Air conditioning temperature control method, server, vehicle terminal and system

Abstract

This invention provides an air conditioning temperature control method, server, vehicle terminal, and system. The air conditioning temperature control method includes: receiving user information, vehicle operating parameters, and vehicle environmental parameters sent by the vehicle terminal; inputting the user information, vehicle operating parameters, and vehicle environmental parameters into an air conditioning temperature prediction model to determine a predicted air conditioning temperature. The air conditioning temperature prediction model is used to predict the probability of the user setting each air conditioning temperature level, obtaining the predicted air conditioning temperature, and sending the predicted air conditioning temperature to the vehicle terminal for the vehicle terminal to control the air conditioning temperature. The air conditioning temperature control method of this invention, through a trained air conditioning temperature prediction model, can predict the optimal air conditioning temperature that meets the user's settings based on user information, vehicle operating parameters, and vehicle environmental parameters, thereby controlling the vehicle's air conditioning to the predicted temperature and avoiding the cumbersome process of users adjusting the air conditioning temperature according to their own needs.

Description

Technical Field

[0001] This invention relates to the field of vehicle air conditioning technology, and in particular to an air conditioning temperature control method, server, vehicle terminal and system. Background Technology

[0002] To enhance driving or passenger comfort, it's essential to set a suitable temperature when using vehicle air conditioning to provide a comfortable environment. Therefore, setting the air conditioning temperature to suit the user's comfort level is a key focus in the field of vehicle air conditioning.

[0003] Currently, the main method for setting vehicle air conditioning temperature is for users to set it according to the current environment and their own needs. However, using this method, users have to adjust the air conditioning temperature to their own comfortable temperature every time they get in the car, and they need to adjust it repeatedly to find the temperature that suits their needs, making the process of setting the air conditioning temperature rather cumbersome. Summary of the Invention

[0004] In view of this, the present invention aims to provide an air conditioning temperature control method, server, vehicle terminal and system to solve the problem that users need to manually set the air conditioning temperature to their needs every time they take a ride, which is cumbersome.

[0005] To achieve the above objectives, the technical solution of the present invention is implemented as follows:

[0006] In a first aspect, embodiments of this application provide an air conditioning temperature control method, applied to a server, comprising:

[0007] Receive user information, vehicle operating parameters, and vehicle environmental parameters sent by the vehicle terminal;

[0008] The user information, vehicle operating parameters, and vehicle environmental parameters are input into the air conditioning temperature prediction model to determine the predicted air conditioning temperature; wherein, the air conditioning temperature prediction model is used to predict the probability of the user setting each air conditioning temperature level, and predict the predicted air conditioning temperature.

[0009] The predicted air conditioning temperature is sent to the vehicle terminal so that the vehicle terminal can control the air conditioning temperature.

[0010] Furthermore, the step of inputting the user information, vehicle operating parameters, and vehicle environmental parameters into the air conditioning temperature prediction model to determine the predicted air conditioning temperature includes:

[0011] Based on the user information, vehicle operating parameters, and vehicle environmental parameters, match historical air conditioning temperatures, historical vehicle operating parameters, and historical vehicle environmental parameters.

[0012] Obtain derived features by performing feature engineering on the historical air conditioning temperature, historical vehicle operating parameters, and historical vehicle environmental parameters;

[0013] Based on the user information and the derived features, predict the probability of the user setting each air conditioner temperature level;

[0014] The air conditioner temperature corresponding to the highest probability among the predicted probabilities is determined as the predicted air conditioner temperature.

[0015] Furthermore, the training process of the air conditioning temperature prediction model includes:

[0016] Receive historical air conditioning temperatures, historical vehicle operating parameters, and historical vehicle environmental parameters sent by each vehicle terminal;

[0017] The historical air conditioning temperatures are divided into time segments, and the historical air conditioning temperatures with a set duration greater than or equal to a preset duration are marked to obtain positive sample data of the historical air conditioning temperatures.

[0018] A predetermined number of data points are extracted from the positive sample data to form a validation set, and the remaining positive sample data is used as a training set.

[0019] Feature engineering is performed on the historical air conditioning temperature, historical vehicle operating parameters, and historical vehicle environmental parameters to obtain the derived features of the feature engineering.

[0020] The derived features are concatenated into the training set and the validation set to obtain the processed training set and validation set.

[0021] The processed training set is fitted, and the AUC value of the processed validation set is obtained. When the AUC value is greater than or equal to a preset parameter, the training of the air conditioning temperature prediction model is completed. The AUC value is a probability index for judging the training effect of the air conditioning temperature prediction model.

[0022] Furthermore, feature engineering is performed on the historical air conditioning temperature, historical vehicle operating parameters, and historical vehicle environmental parameters to obtain derived features from the feature engineering, including:

[0023] Characteristic engineering is performed on the frequency of group air conditioning temperature settings based on the historical air conditioning temperatures to determine the group air conditioning temperature setting characteristics;

[0024] The frequency of user air conditioning temperature settings based on the historical air conditioning temperatures is subjected to feature engineering to obtain user air conditioning temperature setting characteristics;

[0025] Based on the group air conditioning temperature setting characteristics and the user air conditioning temperature setting characteristics, derived characteristics are obtained.

[0026] Furthermore, the step of waiting for the AUC value to be greater than or equal to a preset parameter before the air conditioning temperature prediction model training is completed includes:

[0027] When the AUC value is less than the preset parameter, the air conditioning temperature prediction model is iteratively trained based on the current AUC value and the feature importance of the feature engineering.

[0028] Secondly, embodiments of this application provide an air conditioning temperature control method, applied to a vehicle terminal, comprising:

[0029] Obtain user information, vehicle operating parameters, and vehicle environmental parameters;

[0030] Send the user information, vehicle operating parameters, and vehicle environmental parameters to the server;

[0031] The system receives the predicted air conditioning temperature returned by the server. The server stores historical air conditioning temperature data corresponding to the user information, as well as an air conditioning temperature prediction model. The air conditioning temperature prediction model is used to predict the probability of the user setting each air conditioning temperature level, and to predict the predicted air conditioning temperature.

[0032] The air conditioning temperature is controlled based on the predicted air conditioning temperature.

[0033] Furthermore, before obtaining user information and vehicle environmental parameters, the process includes:

[0034] The system acquires the user information, historical air conditioning temperature data, historical vehicle operating parameters, and historical vehicle environmental parameters.

[0035] The system sends the user information, the historical air conditioning temperature data, the historical vehicle operating parameters, and the historical vehicle environmental parameters to the server.

