Battery replacing method of battery replacing station with intelligent battery replacing guiding function

Abstract

The invention provides a battery swap station battery swap method with an intelligent battery swap guiding function, and the method comprises the steps: obtaining the navigation information of a user navigating to the vicinity of a battery swap station, stripping the factors capable of accurately predicting the arrival battery swap probability from big data, carrying out the independent prediction, and solving a problem that the battery swap station does not use a map to navigate to the station in history. However, data which cannot be accurately predicted, such as user data of station-arriving battery replacement, are subjected to fuzzy prediction and then are processed through the fusion model, so that prediction of the battery replacement peak of the battery replacement flow of the battery replacement station is integrally improved, and the user experience is improved by intelligently guiding the user to perform station-arriving battery replacement.

Description

Technical Field

[0001] The present invention relates to the battery swapping guidance of an electric vehicle swapping station, and particularly to a battery swapping method for an electric vehicle swapping station with an intelligent battery swapping guidance function. Background Art

[0002] In the prior art, predicting the battery swapping flow of an electric vehicle swapping station is based on the battery swapping flow data at historical time nodes or on the number of electric vehicles near the swapping station and the battery power to predict the incoming battery swapping flow. The accuracy of the prediction is relatively low, and it has little practical reference significance. When the predicted incoming battery swapping flow is large, it is unable to intelligently guide users to swap batteries nearby or at off-peak times, which will lead to battery swapping congestion and waste a lot of time for users. Therefore, there is an urgent need to provide a battery swapping method with a specific intelligent battery swapping guidance function to accurately predict the peak battery swapping period of the swapping station, intelligently guide users to avoid the peak, and improve the user experience. Summary of the Invention

[0003] The object of the present invention is to provide a battery swapping method for an electric vehicle swapping station with an intelligent battery swapping guidance function. The swapping station includes a swapping station control system, which is used to control the battery replacement of electric vehicles and guide users to replace batteries at off-peak times. The method steps are as follows: S1: The swapping station control system obtains in real time user data navigated to the vicinity of the swapping station by multiple map navigation platforms and the types of user navigation devices. The multiple map navigation platforms include maps such as Baidu, Amap, and Tencent Maps. The types of navigation devices include networked vehicle-mounted terminals, mobile phone terminals, and professional navigation devices. S1-1 When the type of navigation device is a vehicle-mounted terminal, obtain the vehicle version model through the map navigation platform, and search for vehicle information in the vehicle information database according to the vehicle version model. Determine whether it is an electric vehicle. When it is an electric vehicle, obtain the remaining battery power, location information, and weather information of the vehicle through the map navigation platform. Input the vehicle information, remaining battery power, location information, and weather information data into the first battery swapping prediction model, and predict and output the incoming battery swapping probability, arrival time, remaining battery power upon arrival, and charging curve. Calculate the number of incoming battery swapping users according to the number of vehicle-mounted terminals with the type of navigation device and the predicted incoming battery swapping probability. The prediction of the incoming battery swapping probability by the first battery swapping prediction model is obtained by modeling manually according to the influence of each factor on the possibility of incoming battery swapping.

[0004] Furthermore, the first battery swapping prediction model is obtained by training the information of historical data navigated to the vicinity of the swapping station and the data information of incoming battery swapping by inputting them into the deepseek large model.

[0005] The first battery swapping prediction model can call the arrival time prediction module of the map navigation platform to predict the arrival time and driving speed, and then predict the remaining battery power upon arrival according to the remaining battery power of the vehicle, power consumption, driving speed, and weather temperature information.

[0006] S1-2: When the navigation device type is a mobile phone terminal, obtain the vehicle information data, historical navigation destination data, historical charging orders, historical charging power, terminal location information, and weather information bound to the user's navigation terminal through the map navigation platform; input the vehicle information data, historical navigation destination data, historical charging orders, historical charging power, terminal location information, and weather information into the second battery swapping prediction model, and predict and output the probability of arriving at the station for battery swapping, arrival time, and battery swapping power. Calculate the number of users arriving at the station for battery swapping based on the number of mobile phone terminals as the navigation device type and the predicted probability of arriving at the station for battery swapping.

