A driving data recording method special for new energy vehicles and a tire

Abstract

The application discloses a new energy automobile special driving data recording method and a tire, and belongs to the technical field of new energy automobile tires. The method comprises the following steps: deploying a monitoring module to a tire of a new energy automobile to collect tire driving data of the tire; transmitting the tire driving data to a vehicle control unit and a preset cloud in real time based on a preset wireless transmission protocol; acquiring vehicle driving parameters, environment sensing data and historical maintenance records based on the new energy automobile, and processing the tire driving data, the vehicle driving parameters, the environment sensing data and the historical maintenance records based on a tire state analysis model to generate a tire health degree and a risk warning; when the risk warning exceeds a preset threshold, processing the new energy automobile based on a preset vehicle control strategy; and estimating the remaining service life of the tire based on the tire health degree and a preset tire material degradation curve. The application can record tire driving data and recommend tires according to the tire driving data.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent tires, and in particular to a driving data recording method for new energy vehicles and a tire. BACKGROUND

[0002] With the rapid development of new energy vehicles, tires, as the key components directly contacting the road, directly affect the safety, energy efficiency and driving experience of vehicles.

[0003] Traditional tire monitoring technology mainly relies on a single sensor (such as a tire pressure sensor) to obtain data and performs early warning through simple threshold judgment, which has the problems of single data dimension and insufficient analysis accuracy. For example, the existing tire pressure monitoring system (TPMS) can real-time feedback tire pressure and temperature, but cannot comprehensively evaluate the health of the vehicle in multiple dimensions by combining vehicle dynamic parameters and environmental factors.

[0004] And the software-defined tire (STPMS) technology is gradually emerging, which indirectly infers the tire state by fusing vehicle dynamic data, but it relies on the generalization ability of the algorithm model, which is prone to misjudgment under complex working conditions. For example, the wheel speed difference method is difficult to accurately identify tire pressure abnormalities when the vehicle load changes or the road is uneven. At the same time, the existing tire recommendation is mainly based on static parameter matching, and lacks dynamic optimization of driving habits and environmental adaptability, resulting in insufficient adaptability.

[0005] Therefore, how to record tire driving data and tires with tire driving data have become a technical problem to be solved. SUMMARY

[0006] The embodiments of the present application provide a driving data recording method for new energy vehicles and a tire, which solve the technical problem of how to record tire driving data and tires with tire driving data.

[0007] In a first aspect, the embodiments of the present application provide a driving data recording method special for new energy vehicles, applied to a new energy vehicle, and the method comprises: deploying a monitoring module to a tire of the new energy vehicle to collect tire driving data of the tire; wherein the monitoring module comprises a temperature sensor, a pressure sensor, a three-axis acceleration sensor, and a wireless communication unit, and the tire driving data comprises temperature, pressure, and three-axis acceleration; transmitting the tire driving data to a vehicle control unit and a preset cloud in real time based on a preset wireless transmission protocol; wherein the vehicle control unit comprises a preset tire state analysis model; obtaining vehicle driving parameters, environment perception data, and historical maintenance records based on the new energy vehicle, and processing the tire driving data, the vehicle driving parameters, the environment perception data, and the historical maintenance records based on the tire state analysis model to generate a tire health degree and a risk warning; when the risk warning exceeds a preset threshold, processing the new energy vehicle based on a preset vehicle control strategy; and estimating a remaining service life of the tire based on the tire health degree and a preset tire material degradation curve.

[0008] In an implementation manner of the present application, the monitoring module is deployed to the tire of the new energy vehicle to collect the tire driving data of the tire, and specifically comprises: encapsulating the temperature sensor, the pressure sensor, and the three-axis acceleration sensor to generate a first monitoring module; connecting the first monitoring module with the wireless communication unit to generate the monitoring module; deploying the monitoring module to a preset position on an inner side of a crown of the tire to generate a monitoring band consistent with a curvature of the tire, and embedding the monitoring band between structure layers of a tire body in a tire forming stage; when the three-axis acceleration sensor monitors that an acceleration exceeds a preset vehicle speed threshold, increasing a temperature sampling frequency of the temperature sensor, and performing drift compensation on the pressure sensor based on a reference pressure value of a static state of the tire; and integrating acceleration, temperature, and pressure based on time sequence to generate the tire driving data.

[0009] In an implementation form of the application, the vehicle driving parameters, the environment perception data and the historical maintenance records are acquired based on the new energy vehicle, and the tire driving data, the vehicle driving parameters, the environment perception data and the historical maintenance records are processed based on the tire state analysis model to generate the tire health degree and the risk warning, specifically comprising: acquiring the vehicle driving parameters from the CAN bus of the new energy vehicle; wherein the vehicle driving parameters at least include real-time vehicle speed, motor output torque and battery load state; acquiring the environment perception data through a preset vehicle-mounted environment sensor; wherein the environment perception data at least include road surface humidity, external temperature and air density; calling the historical maintenance records through the database of the cloud; wherein the historical maintenance records include tire replacement cycle, tire pressure calibration records and wear repair log; preprocessing the tire driving data, the vehicle driving parameters and the environment perception data to generate a standardized input vector; inputting the standardized input vector into the tire state analysis model to calculate the tire health degree score; determining the risk warning based on the environment perception data, the vehicle driving parameters and the tire health degree score.

[0010] In an implementation form of the application, the construction of the tire state analysis model specifically comprises: constructing a hybrid model architecture comprising an LSTM network and a fully connected layer; wherein the LSTM network is used to process the time series tire driving data, and the fully connected layer is used to fuse the vehicle driving parameters and the environment perception data; determining training data; wherein the training data includes historical tire failure samples, normal wear samples and artificially labeled health degree labels; training the hybrid model architecture through a preset supervised learning method and the training data; minimizing the mean square error loss between the predicted health degree and the actual health degree label based on a preset Adam optimizer; when the mean square error loss is less than a preset success threshold, outputting the tire state analysis model.

[0011] In an implementation form of the application, when the risk warning exceeds a preset threshold, the new energy vehicle is processed based on a preset vehicle control strategy, specifically comprising: calculating a suspension damping coefficient adjustment amount according to the pressure and temperature, and adjusting the tire support stiffness through the electric control suspension system of the new energy vehicle; communicating the risk warning with the motor control module of the new energy vehicle based on the risk warning, and reducing the motor output power to a safety threshold range by a preset proportion; generating a visual warning icon on the vehicle-mounted interactive interface of the new energy vehicle, and playing real-time risk prompt information.

[0012] In an implementation form of the present application, the remaining service life of the tire is calculated based on the tire health degree and a preset tire material degradation curve, specifically comprising: matching a material degradation curve corresponding to the rubber formula of the tire based on a preset material database; wherein the material degradation curve is used to describe the relationship between the elastic modulus and the temperature and time; superimposing the tire health degree and the material degradation curve for analysis to calculate the theoretical remaining wear thickness of the tire; processing the remaining wear thickness based on the average speed and load data in the vehicle driving parameters to determine the remaining service life.

