Electric vehicle charging time prediction methods, devices, systems, and storage media

Abstract

This invention discloses a method, device, system, and storage medium for predicting electric vehicle charging time, including: establishing a scatter plot of the vehicle's power distribution, a power interval distribution plot with 10kW intervals, and a power interval distribution plot with 1kW intervals, and analyzing them to obtain the boundary lines between fast charging and slow charging, and between supercharging and fast charging; based on the boundary lines between fast charging and slow charging, and between supercharging and fast charging, classifying the vehicle's historical charging data into slow charging data, fast charging data, and supercharging data; step S3, establishing a slow charging model, a fast charging model, and a supercharging model based on the slow charging data, fast charging data, and supercharging data; and predicting the possible charging mode when the vehicle arrives at any charging station based on the vehicle's historical charging and current charging geographical location information, time information, and driver driving model information. Using the technical solution of this invention, the charging time of electric vehicles can be predicted more accurately.

Description

Technical Field

[0001] This invention belongs to the field of electric vehicle technology, and particularly relates to a method, device, system, and storage medium for predicting electric vehicle charging time. Background Technology

[0002] With the continuous development of the electric vehicle industry, fast charging and supercharging technologies are constantly evolving, leading to faster charging speeds and more diverse charging modes. Currently, most electric vehicle charging time prediction methods only target slow charging and typically rely on a single factor such as the State of Charge (SOC) threshold to build models. However, in practical applications, slow charging time is influenced by multiple factors, including battery temperature and SOC, resulting in various slow charging models. Using a single model is insufficient to accurately reflect the actual charging process. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method, device, system and storage medium for predicting electric vehicle charging time. By selecting to distinguish the vehicle charging time prediction model into slow charging, fast charging and supercharging models, it can better fit the current development of electric vehicles and more accurately predict the charging time of electric vehicles.

[0004] To achieve the above objectives, the present invention adopts the following technical solution:

[0005] A method for predicting electric vehicle charging time includes:

[0006] Step S1: Establish a scatter plot of the vehicle's power distribution, a power interval distribution plot with an interval of 10 kW, and a power interval distribution plot with an interval of 1 kW, and analyze them to obtain the boundary line between fast charging and slow charging, and the boundary line between supercharging and fast charging.

[0007] Step S2: Based on the boundary between fast charging and slow charging, and the boundary between supercharging and fast charging, the vehicle's historical charging data is divided into slow charging data, fast charging data, and supercharging data.

[0008] Step S3: Based on the slow charging data, fast charging data, and supercharging data, establish the slow charging model, fast charging model, and supercharging model;

[0009] Step S4: Based on the vehicle's historical charging and current charging location information, time information, and driver driving model information, predict the possible charging mode when the vehicle arrives at any charging station.

[0010] The present invention also provides an electric vehicle charging time prediction device, comprising:

[0011] The first processing module is used to establish a scatter plot of the vehicle's power distribution, a power interval distribution plot with an interval of 10 kW, and a power interval distribution plot with an interval of 1 kW, and analyze them to obtain the boundary line between fast charging and slow charging, and the boundary line between supercharging and fast charging.

[0012] The second processing module is used to classify the vehicle's historical charging data into slow charging data, fast charging data, and supercharging data based on the boundary between fast charging and slow charging, and the boundary between supercharging and fast charging.

[0013] The third processing module is used to establish slow charging models, fast charging models, and supercharging models based on slow charging data, fast charging data, and supercharging data.

[0014] The fourth processing module is used to predict the possible charging mode when the vehicle arrives at any charging station based on the vehicle's historical charging and current charging location information, time information, and driver driving model information.

[0015] The present invention also provides an electric vehicle charging time prediction system, comprising: a memory and a processor, wherein the memory stores a computer program executed by the processor, and the computer program executes an electric vehicle charging time prediction method when executed by the processor.

[0016] The present invention also provides a storage medium storing a computer program that executes an electric vehicle charging time prediction method when running.