[0036] Compared with existing technologies, the air conditioning temperature control method of the present invention has the following advantages:

[0037] In this embodiment of the invention, the server receives user information, vehicle operating parameters, and vehicle environmental parameters obtained and sent by the vehicle terminal. This information is then input into a trained air conditioning temperature prediction model. Based on the model, a predicted air conditioning temperature is determined and sent to the vehicle terminal for control. The trained model predicts the air conditioning temperature to meet the user's settings, eliminating the cumbersome process of manually adjusting the temperature. The vehicle terminal receives the same information and sends it to the server. It then receives the predicted temperature and controls the air conditioning accordingly, directly adjusting it to the user's preferred temperature, thus achieving accurate temperature control.

[0038] This invention uses information interaction between a server and a vehicle terminal to obtain a suitable air conditioning temperature based on user information and vehicle environmental parameters from a trained air conditioning temperature prediction model. This allows the vehicle terminal to directly provide the user with the desired air conditioning temperature, avoiding the cumbersome adjustment process for the user.

[0039] Thirdly, embodiments of the present invention also provide a server, comprising:

[0040] The information receiving module is used to receive user information and vehicle environmental parameters sent by the vehicle terminal;

[0041] The information receiving module is used to receive user information, vehicle operating parameters, and vehicle environmental parameters sent by the vehicle terminal.

[0042] The predicted temperature module is used to input the user information, vehicle operating parameters, and vehicle environmental parameters into the air conditioning temperature prediction model to determine the predicted air conditioning temperature; wherein, the air conditioning temperature prediction model is used to predict the probability of the user setting each air conditioning temperature level, and predict the predicted air conditioning temperature.

[0043] The information sending module is used to send the predicted air conditioning temperature to the vehicle terminal so that the vehicle terminal can control the air conditioning temperature.

[0044] The advantages of the server and the air conditioning temperature control method mentioned above compared to the prior art are the same, and will not be repeated here.

[0045] Fourthly, embodiments of the present invention also provide a vehicle terminal, comprising:

[0046] The parameter acquisition module is used to acquire user information, vehicle operating parameters, and vehicle environmental parameters.

[0047] The information sending module is used to send the user information, vehicle operating parameters, and vehicle environmental parameters to the server;

[0048] The information receiving module is used to receive the predicted air conditioning temperature returned by the server. The server stores historical air conditioning temperature data corresponding to the user information, as well as an air conditioning temperature prediction model. The air conditioning temperature prediction model is used to predict the probability of the user setting each air conditioning temperature level, and to predict the predicted air conditioning temperature.

[0049] An air conditioning control module is used to control the air conditioning temperature based on the predicted air conditioning temperature.

[0050] The advantages of the vehicle terminal and the aforementioned air conditioning temperature control method compared to the prior art are the same, and will not be repeated here.

[0051] Fifthly, embodiments of the present invention also provide an air conditioning temperature control system, comprising:

[0052] The server is used to receive user information, vehicle operating parameters, and vehicle environmental parameters sent by the vehicle terminal, input the user information, vehicle operating parameters, and vehicle environmental parameters into the air conditioning temperature prediction model, and determine the predicted air conditioning temperature; wherein, the air conditioning temperature prediction model is used to predict the probability of the user setting each air conditioning temperature level, predict the predicted air conditioning temperature, and send the predicted air conditioning temperature to the vehicle terminal for the vehicle terminal to control the air conditioning temperature.

[0053] The vehicle terminal is used to acquire user information, vehicle operating parameters, and vehicle environmental parameters, send the user information, vehicle operating parameters, and vehicle environmental parameters to the server, and receive the predicted air conditioning temperature returned by the server. The server stores historical air conditioning temperature data corresponding to the user information and an air conditioning temperature prediction model. The air conditioning temperature prediction model is used to predict the probability of the user setting each air conditioning temperature level, predict the predicted air conditioning temperature, and control the air conditioning temperature based on the predicted air conditioning temperature.

[0054] The advantages of the system and the air conditioning temperature control method described above compared to the prior art are the same, and will not be repeated here. Attached Figure Description

[0055] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0056] Figure 1 Flowchart of the air conditioning temperature control method provided in the embodiments of the present invention Figure 1 ;

[0057] Figure 2 for Figure 1 The illustrated embodiment provides a flowchart of the steps in the air conditioning temperature control method.

[0058] Figure 3 Flowchart of the air conditioning temperature control method provided in the embodiments of the present invention Figure 2 ;

[0059] Figure 4 This is a schematic diagram of the server structure provided in an embodiment of the present invention;

[0060] Figure 5 This is a schematic diagram of the structure of a vehicle terminal provided in an embodiment of the present invention;

[0061] Figure 6 This is a schematic diagram of the structure of an air conditioning temperature control system provided in an embodiment of the present invention. Detailed Implementation

[0062] It should be noted that, unless otherwise specified, the embodiments and features described in this invention can be combined with each other. To make the above-mentioned objectives, features, and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0063] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0064] See Figure 1 , Figure 1 This is the flow chart of the air conditioning temperature control method provided in the embodiments of this application. Figure 1 In this embodiment, Figure 1 The air conditioning temperature control method shown is applied in Figure 6 The server side shown is as follows: Figure 1 As shown, it includes the following steps:

[0065] Step 101: Receive user information, vehicle operating parameters, and vehicle environmental parameters sent by the vehicle terminal.

[0066] In this embodiment, the vehicle terminal's electronic device unit (ECU) acquires user information, as well as vehicle operating parameters and vehicle environmental parameters. The user information is used to identify different users. User information within the vehicle is comprehensively collected and analyzed. User information may include the user's username, user ID, facial information, etc. Optionally, the vehicle terminal's user information may be the username and user ID used when logging into the system to control the vehicle's air conditioning, or it may be the user's facial information obtained through facial recognition by an in-vehicle camera. Generally, the driver's facial information is used as the default user information.

[0067] Vehicle operating parameters and environmental parameters are detected and acquired by vehicle sensors and ECUs. Vehicle operating parameters include parameters such as vehicle speed and engine output power, while vehicle environmental parameters include environmental information such as temperature at the vehicle's location or area. For example, the location or area of ​​each vehicle is determined by the vehicle location information detected by the GPS receiver, and meteorological information is obtained by the communication unit. That is, the vehicle environmental information includes not only temperature and air temperature, but also weather, humidity, wind speed, etc. The vehicle terminal sends the acquired information to the server through the communication link, and the server receives the user information, vehicle operating parameters, and vehicle environmental parameters sent by the vehicle terminal.

[0068] It should be noted that, in order to meet users' preferences for setting the air conditioning temperature in different environments, the server needs to analyze and process user information, vehicle operating parameters, and vehicle environmental parameters. The user information and vehicle environmental parameters obtained by the server are collected through the vehicle terminal, which places certain performance requirements on the vehicle terminal. For example, in the scenario based on the B07 / B16 model, the vehicle environmental parameters need to be detected.