[0013] In an implementation form of the present application, the method further comprises: defining a multi-objective optimization function; wherein the objectives of the multi-objective optimization function include minimizing rolling resistance, maximizing wet road grip and balancing wear distribution; setting constraint conditions; wherein the constraint conditions include tire size matching range, maximum load capacity and speed level limit; obtaining driving habits of the new energy vehicle; adjusting the weight coefficient of the multi-objective optimization function according to the driving habit data; searching for candidate tire models that meet the constraint conditions in the tire recommendation database through a preset genetic algorithm; processing each candidate tire model based on the multi-objective optimization function to calculate the adaptation score corresponding to each candidate tire model.

[0014] In an implementation form of the present application, after processing each candidate tire model based on the multi-objective optimization function to calculate the adaptation score corresponding to each candidate tire model, the method further comprises: setting a remaining service life threshold; when the remaining service life of the tire is lower than the remaining service life threshold, arranging the candidate tires in descending order according to the adaptation score, and eliminating the candidate tire models that do not meet the environmental temperature and / or road humidity requirements to generate a first alternative set; weighting matching the brand preference data in the user historical replacement record with the first alternative set to generate a recommended tire model set containing recommended models, performance comparison data and replacement urgency; synchronizing the recommended tire model set to the vehicle-mounted interaction interface of the new energy vehicle and the user account in the cloud.

[0015] In a second aspect, the embodiments of the present application also provide a special driving data recording tire for new energy vehicles, which is applied to the special driving data recording method for new energy vehicles, and comprises: a flexible monitoring belt embedded between the tire body structure layers of the tire; the flexible monitoring belt comprises a flexible circuit board base body, a temperature sensor, a pressure sensor, a three-axis acceleration sensor and a wireless communication unit packaged on the flexible circuit board base body; the temperature sensor is used for monitoring the internal temperature of the tire, the pressure sensor is used for measuring the air pressure in the tire, and the three-axis acceleration sensor is used for detecting the longitudinal, lateral and vertical accelerations of the tire; the wireless communication unit comprises a dual-mode transmission module, supports a main channel of a short-distance wireless protocol and a standby channel of a long-distance wireless protocol, and is used for transmitting the data of the temperature sensor, the pressure sensor and the three-axis acceleration sensor to a vehicle control unit and a cloud.

[0016] In an implementation manner of the present application, the three-axis acceleration sensor is connected with the temperature sensor, and when the longitudinal acceleration is monitored to exceed a preset vehicle speed threshold, the sampling frequency of the temperature sensor is increased.

[0017] The special driving data recording method and tire for new energy vehicles provided by the embodiments of the present application at least have the following technical effects:

[0018] Data acquisition and transmission: the sensors are integrated into the tire body interior by the flexible circuit board packaging technology, so that the signal distortion problem caused by the deformation of the traditional external sensor is avoided; the dynamic sampling frequency adjustment and the pressure drift compensation mechanism are combined, so that the acquisition accuracy of the temperature, pressure and acceleration data is significantly improved. The redundant transmission channel design and the block check mechanism effectively guarantee the integrity and stability of data transmission, and solve the communication reliability problem under complex working conditions.

[0019] Risk assessment and control: the multi-source data fusion analysis based on the hybrid model architecture (LSTM network and full connection layer) can accurately identify the risks such as tire wear and temperature anomaly to a certain extent, and adjust the suspension stiffness and motor output power in real time through the linkage vehicle control system, so as to improve the control stability and safety of the vehicle under abnormal state.

[0020] Life prediction and recommendation: the superimposed analysis of the tire material degradation characteristics and real-time driving data is combined to realize the fine prediction of the remaining service life; through the multi-objective optimization algorithm, the rolling resistance, grip and wear balance indexes are dynamically balanced, and the personalized recommendation scheme is generated combined with the user driving habit and environmental adaptability constraint, so as to improve the scientificity of tire selection and user adaptation degree.

[0021] In summary, from multi-dimensional data acquisition, health status evaluation to intelligent decision-making, a complete technical system is realized to upgrade tire management from passive response to active prevention. Through the cooperation of the cloud and the vehicle terminal, real-time synchronization of risk warning, control strategy and replacement recommendations is ensured, effectively reducing the safety hazards caused by tire failure, optimizing vehicle energy efficiency and endurance performance, and promoting the intelligent operation and maintenance level of new energy vehicles. BRIEF DESCRIPTION OF DRAWINGS

[0022] The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and serve to explain the principles of the application. In the drawings:

[0023] Figure 1 A new energy vehicle special driving data recording method and tire flowchart are provided for the embodiments of the application. DETAILED DESCRIPTION

[0024] To make the objectives, technical solutions and advantages of the application clearer, the technical solutions of the application will be described below in conjunction with the specific embodiments of the application and corresponding drawings. Obviously, the described embodiments are only some of the embodiments of the application, not all. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the application.

[0025] The embodiments of the application provide a new energy vehicle special driving data recording method and tire, to solve the following technical problems: how to record tire driving data, and tire with tire driving data.

[0026] The technical solutions of the embodiments of the application will be described in detail below with reference to the drawings.

[0027] Figure 1 A new energy vehicle special driving data recording tire recording flowchart is provided for the embodiments of the application. As shown in Figure 1 The new energy vehicle special driving data recording method provided by the embodiments of the application specifically includes the following steps:

[0028] Step 1, deploy a monitoring module to the tire of the new energy vehicle to collect tire driving data of the tire; wherein the monitoring module includes a temperature sensor, a pressure sensor, a three-axis acceleration sensor and a wireless communication unit, and the tire driving data includes temperature, pressure and three-axis acceleration.

[0029] Step 1.1, encapsulate the temperature sensor, pressure sensor and three-axis acceleration sensor by flexible circuit board packaging technology to generate a first monitoring module.

[0030] Flexible circuit board packaging technology refers to the use of flexible and deformation-resistant circuit board (such as polyimide material) to integrate sensors into the same flexible carrier, so that it can be attached to the tire curve. Temperature sensors are used to monitor the internal temperature changes of the tire in real time, pressure sensors are used to measure the tire pressure, and three-axis acceleration sensors are used to detect the lateral, longitudinal and vertical acceleration of the tire.

[0031] In a specific example, the pins of the temperature sensor (such as NTC thermistor), pressure sensor (such as MEMS piezoresistive sensor) and three-axis acceleration sensor (such as ADXL345 chip) are welded to the corresponding interfaces of the flexible circuit board.

[0032] The above three sensors are packaged with the flexible substrate by vacuum lamination process to form a sheet-like structure (i.e. the first monitoring module).

[0033] Step 1.2, connect the first monitoring module with the wireless communication unit to generate the monitoring module.

[0034] The wireless communication unit refers to a communication module that supports Bluetooth Low Energy (BLE) or LoRa protocol, which is used to transmit sensor data in real time to the vehicle terminal or cloud.

