A wheel intelligent monitoring method and system with intelligent early warning

Abstract

The present application relates to a wheel intelligent monitoring method and system with intelligent early warning, and relates to the technical field of vehicles. The method comprises: obtaining vibration data of a brake disc during braking by means of a non-contact laser displacement sensor; obtaining pressure data of a brake pedal during braking by means of a pressure sensor; obtaining driving data of the vehicle by means of an ECU of the vehicle; inputting the vibration data, pressure data and driving data into a brake pad life prediction model; performing a secondary correction on the remaining predicted mileage of the brake pad fed back by the brake pad life prediction model and combining the expert knowledge rules to obtain the remaining mileage; when the remaining mileage is less than a preset value, sending an early warning message to the ECU of the vehicle for displaying the early warning message.

Description

Technical Field

[0001] The present application relates to the technical field of vehicles, and in particular to a method and system for intelligent wheel monitoring with intelligent early warning. Background Art

[0002] Existing vehicle health monitoring systems usually rely on threshold detection of sensor data. This method is prone to false positives or false negatives under complex working conditions and it is difficult to provide intelligent warnings based on the vehicle's status.

[0003] How to solve this technical problem is a technical difficulty that technicians in this field need to overcome. Summary of the invention

[0004] In order to at least partially solve the above technical problems, the present application provides a wheel intelligent monitoring method and system with intelligent early warning.

[0005] In a first aspect, the present application provides an intelligent wheel monitoring method with intelligent early warning, which adopts the following technical solution.

[0006] A wheel intelligent monitoring method with intelligent early warning, characterized by comprising:

[0007] The vibration data of the brake disc during braking is obtained through a non-contact laser displacement sensor;

[0008] The pressure data of the brake pedal during braking is obtained through the pressure sensor;

[0009] Obtain the vehicle's driving data through the vehicle's ECU;

[0010] The vibration data, pressure data and driving data are input into a brake pad life prediction model; the brake pad life prediction model feeds back the remaining predicted mileage of the brake pad and combines the expert knowledge rule to perform a secondary correction on the remaining predicted mileage to obtain the remaining mileage;

[0011] When the remaining mileage is less than a preset value, a warning message is sent to the vehicle's ECU for displaying the warning message;

[0012] The training steps of the brake pad life prediction model include:

[0013] Construct a multi-layer LSTM network, where each layer contains a certain number of memory units; the input layer is used to receive pre-processed vibration data, pressure data, and driving data; the hidden layer is responsible for long-term dependencies and filters irrelevant information; the output layer is used to output the remaining predicted mileage of the brake pad;

[0014] In the feature extraction stage, the vibration data, pressure data and driving data are combined into a high-dimensional vector;

[0015] The initial brake pad life prediction model is trained and adjusted based on the confidence of the model until a brake pad life prediction model that meets the confidence requirement is obtained.

[0016] Optionally, in the feature extraction stage, the vibration data, pressure data and driving data are combined into a high-dimensional vector, including:

[0017] Extract frequency spectrum, amplitude and peak-to-peak value based on vibration data;

[0018] Extract the maximum pressure, average pressure and pressure change rate based on pressure data;

[0019] Extract average speed, number of emergency brakes and acceleration change rate based on driving data;

[0020] Normalizing each feature value to place each feature value within a preset range;

[0021] All standardized feature vectors are horizontally concatenated to form a high-dimensional feature vector.

[0022] Optionally, the configuration method of the expert knowledge rule includes:

[0023] Based on industry standards and technical reports, data on material characteristics and influencing factors of brake pads are obtained, which are recorded as first data;

[0024] Obtaining wear rate data of brake pads under different driving styles and road conditions, recorded as second data;

[0025] Obtain data on the effects of temperature, humidity and altitude on the life of brake pads, recorded as third data;

[0026] The expert knowledge rules are obtained by combining the first data, the second data and the third data.