[0069] In this embodiment of the invention, the vehicle terminal can obtain vehicle environment information and user identification information. The vehicle terminal can obtain information when the user starts the vehicle, or it can automatically start to obtain information at a predetermined time according to the user's settings.

[0070] Of course, the above are just specific examples. In actual use, the vehicle terminal can be any vehicle model, which will not be elaborated here. This implementation method does not limit the vehicle model. In actual use, the vehicle model can be any vehicle model that can obtain user information and vehicle environmental parameters.

[0071] In this embodiment, the server and the vehicle terminal interact with each other to obtain the air conditioning temperature that suits the user's needs based on user information and vehicle environmental parameters. By combining user information and current environmental parameters, the system can accurately analyze the user's needs for setting the air conditioning temperature.

[0072] Step 102: Input user information, vehicle operating parameters, and vehicle environmental parameters into the air conditioning temperature prediction model to determine the predicted air conditioning temperature; wherein, the air conditioning temperature prediction model is used to predict the probability of the user setting each air conditioning temperature level, and predict the predicted air conditioning temperature.

[0073] In this embodiment, the server stores a trained air conditioning temperature prediction model. After receiving user information, vehicle operating parameters, and vehicle environmental parameters, the server performs feature engineering on the data and inputs it into the air conditioning temperature prediction model.

[0074] It should be noted that the air conditioning temperature prediction model trained on the server is trained based on user information obtained and sent by the vehicle terminal, historical vehicle environmental parameters, and historical air conditioning temperatures set by the user under various vehicle environmental conditions. In other words, the server receives a large number of discrete vehicle environmental parameters obtained by the vehicle terminal.

[0075] Specifically, the server receives historical air conditioning temperature and historical vehicle environmental parameter data sent by the vehicle terminal, obtains positive and negative samples of historical air conditioning temperature, extracts a training set for feature engineering, and fits the processed training set using a Light GMB model. The air conditioning temperature prediction model training is complete when the AUC value of the validation set is greater than or equal to a preset parameter. It should be noted that in this embodiment, the AUC value is a probability metric used to judge the training effect of the air conditioning temperature prediction model. The AUC value is calculated as follows: based on the positive and negative samples of the data, the probability that a positive sample ranks ahead of a negative sample is calculated according to a classification algorithm; for example, a classifier can be trained to predict the positive and negative samples of historical air conditioning temperature. The probability that the probability of predicting a positive sample is greater than the probability of predicting a negative sample is the AUC value. In other words, the AUC value can determine the prediction accuracy of the prediction model, thus being used to evaluate the training effect of the prediction model.

[0076] This embodiment predicts the optimal vehicle air conditioning temperature based on user information and the current vehicle environment by inputting user information and vehicle environmental parameters into the air conditioning temperature prediction model.

[0077] In this embodiment, the predicted air conditioning temperature is determined based on the trained air conditioning temperature prediction model and the user information and vehicle environmental parameters input into the model.

[0078] It should be noted that the air conditioner temperature prediction model is a recommendation ranking architecture. By combining user personalized preferences, group preference statistics, and time series features, it learns the user's personalized habits for setting air conditioner temperature parameters. The discrete temperature parameters are matched and recommended to each user in turn, and the model returns the air conditioner temperature that the user is most likely to set under the current environmental conditions.

[0079] Specifically, the air conditioning temperature prediction model is trained through feature engineering and model fitting based on received historical air conditioning temperature data and historical vehicle environmental parameter data. The server receives real-time user information, vehicle operating parameters, and ambient temperature from the vehicle terminal. It then compares the received information with historical information. When the user information, vehicle operating parameters, and vehicle environmental parameters match the historical information stored on the server, the server predicts the probability of the user setting each air conditioning temperature level based on the matched historical vehicle environmental information. Thus, the server determines the recommended air conditioning temperature setting that best suits the user's habits under the current vehicle environmental conditions.

[0080] In this embodiment of the invention, after obtaining user information and historical vehicle environment information corresponding to historical air conditioning temperature data, the vehicle terminal can send the historical air conditioning temperature data, the corresponding historical vehicle environment information, and the user information to the server. The server can perform feature engineering and model prediction analysis on the historical air conditioning temperature data and the corresponding historical vehicle environment information based on the user information, thereby obtaining the predicted air conditioning temperature corresponding to the vehicle user information based on the actual situation of the current environmental information, so that the vehicle terminal can control the vehicle's air conditioning parameters.

[0081] Step 103: Send the predicted air conditioning temperature to the vehicle terminal so that the vehicle terminal can control the air conditioning temperature.

[0082] In this embodiment, after the server determines the predicted air conditioning temperature, it sends it to the vehicle terminal. The vehicle terminal obtains the predicted air conditioning temperature and then controls the vehicle air conditioning to the predicted air conditioning temperature.

[0083] It should be noted that after determining the predicted air conditioning temperature, the server can send the predicted air conditioning temperature to the vehicle terminal based on user information, so that the vehicle terminal can control the vehicle's air conditioning settings. Alternatively, it can send it directly to the receiving end of the vehicle's air conditioning system for direct adjustment of the vehicle's air conditioning settings. The server sends the predicted air conditioning temperature to the corresponding vehicle terminal for the vehicle terminal to control the air conditioning temperature; this embodiment of the invention does not impose specific limitations on this.

[0084] In this embodiment of the invention, the server determines the predicted air conditioning temperature based on user information and current vehicle environmental parameters; the vehicle terminal controls the air conditioning based on the predicted air conditioning temperature, thereby enabling the vehicle air conditioning to determine the air conditioning temperature that best meets the user's settings based on the user's air conditioning settings habits and the influence of current environmental factors, avoiding the tedious operation of the user repeatedly adjusting the air conditioning temperature according to the environment.

[0085] Specifically, in this embodiment, user information, vehicle operating parameters, and vehicle environmental parameters are input into the air conditioning temperature prediction model to determine the predicted air conditioning temperature, including the following steps:

[0086] Based on user information, vehicle operating parameters, and vehicle environmental parameters, historical air conditioning temperatures, historical vehicle operating parameters, and historical vehicle environmental parameters are matched.

[0087] In this embodiment, when the server receives real-time user information, vehicle operating parameters, and vehicle ambient temperature from the vehicle terminal, it matches the user information with historical vehicle operating parameters and historical vehicle ambient parameters. In a specific implementation, the server may store multiple user information entries, along with corresponding historical air conditioning temperatures, historical operating parameters, and historical vehicle ambient parameters for each user. Once the user information, vehicle operating parameters, and vehicle ambient parameters are obtained, the historical air conditioning temperatures, historical vehicle operating parameters, and historical vehicle ambient parameters are matched from the server.

[0088] Derived features are obtained by feature engineering historical air conditioning temperatures, historical vehicle operating parameters, and historical vehicle environmental parameters.