[0035] In a specific example, a communication interface is reserved on the flexible circuit board of the first monitoring module, the pins of the wireless communication unit (such as ESP32 chip) are connected to the interface through conductive silver paste, and epoxy resin glue is used to fix the connection part to prevent falling caused by vibration.

[0036] It can be understood that after the connection is completed, the signal stability of the communication link can be tested, and when the data transmission error rate is lower than a certain threshold (usually set to 0.1%) when the tire is rotating at high speed, it means that the communication module is functioning properly.

[0037] Step 1.3, deploy the monitoring module to the inside of the tire crown at a predetermined position to generate a monitoring band that conforms to the curvature of the tire, and embed it between the tire body structure layers during the tire forming stage.

[0038] The inside of the tire crown refers to the inner surface of the tire contact area with the ground; the monitoring band is a strip-shaped monitoring area arranged along the circumference of the tire, and its length is slightly shorter than the outer circumference of the tire.

[0039] In a specific example, during the tire forming process, the monitoring module is attached to the tire body ply and belt layer on the inside of the tire crown in a circumferential arrangement, and the monitoring module is combined with the rubber material through the vulcanization process. After completion, the monitoring module is entirely covered with rubber.

[0040] Step 1.4, when the triaxial acceleration sensor detects acceleration exceeding a preset vehicle speed threshold, increase the temperature sampling frequency of the temperature sensor, and compensate for the drift of the pressure sensor based on the reference pressure value of the tire static state.

[0041] The preset vehicle speed threshold is set according to the tire specification (for example, the acceleration value corresponding to 80 km / h); the reference pressure value is the standard tire pressure value when the vehicle is static and the tire is in a cooling state.

[0042] In a specific example, the longitudinal acceleration threshold of the triaxial acceleration sensor is set to 3 m / s² (corresponding to a vehicle speed of 80 km / h). When the acceleration exceeds the threshold, the sampling frequency of the temperature sensor is increased from 1 Hz to 10 Hz to capture the temperature rise caused by high-speed driving. At the same time, the pressure sensor reads the reference pressure value (such as 2.5 bar) in the static state every 5 minutes, and takes this value as the reference to eliminate the zero drift error caused by long-term work of the sensor by Kalman filtering algorithm. Kalman filtering algorithm is a prior art, which will not be described here.

[0043] Step 1.5, integrating acceleration, temperature and pressure based on time sequence to generate the tire driving data.

[0044] Time sequence integration refers to adding time stamp to the data of each sensor and aligning them on time axis before packing them as structured data.

[0045] In a specific example, a real-time clock (RTC) module is embedded in the wireless communication unit to mark the collection time of temperature, pressure and acceleration data respectively. Through data fusion algorithm, multi-dimensional data in the same time window are combined into a record, for example, temperature (70℃), pressure (2.8 bar) and longitudinal acceleration (2.5 m / s²) at t=10:00:00 are associated as a piece of tire driving data.

[0046] Step 2, transmitting the tire driving data in real time to the vehicle control unit and the preset cloud based on the preset wireless transmission protocol; wherein the vehicle control unit includes a preset tire state analysis model.

[0047] Step 2.1, constructing a redundant transmission channel; wherein the redundant transmission channel includes a main channel and a backup channel, the main channel is a short-distance wireless protocol channel, and the backup channel is a long-distance wireless protocol transmission channel.

[0048] The main channel is paired with the Bluetooth receiving end of the vehicle control unit, and the backup channel is connected to the cloud server through the SIM card. The main channel and the backup channel work in parallel, and the main channel is used by default to transmit data.

[0049] The dual-mode communication module is integrated in the vehicle control unit (such as the vehicle-mounted ECU), the main channel is configured as the BLE protocol, and the standby channel is configured as the LoRa protocol. The main channel is bound with the vehicle-mounted terminal, and the standby channel is directly connected to the preset cloud. The main channel and the standby channel share the same data source (monitoring module), but transmit through independent antennas and frequency bands.

[0050] In a specific example, a new energy vehicle enterprise A sets the transmission power of the BLE main channel to 4dBm (coverage radius of 30 meters) and the transmission power of the LoRa standby channel to 20dBm (coverage radius of 1.2 kilometers) when designing the redundant transmission channel. When the vehicle is driving in the urban area, the BLE channel is preferred for communication with the vehicle-mounted terminal; when the vehicle enters the signal shielding area (such as the underground garage), the LoRa channel is automatically enabled to upload data to the cloud.

[0051] Step 2.2, based on the preset signal strength threshold, monitor the transmission stability of the main channel, and when the signal strength of the main channel is lower than the signal strength threshold, start the standby channel.

[0052] The signal strength threshold refers to the critical signal strength value that triggers the channel switching, which is usually set according to the minimum stable transmission requirement of the communication protocol. For example, the signal strength threshold of the BLE protocol can be set to -80dBm.

[0053] The transmission stability monitoring is realized by calculating the received signal strength indication (RSSI) and the bit error rate (BER) in real time to determine whether the main channel meets the data transmission requirements.

[0054] When RSSI is detected below the threshold for 3 consecutive times, the channel start instruction is triggered.

[0055] In a specific example, the vehicle control unit of a certain vehicle model B sets the BLE signal strength threshold to -80dBm. When the vehicle enters the tunnel and the BLE signal strength drops to -85dBm, the LoRa standby channel is started.

[0056] Step 2.3, divide the tire driving data into multiple data packets and add a check code in each data packet.

[0057] The block packaging refers to dividing the complete tire driving data into multiple data packets according to the preset size, for example, each data packet contains 256 bytes of payload.

[0058] The check code is used to verify the integrity of the data encoding.

[0059] A single piece of tire driving data (such as timestamp + temperature + pressure + acceleration) is sequentially split into three data packets: header (timestamp), middle (sensor value), and tail (check code).

[0060] Add a checksum: Add a checksum to each data packet and append it to the end of the data packet.

[0061] In a specific example, a measurement module generated single tire driving data as "2023-10-01 10:00:00, 70℃, 2.8bar, 2.5m / s²", with a total length of 64 bytes. This can be broken down into:

[0062] Header data packet (20 bytes): contains timestamp and checksum;

[0063] Middle data packet (40 bytes): contains temperature, pressure, acceleration data and checksum;

[0064] Tail data packet (4 bytes): Reserved for redundant check bits.

[0065] Step 2.4: Transmit different data packets synchronously through the main channel and the backup channel, and verify the data integrity at the receiving end of the vehicle control unit and the cloud through a preset check code.

[0066] Once the backup channel is activated, synchronous transmission is initiated. Synchronous transmission refers to the simultaneous transmission of different data packets by the primary channel and the backup channel. For example, the primary channel transmits the header and middle data packets, while the backup channel transmits the tail and redundant backup packets.

[0067] Data integrity verification involves the receiving end verifying the integrity of the data packet using a checksum. If the verification fails, a retransmission is requested.

