Judgment method, system and equipment for charging demand of electric vehicle and medium

Abstract

The invention discloses a method, a system and equipment for judging the charging demand of an electric automobile and a medium, and relates to the technical field of automobile batteries. Comprising the following steps: acquiring historical charging behavior data and dynamically updated state data of the electric vehicle, and determining charging data including single driving data and single charging data; the destination position is determined through the single-time driving data, and the charge state before charging is determined through the single-time charging data; according to the remaining mileage and the charge state before charging, the distance from the electric vehicle to the nearest charging facility is determined, so that the charging demand and energy consumption of the electric vehicle are obtained; carrying out prediction analysis to obtain charging demand space-time distribution characteristics; and according to the charging demand space-time distribution characteristics, determining conditions of emergency charging demands, and determining the charging demand state of the electric vehicle. According to the invention, the emergency charging demand of the electric vehicle is determined more scientifically, accurately and efficiently.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile batteries, and in particular to a method, system, device and medium for determining charging demand of an electric vehicle. Background Art

[0002] As the world pays more and more attention to environmental protection and energy conservation, the popularity and application of electric vehicles (EVs) as clean and efficient means of transportation are gradually becoming a development trend in the global automotive industry. In actual use, the range of electric vehicles and the distribution of charging facilities are still key factors limiting their further promotion. Especially in remote areas, highway service areas and emergencies, electric vehicles may face charging difficulties, which affects the user's driving experience and the widespread application of electric vehicles.

[0003] In the existing technology, the battery capacity, mileage and remaining power of the electric vehicle are clearly defined. According to the travel plan, schedule and charging facilities, the appropriate charging method is selected. During the charging process, the charging status, power growth and charging speed are monitored in real time through the display screen of the charging device or the electric vehicle. After charging is completed, the status of the charging device and the electric vehicle is checked to ensure that everything is normal before disconnecting the charging device.

[0004] The defects of the above-mentioned prior art are that the determination of the charging demand of electric vehicles is affected by external factors (mileage, driving speed, remaining capacity, etc.), and it is difficult to accurately predict and judge the emergency charging demand of electric vehicles during actual driving. Summary of the invention

[0005] Based on this, it is necessary to provide a method, system, device and medium for determining the charging demand of electric vehicles in response to the above technical problems.

[0006] An embodiment of the present invention provides a method for determining charging demand of an electric vehicle, comprising:

[0007] Acquire historical charging behavior data and dynamically updated status data of the electric vehicle, and determine charging data of the electric vehicle including single driving data and single charging data according to the status data;

[0008] The destination location of the electric vehicle is determined through single driving data, and the state of charge of the electric vehicle before charging is determined through single charging data;

[0009] According to the remaining mileage of the electric vehicle to the destination and the state of charge before charging, the distance of the electric vehicle to the nearest charging facility is determined to obtain the charging demand and energy consumption of the electric vehicle;

[0010] Forecast and analyze the charging demand and energy consumption of electric vehicles to obtain the spatiotemporal distribution characteristics of charging demand;

[0011] According to the spatiotemporal distribution characteristics of charging demand, the conditions for determining emergency charging demand are formulated to determine the charging demand status of electric vehicles.

[0012] Optionally, determining the distance of the electric vehicle to the nearest charging facility specifically includes:

[0013] Collect vehicle GPS data, including latitude and longitude, driving speed and driving direction;

[0014] Use map matching algorithms to match GPS data with digital maps to confirm the vehicle's real-time location;

[0015] Set the search radius based on the real-time location of the vehicle to determine the charging facilities that exist within the search range;

[0016] The distance of the electric vehicle to the nearest charging facility is determined by the dynamic distance of the current remaining range.

[0017] Optionally, the risk of breakdown is assessed when determining the distance of the electric vehicle to the nearest charging facility, including:

[0018] The anchoring risk assessment indicators include throttle opening, vehicle speed preset value and output torque fed back by the motor controller;

[0019] Determine whether the electric vehicle is in a drivable state. If it is in a drivable state, further determine whether there are conditions that affect power. If there are conditions that affect power, the electric vehicle is at risk of breaking down;

[0020] If there is no condition that affects the power, it is determined whether the throttle opening is greater than the throttle opening setting value;

[0021] If the throttle opening is greater than the throttle opening setting value, it is further determined whether the vehicle speed is lower than the vehicle speed setting value. If the vehicle speed is lower than the vehicle speed setting value within the first setting time, the electric vehicle is at risk of breaking down.