[0027] Optionally, the method further includes:

[0028] Obtain the driver's acceleration data, deceleration data, steering data, braking force data, and driving data through the vehicle's ECU;

[0029] Calculate the average acceleration, maximum acceleration and acceleration frequency during the acceleration process to obtain the acceleration characteristics;

[0030] The average deceleration, maximum deceleration and deceleration frequency during the deceleration process are calculated to obtain the deceleration characteristics;

[0031] Based on the steering angle data, the sharp turning frequency is identified to obtain the steering characteristics;

[0032] Based on the braking force data, the emergency braking frequency is identified to obtain the braking characteristics;

[0033] Generate a driving behavior model through training of the acceleration characteristics, deceleration characteristics, steering characteristics and braking characteristics; the driving behavior model is used to identify and classify different driving styles;

[0034] Acquiring driving behavior data of a vehicle driver during vehicle driving; inputting the driving behavior data into a driving behavior model to obtain driving style characteristics;

[0035] The driving style feature is combined with the brake pad prediction model as one of the input features of the brake pad prediction model to obtain the remaining predicted mileage.

[0036] Optionally, the method further includes:

[0037] Obtain driving route data through map software;

[0038] Obtaining road condition data in the driving road bureau according to the driving route data;

[0039] The road condition data is combined with the brake pad prediction model as one of the input features of the brake pad prediction model to obtain the remaining predicted mileage.

[0040] Optionally, send warning information to the vehicle's ECU for warning information display, including:

[0041] Generate a prompt icon;

[0042] The prompt icon is displayed on an interactive screen of the vehicle.

[0043] In the second aspect, the present application provides an intelligent wheel monitoring system with intelligent warning, which adopts the following technical solution.

[0044] An intelligent wheel monitoring system with intelligent early warning, comprising:

[0045] The first processing module is used to obtain the vibration data of the brake disc during the braking process through a non-contact laser displacement sensor;

[0046] The second processing module is used to: obtain the pressure data of the brake pedal during braking through the pressure sensor;

[0047] The third processing module is used to: obtain the driving data of the vehicle through the ECU of the vehicle;

[0048] The fourth processing module is used to: input the vibration data, pressure data and driving data into the brake pad life prediction model; the brake pad life prediction model feeds back the remaining predicted mileage of the brake pad and combines the expert knowledge rule to perform a secondary correction on the remaining predicted mileage to obtain the remaining mileage;

[0049] The fifth processing module is used to: when the remaining mileage is less than a preset value, send a warning message to the ECU of the vehicle for displaying the warning message;

[0050] The training steps of the brake pad life prediction model include:

[0051] Construct a multi-layer LSTM network, where each layer contains a certain number of memory units; the input layer is used to receive pre-processed vibration data, pressure data, and driving data; the hidden layer is responsible for long-term dependencies and filters irrelevant information; the output layer is used to output the remaining predicted mileage of the brake pad;

[0052] In the feature extraction stage, the vibration data, pressure data and driving data are combined into a high-dimensional vector;

[0053] The initial brake pad life prediction model is trained and adjusted based on the confidence of the model until a brake pad life prediction model that meets the confidence requirement is obtained.

[0054] In a third aspect, the present application discloses an electronic device, comprising a memory and a processor, wherein the memory stores a computer program that is loaded by the processor and executes any of the above methods.

[0055] In a fourth aspect, the present application discloses a computer-readable storage medium storing a computer program that can be loaded by a processor and execute any of the above methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 This is a flow chart of a wheel intelligent monitoring method with intelligent warning according to an embodiment of the present application;

[0057] Figure 2 This is a system block diagram of a wheel intelligent monitoring method with intelligent early warning according to an embodiment of the present application;

[0058] In the figure, 201 is a first processing module; 202 is a second processing module; 203 is a third processing module; 204 is a fourth processing module; 205 is a fifth processing module. DETAILED DESCRIPTION

[0059] The following is combined with Figure 1-2 The present application is further described with specific embodiments:

[0060] First of all, it should be noted that in the description of this application, if the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inside", "outside" and other directional words appear, the orientation or position relationship indicated is based on the orientation or position relationship shown in the drawings, which is only for the convenience of description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of this application; in addition, if the terms "first", "second", "third" and other numerical quantifiers appear, they are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in this application, unless otherwise clearly specified and limited, if the terms "installed", "connected", and "connected" appear, they should be understood in a broad sense, for example, it can be a fixed connection, or a detachable connection, a limited connection such as an interference fit, a transition fit, or an integral connection; it can be directly connected or indirectly connected through an intermediate medium; therefore, for ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to the specific circumstances.