[0068] The transmission strategy prioritizes the transmission of critical data (such as headers and middle sections) through the main channel, while the backup channel transmits auxiliary data (such as tails and redundant packets).

[0069] The receiving end processes the data packets by receiving them from both the vehicle-mounted terminal and the cloud server, and verifying their integrity using a checksum. If a data packet fails to be verified, a retransmission request is sent to the sending end.

[0070] In a specific example, a cloud server C receives a header data packet (CRC check passed) from the primary channel and a tail data packet (CRC check failed) from the backup channel. The server then extracts the tail data packet from the redundant backup of the backup channel, re-verifies it, and completes the data splicing to ultimately generate a complete tire driving data record.

[0071] Step 3: Based on the new energy vehicle, obtain vehicle driving parameters, environmental perception data and historical maintenance records, and process the tire driving data, vehicle driving parameters, environmental perception data and historical maintenance records based on the tire condition analysis model to generate tire health and risk warnings.

[0072] Step 3.1, Obtain vehicle driving parameters from the CAN bus of the new energy vehicle; wherein the vehicle driving parameters at least include real-time vehicle speed, motor output torque, and battery load state.

[0073] The CAN bus is a communication network between internal electronic control units (ECUs) of a vehicle, used to transmit real-time vehicle state information.

[0074] In a specific example:

[0075] Data collection: access the CAN bus through the OBD-II interface and read the following parameters in real time:

[0076] Real-time vehicle speed: analyze CAN message ID 0x0CFF0000 and convert it to a physical quantity (km / h) with a sampling frequency of 10 Hz.

[0077] Motor output torque: analyze ID 0x0CF00400 and convert it to a torque value (Nm) with an accuracy of ±2%.

[0078] Battery load state: analyze ID 0x0CF00300 to obtain the current load percentage of the battery.

[0079] Abnormality detection: if the vehicle speed exceeds the design limit value (e.g. 300 km / h) for 5 consecutive times, it is determined to be an abnormal signal and the cached data is used instead.

[0080] Step 3.2, Obtain environmental perception data through pre-set vehicle-mounted environmental sensors; wherein the environmental perception data at least includes road surface humidity, external temperature, and air density.

[0081] Vehicle-mounted environmental sensors are used to monitor external environmental parameters and assess their real-time impact on tire performance.

[0082] It can be understood that the environmental data and vehicle driving parameters are aligned by a unified timestamp.

[0083] Step 3.3, Retrieve historical maintenance records from the cloud database; wherein the historical maintenance records include tire replacement cycle, tire pressure calibration records, and wear repair logs.

[0084] Historical maintenance records are used to correct the timeliness of the health assessment and reflect the actual use history of the tires.

[0085] In a specific example:

[0086] Data query and matching:

[0087] Request the following data from the cloud through the vehicle VIN code:

[0088] Tire replacement cycle: Record tire brand, installation date, cumulative mileage, and replacement reason (e.g., wear, bulge).

[0089] Tire pressure calibration record: Time of the last 5 calibrations, calibration value (bar), and operating equipment model.

[0090] Wear repair log: Repair location (e.g., crown left area), repair technology (patch / mushroom nail), and post-repair test results.

[0091] Weight distribution: Assign weight coefficients based on the time of maintenance records (e.g., 1.0 for records within 1 year, and 0.15 for each additional year).

[0092] Step 3.4, preprocess the tire driving data, vehicle driving parameters, and environmental perception data to generate standardized input vectors.

[0093] Standardized input vectors convert multi-source heterogeneous data into a unified format that the model can process.

[0094] In one specific example:

[0095] Data cleaning and completion:

[0096] Median filtering (window size 3 seconds) is used on tire temperature data to eliminate impulse noise.

[0097] If the road surface humidity data is missing, use the adjacent vehicle data in the same time period to interpolate and fill in.

[0098] Vehicle speed is normalized to 0-1 (based on the maximum design speed of the vehicle, such as 200 km / h).

[0099] Road surface humidity is discretely encoded as 0 (dry, 0-1 mm), 1 (wet, 1-3 mm), and 2 (water accumulation, >3 mm).

[0100] The historical maintenance weight is directly used as an input feature according to the time decay coefficient (e.g., a record 2 years ago has a weight of 0.7).

[0101] Time alignment: Align the 60 time steps (1 point per minute) of the tire data with the vehicle parameters and environmental data on the time axis.

[0102] Step 3.5, input the standardized input vectors into the tire state analysis model to calculate the tire health score.

[0103] Model processing flow:

[0104] LSTM layer input: 60 time steps x 5 features (temperature, pressure, three-axis acceleration).

[0105] Fully connected layer input: LSTM output features (128 dimensions) + vehicle parameters (3 dimensions) + environmental data (3 dimensions) + maintenance weight (1 dimension), totaling 135 dimensions.

[0106] In one specific example, a dynamic correction rule is also included:

[0107] If the historical maintenance record shows that the tire has been used for more than the manufacturer's recommended life (e.g., 5 years), the health score is multiplied by a time decay coefficient of 0.8.

[0108] If there is a tire pressure calibration record within the last 3 months, the score is additionally increased by 5 points.

[0109] Step 3.6, determine the risk warning based on environmental perception data, vehicle driving parameters and tire health score.

[0110] Risk warning triggers graded alerts according to the combination of health score and real-time environmental / vehicle state.

[0111] In one specific example:

[0112] Level 1 warning (emergency): health score ≤ 40 points and (vehicle speed ≥ 100 km / h or road surface wetness = level 2).

[0113] Level 2 warning (high risk): health score ≤ 60 points and (battery load ≥ 90% or external temperature ≥ 45°C).

[0114] Level 3 warning (prompt): health score ≤ 80 points and cumulative mileage exceeds 80% of the recommended replacement mileage.

[0115] The construction process of the tire state analysis model is:

[0116] A1, build a hybrid model architecture including LSTM network and fully connected layer; wherein the LSTM network is used to process time series tire driving data, and the fully connected layer is used to fuse vehicle driving parameters and environmental perception data.

[0117] LSTM network (Long Short-Term Memory Network) is a variant of recurrent neural network, which controls information flow through forget gate, input gate and output gate, and is suitable for processing dynamic features of time series data such as tire temperature and pressure.

[0118] The input dimension is set to time step (e.g., 60 seconds of sensor data) and feature number (temperature, pressure, three-axis acceleration, a total of 5 features).

[0119] A 3-layer LSTM network is built, each layer containing 128 neurons, with the initial forget gate bias set to 1.0 to prevent gradient vanishing, and the time step fixed at 60 (corresponding to data sampled once every minute).

[0120] The LSTM layer outputs the hidden state of each time step, which is compressed into a single feature vector (dimension 128) by an average pooling layer.

[0121] The fully connected layer is used to fuse vehicle driving parameters (e.g. speed, torque) and environmental perception data (e.g. road wetness), combining them with the time-series features extracted by LSTM.

