Charging load prediction method, device and equipment of electric vehicle driving mode and medium

Abstract

The invention relates to the technical field of electric vehicle charging load prediction, in particular to a charging load prediction method, device and equipment for an electric vehicle driving mode and a medium, and the method comprises the steps: constructing a road traffic impedance model and an electric vehicle power consumption model; dividing the electric vehicles into private vehicles and online hailed vehicles according to travel rules of the electric vehicles, and constructing an electric vehicle travel behavior model; constructing a charging demand model of the electric vehicle; various parameters of the electric vehicle are generated according to probability distribution by adopting a Monte Carlo method, and a charging load prediction process of the electric vehicle is provided; the charging load is predicted by comprehensively considering the real-time change of traffic flow and the travel behavior modes of different types of electric vehicles, the travel behavior modes of different types of electric vehicles may cause the charging load at a specific time and place to reach a peak value, the possible time and place are found, and the charging load is predicted. And meanwhile, the accuracy of overall prediction can be improved for guiding the charging behaviors of different types of electric vehicles.

Description

Technical Field

[0001] The present invention relates to the technical field of electric vehicle charging load prediction, and in particular to a charging load prediction method, device, equipment and medium for an electric vehicle driving mode. Background Art

[0002] In the context of countries actively promoting the application of new energy, electric vehicles can absorb new energy well and gradually replace traditional fuel vehicles that produce pollution with their advantages of no pollution and low cost. The number of electric vehicles is growing explosively every year, and with it comes a huge demand for charging load. These increasing charging loads will have a significant impact on the stability and economy of the power grid. The charging behavior of electric vehicles is random in space and time. Therefore, it is of great significance to accurately model the charging behavior of electric vehicles and make accurate charging load predictions by considering all influencing factors.

[0003] Existing studies have focused on obtaining a more accurate model of EV travel behavior by more accurately depicting EV traffic flow and fully considering the impact of external factors on traffic flow. However, in addition to the study of traffic flow, the charging mode of EVs and the impact of different vehicle travel patterns on charging demand should also be studied in more depth.

[0004] The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present invention, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art already known to those skilled in the art. Summary of the invention

[0005] The present invention provides a charging load prediction method, device, equipment and medium for an electric vehicle driving mode, thereby effectively solving the problems in the background technology.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is: a charging load prediction method for an electric vehicle driving mode, comprising the following steps:

[0007] S10: Obtaining road section traffic flow information, traffic light signal information, road topology information, and road fixed parameter information, and constructing a road traffic impedance model and an electric vehicle power consumption model;

[0008] S20: Based on the road traffic impedance model and the electric vehicle power consumption model, private car travel chain information, online car-hailing state transition probability information, and electric vehicle fixed parameter information are obtained, and electric vehicles are divided into private cars and online car-hailing vehicles according to the travel rules of electric vehicles, and travel behavior models of electric vehicles are constructed respectively;

[0009] S30: Based on the electric vehicle travel behavior model, obtain the charging station location information, the charging station charging power information, and the electric vehicle trajectory information to build a charging demand model for the electric vehicle;

[0010] S40: Based on the charging demand model of the electric vehicle, various parameters of the electric vehicle are generated according to probability distribution using the Monte Carlo method, and a charging load prediction process of the electric vehicle is proposed.

[0011] Furthermore, in step S10, a road traffic impedance model is constructed, including:

[0012] R ij (t) = N ij (t)+H ij (t);

[0013] Where t is each time period, t = 1, 2, ..., T, where T represents a total time period; R ij (t) is the road impedance; N ij (t) is the node impedance; H ij (t) is the section impedance.

[0014] Furthermore, the road section impedance includes:

[0015]

[0016] In the formula, H ij (t) is the road section impedance; S is the road congestion level; T 0 is the road travel time when the road traffic flow is zero; α and β are impedance influencing factors; D is the real-time road traffic density; D v0 is the road traffic density when the vehicle speed is zero.

[0017] Furthermore, the node impedance model includes:

[0018]

[0019] Where N ij (t) is the node impedance; T cyc is the time of one cycle of the traffic light; λ is the ratio of green traffic lights to one cycle of the traffic light; q is the ratio of electric vehicles arriving at the road section; y is a Boolean variable, which takes 1 when the electric vehicle goes straight or turns left at the intersection, and takes 0 when turning right; T rec is the psychological delay time of electric vehicles at intersections, which is related to the driver's reaction speed; S is the degree of road congestion.

[0020] Furthermore, in step S10, the electric vehicle power consumption model includes:

[0021]

[0022] Where P use V is the power consumption of electric vehicles per kilometer; l is the average speed of the electric vehicle on road section l. The speeds corresponding to different levels of roads and different road conditions are different; a, b and c are power consumption coefficients.

