Mountain city charging load probability prediction system and method

Abstract

The application provides a mountain city charging load probability prediction system and method, the system comprises: a road side camera module that collects vehicle pictures of passing vehicles; a road side GPS positioning module that obtains real-time positions of the passing vehicles; a road side temperature acquisition module that acquires real-time environmental temperatures of environments in which the passing vehicles are located; a cloud platform data center that stores charging station position data, road data of the mountain city, historical charging information, historical air conditioner use data, unit mileage driving power consumption parameters, vehicle-mounted equipment power consumption parameters, and battery charging power parameters; a cloud processing platform that determines driver mileage anxiety of the electric vehicle according to the above information, calculates travel power consumption, equipment power consumption, charging probability, and shortest time consumption of the electric vehicle to reach a target charging station, determines a charging load value according to the battery charging power parameters and charging mode data of the target charging station, and generates a charging load prediction result of the electric vehicle.

Description

TECHNICAL FIELD

[0001] The present application relates to the electric vehicle charging technical field, especially to a mountain city electric vehicle charging load probability prediction system and a mountain city electric vehicle charging load probability prediction method. BACKGROUND

[0002] Electric vehicle charging load prediction is the basis for analyzing the influence of electric vehicle access on power grid, power distribution network planning and control operation, electric vehicle and power grid two-way interaction, and electric vehicle and other energy, transportation, and other system coordination research. Due to the randomness of electric vehicle charging behavior in time and space, the prediction of charging load and probability involves very complex influencing factors, and different consideration angles will form different load prediction models and results. In order to ensure the normal and reliable operation of urban power grid, the future charging load and probability of electric vehicle should be accurately predicted, and an adaptive scheme for coordinated development of power grid and electric vehicle should be formed, so as to more effectively promote the popularization and application of new energy electric vehicle.

[0003] At present, the load and probability prediction method of electric vehicle includes power system short-term load prediction method, Monte Carlo simulation method and other new electric vehicle load prediction methods. However, from the current research method, the influence of driver's "mileage anxiety" and mountain city road characteristics on load is not considered, which leads to inaccurate calculation of electric vehicle power consumption, and further leads to inaccurate prediction of electric vehicle charging probability and load. SUMMARY

[0004] Therefore, the present application provides a mountain city charging load probability prediction system and a mountain city charging load probability prediction method, which combines the objective reality of vehicle remaining power with the subjective will of vehicle owner's mileage anxiety degree, so that the load prediction is more realistic, and the road characteristics of mountain city are considered, so that the load prediction of electric vehicle in mountain city is more accurate.

[0005] In the first aspect, the present application provides a mountain city electric vehicle charging load probability prediction system, which comprises:

[0006] A road side camera module is arranged on each traffic road, and is used to collect vehicle pictures of passing vehicles on the traffic road.

[0007] A road side GPS positioning module is installed on the road side camera module, and is used to obtain the real-time position of the passing vehicle.

[0008] A road side temperature acquisition module is arranged on each traffic road, and is used to collect the real-time environmental temperature of the environment where the passing vehicle is located.

[0009] a cloud platform data center, configured to store charging station location data, charging mode data of the charging station, road data of the mountain city, historical charging information, historical air conditioner usage data, and unit mileage driving power consumption parameters corresponding to different vehicle models, power consumption parameters of the vehicle-mounted equipment, and battery charging power parameters;

[0010] a cloud processing platform, in communication connection with the road side camera module, the road side GPS positioning module, the road side temperature acquisition module, and the cloud platform data center, and configured to:

[0011] perform image processing on the vehicle pictures of the passing vehicles, identify the electric vehicles in the passing vehicles and facial information of the drivers of the electric vehicles, and determine the driver mileage anxiety degree according to the facial information of the drivers;

[0012] according to the vehicle model data of the electric vehicle, search for, in the cloud platform data center, the unit mileage driving power consumption parameters corresponding to the vehicle model of the electric vehicle, the power consumption parameters of the vehicle-mounted equipment, and the battery charging power parameters, acquire, in the cloud platform data center, the historical charging information corresponding to the electric vehicle, the charging station location data, and the charging mode data of the charging station, and acquire real-time information of the electric vehicle, the real-time information including real-time position and real-time environmental temperature;

[0013] calculate, according to the real-time position, the historical charging information, the road data of the mountain city in the cloud platform data center, the unit mileage driving power consumption parameters, and a travel power consumption formula, the travel power consumption of the electric vehicle in the mountain city, predict, according to a corresponding relationship between the real-time environmental temperature and the environmental data in the cloud platform data center and the historical air conditioner usage data of the mountain city, the vehicle air conditioner usage data of the electric vehicle, and calculate, according to the historical charging information, the vehicle air conditioner usage data, the power consumption parameters of the vehicle-mounted equipment, and a device power consumption formula, the device power consumption of the electric vehicle;

[0014] calculate, according to the travel power consumption, the device power consumption, and the battery state of charge at the last time of leaving the charging station in the historical charging information, the remaining battery power of the electric vehicle, and obtain, according to the remaining battery power and the driver mileage anxiety degree, the charging probability of the electric vehicle;

[0015] according to the real-time position in the real-time information and the charging station location data, take the nearest charging station of the real-time position as a destination, take a shortest path as an objective function, input a path optimization algorithm mathematical model, output the nearest target charging station of the electric vehicle, an optimal path of the electric vehicle to the target charging station, and calculate the shortest time consumption of the electric vehicle to the target charging station;

[0016] determine, according to the battery charging power parameters and the charging mode data of the target charging station, the charging load value of the electric vehicle;

[0017] The charging load prediction result of the electric vehicle is generated, and the charging load prediction result includes: a license plate number of the electric vehicle, a charging probability, a target charging station, a shortest time consumption of the electric vehicle to the target charging station, and a charging load value of the electric vehicle.

[0018] In a second aspect, the embodiments of the present application provide a charging load probability prediction method for electric vehicles in a mountain city, which comprises:

[0019] A vehicle picture of a passing vehicle on a traffic road is obtained by a road side camera module arranged on the traffic road, a real-time position of the passing vehicle is obtained by a road side GPS positioning module arranged on the road side camera module, and a real-time environmental temperature of an environment where the passing vehicle is located is obtained by a road side temperature acquisition module arranged on the traffic road.

[0020] The vehicle picture of the passing vehicle is processed to identify an electric vehicle in the passing vehicle and facial information of a driver of the electric vehicle, and the driver's range anxiety is determined according to the facial information of the driver.

[0021] According to the vehicle type data of the electric vehicle, a unit mileage running power consumption parameter corresponding to the vehicle type of the electric vehicle, a power consumption parameter of a vehicle-mounted device, and a battery charging power parameter are searched in an electric vehicle database, historical charging information corresponding to a record of the electric vehicle is obtained in a charging database, and charging station position data and a charging method data of the charging station are obtained in a charging station database.