[0022] If the vehicle speed is not lower than the vehicle speed setting value, it is determined whether the output torque feedback from the motor controller is lower than the output torque setting value. If the output torque feedback from the motor controller is lower than the output torque setting value within the second setting time, the electric vehicle is at risk of breaking down.

[0023] Optionally, the over-discharge risk is assessed when determining the distance of the electric vehicle to the nearest charging facility, including:

[0024] Over-discharge risk assessment indicators include battery SOC change rate, single cell voltage change rate, total battery voltage change rate, single cell temperature change rate and battery current change rate;

[0025] Establish an over-discharge risk assessment model to predict the over-discharge risk level based on the input features;

[0026] Obtain the data of various indicators during the discharge process of electric vehicles within a week, with the charging time as the horizontal axis and the corresponding indicator as the vertical axis, determine the threshold range of each indicator when the discharge is normal, and compare it with the abnormal data information of the day;

[0027] Determine the normal range of change rates for each indicator, compare the normal and abnormal indicator change rates, and determine risk identification indicators through data analysis;

[0028] The data of the vehicle-side battery is used for analysis, and the over-discharge risk status of the electric vehicle is evaluated within the specified threshold range by analyzing the data.

[0029] Optionally, the charging demand of the electric vehicle is obtained, specifically including:

[0030] The random forest algorithm RandomFrost based on the bagging algorithm Bagging and the lightweight gradient boosting machine LGBM based on the ensemble algorithm Boosting are used as the base models of the stacking ensemble learning method Stacking, and the lasso regression Lasso is used as the meta-model of the stacking ensemble learning method Stacking to establish a charging energy prediction model;

[0031] The Bayesian-Gaussian mixture algorithm is used to mark and delete abnormal charging energy values ​​in the vehicle charging energy, and the distribution statistics of the vehicle charging energy after deleting the abnormal charging energy values ​​are performed, so that the overall distribution of the vehicle charging energy approximately obeys the normal distribution; the calculation of the vehicle charging energy is expressed as follows:

[0032]

[0033] Among them, SOC (t) and SOC (t0) are the SOC values ​​at the current time and the time when charging starts, respectively, c(t) is the Coulomb efficiency, C n is the capacity of the battery; C energy is the cumulative charging energy of the first k charging moments, U (t) is the battery charging voltage at the current moment, I (t) The battery charging current at the current moment;

[0034] The random forest and L1 regularization method suitable for linear models in the Embedded method are used to perform feature screening on the calculated vehicle charging energy to obtain the screening features;

[0035] The screening features are predicted through the charging energy prediction model to obtain the charging demand of electric vehicles.

[0036] Optionally, the energy consumption of the electric vehicle is obtained, specifically including:

[0037] The driving data set is composed of the vehicle's charging start time, charging completion time data information, average daily charging frequency, charging duration, driving route information, battery capacity attenuation and vehicle operation data;

[0038] The multi-faceted energy consumption factors of the driving data set are constructed using data mining methods, and the vehicle characteristics are classified and screened by combining the Pearson correlation coefficient and random forest, which are divided into vehicle status characteristics and road condition dynamic characteristics;

[0039] The CatBoost algorithm, a category feature boosting algorithm under the Boosting framework of the integrated algorithm, is used to build a path driving state prediction model, which takes vehicle state characteristics and road condition dynamic characteristics as input and outputs the path vehicle energy consumption;

[0040] The core energy consumption prediction model is constructed using the extreme gradient boosting algorithm XGBoost under the integrated algorithm Boosting framework. The actual energy consumption of the vehicle during driving is calculated using the interval mapping energy consumption estimation method. The output of the path driving state prediction model and the vehicle state characteristics are integrated as input, and the vehicle energy consumption prediction result is output to obtain the energy consumption of the electric vehicle.