[0017] The beneficial effects of this invention are as follows:

[0018] 1) Based on various phenomena observed during the slow charging process of electric vehicles, multiple slow charging models have been established, thus encompassing a wider range of actual vehicle charging conditions. This allows for broader applicability to predict slow charging times for different types of vehicles under various driving environments, resulting in greater applicability.

[0019] 2) For the same vehicle, not only were multiple slow charging models established, but also fast charging and supercharging models were built based on fast charging and supercharging data differentiated from historical charging data. The fast charging and supercharging models, built by differentiating the data, have higher prediction accuracy. Establishing multiple models comprehensively covers various scenarios in actual vehicle charging, allowing for more accurate predictions of charging time.

[0020] 3) The pattern prediction model that combines the driver's historical charging location information, charging time information, and driver model selection preference information can continuously learn user habits and gradually achieve higher prediction accuracy. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0022] Figure 1 This is a flowchart of the electric vehicle charging time prediction method according to an embodiment of the present invention.

[0023] Figure 2 Flowchart for slow charging time prediction model I;

[0024] Figure 3 Flowchart for slow charging time prediction model II;

[0025] Figure 4 Flowchart for slow charging time prediction model III;

[0026] Figure 5 Flowchart for slow charging time prediction model IV;

[0027] Figure 6 The flowchart for slow charging time prediction model V;

[0028] Figure 7 Flowchart for slow charging time prediction model VI;

[0029] Figure 8 Flowchart for a fast charging time prediction model;

[0030] Figure 9 The flowchart for the fast charging current prediction model. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0033] Example 1:

[0034] like Figure 1 As shown, this embodiment of the invention provides a method for predicting electric vehicle charging time, including:

[0035] Step S1: Establish a scatter plot of the vehicle's power distribution, a power interval distribution plot with an interval of 10 kW, and a power interval distribution plot with an interval of 1 kW, and analyze them to obtain the boundary line between fast charging and slow charging, and the boundary line between supercharging and fast charging.

[0036] Step S2: Based on the boundary between fast charging and slow charging, and the boundary between supercharging and fast charging, the vehicle's historical charging data is divided into slow charging data, fast charging data, and supercharging data.

[0037] Step S3: Based on the slow charging data, fast charging data, and supercharging data, establish the slow charging model, fast charging model, and supercharging model;

[0038] Step S4: Based on the vehicle's historical charging and current charging location information, time information, and driver driving model information, predict the possible charging mode when the vehicle arrives at any charging station; increase the accuracy of the mode used when predicting vehicle charging time, thereby improving the accuracy of the prediction time.