[0022] According to the real-time position of the electric vehicle, the historical charging information, road data of the mountain city in a traffic road network database, the unit mileage running power consumption parameter, and a travel power consumption formula, a travel power consumption of the electric vehicle in the mountain city is calculated, vehicle air conditioner usage data of the electric vehicle is predicted according to a corresponding relationship between the real-time environmental temperature and environmental data in an air conditioner database and historical air conditioner usage data of the mountain city, and a device power consumption of the electric vehicle is calculated according to the historical charging information, the vehicle air conditioner usage data, the power consumption parameter of the vehicle-mounted device, and a device power consumption formula.

[0023] According to the travel power consumption, the device power consumption, and a battery state of charge in the historical charging information when the electric vehicle last left a charging station, a remaining battery capacity of the electric vehicle is calculated, and a charging probability of the electric vehicle is obtained according to the remaining battery capacity and the driver's range anxiety.

[0024] According to the real-time position and the charging station position data, a nearest charging station of the real-time position is taken as a destination, a shortest path is taken as a target function, a path optimization algorithm mathematical model is input, a nearest target charging station of the electric vehicle, an optimal path of the electric vehicle to the target charging station, and a shortest time consumption of the electric vehicle to the target charging station are output, and the shortest time consumption of the electric vehicle to the target charging station is calculated.

[0025] determine the charging load value of the electric vehicle according to the battery charging power parameter and the charging mode data of the target charging station;

[0026] generate the charging load prediction result of the electric vehicle, the charging load prediction result comprising: the license plate number of the electric vehicle, the charging probability, the target charging station, the shortest time consumption of the electric vehicle to the target charging station, and the charging load value of the electric vehicle.

[0027] The charging load prediction scheme of the electric vehicle in the mountain city of the application combines the objective reality of the remaining vehicle power with the subjective desire of the vehicle owner's range anxiety, and combines the power grid, the traffic network and the information network, so that the load prediction is more realistic, and the road characteristics of the mountain city are considered, so that the load prediction of the electric vehicle in the mountain city is more accurate. Moreover, the prediction result is more numerical, and the license plate number, charging probability, target charging station, time of reaching the target charging station and specific size of the charging load of each electric vehicle are given, so that the future load prediction growth curve of each charging station can be obtained more conveniently, and subsequent power grid economic scheduling, power flow optimization and the like can be facilitated.

[0028] In addition, the electric vehicle adopts a non-intrusive detection method, does not need to be associated with the equipment in the vehicle, collects data information from the outside, and avoids affecting the mood of the vehicle owner to cause a large error in the prediction result.

[0029] The above description is only a summary of the technical scheme of the application, in order to more clearly understand the technical means of the application, and can be implemented according to the content of the specification, and in order to make the above and other purposes, characteristics and advantages of the application more obvious and easy to understand, the following specific embodiments of the application are described. BRIEF DESCRIPTION OF DRAWINGS

[0030] The drawings described herein are used to provide further understanding of the application, and form a part of the application. The schematic embodiments of the application and the description thereof are used to explain the application, and do not constitute an improper limitation on the application. In the drawings:

[0031] Figure 1 The structure block diagram of the charging load probability prediction system of the electric vehicle in the mountain city based on non-intrusive detection of the embodiment of the application is shown;

[0032] Figure 2 The flowchart of the charging load probability prediction method of the electric vehicle in the mountain city based on non-intrusive detection of the embodiment of the application is shown;

[0033] Figure 3 The flowchart of the processing work of the cloud processing platform of the embodiment of the application is shown. DETAILED DESCRIPTION

[0034] With reference to the drawings and the embodiments of the present application, the technical solutions in the embodiments of the present application will be described clearly. Obviously, the described embodiments are only some of the embodiments of the present application, but not all of them. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art are within the scope of protection of the present application.

[0035] The terms "first", "second", and the like in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are generally a category and do not limit the number of objects, for example, the first object can be one or more. In addition, "and / or" in the specification and claims indicates at least one of the connected objects, and the character " / ", generally indicates that the objects before and after are in an "or" relationship.

[0036] The charging load prediction system for electric vehicles in mountainous cities and the charging load prediction method for electric vehicles in mountainous cities provided by the embodiments of the present application will be described in detail below with reference to the drawings, specific embodiments and application scenarios.

[0037] Embodiment one

[0038] The charging load probability prediction system for electric vehicles in mountainous cities provided by the embodiments of the present application is shown in Figure 1 The charging load probability prediction system includes an electric vehicle data acquisition and calculation system, a cloud platform data center, and a cloud processing platform. The electric vehicle data acquisition and calculation system includes:

[0039] A road-side camera module is deployed on each traffic road to collect vehicle pictures of passing vehicles on the traffic road.

[0040] A road-side GPS positioning module is installed on the road-side camera module to obtain real-time positions of the passing vehicles.

[0041] A road-side temperature acquisition module is deployed on each traffic road to collect real-time environmental temperatures of the environment in which the passing vehicles are located.

[0042] An edge server calculation module is used to pre-process the information collected by the road-side camera module, the road-side GPS positioning module, and the road-side temperature acquisition module.

[0043] An electric vehicle communication module is used to send the pre-processed information to the cloud processing platform.

[0044] The cloud platform data center is used for storing charging station position data, charging mode data of the charging station, road data of the mountain city, historical charging information, historical air conditioner use data, and unit mileage driving power consumption parameters corresponding to different vehicle models, power consumption parameters of the vehicle-mounted equipment, and battery charging power parameters;

[0045] The cloud processing platform is in communication connection with the electric vehicle data acquisition and calculation system and the cloud platform data center, and is used for: performing image processing on the vehicle pictures of the passing vehicles, identifying the electric vehicles in the passing vehicles and face information of drivers of the electric vehicles, and determining a driver mileage anxiety degree according to the face information of the drivers; according to the vehicle model data of the electric vehicles, searching for, in the cloud platform data center, unit mileage driving power consumption parameters corresponding to the vehicle model of the electric vehicles, power consumption parameters of the vehicle-mounted equipment, and battery charging power parameters, and obtaining, in the cloud platform data center, historical charging information corresponding to the electric vehicles, charging station position data and charging mode data of the charging station, and obtaining real-time information of the electric vehicles, the real-time information including real-time position and real-time environmental temperature; according to the real-time position, the historical charging information, road data of the mountain city in the cloud platform data center, the unit mileage driving power consumption parameters and a travel power consumption formula, calculating travel power consumption of the electric vehicles in the mountain city, and according to a corresponding relationship between the real-time environmental temperature and environmental data in the cloud platform data center and historical air conditioner use data of the mountain city, predicting vehicle air conditioner use data of the electric vehicles, and according to the historical charging information, the vehicle air conditioner use data, the power consumption parameters of the vehicle-mounted equipment and an equipment power consumption formula, calculating equipment power consumption of the electric vehicles; according to the travel power consumption, the equipment power consumption, a battery state of charge of the last time when the electric vehicles are separated from the charging station in the historical charging information, calculating a battery residual capacity of the electric vehicles, and according to the battery residual capacity and the driver mileage anxiety degree, obtaining a charging probability of the electric vehicles; according to the real-time position in the real-time information and the charging station position data, taking the nearest charging station of the real-time position as a destination, taking a shortest path as a target function, inputting into a path optimization algorithm mathematical model, outputting the nearest charging station of the electric vehicles, an optimal path of the electric vehicles to the target charging station, and calculating a shortest time consumption of the electric vehicles to the target charging station; according to the battery charging power parameters and the charging mode data of the target charging station, determining a charging load value of the electric vehicles; and generating a charging load prediction result of the electric vehicles, the charging load prediction result including: a license plate number of the electric vehicles, the charging probability, the target charging station, the shortest time consumption of the electric vehicles to the target charging station, and the charging load value of the electric vehicles.