[0041] The embodiment of the present invention further provides a system for determining charging demand of an electric vehicle, comprising:

[0042] A data acquisition module, used to acquire historical charging behavior data and dynamically updated status data of the electric vehicle, and determine charging data of the electric vehicle including single driving data and single charging data according to the status data;

[0043] A data determination module, used to determine the destination position of the electric vehicle through single driving data, and to determine the state of charge of the electric vehicle before charging through single charging data;

[0044] A calculation module, used to determine the distance of the electric vehicle to the nearest charging facility according to the remaining mileage of the electric vehicle from the destination and the charge state before charging, so as to obtain the charging demand and energy consumption of the electric vehicle;

[0045] The prediction and analysis module is used to predict and analyze the charging demand and energy consumption of electric vehicles and obtain the spatiotemporal distribution characteristics of charging demand;

[0046] The determination module is used to formulate the determination conditions of the emergency charging demand according to the spatiotemporal distribution characteristics of the charging demand and to determine the charging demand status of the electric vehicle.

[0047] An embodiment of the present invention further provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of the above-mentioned electric vehicle charging demand determination method are implemented.

[0048] An embodiment of the present invention further provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned method for determining the charging demand of an electric vehicle are implemented.

[0049] Compared with the prior art, the above-mentioned method, system, device and medium for determining the charging demand of an electric vehicle provided by the embodiment of the present invention have the following beneficial effects:

[0050] The present invention obtains the historical charging behavior data and real-time status data of electric vehicles to accurately analyze the details of a single trip and charging, including key information such as the destination location and the state of charge before charging. Combining the remaining mileage and the power before charging, it can overcome the influence of external factors on the charging demand determination process, dynamically calculate the distance from the electric vehicle to the nearest charging station, and accurately evaluate the charging demand and energy consumption.

[0051] Furthermore, historical charging data is analyzed to determine whether the power level during charging exceeds the emergency threshold, revealing the user's charging habits and providing data support for the determination of emergency charging needs; based on the vehicle's real-time location and power level, the risk of breakdown and over-discharge is assessed, and endurance problems are discovered in a timely manner to prevent breakdowns caused by power exhaustion.

[0052] This not only improves the level of intelligent charging of electric vehicles, but also provides accurate decision-making support for charging service providers and users, promotes the sustainable development of the electric vehicle industry, reduces the risks and costs caused by improper charging, and makes the determination of emergency charging needs of electric vehicles more scientific, accurate and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 A schematic diagram of a method for determining charging demand of an electric vehicle provided in an embodiment;

[0054] Figure 2 A charging demand prediction flow chart of a method for determining charging demand of an electric vehicle provided in an embodiment. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0056] In the technical solutions disclosed in one or more embodiments, Figure 1As shown, a method for determining the charging demand of an electric vehicle comprises:

[0057] Acquire historical charging behavior data and dynamically updated status data of the electric vehicle, and determine charging data of the electric vehicle including single driving data and single charging data according to the status data;

[0058] The destination location of the electric vehicle is determined through single driving data, and the state of charge of the electric vehicle before charging is determined through single charging data;

[0059] According to the remaining mileage of the electric vehicle to the destination and the state of charge before charging, the distance of the electric vehicle to the nearest charging facility is determined to obtain the charging demand and energy consumption of the electric vehicle;

[0060] Forecast and analyze the charging demand and energy consumption of electric vehicles to obtain the spatiotemporal distribution characteristics of charging demand;

[0061] According to the spatiotemporal distribution characteristics of charging demand, the conditions for determining emergency charging demand are formulated to determine the charging demand status of electric vehicles.

[0062] By way of example, a specific embodiment of the present invention is presented.

[0063] Step 1

[0064] Obtain the historical charging behavior data and dynamically updated status data of the electric vehicle, and determine the charging data of the electric vehicle including single driving data and single charging data according to the status data. Determine the destination location of the electric vehicle through the single driving data, and determine the charge state of the electric vehicle before charging through the single charging data. The following steps are included:

[0065] Step 1.1 extracts static and dynamically updated status data of the electric vehicle from a designated data platform, and determines charging data of the electric vehicle based on the status data.

[0066] The charging data includes the single driving data and single charging data of the electric vehicle, wherein the single driving data includes the driving start time, driving end time, driving duration, driving start longitude, driving start latitude, driving end longitude, driving end latitude, driving start state of charge, driving end state of charge. The single charging data includes the electric vehicle unique identification, charging start time, charging end time, charging duration, charging start longitude, charging start latitude, charging start state of charge, charging end state of charge.

[0067] The destination location of the electric vehicle is determined through single driving data, and the charge state of the electric vehicle before charging is determined through single charging data.

[0068] Step 1.2 sets the emergency charging power threshold based on the total driving range of the electric vehicle and the warning factor.