[0023] Furthermore, in step S20, based on the road traffic impedance model and the electric vehicle power consumption model, private car travel chain information, online car-hailing state transition probability information, and electric vehicle fixed parameter information are obtained, and electric vehicles are divided into private cars and online car-hailing vehicles according to the travel rules of electric vehicles, and electric vehicle travel behavior models are constructed respectively. The steps include:

[0024] S21: The travel behavior model of private cars is constructed based on a travel chain with residential areas as the starting and ending points. The first trip time and stay time in the private car travel chain follow a normal distribution;

[0025] S22: The travel behavior model of online ride-hailing vehicles is described by the travel state transfer matrix. The first trip time and the number of trips of online ride-hailing vehicles follow a normal distribution.

[0026] S23: Based on the travel behavior model of the private car and the travel behavior model of the online car-hailing service, a real-time Dijkstra dynamic path search algorithm is used for path optimization. The Dijkstra dynamic path search algorithm screens and compares urban road impedances, with the goal of searching for the shortest road travel time and constructing a total time consumption model between the departure node and the destination node.

[0027] Furthermore, in step S23, a total time consumption model between the departure node and the destination node is constructed, including:

[0028]

[0029] In the formula, S od is the total time between the departure node and the destination node; R ij (t) is the road impedance; E is the road node set; d is the actual driving trajectory of the electric vehicle; x ij It is a Boolean variable. When its value is 1, it means that this road is included in the driving path of the electric vehicle.

[0030] Furthermore, in step S30, based on the electric vehicle travel behavior model, the charging station location information, the charging station charging power information, and the electric vehicle trajectory information are acquired to construct a charging demand model for the electric vehicle, and the steps include:

[0031] S31: Constructing a charging demand model for private cars:

[0032] When a private car arrives at the destination, it is determined whether its remaining power is lower than the preset power threshold and whether the stay time exceeds the time threshold for the vehicle to be slowly charged to the ideal power; if so, the vehicle is slowly charged at the current node, otherwise it is fast charged or not charged;

[0033] When a private car arrives at a non-destination node, if the remaining power is lower than the preset power threshold, it will be fast charged at the current node, otherwise it will not be charged; when a private car returns to a residential area, if the remaining power is lower than the preset target value, it will be slow charged, otherwise it will not be charged;

[0034] S32: Constructing a charging demand model for online ride-hailing vehicles:

[0035] When the remaining power of an online-hailing vehicle is lower than a preset power threshold during driving, the vehicle will immediately perform fast charging at the node of the current driving route; if the remaining power is not lower than the threshold, no charging operation will be performed;

[0036] S33: Based on the charging demand model of the private car and the charging demand model of the online car-hailing service, respectively determining whether the electric vehicle needs to be charged and the charging type, and calculating the charging time of the electric vehicle;

[0037] S34: Calculate the charging load at the electric vehicle node based on the charging time of the electric vehicle and the charging power at the node.

[0038] Further, in step S34, the model for calculating the charging load at the electric vehicle node includes:

[0039]

[0040] Where P t,k is the charging load demand of node k in period t; P i,t,k Charging load demand of electric vehicle i at node k and time period t; N is the total number of electric vehicles.

[0041] Further, in step S40, based on the charging demand model of the electric vehicle, the Monte Carlo method is used to generate various parameters of the electric vehicle according to the probability distribution, and a load forecasting process of the electric vehicle is proposed, the steps comprising:

[0042] S41: Import all electric vehicle numbers, road network structure, and initial traffic flow data, generate random samples for electric vehicles based on the Monte Carlo method, and generate corresponding driving parameters and charging parameters for each sample;

[0043] S42: Based on the driving parameters and charging parameters, extract the parameters of the i-th electric vehicle, the private car based on the travel chain model, and the online car-hailing based on the state transfer matrix, extract the first travel time and destination functional area of ​​the vehicle; use the real-time Dijkstra algorithm to minimize the driving time, plan the vehicle's driving path, and dynamically update the vehicle's driving characteristic parameters;

[0044] S43: Based on the driving path and driving characteristic parameters of the vehicle, according to the remaining power of the vehicle, the destination and the itinerary, determine whether charging is required, charge the vehicle according to the corresponding charging standard, record the charging time of the vehicle, and update the remaining power; if charging is not required, update the time parameter;

[0045] S44: Based on the updated remaining power and the updated time parameters, determine whether the private car has reached the end of the residential area or whether the online car-hailing service has completed its maximum number of trips; if so, end the load simulation of the private car or the online car-hailing service; if not, return to step S42 to continue simulating the travel behavior of the next electric car; by dividing the functional types of different power consumption areas, accumulating the charging power of electric vehicles at each node in the area, and obtaining the load of the regional distribution node, the spatiotemporal prediction of the overall charging load is completed.