[0122] Data alignment: The feature vector (dimension 128) output by LSTM is concatenated with vehicle driving parameters (3 features), environmental data (3 features), and maintenance weights (1 dimension) to form a unified input vector (dimension 135).

[0123] First layer: 64 neurons, activation function using ReLU (Rectified Linear Unit), used to extract non-linear features.

[0124] Second layer: 32 neurons, activation function using ReLU, further dimension reduction.

[0125] Output layer: 1 neuron, activation function using Sigmoid, scaling the output value to a 0-100 health score.

[0126] A2, determine the training data; wherein the training data includes historical tire failure samples, normal wear samples and artificially labeled health labels.

[0127] The training data needs to cover the whole life cycle state of the tire, including extreme failure and normal wear scenarios, and the health labels are based on expert experience to quantify the remaining life of the tire.

[0128] Extract time-series data of tire temperature, pressure, and acceleration from the vehicle terminal, and simultaneously record vehicle speed, torque from the OBD interface, and road wetness collected by environmental sensors.

[0129] A3, train the hybrid model architecture through a preset supervised learning method and the training data.

[0130] In a specific example, 10,000 samples are divided into training set, validation set, and test set according to 7:2:1, and mean square error (MSE) is used to calculate the difference between predicted health and actual label.

[0131] The training strategy can be set as:

[0132] Batch size: 32 samples per batch.

[0133] Learning rate: initial value 0.001, decay 50% every 10 epochs.

[0134] Early stopping mechanism: terminate training if validation loss does not decrease for 5 consecutive epochs.

[0135] A4、Minimize the mean squared error loss between the predicted health and the actual health label based on the preset Adam optimizer.

[0136] Adam optimizer combines momentum method and adaptive learning rate, suitable for optimization of non-stationary objective function.

[0137] In one specific example:

[0138] Parameter setting: first moment estimation decay rate β1=0.9, second moment estimation decay rate β2=0.999, epsilon=1e-7.

[0139] Gradient clipping: set the gradient norm threshold to 5.0 to prevent gradient explosion.

[0140] Weight regularization: apply L2 regularization (coefficient 0.01) to the weights of the fully connected layer to suppress overfitting.

[0141] A5、When the mean squared error loss is less than the preset success threshold, output the tire state analysis model.

[0142] The preset success threshold is that the mean squared error of the validation set is less than 1.0 points (i.e., the average deviation between the predicted health and the actual label is less than 1 point).

[0143] For example, if the validation set loss fluctuation range is less than 0.1 for 3 consecutive epochs, the model is considered to have converged. The model parameters with the minimum validation set loss are selected as the tire state analysis model.

[0144] Step 4, when the risk warning exceeds the preset threshold, handle the new energy vehicle based on the preset vehicle control strategy.

[0145] Step 4.1, calculate the suspension damping coefficient adjustment amount according to the pressure and temperature, and adjust the tire support stiffness through the electric control suspension system of the new energy vehicle.

[0146] Suspension damping coefficient adjustment amount refers to the suspension system stiffness adjustment parameter dynamically calculated according to the real-time pressure and temperature of the tire, used to optimize the contact performance of the tire and the road.

[0147] In one specific example:

[0148] Based on the tire pressure (unit: bar) and temperature (unit: ℃), the basic damping coefficient (for example, pressure 2.5 bar and temperature 70 ℃ correspond to damping coefficient C0) is determined through the preset mapping table.

[0149] The compensation coefficient a is calculated according to the pressure change rate (ΔP / Δt) and the temperature gradient (ΔT / Δt), and the adjustment amount is C = C0 × (1 + a).

[0150] Send adjustment instructions to the electronic control suspension system through the CAN bus, and write the target damping coefficient into the suspension control module (such as CDC continuous damping control system).

[0151] Step 4.2, based on risk early warning and motor control module of the new energy vehicle, reduce the motor output power to the safety threshold range according to the preset proportion.

[0152] The safety threshold range refers to the maximum allowed output power dynamically calculated according to the current state of the vehicle (such as vehicle speed, battery load).

[0153] In a specific example:

[0154] High risk warning: immediately reduce the motor output power to 50% of the current value.

[0155] Medium risk warning: reduce power in stages, reduce by 10% every 5 seconds, until the safety threshold (such as 70% of the maximum power) is reached.

[0156] Step 4.3, generate a visual warning icon on the vehicle-mounted interaction interface of the new energy vehicle, and broadcast real-time risk prompt information.

[0157] The visual warning icon dynamically reflects the risk level through color (red / yellow / green) and graphics, and the voice prompt information contains specific risk types and suggestions.

[0158] In a specific example:

[0159] High risk warning: the central control screen displays a flashing red tire icon, with the text "stop and check immediately".

[0160] Medium risk warning: display a yellow exclamation mark icon and the text "suggest reducing speed and contacting service".

[0161] Play synthesized voice through the vehicle-mounted audio system, including risk location (such as "left front tire pressure anomaly") and recommended operation (such as "please reduce the vehicle speed to 60km / h").

[0162] Step 5, calculate the remaining service life of the tire based on the tire health and the preset tire material degradation curve.

[0163] Step 5.1, match the material degradation curve corresponding to the rubber formula of the tire based on the preset material database; wherein the material degradation curve is used to describe the relationship between elastic modulus and temperature and time.

[0164] The material database stores experimental data curves of elastic modulus vs. temperature and time for different rubber formulations, each curve corresponding to a specific tire material aging law.

[0165] In one specific example:

[0166] The tire model and rubber formulation code (e.g. "NR-SBR-70A") is obtained through the RFID tag or QR code on the tire sidewall, and the corresponding degradation curve is retrieved from the material database.

[0167] The degradation curve is stored in a two-dimensional chart format, with the horizontal axis representing cumulative usage time (hours) and the vertical axis representing the elastic modulus decay rate (%), and accompanied by a temperature correction coefficient table (e.g. the aging rate at 70°C is 3.2 times that at 25°C).

[0168] The degradation curve is weighted according to the tire historical temperature data (collected in step 3), for example, a certain tire runs at 50°C for 70% of the time and at 30°C for 30% of the time, then the equivalent aging time is calculated.

[0169] Step 5.2, superimpose the tire health score with the material degradation curve to calculate the theoretical remaining wear thickness of the tire.

[0170] Superimposition analysis refers to combining the health score (reflecting the current wear state) with the material degradation curve (predicting future aging rate) to calculate the remaining usable thickness of the tire tread.

[0171] In one specific example:

[0172] The health score (0-100 points) is linearly mapped to the current tread remaining thickness percentage (for example, a health score of 80 points corresponds to a remaining thickness of 80% of the initial value).

[0173] According to the material degradation curve, the time integral of the elastic modulus decay rate is performed from the current time, to predict the thickness reduction per 1000 kilometers of future driving.