[0069] The emergency charging power threshold is determined based on the total driving range of the electric vehicle and the warning coefficient.

[0070] The warning coefficient is determined in real time by the driver's driving attributes and dynamic driving data.

[0071] The driver's driving attributes include the driver's physical condition, driving skills, mentality and emotions, and reaction characteristics. The dynamic driving data of electric vehicles includes driving speed, acceleration, and acceleration distribution rate.

[0072] Determine whether the charge state of the electric vehicle before charging is greater than the emergency charging power threshold. If so, issue a charging warning prompt.

[0073] Step 2

[0074] According to the remaining mileage of the electric vehicle from the destination and the charge state before charging, the distance of the electric vehicle to the nearest charging facility is determined to obtain the charging demand and energy consumption of the electric vehicle, including the following steps:

[0075] Step 2.1 Accurately obtain the real-time geographic location of the vehicle and perform charging facility data retrieval.

[0076] In complex urban environments, GPS signals are easily blocked or interfered with, which may cause deviations in GPS positioning. A map matching algorithm is used to match the driving trajectory with the digital map to correct and confirm the real-time position of the vehicle.

[0077] Collect the vehicle's GPS data, including latitude and longitude, driving speed, driving direction, etc., find the road closest to the GPS data on the digital map, and combine multiple factors such as road network topology, vehicle driving direction and speed to determine the most likely precise location of the vehicle.

[0078] Based on the dynamic distance of the current remaining cruising range, the search radius is set according to the vehicle's location to determine the charging facilities within the search range. The system performs a search query in its established charging facility database, including location, type, power, and idle status. By comparing the vehicle location and the charging pile location, all charging facilities within the search radius are retrieved.

[0079] Step 2.2 Assess the risk of breakdown when determining the distance of the electric vehicle to the nearest charging facility.

[0080] The anchoring risk assessment indicators include throttle opening, vehicle speed preset value, and output torque feedback from the motor controller.

[0081] On this basis, the maximum-minimum method is used to normalize the risk assessment indicators to obtain the normalized decision matrix G = [g ij ],g ij ∈[0,1],g ij It is a positive and negative indicator.

[0082] Input anchoring risk assessment index weight value W = [w 1 ,w 2 ,…,w i ,…,w n ], calculate the weighted normalized decision matrix X for risk assessment = [x ij ], x ij According to the formula x ij =w j g ij calculate.

[0083] Determine the negative ideal solution of risk assessment and form the negative ideal solution vector F = [f j ], f j By formula calculate.

[0084] Calculate the Manhattan distance between the anchoring risk assessment index and the negative ideal solution:

[0085] The steps for judging the risk of an electric vehicle breaking down are as follows: judging whether the vehicle is in a drivable state; if so, further judging whether there are conditions that affect power; if so, the electric vehicle is at risk of breaking down; if there are no conditions that affect power, judging whether the throttle opening is greater than the throttle opening setting value; if the throttle opening is greater than the throttle opening setting value, further judging whether the vehicle speed is lower than the vehicle speed setting value; if the vehicle speed is lower than the vehicle speed setting value within the first setting time, the electric vehicle is at risk of breaking down; if the vehicle speed is not lower than the vehicle speed setting value, judging whether the output torque fed back by the motor controller is lower than the output torque setting value; if the output torque fed back by the motor controller is lower than the output torque setting value within the second setting time, the electric vehicle is at risk of breaking down.

[0086] Step 2.3: Electric vehicle over-discharge risk assessment based on multi-dimensional indicator data.

[0087] Over-discharge risk assessment indicators include battery SOC change rate, single cell voltage change rate, total battery voltage change rate, single cell temperature change rate, and battery current change rate.

[0088] Obtain data on various indicators during the discharge process of electric vehicles within a week, with charging time as the horizontal axis and the corresponding indicator as the vertical axis, determine the threshold range of each indicator when discharge is normal, and compare it with the abnormal data information of the day.

[0089] Determine the normal range of change rates for each indicator, compare the normal and abnormal indicator change rates, and determine risk identification indicators through data analysis.

[0090] The data of the vehicle-side battery is used for analysis, and the over-discharge risk status of the electric vehicle is evaluated within the specified threshold range by analyzing the data.