[0046] In an embodiment of the present application, the road side camera module comprises: a traffic camera, used for shooting vehicle pictures of passing vehicles; a first communication module, in communication connection with the image processing module and the edge server calculation module, used for uploading data obtained according to the vehicle pictures to the edge server calculation module for preprocessing.

[0047] The road-side GPS positioning module comprises a positioning module for positioning the real-time position of the passing vehicle; a second communication module in communication connection with the positioning module and the edge server computing module, for uploading the real-time position of the passing vehicle to the edge server computing module for preprocessing.

[0048] The road-side temperature collecting module comprises a temperature sensor arranged on each traffic road for collecting the real-time environmental temperature of the environment where the passing vehicle is located; a third communication module in communication connection with the temperature sensor and the edge server computing module, for uploading the real-time environmental temperature to the edge server computing module for preprocessing.

[0049] Finally, the preprocessed data is uploaded to the cloud processing platform through the electric vehicle communication module by the edge server computing module.

[0050] In an embodiment of the present application, the cloud platform data center comprises an electric vehicle database for storing the unit mileage driving power consumption parameters corresponding to different vehicle models, the power consumption parameters of the on-board equipment and the battery charging power parameters; a charging database for storing the historical charging information of various electric vehicles; a charging station database for storing the charging station location data and the charging mode data of the charging station; a traffic road network database for storing the road data of the mountain city; an air conditioner database for storing the historical air conditioner usage data of the mountain city; and a fourth communication module in communication connection with the cloud processing platform, for uploading the unit mileage driving power consumption parameters corresponding to different vehicle models, the power consumption parameters of the on-board equipment and the battery charging power parameters, as well as the historical charging information of various electric vehicles, the charging station location data, the charging mode data of the charging station, the road data of the mountain city and the historical air conditioner usage data of the mountain city to the cloud processing platform.

[0051] In an embodiment of the present application, as shown in Figure 1 The cloud processing platform comprises an image recognition module for recognizing the electric vehicle in the passing vehicle and the face information of the driver of the electric vehicle according to the vehicle picture; a charging probability prediction module for calculating the charging probability of the electric vehicle; a charging load value prediction module for calculating the charging load value of the electric vehicle; a path optimization module for calculating the target charging station closest to the electric vehicle and the shortest time consumption of the electric vehicle to reach the target charging station; and a fifth communication module in communication connection with the road-side camera module, the road-side GPS positioning module, the road-side temperature collecting module and the cloud platform data center.

[0052] Embodiment Two

[0053] The embodiment of the present application provides a charging load probability prediction method for electric vehicles in a mountain city, as shown in Figure 2 The method comprises:

[0054] The system uses roadside camera modules to collect images of vehicles passing through traffic, identifies license plate information based on the images, and determines whether a vehicle is an electric vehicle based on the license plate information.

[0055] For electric vehicles, image processing is performed on vehicle images. An OCR recognition model is used to extract the driver's facial information. A comprehensive evaluation function mathematical model is used to determine the driver's range anxiety level based on this facial information. Real-time location, historical charging information, road data for mountainous cities from the cloud platform data center, power consumption parameters per unit mileage, real-time ambient temperature, historical charging information, power consumption parameters of onboard devices, vehicle air conditioning usage data, and power consumption parameters of onboard devices are acquired. A travel power consumption model is used to calculate the electric vehicle's travel power consumption in mountainous cities, and a device power consumption model is used to calculate the electric vehicle's device power consumption. The remaining battery power of the electric vehicle is calculated. A charging probability prediction model is then trained and calculated, and an adaptive model is trained using a neural network to finally predict the charging probability of the electric vehicle.

[0056] It also acquires real-time location data, charging station location data, and urban road network data; uses path optimization algorithms to determine the location of the target charging station and the shortest time to reach the target charging station; and acquires charging method data provided by the target charging station and battery charging power parameters of the electric vehicle to calculate the charging load value of the electric vehicle.

[0057] Finally, the system outputs the electric vehicle's license plate number, charging probability, target charging station, shortest time for the electric vehicle to reach the target charging station, and charging load value.

[0058] Finally, cluster analysis is performed on the charging load forecast results of all electric vehicles heading to the same charging station to generate a charging load growth forecast curve for each charging station. Then, returning to the first step, rolling forecasts are implemented to continuously update the forecast curves.

[0059] In one embodiment of this application, the method for predicting the charging load probability of electric vehicles specifically includes:

[0060] The vehicle picture of the passing vehicle is obtained by the road side camera module arranged on the traffic road, the real-time position of the passing vehicle is obtained by the road side GPS positioning module arranged on the road side camera module, and the real-time environmental temperature of the environment where the passing vehicle is located is obtained by the road side temperature collection module arranged on the traffic road; the vehicle picture of the passing vehicle is processed, the electric vehicle in the passing vehicle and the face information of the driver of the electric vehicle are recognized, and the driver's range anxiety is determined according to the face information of the driver; the unit mileage driving power consumption parameter corresponding to the vehicle model of the electric vehicle, the power consumption parameter of the vehicle-mounted equipment and the battery charging power parameter are searched in the electric vehicle database according to the vehicle model data of the electric vehicle, the historical charging information corresponding to the record of the electric vehicle is obtained in the charging database, and the charging station position data and the charging mode data of the charging station are obtained in the charging station database; the travel power consumption of the electric vehicle in the mountain city is calculated according to the real-time position of the electric vehicle, the historical charging information, the road data of the mountain city in the traffic road network database, the unit mileage driving power consumption parameter and the travel power consumption formula, and the vehicle air conditioner use data of the electric vehicle is predicted according to the corresponding relationship between the real-time environmental temperature and the environmental data in the air conditioner database and the historical air conditioner use data of the mountain city, and the device power consumption of the electric vehicle is calculated according to the historical charging information, the vehicle air conditioner use data, the power consumption parameter of the vehicle-mounted equipment and the device power consumption formula; the battery residual capacity of the electric vehicle is calculated according to the travel power consumption, the device power consumption and the battery state of charge when the last time the electric vehicle is separated from the charging station, and the charging probability of the electric vehicle is obtained according to the battery residual capacity and the driver's range anxiety; the nearest target charging station of the electric vehicle, the optimal path of the electric vehicle to the target charging station and the shortest time consumption of the electric vehicle to the target charging station are output by inputting the nearest target charging station of the electric vehicle, the optimal path of the electric vehicle to the target charging station and the shortest time consumption of the electric vehicle to the target charging station into the path optimization algorithm mathematical model, and the charging load value of the electric vehicle is determined according to the battery charging power parameter and the charging mode data of the target charging station; the charging load prediction result of the electric vehicle is generated, and the charging load prediction result includes the license plate number of the electric vehicle, the charging probability, the target charging station, the shortest time consumption of the electric vehicle to the target charging station and the charging load value of the electric vehicle.