[0174] Assuming the initial thickness of the tire is 8mm, the current health score is 60 points, corresponding to a remaining thickness of 4.8mm, and the superimposed prediction is that the thickness will decrease by 1.2mm every 5000 kilometers of future driving, then the theoretical remaining wear thickness is 3.6mm.

[0175] Step 5.3, process the remaining wear thickness based on the average speed and load data in the vehicle driving parameters to determine the remaining service life.

[0176] The remaining service life needs to be corrected according to the actual use conditions (such as accelerated wear due to high-speed driving, increased deformation due to heavy load) to the theoretical remaining thickness.

[0177] In one specific example:

[0178] Vehicle speed correction factor: preset vehicle speed wear factor table, for example, the wear rate is 1.8 times that of 40 km / h when the average vehicle speed is 80 km / h.

[0179] According to the ratio of the vehicle load to the rated load of the tire, the wear rate is linearly corrected (for example, the wear rate is increased by 30% when the overload is 20%).

[0180] Divide the theoretical residual thickness 3.6 mm by the corrected single-kilometer wear amount (for example, 0.00024 mm / km), to obtain the remaining driving mileage of 15,000 kilometers, which is converted into the remaining service life (assuming an average daily driving of 200 kilometers, the remaining life is 75 days).

[0181] In the overall case of this step, the tire two-dimensional code is scanned to obtain the formula code "NR-EPDM-65B", and the corresponding degradation curve in the database is called. The formula is that the elastic modulus is attenuated by 12% per 1000 hours at 60°C. Combined with historical data, the tire is operated at-18°C (cold chain environment) for 70% of the time and at 25°C (loading and unloading) for 30% of the time, and the equivalent aging rate is 0.35 times the standard curve.

[0182] The current health score is 72 points, which maps to a residual thickness of 5.76 mm (initial 8 mm x 72%).

[0183] According to the equivalent degradation curve, the thickness is reduced by 0.48 mm per 1000 kilometers of driving.

[0184] The vehicle average speed is 65 km / h (correction factor 1.5), and the load is 110% of the rated value (correction factor 1.2), and the comprehensive wear rate correction is 0.48 mm / 1000 km x 1.5 x 1.2 = 0.864 mm / 1000 km.

[0185] Remaining driving mileage = 5.76 mm ÷ 0.864 mm / 1000 km ≈ 6,667 kilometers, and the remaining service life is about 22 days according to the average daily driving of 300 kilometers.

[0186] The vehicle-mounted system prompts "the remaining life of the left front tire is 22 days", and suggests to make an appointment for replacement within 15 days.

[0187] After determining that the tire needs to be replaced, the present application also includes a method for recommending suitable tires according to the driving habits and driving environment of the customer.

[0188] Step B, processing the tire driving data, vehicle driving parameters, and environment perception data based on a preset multi-objective optimization algorithm and a tire recommendation database to generate an adaptation score.

[0189] Step B1, define a multi-objective optimization function; wherein the objectives of the multi-objective optimization function include minimizing rolling resistance, maximizing wet road grip, and balancing wear distribution.

[0190] The multi-objective optimization function is a mathematical model that simultaneously optimizes multiple performance indicators, used to quantify the overall performance of the tire in terms of rolling resistance, grip, and wear balance.

[0191] In a specific example:

[0192] Minimize rolling resistance: Calculate the energy consumption coefficient based on tire tread material and pattern design parameters, the lower the target value, the greater the potential for improved endurance.

[0193] Maximize wet road grip: Calculate the friction coefficient based on parameters such as groove depth and rubber hardness, the higher the target value, the better the braking performance.

[0194] Balancing wear distribution: Evaluate through tire temperature distribution variance and pressure distribution uniformity indicators, the smaller the variance, the more uniform the wear.

[0195] Function construction: Normalize the three objectives to 0-1 scores, and generate a comprehensive optimization target value by weighted summation, with initial weights set to rolling resistance 40%, grip 40%, and wear balance 20%.

[0196] Step B2, set constraints; wherein the constraints include tire size matching range, maximum load capacity, and speed rating limit.

[0197] Constraints are hard parameter limits that the tire must meet to ensure compatibility with the recommended model and vehicle.

[0198] Size matching range: Allow wheel diameter deviation of ±0.5 inches and cross-sectional width deviation of ±10%.

[0199] Maximum load capacity: The rated load of the candidate tire must be ≥ 110% of the vehicle's maximum total mass (including load).

[0200] Speed rating limit: The speed rating of the candidate tire (e.g., V-class 240km / h) must be ≥ 120% of the vehicle's maximum design speed.

[0201] Step B3, obtain the driving habits of the new energy vehicle.

[0202] Driving habits include acceleration frequency, average speed, braking intensity, and other characteristics, reflecting the actual use mode of the vehicle.

[0203] Data collection: Extract 90 days of driving data from the vehicle terminal, including:

[0204] The proportion of sudden acceleration (acceleration ≥ 2m / s²) times.

[0205] Average daily high-speed driving time (≥ 80 km / h).

[0206] Average braking deceleration and frequency.

[0207] Classify driving habits into "aggressive", "moderate", and "conservative", for example, if the proportion of sudden acceleration is > 15%, it is classified as aggressive.

[0208] Step B4: Adjust the weight coefficients of the multi-objective optimization function according to the driving habit data.

[0209] The weight coefficient adjustment is used to prioritize the core needs of different driving habits.

[0210] Aggressive driving: increase the grip weight to 50% and reduce the rolling resistance weight to 30%.

[0211] Moderate driving: maintain the initial weight (rolling resistance 40%, grip 40%, wear balance 20%).

[0212] Conservative driving: increase the wear balance weight to 30% and reduce the grip weight to 30%.

[0213] Step B5: Search for candidate tire models that meet the constraint conditions in the tire recommendation database through a pre-set genetic algorithm.

[0214] The genetic algorithm simulates the biological evolution process to select candidate tires that meet the constraint conditions and have high optimization target values.

[0215] Randomly select any model from the tire database as the initial population.

[0216] Calculate the constraint compliance of each model (such as size matching, load capacity) and eliminate those that do not meet the requirements.

[0217] Simulate gene crossing and random variation of the parameters of the remaining models (such as tread depth, rubber formula) to generate a new generation of population.

[0218] Repeat the fitness evaluation and population update until the optimal solution changes rate of consecutive iterations is less than the pre-set threshold.

[0219] Step B6: Process each candidate tire model based on the multi-objective optimization function to calculate the adaptation score corresponding to each candidate tire model.

[0220] Extract the rolling resistance coefficient of the candidate tire (such as EU label grade C), wet grip index (such as 1.2G), and wear test data (such as 5 million kilometers of wear 1.5mm).

[0221] The three indicators are put into the optimization function, for example, a certain tire rolling resistance gets 0.8 points, grip gets 0.9 points, wear balance gets 0.7 points, and the comprehensive score is 0.8*0.4+0.9*0.4+0.7*0.2=0.82 points.

[0222] The candidate tires are ranked in descending order according to the adaptation score, and the key performance parameters are marked.