[0061] The embodiment of the present application discloses a wheel intelligent monitoring method with intelligent early warning. Figure 1 As an implementation method of a wheel intelligent monitoring method with intelligent early warning, a wheel intelligent monitoring method with intelligent early warning includes the following steps:

[0062] Step 101: Acquire vibration data of the brake disc during braking by using a non-contact laser displacement sensor.

[0063] Step 102: Obtain pressure data of the brake pedal during braking through a pressure sensor.

[0064] Step 103: Acquire the vehicle's driving data through the vehicle's ECU.

[0065] Step 104: input the vibration data, pressure data and driving data into a brake pad life prediction model; the brake pad life prediction model feeds back the remaining predicted mileage of the brake pad and performs a secondary correction on the remaining predicted mileage in combination with expert knowledge rules to obtain the remaining mileage.

[0066] Step 105: When the remaining mileage is less than a preset value, a warning message is sent to the vehicle's ECU for displaying the warning message.

[0067] The training steps of the brake pad life prediction model include:

[0068] Construct a multi-layer LSTM network, where each layer contains a certain number of memory units; the input layer is used to receive pre-processed vibration data, pressure data, and driving data; the hidden layer is responsible for long-term dependencies and filters irrelevant information; the output layer is used to output the remaining predicted mileage of the brake pad;

[0069] In the feature extraction stage, the vibration data, pressure data and driving data are combined into a high-dimensional vector;

[0070] The initial brake pad life prediction model is trained and adjusted based on the confidence of the model until a brake pad life prediction model that meets the confidence requirement is obtained.

[0071] Specifically, the non-contact laser displacement sensor can capture the tiny vibrations of the brake disc with high accuracy;

[0072] The pressure sensor records the pressure change of the brake pedal, reflects the intensity of the driver's braking action, and provides key parameters for the evaluation of brake pad wear. The ECU (electronic control unit) simultaneously collects vehicle driving data such as speed, acceleration, mileage, etc. The frequency spectrum, amplitude, and peak-to-peak value are extracted from the vibration data; the maximum pressure, average pressure, and pressure change rate are extracted from the pressure data; and the average speed, number of emergency braking, and acceleration change rate are extracted from the driving data. All features are standardized and horizontally spliced ​​to form a high-dimensional feature vector to provide input for the deep learning model. The design of the LSTM (long short-term memory) network enables the model to handle long-term dependencies in time series data, and can effectively avoid the gradient vanishing problem even when the data sequence is very long, which is crucial for predicting wear processes such as brake pad life that gradually accumulate over time. The introduction of expert knowledge rules, such as the characteristics of brake pad materials, the influence of driving style and road conditions, and the role of environmental factors, can make up for the shortcomings of data-driven models and make the prediction results more in line with actual conditions. The secondary correction ensures that the prediction results are not only based on data statistics, but also take into account industry experience and domain knowledge, improving the stability and robustness of the prediction. When the predicted remaining life of the brake pad is lower than the preset value, the system automatically warns the driver to check or replace it, thereby avoiding accidents and improving driving safety.

[0073] As a specific implementation of a wheel intelligent monitoring method for intelligent early warning, in the feature extraction stage, the vibration data, pressure data and driving data are combined into a high-dimensional vector, including:

[0074] Extract frequency spectrum, amplitude and peak-to-peak value based on vibration data;

[0075] Extract the maximum pressure, average pressure and pressure change rate based on pressure data;

[0076] Extract average speed, number of emergency brakes and acceleration change rate based on driving data;

[0077] Normalizing each feature value to place each feature value within a preset range;

[0078] All standardized feature vectors are horizontally concatenated to form a high-dimensional feature vector.