[0061] In addition, other real-time information of the electric vehicle can also be obtained, such as real-time speed, real-time time, real-time motion trajectory, etc., and the historical charging information includes the time when the last time the electric vehicle is separated from the charging station, the charging station position when the last time the electric vehicle is separated from the charging station, the battery state of charge when the last time the electric vehicle is separated from the charging station, and all travel data from the last time the electric vehicle is separated from the charging station to the real-time position of the electric vehicle, the travel data including speed, time, position, temperature, motion trajectory, and the environmental data including season, weather, time, temperature and humidity data.

[0062] As Figure 3 shown, the processing work of the cloud processing platform includes range anxiety calculation, residual power calculation, charging probability calculation, recent charging station path optimization, and charging load prediction. The specific content of each processing work is given below:

[0063] I. Range anxiety calculation

[0064] 1. Traffic cameras are installed along the road to record the driving conditions of vehicles on the road. Using machine vision technology, electric vehicles are selected, and the license plate number and driver's face information of the electric vehicles are identified and recorded in the cloud storage database for residual power calculation.

[0065] Using the vehicle big data recorded by the traffic cameras installed along the road as the training set, a target detection model based on YOLOv3 is used to design the loss function:

[0066] L cls = -log p u

[0067]

[0068] L cls represents the loss-cls cost function of vehicle classification loss, L loc represents the loss-bbox cost function of predicted frame regression loss, p represents the predicted class, u represents the true class, t represents the coordinate information of the predicted frame, v represents the coordinate information of the true frame, i represents the traversal index of each car in an image, g represents the gradient weight, which is used to weaken the predicted vehicle position information. The error between the predicted data and the test set data is calculated by the loss function, and all weight parameters in the YOLOv3 target detection model are updated by gradient descent method. The gradient descent method is to calculate the gradient size and direction corresponding to the current network weight parameters, and update the network parameters in the opposite direction of the gradient according to the set step size. After multiple iterations of training, a network parameter that minimizes the loss function error can be obtained.

[0069] After loading the trained network parameters using the YOLOv3 target detection model, input the recorded vehicle pictures, and output the license plate region and driver face region after model prediction calculation.

[0070] 2. After obtaining the license plate region and driver face region, use the OCR recognition model to extract the license plate information and facial micro-expression information. According to the time sequence, record the facial micro-expression information, construct a comprehensive evaluation function, and calculate the driver's range anxiety simulation value by substituting the facial micro-expression information into the comprehensive evaluation function. The comprehensive evaluation function is:

[0071]

[0072] wherein, score represents the driver mileage anxiety simulation value, a i represents the weight value of the i-th facial micro-expression key point, Z i represents the i-th facial micro-expression key point coordinate, n represents the number of facial micro-expression key points.

[0073] Then, the driver mileage anxiety simulation value is converted by using the conversion formula to obtain the principal component risk value of the driver mileage anxiety. The conversion formula is:

[0074] risk_value = [score + abs(min(score))] x 10

[0075] wherein, risk_value represents the principal component risk value of the driver mileage anxiety, and score represents the driver mileage anxiety simulation value.

[0076] II. Residual power calculation

[0077] 1. The traffic camera installed along the road records the driving conditions of the vehicles on the road. Using machine vision technology, the electric vehicles are screened out, the license plate information of the electric vehicles is identified, the vehicle model data in the electric vehicle database is compared, the unit mileage driving power consumption parameters of the battery used by the vehicle model, the power consumption parameters of the vehicle-mounted equipment (mainly air conditioner) and the battery charging power parameters are called, and the real-time speed, real-time time, real-time location, real-time environmental temperature near the real-time location, motion trajectory (including driving direction) are recorded to the cloud storage database.

[0078] 2. Taking the license plate information of the electric vehicle as the retrieval keyword, the historical charging information corresponding to the record of the electric vehicle is extracted from the charging database. The historical charging information is a linked list information, and the linked list is a storage method, for example, using the license plate number as the head of the chain, and the subsequent chain information is all the information corresponding to this license plate number, which is convenient for retrieval. The historical charging information includes the time of the last disengagement from the charging station, the charging station location of the last disengagement from the charging station, the state of charge of the battery at the last disengagement from the charging station, and all travel data (travel data including speed, time, location, temperature, motion trajectory) from the last disengagement from the charging station to the real-time location of the electric vehicle.

[0079] 3. According to real-time information and historical charging information, combined with road data of mountainous city in traffic network database (mainly according to past records of vehicle time, position, motion trajectory, above last departure charging station charging station position as the starting point, real-time position as the destination, simulate the complete driving path of electric vehicle), using the battery travel power consumption formula, the travel power consumption of the battery in the driving path is calculated (the vehicle speed, time, road length and slope in the traffic network database, and the unit mileage driving power consumption parameter of the battery in each section of the past driving trajectory are needed).

[0080] In plain city, the travel power consumption of electric vehicle can be calculated by the horizontal distance between two points, and the calculation formula is:

[0081]

[0082] In the formula, Q O,D is the travel power consumption of electric vehicle from the starting point to the destination, L is the driving path from the starting point to the destination, i, j are two adjacent nodes on the driving path L, X' i,j is the horizontal distance between adjacent two points i, j, N R represents the city road network set, p s is the unit mileage driving power consumption parameter of the battery.

[0083] Considering the relative height of the road in mountainous city, the travel power consumption of electric vehicle depends not only on the horizontal power consumption, but also on the work done in the vertical direction to overcome gravity. At the same time, considering the rugged terrain and undulating road in mountainous city, the driving path between two points is no longer a smooth straight line, but a sloping road with a certain slope. The final travel power consumption of electric vehicle in mountainous city is:

[0084]

[0085] In the formula, E O, is the travel power consumption, p s is the unit mileage driving power consumption parameter of electric vehicle, L is the driving path of electric vehicle, X' i,j is the driving mileage of adjacent two points i, j on the driving path, i, j ∈ N R , N R is the city road network set, H i,j is the relative height difference of adjacent two points i, j in the vertical direction, H i,j = h j - h i , h i is the height of i point, h j is the height of j point, when H i,jWhen <0, path[i, j] is a downhill section, and when H i,j >0, path[i, j] is an uphill section, and a i,j is a climbing coefficient or an energy recovery efficiency coefficient of the electric vehicle overcoming gravity in unit relative height, with a unit of m / kWh.

[0086] p s ∑ i,j∈L X i,j represents power consumption when driving on flat ground, which is a driving power consumption parameter per unit distance multiplied by driving distance; ∑ i,j∈L a i,j H i,j is an additional power consumption or energy recovery on an uphill or downhill, considering the characteristics of mountainous cities, where many roads are not flat.

[0087] a i,j The calculation formula of a

[0088]

[0089] wherein a c is a climbing coefficient, a d is an energy recovery efficiency coefficient. When the electric vehicle drives on an uphill section, it needs to overcome gravity to do work, so a i,j is a climbing coefficient a c When the electric vehicle drives on a downhill section, it can obtain part of the energy recovery due to the braking state, so a i,j is an energy recovery efficiency coefficient a d .