[0007] The second battery swapping prediction model predicts the probability of arriving at the station for battery swapping based on the above factors. The second battery swapping prediction model calls the map navigation platform to predict the arrival time, and predicts the battery swapping power based on the historical charging orders and charging power.

[0008] The second battery swapping prediction model calls the map navigation platform to predict the arrival time, and predicts the battery swapping power based on the historical charging orders and charging power.

[0009] The prediction of the battery swapping power based on the historical charging orders and charging power is as follows: obtain the charging time, charging power, and charging station location information from the historical charging orders, merge the charging powers within the same charging station during the same charging time period into one charging order, merge the charging orders of charging stations within a certain distance during the same charging time period into one charging order, and calculate the mean value of the merged charging orders to obtain the predicted battery swapping power.

[0010] S1-3: When the navigation device type is a professional navigation device, obtain its historical navigation destination data through the map navigation platform, identify the number of times the destination is a charging and swapping station and the arrival time and departure time, and obtain its location information through the map navigation platform; input the historical navigation destination data, arrival time at the destination, departure time from the destination, and professional navigation device location information into the third battery swapping prediction model to predict and output the probability of arriving at the station for battery swapping, arrival time, and battery swapping power. Calculate the number of users arriving at the station for battery swapping based on the number of professional navigation devices as the navigation device type and the predicted probability of arriving at the station for battery swapping.

[0011] S2: The battery swapping station control system obtains the historical battery swapping user data of the battery swapping station, identifies the non-map navigation arrival user data, and inputs it into the fourth battery swapping prediction model to predict and output the number of users arriving at the station for battery swapping, battery swapping power, and arrival time for battery swapping.

[0012] S3: The battery swapping station control system inputs the prediction output data of the first, second, third, and fourth battery swapping prediction models, as well as the number of users arriving at the station for battery swapping, into the arrival battery swapping fusion prediction model to predict and output the time series of the arrival battery swapping power and the number of battery swapping users.

[0013] S4: The battery swapping station control system guides users to swap batteries nearby or at off-peak times according to the predicted battery swapping power, the time series of the number of battery swapping users, and the battery swapping capacity output by the arrival battery swapping integration prediction model.

[0014] The battery swapping capacity is predicted based on the number of fully charged batteries and the maximum charging power of the battery swapping station; when the predicted battery swapping power for a certain time period by the arrival battery swapping integration prediction model exceeds the battery swapping capacity of the battery swapping station, corresponding numbers of users are selected to push reminder messages according to the exceeded battery swapping power. The reminder messages include battery swapping tension reminders, as well as predicted battery swapping low-peak times, nearby shopping malls, parks and other information.

[0015] Beneficial effects: The present invention provides a battery swapping method for a battery swapping station with an intelligent battery swapping guidance function. By obtaining the navigation information of users navigating to the vicinity of the battery swapping station, the factors that can accurately predict the probability of arriving at the station for battery swapping are stripped from big data for separate prediction, and the user data that cannot be accurately predicted, such as users who have arrived at the station for battery swapping without using map navigation in the history of the battery swapping station, are fuzzily predicted. Then, through the fusion model, the prediction of the battery swapping flow peak of the battery swapping station is improved as a whole, and the user experience is improved by intelligently guiding users to arrive at the station for battery swapping.

[0016] In the present invention, during the prediction process of arriving at the station for battery swapping and the remaining battery power and battery swapping amount of the vehicle when arriving at the station, various factors considered are the laws summarized from long-term engagement in battery swapping work, and the various factors considered in the prediction process are the innovative contributions to the technology. Brief Description of the Drawings

[0017] Figure 1 It is a flowchart of intelligent battery swapping guidance. Detailed Embodiment

[0018] The embodiment of the present invention provides a battery swapping method for a battery swapping station with an intelligent battery swapping guidance function. The battery swapping station includes a battery swapping station control system, which is used to control electric vehicles to replace batteries and guide users to replace batteries during off-peak hours. The guiding method steps are as follows: S1: The battery swapping station control system real-time obtains user data and user navigation device types that navigate to the vicinity of the battery swapping station from multiple map navigation platforms; the multiple map navigation platforms include maps such as Baidu, Amap, and Tencent; the navigation device types include networked in-vehicle terminals, mobile phone terminals, and professional navigation devices. The vicinity of the battery swapping station may be within a radius of 200 m around the battery swapping station. In most cases, when a user navigates to the battery swapping station for battery swapping, the destination selected through the map is usually near the destination and cannot be accurately located at the destination. Therefore, when making predictions in the present invention, a certain distance range near the battery swapping station is selected. The user data obtained by the control system of the battery swapping station is obtained with the user's authorization and consent.