[0079] Specifically, by extracting the frequency spectrum, amplitude and peak-to-peak value of vibration data, the maximum pressure, average pressure and pressure change rate of pressure data, and the average speed, number of emergency braking times and acceleration change rate of driving data, the use condition and wear degree of the brake pads can be more comprehensively reflected. Standardizing each eigenvalue to a preset range eliminates the dimensional influence between features and ensures that different features are compared on the same scale, which is conducive to the model learning the true relationship between features rather than being misled by the numerical value, thereby improving the model's interpretation and generalization capabilities. The standardized feature vectors are horizontally spliced ​​to form a high-dimensional feature vector. This operation integrates all relevant braking information into one vector, providing rich input information for the deep learning model and helping the model capture complex and subtle brake pad wear patterns, especially recurrent neural networks such as LSTM, which can effectively process time series data and long-term dependencies. LSTM networks can handle the long-term dependencies of time series data and are suitable for processing processes such as brake pad wear that gradually accumulate over time. Through the input of high-dimensional feature vectors, the model can learn the combined influence of various factors in the braking process, thereby making more accurate predictions of brake pad life.

[0080] Through feature extraction, standardization processing, construction of high-dimensional feature vectors and training of deep learning models, combined with the correction of expert knowledge rules, the entire intelligent warning wheel intelligent monitoring method can achieve accurate prediction of the brake pad wear status, timely warning, and improve the driver's driving experience.

[0081] As a specific implementation method of a wheel intelligent monitoring method with intelligent early warning, the configuration method of expert knowledge rules includes:

[0082] Based on industry standards and technical reports, data on material characteristics and influencing factors of brake pads are obtained, which are recorded as first data;

[0083] Obtaining wear rate data of brake pads under different driving styles and road conditions, recorded as second data;

[0084] Obtain data on the effects of temperature, humidity and altitude on the life of brake pads, recorded as third data;

[0085] The expert knowledge rules are obtained by combining the first data, the second data and the third data.

[0086] Specifically, by embedding these expert knowledge rules into the brake pad life prediction model, the output of the model can be calibrated and corrected, improving the robustness and decision support capabilities of the model.

[0087] As a specific implementation of a wheel intelligent monitoring method with intelligent early warning, the method further includes:

[0088] Obtain the driver's acceleration data, deceleration data, steering data, braking force data, and driving data through the vehicle's ECU;

[0089] Calculate the average acceleration, maximum acceleration and acceleration frequency during the acceleration process to obtain the acceleration characteristics;

[0090] The average deceleration, maximum deceleration and deceleration frequency during the deceleration process are calculated to obtain the deceleration characteristics;

[0091] Based on the steering angle data, the sharp turning frequency is identified to obtain the steering characteristics;

[0092] Based on the braking force data, the emergency braking frequency is identified to obtain the braking characteristics;

[0093] Generate a driving behavior model through training of the acceleration characteristics, deceleration characteristics, steering characteristics and braking characteristics; the driving behavior model is used to identify and classify different driving styles;

[0094] Acquiring driving behavior data of a vehicle driver during vehicle driving; inputting the driving behavior data into a driving behavior model to obtain driving style characteristics;

[0095] The driving style feature is combined with the brake pad prediction model as one of the input features of the brake pad prediction model to obtain the remaining predicted mileage.

[0096] Specifically, the vehicle's ECU (electronic control unit) collects the driver's acceleration, deceleration, steering, braking force and driving data in real time. Based on these data, the acceleration characteristics (average acceleration, maximum acceleration, acceleration frequency), deceleration characteristics (average deceleration, maximum deceleration, deceleration frequency), steering characteristics (sharp turning frequency) and braking characteristics (sudden braking frequency) are calculated. The driving behavior model is generated by training using the above driving characteristics, and the model can effectively identify and classify different driving styles, such as aggressive, stable, and economical. During the vehicle's driving process, the driver's driving behavior data is obtained in real time and input into the driving behavior model, which can provide instant feedback on driving style characteristics. The driving style characteristics are combined with the brake pad prediction model as one of the important input features of the prediction model. The brake pad prediction model dynamically adjusts the prediction results based on the driver's individual operating habits, realizes personalized prediction of the remaining predicted mileage of the brake pad, and enhances the accuracy and reliability of the prediction.