[0090] When the electric vehicle drives on a downhill braking section, the electric motor will be converted into a generator operation state, which can assist the on-board battery to recover part of the energy, effectively improve the energy utilization efficiency, and increase the cruising range of the electric vehicle. However, the running process of the electric vehicle generally includes acceleration, driving, deceleration and braking states, and the energy recovery is related to the braking time, acceleration and driving speed, etc. It is difficult to accurately calculate and collect the running state in one trip, so a d is assumed to be a constant average constant when calculating using mathematical formula, and its value can be measured during daily operation of the electric taxi. Specifically, the energy consumption data measured by the electric vehicle during daily operation on various road sections at different speeds can be obtained by subtracting the calculated energy consumption of the same length and speed on flat ground corresponding to these road sections, and the uphill additional energy consumption or downhill energy recovery is obtained. The uphill additional energy consumption or downhill energy recovery is divided by the height difference, and the climbing coefficient a c or the energy recovery efficiency coefficient a d corresponding to different speeds on the same road section is obtained.

[0091] A database of the climbing coefficient α c and the energy recovery efficiency coefficient α d corresponding to different vehicle speeds, when the electric vehicle passes an uphill or downhill road section, the database is searched for the climbing coefficient α c or the energy recovery efficiency coefficient α d corresponding to the real-time vehicle speed of the electric vehicle.

[0092] In addition, the climbing coefficient α c of the path [i, j] is determined by the actual road slope grade of each section in the path.

[0093] For example, the climbing coefficient α c is:

[0094]

[0095] where v max is the maximum vehicle driving speed allowed in the city, l v represents the energy conversion efficiency of the power battery, v n is the normal driving speed of the electric vehicle, M v is a parameter that effectively represents the road slope grade, the larger the value, the more severe the road undulation, and the more steep the undulation slope of the road, and vice versa. Because the roads in the actual traffic network are very complex, it is difficult to describe in detail, therefore, different road slope angles can be one-to-one corresponding to the setting of M v is a constant value.

[0096] For example, a section is an uphill section, first calculate how much electricity is used if it is normal flat driving, then multiply the corresponding height of the uphill section by the climbing coefficient to calculate how much additional electricity consumption is required for the uphill section compared to normal flat driving. If it is a downhill section, ∑ i,j∈L α i,j H i,j The calculation result is negative, and the final flat section power consumption minus the downhill recovered power is the actual power consumption of the downhill.

[0097] In summary, after all the parameters in the mathematical formula are given, the travel power consumption E O, of the electric vehicle can be calculated.

[0098] 4. Based on real-time and historical charging information, combined with vehicle air conditioning usage data corresponding to different parameter values ​​such as season, weather, time, temperature, and humidity in the air conditioning database—that is, the usage status of the air conditioning equipment—the power consumption of the electric vehicle is calculated using the equipment power consumption formula. For example, in summer, on a sunny day at 2 PM, the average probability of the air conditioning being on is 98% when the temperature is 39 degrees Celsius. An air conditioning being on is considered to be on when the probability of it being on exceeds a certain threshold.

[0099] To determine which road segments along the vehicle's past travel route (starting from the last charging station disconnected from the grid and ending at the real-time location), the air conditioning was on, and the total mileage of these segments with the air conditioning on was calculated. This requires using previously recorded time and temperature data for each segment, and then applying formulas derived from big data analysis relating time, temperature, weather, season, and the probability of air conditioning operation. By connecting all past road segments, a simulation of the electric vehicle's air conditioning operation along the travel route is created. This simulation yields the distance traveled with the air conditioning on, which is then used to calculate the device's power consumption.

[0100] The power consumption of an electric vehicle's air conditioning is entirely supplied by the battery, and the statistical mathematical formula relating the probability of the air conditioning starting to temperature is:

[0101]

[0102] Among them, P ac Let T be the probability of the air conditioner starting, and T be the temperature. Correspondingly, season, weather, time, and humidity can be calculated using similar statistical mathematical formulas. By comprehensively considering all factors, the probability of the air conditioner starting during the operation of an electric vehicle can be obtained, and the power consumption of the device can be calculated based on the probability of the air conditioner starting.

[0103] Because air conditioning power consumption is affected by factors such as vehicle size and insulation, it's impossible to accurately define how air conditioning power consumption changes with temperature. Instead, we can only analyze the relationship between average air conditioning power consumption and ambient temperature for various electric vehicle types based on extensive statistical data. Different types of electric vehicles exhibit significant differences in air conditioning usage (primarily duration). For example, most taxis and normally operating electric buses keep their air conditioning on throughout their journeys. Vehicles typically turn on the air conditioning as needed. When setting up simulations, we can first consider vehicle model factors (i.e., the power consumption parameters of onboard equipment) and then consider external factors such as season, weather, temperature, and time.

[0104] In actual running road, electric vehicles have various types and models, and different models have different parameters. When calculating the power consumption of the equipment, the corresponding power consumption parameters of the on-board equipment need to be called. For example, the rated endurance mileage of Geely Emgrand EV4 is 0 km, the battery capacity is 52 kWh, and the power consumption per 100 km before and after starting the air conditioner for cooling is 13.12 kWh and 19.06 kWh, respectively. The power consumption per 100 km before and after starting the air conditioner for heating is 13.12 kWh and 19.06 kWh, respectively. The mathematical formula of the equipment power consumption is:

[0105]

[0106] wherein, E T is the equipment power consumption of the electric vehicle from the starting point to the destination, L i,j is the driving path of the electric vehicle with the air conditioner turned on, X i,j is the driving distance between adjacent points i and j on the driving path with the air conditioner turned on, X R is determined according to the vehicle air conditioner usage data of the mountain city, i,j∈N R N R / L is the set of urban road networks, E R / L is the power consumption per 100 km of the electric vehicle under the condition of air conditioner heating / cooling, and E0 is the power consumption per 100 km of the electric vehicle under the condition of air conditioner not turned on.

[0107] wherein, E R / L and E0 are the power consumption parameters of the on-board equipment, which are related to the vehicle model.

[0108] 5. According to the travel power consumption, the equipment power consumption, and the battery state of charge (i.e. the starting power) when the last time the electric vehicle leaves the charging station, the remaining power of the battery at the moment is calculated. The calculation formula of the remaining power is:

[0109] E Z = E S -E O,D -E T

[0110] wherein, E Z is the remaining power, E S is the starting power, E O,D is the travel power consumption, and E T is the equipment power consumption.

[0111] Three. Charging probability calculation

[0112] The remaining power and the range anxiety are used as input variables, and the joint distribution function or probability density function of the charging demand is trained using probability theory knowledge, i.e. the charging probability of the driver of the electric vehicle under different remaining power and range anxiety is predicted.

[0113] The prediction method can use existing research model in early use, and after accumulating enough historical data, neural network algorithm can be used to train unique and more applicable joint distribution function or joint probability density function under mountainous city conditions.

[0114] Exemplarily, assuming that the remaining power fits a normal distribution and the mileage anxiety degree fits a lognormal distribution, the marginal probability density function of the remaining power is:

[0115]

[0116] wherein f(E Z ) is the charging probability corresponding to the battery remaining power, E Z is the battery remaining power, u = 17.6, and σ = 3.4.