[0046] The present invention also includes a charging load prediction device for an electric vehicle driving mode, using the above method, including:

[0047] A data acquisition unit is used to acquire road section traffic flow information, traffic light signal information, road topology information, road fixed parameter information, and construct a road traffic impedance model and an electric vehicle power consumption model;

[0048] A travel behavior model building unit is used to obtain private car travel chain information, online car-hailing state transition probability information, and electric vehicle fixed parameter information based on the road traffic impedance model and the electric vehicle power consumption model, and divide electric vehicles into private cars and online car-hailing vehicles according to the travel rules of electric vehicles, and build travel behavior models for electric vehicles respectively;

[0049] A charging demand model building unit, used to acquire charging station location information, charging station charging power information, and electric vehicle trajectory information based on the electric vehicle travel behavior model, and build a charging demand model for the electric vehicle;

[0050] The load prediction module is used to generate various parameters of the electric vehicle according to probability distribution based on the charging demand model of the electric vehicle by using the Monte Carlo method, and propose a charging load prediction process for the electric vehicle.

[0051] The present invention also includes a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method described above is implemented.

[0052] The present invention also includes a storage medium on which a computer program is stored. When the computer program is executed by a processor, the method described above is implemented.

[0053] The beneficial effects of the present invention are:

[0054] By comprehensively considering the real-time changes in traffic flow and the travel behavior patterns of different types of electric vehicles to predict the charging load, the travel behavior patterns of different types of electric vehicles may cause the charging load to reach a peak at a specific time and place, and the possible time and place can be found. At the same time, the present invention can guide the charging behavior of different types of electric vehicles in the future. Compared with the prediction method of the unified model, the technical solution is more in line with the actual situation, thereby improving the accuracy of the overall prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0056] Figure 1 A flow chart of a charging load prediction method for an electric vehicle driving mode;

[0057] Figure 2 A flowchart of a charging load prediction method for an electric vehicle driving mode;

[0058] Figure 3 Flowchart for charging load prediction for electric vehicle driving mode;

[0059] Figure 4 A traffic node network diagram of an embodiment;

[0060] Figure 5 It is a structural schematic diagram of a charging load prediction device for an electric vehicle driving mode;

[0061] Figure 6 A schematic diagram of the structure of a computer device. DETAILED DESCRIPTION

[0062] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0063] like Figures 1 to 4 As shown: A charging load prediction method for an electric vehicle driving mode comprises the following steps:

[0064] S10: Obtaining road section traffic flow information, traffic light signal information, road topology information, and road fixed parameter information, and constructing a road traffic impedance model and an electric vehicle power consumption model;

[0065] S20: Based on the road traffic impedance model and the electric vehicle power consumption model, the private car travel chain information, the online car-hailing state transition probability information, and the electric vehicle fixed parameter information are obtained. According to the travel rules of electric vehicles, electric vehicles are divided into private cars and online car-hailing vehicles, and travel behavior models of electric vehicles are constructed respectively;

[0066] S30: Based on the electric vehicle travel behavior model, the charging station location information, the charging power information of the charging station, and the electric vehicle trajectory information are obtained to build a charging demand model for the electric vehicle;

[0067] S40: Based on the charging demand model of electric vehicles, the Monte Carlo method is used to generate various parameters of electric vehicles according to probability distribution, and a charging load prediction process for electric vehicles is proposed.

[0068] The present invention predicts the charging load by comprehensively considering the real-time changes in traffic flow and the travel behavior patterns of different types of electric vehicles. The travel behavior patterns of different types of electric vehicles may cause the charging load to reach a peak at a specific time and place, and the possible time and place can be found. At the same time, the present invention can lay a foundation for subsequent research on guiding the charging behavior of different types of electric vehicles, thereby reducing the impact of electric vehicle charging behavior on the power grid.

[0069] According to the travel patterns of electric vehicles, the vehicles are divided into two categories: private cars and online-hailing vehicles, and their travel behavior models are constructed separately. The driving patterns and charging requirements of private cars and online-hailing vehicles are significantly different. Distinguishing these two types of vehicles can improve the accuracy of charging demand prediction for different types of vehicles. Compared with the prediction method of a unified model, this technical solution is more in line with the actual situation, thereby improving the accuracy of the overall prediction.

[0070] By using the Monte Carlo method to randomly sample the parameters of electric vehicles and perform load forecasting based on the charging demand model, this probability distribution-based simulation method can effectively deal with the randomness in electric vehicle charging behavior, and thus more accurately reflect the changes in charging demand for electric vehicles in space and time. The introduction of the Monte Carlo method has significantly improved the ability to handle uncertainty factors in complex systems.