[0089] In this embodiment, after the server matches the user information, vehicle operating parameters, and vehicle ambient temperature sent by the vehicle terminal in real time with the corresponding historical parameters, it obtains the derived features obtained by feature engineering based on the historical air conditioning temperature, historical vehicle operating parameters, and historical vehicle ambient parameters. This allows the server to obtain the derived features of the historical data that match the current parameters, so as to predict the probability of setting the air conditioning temperature.

[0090] Based on user information and derived characteristics, predict the probability of users setting each air conditioner temperature level.

[0091] In this embodiment, based on user information and derived features, the probability of the user setting each air conditioning temperature level is predicted one by one in the air conditioning temperature prediction model, so that the server determines the air conditioning temperature that conforms to the user's habits under the current vehicle environment information as the recommended air conditioning setting parameter.

[0092] The air conditioner temperature corresponding to the highest probability among the predicted probabilities is determined as the predicted air conditioner temperature.

[0093] It should be noted that the model predicts multiple air conditioning temperatures that meet the user's needs and preferences. Based on the frequency statistics calculated by feature engineering, the probability of the user setting the air conditioning temperature is obtained. The predicted air conditioning temperatures are then compared and ranked according to their probabilities. The air conditioning temperature with the highest probability among the predicted probabilities is determined as the final air conditioning temperature recommended to the vehicle terminal, thereby satisfying the user's setting preferences.

[0094] In this embodiment of the invention, the server receives user information and vehicle environmental parameters acquired and sent by the vehicle terminal, inputs them into a trained air conditioning temperature prediction model, determines the predicted air conditioning temperature based on the model, and sends the predicted air conditioning temperature to the vehicle terminal for the vehicle terminal to control the air conditioning temperature. Through the trained air conditioning temperature prediction model, the system can predict the air conditioning temperature that meets the user's settings based on user information and vehicle environmental parameters, thereby controlling the vehicle's air conditioning to the predicted temperature and avoiding the cumbersome process of users adjusting the air conditioning temperature according to their own needs. This invention, through information interaction between the server and the vehicle terminal, obtains a suitable air conditioning temperature based on user information and vehicle environmental parameters from the trained air conditioning temperature prediction model, allowing the vehicle terminal to directly provide the user with the desired air conditioning temperature, avoiding the tedious adjustment process for the user.

[0095] To better understand the technical solution provided in this example, an example will be used for further explanation below. Figure 2 , Figure 2 for Figure 1 The illustrated embodiment provides a flowchart of the steps in the air conditioning temperature control method. The flowchart of the training steps for the air conditioning temperature prediction model includes:

[0096] S201 receives historical air conditioning temperatures, historical vehicle operating parameters, and historical vehicle environmental parameters sent by each vehicle terminal.

[0097] Specifically, the air conditioning temperature and vehicle environmental parameters acquired by each vehicle terminal are stored and sent to the server. A large amount of acquired data is used as historical air conditioning data and historical vehicle environmental parameter data to provide sample parameters for training the model on the server.

[0098] In this embodiment, the historical air conditioning temperature sent by the vehicle terminal is any integer within a closed range of 16°C to 38°C. The historical vehicle environmental parameter data includes vehicle speed, weather, humidity, ambient temperature, etc. It should be noted that the vehicle environmental parameters can be obtained by the vehicle terminal based on data collected by the sensor device.

[0099] In this embodiment of the invention, the vehicle terminal acquires historical air conditioning temperature and historical vehicle environmental parameter data. The historical air conditioning temperature can be a temperature previously set by the user, such as settings made by the user while riding or driving the vehicle via the vehicle terminal, or subsequent adjustments made by the user to the vehicle's air conditioning temperature. The historical vehicle environmental parameter data can be the vehicle's environmental parameters at the time the air conditioning was set to that temperature. For example, if the user sets the air conditioning temperature to 20°C, the ambient temperature is 30°C, the weather is sunny, and the humidity is 50%, then 20°C is stored as the historical air conditioning temperature, and the corresponding historical vehicle environmental parameter data is stored as an ambient temperature of 30°C, a sunny day, and 50% humidity. Additionally, the vehicle environmental parameters may also include user behavioral data, such as user preferences.

[0100] In this embodiment, the server and the vehicle terminal interact with each other to obtain the air conditioning temperature that suits the user's needs based on user information and vehicle environmental parameters. By combining user information and current environmental parameters, the system can accurately analyze the user's desired air conditioning temperature setting.

[0101] S202, the historical air conditioning temperatures are divided according to time, and the historical air conditioning temperatures with a set duration greater than or equal to the preset duration are marked to obtain positive sample data of historical air conditioning temperatures.

[0102] Specifically, the received historical air conditioning temperatures are sliced ​​into time segments and categorized according to preset durations. It should be noted that these historical air conditioning temperatures are based on user settings made to the vehicle's air conditioning via the vehicle's terminal while the user is riding or driving, as well as subsequent adjustments made by the user. To statistically analyze user preferences for air conditioning temperature settings, all valid historical air conditioning temperature data needs to be segmented.

[0103] It should be noted that the validity of the air conditioner temperature setting depends on the duration for which the set temperature is maintained. If the user maintains a certain air conditioner temperature for a long time, then that temperature is the user's preferred setting. If the user sets the air conditioner temperature and maintains it for only a few milliseconds, seconds, or even minutes, then that temperature could be due to the user setting it to be unsuitable for their needs, the temperature that the user experienced during the temperature adjustment process, or the temperature that the user briefly turned on and off. Therefore, historical air conditioner temperature data is sliced ​​over time, with a preset duration of 15 minutes. When the air conditioner temperature setting time exceeds the preset duration, the air conditioner temperature is marked as a positive sample; otherwise, the air conditioner temperature is treated as a negative sample.

[0104] This embodiment classifies the acquired historical data to obtain positive sample data of historical air conditioning temperatures, which is convenient for training the prediction model. Furthermore, the filtered historical data has high accuracy, which improves the prediction accuracy of the prediction model trained on the server side.

[0105] S203, a preset number of data points are extracted from the positive sample data to form the validation set, and the remaining positive sample data is used as the training set.

[0106] Specifically, it's typically necessary to partition the dataset before building the model to prevent data snooping bias and to avoid selecting models that favor test data, which could distort the training results. This involves dividing the positive sample data into a validation set and a training set. The training set contains data samples used for model fitting, primarily for training parameters. The validation set is a separate set of samples reserved during model training to adjust hyperparameters and provide an initial assessment of the model's capabilities. Generally, 80% of the dataset is used as the training set and 20% as the test set. In this example, 1500 data points can be randomly selected from the positive sample data as the validation set, with the remaining data used as the training set.