[0223] In a specific example, a certain automobile platform B recommends suitable tires for vehicles:

[0224] Analysis shows that the proportion of vehicle sudden acceleration is 18%, and the daily average high-speed driving is 2 hours, which is determined as "aggressive driving".

[0225] The grip weight is increased to 50%, the rolling resistance weight is reduced to 30%, and the wear balance remains 20%.

[0226] From 1,000 tire models, 42 candidate models that meet the size of 215 / 55R17, load ≥750kg, and speed level ≥V level are selected.

[0227] Model X rolling resistance gets 0.75 points (C level), grip gets 0.95 points (AA level), and wear balance gets 0.65 points, with a comprehensive score of 0.75*0.3+0.95*0.5+0.65*0.2=0.815 points, ranking first.

[0228] Platform preferentially purchases model X.

[0229] Step C, determining the recommended tire model set according to the ranking result of the adaptation score and the remaining service life.

[0230] Step C1, set the remaining service life threshold.

[0231] The remaining service life threshold is a tire replacement trigger threshold set according to safety standards or user-defined requirements, usually based on the remaining driving distance or days.

[0232] Industry standard setting: refer to the replacement standard recommended by the tire manufacturer (such as the remaining tread thickness ≤1.6mm), and set the remaining service life threshold to 15 days (calculated based on daily driving 200km).

[0233] According to the vehicle usage frequency, dynamically correct the threshold, for example, if the daily average driving is 300km in the last 30 days, then the threshold is shortened to 10 days.

[0234] Step C2, when the remaining service life of the tire is lower than the remaining service life threshold, the candidate tires are ranked in descending order according to the adaptation score, and the candidate tire models that do not meet the environmental temperature and / or road humidity requirements are removed to generate the first alternative set.

[0235] The first alternative set is a list of candidate tires that meet the environmental adaptability requirements (such as temperature, humidity) and are ranked in descending order of adaptation score.

[0236] In a specific example,

[0237] Eliminate models whose applicable temperature range does not include the current annual average temperature ± 10°C (for example, a certain tire has an applicable temperature range of -20°C to 50°C, and the current annual average temperature is 35°C, so it is retained).

[0238] Eliminate models whose wet grip level is lower than the requirement corresponding to the current annual rainfall (for example, in an area with annual rainfall > 1000 mm, the requirement for wet grip level is ≥ B level).

[0239] Rank in descending order of adaptation score, and in case of the same score, prefer models with lower rolling resistance.

[0240] Step C3, weight match the brand preference data in the user's historical replacement record with the first alternative set to generate a recommended tire model set containing recommended models, performance comparison data, and replacement urgency.

[0241] User brand preference data is extracted based on historical replacement records, and the ranking priority of preferred brands is improved through weighting.

[0242] Preference weight calculation: Calculate the proportion of each brand in the user's past 5 replacement records, for example, brand X accounts for 80%, and its weight coefficient is 0.8.

[0243] Multiply the adaptation score of the candidate tire by (1 + brand weight), for example, brand X tire original score 0.85, adjusted to 0.85 x 1.8 = 1.53.

[0244] According to the difference between the remaining service life and the threshold value, the urgency is divided into:

[0245] Urgent (difference ≤ 3 days): red mark, recommended to replace immediately.

[0246] Suggestion (3 days < difference ≤ 7 days): yellow mark, recommended to replace this week.

[0247] Observation (difference > 7 days): green mark, recommended to replace next time.

[0248] Step C4, synchronize the recommended tire model set to the vehicle-mounted interactive interface of the new energy vehicle and the user's account in the cloud.

[0249] Data synchronization ensures that users get consistent recommended information on multiple terminals, supporting online and offline replacement service connection.

[0250] In a specific example,

[0251] Display the top three recommended models' fit score, price range, and urgency level. Click on the model to view the performance comparison radar chart (rolling resistance, grip, wear resistance).

[0252] The recommendation list is associated with nearby cooperative service network inventory information, displaying available models and reservation status in real time. The user's account records the recommendation history for subsequent service tracking.

[0253] In a specific example, user C's vehicle tire has a remaining service life of 8 days (threshold 15 days):

[0254] From the top 50 candidate tires ranked by fit score, eliminate 3 models that are not suitable for high temperature and high humidity environments in the south (maximum applicable temperature 45℃ < local summer extreme temperature 48℃).

[0255] The remaining 47 models are ranked in descending order of score, with the top five scores being 0.92, 0.89, 0.87, 0.85, and 0.83.

[0256] Brand Y accounts for 60% of user C's historical replacement records, and brand Z accounts for 40%.

[0257] Brand Y tire (original score 0.89) has an adjusted score of 0.89 x 1.6 = 1.424, rising to the first place.

[0258] Brand Z tire (original score 0.87) has an adjusted score of 0.87 x 1.4 = 1.218, remaining in the third place.

[0259] The urgency level is marked as "recommended" (difference 7 days), and the top three recommended models are brand Y-A (1.424 points), brand X-B (0.92 points), and brand Z-C (1.218 points).

[0260] The performance comparison data marks brand Y-A's wet grip as AA level and rolling resistance as B level.

[0261] The vehicle-mounted central control screen displays the recommendation list, with brand Y-A at the top and marked as "frequently selected brand preference".

[0262] The user's mobile phone app synchronously pushes: "recommend replacing brand Y-A, there is stock at nearby D service network".

[0263] The present application also includes a new energy vehicle special driving data recording tire, which comprises:

[0264] A flexible monitoring belt is embedded between the tire body structure layers of the tire.

[0265] The flexible monitoring belt comprises a flexible circuit board base, a temperature sensor, a pressure sensor, a three-axis acceleration sensor and a wireless communication unit encapsulated on the flexible circuit board base.

[0266] The temperature sensor is used for monitoring the internal temperature of the tire, the pressure sensor is used for measuring the air pressure in the tire, and the three-axis acceleration sensor is used for detecting the longitudinal, lateral and vertical accelerations of the tire.

[0267] The wireless communication unit comprises a dual-mode transmission module, a main channel supporting a short-distance wireless protocol and a standby channel supporting a long-distance wireless protocol, and is used for transmitting the data of the temperature sensor, the pressure sensor and the three-axis acceleration sensor to a vehicle control unit and a cloud.

[0268] The three-axis acceleration sensor is connected with the temperature sensor, and when the longitudinal acceleration exceeds a preset vehicle speed threshold, the sampling frequency of the temperature sensor is increased.