[0117] The marginal probability density function of the mileage anxiety degree is:

[0118]

[0119] wherein f(S) is the charging probability corresponding to the driver mileage anxiety degree, S is the driver mileage anxiety degree, u = 3.2, and σ = 0.88.

[0120] Assuming that the remaining power and the mileage anxiety degree are independent of each other, the joint probability density function is:

[0121] f(E Z , S) = f(E Z )f(S).

[0122] wherein f(E Z , S) is the charging probability of the electric vehicle.

[0123] IV. Recent charging station path optimization

[0124] According to the real-time position of the current electric vehicle, combining with the traffic network data, taking the recent charging station near the real-time position as the destination and taking the shortest path or the shortest time consumption as the objective function, the path optimization algorithm mathematical model is inputted, and the optimal path and the shortest time consumption to the nearest charging station are outputted.

[0125] Dijkstra algorithm is to start node as the starting point, in strict incremental manner of distance search node, when search all nodes, algorithm ends. Dijkstra algorithm is very representative algorithm in solving shortest path problem, the precondition of this algorithm is that there is no negative weight in all edges of graph, road network can meet this condition. When selecting a node in road network as reference point, it will search the remaining nodes in order of increasing path distance from reference point, in actual operation, it is not necessary to search all the remaining nodes in road network, and the target node can be searched to stop the algorithm in advance.

[0126] Let the real-time position of electric vehicle be V0, the charging station near the real-time position be other nodes, establish an array Dis, the serial number of the elements of array Dis corresponds to the node number, and the value stored in each element of array Dis is the path length of the corresponding node and V0. The nodes whose shortest path to V0 have been found are stored in array S in order, and all the successor nodes of the nodes that have been found are stored in array T. The specific execution sequence of the mathematical model of the shortest path planning and path optimization algorithm with V0 as the initial node is as follows:

[0127] (1) move all the successor nodes of V0 to array T, and store the path length of the node and V0 in the corresponding element of array Dis;

[0128] (2) move the node with the minimum path length between V0 in array T to array S, and let this node be Vi;

[0129] (3) if Vi is the target node, execute step (6);

[0130] (4) expand Vi, add all the successor nodes of Vi which do not belong to array S and array T to array T, calculate the path length between them and V0 and store it in the corresponding element of array Dis, if a successor node has been in array T before, compare the new path length between this successor node and V0 with the path length already existing in array Dis, if the new path length is smaller, update the value in the element of array Dis, if the successor node has been in array S before, skip;

[0131] (5) if there is no node in array T, the algorithm ends, indicating that there is no passable path between V0 and the target node, otherwise, go to step (2) and continue to execute;

[0132] (6) find the optimal path Xmin between V0 and the target node, and set the target node as the nearest charging station Vmin, and the algorithm ends.

[0133] According to the recent charging station Vmin and the corresponding relationship of nearby charging stations and traffic network data, the specific location of the nearest charging station Vmin is obtained. And the real-time vehicle speed v is the average speed of the electric vehicle to the nearest charging station Vmin, and the time consumption is calculated The expected arrival time of the charging station is obtained by delaying the real-time time by Δt.

[0134] The distance between the nearest target charging station and the location where the electric vehicle generates charging demand is less than or equal to the cruising range of the electric vehicle, and the constraint condition is:

[0135] 0≤x total (t)≤M i (t)

[0136] Wherein, x total (t) represents the distance between the target charging station and the location where the electric vehicle generates charging demand, M i (t) represents the cruising range of the ith electric vehicle;

[0137] The shortest time consumption of the electric vehicle to reach the target charging station is:

[0138]

[0139] Wherein, t0 represents the time when the electric vehicle generates charging demand, t1 represents the shortest time consumption of the electric vehicle to access the target charging station, V0 represents the average speed of the electric vehicle, X" represents the path from the starting point to the destination, L' i,j represents the distance between adjacent points in the path, and H' i,j represents the relative height difference between adjacent points in the path in the vertical direction.

[0140] Five. Determine the charging load value

[0141] The battery charging power parameter corresponding to the model of the electric vehicle is called, the destination charging station predicted by the above path optimization algorithm is called, the charging mode data that can be provided by this charging station (that is, whether the charging pile provided by the charging station is slow charging or fast charging) is called, and finally the charging load size is determined according to the battery charging power parameter and the charging mode data.

[0142] The constraint condition of the charging load value of the electric vehicle is:

[0143]

[0144] Wherein, represents the minimum amount of the battery discharge warning of the electric vehicle, represents the maximum amount of the battery charging of the electric vehicle, represents the current state of charge of the battery of the ith electric vehicle.

[0145] VI. Charging load prediction

[0146] The final charging load prediction result of an electric vehicle is given, including: license plate number of the electric vehicle, charging probability, target charging station, shortest time consumption of the electric vehicle to the target charging station, and charging load value of the electric vehicle. For example: the electric vehicle with license plate number "xxxxxx" will have an 85% probability of arriving at Chaoshimen Lafayette underground parking lot charging station for charging at 13:00, and the charging power is 7kw.

[0147] Finally, it should be noted that the electric vehicle travel trajectory depends on the driving decision of the vehicle owner (i.e. the driver), and different vehicle owners have different charging decisions, making the electric vehicle charging behavior itself random, so the future charging load prediction is a probabilistic prediction; at the same time, the image data is collected based on the non-intrusive detection method, and the charging demand implied in the image data is mined and analyzed, which is affected by the coupling degree between the image data and the real charging decision of the vehicle owner, making the analysis result have fitting deviation with the real charging behavior in the future, so the charging load prediction method based on non-intrusive detection is a probabilistic prediction method.

[0148] Example III

[0149] The charging load prediction results of all electric vehicles going to the same charging station are clustered and analyzed, and the future charging load growth prediction curve of each charging station is given, which can be used for subsequent planning and dispatch personnel to carry out power flow optimization, unit combination, economic dispatch, etc.

[0150] Wherein, the electric quantity of the electric vehicle after charging and discharging is connected to the grid to meet the customer demand, and the constraint condition of the electric quantity connected to the grid is:

[0151]

[0152] Wherein, represents the charging electric quantity during the electric vehicle accessing the charging pile, represents the discharging electric quantity during the electric vehicle accessing the charging pile, represents the expected charging electric quantity of the electric vehicle user, represents the initial electric quantity when the electric vehicle accesses the charging pile, η cha represents the charging efficiency of the electric vehicle, η dis represents the discharging efficiency of the battery, B i represents the battery capacity of the i-th electric vehicle, ε bat represents the battery loss coefficient of the electric vehicle;

[0153] Whether the electric vehicle accessing the charging pile meets the time constraint condition is:

[0154]

[0155] wherein, t off denotes the time when the electric vehicle accesses the charging pile, t on denotes the time when the electric vehicle user sets the expected leaving time of the charging pile, denotes the maximum charging power.