[0107] In this embodiment, the training set can fit the model and adjust the weights, while the validation set can quickly tune parameters. That is, the validation set can be used to select and adjust parameters (number of network layers, number of network nodes, number of iterations in epochs, learning rate, optimizer, etc.). The validation set can be used to check the training effect of the model after iterative training and to monitor whether there are any abnormalities in the training network, thereby adjusting the model training parameters.

[0108] S204. Characteristic engineering is performed on historical air conditioning temperatures, historical vehicle operating parameters, and historical vehicle environmental parameters to obtain derived features from the characteristic engineering.

[0109] Specifically, feature engineering is the process of transforming raw data into features that better represent the essence of the problem, enabling the application of these features to predictive models and improve the accuracy of model predictions for data that is not readily available. Feature engineering aims to discover features that have a significant impact on the dependent variable y, usually referred to as features in the context of the independent variable x. The goal of feature engineering is to discover important features.

[0110] In this implementation, to obtain users' personalized preferences for air conditioning temperature settings, feature engineering is performed on the sample data of the training set to further train the prediction model. Through feature engineering, the group's air conditioning temperature setting preferences are calculated, including: frequency statistics of each air conditioning temperature setting for each group in each city; frequency statistics of each air conditioning temperature setting for each group in each city each month; frequency statistics of each air conditioning temperature setting for each group each month; frequency statistics of each air conditioning temperature setting for each group under each ambient temperature; frequency statistics of each air conditioning temperature setting for each group under each humidity level; frequency statistics of each air conditioning temperature setting for each group under each weather condition; and frequency statistics of each air conditioning temperature setting for each group under each vehicle speed. It should be noted that setting the air conditioning temperature to any integer within a closed range of 16℃ to 38℃ divides the air conditioning temperature into multiple groups. For example, Group 1: 16℃ to 17℃, Group 2: 18℃ to 19℃, Group 3: 20℃ to 21℃, and so on. This will not be elaborated further here. The feature engineering calculation of the group's air conditioning temperature setting preferences is based on the frequency statistics of each group's air conditioning temperature settings.

[0111] Through feature engineering, the calculation of users' personalized air conditioning temperature setting preferences includes: the frequency statistics of each car owner's air conditioning temperature settings for each group; the frequency statistics of each car owner's air conditioning temperature settings for each group per month; the frequency statistics of each car owner's air conditioning temperature settings for each group per day; the frequency statistics of each car owner's air conditioning temperature settings for each group under each ambient temperature; the frequency statistics of each car owner's air conditioning temperature settings for each group under each humidity; the frequency statistics of each car owner's air conditioning temperature settings for each group under each weather; and the frequency statistics of each car owner's air conditioning temperature settings for each group under each vehicle speed.

[0112] S205, the derived features are concatenated to the training set and the validation set to obtain the processed training set and validation set.

[0113] The derived features obtained from feature engineering are concatenated to the training and validation sets to obtain the processed training and validation sets. For example, the derived feature "User 001 in Beijing in July, with an ambient temperature of 35℃, sunny weather, and 40% humidity, sets the air conditioner temperature to 25 degrees 5 times" is concatenated to the training and validation sets.

[0114] It should be noted that multi-granularity and multi-level features are extracted from historical air conditioning temperatures, historical vehicle operating parameters, and historical vehicle environmental parameters, including monthly features, ambient temperature features, humidity features, weather features, and vehicle speed features. Among the extracted static features, features with string values ​​are hashed and mapped to a vector of a certain dimension to obtain numerical features. Numerical features are hashed and mapped to a vector of a certain dimension to obtain feature value vectors. For both numerical and string-type features, the feature values ​​are concatenated to form feature vectors. These feature vectors are then added to the training and validation sets, thus completing the concatenation of derived features into the training and validation sets, resulting in processed training and validation set data.

[0115] In this embodiment, feature engineering is performed on the frequency of group air conditioning temperature settings in the training set to determine the group air conditioning temperature setting characteristics. Feature engineering is also performed on the frequency of user air conditioning temperature settings in the training set to obtain user air conditioning temperature setting characteristics. Derived features are then obtained based on the group air conditioning temperature setting characteristics and the user air conditioning temperature setting characteristics. By processing a large amount of discrete data through feature engineering, derived features are obtained, which transforms the data into feature data that is easy for the model to process, thereby improving the speed of model training.

[0116] S206, Fit the processed training set and obtain the AUC value of the processed validation set. When the AUC value is greater than or equal to the preset parameter, the training of the air conditioning temperature prediction model is completed. The AUC value is a probability index for judging the training effect of the air conditioning temperature prediction model.

[0117] It should be noted that, to achieve iterative training to obtain the optimal model, this embodiment uses the Light GBM model to fit the feature-engineered training set. The Light GBM model (Light Gradient Boosting Machine) is a framework for implementing the GBDT (Gradient Boosting Decision Tree) algorithm, supporting efficient parallel training and offering advantages such as faster training speed, lower memory consumption, better accuracy, and support for distributed processing of massive amounts of data. The main idea of ​​Light GBM is to use weak classifiers (decision trees) for iterative training to obtain the optimal model, which has advantages such as good training performance and low overfitting risk. Light GBM uses a histogram algorithm to transform traversing samples into traversing histograms, greatly reducing time complexity. During training, a one-sided gradient algorithm is used to filter out samples with small gradients, reducing a significant amount of computation. Optimized feature parallelism and data parallelism methods are employed to accelerate computation. When the data volume is very large, a voting parallelism strategy can also be used. Cache optimization has also been implemented to increase cache hit rate.

[0118] Specifically, Light GBM can use the Histogram algorithm to traverse the training data and count the cumulative statistics of each discrete value in the histogram. When performing feature selection, it only needs to traverse the discrete values ​​of the histogram to find the optimal split point, that is, use the light GBM model to fit the training set and set the optimal parameters.

[0119] Obtain the AUC value of the processed validation set, since the dataset includes a training set, a validation set, and a prediction set; use the training set to train the model, and use the validation set to detect whether the currently trained model has overfitted. If it is overfitted, stop training; otherwise, continue training.

[0120] It's important to note that the AUC value is a probability value used to assess the training performance of an air conditioning temperature prediction model. Based on the positive and negative samples in the data, the probability that a positive sample will precede a negative sample is calculated using a classification algorithm. The higher the AUC value, the more likely the current classification algorithm is to rank positive samples ahead of negative samples, thus achieving better classification. For example, the criteria for judging the quality of a prediction model based on AUC are: AUC = 1, a perfect prediction model; AUC = [0.85, 0.95], very good performance; AUC = [0.7, 0.85], average performance; AUC = [0.5, 0.7], low performance; AUC = 0.5, the model has no predictive value.