[0091] Step 3

[0092] The charging demand and energy consumption of electric vehicles are predicted and analyzed to obtain the spatiotemporal distribution characteristics of the charging demand; the judgment conditions for emergency charging demand are formulated according to the spatiotemporal distribution characteristics of the charging demand, and the charging demand status of the electric vehicle is judged, including the following steps:

[0093] Step 3.1 uses the random forest algorithm RandomFrost based on the bagging algorithm Bagging and the lightweight gradient boosting machine LGBM (Light Gradient Boosting Machine) based on the boosting algorithm Boosting as the base model of the stacking ensemble learning method Stacking, and lasso regression Lasso as the meta-model of the stacking ensemble learning method Stacking to establish a charging energy prediction model.

[0094] The Bayesian-Gaussian mixture algorithm is used to mark and delete abnormal charging energy values ​​in the vehicle charging energy, and the distribution statistics of the vehicle charging energy after deleting the abnormal charging energy values ​​are performed, so that the overall distribution of the vehicle charging energy approximately obeys the normal distribution. The calculation of the vehicle charging energy is expressed as follows:

[0095]

[0096]

[0097] Among them, SOC (t) and SOC (t0) are the SOC values ​​at the current time and the time when charging starts, respectively, c(t) is the Coulomb efficiency, C n is the capacity of the battery; C energy is the cumulative charging energy of the first k charging moments, U (t) is the battery charging voltage at the current moment, I (t) The battery charging current at the current moment.

[0098] Then, the random forest and L1 regularization method suitable for linear models in the Embedded method are used to perform feature screening on the calculated vehicle charging energy to obtain the screening features. The screening features include battery SOC, battery capacity value, battery average temperature value, battery consistency, charging power level, charging average current, and charging start time. The features screened out include battery SOC, capacity value, and charging average current.

[0099] The screening features are predicted through the charging energy prediction model to obtain the charging demand of electric vehicles.

[0100] Step 3.2 is based on the time-informed interval driving data set composed of the vehicle's charging start time, charging completion time data information, average daily charging frequency, charging duration, driving route information, battery capacity attenuation, and vehicle operation data, and combines the CatBoost algorithm and the XGBoost algorithm to build a vehicle trip energy consumption prediction model to obtain the vehicle trip energy consumption prediction result.

[0101] Specifically, the vehicle trip energy consumption prediction model includes a path driving state prediction model and a core energy consumption prediction model;

[0102] The construction method of the path driving state prediction model is as follows: use data mining methods to construct multi-faceted energy consumption factors, combine the Pearson correlation coefficient and random forest to classify and screen vehicle characteristics, and divide them into vehicle state characteristics and road condition dynamic characteristics. Vehicle state characteristics include static information of vehicle use including the vehicle's current mileage and longitude and latitude, and current battery information including current battery capacity, current battery SOC, and current battery SOH. Dynamic road condition characteristics refer to the average battery current and average speed of the vehicle on the current road section. Construct a path driving state prediction model based on the category feature boosting algorithm CatBoost under the integrated algorithm Boosting framework, take vehicle state characteristics and road condition dynamic characteristics as input, and predict the path vehicle energy consumption.

[0103] The construction method of the core energy consumption prediction model is as follows: use the interval mapping energy consumption estimation method to calculate the actual energy consumption of the vehicle during driving, integrate the output of the path driving state prediction model and the vehicle state characteristics as input, use the extreme gradient boosting algorithm XGBoost under the integrated algorithm Boosting framework, establish the core energy consumption prediction model, output the vehicle energy consumption prediction results, and obtain the energy consumption of the electric vehicle.

[0104] Step 3.3 predicts and analyzes the charging demand and energy consumption of electric vehicles to make spatiotemporal predictions of the charging demand in the target area and obtain the spatiotemporal distribution characteristics of the charging demand.

[0105] The process of spatiotemporal prediction of charging demand is as follows: Figure 2As shown, specifically, charging needs include recharging after short-distance travel, short-time charging during parking, preventive charging, emergency and backup charging.

[0106] Furthermore, the energy consumption of the vehicle under specific routes and driving conditions is predicted through the vehicle trip energy consumption prediction results. Combined with the prediction results of the vehicle battery charging energy, a spatiotemporal prediction model for charging demand is constructed. Taking into account the vehicle's driving route, energy consumption, battery charging energy, peak and trough periods of demand, and time and space changes, the charging demand of the target area in different time periods and locations is predicted.