[0019] S1-1 When the navigation device type is an in-vehicle terminal, obtain the vehicle version model through the map navigation platform, and search for vehicle information in the vehicle information database according to the vehicle version model; the vehicle information includes vehicle type and vehicle power battery information, and the vehicle type includes fuel vehicles and electric vehicles; the battery information includes battery capacity and charging characteristics; determine whether it is an electric vehicle, and when it is an electric vehicle, obtain the remaining battery power, location information, and weather information of the vehicle through the map navigation platform; input the data such as the vehicle information, remaining battery power, location information, and weather information into the first battery swapping prediction model, and predict and output the probability of arriving at the station for battery swapping, arrival time, remaining battery power at arrival, and charging curve. Calculate the number of users arriving at the station for battery swapping according to the number of in-vehicle terminals as the navigation device type and the predicted probability of arriving at the station for battery swapping. The calculation method is that when the predicted vehicle arrival battery swapping probability is greater than the threshold, it is considered that the user arrives at the station for battery swapping, and the number of users arriving at the station for battery swapping is incremented by 1, and the threshold is 80%.

[0020] The vehicle information database is used to collect the vehicle configuration information of various brands, models, and years, and can be obtained through public channels such as the website of the Ministry of Industry and Information Technology, automobile sales websites, and automobile manufacturer websites.

[0021] Information such as the vehicle battery size, charging characteristics, remaining battery level, location information, and weather is related to the battery replacement capacity of the battery swapping station. The battery size and remaining battery level determine the size of the battery to be replaced and how long it takes for the replaced battery to be fully charged and ready for reuse. The charging characteristics are used to predict how long it takes for the replaced battery to be fully charged. Since the charging power of vehicle lithium-ion batteries is not linear, usually when the remaining battery level is within 10%, a small current is used for charging; when it is between 10% and 80%, a large current is used for charging; and after 80%, the charging current gradually decreases until it is fully charged. This charging method can extend the battery life and cause less damage to the battery. Location information is used to determine how long it takes for the vehicle to reach the battery swapping station and what the remaining battery level of the vehicle is when it arrives at the station. Since the weather temperature affects the battery activity and has a significant impact on the charge and discharge loss of the battery, there is a large difference in the power consumption of electric vehicles when the weather is 25 degrees Celsius and -10 degrees Celsius. Therefore, the weather temperature is an important factor in predicting the per 100-kilometer power consumption of electric vehicles and the remaining battery level when the vehicle arrives at the station. The prediction of the probability of battery swapping at the station by the first battery swapping prediction model can be obtained by manually modeling according to the influence of each factor on the possibility of battery swapping at the station. For example, for an electric vehicle near the battery swapping station with a low remaining battery level and a short distance from the battery swapping station, a higher weight is assigned to predict that the probability of it swapping batteries at the station is very high. For a vehicle with a relatively high remaining battery level, a lower weight is assigned, and for a fuel vehicle, a lower weight is assigned to predict that the probability of it swapping batteries at the station is extremely low or zero. The reason for predicting it as extremely low is that it cannot be excluded that a fuel vehicle may tow an electric vehicle with a dead battery or act as a guide to lead another electric vehicle to the battery swapping station for battery swapping. Further, the parameters of the prediction model can be continuously optimized based on the data accumulated every day. The first battery swapping prediction model can also be obtained by training the information of historical data navigating to the vicinity of the battery swapping station and the data information of battery swapping at the station by inputting them into the DeepSeek large model. The training and operation of the first battery swapping prediction model are completed by the battery swapping station control system.

[0022] The first battery swapping prediction model can call the arrival time prediction module of the map navigation platform to predict the arrival time and driving speed, and then predict the remaining battery level when it arrives at the station based on the vehicle's remaining battery level, power consumption, driving speed, and weather temperature information. The prediction of the charging curve by the first battery swapping prediction model can be obtained by predicting based on the standard charging curve of the vehicle model, weather temperature information, predicted remaining battery level when arriving at the station, and battery capacity size.