[0121] Specifically, the prediction parameters can be set such that the validation set AUC value reaches above 0.90 and the average error is below 1 degree Celsius. When the AUC value does not meet the preset parameters, the air conditioning temperature prediction model is iteratively trained based on the current AUC value and the feature importance of the feature engineering.

[0122] In this embodiment, the AUC value of the validation set parameter reflects the performance of model training. When the model training does not reach the best effect, iterative training is carried out until the accuracy of the model reaches the best and the error is minimized, thus ensuring that the model has better accuracy.

[0123] By training an air conditioning temperature prediction model, this invention can predict the air conditioning temperature that meets the user's set requirements based on user information and vehicle environmental parameters, thereby controlling the vehicle's air conditioning to the predicted temperature and avoiding the tedious process of adjusting the air conditioning temperature according to the user's own needs.

[0124] See Figure 3 , Figure 3 Flowchart of the air conditioning temperature control method provided in the embodiments of the present invention Figure 2 In this embodiment, Figure 3 The air conditioning temperature control method shown is applied in Figure 6 The vehicle terminal shown is as follows: Figure 3 As shown, it includes the following steps:

[0125] Step 301: Obtain user information, vehicle operating parameters, and vehicle environmental parameters.

[0126] In this embodiment, the user information obtained by the vehicle terminal can be the user account stored in the vehicle terminal, or the username, user serial number, etc. used by the user when logging into the system through the vehicle terminal to control the vehicle air conditioning, or the user's facial information obtained by performing facial recognition on the user through the in-vehicle camera. Generally, the driver's facial information can be obtained by default as the user information.

[0127] Specifically, the vehicle terminal acquires vehicle environmental parameters. This acquisition can occur when the user starts the vehicle, or it can automatically acquire information at a predetermined time based on user settings. Vehicle environmental parameters can include current interior and exterior environmental information. Interior environmental information includes current ambient temperature, humidity, and vehicle speed. Furthermore, it can also include information about the users currently in the vehicle, such as the number of drivers and passengers. Exterior environmental information can include current ambient temperature, humidity, weather, wind speed, and time. Users can configure the types of environmental information acquired based on their needs.

[0128] It should be noted that the vehicle terminal acquires environmental information through different types of sensors installed in the vehicle, such as temperature sensors, humidity sensors, and facial recognition sensors. Optionally, it can also obtain weather information or other information released through specific channels via the vehicle network or a networked backend to obtain current environmental information. In this embodiment of the invention, no specific limitation is made on the method of obtaining current environmental information.

[0129] The vehicle terminal, based on user information and vehicle environmental parameters, avoids the memory cost of a large number of settings for users, as well as the time wasted adjusting different settings according to the current environment, so that the server can accurately predict the target air conditioning temperature based on the current parameters.

[0130] Step 302: Send user information, vehicle operating parameters, and vehicle environmental parameters to the server.

[0131] Specifically, after obtaining user information and vehicle environmental parameters, the vehicle terminal sends them to the server. The server can then use the user information to determine the recommended air conditioning temperature based on the actual environmental conditions, allowing the vehicle terminal to control the vehicle's air conditioning temperature.

[0132] In this embodiment of the invention, the vehicle terminal sends the acquired user information and vehicle environmental parameters to the server. The server determines the corresponding historical air conditioning temperature setting statistics based on the current user information and vehicle environmental parameters, determines the corresponding recommended air conditioning temperature based on the current environmental information, and returns it to the vehicle terminal, thereby ensuring that the vehicle terminal can obtain the air conditioning setting parameters preferred by the user in the current environment.

[0133] Step 303: Receive the predicted air conditioning temperature returned by the server. The server stores historical air conditioning temperature data corresponding to user information, as well as an air conditioning temperature prediction model. The air conditioning temperature prediction model is used to predict the probability of the user setting each air conditioning temperature level, and predict the predicted air conditioning temperature.

[0134] Specifically, the server stores historical air conditioning temperature data corresponding to user information, as well as an air conditioning temperature prediction model, which predicts how different users or vehicles have previously preferred to set the vehicle's air conditioning. The predicted air conditioning temperature is then sent to the vehicle terminal, which receives the predicted air conditioning temperature.

[0135] It should be noted that after determining the predicted air conditioning temperature, the server can send the predicted air conditioning temperature to the vehicle terminal based on user information, so that the vehicle terminal can control the vehicle's air conditioning settings. Alternatively, it can send it directly to the vehicle's air conditioning receiver to directly adjust the vehicle's air conditioning settings. The vehicle terminal receives the predicted air conditioning temperature returned by the server; this embodiment of the invention does not impose specific limitations on this.

[0136] In this embodiment of the invention, the server determines the predicted air conditioning temperature based on user information and current vehicle environmental parameters; the vehicle terminal controls the air conditioning based on the predicted air conditioning temperature, thereby enabling the vehicle air conditioning to determine the air conditioning temperature that best meets the user's settings based on the user's air conditioning settings habits and the influence of current environmental factors, avoiding the tedious operation of the user repeatedly adjusting the air conditioning temperature according to the environment.

[0137] Step 304: Control the air conditioning temperature based on the predicted air conditioning temperature.

[0138] In this embodiment, before obtaining user information, vehicle operating parameters, and vehicle environmental parameters, the following steps are also included:

[0139] It acquires user information, historical air conditioning temperature data, historical vehicle operating parameters, and historical vehicle environmental parameters, and sends these data to the server.

[0140] In this invention, the vehicle terminal acquires user information and vehicle environmental parameters, sends these parameters to a server, receives a predicted air conditioning temperature from the server, and controls the air conditioning temperature accordingly. This means the air conditioning temperature is directly controlled to meet the user's needs and preferences, achieving accurate temperature control. Through a trained air conditioning temperature prediction model, the system can predict the desired temperature based on user information and vehicle environmental parameters, thereby controlling the vehicle's air conditioning to the predicted temperature and eliminating the cumbersome process of adjusting the temperature manually. This invention, through information interaction between the server and the vehicle terminal, obtains a suitable air conditioning temperature from the trained model based on user information and vehicle environmental parameters, allowing the vehicle terminal to directly provide the user with the appropriate temperature, avoiding the tedious adjustment process.

[0141] See Figure 4 , Figure 4 This is a schematic diagram of the server structure provided in an embodiment of the present invention, such as... Figure 4 As shown, the server in this embodiment of the invention includes:

[0142] The information receiving module 401 is used to receive user information, vehicle operating parameters and vehicle environmental parameters sent by the vehicle terminal.

[0143] The predicted temperature module 402 is used to input user information, vehicle operating parameters and vehicle environmental parameters into the air conditioning temperature prediction model to determine the predicted air conditioning temperature; wherein, the air conditioning temperature prediction model is used to predict the probability of the user setting each air conditioning temperature level, and predict the predicted air conditioning temperature.