Claims

1. A method for recording driving data specifically for new energy vehicles, applied to new energy vehicles, characterized in that, The method includes: A monitoring module is deployed to the tires of the new energy vehicle to collect tire driving data; wherein, the monitoring module includes a temperature sensor, a pressure sensor, a triaxial acceleration sensor and a wireless communication unit, and the tire driving data includes temperature, pressure and triaxial acceleration; The tire driving data is transmitted in real time to the vehicle control unit and the preset cloud based on a preset wireless transmission protocol; wherein the vehicle control unit includes a preset tire condition analysis model. Based on the new energy vehicle, vehicle driving parameters, environmental perception data and historical maintenance records are obtained, and the tire driving data, vehicle driving parameters, environmental perception data and historical maintenance records are processed based on the tire condition analysis model to generate tire health and risk warnings. When the risk warning exceeds a preset threshold, the new energy vehicle is processed based on a preset vehicle control strategy; Based on the tire health status and the preset tire material degradation curve, the remaining service life of the tire is estimated. When the triaxial accelerometer detects that the acceleration exceeds the preset vehicle speed threshold, the temperature sampling frequency of the temperature sensor is increased, and the pressure sensor is drift compensated based on the reference pressure value of the tire in a stationary state. Define a multi-objective optimization function; wherein the objectives of the multi-objective optimization function include minimizing rolling resistance, maximizing grip on wet surfaces, and equalizing wear distribution; Obtain the driving habits of the new energy vehicles; The weight coefficients of the multi-objective optimization function are adjusted based on the driving habit data.

2. The method for recording driving data specifically for new energy vehicles according to claim 1, characterized in that, Deploying a monitoring module to the tires of the new energy vehicle to collect tire driving data specifically includes: The temperature sensor, pressure sensor, and triaxial accelerometer are packaged to form the first monitoring module; The first monitoring module is connected to the wireless communication unit to generate the monitoring module; The monitoring module is deployed to a preset position inside the tire crown to generate a monitoring strip that matches the tire curvature, and is embedded between the tire body structure layers during the tire forming stage; The tire driving data is generated by integrating acceleration, temperature, and pressure in a time sequence.

3. The method for recording driving data specifically for new energy vehicles according to claim 1, characterized in that, Based on the new energy vehicle, vehicle driving parameters, environmental perception data, and historical maintenance records are acquired, and the tire driving data, vehicle driving parameters, environmental perception data, and historical maintenance records are processed based on the tire condition analysis model to generate tire health and risk warnings, specifically including: Vehicle driving parameters are obtained from the CAN bus of the new energy vehicle; wherein, the vehicle driving parameters include at least real-time vehicle speed, motor output torque and battery load status; Environmental perception data is acquired through preset vehicle-mounted environmental sensors; wherein, the environmental perception data includes at least road surface humidity, outside temperature, and air density; Historical maintenance records are retrieved from the cloud-based database; these records include tire replacement cycles, tire pressure calibration records, and wear repair logs. The tire driving data, vehicle driving parameters, and environmental perception data are preprocessed to generate standardized input vectors; The standardized input vector is input into the tire condition analysis model to calculate the tire health score; Risk warnings are determined based on environmental perception data, vehicle driving parameters, and tire health scores.

4. The method for recording driving data specifically for new energy vehicles according to claim 3, characterized in that, The construction of the tire condition analysis model specifically includes: A hybrid model architecture comprising an LSTM network and a fully connected layer is constructed; wherein the LSTM network is used to process time-series tire driving data, and the fully connected layer is used to fuse vehicle driving parameters and environmental perception data; Determine the training data; wherein, the training data includes historical tire failure samples, normal wear samples, and manually labeled health labels; The hybrid model architecture is trained using a pre-defined supervised learning method and the training data. The mean squared error loss between the predicted health score and the actual health score label is minimized based on the preset Adam optimizer. When the mean square error loss is less than a preset success threshold, the tire condition analysis model is output.

5. The method for recording driving data specifically for new energy vehicles according to claim 1, characterized in that, When the risk warning exceeds a preset threshold, the new energy vehicle is processed based on a preset vehicle control strategy, specifically including: The suspension damping coefficient adjustment is calculated based on the pressure and temperature, and the tire support stiffness is adjusted through the electronically controlled suspension system of the new energy vehicle. Based on risk warning and communication with the motor control module of the new energy vehicle, the motor output power is reduced to a safe threshold range according to a preset ratio; The system generates visual warning icons on the in-vehicle interactive interface of the new energy vehicle and broadcasts real-time risk warning information.

6. The method for recording driving data specifically for new energy vehicles according to claim 1, characterized in that, The remaining service life of the tire is calculated based on the tire health status and a preset tire material degradation curve, specifically including: The material degradation curve corresponding to the rubber formulation of the tire is matched based on a preset material database; wherein the material degradation curve is used to describe the relationship between the elastic modulus and temperature and time. The tire health status and material degradation curve are superimposed and analyzed to calculate the theoretical remaining wear thickness of the tire. The remaining wear thickness is processed based on the average vehicle speed and load data in the vehicle's driving parameters to determine the remaining service life.

7. The method for recording driving data specifically for new energy vehicles according to claim 1, characterized in that, The method further includes: Set constraints; wherein the constraints include tire size matching range, maximum load capacity and speed rating limit; The preset genetic algorithm searches the tire recommendation database for candidate tire models that meet the constraints. Each candidate tire model is processed based on the multi-objective optimization function to calculate the fit score corresponding to each candidate tire model.

8. The method for recording driving data for new energy vehicles according to claim 7, characterized in that, After processing each candidate tire model based on the multi-objective optimization function to calculate the fit score corresponding to each candidate tire model, the method further includes: Set the remaining service life threshold; When the remaining service life of the tire is lower than the remaining service life threshold, the candidate tires are sorted in descending order according to the fit score, and the candidate tire models that do not meet the requirements of ambient temperature and / or road surface humidity are eliminated to generate a first candidate set. The brand preference data in the user's historical replacement records is weighted and matched with the first candidate set to generate a set of recommended tire models that includes recommended models, performance comparison data and replacement urgency. The recommended tire models are synchronized to the in-vehicle interface of the new energy vehicle and the user account in the cloud.

9. A tire for recording driving data specifically for new energy vehicles, applied to the driving data recording method for new energy vehicles as described in any one of claims 1-8, characterized in that, The tire includes: A flexible monitoring strip is embedded between the tire's carcass structural layers; The flexible monitoring strip includes a flexible circuit board substrate, and a temperature sensor, a pressure sensor, a triaxial accelerometer, and a wireless communication unit encapsulated on the flexible circuit board substrate. The temperature sensor is used to monitor the internal temperature of the tire, the pressure sensor is used to measure the internal air pressure of the tire, and the triaxial accelerometer is used to detect the longitudinal, lateral, and vertical acceleration of the tire. The wireless communication unit includes a dual-mode transmission module that supports a main channel for short-range wireless protocols and a backup channel for long-range wireless protocols, used to transmit data from temperature sensors, pressure sensors, and triaxial accelerometers to the vehicle control unit and the cloud.

10. A tire for recording driving data for new energy vehicles according to claim 9, characterized in that: The triaxial accelerometer is connected to the temperature sensor. When the longitudinal acceleration is detected to exceed the preset vehicle speed threshold, the sampling frequency of the temperature sensor is increased.

Images