[0039] As one embodiment of the present invention, in step S3, various slow charging models are established based on the various phenomena existing in the slow charging process of electric vehicles, thereby making them applicable to a wider range of actual vehicle charging situations. This allows for broader application to predicting slow charging time for different types of vehicles under different driving environments. The slow charging process of electric vehicles exhibits the following phenomena: 1. Charging is always performed with a constant current. 2. At the beginning of charging, a lower current value is used initially, during which the battery temperature gradually rises. After the battery temperature reaches a certain level, charging switches to a higher current, continuing until fully charged. 3. Charging also begins with a low current at the beginning, switching to a higher current after the battery temperature reaches a certain level. During the high-current charging phase, the battery temperature gradually decreases. When the temperature drops to a decreasing temperature threshold, the vehicle switches back to a low current, during which the battery temperature rises again. When the temperature rises to a increasing temperature threshold, charging switches back to a high current. This cycle continues until the vehicle is fully charged. 4. When the vehicle starts charging, it is charged with a certain current. When the vehicle reaches a certain SOC value, such as 80% or 90%, the charging current switches to a constant current until the vehicle is fully charged. Different vehicles have different slow charging modes due to differences in charging protocols, usage conditions, etc. Based on the analysis of the above four phenomena, this invention establishes multiple slow charging models. For example, Model I, which only considers phenomenon 1, involves the vehicle charging with a constant current. Charge the car until it is fully charged, as shown in the flowchart below. Figure 2 As shown. Only Model II exhibits phenomenon 2: the vehicle starts with a low current. During charging, the battery temperature begins to rise. The temperature rise coefficient is then used. Calculate the temperature at any given time: , (in the formula) Represents the temperature at any given time. This indicates the initial temperature at the beginning. This represents the time from the start to the target time. When the temperature rises to the rising temperature threshold... At that time, the vehicle switches to a high current. During charging, the vehicle uses current. The charging process is shown in the flowchart below. Figure 3 As shown. Only Model III exhibits phenomenon 3: the vehicle begins with a low current. During charging, the battery temperature begins to rise. The temperature rise coefficient is then used. Calculate the temperature at any given time: , (in the formula) Represents the temperature at any given time. This indicates the initial temperature at the beginning. This represents the time from the start to the target time. When the temperature rises to the rising temperature threshold... At that time, the vehicle switches to a high current. Charging begins. During the high-current charging phase, the battery temperature gradually decreases, and at this point, the temperature drop coefficient is applied. Calculate any temperature , (in the formula) Represents the temperature at any given time. This indicates the initial temperature when the temperature began to drop. This represents the time from when the temperature begins to decrease to the target time, when the temperature decreases to the temperature threshold. Then it will come from high current Charging turns into low current Charging begins, then the temperature rises again, and this cycle continues until charging is complete. A detailed flowchart is shown below. Figure 4 As shown. Charging Model IV: The vehicle begins charging with current. Charging begins, but the SOC reaches the threshold. At that time, the charging current will change, and the vehicle will charge with the changed current. Charge the vehicle until it is fully charged, as shown in the flowchart below. Figure 5 As shown. Charging Model V: The vehicle starts with a low current. During charging, the battery temperature begins to rise. The temperature rise coefficient is then used. Calculate the temperature at any given time: , (in the formula) Represents the temperature at any given time. This indicates the initial temperature at the beginning. This represents the time from the start to the target time. When the temperature rises to the rising temperature threshold... At that time, the vehicle switches to a high current. Charge until the SOC reaches the threshold. When the current becomes a constant value The vehicle is charged using this charging current until it is fully charged. A detailed flowchart is shown below. Figure 6 As shown. Charging Model VI: The vehicle starts with a low current. During charging, the battery temperature begins to rise. The temperature rise coefficient is then used. Calculate the temperature at any given time: , (in the formula) Represents the temperature at any given time. This indicates the initial temperature at the beginning. This represents the time from the start to the target time. When the temperature rises to the rising temperature threshold... At that time, the vehicle switches to a high current. Charging begins. During the high-current charging phase, the battery temperature gradually decreases, and at this point, the temperature drop coefficient is applied. Calculate any temperature , (in the formula) Represents the temperature at any given time. This indicates the initial temperature when the temperature began to drop. This represents the time from when the temperature begins to decrease to the target time, when the temperature decreases to the temperature threshold. Then it will come from high current Charging turns into low current The system charges, then the temperature rises again, and this cycle continues until the State of Charge (SOC) reaches the threshold. At this point, the current becomes a constant value. The vehicle is charged using this charging current until it is fully charged. A detailed flowchart is shown below. Figure 7 As shown.

[0040] Example 2:

[0041] This invention also provides a method for predicting electric vehicle charging time, comprising:

[0042] Step 1: The vehicle transmits its historical charging data to the target cloud. The cloud performs some preprocessing on the vehicle's historical charging data, such as: deleting data with abnormal SOC values, filling missing data with nearest neighbor values, using low-pass filtering to reduce high-frequency fluctuations in the data, and using median filtering to reduce accidental abrupt changes in the data.

