[0026] Reference numerals:

[0027] Among them, 101, model building module; 102, path prediction module; 103, function establishment module; 104, capacity determination module; 5000, electronic device; 5001, processor; 5002, bus; 5003, memory; 5004, transceiver. DETAILED DESCRIPTION

[0028] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings and embodiments. Obviously, the embodiments described are only some embodiments of the present invention, rather than all embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0029] In the description of this application, the terms "first," "second," "third," etc. are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first," "second," "third," etc. may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0030] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or an indirect connection through an intermediate medium, or it can be a communication between the two components. The terms "vertical", "horizontal", "left", "right", "up", "down" and similar expressions used herein are for illustrative purposes only, and do not indicate or imply that the system or component referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to the specific circumstances.

[0031] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in this application have the same meanings as those commonly understood by those skilled in the art. The terms used in this specification are only for describing specific embodiments and are not intended to limit the present invention. Those skilled in the art will understand the specific meanings of the above terms in this application according to specific circumstances.

[0032] In one embodiment, if Figure 1 As shown, the first aspect of the present invention provides a method for predicting the dispatchable capacity of electric vehicles, comprising:

[0033] S1. Constructing a travel chain model for electric vehicles in a target area, and using a Markov chain to model the transfer probability of each electric vehicle in the target area based on the travel chain model to obtain a travel spatiotemporal model of each electric vehicle;

[0034] In one embodiment, constructing a travel chain model of electric vehicles in a target area includes:

[0035] Dividing the target area according to land use types to obtain a plurality of zones; the zones include residential areas, commercial areas and industrial areas;

[0036] Taking the residential area as the initial node and the end node, and the commercial area and the industrial area as the intermediate nodes, a travel chain model of each electric vehicle in the target area is constructed.

[0037] Specifically, the travel chain of electric vehicles can be divided into intercity travel chains and urban travel chains according to the size of the travel range. The intercity travel chain is generally for the purpose of tourism, business trips and visiting relatives, and the distance is longer than the urban travel chain. Therefore, the present invention focuses on the dispatchable capacity of electric vehicles in the city and only considers the urban travel chain. Then the target area can be a certain city or a certain district in a certain city, etc.

[0038] The present invention divides the target area according to the land use type to obtain a number of partitions; wherein the partitions include residential areas (such as high-density residential areas), commercial areas (such as shopping centers, office building concentrated areas) and industrial areas (such as factories, logistics parks, etc.). Therefore, the number of each partition in the target area is not unique, and can be one or more, and they are numbered.

[0039] Generally, urban travel chains can be divided into simple and complex chains based on the number of stops. However, this invention uses residential areas as the initial and final nodes, and commercial and / or industrial areas as intermediate nodes to construct travel chain models for each electric vehicle within the target area. Consequently, these travel chain models can be categorized into four types: 1. Residential-Industrial-Residential; 2. Residential-Commercial-Residential; 3. Residential-Industrial-Commercial-Residential; and 4. Residential-Commercial-Industrial-Residential. It should be noted that these four travel chain models describe the travel trajectories of electric vehicles throughout the day.

[0040] By dividing the target area according to land use type and constructing a travel chain model, the present invention can more carefully and accurately characterize the travel characteristics of each electric vehicle in the area, providing detailed basic data for subsequent scheduling judgments; the travel chain model construction method based on land use type and specific nodes (residential area, commercial area, industrial area) facilitates the classification and analysis of electric vehicle travel behavior in different types of areas and different travel destinations, helps to discover travel patterns and differences in different scenarios, and thus formulate more targeted scheduling measures.

[0041] In one embodiment, the transition probability of each electric vehicle in the target area is modeled using a Markov chain based on the travel chain model to obtain a travel spatiotemporal model of each electric vehicle, including:

[0042] Acquiring historical travel trajectory data of each of the electric vehicles in the target area, and dividing the historical travel trajectory data into a plurality of single travel chains;

[0043] Taking each node in each single trip chain as a state, traversing all single trip chains of each electric vehicle, extracting the number of transitions between each pair of adjacent states, and constructing a transition probability matrix for each electric vehicle;

[0044] Performing Laplace smoothing on each of the transfer probability matrices to obtain a smoothed transfer probability matrix of each of the electric vehicles;

[0045] The current position of each electric vehicle is obtained, and a travel space model of each electric vehicle is constructed according to each current position and each smooth transition probability matrix.

[0046] Specifically, modeling a single electric vehicle accurately describes its travel behavior in both time and space. Temporal characteristics include the time of each electric vehicle's first trip, the duration of its stay at the destination, the departure time of the next leg of its journey, and the charging time of each electric vehicle. Spatial characteristics include the purpose of each electric vehicle's trip, the length of its trip chain, and the probability of transitions between different destinations. Therefore, this paper constructs a spatiotemporal travel model for each electric vehicle from both temporal and spatial perspectives.

[0047] For the travel space characteristics of electric vehicles, the next destination selected by each electric vehicle is determined only by its current location. Therefore, the present invention uses a Markov chain model to model the transition probability of each electric vehicle between different destinations, thereby obtaining a travel space model for each electric vehicle. The modeling process specifically includes:

[0048] The historical travel trajectory data of each electric vehicle in the target area is obtained, such as GPS trajectory data (timestamp, latitude and longitude, speed, direction, etc.), charging records (charging time, location, power change, etc.), and user tags (vehicle ID, commuting time, etc.). Based on the partition type involved in the historical travel trajectory data, it is divided into several single travel chains. For example, if the historical operation trajectory of electric vehicle A is residential area H1-commercial area C3-industrial area I2-commercial area C5-residential area H1, then it can be divided into four single travel chains: residential area H1-commercial area C3, commercial area C3-industrial area I2, industrial area I2-commercial area C5, and commercial area C5-residential area H1.

[0049] Each node in the travel chain model is considered a state of the Markov chain. Still taking electric car A as an example, its states include residential area H1, commercial area C3, industrial area I2, and commercial area C5. Although electric car A's travel trajectory includes two commercial areas, these two commercial areas cannot be merged due to their different locations and travel processes in a single travel chain. Then, all single trip chains of each electric car are traversed, the number of transitions between each pair of adjacent states is counted, and the transition probability is calculated to obtain the transition probability matrix of each electric car.

[0050] Since each electric vehicle has not been to each partition of the target area during its historical travel, there may be some transitions that have never occurred and their probability is 0, which makes the constructed transition probability matrix unable to be generalized. Therefore, the present invention obtains the smoothed transition probability matrix of each electric vehicle by performing Laplace smoothing on the calculated transition probability matrix.