[0107] Through the charging demand forecast results, the vehicle charging demand is determined, and the spatiotemporal statistics of the points with demand are carried out to obtain the time distribution of scattered charging demand during the peak charging period; combined with the charging demand decision results and the charging energy prediction model, the charging demand for different charging power levels is calculated, and the peak load charging amount of each power level in the daily average scattered charging demand is obtained, and the spatiotemporal distribution characteristics of the charging demand are obtained.

[0108] Step 4

[0109] According to the spatiotemporal distribution characteristics of charging demand, the determination conditions of emergency charging demand are formulated to determine the charging demand status of electric vehicles, including the following steps:

[0110] Step 4.1: Based on the spatiotemporal distribution characteristics of charging demand, formulate the criteria for determining emergency charging demand.

[0111] Collect spatiotemporal data on charging demand in the target area, including charging time, charging location, charging amount, traffic flow, parking lot usage, distance to charging stations, peak charging demand period, and remaining vehicle power.

[0112] Based on the operation data, the total number of electric vehicle users N and the total number of charging events M are extracted, thereby extracting the user's charging segment information.

[0113] Based on the charging segment information, the spatiotemporal characteristics of electric vehicle users are analyzed, and the spatiotemporal characteristics of charging behavior of electric vehicle users are explored.

[0114] Specifically, based on the extracted spatiotemporal distribution characteristics of charging demand, combined with the operating conditions of the power grid, the distribution of charging facilities, and user charging habits, the judgment conditions for emergency charging demand are formulated to accurately identify charging demand that may cause charging congestion, power grid overload, etc.

[0115] During peak hours of charging demand, since charging stations face greater load pressure, more sensitive judgment conditions are set to identify and respond to emergency charging needs.

[0116] The real-time load rate of the charging station reaches the preset peak load threshold, the average waiting time of users at the charging station exceeds the preset time threshold, the grid load reaches the limit of safe operation, the number of nearby charging stations is insufficient or the distance is too far, making it difficult for users to charge in time, which means that there is an emergency charging demand.

[0117] During non-peak hours of charging demand, ensure the continuity and stability of charging services, set relatively loose judgment conditions, monitor the usage of charging stations, grid load and user charging behavior, and set flexible judgment strategies to balance the load of charging stations at different times.

[0118] The remaining power of the user's vehicle is lower than the preset safety power threshold, and charging stations are scarce or far away in remote areas. The remaining power of the vehicle is insufficient to support the next charging station, which should be judged as an emergency charging demand.

[0119] Step 4.2 comprehensively considers time periods, facility distribution, special vehicle needs, and user personalized habits, and fine-tunes and optimizes the emergency charging determination model dynamically.

[0120] Specifically, the model inputs include the vehicle’s current power level, the location of the nearest charging station, and the time, and outputs the determination of emergency charging needs;

[0121] Dynamically adjust the emergency charging power threshold according to actual conditions. During peak hours or in areas with fewer charging facilities, when users have a more aggressive power management strategy, lower the emergency charging power threshold.

[0122] During non-peak hours of charging demand, in areas with dense charging stations and sufficient charging facilities, the load on charging stations and power grids is relatively small; for some special vehicles of electric vehicle rental services and shared travel services, their usage frequency and mileage are high; based on user's personalized needs, users prefer conservative power management strategies, and users continue to have concerns about cruising range, the emergency charging power threshold is increased.

[0123] Furthermore, by analyzing users' historical data, considering their unique driving and charging habits and incorporating them into the judgment model, a feedback loop mechanism is established to provide feedback based on actual usage, further optimize and adjust the judgment model, and improve its accuracy and practicality.

[0124] Step 4.3 Determine the charging demand status based on real-time data and personalized strategies on a typical day.

[0125] On a typical day, the current power of electric vehicles in the target area, the distance to the charging station, the peak period of charging demand, the remaining power data of the vehicle, the charging waiting time, and the use of facilities are updated in real time. According to the user's personalized power management strategy and driving habit data, the real-time distance between the electric vehicle's location and the nearest charging station, and whether the collection period is in peak special vehicle usage status.

[0126] The collected real-time data is input into the optimized emergency charging determination model. The model determines whether there is an emergency charging demand at present. The output results of the model include the determination result of the emergency charging demand and the corresponding emergency charging power threshold.

[0127] Predict the power consumption of electric vehicles on a typical day, as well as the number of charging times and charging amounts required, and comprehensively and accurately determine the charging demand status of electric vehicles on a typical day.