[0043] Step 2: After preprocessing the data, plot a scatter plot of the vehicle's power distribution, a power interval distribution plot with 10kW intervals, and a power interval distribution plot with 1kW intervals. Analyze these power distribution plots: Based on the aggregation of charging data, obtain the power boundary lines between slow charging and fast charging, and between fast charging and supercharging. This divides the vehicle's historical charging data into slow charging data, fast charging data, and supercharging data. Then, analyze the vehicle's historical charging data to obtain the vehicle's OCV curve (the curve corresponding to the state of charge (SOC)), the vehicle's temperature rise coefficient, and the temperature drop coefficient.

[0044] Step 3: Construct vehicle slow charging, fast charging, and supercharging models.

[0045] The slow charging model for electric vehicles is established as follows: 1. A constant current is used throughout the charging process. 2. Initially, a lower current is used, causing the battery temperature to gradually rise. Once the battery temperature reaches a certain level, the current increases until the battery is fully charged. 3. Again, a low current is used initially, then increases until the battery temperature reaches a certain level. During this high-current phase, the battery temperature gradually decreases. When the temperature drops to a threshold, the current decreases again, causing the battery temperature to rise again. This process repeats until the battery is fully charged. 4. Initially, a certain current is used. When the battery reaches a certain State of Charge (SOC), such as 80% or 90%, the charging current changes to a constant current until the battery is fully charged. Different vehicles exhibit different slow charging modes due to variations in charging protocols and usage conditions. Based on the analysis of these four phenomena, this invention establishes multiple slow charging models. For example, charging model I, which only exhibits phenomenon 1: the vehicle operates at a constant current. Charge the car until it is fully charged, as shown in the flowchart below. Figure 2 As shown. Only charging model II for phenomenon 2: The vehicle starts with a low current. During charging, the battery temperature begins to rise. The temperature rise coefficient is then used. Calculate the temperature at any given time: , (in the formula) Represents the temperature at any given time. This indicates the initial temperature at the beginning. This represents the time from the start to the target time. When the temperature rises to the rising temperature threshold... At that time, the vehicle switches to a high current. During charging, the vehicle uses current. The charging process is shown in the flowchart below. Figure 3As shown. Only charging model III with phenomenon 3: The vehicle starts with a low current. During charging, the battery temperature begins to rise. The temperature rise coefficient is then used. Calculate the temperature at any given time: , (in the formula) Represents the temperature at any given time. This indicates the initial temperature at the beginning. This represents the time from the start to the target time. When the temperature rises to the rising temperature threshold... At that time, the vehicle switches to a high current. Charging begins. During the high-current charging phase, the battery temperature gradually decreases, and at this point, the temperature drop coefficient is applied. Calculate any temperature , (in the formula) Represents the temperature at any given time. This indicates the initial temperature when the temperature began to drop. This represents the time from when the temperature begins to decrease to the target time, when the temperature decreases to the temperature threshold. Then it will come from high current Charging turns into low current Charging begins, then the temperature rises again, and this cycle continues until charging is complete. A detailed flowchart is shown below. Figure 4 As shown. Charging Model IV: The vehicle begins charging with current. Charging begins, but the SOC reaches the threshold. At that time, the charging current will change, and the vehicle will charge with the changed current. Charge the vehicle until it is fully charged, as shown in the flowchart below. Figure 5 As shown. Charging Model V: The vehicle starts with a low current. During charging, the battery temperature begins to rise. The temperature rise coefficient is then used. Calculate the temperature at any given time: , (in the formula) Represents the temperature at any given time. This indicates the initial temperature at the beginning. This represents the time from the start to the target time. When the temperature rises to the rising temperature threshold... At that time, the vehicle switches to a high current. Charge until the SOC reaches the threshold. When the current becomes a constant value The vehicle is charged using this charging current until it is fully charged. A detailed flowchart is shown below. Figure 6 As shown. Charging Model VI: The vehicle starts with a low current. During charging, the battery temperature begins to rise. The temperature rise coefficient is then used. Calculate the temperature at any given time: , (in the formula) Represents the temperature at any given time. This indicates the initial temperature at the beginning. This represents the time from the start to the target time. When the temperature rises to the rising temperature threshold... At that time, the vehicle switches to a high current. Charging begins. During the high-current charging phase, the battery temperature gradually decreases, and at this point, the temperature drop coefficient is applied. Calculate any temperature , (in the formula) Represents the temperature at any given time. This indicates the initial temperature when the temperature began to drop. This represents the time from when the temperature begins to decrease to the target time, when the temperature decreases to the temperature threshold. Then it will come from high current Charging turns into low current The system charges, then the temperature rises again, and this cycle continues until the State of Charge (SOC) reaches the threshold. At this point, the current becomes a constant value. The vehicle is charged using this charging current until it is fully charged. A detailed flowchart is shown below. Figure 7 As shown.