[0156] In the process of applying the charging load growth prediction curve to the staff for power grid scheduling, a scheduling method for electric vehicle flexible air conditioning load to suppress substation overload, and a minimum temperature control plan scheduling deviation model are as follows:

[0157]

[0158] wherein, denotes the charging and discharging power of the electric vehicle at time t in the actual scheduling process, denotes the electricity consumption of the user after the air conditioner is regulated at time t in the data scheduling process, denotes the ordered charging and discharging and air conditioner temperature control plan prediction power of the electric vehicle obtained by the scheduling center according to each prediction curve;

[0159] The scheduling constraint condition is that the substation cannot be overloaded, that is, the substation load cannot exceed the substation capacity, and the formula is as follows:

[0160]

[0161] wherein, denotes the basic load of the regional power grid at time t, denotes the charging load of the newly added electric vehicle in the region at time t, denotes the line loss at time t, denotes the reduction of the room air conditioning load at time t, denotes the discharging power of the electric vehicle at time t, S N denotes the rated power of the transformer, and cosψ denotes the power factor of the transformer.

[0162] It should be noted that, in the present document, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the presence of additional identical elements in the process, method, article, or apparatus that comprises the element. Furthermore, it should be noted that the methods and apparatus of the present embodiments are not limited by the order of the steps or the sequence for performing the steps, as some steps can occur in different orders and / or concurrently with one another; for example, described methods can be performed in an order other than that described, and / or additional steps can be added, or steps can be omitted, or a combination thereof. Also, characteristics described in relation to certain examples can be combined in other examples.

[0163] The embodiments of the present application described above are merely illustrative, and the present application is not limited to the above-described specific embodiments, which are merely illustrative, but not restrictive, and a person of ordinary skill in the art can make many forms under the inspiration of the present application without departing from the purpose of the present application and the scope protected by the claims.

Claims

1. A mountain city charging load probability prediction system, characterized in that, The system comprises: a road-side camera module arranged on each traffic road to collect vehicle pictures of passing vehicles on the traffic road; a road-side GPS positioning module installed on the road-side camera module to obtain real-time positions of the passing vehicles; a road-side temperature collection module arranged on each traffic road to collect real-time ambient temperatures of environments in which the passing vehicles are located; a cloud platform data center configured to store charging station position data, charging modes of charging stations, road data of mountain cities, historical charging information, historical air conditioner usage data, and unit mileage driving power consumption parameters corresponding to different vehicle models, power consumption parameters of vehicle-mounted devices, and battery charging power parameters; a cloud processing platform in communication connection with the road-side camera module, the road-side GPS positioning module, the road-side temperature collection module, and the cloud platform data center, and configured to: perform image processing on the vehicle pictures of the passing vehicles to identify electric vehicles in the passing vehicles and facial information of drivers of the electric vehicles, and determine driver mileage anxiety degrees according to the facial information of the drivers; according to vehicle model data of the electric vehicles, find, in the cloud platform data center, unit mileage driving power consumption parameters corresponding to the vehicle models of the electric vehicles, power consumption parameters of vehicle-mounted devices, and battery charging power parameters, obtain, in the cloud platform data center, historical charging information corresponding to the electric vehicles, charging station position data, and charging modes of charging stations, and obtain real-time information of the electric vehicles, the real-time information including real-time positions and real-time ambient temperatures; according to the real-time positions, the historical charging information, road data of mountain cities in the cloud platform data center, the unit mileage driving power consumption parameters, and a travel power consumption formula, calculate travel power consumption of the electric vehicles in mountain cities, according to a corresponding relationship between the real-time ambient temperatures and environmental data in the cloud platform data center and historical air conditioner usage data of the mountain cities, predict vehicle air conditioner usage data of the electric vehicles, and according to the historical charging information, the vehicle air conditioner usage data, the power consumption parameters of the vehicle-mounted devices, and a device power consumption formula, calculate device power consumption of the electric vehicles; according to the travel power consumption, the device power consumption, and a battery state of charge at a last time when the electric vehicles leave charging stations in the historical charging information, calculate a battery residual power of the electric vehicles, and according to the battery residual power and the driver mileage anxiety degrees, obtain a charging probability of the electric vehicles; according to the real-time positions in the real-time information and the charging station position data, take the nearest charging station of the real-time positions as a destination, take a shortest path as an objective function, input a path optimization algorithm mathematical model, and output the nearest target charging station of the electric vehicles, an optimal path of the electric vehicles to the target charging station, and a shortest time consumption of the electric vehicles to the target charging station. determining a charging load value of the electric vehicle according to the battery charging power parameter and the charging mode data of the target charging station; generating a charging load prediction result of the electric vehicle, the charging load prediction result including a license plate number of the electric vehicle, the charging probability, the target charging station, a shortest time consumption of the electric vehicle to the target charging station, and the charging load value of the electric vehicle.

2. The system of claim 1, wherein, The road-side camera module comprises: a traffic camera configured to capture vehicle pictures of the passing vehicles; a first communication module in communication connection with the image processing module and configured to upload data obtained from the vehicle pictures to the cloud processing platform; The road-side GPS positioning module comprises: a positioning module configured to locate real-time positions of the passing vehicles; a second communication module in communication connection with the positioning module and configured to upload the real-time positions of the passing vehicles to the cloud processing platform; The road-side temperature collection module comprises: a temperature sensor deployed on each traffic road and configured to collect real-time environmental temperatures of environments where the passing vehicles are located; a third communication module in communication connection with the temperature sensor and configured to upload the real-time environmental temperatures to the cloud processing platform.

3. The system of claim 1, wherein, The cloud platform data center comprises: an electric vehicle database configured to store unit mileage driving power consumption parameters corresponding to different vehicle models, power consumption parameters of vehicle-mounted devices, and battery charging power parameters; a charging database configured to store historical charging information of various electric vehicles; a charging station database configured to store charging station location data and charging mode data of charging stations; a traffic road network database configured to store road data of the mountainous city; an air conditioner database configured to store historical air conditioner usage data of the mountainous city; a fourth communication module in communication connection with the cloud processing platform and configured to upload the unit mileage driving power consumption parameters corresponding to different vehicle models, the power consumption parameters of vehicle-mounted devices, and the battery charging power parameters, as well as the historical charging information of various electric vehicles, the charging station location data, the charging mode data of charging stations, the road data of the mountainous city, and the historical air conditioner usage data of the mountainous city to the cloud processing platform.

4. The system of claim 1, wherein, The cloud processing platform comprises: an image recognition module configured to recognize electric vehicles in the passing vehicles and facial information of drivers of the electric vehicles according to the vehicle pictures; a charging probability prediction module configured to calculate a charging probability of the electric vehicle; a charging load value prediction module configured to calculate a charging load value of the electric vehicle; a path optimization module configured to calculate the target charging station closest to the electric vehicle and a shortest time consumption of the electric vehicle to the target charging station; a fifth communication module in communication connection with the road-side camera module, the road-side GPS positioning module, the road-side temperature collection module, and the cloud platform data center.