[0071] Since the temporal and spatial randomness of electric vehicle charging behavior will have a significant impact on the stability and economy of the power grid, the charging load prediction method of this technical solution can identify possible peak charging periods and locations in advance. Through these prediction results, it can help power grid operators better manage power resources, allocate power loads, avoid power grid overload problems, and improve the stability and operating efficiency of the power grid.

[0072] As a preference of the above embodiment, in step S10, building a road traffic impedance model includes:

[0073] R ij (t) = N ij (t)+H ij (t);

[0074] Where t is each time period, t = 1, 2, ..., T, where T represents a total time period; R ij (t) is the total travel time between traffic node i and traffic node j, that is, road impedance; N ij (t) is the total travel time of the road intersection between traffic node i and traffic node j, that is, the node impedance; H ij (t) is the total travel time of the road section between traffic node i and traffic node j, that is, the road section impedance.

[0075] By constructing a detailed road traffic impedance model, this method can take into account multiple factors such as traffic flow, signal light control, road section impedance, node impedance, etc., which helps to more accurately predict the travel time of electric vehicles in different road sections and time periods, thereby providing a basis for the time prediction of charging behavior;

[0076] The section impedance includes:

[0077]

[0078] In the formula, H ij (t) is the total travel time of the road section between traffic node i and traffic node j, that is, the road section impedance; S is the road saturation that characterizes the traffic conditions of the road section, that is, the road congestion level; T 0 is the road travel time when the road traffic flow is zero; α and β are impedance influencing factors; D is the real-time road traffic density; D v0is the road traffic density when the vehicle speed is zero.

[0079] The model can accurately estimate the driving time of electric vehicles on various road sections based on different road conditions, time periods and traffic flows. Accurate driving time estimation is crucial for subsequent energy consumption calculations and charging demand predictions.

[0080] Among them, the node impedance model includes:

[0081]

[0082] Where N ij (t) is the total travel time of the road intersection between traffic node i and traffic node j, that is, the node impedance; T cyc is the time of one cycle of the traffic light; λ is the ratio of green traffic lights to one cycle of the traffic light; q is the ratio of electric vehicles arriving at the road section; y is a Boolean variable, which takes 1 when the electric vehicle goes straight or turns left at the intersection, and takes 0 when turning right; T rec is the psychological delay time of electric vehicles at intersections, which is related to the driver's reaction speed; S is the road saturation that characterizes the traffic conditions of the road section, that is, the degree of road congestion.

[0083] By modeling node impedance and link impedance separately, the method can analyze in detail the impact of different parts of the traffic network on the passage of electric vehicles. Node impedance mainly reflects the delay at the intersection, while link impedance takes into account the congestion and traffic capacity of the link itself. This detailed decomposition can more accurately simulate the impact of different road conditions on the travel of electric vehicles.

[0084] In this embodiment, in step S10, the electric vehicle power consumption model includes:

[0085]

[0086] Where P use V is the power consumption of electric vehicles per kilometer; l is the average speed of the electric vehicle on road section l. The speeds corresponding to different levels of roads and different road conditions are different; a, b and c are power consumption coefficients.

[0087] The model combines the power consumption of an electric vehicle with its driving speed V on a specific road section. l In association, by considering the speed differences on different road sections, the energy consumption of electric vehicles under various traffic conditions can be accurately simulated, which helps to dynamically adjust the energy consumption prediction of electric vehicles according to factors such as road grade and traffic flow, thereby improving accuracy.

[0088] As a preferred embodiment of the above embodiment, in step S20, based on the road traffic impedance model and the electric vehicle power consumption model, private car travel chain information, online car-hailing state transition probability information, and electric vehicle fixed parameter information are obtained, and according to the travel rules of electric vehicles, electric vehicles are divided into private cars and online car-hailing vehicles, and electric vehicle travel behavior models are constructed respectively. The steps include:

[0089] S21: Private cars tend to start from home and have a fixed itinerary. They will not leave immediately after arriving at the destination. Therefore, the travel behavior model of private cars is based on a travel chain with residential areas as the starting point and the end point. The first trip time and the stay time in the private car travel chain follow a normal distribution. This reflects the travel mode that private cars tend to start from home and have a fixed itinerary.

[0090] S22: The travel behavior model of online ride-hailing vehicles is described by the travel state transfer matrix. The first trip time and the number of trips of online ride-hailing vehicles follow a normal distribution. After each trip, online ride-hailing vehicles will not stay at the destination but go to the next destination. Therefore, their travel mode is more frequent than that of private cars. The travel state transfer matrix is ​​expressed as:

[0091]

[0092] In the formula, p t is the state transfer matrix of electric vehicles in each period of 24 hours a day; p mn (m,n=1,2,3) is the transfer probability of electric vehicles between various functional areas; R, W and O represent other areas such as residential areas, work areas and commercial areas respectively;

[0093] S23: Based on the travel behavior model of private cars and the travel behavior model of online ride-hailing vehicles, the real-time Dijkstra dynamic path search algorithm is used for path optimization. The Dijkstra dynamic path search algorithm screens and compares the impedance of urban roads, with the goal of searching for the shortest road travel time and constructing a total time consumption model between the departure node and the destination node.