[0023] S1-2: When the navigation device type is a mobile phone terminal, obtain the vehicle information data, historical navigation destination data, historical charging orders, historical charging power, terminal location information, and weather information bound to the user's navigation terminal through the map navigation platform; input the vehicle information data, historical navigation destination data, historical charging orders, historical charging power, terminal location information, and weather information into the second battery swapping prediction model, and predict and output the probability of arriving at the station for battery swapping, arrival time, and battery swapping power. Calculate the number of users arriving at the station for battery swapping based on the number of mobile phone terminals with the navigation device type and the predicted probability of arriving at the station for battery swapping.

[0024] Currently, each mobile phone software of the map navigation platform has an account, and users can bind their own vehicles and also use services such as navigation, refueling, charging, accommodation, dining, bus, and taxi on the map navigation platform. By judging information such as the vehicle information, time, historical navigation destination, historical charging order, and charging power bound by the user, the probability of the user driving an electric vehicle for battery swapping can be judged. For example, if the map navigation platform account used by the user binds a BYD pure electric vehicle "Han", the number of times the historical navigation destination is a charging and swapping station is 30, the historical charging order is 35, the historical average charging amount is 50 degrees, and the distance from the charging and swapping station is 2Km, then a relatively high probability of arriving at the station for battery swapping can be given. The second battery swapping prediction model predicts the probability of arriving at the station for battery swapping based on the above factors. The second battery swapping prediction model calls the map navigation platform to predict the arrival time and predicts the battery swapping power based on the historical charging order and charging power.

[0025] The prediction of the battery swapping power based on the historical charging order and charging power is as follows: Obtain the charging time, charging power, and charging station location information from the historical charging order. Combine the charging powers within the same charging station within the same charging time period into one charging order, and combine the charging orders of charging stations within a certain distance within the same charging time period into one charging order. Take the average value of the combined charging orders to obtain the predicted battery swapping power. The same charging time period refers to 1.5 hours, and the same distance can be 2km. The reason is that generally, fast charging is used, and the charging time of electric vehicles is 1 - 1.5 hours. Due to various reasons, the charging pile may jump the gun during charging, and the order will automatically end. The user needs to reinsert the gun to charge and generate a new order. In fact, it is still one charging behavior. Only by combining the two orders can the power of one charging be truly reflected, which is used to roughly predict the battery size. The reason for "combining the charging orders of charging stations within a certain distance within the same charging time period into one charging order" is that sometimes the charging speed of the user at a charging station is slow, or the charging time period standards of each charging station are different, or the charging gun of this charging station is not well matched and prone to jumping the gun. In this case, the user will change to another charging station to charge.

[0026] The construction of the second battery swap prediction model is the same as the construction of the first battery swap prediction model, and is constructed through artificial mathematical modeling, and can also be obtained by inputting relevant information into the deepseek large model for training.

[0027] Similarly, the user data obtained by the battery swap station control system through the map navigation platform is obtained with the user's authorization and consent.

[0028] S1-3: When the navigation device type is a professional navigation device, obtain its historical navigation destination data through the map navigation platform, identify the number of times the destination is a charging and swapping station and the arrival time and departure time, and obtain its location information through the map navigation platform; input the historical navigation destination data, arrival time, departure time, and professional navigation device location information into the third battery swap prediction model to predict and output the probability of battery swapping at the station, arrival time, and battery swapping amount. Calculate the number of users who arrive at the station to swap batteries based on the number of professional navigation devices and the predicted probability of battery swapping at the station.

[0029] Professional navigation equipment is generally purchased and installed by users from shopping websites or from auto electronics or auto accessories stores. It is a separate system that is not bound to the vehicle. However, the probability that the vehicle is an electric vehicle can be judged by the number of historical navigations to the destination of a charging station, as well as the length of stay at the charging station, and the charging power can be roughly judged.

[0030] The third battery swap prediction model predicts the probability of battery swapping at the station based on the number of times and the stay time at the charging and swapping station as the historical navigation destination, calls the map navigation platform to predict the arrival time, and predicts the battery swapping amount based on the stay time at the battery swapping station, the time interval between two historical visits to the battery swapping station, and the driving distance.