[0144] The information sending module 403 is used to send the predicted air conditioning temperature to the vehicle terminal so that the vehicle terminal can control the air conditioning temperature.

[0145] Furthermore, the temperature prediction module 402 includes:

[0146] The parameter processing submodule is used to match historical air conditioning temperatures, historical vehicle operating parameters, and historical vehicle environmental parameters based on user information, vehicle operating parameters, and vehicle environmental parameters.

[0147] The information acquisition submodule is used to acquire derived features obtained by feature engineering of historical air conditioning temperatures, historical vehicle operating parameters, and historical vehicle environmental parameters.

[0148] The temperature prediction sub-model is used to predict the probability of a user setting each air conditioner temperature level based on user information and derived features.

[0149] The temperature determination submodule is used to determine the air conditioner temperature corresponding to the highest probability among the predicted probabilities as the predicted air conditioner temperature.

[0150] See Figure 5 , Figure 5 This is a schematic diagram of the structure of a vehicle terminal provided in an embodiment of the present invention, such as... Figure 4 As shown, the vehicle terminal in this embodiment of the invention includes:

[0151] The parameter acquisition module 501 is used to acquire user information, vehicle operating parameters, and vehicle environmental parameters.

[0152] The information sending module 502 is used to send user information, vehicle operating parameters, and vehicle environmental parameters to the server;

[0153] The receiving information module 503 is used to receive the predicted air conditioning temperature returned by the server. The server stores historical air conditioning temperature data corresponding to user information, as well as an air conditioning temperature prediction model. The air conditioning temperature prediction model is used to predict the probability of the user setting each air conditioning temperature level, and to predict the predicted air conditioning temperature.

[0154] The air conditioning control module 504 is used to control the air conditioning temperature based on the predicted air conditioning temperature.

[0155] This application provides a remote charging system, see [link to relevant documentation]. Figure 6 As shown, it includes: a vehicle terminal and a server, including:

[0156] Server 601 is used to receive user information, vehicle operating parameters and vehicle environmental parameters sent by the vehicle terminal, input the user information, vehicle operating parameters and vehicle environmental parameters into the air conditioning temperature prediction model, and determine the predicted air conditioning temperature; wherein, the air conditioning temperature prediction model is used to predict the probability of the user setting each air conditioning temperature level, predict the predicted air conditioning temperature, and send the predicted air conditioning temperature to the vehicle terminal for the vehicle terminal to control the air conditioning temperature.

[0157] The vehicle terminal 602 is used to acquire user information, vehicle operating parameters, and vehicle environmental parameters, send user information, vehicle operating parameters, and vehicle environmental parameters to the server, and receive the predicted air conditioning temperature returned by the server. The server stores historical air conditioning temperature data corresponding to user information and an air conditioning temperature prediction model. The air conditioning temperature prediction model is used to predict the probability of the user setting each air conditioning temperature level, predict the predicted air conditioning temperature, and control the air conditioning temperature based on the predicted air conditioning temperature.

[0158] The specific implementation method of the air conditioning temperature control system provided in this embodiment can be combined with the content of the air conditioning temperature control method provided in the above embodiments, and will not be repeated here.

[0159] Those skilled in the art will understand that the above embodiments are specific examples of implementing the present invention, and in practical applications, various changes in form and detail may be made without departing from the spirit and scope of the present invention.

[0160] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0161] Although preferred embodiments of the present invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present invention.

[0162] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes the aforementioned element.

[0163] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An air conditioning temperature control method, characterized in that, Applied to servers, including: Receive user information, vehicle operating parameters, and vehicle environmental parameters sent by the vehicle terminal; The user information, vehicle operating parameters, and vehicle environmental parameters are input into the air conditioning temperature prediction model to determine the predicted air conditioning temperature. The air conditioning temperature prediction model is used to predict the probability of a user setting each air conditioning temperature level, and the training process of the air conditioning temperature prediction model includes: receiving historical air conditioning temperatures, historical vehicle operating parameters, and historical vehicle environmental parameters sent from each vehicle terminal; dividing the historical air conditioning temperatures by time; marking historical air conditioning temperatures with a setting duration greater than or equal to a preset duration to obtain positive sample data of the historical air conditioning temperatures; and... In this dataset, a preset number of data points are extracted as the validation set, and the remaining positive sample data is used as the training set. Feature engineering is performed on the historical air conditioning temperatures, historical vehicle operating parameters, and historical vehicle environmental parameters to obtain derived features. These derived features are then concatenated to the training set and the validation set to obtain processed training and validation sets. The processed training set is fitted, and the AUC value of the processed validation set is obtained. When the AUC value is greater than or equal to a preset parameter, the air conditioning temperature prediction model training is complete. The AUC value is an indicator used to judge the training effect of the air conditioning temperature prediction model. The predicted air conditioning temperature is sent to the vehicle terminal so that the vehicle terminal can control the air conditioning temperature.

2. The method according to claim 1, characterized in that, The step of inputting the user information, vehicle operating parameters, and vehicle environmental parameters into the air conditioning temperature prediction model to determine the predicted air conditioning temperature includes: Based on the user information, vehicle operating parameters, and vehicle environmental parameters, match historical air conditioning temperatures, historical vehicle operating parameters, and historical vehicle environmental parameters. Obtain derived features by performing feature engineering on the historical air conditioning temperature, historical vehicle operating parameters, and historical vehicle environmental parameters; Based on the user information and the derived features, predict the probability of the user setting each air conditioner temperature level; The air conditioner temperature corresponding to the highest probability among the predicted probabilities is determined as the predicted air conditioner temperature.

3. The method according to claim 1, characterized in that, The process of performing feature engineering on the historical air conditioning temperature, historical vehicle operating parameters, and historical vehicle environmental parameters to obtain derived features includes: Characteristic engineering is performed on the frequency of group air conditioning temperature settings for the historical air conditioning temperatures to determine the group air conditioning temperature setting characteristics. The frequency of user air conditioning temperature settings based on the historical air conditioning temperatures is characterized to obtain user air conditioning temperature setting characteristics. Based on the group air conditioning temperature setting characteristics and the user air conditioning temperature setting characteristics, derived characteristics are obtained.

4. The method according to claim 1, characterized in that, Before the air conditioning temperature prediction model training is completed when the AUC value is greater than or equal to a preset parameter, the following steps are included: When the AUC value is less than the preset parameter, the air conditioning temperature prediction model is iteratively trained based on the AUC value and the feature importance of the feature engineering.