[0094] The normal distribution is used to simulate the travel chain information of private cars and the travel times of online ride-hailing vehicles respectively. This method can better capture the randomness of travel time and frequency, making the charging demand prediction closer to the actual situation and improving the accuracy of the model. Through the real-time Dijkstra algorithm, the traffic impedance model of urban roads is screened and compared to optimize the driving path of electric vehicles, and the shortest driving time is taken as the goal, so that electric vehicles can choose the optimal route to reduce energy consumption during driving, thereby more accurately predicting the power consumption and charging needs of electric vehicles.

[0095] In step S23, a total time consumption model between the departure node and the destination node is constructed, including:

[0096]

[0097] In the formula, S od is the total time between the departure node and the destination node; R ij (t) is the total travel time between traffic node i and traffic node j, that is, road impedance; E is the set of road nodes; d is the actual driving trajectory of the electric vehicle; x ij It is a Boolean variable. When its value is 1, it means that this road is included in the driving path of the electric vehicle.

[0098] Road resistance R ij The introduction of (t) enables the model to dynamically capture the changes in road impedance at different times, especially the differences in travel time caused by changes in traffic flow and traffic lights. The model can adjust the driving path of electric vehicles in real time to ensure that the estimated driving time is closer to the actual situation. Accurate total time estimation directly affects the energy consumption prediction and charging demand of electric vehicles. If the electric vehicle travels for a long time, its battery will consume more and the charging demand will increase accordingly. Through accurate time estimation, the model can provide more reliable data support for charging load prediction.

[0099] As a preferred embodiment of the above embodiment, in step S30, based on the electric vehicle travel behavior model, the charging station location information, the charging station charging power information, and the electric vehicle trajectory information are obtained to construct a charging demand model for the electric vehicle, and the steps include:

[0100] S31: Constructing a charging demand model for private cars:

[0101] When a private car arrives at the destination, it is determined whether its remaining power is lower than the preset power threshold and whether the stay time exceeds the time threshold for the vehicle to be slowly charged to the ideal power; if so, the vehicle is slowly charged at the current node, otherwise it is fast charged or not charged;

[0102] When a private car arrives at a non-destination node, if the remaining power is lower than the preset power threshold, it will be fast charged at the current node, otherwise it will not be charged; when a private car returns to a residential area, if the remaining power is lower than the preset target value, it will be slow charged, otherwise it will not be charged;

[0103] S32: Constructing a charging demand model for online ride-hailing vehicles:

[0104] When the remaining power of an online-hailing vehicle is lower than a preset power threshold during driving, the vehicle will immediately perform fast charging at the node of the current driving route; if the remaining power is not lower than the threshold, no charging operation will be performed;

[0105] S33: Based on the charging demand model of the private car and the charging demand model of the online car-hailing service, respectively determine whether the electric vehicle is charged and the charging type, and calculate the charging time of the electric vehicle;

[0106] S34: Calculate the charging load at the electric vehicle node based on the charging time of the electric vehicle and the charging power at the node.

[0107] By carefully distinguishing the charging demand behavior models of private cars and online ride-hailing vehicles, this technical solution can more accurately predict the charging needs of different types of electric vehicles. The charging decisions of private cars depend on multiple factors such as remaining power and stay time, while online ride-hailing vehicles pay more attention to the demand for fast charging. Through this differentiated processing, the accuracy of the prediction can be greatly improved and the waste of charging resources can be avoided. By calculating the charging time of electric vehicles and combining the charging power information at the node, the charging load of each node can be accurately calculated. The precise load forecast of this solution helps power grid managers to estimate the electricity demand in different regions in advance, prevent overload or insufficient power supply to the power grid during charging peak periods, and is beneficial to the stability and economic management of the power grid.

[0108] In this embodiment, in step S34, the model of the charging load at the electric vehicle node is calculated, including:

[0109]

[0110] Where P t,k is the charging load demand of node k in period t; P i,t,k Charging load demand of electric vehicle i at node k and time period t; N is the total number of electric vehicles.

[0111] The model accumulates the individual charging demands of each electric vehicle to accurately calculate the total charging load of each node in a specific time period, providing grid managers with an accurate basis for load forecasting and effectively avoiding grid overload or resource waste.