[0046] Vehicle fast charging model establishment: Data analysis and theoretical analysis revealed a correlation between vehicle charging current and factors such as voltage, SOC, and temperature. Analysis of vehicle fast charging data did not reveal obvious model characteristics, so machine learning methods were used to predict the charging current, selecting voltage, SOC, minimum temperature, and maximum temperature as influencing factors. Comparison of the performance of various machine learning algorithms, including AdaBoost regression, K-Nearest Neighbor (KNN) regression, XGBoost regression, and CatBoost regression, showed that KNN and AdaBoost regression performed better for most vehicles. Therefore, the more universally applicable KNN and AdaBoost regression algorithms were chosen to predict the charging current. The fast charging model was trained using fast charging data previously extracted from historical vehicle charging data. The steps for obtaining the vehicle's charging time using the fast charging model are as follows: First, collect the vehicle's voltage, current, SOC, minimum temperature, maximum temperature, and battery capacity at the start of charging. The step size is set to 1 second. The current multiplied by 1 second gives the amount of charge charged during this time period. ,in , (in the formula) This represents the charging current during this time period (t represents time). The amount of charge being charged. Divide by vehicle capacity In addition to the vehicle's previous moment The vehicle can be obtained at the next moment. Value. The formula is: . ( It is the vehicle from the previous moment. , The vehicle at the next moment value, This refers to the amount of charge added during this time period. (This refers to the vehicle's capacity). According to... By analyzing the vehicle's OCV curve, the voltage value of the vehicle at the next moment can be obtained. Then, based on the vehicle's temperature rise coefficient and temperature fall coefficient, the highest and lowest temperatures at the next moment can be calculated. At this point, the vehicle's voltage, SOC, minimum temperature, and maximum temperature at the next moment are known. Using the established fast-charging model as input, the fast-charging model will output the current for the next moment. The flowchart for fast-charging current prediction is shown below. Figure 9 As shown. At this point, the voltage, current, vehicle SOC, minimum temperature, and maximum temperature for the next time step have been obtained. Repeating the above steps will yield the vehicle data for the time step after that. Continue executing the above steps until the vehicle's SOC reaches the target SOC, and count the number of repetitions. This refers to the charging time. A flowchart illustrating the prediction of fast charging time is shown below. Figure 8 As shown.