[0128] Based on the same inventive concept, an embodiment of the present invention further provides a system for determining charging demand of an electric vehicle, comprising:

[0129] The data acquisition module is used to acquire the historical charging behavior data and dynamically updated status data of the electric vehicle, and determine the charging data of the electric vehicle including single driving data and single charging data according to the status data.

[0130] The data determination module is used to determine the destination position of the electric vehicle through single driving data, and to determine the charge state of the electric vehicle before charging through single charging data.

[0131] The calculation module is used to determine the distance of the electric vehicle to the nearest charging facility according to the remaining mileage of the electric vehicle to the destination and the charge state before charging, so as to obtain the charging demand and energy consumption of the electric vehicle.

[0132] The prediction and analysis module is used to predict and analyze the charging demand and energy consumption of electric vehicles and obtain the spatiotemporal distribution characteristics of charging demand.

[0133] The determination module is used to formulate the determination conditions of the emergency charging demand according to the spatiotemporal distribution characteristics of the charging demand and to determine the charging demand status of the electric vehicle.

[0134] Based on the same inventive concept, the present invention also provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of the method for determining the charging demand of an electric vehicle are implemented.

[0135] Based on the same inventive concept, the present invention also provides a computer-readable storage medium containing instructions, and a computer program is stored on the storage medium. For example, a memory containing instructions, the above instructions can be executed by a processor of a computer device to complete the above method. For example, a non-temporary computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device. When the computer program is executed by the processor, the steps in the embodiment of the method for determining the charging demand of an electric vehicle can be implemented. The specific implementation method can be found in the method embodiment, which will not be repeated here.

[0136] The above embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for those of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for determining the charging demand of an electric vehicle, characterized in that: include: Acquire historical charging behavior data and dynamically updated status data of the electric vehicle, and determine charging data of the electric vehicle including single driving data and single charging data according to the status data; The destination location of the electric vehicle is determined through single driving data, and the state of charge of the electric vehicle before charging is determined through single charging data; According to the remaining mileage of the electric vehicle to the destination and the state of charge before charging, the distance of the electric vehicle to the nearest charging facility is determined to obtain the charging demand and energy consumption of the electric vehicle; Forecast and analyze the charging demand and energy consumption of electric vehicles to obtain the spatiotemporal distribution characteristics of charging demand; According to the spatiotemporal distribution characteristics of charging demand, the conditions for determining emergency charging demand are formulated to determine the charging demand status of electric vehicles.

2. A method for determining charging demand of an electric vehicle as claimed in claim 1, characterized in that: Determining the distance from the electric vehicle to the nearest charging facility specifically includes: Collecting GPS data of the vehicle, wherein the GPS data includes latitude and longitude, driving speed and driving direction; Use map matching algorithms to match GPS data with digital maps to confirm the vehicle's real-time location; Set the search radius based on the real-time location of the vehicle to determine the charging facilities that exist within the search range; The distance of the electric vehicle to the nearest charging facility is determined by the dynamic distance of the current remaining range.

3. A method for determining charging demand of an electric vehicle as claimed in claim 2, characterized in that: Assess the risk of breakdown when determining the distance of the electric vehicle to the nearest charging facility, including: The anchoring risk assessment indicators include throttle opening, vehicle speed preset value and output torque fed back by the motor controller; Determine whether the electric vehicle is in a drivable state. If it is in a drivable state, further determine whether there are conditions that affect power. If there are conditions that affect power, the electric vehicle is at risk of breaking down; If there is no condition that affects the power, it is determined whether the throttle opening is greater than the throttle opening setting value; If the throttle opening is greater than the throttle opening setting value, it is further determined whether the vehicle speed is lower than the vehicle speed setting value. If the vehicle speed is lower than the vehicle speed setting value within the first setting time, the electric vehicle is at risk of breaking down. If the vehicle speed is not lower than the vehicle speed setting value, it is determined whether the output torque feedback from the motor controller is lower than the output torque setting value. If the output torque feedback from the motor controller is lower than the output torque setting value within the second setting time, the electric vehicle is at risk of breaking down.