[0112] As a preferred embodiment of the above, in step S40, based on the charging demand model of the electric vehicle, the Monte Carlo method is used to generate various parameters of the electric vehicle according to the probability distribution, and a load prediction process of the electric vehicle is proposed, such as Figure 3 As shown, the steps include:

[0113] S41: Import all electric vehicle numbers, road network structure, and initial traffic flow data, generate random samples for electric vehicles based on the Monte Carlo method, and generate corresponding driving parameters and charging parameters for each sample; this step is the starting point of load forecasting, imports basic data and generates random samples to ensure that the driving and charging parameters of each vehicle are random;

[0114] S42: Based on driving parameters and charging parameters, extract the parameters of the i-th electric vehicle. For private cars, the first travel time and destination functional area are extracted based on the travel chain model, and for online car-hailing, the state transfer matrix. Use the real-time Dijkstra algorithm to plan the vehicle's driving path with the goal of minimizing the driving time, and dynamically update the vehicle's driving characteristic parameters. This step reasonably plans the driving path based on the vehicle's driving status, road network structure, and charging needs, which is the basis for subsequent judgment of the vehicle's charging needs.

[0115] S43: Based on the vehicle's driving path and driving characteristic parameters, according to the vehicle's remaining power, destination and itinerary, determine whether charging is required, charge the vehicle according to the corresponding charging standard, record the vehicle's charging time, and update its remaining power; if charging is not required, update the time parameter; this step is the core of charging load prediction, and provides key data for calculating load by determining whether charging is required and updating parameters;

[0116] S44: Based on updating its remaining power and updating time parameters, determine whether the private car has reached the end of the residential area or whether the online car-hailing service has completed its maximum number of trips; if so, end the load simulation of the private car or the online car-hailing service; if not, return to step S42 to continue simulating the travel behavior of the next electric car; by dividing the functional types of different power consumption areas, accumulating the charging power of electric vehicles at each node in the area, and obtaining the load of the regional distribution node, the spatiotemporal prediction of the overall charging load is completed.

[0117] By using the Monte Carlo method to generate random samples, the randomness of vehicle travel is taken into account, making the simulation closer to reality. The travel time, path, and charging demand of vehicles are random, which enables this method to better reflect the behavior of electric vehicles and load fluctuations in different situations;

[0118] The real-time Dijkstra algorithm is used to dynamically plan the shortest path for vehicles to ensure that the vehicles travel along the optimal path. This dynamic path planning combined with actual traffic conditions and grid demand can more accurately predict vehicle travel time and charging demand, which helps to improve the accuracy of electric vehicle load prediction;

[0119] By judging the remaining power of the vehicle, the destination and the itinerary, it dynamically decides whether to charge, and selects fast or slow charging according to the different charging needs of the vehicle. This detailed judgment mechanism can optimize the use of charging resources and avoid overcharging or waste of resources;

[0120] This solution is not only applicable to private cars, but also to online ride-hailing vehicles. Through the travel chain model of private cars and the state transfer matrix of online ride-hailing vehicles, the travel behaviors of different vehicles are constructed respectively. The travel modes and charging requirements of different vehicles are personalized to meet the prediction needs in different application scenarios.

[0121] By calculating the spatiotemporal distribution of charging loads in different power consumption areas, it can provide support for regional power grid dispatching, accurately predict the load of electric vehicles at different times and nodes, and help to rationally allocate power grid resources and avoid power grid overload caused by excessive charging demand;

[0122] By judging whether the vehicle has completed the trip (private cars return to the residential area, online ride-hailing vehicles complete the maximum number of trips), if not, it returns to S42 to continue simulating the next stage of travel behavior. The multi-round iterative mechanism ensures that the complete travel behavior of the vehicle is fully simulated, thereby improving the integrity of the load forecast.

[0123] like Figure 4 As shown, it is a traffic node network diagram of the embodiment, R indicates that the functional area attribute of this grid is a residential area, W indicates that the functional area attribute of this grid is a work area, and O indicates that the functional area attribute of this grid is a commercial area or other areas. Each traffic node is equipped with a charging station, the slow charging power of electric vehicles is 3.5kw / h, and the fast charging power of electric vehicles is 37.5kw / h. Each charging station has fast charging and slow charging piles, and the roads between traffic nodes are all two-way traffic.

[0124] The present invention also includes a charging load prediction device for an electric vehicle driving mode, using the above method, such as Figure 5 As shown, including:

[0125] A data acquisition unit is used to acquire road section traffic flow information, traffic light signal information, road topology information, road fixed parameter information, and construct a road traffic impedance model and an electric vehicle power consumption model;

[0126] A travel behavior model building unit is used to obtain private car travel chain information, online car-hailing state transition probability information, and electric vehicle fixed parameter information based on the road traffic impedance model and the electric vehicle power consumption model, and divide electric vehicles into private cars and online car-hailing vehicles according to the travel rules of electric vehicles, and build travel behavior models for electric vehicles respectively;

[0127] A charging demand model building unit is used to obtain the location information of the charging station, the charging power information of the charging station, and the trajectory information of the electric vehicle based on the travel behavior model of the electric vehicle, and to build a charging demand model for the electric vehicle;

[0128] The load forecasting module is used to generate various parameters of electric vehicles according to probability distribution based on the charging demand model of electric vehicles, and proposes a charging load forecasting process for electric vehicles.