[0047] Vehicle Supercharging Model Establishment: The charging time prediction process of the vehicle supercharging model is basically the same as that of fast charging. Regarding the selection of machine learning methods, after comparing several methods again, K-nearest neighbor regression and AdaBoost regression were still chosen to predict the charging current. The steps for obtaining the vehicle's charging time using the supercharging model are similar to those for fast charging: First, the vehicle's voltage, current, SOC, minimum temperature, maximum temperature, and battery capacity at the start of charging are collected. The step size is set to 1 second. The current multiplied by 1 second gives the amount of charge charged during this time period. ,in , (in the formula) This represents the charging current during this time period (t represents time). The amount of charge being charged. Divide by vehicle capacity In addition to the vehicle's previous moment The vehicle can be obtained at the next moment. Value. The formula is: . ( It is the vehicle from the previous moment. , The vehicle at the next moment value, This refers to the amount of charge added during this time period. (This refers to the vehicle's capacity). According to... By analyzing the vehicle's OCV curve, the voltage value of the vehicle at the next moment can be obtained. Then, based on the vehicle's temperature rise coefficient and temperature fall coefficient, the highest and lowest temperatures at the next moment can be calculated. At this point, the vehicle's voltage, SOC, minimum temperature, and maximum temperature for the next moment are known. Next, the established vehicle fast-charging model is input with these values, and the fast-charging model will output the current for the next moment. At this point, the voltage, current, vehicle SOC, minimum temperature, and maximum temperature for the next moment are all obtained. Repeating the above steps yields the vehicle data for the moment after that. This process continues until the vehicle's SOC reaches the target SOC, and the number of repetitions is counted. This refers to the charging time.

[0048] Based on the established six slow charging models, the slow charging model parameters for each vehicle are found from the vehicle's slow charging data, and a slow charging model for each vehicle is established. Then, using the vehicle's fast charging and supercharging data, fast charging and supercharging models for each vehicle are established respectively.

[0049] Step 4: When the vehicle arrives at any charging station, upload the vehicle's geographical information, time information, and charging station current information. Based on the uploaded geographical and time information, search the vehicle's historical charging records to determine the most likely charging mode at that moment. Then, during charging, compare the charging data with the initially selected charging mode for compatibility assessment. If subsequent charging data, such as current, voltage, and SOC, are close to the data predicted by the selected mode, continue using the selected model to predict charging time. If the subsequent actual charging data deviates significantly from the data predicted by the selected model, switch models for deviation analysis and select the model with smaller deviation for subsequent charging time prediction. After charging is complete, upload the charging process data to the cloud. Use the newly added charging data to optimize the vehicle's charging model, increasing the accuracy of the initially selected model and the accuracy of charging time prediction.

[0050] Example 3:

[0051] The present invention also provides an electric vehicle charging time prediction device, comprising:

[0052] The first processing module is used to establish a scatter plot of the vehicle's power distribution, a power interval distribution plot with an interval of 10 kW, and a power interval distribution plot with an interval of 1 kW, and analyze them to obtain the boundary line between fast charging and slow charging, and the boundary line between supercharging and fast charging.

[0053] The second processing module is used to classify the vehicle's historical charging data into slow charging data, fast charging data, and supercharging data based on the boundary between fast charging and slow charging, and the boundary between supercharging and fast charging.

[0054] The third processing module is used to establish slow charging models, fast charging models, and supercharging models based on slow charging data, fast charging data, and supercharging data.

[0055] The fourth processing module is used to predict the possible charging mode when the vehicle arrives at any charging station based on the vehicle's historical charging and current charging location information, time information, and driver driving model information.

[0056] Example 4:

[0057] This invention also provides an electric vehicle charging time prediction system, including: a memory and a processor, wherein the memory stores a computer program executed by the processor, and the computer program executes an electric vehicle charging time prediction method when run by the processor.

[0058] Example 5:

[0059] This invention also provides a storage medium storing a computer program that executes an electric vehicle charging time prediction method during runtime.