[0031] The construction of the third battery swap prediction model is the same as the construction of the first battery swap prediction model. It can be constructed through artificial mathematical modeling, or it can be obtained by inputting relevant information into the deepseek large model for training.

[0032] S2: The battery swap station control system obtains the historical battery swap user data of the battery swap station, identifies the non-map navigation to the station user data and inputs it into the fourth battery swap prediction model, and predicts the output of the number of battery swap users at the station, the battery swap amount, and the battery swap time at the station. The construction of the fourth battery swap prediction model is obtained by inputting relevant information into the deepseek large model for training. The historical battery swap user data includes the user's battery swap time, whether the user uses navigation to the station, the battery swap vehicle model, and the battery swap amount.

[0033] In the prior art, one method is to completely predict the battery replacement current based on historical time nodes, and the other is to predict the battery replacement current when arriving at the station based on the number of electric vehicles near the battery replacement station and the battery power. In this way, the accuracy of predicting the probability of arriving at the station for battery replacement is relatively low, and the reference significance is not great.

[0034] In the present invention, the factors that can be more accurately predicted when the user navigates to near the battery replacement station are stripped from the big data for separate prediction. For the user data that the battery replacement station has not used map navigation to arrive at the station in history but has arrived at the station for battery replacement, etc., which cannot be accurately predicted temporarily, fuzzy prediction is carried out, and then the prediction model fusion processing is carried out to overall improve the prediction accuracy.

[0035] S3: The battery replacement station control system inputs the prediction output data of the first, second, third, and fourth battery replacement prediction models, as well as the number of users arriving at the station for battery replacement, into the arrival battery replacement fusion prediction model, and outputs the time series of the battery replacement power and the number of battery replacement users when arriving at the station.

[0036] S4: The battery replacement station control system guides the user to replace the battery nearby or at off-peak times according to the time series of the battery replacement power and the number of battery replacement users when arriving at the station output by the arrival battery replacement fusion prediction model and the battery replacement capacity of the battery replacement station.

[0037] The battery replacement capacity is predicted based on the number of fully charged batteries and the maximum charging power of the battery replacement station; when the battery replacement power predicted by the arrival battery replacement fusion prediction model for a certain time period exceeds the battery replacement capacity of the battery replacement station, according to the exceeded battery replacement power, select the corresponding number of users to push reminder information, and the reminder information includes a reminder of tight battery replacement, as well as information such as the predicted off-peak time for battery replacement, nearby shopping malls, parks, etc. This is convenient for users to plan their time arrangements and avoid long queuing waiting times.

[0038] The pushed reminder information is pushed through the map navigation platform, and the pushed objects are users who use the map navigation platform. For users who do not arrive at the station through navigation, there is no need to push because they do not master the push platform.

[0039] It should be noted that the above embodiments illustrate the present invention rather than limit the present invention, and those skilled in the art can design alternative embodiments without departing from the scope of the appended claims. Any person skilled in the art in the technical field disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. The protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for battery swapping at a battery swapping station with an intelligent battery swapping guidance function, the steps of which are as follows: S1: The battery swapping station control system obtains in real time user data and user navigation device types that are navigated to near the battery swapping station by multiple map navigation platforms; the navigation device types include in-vehicle terminals, mobile phone terminals, and professional navigation devices; S1-1: When the navigation device type is an in-vehicle terminal, obtain the vehicle version model through the map navigation platform, search for vehicle information in the vehicle information database according to the vehicle version model; determine whether it is an electric vehicle, and when it is an electric vehicle, obtain the remaining battery power, location information, and weather information of the vehicle through the map navigation platform; input the vehicle information, remaining battery power, location information, and weather information data into the first battery swapping prediction model, and predict and output the probability of arriving at the station for battery swapping, arrival time, remaining battery power upon arrival, and charging curve; S1-2: When the navigation device type is a mobile phone terminal, obtain the vehicle information data, historical navigation destination data, historical charging orders, historical charging power, terminal location information, and weather information bound to the user navigation terminal through the map navigation platform; input the vehicle information data, historical navigation destination data, historical charging orders, historical charging power, terminal location information, and weather information into the second battery swapping prediction model, and predict and output the probability of arriving at the station for battery swapping, arrival time, and battery swapping power; calculate the number of users arriving at the station for battery swapping according to the number of mobile phone terminal navigation device types and the predicted probability of arriving at the station for battery swapping; S1-3: When the navigation device type is a professional navigation device, obtain its historical navigation destination data through the map navigation platform, identify the number of times the destination is a charging and battery swapping station and the arrival time and departure time, and obtain its location information through the map navigation platform; input the historical navigation destination data, arrival time at the destination, departure time from the destination, and professional navigation device location information into the third battery swapping prediction model to predict and output the probability of arriving at the station for battery swapping, arrival time, and battery swapping power; S2: The battery swapping station control system obtains the historical battery swapping user data of the battery swapping station, identifies the non-map navigation arrival user data and inputs it into the fourth battery swapping prediction model, and predicts and outputs the number of users arriving at the station for battery swapping, battery swapping power, and arrival time for battery swapping; S3: The battery swapping station control system inputs the prediction output data of the first, second, third, and fourth battery swapping prediction models, as well as the number of users arriving at the station for battery swapping, into the arrival battery swapping fusion prediction model, and predicts and outputs the time series of the arrival battery swapping power and the number of battery swapping users; S4: The battery swapping station control system guides users to swap batteries nearby or at staggered times according to the time series of the arrival battery swapping power, the number of battery swapping users, and the battery swapping capacity of the battery swapping station output by the arrival battery swapping fusion prediction model.