5. An air conditioning temperature control method, characterized in that, Applied to vehicle terminals, including: Obtain user information, vehicle operating parameters, and vehicle environmental parameters; Send the user information, vehicle operating parameters, and vehicle environmental parameters to the server; The system receives predicted air conditioning temperatures returned by the server, which stores historical air conditioning temperature data corresponding to the user information, as well as an air conditioning temperature prediction model. The air conditioning temperature prediction model is used to predict the probability of a user setting each air conditioning temperature level, and to predict the predicted air conditioning temperature. The training process of the air conditioning temperature prediction model includes: receiving historical air conditioning temperatures, historical vehicle operating parameters, and historical vehicle environmental parameters sent by each vehicle terminal; segmenting the historical air conditioning temperatures by time; marking historical air conditioning temperatures with a setting duration greater than or equal to a preset duration to obtain positive sample data of the historical air conditioning temperatures. A preset number of data points are extracted from the positive sample data to form a validation set, and the remaining positive sample data forms a training set. Feature engineering is performed on the historical air conditioning temperature, historical vehicle operating parameters, and historical vehicle environmental parameters to obtain derived features. The derived features are then concatenated to the training set and the validation set to obtain processed training and validation sets. The processed training set is fitted, and the AUC value of the processed validation set is obtained. When the AUC value is greater than or equal to a preset parameter, the air conditioning temperature prediction model training is complete. The AUC value is an indicator used to judge the training effect of the air conditioning temperature prediction model. The air conditioning temperature is controlled based on the predicted air conditioning temperature.

6. The method according to claim 5, characterized in that, Before obtaining user information and vehicle environmental parameters, the following steps are included: The system acquires the user information, historical air conditioning temperature data, historical vehicle operating parameters, and historical vehicle environmental parameters. The system sends the user information, the historical air conditioning temperature data, the historical vehicle operating parameters, and the historical vehicle environmental parameters to the server.

7. A server, characterized in that, include: The information receiving module is used to receive user information, vehicle operating parameters, and vehicle environmental parameters sent by the vehicle terminal. The temperature prediction module is used to input the user information, vehicle operating parameters, and vehicle environmental parameters into the air conditioning temperature prediction model to determine the predicted air conditioning temperature. The air conditioning temperature prediction model is used to predict the probability of the user setting each air conditioning temperature level, and to predict the predicted air conditioning temperature. The training process of the air conditioning temperature prediction model includes: receiving historical air conditioning temperatures, historical vehicle operating parameters, and historical vehicle environmental parameters sent from each vehicle terminal; dividing the historical air conditioning temperatures by time; marking historical air conditioning temperatures with a setting duration greater than or equal to a preset duration to obtain positive sample data of the historical air conditioning temperatures. A preset number of data points are extracted from the positive sample data to form a validation set, and the remaining positive sample data forms a training set. Feature engineering is performed on the historical air conditioning temperature, historical vehicle operating parameters, and historical vehicle environmental parameters to obtain derived features. The derived features are then concatenated to the training set and the validation set to obtain processed training and validation sets. The processed training set is fitted, and the AUC value of the processed validation set is obtained. When the AUC value is greater than or equal to a preset parameter, the air conditioning temperature prediction model training is complete. The AUC value is an indicator used to judge the training effect of the air conditioning temperature prediction model. The information sending module is used to send the predicted air conditioning temperature to the vehicle terminal so that the vehicle terminal can control the air conditioning temperature.

8. A vehicle terminal, characterized in that, include: The parameter acquisition module is used to acquire user information, vehicle operating parameters, and vehicle environmental parameters. The information sending module is used to send the user information, vehicle operating parameters, and vehicle environmental parameters to the server; The information receiving module is used to receive the predicted air conditioning temperature returned by the server. The server stores historical air conditioning temperature data corresponding to the user information, as well as an air conditioning temperature prediction model. The air conditioning temperature prediction model is used to predict the probability of the user setting each air conditioning temperature level, and to predict the predicted air conditioning temperature. The training process of the air conditioning temperature prediction model includes: receiving historical air conditioning temperatures, historical vehicle operating parameters, and historical vehicle environmental parameters sent by each vehicle terminal; dividing the historical air conditioning temperatures by time; marking historical air conditioning temperatures with a setting duration greater than or equal to a preset duration; and obtaining the positive value of the historical air conditioning temperature. The sample data is used to extract a preset number of data points from the positive sample data as a validation set, and the remaining positive sample data as a training set. Feature engineering is performed on the historical air conditioning temperature, historical vehicle operating parameters, and historical vehicle environmental parameters to obtain derived features. The derived features are then concatenated to the training set and the validation set to obtain processed training and validation sets. The processed training set is fitted, and the AUC value of the processed validation set is obtained. When the AUC value is greater than or equal to a preset parameter, the air conditioning temperature prediction model training is complete. The AUC value is an indicator used to judge the training effect of the air conditioning temperature prediction model. An air conditioning control module is used to control the air conditioning temperature based on the predicted air conditioning temperature.

9. An air conditioning temperature control system, characterized in that, include: The server is used to receive user information, vehicle operating parameters, and vehicle environmental parameters sent by the vehicle terminal, input the user information, vehicle operating parameters, and vehicle environmental parameters into the air conditioning temperature prediction model, and determine the predicted air conditioning temperature; wherein, the air conditioning temperature prediction model is used to predict the probability of the user setting each air conditioning temperature level, predict the predicted air conditioning temperature, and send the predicted air conditioning temperature to the vehicle terminal for the vehicle terminal to control the air conditioning temperature. The vehicle terminal is used to acquire user information, vehicle operating parameters, and vehicle environmental parameters, send the user information, vehicle operating parameters, and vehicle environmental parameters to the server, and receive the predicted air conditioning temperature returned by the server. The server stores historical air conditioning temperature data corresponding to the user information and an air conditioning temperature prediction model. The air conditioning temperature prediction model is used to predict the probability of the user setting each air conditioning temperature level, predict the predicted air conditioning temperature, and control the air conditioning temperature based on the predicted air conditioning temperature. The training process of the air conditioning temperature prediction model includes: receiving historical air conditioning temperatures, historical vehicle operating parameters, and historical vehicle environmental parameters sent by various vehicle terminals; dividing the historical air conditioning temperatures according to time; marking the historical air conditioning temperatures with a set duration greater than or equal to a preset duration to obtain positive sample data of the historical air conditioning temperatures; extracting a preset number of data from the positive sample data as a validation set, and using the remaining positive sample data as a training set; performing feature engineering on the historical air conditioning temperatures, historical vehicle operating parameters, and historical vehicle environmental parameters to obtain derived features from the feature engineering; concatenating the derived features to the training set and the validation set to obtain processed training set and validation set; fitting the processed training set and obtaining the AUC value of the processed validation set; when the AUC value is greater than or equal to a preset parameter, the training of the air conditioning temperature prediction model is completed, wherein the AUC value is an indicator used to judge the training effect of the air conditioning temperature prediction model.

Images