[0051] By obtaining the current position of each electric vehicle and using it as the starting state, combined with the smooth transition probability matrix, the travel space model of each electric vehicle can be constructed. The model can be expressed as a state transition graph, where nodes represent states (residential area, commercial area, industrial area, etc.) and edges represent the transition probability between states. Based on the current position of each electric vehicle and the transition probability matrix, the possible travel path of the electric vehicle in the future and the probability of reaching each state can be predicted, thereby constructing the travel space model of each electric vehicle.

[0052] Based on the historical trajectories and smoothed probability matrices of each electric vehicle, the present invention accurately predicts the user's future travel path, improving the reliability of route planning and charging recommendations. Combined with the real-time location data of each electric vehicle, the travel space model is dynamically adjusted to adapt to changing traffic conditions and user behavior preferences. Laplace smoothing effectively solves the zero-probability problem, enhancing the model's generalization ability for sparse data.

[0053] In one embodiment, the method of using a Markov chain to model the transition probability of each electric vehicle in the target area based on the travel chain model to obtain a travel spatiotemporal model of each electric vehicle further includes:

[0054] Segmenting the travel space model of each electric vehicle to obtain a plurality of travel segments;

[0055] quantifying the stay time of each electric vehicle in each zone in each travel itinerary and the travel time in the next travel itinerary by a dynamic time stay model;

[0056] At the end of each segment of the travel, the remaining power of each electric vehicle in the current segment of the travel and the distance to the next segment of the travel are obtained to quantify the charging anxiety coefficient of each electric vehicle in the current segment of the travel;

[0057] comparing the charging anxiety coefficient with a preset anxiety threshold, and when the charging anxiety coefficient exceeds the preset anxiety threshold, quantifying the charging time of each electric vehicle during the current travel segment based on the remaining power of each electric vehicle;

[0058] Constructing a travel time model for each electric vehicle based on the stay time of each electric vehicle in each zone in each travel itinerary, the travel time of each electric vehicle in the next travel itinerary, and the charging time of each electric vehicle in each travel itinerary;

[0059] The travel space models and the travel time models are integrated to obtain the travel space-time models of the electric vehicles.

[0060] Specifically, with residential areas as initial nodes and end nodes as limiting rules, the travel probabilities of multiple initial travel paths contained in the model are calculated based on the travel space model of each electric vehicle. If the travel probability of one and only one initial travel path is greater than a preset probability threshold, it means that the initial travel path includes the partition that the electric vehicle will pass through in the next 24 hours; the preset probability threshold is determined by historical data or manual experience and is not limited to a specific value here; if the travel probability of multiple or no initial travel paths of an electric vehicle is greater than the preset probability threshold, it means that the accuracy of the travel space model of the electric vehicle is insufficient, and the transfer probability of the electric vehicle in the target area needs to be remodeled until the probability threshold condition is met; and the initial travel paths that meet the threshold condition are segmented to obtain several travel itineraries. For example, if the initial travel route of electric vehicle A is residential area H1-industrial area I1-commercial area C1-residential area H1, and the travel probability of electric vehicle A only on this route meets the probability threshold, then it can be divided into travel route 1, residential area H1-industrial area I1, travel route 2, industrial area I1-commercial area C1, and travel route 3, commercial area C1-residential area H1.

[0061] For the travel time characteristics of electric vehicles, the first travel time satisfies the normal distribution:

[0062]

[0063] Where, f(t first ) is the probability density function of the first trip time of electric vehicles; σ is the variance of the first trip time of electric vehicles, which is 7.32; t first is the first trip time of electric vehicles; μ is the mean first trip time of electric vehicles, which is 1.34.

[0064] This paper uses a dynamic time dwell model to quantify the dwell time of each electric vehicle in each zone during different trips. It also introduces a regional traffic impact index τ, which reflects the degree of regional congestion. The value of this index ranges from [0, 1], and the closer it is to 1, the more congested the region. The index is specifically expressed as follows:

[0065]

[0066] Where q i (t) is the number of vehicles in the current area at time t; Q max The index indicates that the different carrying capacities of electric vehicles in different regions will also affect the residence time of electric vehicles.

[0067] The residence time in the residential area satisfies the exponential distribution, so the dynamic time residence model corresponding to the residential area is expressed as follows:

[0068]

[0069] Where, f(t s,res ) is the probability density function of the residence time of electric vehicles in residential areas; λ e is the reciprocal of the average stay time of electric vehicles without considering the impact of traffic; t s,res is the residence time of electric vehicles in residential areas.

[0070] The residence time in the commercial area satisfies the Weibull distribution, so the dynamic residence time model corresponding to the commercial area is expressed as follows:

[0071]

[0072] Where, f(t s,com ) is the probability density function of the residence time of electric vehicles in the commercial area; k w is the morphological parameter, k w >0;λ w is the proportionality coefficient, λ w >0;k w and λ w The value of is determined by fitting the historical residence time data of electric vehicle users in commercial areas, where the morphological parameter is used to control the shape of the residence time distribution to reflect the concentration trend of user behavior. The larger the value, the more concentrated the residence time. In this invention, k w Take 1.8; and the proportional coefficient is used to determine the benchmark scale of the residence time to reflect the length of the typical residence time. The larger the value, the longer the overall level of the user's stay here. In the present invention, λ w Take 2.2; t s,com is the dwelling time of electric vehicles in the commercial area.

[0073] The residence time in the industrial zone satisfies the normal distribution, so the dynamic time residence model corresponding to the industrial zone is expressed by the following formula:

[0074]

[0075] Where, f(t s,ind ) is the probability density function of the residence time of electric vehicles in the industrial area; ts,ind is the residence time of electric vehicles in the industrial area; μ g is the mean stay time of electric vehicles without considering traffic impact; σ g is the standard deviation of the electric vehicle’s residence time when traffic impact is not considered.

[0076] It should be noted that regardless of whether the electric vehicle is located in residential area H1 or residential area H2, it satisfies the exponential distribution in the residential area. Although the locations of the residential areas are different, the residence time of each electric vehicle in the same partition satisfies the dynamic time residence model of the corresponding partition.