[0060] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for predicting charging time for electric vehicles, characterized in that, include: Step S1: Establish a scatter plot of the vehicle's power distribution, a power interval distribution plot with an interval of 10 kW, and a power interval distribution plot with an interval of 1 kW, and analyze them to obtain the boundary line between fast charging and slow charging, and the boundary line between supercharging and fast charging. Step S2: Based on the boundary between fast charging and slow charging, and the boundary between supercharging and fast charging, the vehicle's historical charging data is divided into slow charging data, fast charging data, and supercharging data. Step S3: Based on the slow charging data, fast charging data, and supercharging data, establish the slow charging model, fast charging model, and supercharging model; The establishment of the vehicle slow charging model includes: establishing multiple slow charging models. Charging Model I: The vehicle is charged with a constant current until it is fully charged; Charging Model II: The vehicle starts charging with a low current, at which point the battery temperature begins to rise. The temperature at any given time is calculated using the temperature rise coefficient. When the temperature rises to the temperature rise threshold, the vehicle switches to high current for charging. The vehicle charges with current until charging is complete. Charging Model III: The vehicle starts charging with a low current, at which point the battery temperature begins to rise. The temperature at any given time is calculated using the temperature rise coefficient. When the temperature rises to the rising temperature threshold, the vehicle switches to high current charging. During the high current charging phase, the battery temperature gradually decreases. At this time, the temperature at any given time is calculated based on the temperature drop coefficient. When the temperature drops to the falling temperature threshold, the charging will switch from high current charging to low current charging. Then the temperature rises again, and the cycle continues until charging is complete. Charging Model IV: The vehicle begins charging with current. I 1 Charging begins, but when the State of Charge (SOC) reaches the threshold, the charging current changes, and the vehicle charges with the changed current. I 2 Charge the vehicle until it is fully charged; Charging Model V: The vehicle starts with a low current. I 1 During charging, the battery temperature begins to rise. The temperature at any given moment is calculated using a temperature rise coefficient. When the temperature reaches a threshold value, the vehicle switches to a high current. I 2 During charging, the current becomes constant when the SOC reaches the threshold. I 3 Charge the vehicle with this charging current until it is fully charged. Charging Model VI: The vehicle starts with a low current. I 1 During charging, the battery temperature begins to rise. The temperature at any given moment is calculated using a temperature rise coefficient. When the temperature reaches a threshold value, the vehicle switches to a high current. I 2 During charging, the battery temperature gradually decreases during the high-current charging phase. Based on the temperature drop coefficient, the temperature at any given moment is calculated. When the temperature drops to a threshold value, the high-current charging will resume. I 2 Charging turns into low current I 1 The system begins charging, then the temperature rises again, and this cycle continues until the State of Charge (SOC) reaches the threshold, at which point the current becomes constant. I 3 Charge the vehicle with this charging current until it is fully charged. The fast charging model for vehicles includes: training the fast charging model using fast charging data differentiated from historical vehicle charging data; selecting voltage, SOC, minimum temperature, and maximum temperature as influencing factors to predict current; and using the fast charging model to obtain the vehicle's charging time. The vehicle supercharging model establishment includes: constructing a supercharging model using fast charging data differentiated from historical vehicle charging data, and predicting charging current and charging time; Step S4: Based on the vehicle's historical charging and current charging location information, time information, and driver driving model information, predict the possible charging mode when the vehicle arrives at any charging station. Specifically, when the vehicle arrives at any charging station, upload the vehicle's geographical information, time information, and charging station current information. Based on the uploaded vehicle geographical and time information, search in the vehicle's historical charging records to determine the most likely charging mode for the vehicle at this time. Then, during charging, compare the charging data with the initially selected charging mode for compatibility assessment. If the subsequent charging data is close to the data predicted by the selected mode, continue to use the selected model to predict the charging time. If the subsequent actual charging data deviates significantly from the data predicted by the selected model, change the model for deviation analysis and select the model with smaller deviation for subsequent charging time prediction.