5. A mountain city charging load probability prediction method, characterized in that The method comprises: The system acquires vehicle images of passing vehicles by using a roadside camera module installed on the road, obtains the real-time location of the passing vehicles by using a roadside GPS positioning module installed on the roadside camera module, and obtains the real-time ambient temperature of the environment in which the passing vehicles are located by using a roadside temperature acquisition module installed on the road. Image processing is performed on the vehicle images of the passing vehicles to identify electric vehicles and the facial information of the drivers of the electric vehicles, and the driver's range anxiety level is determined based on the driver's facial information. Based on the electric vehicle model data, the system searches the electric vehicle database for the unit mileage power consumption parameters, on-board equipment power consumption parameters, and battery charging power parameters corresponding to the electric vehicle model. It also retrieves the historical charging information of the electric vehicle in the charging database and the charging station location data and charging method data of the charging station in the charging station database. Based on the real-time location of the electric vehicle, the historical charging information, road data of mountainous cities in the traffic network database, the power consumption parameters per unit mileage, and the travel power consumption formula, the travel power consumption of the electric vehicle in mountainous cities is calculated. Based on the correspondence between the real-time ambient temperature and environmental data in the air conditioning database and the historical air conditioning usage data of the mountainous cities, the vehicle air conditioning usage data of the electric vehicle is predicted. Based on the historical charging information, the vehicle air conditioning usage data, the power consumption parameters of the on-board equipment, and the equipment power consumption formula, the equipment power consumption of the electric vehicle is calculated. The remaining battery power of the electric vehicle is calculated based on the power consumption of the trip, the power consumption of the device, and the battery state of charge at the last time it left the charging station in the historical charging information. The charging probability of the electric vehicle is obtained based on the remaining battery power and the driver's range anxiety. Based on the real-time location and the charging station location data, with the nearest charging station at the real-time location as the destination and the shortest path as the objective function, the mathematical model of the path optimization algorithm is input, and the nearest target charging station of the electric vehicle, the optimal path of the electric vehicle to the target charging station, and the shortest time for the electric vehicle to reach the target charging station are output. Based on the battery charging power parameters and the charging method data of the target charging station, the charging load value of the electric vehicle is determined; Generate a charging load prediction result for the electric vehicle, the charging load prediction result including: the license plate number of the electric vehicle, the charging probability, the target charging station, the shortest time for the electric vehicle to reach the target charging station, and the charging load value of the electric vehicle.

6. The method of claim 5, wherein, The step of image processing of the vehicle images of the passing vehicles to identify electric vehicles among the passing vehicles and the facial information of the drivers of the electric vehicles, and determining the driver's range anxiety level based on the driver's facial information, includes: The vehicle picture is input into a trained YOLOv3 target detection model, and the license plate region and the driver face region of the past vehicle are output; The license plate information in the license plate region and the facial micro-expression information in the driver face region are extracted by using a trained OCR recognition model, the electric vehicle in the past vehicle is determined according to the license plate information, the driver range anxiety simulation value is calculated by inputting the facial micro-expression information into a comprehensive evaluation function, and the driver range anxiety principal component risk value is obtained by converting the driver range anxiety simulation value by using a conversion formula. The formula of the comprehensive evaluation function is: wherein score represents the driver range anxiety simulation value, a i represents the weight value of the i-th facial micro-expression key point, Z i represents the i-th facial micro-expression key point coordinate, and n represents the number of facial micro-expression key points; The conversion formula is: risk_value=[score+abs(min(score))]×10 Wherein, risk_value represents the driver range anxiety principal component risk value, and score represents the driver range anxiety simulation value.

7. The method of claim 5, wherein, The travel power consumption formula is: wherein E O,D is the travel power consumption, p s is the unit mileage travel power consumption parameter of the electric vehicle, L is the travel path of the electric vehicle, X i,j is the travel mileage of adjacent two points i, j on the travel path, i, j ∈ N R , N R is a city road network set, H i,j is the relative height difference of adjacent two points i, j in the vertical direction, H i,j = h j - h i , h i is the height of point i, h j is the height of point j, when H i,j < 0, the path [i, j] is a downhill section, when H i,j > 0, the path [i, j] is an uphill section, α i,j is the climbing coefficient or energy recovery efficiency coefficient of the electric vehicle overcoming gravity in unit relative height. a i,j The calculation formula is: wherein a c is the ramping coefficient, a d is the energy recovery efficiency coefficient; The device power consumption formula is: wherein, E T is the equipment power consumption of the electric vehicle, L is the driving path of the electric vehicle with the air conditioner turned on, X i,j is the driving distance of adjacent two points i, j on the driving path with the air conditioner turned on, X i,j is determined according to the vehicle air conditioner usage data of the mountain city, i, j ∈ N R , N R is the urban road network set, E R / L is the 100 km power consumption parameter of the electric vehicle under the condition of air conditioner heating / cooling, E0 is the 100 km power consumption parameter of the electric vehicle under the condition of air conditioner not turned on; The remaining power calculation formula is: E Z = E S - E O,D - E T where E Z is the remaining power, E S is the initial power, E O,D is the travel power consumption, and E T is the device power consumption.

8. The method of claim 5, wherein, The distance between the target charging station and the position where the electric vehicle generates charging demand is less than or equal to the cruising range of the electric vehicle, and the constraint condition is: 0 < x total (t) < M i (t) wherein x total (t) represents the distance between the target charging station and the location where the electric vehicle generates the charging demand, M i (t) represents the cruising range of the i-th electric vehicle; The shortest time consumption of the electric vehicle to reach the target charging station is: Wherein, t0 represents the time when the electric vehicle generates charging demand, t1 represents the shortest time consumption of the electric vehicle accessing the target charging station, V0 represents the average driving speed of the electric vehicle, X" represents the path from the starting point to the destination, L' i,j represents the distance between two adjacent points in the path, H' i,j represents the relative height difference between two adjacent points in the path in the vertical direction; The constraint condition of the charging load value of the electric vehicle is: wherein, represents the minimum amount of electricity for a battery discharge warning of the electric vehicle, represents the maximum amount of electricity for a battery charge of the electric vehicle, represents the current state of charge of the battery of the i-th electric vehicle.

9. The method of claim 5, wherein, Further comprising: The charging load growth prediction curve of each charging station is given by clustering analysis of the charging load prediction results of all electric vehicles going to the same charging station; The power output after the electric vehicle charging and discharging meets the customer demand, and the constraint condition of the power output is: wherein, represents the charging power during the electric vehicle accessing the charging pile, represents the discharging power during the electric vehicle accessing the charging pile, represents the expected charging power of the electric vehicle user, represents the initial power when the electric vehicle accesses the charging pile, η cha represents the charging efficiency of the electric vehicle, η dis represents the discharging efficiency of the battery, B i represents the battery capacity of the i-th electric vehicle, ε bat represents the battery loss coefficient of the electric vehicle; Whether the electric vehicle accessing the charging pile meets the time constraint condition is: Wherein, t off represents the time when the electric vehicle accesses the target charging pile, t on represents the expected time when the electric vehicle user sets to leave the charging pile, represents the maximum charging power.

10. The method of claim 5, wherein, The charging load prediction of the electric vehicle in the mountain city is coupled with the image data and the charging decision of the driver, which is a probability prediction method; After a large amount of electric vehicle data and a large amount of electric vehicle charging behavior data are collected in the mountain city, a charging probability prediction model of the electric vehicle used in the mountain city is trained by using a neural network algorithm, and the charging probability prediction model is used to predict the charging probability of the electric vehicle.

Images