[0077] When each electric vehicle arrives at the destination, the dwell time is generated according to the dynamic time dwell model of the corresponding partition. After the dwell time ends, the next trip begins. Therefore, the travel time (also known as the departure time) of the next trip can be expressed as:

[0078]

[0079] Where, t k+1 is the travel time of the electric vehicle in the next trip; K is the total number of trips; t s,k is the dwell time of the electric vehicle at the destination of the kth trip.

[0080] At the end of a trip, the remaining power of each electric vehicle in the current trip and the distance to the next trip are obtained, and the charging anxiety coefficient of each electric vehicle in the current trip is quantified. The charging anxiety coefficient can be expressed by the following formula:

[0081]

[0082] Where, ω c is the charging anxiety coefficient; a and Θ are psychological threshold parameters, where Θ varies from user to user, i.e., the remaining power of the electric vehicle when it needs to be charged. When the SOC is less than this value, the anxiety coefficient will increase significantly as the power decreases; a is used to control the rate of change of the anxiety coefficient near the value of Θ. The larger the value, the more obvious the change of the anxiety coefficient when it approaches the value of Θ, which is reflected in the rapid increase of anxiety. In the present invention, a is taken as 0.3; SOC is the remaining power of the electric vehicle during the current travel period; d i+1 is the number of kilometers of the next trip, that is, the distance between the current location and the target node of the next trip; r i,Z is the energy consumption per kilometer of electric vehicles; E s is the battery capacity.

[0083] When the remaining power of each electric vehicle is high and the remaining endurance is far greater than the distance of the next trip, the charging anxiety coefficient is small; when the remaining power of each electric vehicle is less than a certain value, the charging anxiety coefficient will rise rapidly, and when the remaining endurance of each electric vehicle is close to the distance of the next trip, the value of the charging anxiety coefficient approaches the maximum value, which is the maximum anxiety value.

[0084] The calculated charging anxiety coefficient is compared with a preset anxiety threshold (which can be determined based on the remaining power of each electric vehicle when charging based on historical running trajectory data, and the preset anxiety threshold corresponding to each electric vehicle is different). When the charging anxiety coefficient exceeds the preset anxiety threshold, it indicates that the electric vehicle needs to be charged at that location. In addition, a judgment can also be made based on the remaining power of the electric vehicle. If the remaining power of an electric vehicle is less than or equal to 20% of the total power, it indicates that the electric vehicle needs to be charged immediately in that zone. Then the charging time of the electric vehicle in the current travel segment can be expressed by the following formula:

[0085]

[0086] Where, t c The charging time of the electric vehicle during the current travel period; SOC cur is the SOC of the electric vehicle at the current moment; η c is the charging efficiency; P c is the charging power.

[0087] If it is determined that the charging anxiety coefficient of an electric vehicle does not exceed the preset anxiety threshold, or the remaining power is greater than 20% of the total power, it means that the electric vehicle has sufficient power and does not need to be charged in this section, then the charging time of the electric vehicle in the current travel period is 0.

[0088] The charging time of each electric vehicle in each trip segment is calculated and integrated with the dwell time of each segment in each trip segment and the travel time of the corresponding next trip segment to construct a travel time model for each electric vehicle. This model can be represented as a time series, where each element corresponds to the total time of a trip segment (dwell time + travel time + charging time). It should be noted that if a trip segment is the final trip, the travel time of the corresponding next trip segment is 0. Finally, the travel space model of each electric vehicle is integrated with the travel time model of the electric vehicle to obtain its travel space-time model. This model can also be represented as a space-time dataset that contains the travel information of the electric vehicle in different time and space, such as location, time, and power status.

[0089] The present invention can more carefully and accurately characterize the travel characteristics of electric vehicles in the target area, including the division of travel itineraries, stay time, travel time, charging demand, etc., through segmented processing of the electric vehicle travel space model, combined with the construction of a dynamic time residence model, a charging anxiety coefficient and a travel time model, to provide comprehensive and accurate data support for subsequent scheduling capacity prediction; by quantifying the charging anxiety coefficient and determining the charging time based on it, it helps to reasonably arrange the charging time and location of electric vehicles, avoid travel inconveniences caused by insufficient power, improve the convenience of use and user satisfaction of electric vehicles, and also facilitate the efficient use of charging facilities; the constructed travel time and space model can comprehensively consider the travel behavior of electric vehicles in space and time, thereby improving the accuracy of predictions of future electric vehicle travel.

[0090] S2. Constructing a transportation network model for the target area, and based on each of the travel spatiotemporal models and the transportation network model, using a multi-objective optimization algorithm to search for travel paths for each of the electric vehicles, to obtain a predicted travel path for each of the electric vehicles;

[0091] In one embodiment, constructing the traffic network model of the target area includes:

[0092] Acquire historical traffic flow data of the target area and historical charge and discharge data of the power distribution network in the target area;

[0093] Classifying the roads according to the historical traffic flow data to obtain primary roads and secondary roads, and calculating the unit driving energy consumption data of each electric vehicle on the primary roads and the secondary roads respectively;

[0094] Determining a plurality of centers based on the connection relationship between the primary road and the secondary road, and performing energy flow mapping between the distribution network and the target area based on the historical charging and discharging data to classify the centers into charging centers and non-charging centers;

[0095] A traffic network topology is constructed with the charging center and the non-charging center as nodes and the primary road and the secondary road as edges, and the traffic network model is generated by annotating the unit driving energy consumption data.

[0096] Specifically, the present invention obtains historical traffic flow data of the target area, such as traffic flow information in different time periods (peak period, off-peak period, off-peak period) and different roads (such as main roads, secondary roads, and branch roads), and obtains historical charging and discharging data of the distribution network in the target area from the distribution network's monitoring system or related database, including charging records of charging piles, such as charging time, charging power, charging location, etc.

[0097] Based on historical traffic flow data, certain traffic flow thresholds are set to classify roads. For example, roads with daily average traffic flow exceeding a certain threshold are classified as Class I roads, while roads with daily average traffic flow below the threshold are classified as Class II roads. Dynamic classification can also be adopted to consider changes in traffic flow over different time periods. Road grades can also be divided by speed limit, such as roads with speed limits exceeding 50 km / h are Class I roads, while roads with speed limits not exceeding 50 km / h are Class II roads. (This is just an example; speed limits can also be set based on historical data or manual experience.) However, it should be noted that there are two or more Class I and Class II roads in the target area.