[0129] See also Figure 6A computer device 400 provided in an embodiment of the present application includes: a processor 410 and a memory 420, wherein the memory 420 stores a computer program executable by the processor 410, and when the computer program is executed by the processor 410, the above method is executed.

[0130] The embodiment of the present application further provides a storage medium 430 on which a computer program is stored. When the computer program is run by the processor 410, the above method is executed.

[0131] Among them, the storage medium 430 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable red-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, disk or optical disk.

[0132] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. "Multiple" means two or more, unless otherwise clearly and specifically defined.

[0133] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like 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, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0134] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0135] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0136] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0137] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0138] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0139] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present invention. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A charging load prediction method for an electric vehicle driving mode, characterized in that: The steps include: S10: Obtaining road section traffic flow information, traffic light signal information, road topology information, and road fixed parameter information, and constructing a road traffic impedance model and an electric vehicle power consumption model; S20: Based on the road traffic impedance model and the electric vehicle power consumption model, private car travel chain information, online car-hailing state transition probability information, and electric vehicle fixed parameter information are obtained, and electric vehicles are divided into private cars and online car-hailing vehicles according to the travel rules of electric vehicles, and travel behavior models of electric vehicles are constructed respectively; S30: Based on the electric vehicle travel behavior model, obtain the charging station location information, the charging station charging power information, and the electric vehicle trajectory information to build a charging demand model for the electric vehicle; S40: Based on the charging demand model of the electric vehicle, various parameters of the electric vehicle are generated according to probability distribution using the Monte Carlo method, and a charging load prediction process of the electric vehicle is proposed.

2. The charging load prediction method for electric vehicle driving mode according to claim 1, characterized in that: In step S10, a road traffic impedance model is constructed, including: Where t is each time period, t = 1, 2, ..., T, where T represents a total time period; R ij (t) is the road impedance; N ij (t) is the node impedance; H ij (t) is the section impedance.

3. The charging load prediction method for electric vehicle driving mode according to claim 2, characterized in that: The road section impedance includes: In the formula, H ij (t) is the road impedance; S is the road congestion level; T0 is the road travel time when the road traffic flow is zero; α and β are the impedance influencing factors; D is the real-time road traffic density; D v0 is the road traffic density when the vehicle speed is zero.

4. The charging load prediction method for electric vehicle driving mode according to claim 2, characterized in that: The node impedance model includes: Where N ij (t) is the node impedance; T cyc is the time of one cycle of the traffic light; λ is the ratio of green traffic lights to one cycle of the traffic light; q is the ratio of electric vehicles arriving at the road section; y is a Boolean variable, which takes 1 when the electric vehicle goes straight or turns left at the intersection, and takes 0 when turning right; T rec is the psychological delay time of electric vehicles at intersections, which is related to the driver's reaction speed; S is the degree of road congestion.

5. The charging load prediction method for electric vehicle driving mode according to claim 1, characterized in that: In step S10, the electric vehicle power consumption model includes: Where P use V is the power consumption of electric vehicles per kilometer; l is the average speed of the electric vehicle on road section l. The speeds corresponding to different levels of roads and different road conditions are different; a, b and c are power consumption coefficients.

6. The charging load prediction method for electric vehicle driving mode according to claim 1, characterized in that: In step S20, based on the road traffic impedance model and the electric vehicle power consumption model, private car travel chain information, online car-hailing state transition probability information, and electric vehicle fixed parameter information are obtained, and electric vehicles are divided into private cars and online car-hailing vehicles according to the travel rules of electric vehicles, and electric vehicle travel behavior models are constructed respectively. The steps include: S21: The travel behavior model of private cars is constructed based on a travel chain with residential areas as the starting and ending points. The first trip time and stay time in the private car travel chain follow a normal distribution; S22: The travel behavior model of online ride-hailing vehicles is described by the travel state transfer matrix. The first trip time and the number of trips of online ride-hailing vehicles follow a normal distribution. S23: Based on the travel behavior model of the private car and the travel behavior model of the online car-hailing service, a real-time Dijkstra dynamic path search algorithm is used for path optimization. The Dijkstra dynamic path search algorithm screens and compares urban road impedances, with the goal of searching for the shortest road travel time and constructing a total time consumption model between the departure node and the destination node.