2. An electric vehicle charging time prediction device, characterized in that, include: The first processing module is used to establish a scatter plot of the vehicle's power distribution, a power interval distribution plot with an interval of 10 kW, and a power interval distribution plot with an interval of 1 kW, and analyze them to obtain the boundary line between fast charging and slow charging, and the boundary line between supercharging and fast charging. The second processing module is used to classify the vehicle's historical charging data into slow charging data, fast charging data, and supercharging data based on the boundary between fast charging and slow charging, and the boundary between supercharging and fast charging. The third processing module is used to establish slow charging models, fast charging models, and supercharging models based on slow charging data, fast charging data, and supercharging data. The establishment of the vehicle slow charging model includes: establishing multiple slow charging models. Charging Model I: The vehicle is charged with a constant current until it is fully charged; Charging Model II: The vehicle starts charging with a low current, at which point the battery temperature begins to rise. The temperature at any given time is calculated using the temperature rise coefficient. When the temperature rises to the temperature rise threshold, the vehicle switches to high current for charging. The vehicle charges with current until charging is complete. Charging Model III: The vehicle starts charging with a low current, at which point the battery temperature begins to rise. The temperature at any given time is calculated using the temperature rise coefficient. When the temperature rises to the rising temperature threshold, the vehicle switches to high current charging. During the high current charging phase, the battery temperature gradually decreases. At this time, the temperature at any given time is calculated based on the temperature drop coefficient. When the temperature drops to the falling temperature threshold, the charging will switch from high current charging to low current charging. Then the temperature rises again, and the cycle continues until charging is complete. Charging Model IV: The vehicle begins charging with current. I 1 Charging begins, but when the State of Charge (SOC) reaches the threshold, the charging current changes, and the vehicle charges with the changed current. I 2 Charge the vehicle until it is fully charged; Charging Model V: The vehicle starts with a low current. I 1 During charging, the battery temperature begins to rise. The temperature at any given moment is calculated using a temperature rise coefficient. When the temperature reaches a threshold value, the vehicle switches to a high current. I 2 During charging, the current becomes constant when the SOC reaches the threshold. I 3 Charge the vehicle with this charging current until it is fully charged. Charging Model VI: The vehicle starts with a low current. I 1 During charging, the battery temperature begins to rise. The temperature at any given moment is calculated using a temperature rise coefficient. When the temperature reaches a threshold value, the vehicle switches to a high current. I 2 During charging, the battery temperature gradually decreases during the high-current charging phase. Based on the temperature drop coefficient, the temperature at any given moment is calculated. When the temperature drops to a threshold value, the high-current charging will resume. I 2 Charging turns into low current I 1 The system begins charging, then the temperature rises again, and this cycle continues until the State of Charge (SOC) reaches the threshold, at which point the current becomes constant. I 3 Charge the vehicle with this charging current until it is fully charged. The fast charging model for vehicles includes: training the fast charging model using fast charging data differentiated from historical vehicle charging data; using machine learning methods to select voltage, SOC, minimum temperature, and maximum temperature as influencing factors to predict current; and using the fast charging model to obtain the vehicle's charging time. The vehicle supercharging model establishment includes: constructing a supercharging model using fast charging data differentiated from historical vehicle charging data, and predicting charging current and charging time; The fourth processing module is used to predict the possible charging mode when a vehicle arrives at any charging station based on the vehicle's historical charging and current charging location information, time information, and driver driving model information. Specifically, when a vehicle arrives at any charging station, it uploads the vehicle's geographical information, time information, and charging station current information. Based on the uploaded vehicle geographical and time information, it searches in the vehicle's historical charging records to determine the most probable charging mode for the vehicle at this time. Then, during charging, it compares the charging data with the initially selected charging mode to determine the compatibility. If the subsequent charging data is close to the data predicted by the selected mode, the selected model is used to predict the charging time. If the subsequent actual charging data deviates significantly from the data predicted by the selected model, the model is changed for deviation analysis, and the model with smaller deviation is selected for subsequent charging time prediction.

3. An electric vehicle charging time prediction system, characterized in that, include: The device includes a memory and a processor, wherein the memory stores a computer program that is executed by the processor, the computer program performing the electric vehicle charging time prediction method as described in claim 1 when executed by the processor.

4. A storage medium, characterized in that, The storage medium stores a computer program, which executes the electric vehicle charging time prediction method as described in claim 1 when it runs.

Images