[0098] The traffic status index of each road level is quantified based on the driving speed of electric vehicles when there is no traffic congestion or the traffic volume is very low. The quantification process is expressed by the following formula:

[0099]

[0100] Where V i,Z (t) is the driving speed on the Z-level road at time t; is the vehicle speed when there is no traffic congestion or the traffic volume is very low on the Z-level road, where Z is 1 (for a first-level road) or 2 (for a second-level road); i,Z (t) is the traffic flow of the Z-level road at time t, where t∈N and 1≤t≤96; T i,Z is the maximum traffic capacity of Z-class road; β is the traffic state index of Z-class road; m and n are road parameters, when the road is a first-class road, m = 1.726 and n = 3.150, when the road is a second-class road, m = 2.076 and n = 2.870; α is a fixed parameter 3.

[0101] Then, the unit driving energy consumption data of each electric vehicle on different levels of roads can be expressed by the following formula:

[0102]

[0103] Where, The energy consumption per kilometer of electric vehicles traveling on primary roads and secondary roads respectively; is the average speed of electric vehicles on Z-level roads.

[0104] According to the connection relationship between primary roads and secondary roads, clustering algorithms in graph theory (such as K-means algorithm, DBSCAN algorithm, etc.) are used to determine several centers. These centers can be areas with relatively concentrated traffic flow or key connection points in the road network; and based on historical charging and discharging data, the distribution network and each center in the target area are mapped to energy flow, and the energy interaction between each center and the distribution network is analyzed. That is, based on historical charging and discharging data, the number of charging facilities, charging power, discharged power and other information around each center are counted, and a certain threshold is set according to historical data or manual experience, and the centers with a large number of charging facilities, large charging power and small discharged power are classified as charging centers; otherwise, they are classified as non-charging centers.

[0105] Finally, the charging center and the non-charging center are used as nodes, and the primary roads and secondary roads are used as edges to construct the topology of the transportation network. The topology can be represented by a graph data structure, where the nodes store the relevant information of the center (such as location, type, etc.), and the edges store the relevant information of the roads (such as road grade, length, etc.). It should be noted that the nodes here are not the same as the nodes in the travel itinerary. The nodes here are named for the convenience of constructing the transportation network topology and are used to store the relevant information of the center, while the nodes in the travel itinerary are the destinations in the corresponding partitions. The calculated unit driving energy consumption data is then marked on the edges of the transportation network topology, that is, the unit driving energy consumption of electric vehicles on each road is marked on the edge of the road, and a transportation network model containing multiple information such as traffic flow, road grade, energy consumption, and charging facility distribution can be generated. The fusion structure of the distribution network and the transportation network model is as follows: Figure 2 As shown, Figure 2 The gray-white half on the left is the residential area, the dark gray in the middle is the commercial area, and the light gray on the right is the industrial area. The solid lines in the traffic network model represent the primary roads, and the dotted lines connecting the centers are the secondary roads. Figure 2 It can be seen that there is energy flow between the 23 nodes in the distribution network and the center 7 in the transportation network, which means that the center 7 is the charging center. Similarly, centers 10, 15, 19, 26, and 30 are all charging centers. Of course, Figure 2 This is only an example diagram and does not indicate that the size of the target area, the number of zones, charging centers, etc. are distributed as shown in the diagram.

[0106] By comprehensively considering the historical traffic flow data of the target area and the historical charging and discharging data of the distribution network, the present invention can more comprehensively and accurately reflect the traffic and energy conditions in the area, thereby constructing a more realistic traffic network model and providing a reliable basis for subsequent electric vehicle travel route planning; grading roads and calculating unit driving energy consumption data help to consider energy consumption factors when planning routes, plan more energy-efficient travel routes for electric vehicles, reduce travel costs, and improve energy utilization efficiency; by determining charging centers and non-charging centers, the distribution of charging facilities in the target area can be clarified, providing a basis for the reasonable layout of charging facilities, improving the utilization rate of charging facilities, and alleviating the charging anxiety of electric vehicles.

[0107] In one embodiment, the multi-objective optimization algorithm is used to search for the travel path of each electric vehicle based on each of the travel time-space models and the traffic network model to obtain the predicted travel path of each electric vehicle, including:

[0108] Obtaining the current position of each electric vehicle and inputting it into a travel space-time model of each electric vehicle to predict the travel process of each electric vehicle to obtain a predicted intermediate node, a predicted terminal node, and a predicted travel time;

[0109] Inputting the predicted intermediate node and the predicted terminal node into the traffic network model, and constructing an adjacency matrix of the current position, the predicted intermediate node, and the predicted terminal node;

[0110] Based on the target model of each electric vehicle, a multi-objective optimization algorithm is used to traverse all adjacency matrices of each electric vehicle to determine the target charging center and target non-charging center connecting the current position, predicted intermediate node and predicted terminal node of each electric vehicle, and generate a predicted travel path for each electric vehicle; wherein,

[0111] The process of constructing the target model of each electric vehicle is as follows:

[0112] The predicted total consumption time and the predicted driving energy consumption corresponding to the predicted travel path of each electric vehicle are weighted by an adaptive weight coefficient determined based on real-time time, and a target model of each electric vehicle is constructed with the goal of minimizing the weighted processing result; wherein the predicted total consumption time is the sum of the predicted driving time and the predicted travel time; the predicted driving time and the predicted driving energy consumption are determined by the traffic network model and the travel time-space model.

[0113] Specifically, the current position of each electric vehicle is obtained in real time and input into the travel space-time model of the corresponding electric vehicle, so that the travel space-time model that combines historical travel data, spatial distribution, time patterns and other information simulates the driving process of each electric vehicle, and obtains the predicted intermediate node of each electric vehicle (which can be an industrial area or a commercial area, and the number can be one or more), the predicted terminal node (the residential area corresponding to the electric vehicle) and the predicted travel time; wherein the predicted travel time is the travel time model of the electric vehicle based on the current position, the predicted intermediate node, the charging time in multiple travel segments corresponding to the predicted terminal node, the stay time in each partition contained in each travel segment, and the travel time corresponding to the next travel segment.

[0114] The intermediate nodes and terminal nodes predicted by each electric vehicle are used as input information and imported into the traffic network model. Based on the traffic network model, an adjacency matrix of the current position, predicted intermediate nodes and predicted terminal nodes of each electric vehicle is constructed. It is assumed that the set S contains the optimal road set that has been obtained, which can minimize the target. The predicted total time consumed and the predicted driving energy consumption corresponding to the predicted travel path of each electric vehicle are weighted based on the adaptive weight coefficient determined based on real time, and the goal is to minimize the weighted processing result. The predicted driving time of each electric vehicle and its predicted travel time are added together to obtain the predicted total time consumed by the electric vehicle. The predicted driving time of each electric vehicle in each travel segment of the predicted travel path can be calculated based on the traffic network model and the travel time-space model, and the predicted driving energy consumption can be calculated based on the traffic network model. The present invention introduces an adaptive weight coefficient determined based on real time, which is expressed by the following formula:

[0115]

[0116] Where γ is the adaptive weight coefficient and t is the time.