7. The charging load prediction method for electric vehicle driving mode according to claim 6, characterized in that: In step S23, a total time consumption model between the departure node and the destination node is constructed, including: In the formula, S od is the total time between the departure node and the destination node; R ij (t) is the road impedance; E is the road node set; d is the actual driving trajectory of the electric vehicle; x ij It is a Boolean variable. When its value is 1, it means that this road is included in the driving path of the electric vehicle.

8. The charging load prediction method for electric vehicle driving mode according to claim 1, characterized in that: In step S30, based on the electric vehicle travel behavior model, the charging station location information, the charging station charging power information, and the electric vehicle trajectory information are acquired to construct a charging demand model for the electric vehicle, and the steps include: S31: Constructing a charging demand model for private cars: When a private car arrives at the destination, it is determined whether its remaining power is lower than the preset power threshold and whether the stay time exceeds the time threshold for the vehicle to be slowly charged to the ideal power; if so, the vehicle is slowly charged at the current node, otherwise it is fast charged or not charged; When a private car arrives at a non-destination node, if the remaining power is lower than the preset power threshold, it will be fast charged at the current node, otherwise it will not be charged; when a private car returns to a residential area, if the remaining power is lower than the preset target value, it will be slow charged, otherwise it will not be charged; S32: Constructing a charging demand model for online ride-hailing vehicles: When the remaining power of an online-hailing vehicle is lower than a preset power threshold during driving, the vehicle will immediately perform fast charging at the node of the current driving route; if the remaining power is not lower than the threshold, no charging operation will be performed; S33: Based on the charging demand model of the private car and the charging demand model of the online car-hailing service, respectively determining whether the electric vehicle needs to be charged and the charging type, and calculating the charging time of the electric vehicle; S34: Calculate the charging load at the electric vehicle node based on the charging time of the electric vehicle and the charging power at the node.

9. The charging load prediction method for electric vehicle driving mode according to claim 8, characterized in that: In step S34, the model for calculating the charging load at the electric vehicle node includes: Where P t,k is the charging load demand of node k in period t; P i,t,k Charging load demand of electric vehicle i at node k and time period t; N is the total number of electric vehicles.

10. The charging load prediction method of the electric vehicle driving mode according to claim 1, characterized in that: In step S40, based on the charging demand model of the electric vehicle, the Monte Carlo method is used to generate various parameters of the electric vehicle according to the probability distribution, and a load forecasting process of the electric vehicle is proposed, and the steps include: S41: Import all electric vehicle numbers, road network structure, and initial traffic flow data, generate random samples for electric vehicles based on the Monte Carlo method, and generate corresponding driving parameters and charging parameters for each sample; S42: Based on the driving parameters and charging parameters, extract the parameters of the i-th electric vehicle, the private car based on the travel chain model, and the online car-hailing based on the state transfer matrix, extract the first travel time and destination functional area of ​​the vehicle; use the real-time Dijkstra algorithm to minimize the driving time, plan the vehicle's driving path, and dynamically update the vehicle's driving characteristic parameters; S43: Based on the driving path and driving characteristic parameters of the vehicle, according to the remaining power of the vehicle, the destination and the itinerary, determine whether charging is required, charge the vehicle according to the corresponding charging standard, record the charging time of the vehicle, and update the remaining power; if charging is not required, update the time parameter; S44: Based on the updated remaining power and the updated time parameters, determine whether the private car has reached the end of the residential area or whether the online car-hailing service has completed its maximum number of trips; if so, end the load simulation of the private car or the online car-hailing service; if not, return to step S42 to continue simulating the travel behavior of the next electric car; by dividing the functional types of different power consumption areas, accumulating the charging power of electric vehicles at each node in the area, and obtaining the load of the regional distribution node, the spatiotemporal prediction of the overall charging load is completed.

11. A charging load prediction device for an electric vehicle driving mode, characterized in that: Use of the method according to any one of claims 1 to 10, comprising: A data acquisition unit is used to acquire road section traffic flow information, traffic light signal information, road topology information, road fixed parameter information, and construct a road traffic impedance model and an electric vehicle power consumption model; A travel behavior model building unit is used to obtain private car travel chain information, online car-hailing state transition probability information, and electric vehicle fixed parameter information based on the road traffic impedance model and the electric vehicle power consumption model, and divide electric vehicles into private cars and online car-hailing vehicles according to the travel rules of electric vehicles, and build travel behavior models for electric vehicles respectively; A charging demand model building unit, used to acquire charging station location information, charging station charging power information, and electric vehicle trajectory information based on the electric vehicle travel behavior model, and build a charging demand model for the electric vehicle; The load prediction module is used to generate various parameters of the electric vehicle according to probability distribution based on the charging demand model of the electric vehicle by using the Monte Carlo method, and propose a charging load prediction process for the electric vehicle.

12. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 10 is implemented.

13. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 10 is implemented.

Images