[0097] As one implementation of a wheel intelligent monitoring method with intelligent early warning, the method further includes:

[0098] Obtain driving route data through map software;

[0099] Obtaining road condition data in the driving road bureau according to the driving route data;

[0100] The road condition data is combined with the brake pad prediction model as one of the input features of the brake pad prediction model to obtain the remaining predicted mileage.

[0101] As one implementation of a wheel intelligent monitoring method for intelligent early warning, sending early warning information to the ECU of a vehicle for displaying the early warning information includes:

[0102] Generate a prompt icon;

[0103] The prompt icon is displayed on an interactive screen of the vehicle.

[0104] The present application also provides a wheel intelligent monitoring system with intelligent early warning, comprising:

[0105] The first processing module 201 is used to obtain the vibration data of the brake disc during the braking process through a non-contact laser displacement sensor;

[0106] The second processing module 202 is used to: obtain the pressure data of the brake pedal during braking through the pressure sensor;

[0107] The third processing module 203 is used to: obtain the driving data of the vehicle through the ECU of the vehicle;

[0108] The fourth processing module 204 is used to: input the vibration data, pressure data and driving data into the brake pad life prediction model; perform secondary correction on the remaining predicted mileage of the brake pad fed back by the brake pad life prediction model and in combination with expert knowledge rules to obtain the remaining mileage;

[0109] The fifth processing module 205 is used to: when the remaining mileage is less than a preset value, send a warning message to the ECU of the vehicle for displaying the warning message;

[0110] The training steps of the brake pad life prediction model include:

[0111] Construct a multi-layer LSTM network, where each layer contains a certain number of memory units; the input layer is used to receive pre-processed vibration data, pressure data, and driving data; the hidden layer is responsible for long-term dependencies and filters irrelevant information; the output layer is used to output the remaining predicted mileage of the brake pad;

[0112] In the feature extraction stage, the vibration data, pressure data and driving data are combined into a high-dimensional vector;

[0113] The initial brake pad life prediction model is trained and adjusted based on the confidence of the model until a brake pad life prediction model that meets the confidence requirement is obtained.

[0114] The embodiment of the present application also discloses an electronic device.

[0115] Specifically, the device includes a memory and a processor, and the memory stores a computer program that can be loaded by the processor and execute any one of the above-mentioned intelligent wheel monitoring methods for intelligent warning.

[0116] The embodiment of the present application also discloses a computer-readable storage medium. Specifically, the computer-readable storage medium stores a computer program that can be loaded by a processor and executes any of the above-mentioned intelligent warning wheel intelligent monitoring methods, and the computer-readable storage medium includes, for example: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0117] It should be noted that the above embodiments are only used to illustrate the present application and are not intended to limit the technical solutions described in the present application. Although the present application has been described in detail in this specification with reference to the above embodiments, a person of ordinary skill in the art should understand that a person of ordinary skill in the art can still modify or make equivalent substitutions to the present application, and all technical solutions and improvements thereof that do not depart from the spirit and scope of the present application should be included in the scope of the claims of the present application.

Claims

1. A wheel intelligent monitoring method with intelligent early warning, characterized in that: include: The vibration data of the brake disc during braking is obtained through a non-contact laser displacement sensor; The pressure data of the brake pedal during braking is obtained through the pressure sensor; Obtain the vehicle's driving data through the vehicle's ECU; The vibration data, pressure data and driving data are input into a brake pad life prediction model; the brake pad life prediction model feeds back the remaining predicted mileage of the brake pad and performs secondary correction on the remaining predicted mileage in combination with expert knowledge rules to obtain the remaining mileage; When the remaining mileage is less than a preset value, a warning message is sent to the vehicle's ECU for displaying the warning message; The training steps of the brake pad life prediction model include: Construct a multi-layer LSTM network, where each layer contains a certain number of memory units; the input layer is used to receive pre-processed vibration data, pressure data, and driving data; the hidden layer is responsible for long-term dependencies and filters irrelevant information; the output layer is used to output the remaining predicted mileage of the brake pad; In the feature extraction stage, the vibration data, pressure data and driving data are combined into a high-dimensional vector; Training an initial brake pad life prediction model and adjusting the initial brake pad life prediction model based on the confidence of the model until a brake pad life prediction model that meets the confidence requirement is obtained; The configuration methods of expert knowledge rules include: Based on industry standards and technical reports, data on material characteristics and influencing factors of brake pads are obtained, which are recorded as first data; Obtaining wear rate data of brake pads under different driving styles and road conditions, recorded as second data; Obtain data on the effects of temperature, humidity and altitude on the life of brake pads, recorded as third data; The expert knowledge rules are obtained by combining the first data, the second data and the third data.