[0117] There is only one starting point in the initial set S, which is the current position. The centers (charging centers or non-charging centers) contained in other adjacency matrices are placed in the to-be-traversed set K to ensure that the electric vehicle can complete the predicted optimal path and its corresponding number of charging times as a constraint. A multi-objective optimization algorithm (such as NSGA-II, MOEA / D, etc.) is used to find a charging center or non-charging center that meets the constraints from the to-be-traversed set K as the target charging center or target non-charging center and add it to the set S. For the adjacent centers of the target charging / non-charging center, it is determined whether there is a better path from the starting point through the target center to the adjacent center of the target center. If so, the set K is updated. According to this method, the target charging center and target non-charging center connecting the current position, the predicted intermediate node and the predicted terminal node are iteratively searched until the predicted terminal node is added to the set S, thereby generating the predicted travel path for each electric vehicle.

[0118] Based on trip chain and traffic network modeling, the starting and ending nodes of electric vehicles can be determined. The initial travel time and starting SOC of electric vehicles are generated through a probabilistic model. Then, based on the electric vehicle modeling and algorithm, a path search is performed to obtain the spatiotemporal transfer path of each electric vehicle throughout the next 24 hours. Assuming that electric vehicles all travel at a constant speed, the dwell time at each destination and the determination of whether they are charging can be used to determine their location and remaining SOC in each time period. This provides a foundation for further research on the dispatchable capacity of electric vehicles.

[0119] In addition, the present invention may also use a multi-objective optimization algorithm (such as a genetic algorithm, a particle swarm optimization algorithm, a non-dominated sorting genetic algorithm, etc.) to search for charging centers and non-charging centers connecting each electric vehicle to its current location, predicted intermediate nodes, and predicted terminal nodes from the transportation network model; and calculate the predicted total time and predicted driving energy consumption of the initial path formed by combining the searched charging centers and non-charging centers with the current location, predicted intermediate nodes, and predicted terminal nodes based on the transportation network model and each travel spatiotemporal model; then use the calculated adaptive weight coefficient as the weight of the predicted total time, and use the difference between 1 and the adaptive weight coefficient as the weight of the predicted driving energy consumption to perform a weighted processing on the two, and construct a target model for each electric vehicle with the goal of minimizing the weighted processing result; using the target model of each electric vehicle as the optimization goal, and ensuring that the electric vehicle can complete the journey without charging, after charging once at a charging center, or after charging at multiple charging centers as a constraint, the initial path is iteratively updated, and during the search process, the distribution of charging centers and non-charging centers is considered, and based on the remaining power and charging demand of the electric vehicle, a decision is made as to whether to pass through a charging center in the path and charge. For example, if the remaining power of an electric vehicle is low and there is a suitable charging center on the route, the algorithm may choose to charge at the charging center to avoid insufficient power; ultimately, a set of predicted travel paths that meet the requirements of the target model is generated. These paths achieve a good balance in terms of time and energy consumption, and may include only charging centers or only non-charging centers, or both. Among them, the introduction of adaptive weight coefficients makes the algorithm tend to the optimal time when the time is close to the peak period, such as 7-9 am, 5-7 pm, etc.; otherwise, the algorithm tends to the optimal energy consumption. It should also be noted that the present invention only uses the multi-objective optimization algorithm and does not improve it, so the specific process of the multi-objective optimization algorithm is not expanded.

[0120] The present invention comprehensively considers the travel time-space model and traffic network model of electric vehicles, incorporates multiple objectives such as time and energy consumption into the path search process, and can plan a travel path for electric vehicles that is more in line with actual needs and more optimized, thereby improving travel efficiency and reducing energy consumption; adopts an adaptive weight coefficient determined based on real-time time, so that path planning can be dynamically adjusted according to the actual situation of different time periods, better adapt to changes in traffic conditions and energy demand, and improve the real-time and practicality of path planning; predicts the driving process through the travel time-space model, and calculates the driving time and energy consumption in combination with the traffic network model, which can more accurately predict the travel situation of electric vehicles and provide a reliable basis for subsequent scheduling and decision-making; considering energy consumption factors in path planning helps guide electric vehicles to choose more energy-saving routes, reduce energy waste, and promote the rational use and sustainable development of energy.

[0121] S3. Determine the time cost of introducing a time urgency factor based on each predicted travel path, and combine it with the battery degradation cost, discharge cost, and mood loss cost of each electric vehicle to establish a dispatch willingness function for each electric vehicle;

[0122] In one embodiment, determining the time cost of introducing a time urgency factor based on each predicted travel path includes:

[0123] determining a travel type of each of the electric vehicles based on each of the predicted travel paths;

[0124] Calculating a time urgency factor for each of the electric vehicles based on the predicted total consumed time in each of the predicted travel paths and each of the travel types;

[0125] The time cost of each electric vehicle is determined according to each time urgency factor and each predicted travel path.

[0126] Specifically, the present invention determines the travel type of each electric vehicle (that is, which travel chain model it belongs to) based on the predicted travel path of each electric vehicle, and calculates the time urgency factor of the electric vehicle based on the predicted total consumed time in the predicted travel path of the electric vehicle and its travel type. The time urgency factor is expressed by the following formula:

[0127] σ1=α ij ,i∈{t 0-4 ,t 4-8 ,t 8-12 ,t 12-16 ,t 16-20 ,t 20-24},j∈{A,B,C,D}

[0128] Where σ1 is the time urgency factor, which is affected by both time and trip chain type, and 24 hours is divided into 6 time periods; 0-4 , t 4-8 , t 8-12 , t 12-16 , t 8-12 , t 20-24 These are the six time periods for electric vehicles in a day; A, B, C, and D are four travel chain models. The time urgency factor ranges from 0 to 1, with values ​​closer to 1 indicating a more urgent event. For example, during rush hour, users may face greater commuting pressure, leading to a higher value for this factor. This paper, based on expert experience and historical data analysis, determines a time urgency factor matrix, as shown in the following table:

[0129] Table 1 Time urgency factor matrix

[0130]

[0131]

[0132] From Table 1, we can see that a certain electric car 0-4 If an electric vehicle is only in travel chain model B during a period, its time urgency factor in that period is 0.2, and the time urgency factor corresponding to each electric vehicle can be determined based on this. It should be understood that an electric vehicle may be in one or more different travel chain models in a period, or stay in a travel chain model for one or more periods. In this case, the corresponding time urgency factor should be determined in the corresponding travel chain model or period in accordance with Table 1. The time cost of each electric vehicle can then be determined based on the time urgency factor of each electric vehicle and its predicted travel path, which is expressed as follows:

[0133]

[0134] Where cost time is the time cost of electric vehicles; t cs is the time required to reach the charging station; t discharge is the discharge time; is the time-optimal path, that is, the predicted travel path; is a continuous segment in the time-optimal path; for Length; SOC i,exp is the expected SOC of the electric vehicle; η dis is the discharge efficiency; P dis is the discharge power; C time is the user's unit time value, used to convert cost time The unit of time is converted from yuan, and the average wage of the place where the electric car is located can be taken.

[0135] By combining the time cost of each electric vehicle with its battery degradation cost, discharge cost, and mood loss cost, the dispatch willingness function of each electric vehicle can be established, which is expressed as follows:

[0136] F=cost time +cost dep +cost out +cost emo

[0137]

[0138] cost emo =ε ij ,i∈{W sun ,W rain ,W cloud ,W extreme},j∈{t workday ,tweekend}

[0139] Where F is the scheduling willingness function; cost dep is the battery degradation cost; cost out is the discharge cost, which is also the discharge electricity price in the current period. In this invention, the time-of-use electricity price is adopted, that is, the discharge electricity price is related to time and can be determined by the electricity price information of the charging station or the power supply company; cost emo K is the mood loss cost; i is the linear relationship coefficient, which is -0.015625; is the total amount of electricity discharged by the electric vehicle within time t; is the battery replacement cost; W sun 、W rain 、W cloud 、W extreme There are four weather types, namely sunny, rainy, cloudy and extreme weather; workday , t weekend The mood loss cost value reflects the willingness of electric vehicle users to participate in scheduling. For example, when faced with extreme weather, higher costs are incurred, and electric vehicle owners are willing to participate in scheduling. This invention determines the mood loss factor matrix based on expert experience and historical data analysis, as shown in the following table:

[0140] Table 2 Mood loss factor matrix

[0141]

[0142] As can be seen from Table 2, if a weekday is sunny all day, the mood loss factor of that day is 2, which can be used to determine the time urgency factor corresponding to each electric vehicle. It should be understood that the weather on a certain day is not constant and may experience one or more different weather types. Therefore, it is necessary to combine the corresponding weather types in Table 2 to determine the mood loss factor for the corresponding time period.

[0143] By introducing a time urgency factor that takes into account the travel type and the predicted total time consumption, the present invention can more accurately evaluate the time cost of an electric vehicle under different predicted travel routes, so that the calculation of the time cost is more in line with actual travel needs; accurate time cost evaluation helps electric vehicle users or scheduling systems make more reasonable travel decisions; the time urgency factor that takes into account the travel type can adapt to the needs of different travel scenarios, making the calculation of time cost more targeted and improving the adaptability of the system to different travel scenarios; based on the scheduling willingness function that takes into account conventional costs, time costs with the introduction of time urgency factors, and mood loss costs, judgment can be made, and the scheduling status of the electric vehicle can be flexibly adjusted according to actual conditions, and different scheduling needs can be responded to in a timely manner, thereby improving the flexibility and adaptability of the entire electric vehicle scheduling system.

[0144] S4. Sending a real-time dispatch request to each of the electric vehicles, so that each of the electric vehicles performs a dispatch willingness judgment based on the real-time dispatch request using each of the dispatch willingness functions to obtain the electric vehicle dispatch capacity within the target area;

[0145] Specifically, a real-time dispatch request is sent to all electric vehicles in the target area so that they can obtain the dispatchable capacity of electric vehicles in the target area at the current moment after making a willingness judgment based on their own dispatch willingness function.

[0146] The present invention can classify electric vehicles at a certain moment into three categories according to their behavior; Category A is electric vehicles that are parked at a node but not connected to a charging pile. Users of this type of electric vehicles are considered busy users and do not participate in scheduling; Category B is electric vehicles that are in motion. Users of this type of electric vehicles will make willingness judgments based on their scheduling willingness functions; Category C is electric vehicles that are parked at a node and connected to a charging pile. Users of this type of electric vehicles also need to make willingness judgments. It should be noted that the time required for Category C electric vehicles to reach the charging station is 0, so they are more inclined to participate in scheduling.

[0147] For willingness judgment, the power grid will give an acceptable price based on the active power demand at that moment. This price will be compared with the quantitative price calculated by each electric vehicle based on its dispatch willingness function. If the quantitative price is less than the acceptable price given by the power grid, the electric vehicle will participate in the dispatch, which is expressed by the following formula:

[0148]

[0149] Where, Participation is the scheduling participation index. When its value is 1, it means willingness to participate in scheduling, and when its value is 0, it means refusal to participate in scheduling; C g The electricity price is given to the grid.

[0150] The present invention integrates scattered electric vehicle resources through real-time dispatch requests to improve the flexibility and responsiveness of the power grid; based on the willingness function, it respects user choices, balances power grid demand and user interests, increases the dispatch participation rate, and thus improves the prediction results of the dispatchable capacity of electric vehicles.

[0151] In an embodiment of the present application, a method for predicting the dispatchable capacity of electric vehicles is designed to improve the prediction results of the dispatchable capacity of electric vehicles. The method constructs a travel chain model of electric vehicles in a target area to fully present the travel activity sequence of electric vehicles in the target area, and uses a Markov chain to model the transition probability of each electric vehicle in the target area to obtain a travel spatiotemporal model of each electric vehicle to characterize the travel possibilities of electric vehicles at different times and spaces in the future, providing forward-looking information for traffic management and scheduling. By constructing a traffic network model of the target area, a multi-objective optimization algorithm is used to search the travel paths of each electric vehicle to obtain a predicted travel path for each electric vehicle, which helps to improve travel efficiency, reduce traffic congestion, and reduce energy consumption. A dispatch willingness function is established that comprehensively considers multiple factors such as time cost, battery degradation cost, discharge cost, and mood loss cost, which can more comprehensively reflect the willingness of electric vehicles to participate in scheduling. The method sends dispatch requests in real time and controls electric vehicles to make judgments based on the dispatch willingness function, which can promptly respond to real-time demand changes in the traffic system. This not only improves the prediction results of the dispatchable capacity of electric vehicles, but also helps to reasonably arrange subsequent dispatch tasks, optimize resource allocation, and improve the operating efficiency of the entire electric vehicle dispatch system.