2. The battery swapping method for a battery swapping station with an intelligent battery swapping guiding function according to claim 1, wherein the step S1-1 further includes: Calculate the number of users arriving at the station for battery swapping according to the number of in-vehicle terminal navigation device types and the predicted probability of arriving at the station for battery swapping.

3. The battery swapping method for a battery swapping station with an intelligent battery swapping guiding function according to claim 2, wherein the step S1-1 further includes: The prediction of the probability of arriving at the station for battery swapping by the first battery swapping prediction model is obtained by mathematical modeling based on the influence of various factors on the possibility of arriving at the station for battery swapping in an artificial manner.

4. The battery swapping method for a battery swapping station with an intelligent battery swapping guiding function according to claim 3, wherein the step S1-1 further includes: The first battery swapping prediction model predicts the arrival time by calling the arrival time prediction module of the map navigation platform to predict the arrival time and driving speed, and then predicts the remaining battery power when arriving at the station based on the remaining battery power, power consumption, driving speed, and weather temperature information of the vehicle.

5. The battery swapping method for a battery swapping station with an intelligent battery swapping guiding function according to claim 4, wherein the step S1-3 further includes: Calculate the number of users arriving at the station for battery swapping according to the number of professional navigation devices among the navigation device types and the predicted probability of arriving at the station for battery swapping.

6. The battery swapping method for a battery swapping station with an intelligent battery swapping guidance function according to claim 5, wherein the step S4 further includes that the battery swapping capacity is predicted based on the number of fully charged batteries and the maximum charging power of the battery swapping station.

7. The battery swapping method for a battery swapping station with an intelligent battery swapping guidance function according to claim 6, wherein the step S4 further includes that when the battery swapping power arriving at the station predicted by the arrival-at-station battery swapping fusion prediction model exceeds the battery swapping capacity of the battery swapping station, corresponding numbers of users are selected to push reminder information according to the exceeded battery swapping power.

8. The battery swapping method for a battery swapping station with an intelligent battery swapping guiding function according to claim 7, wherein the step S4 further includes: The reminder information includes a battery swapping tension reminder, as well as predicted battery swapping low valley times, nearby shopping mall, and park information.

9. The battery swapping method for a battery swapping station with an intelligent battery swapping guidance function according to claim 8, wherein the step S1-2 further includes that the second battery swapping prediction model calls the map navigation platform to predict the arrival time and predicts the battery swapping power according to historical charging orders and charging power.

10. The predicting the battery swapping power according to historical charging orders and charging power in the battery swapping method for a battery swapping station with an intelligent battery swapping guidance function according to claim 9 is as follows: obtain the charging time, charging power, and charging station location information from historical charging orders; merge the charging powers within the same charging time period and at the same charging station into one charging order; merge the charging orders of charging stations within a certain distance range within the same charging time period into one charging order; and calculate the average value of the merged charging orders to obtain the predicted battery swapping power.

Images