4. A method for determining charging demand of an electric vehicle as claimed in claim 2, characterized in that: The over-discharge risk assessment when determining the distance of the electric vehicle to the nearest charging facility specifically includes: Over-discharge risk assessment indicators include battery SOC change rate, single cell voltage change rate, total battery voltage change rate, single cell temperature change rate and battery current change rate; Obtain the data of various indicators during the discharge process of electric vehicles within a week, with the charging time as the horizontal axis and the corresponding indicator as the vertical axis, determine the threshold range of each indicator when the discharge is normal, and compare it with the abnormal data information of the day; Determine the normal range of change rates for each indicator, compare the normal and abnormal indicator change rates, and determine risk identification indicators through data analysis; The data of the vehicle-side battery is used for analysis, and the over-discharge risk status of the electric vehicle is evaluated within the specified threshold range by analyzing the data.

5. A method for determining charging demand of an electric vehicle as claimed in claim 1, characterized in that: The obtaining of the charging demand of the electric vehicle specifically includes: The random forest algorithm RandomFrost based on the bagging algorithm Bagging and the lightweight gradient boosting machine LGBM based on the ensemble algorithm Boosting are used as the base models of the stacking ensemble learning method Stacking, and the lasso regression Lasso is used as the meta-model of the stacking ensemble learning method Stacking to establish a charging energy prediction model; The Bayesian-Gaussian mixture algorithm is used to mark and delete abnormal charging energy values ​​in the vehicle charging energy, and the vehicle charging energy after deleting the abnormal charging energy values ​​is distributed statistically, so that the overall distribution of the vehicle charging energy approximately obeys the normal distribution; the calculation of the vehicle charging energy is expressed as follows: Among them, SOC (t) and SOC (t0) are the SOC values ​​at the current time and the time when charging starts, respectively, c(t) is the Coulomb efficiency, C n is the capacity of the battery; C energy is the cumulative charging energy of the first k charging moments, U (t) is the battery charging voltage at the current moment, I (t) The battery charging current at the current moment; The random forest and L1 regularization method suitable for linear models in the Embedded method are used to perform feature screening on the calculated vehicle charging energy to obtain the screening features; The screening features are predicted through the charging energy prediction model to obtain the charging demand of electric vehicles.

6. A method for determining charging demand of an electric vehicle as claimed in claim 1, characterized in that: The energy consumption of the electric vehicle is obtained, specifically including: The driving data set is composed of the vehicle's charging start time, charging completion time data information, average daily charging frequency, charging duration, driving route information, battery capacity attenuation and vehicle operation data; The multi-faceted energy consumption factors of the driving data set are constructed using data mining methods, and the vehicle characteristics are classified and screened by combining the Pearson correlation coefficient and random forest, which are divided into vehicle status characteristics and road condition dynamic characteristics; The CatBoost algorithm, a category feature boosting algorithm under the Boosting framework of the integrated algorithm, is used to build a path driving state prediction model, which takes vehicle state characteristics and road condition dynamic characteristics as input and outputs the path vehicle energy consumption; The core energy consumption prediction model is constructed using the extreme gradient boosting algorithm XGBoost under the integrated algorithm Boosting framework. The actual energy consumption of the vehicle during driving is calculated using the interval mapping energy consumption estimation method. The output of the path driving state prediction model and the vehicle state characteristics are integrated as input, and the vehicle energy consumption prediction result is output to obtain the energy consumption of the electric vehicle.

7. A system for determining charging demand of electric vehicles, characterized in that: include: A data acquisition module, used to acquire historical charging behavior data and dynamically updated status data of the electric vehicle, and determine charging data of the electric vehicle including single driving data and single charging data according to the status data; A data determination module, used to determine the destination position of the electric vehicle through single driving data, and to determine the state of charge of the electric vehicle before charging through single charging data; A calculation module, used to determine the distance of the electric vehicle to the nearest charging facility according to the remaining mileage of the electric vehicle from the destination and the charge state before charging, so as to obtain the charging demand and energy consumption of the electric vehicle; The prediction and analysis module is used to predict and analyze the charging demand and energy consumption of electric vehicles and obtain the spatiotemporal distribution characteristics of charging demand; The determination module is used to formulate the determination conditions of the emergency charging demand according to the spatiotemporal distribution characteristics of the charging demand and to determine the charging demand status of the electric vehicle.

8. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method for determining the charging demand of an electric vehicle as described in any one of claims 1-6 are implemented.

9. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for determining the charging demand of an electric vehicle described in any one of claims 1 to 6 are implemented.

Images