[0152] It should be noted that although the steps in the above flowchart are shown in sequence as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders.

[0153] In another embodiment, Figure 3 As shown, the second aspect of the present invention provides an electric vehicle dispatchable capacity prediction system, comprising:

[0154] A model building module 101 is used to build a travel chain model of electric vehicles in a target area, and based on the travel chain model, use a Markov chain to model the transfer probability of each electric vehicle in the target area to obtain a travel spatiotemporal model of each electric vehicle;

[0155] a path prediction module 102 for constructing a traffic network model of the target area and, based on each of the travel spatiotemporal models and the traffic network model, searching for a travel path for each of the electric vehicles using a multi-objective optimization algorithm to obtain a predicted travel path for each of the electric vehicles;

[0156] A function establishment module 103 is configured to determine a time cost of introducing a time urgency factor based on each predicted travel path, and to establish a dispatch willingness function for each electric vehicle by combining the time cost of introducing a time urgency factor with the time cost of introducing a time urgency factor and ...

[0157] The capacity determination module 104 is configured to send a real-time dispatch request to each of the electric vehicles, so that each of the electric vehicles makes a dispatch willingness judgment based on the real-time dispatch request using each of the dispatch willingness functions to obtain the dispatch capacity of the electric vehicles in the target area.

[0158] It should be noted that the various modules in the above-mentioned electric vehicle dispatchable capacity prediction system can be implemented in whole or in part through software, hardware, or a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above-mentioned modules. For the specific definition of an electric vehicle dispatchable capacity prediction system, please refer to the definition of an electric vehicle dispatchable capacity prediction method above. The two have the same functions and effects and will not be repeated here.

[0159] A third aspect of the present invention provides an electronic device, comprising:

[0160] processor, memory, and bus;

[0161] The bus is used to connect the processor and the memory;

[0162] The memory is used to store operation instructions;

[0163] The processor is used to call the operation instruction, and the executable instruction enables the processor to perform operations corresponding to the electric vehicle dispatchable capacity prediction method shown in the first aspect of the present application.

[0164] In an alternative embodiment, an electronic device is provided, such as Figure 4 As shown, Figure 4 The electronic device 5000 shown includes: a processor 5001 and a memory 5003. The processor 5001 and the memory 5003 are connected, for example, via a bus 5002. Optionally, the electronic device 5000 may further include a transceiver 5004. It should be noted that in actual applications, the number of transceivers 5004 is not limited to one, and the structure of the electronic device 5000 does not constitute a limitation on the embodiments of the present application.

[0165] Processor 5001 may be a CPU, a general-purpose processor, a DSP, an ASIC, an FPGA, or other programmable logic device, a transistor logic device, a hardware component, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 5001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0166] The bus 5002 may include a path for transmitting information between the above components. The bus 5002 may be a PCI bus or an EISA bus, etc. The bus 5002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0167] The memory 5003 may be a ROM or other type of static storage device that can store static information and instructions, a RAM or other type of dynamic storage device that can store information and instructions, or an EEPROM, a CD-ROM or other optical disk storage, an optical disc storage (including a compact disc, a laser disc, an optical disc, a digital versatile disc, a Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to these.

[0168] The memory 5003 is used to store application code for executing the solution of the present application, and the execution is controlled by the processor 5001. The processor 5001 is used to execute the application code stored in the memory 5003 to implement the content shown in any of the above method embodiments.

[0169] Among them, electronic devices include but are not limited to: mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc.

[0170] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the method for predicting the dispatchable capacity of an electric vehicle as shown in the first aspect of the present application is implemented.

[0171] Another embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer-readable storage medium is run on a computer, the computer can execute the corresponding contents of the aforementioned method embodiments.

[0172] In addition, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above method when executed by a processor.

[0173] In summary, the present invention relates to the field of electric vehicles and discloses a method, system, device and medium for predicting the dispatchable capacity of electric vehicles. The method constructs a travel chain model of electric vehicles in a target area and constructs a travel space-time model of each electric vehicle in combination with a Markov chain; constructs a traffic network model of the target area to be combined with each travel space-time model, and adopts a multi-objective optimization algorithm to predict the travel path of each electric vehicle; determines the time cost of introducing a time urgency factor based on the travel path prediction result of each electric vehicle, and establishes a dispatch willingness function in combination with its battery degradation cost, discharge cost and mood loss cost, so that each electric vehicle responds to the received real-time dispatch request based on its own dispatch willingness function, thereby obtaining the dispatch capacity of electric vehicles in the target area; and realizes accurate prediction of the dispatchable capacity of electric vehicles in the target area.

[0174] Each embodiment in this specification is described in a progressive manner, and the same or similar parts of each embodiment can be directly referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. It should be noted that the various technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the various technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0175] The above-described embodiments merely represent several preferred implementations of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent. It should be noted that a person skilled in the art could make several improvements and substitutions without departing from the technical principles of the present invention, and these improvements and substitutions should also be considered within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be based on the scope of protection of the claims.

Claims

1. A method for predicting the dispatchable capacity of electric vehicles, characterized in that: include: Constructing a travel chain model for electric vehicles in a target area, and using a Markov chain to model the transfer probability of each electric vehicle in the target area based on the travel chain model to obtain a travel spatiotemporal model of each electric vehicle; Constructing a transportation network model of the target area, and based on each of the travel spatiotemporal models and the transportation network model, using a multi-objective optimization algorithm to search for travel paths of each of the electric vehicles to obtain predicted travel paths of each of the electric vehicles; Determining a time cost of introducing a time urgency factor based on each predicted travel path, and combining this with a battery degradation cost, a discharge cost, and a mood loss cost of each electric vehicle to establish a dispatch willingness function for each electric vehicle; A real-time dispatch request is sent to each of the electric vehicles, so that each of the electric vehicles performs a dispatch willingness judgment based on the real-time dispatch request using each of the dispatch willingness functions to obtain the electric vehicle dispatch capacity within the target area.