2. The wheel intelligent monitoring method with intelligent early warning according to claim 1 is characterized in that: In the feature extraction stage, the vibration data, pressure data and driving data are combined into a high-dimensional vector, including: Extract frequency spectrum, amplitude and peak-to-peak value based on vibration data; Extract the maximum pressure, average pressure and pressure change rate based on pressure data; Extract average speed, number of emergency brakes and acceleration change rate based on driving data; Normalizing each feature value to place each feature value within a preset range; All standardized feature vectors are horizontally concatenated to form a high-dimensional feature vector.

3. The wheel intelligent monitoring method with intelligent early warning according to claim 2 is characterized in that: The method further comprises: Obtain the driver's acceleration data, deceleration data, steering data, braking force data, and driving data through the vehicle's ECU; Calculate the average acceleration, maximum acceleration and acceleration frequency during the acceleration process to obtain the acceleration characteristics; The average deceleration, maximum deceleration and deceleration frequency during the deceleration process are calculated to obtain the deceleration characteristics; Based on the steering angle data, the sharp turning frequency is identified to obtain the steering characteristics; Based on the braking force data, the emergency braking frequency is identified to obtain the braking characteristics; Generate a driving behavior model through training of the acceleration characteristics, deceleration characteristics, steering characteristics and braking characteristics; the driving behavior model is used to identify and classify different driving styles; Acquiring driving behavior data of a vehicle driver during vehicle driving; inputting the driving behavior data into a driving behavior model to obtain driving style characteristics; The driving style feature is combined with the brake pad life prediction model as one of the input features of the brake pad life prediction model to obtain the remaining predicted mileage.

4. The wheel intelligent monitoring method with intelligent early warning according to claim 3 is characterized in that: The method further comprises: Obtain driving route data through map software; Obtaining road condition data in the driving road bureau according to the driving route data; The road condition data is combined with the brake pad life prediction model as one of the input features of the brake pad life prediction model to obtain the remaining predicted mileage.

5. The wheel intelligent monitoring method with intelligent early warning according to claim 4 is characterized in that: Send warning information to the vehicle's ECU for warning information display, including: Generate a prompt icon; The prompt icon is displayed on an interactive screen of the vehicle.

6. An intelligent wheel monitoring system with intelligent early warning, characterized in that: include: The first processing module is used to obtain the vibration data of the brake disc during the braking process through a non-contact laser displacement sensor; The second processing module is used to: obtain the pressure data of the brake pedal during braking through the pressure sensor; The third processing module is used to: obtain the driving data of the vehicle through the ECU of the vehicle; The fourth processing module is used to: input the vibration data, pressure data and driving data into the brake pad life prediction model; the brake pad life prediction model feeds back the remaining predicted mileage of the brake pad and combines the expert knowledge rule to perform a secondary correction on the remaining predicted mileage to obtain the remaining mileage; The fifth processing module is used to: when the remaining mileage is less than a preset value, send a warning message to the ECU of the vehicle to display the warning message; The training steps of the brake pad life prediction model include: Construct a multi-layer LSTM network, where each layer contains a certain number of memory units; the input layer is used to receive pre-processed vibration data, pressure data, and driving data; the hidden layer is responsible for long-term dependencies and filters irrelevant information; the output layer is used to output the remaining predicted mileage of the brake pad; In the feature extraction stage, the vibration data, pressure data and driving data are combined into a high-dimensional vector; The initial brake pad life prediction model is trained and adjusted based on the confidence of the model until a brake pad life prediction model that meets the confidence requirement is obtained.

7. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program according to any one of the methods of claims 1 to 5 which is loaded and executed by the processor.

8. A computer-readable storage medium, characterized in that: A computer program is stored which can be loaded by a processor and execute the method according to any one of claims 1 to 5.

Images