2. The method for predicting the dispatchable capacity of electric vehicles according to claim 1, characterized in that: The constructing of a travel chain model for electric vehicles in a target area includes: Dividing the target area according to land use types to obtain a plurality of zones; the zones include residential areas, commercial areas and industrial areas; Taking the residential area as the initial node and the end node, and the commercial area and the industrial area as the intermediate nodes, a travel chain model of each electric vehicle in the target area is constructed.

3. The method for predicting the dispatchable capacity of electric vehicles according to claim 1, characterized in that: The method of using a Markov chain to model the transfer probability of each electric vehicle in the target area based on the travel chain model to obtain a travel spatiotemporal model of each electric vehicle includes: Acquiring historical travel trajectory data of each of the electric vehicles in the target area, and dividing the historical travel trajectory data into a plurality of single travel chains; Taking each node in each single trip chain as a state, traversing all single trip chains of each electric vehicle, extracting the number of transitions between each pair of adjacent states, and constructing a transition probability matrix for each electric vehicle; Performing Laplace smoothing on each of the transfer probability matrices to obtain a smoothed transfer probability matrix of each of the electric vehicles; The current position of each electric vehicle is obtained, and a travel space model of each electric vehicle is constructed according to each current position and each smooth transition probability matrix.

4. The method for predicting the dispatchable capacity of electric vehicles according to claim 3, characterized in that: The method further includes: modeling the transfer probability of each electric vehicle in the target area using a Markov chain based on the travel chain model to obtain a travel spatiotemporal model of each electric vehicle; Segmenting the travel space model of each electric vehicle to obtain a plurality of travel segments; quantifying the stay time of each electric vehicle in each zone in each travel itinerary and the travel time in the next travel itinerary by a dynamic time stay model; At the end of each segment of the travel, the remaining power of each electric vehicle in the current segment of the travel and the distance to the next segment of the travel are obtained to quantify the charging anxiety coefficient of each electric vehicle in the current segment of the travel; comparing the charging anxiety coefficient with a preset anxiety threshold, and when the charging anxiety coefficient exceeds the preset anxiety threshold, quantifying the charging time of each electric vehicle during the current travel segment based on the remaining power of each electric vehicle; Constructing a travel time model for each electric vehicle based on the stay time of each electric vehicle in each zone in each travel itinerary, the travel time of each electric vehicle in the next travel itinerary, and the charging time of each electric vehicle in each travel itinerary; The travel space models and the travel time models are integrated to obtain the travel space-time models of the electric vehicles.

5. The method for predicting the dispatchable capacity of electric vehicles according to claim 1, characterized in that: The constructing of the traffic network model of the target area includes: Acquire historical traffic flow data of the target area and historical charge and discharge data of the power distribution network in the target area; Classifying the roads according to the historical traffic flow data to obtain primary roads and secondary roads, and calculating the unit driving energy consumption data of each electric vehicle on the primary roads and the secondary roads respectively; Determining a plurality of centers based on the connection relationship between the primary road and the secondary road, and performing energy flow mapping between the distribution network and the target area based on the historical charging and discharging data to classify the centers into charging centers and non-charging centers; A traffic network topology is constructed with the charging center and the non-charging center as nodes and the primary road and the secondary road as edges, and the traffic network model is generated by annotating the unit driving energy consumption data.

6. The method for predicting the dispatchable capacity of electric vehicles according to claim 5, characterized in that: The method of searching the travel paths of the electric vehicles using a multi-objective optimization algorithm based on the travel time-space model and the traffic network model to obtain the predicted travel paths of the electric vehicles includes: Obtaining the current position of each electric vehicle and inputting it into a travel space-time model of each electric vehicle to predict the travel process of each electric vehicle to obtain a predicted intermediate node, a predicted terminal node, and a predicted travel time; Inputting the predicted intermediate node and the predicted terminal node into the traffic network model, and constructing an adjacency matrix of the current position, the predicted intermediate node, and the predicted terminal node; Based on the target model of each electric vehicle, a multi-objective optimization algorithm is used to traverse all adjacency matrices of each electric vehicle to determine the target charging center and target non-charging center connecting the current position, predicted intermediate node and predicted terminal node of each electric vehicle, and generate a predicted travel path for each electric vehicle; wherein, The process of constructing the target model of each electric vehicle is as follows: The predicted total consumption time and the predicted driving energy consumption corresponding to the predicted travel path of each electric vehicle are weighted by an adaptive weight coefficient determined based on real-time time, and a target model of each electric vehicle is constructed with the goal of minimizing the weighted processing result; wherein the predicted total consumption time is the sum of the predicted driving time and the predicted travel time; the predicted driving time and the predicted driving energy consumption are determined by the traffic network model and the travel time-space model.

7. The method for predicting the dispatchable capacity of electric vehicles according to claim 6, characterized in that: The determining of the time cost of introducing the time urgency factor based on each predicted travel path includes: determining a travel type of each of the electric vehicles based on each of the predicted travel paths; Calculating a time urgency factor for each of the electric vehicles based on the predicted total consumed time in each of the predicted travel paths and each of the travel types; The time cost of each electric vehicle is determined according to each time urgency factor and each predicted travel path.

8. An electric vehicle dispatchable capacity prediction system, characterized in that: include: a model building module, configured to build a travel chain model for electric vehicles in a target area, and to model the transfer probability of each electric vehicle in the target area using a Markov chain based on the travel chain model, thereby obtaining a travel spatiotemporal model for each electric vehicle; a path prediction module, configured to construct a traffic network model of the target area, and based on each of the travel spatiotemporal models and the traffic network model, use a multi-objective optimization algorithm to search for a travel path of each of the electric vehicles to obtain a predicted travel path of each of the electric vehicles; a function establishment module for determining a time cost of introducing a time urgency factor based on each predicted travel path, and combining this with a battery degradation cost, a discharge cost, and a mood loss cost of each electric vehicle to establish a dispatch willingness function for each electric vehicle; The capacity determination module is used to send a real-time scheduling request to each of the electric vehicles, so that each of the electric vehicles performs a scheduling willingness judgment based on the real-time scheduling request through each of the scheduling willingness functions to obtain the electric vehicle scheduling capacity in the target area.

9. An electronic device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the method for predicting the dispatchable capacity of an electric vehicle according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored computer program, wherein when the device where the computer-readable storage medium is located executes the computer program, the electric vehicle dispatchable capacity prediction method according to any one of claims 1 to 7 is implemented.

Images