An AI prediction method for test temperature rise curve of floating caliper automobile brake disc
By employing AI prediction methods, utilizing ROMAI tools and neural network algorithms, the problems of long prediction time and insufficient accuracy in brake disc temperature rise prediction have been solved, achieving fast and accurate temperature rise curve prediction to guide brake disc design.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHENZHI(CHONGQING)BRAKE SYSTEM CO LTD
- Filing Date
- 2026-02-05
- Publication Date
- 2026-06-05
AI Technical Summary
Existing technologies for predicting temperature rise during the brake disc design phase are time-consuming and the results deviate from experimental results, making it difficult to quickly and accurately guide the design process.
An AI prediction method is adopted, utilizing Altair's ROMAI tool and neural network algorithm, and based on a large amount of brake disc test data and design parameters, to establish a temperature rise prediction model. The model parameters are optimized through training and testing to achieve rapid and accurate temperature rise curve prediction.
The brake disc temperature rise curve can be predicted in seconds, significantly shortening the development cycle, improving prediction accuracy, and reducing costs.
Smart Images
Figure CN122154407A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of automotive disc brake technology and relates to an AI prediction method for the test temperature rise curve of a floating caliper automotive brake disc. Background Technology
[0002] Automotive braking is essentially a process of converting the vehicle's kinetic energy (speed) into heat energy (heat) through friction. The vast majority of this heat (over 90%) is absorbed by the brake discs and pads, causing their temperature to rise rapidly. If this heat cannot dissipate in time, it can lead to a series of serious problems. Therefore, the brake disc temperature rise test curve is a direct verification of whether the brake disc's heat capacity meets the requirements. Currently, CAE simulation is generally used in the brake disc design stage to predict the temperature rise of the brake disc. This analysis typically employs a direct thermo-solid coupling method, while also considering the effects of air convection heat transfer, etc. The calculation time is relatively long (several days), and the calculation results often deviate somewhat from the experimental results. Summary of the Invention
[0003] In view of this, the purpose of this invention is to provide an AI prediction method for the temperature rise curve of a floating caliper type automotive brake disc during testing.
[0004] To achieve the above objectives, the present invention provides the following technical solution: An AI prediction method for the temperature rise curve of a floating caliper type automotive brake disc includes the following steps: S1: Data processing: Collect brake disc temperature rise test data and structural design parameters for different products, define time parameters, input variables and output variables, and establish a training data table; S2: Input the training data table into the AI prediction model to train the brake disc temperature rise curve prediction model; S3: Test the prediction model using the new brake disc structural characteristic parameter variables. If the error between the test results and the experimental data meets the conditions, the training is completed; otherwise, adjust the prediction model parameters and retrain. S4: During the brake disc design phase, input the brake disc structural parameters into the prediction model to generate a temperature rise curve, which guides the brake disc design.
[0005] Furthermore, in step S1, brake disc temperature rise test data of different products are collected, and time variables are defined in combination with test conditions. The product structural parameters related to brake disc temperature rise are sorted out, including brake disc inner diameter, outer diameter, brake disc thickness, heat dissipation fin length, heat dissipation fin width, density, specific heat capacity, thermal conductivity, test inertia and effective radius. Input variables are established, and the disc temperature value that changes with time is used as the output variable to establish a training data table.
[0006] Furthermore, the training data table is arranged in a certain order, with the time variable as the first column, the temperature rise data as the second column, and other variables not restricted in order.
[0007] Furthermore, in step S2, Altair's ROMAI tool is used as the basic AI model. ROMAI uses neural network algorithms as its theoretical basis. Biological neurons consist of a cell body, dendrites, and axons. Dendrites are one or more protrusions that radiate from the cell body, resembling tree branches. Each neuron has only one axon. Neurons are connected to other neurons through dendrites and axons to form a biological neural network. For a given neuron j, it may simultaneously receive many input signals Xi. Because biological neurons have different synaptic properties and synaptic strengths, the effects on the neuron vary. This is represented by weights W. ji This indicates that its positive and negative values simulate the excitation and inhibition of synapses in biological neurons, and its magnitude represents different connection strengths at the synapse, θ. j This represents a threshold; The summation and integration of all input signals is equivalent to the membrane potential in biological neurons. Whether a neuron is activated depends on a certain threshold potential. That is, a neuron is only activated and fires a pulse when the total input exceeds the threshold; otherwise, it will not output a signal. Output Indicates net activation The neural network algorithm constructs several simple functions, stacks them according to certain weights, and then fits the complex function.
[0008] Furthermore, the model training settings in the ROMAI tool include: Define input variables, output variables, and state variables. Define input variables D1-D9. Define the brake disc temperature rise data temp(output) as the output variable. The state variable is a time-related quantity. Here, the brake disc temperature rise data is defined as the state variable. The model type is set to non-linear, and the ReLU piecewise linear function is used as the activation function of the neural network. The training parameters are explored using Auto Exploration, allowing ROMAI to automatically batch-process and perform model training under different combinations of training parameters. The post-processing focuses on comparing the impact of training parameters on model accuracy. The number of hidden layers (HiddenLayers) and the number of neurons per layer in the Neurons X Layer are set, and the number of training iterations (Epochs) is used before submission. After the calculation is completed, the training results are displayed under different combinations of hidden layers and number of neurons. The smaller the AVG RMSE value, the higher the relative accuracy of the model. The training result with the smallest AVG RMSE value is selected for the next test.
[0009] Further, in step S3, testing the prediction model includes: inputting new brake disc structural feature parameter variables into the prediction model, generating a set of predicted temperature rise data, comparing the data with the experimental data, and if the relative error is within 5%, the requirement is met; otherwise, return to step S1 and adjust the prediction data and model training algorithm until the error between the prediction result and the experimental result is within 5%.
[0010] The beneficial effects of this invention are as follows: This invention uses AI prediction, based on a large amount of experimental data and brake disc design parameters. Once the prediction model is successfully established, the temperature rise curve can be predicted within seconds by inputting the structural parameters of the brake disc. This results in high consistency with experimental data and greatly shortens the development cycle of the brake disc, saving costs.
[0011] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description
[0012] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein: Figure 1 Flowchart of AI prediction method for temperature rise curve of floating caliper type automotive brake disc; Figure 2 A diagram illustrating data organization; Figure 3 This is a schematic diagram of the output of the ROMAI neural network algorithm; Figure 4 A schematic diagram illustrating the setup for training the ROMAI model; Figure 5 A diagram illustrating the parameter settings for neural network training; Figure 6 The AVG RMSE value is the training result. Figure 7 This is a schematic diagram for testing an AI prediction model. Detailed Implementation
[0013] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.
[0014] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.
[0015] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.
[0016] Example 1: like Figure 1 As shown, this invention provides an AI prediction method for the test temperature rise curve of automotive brake discs (floating caliper type), including the following steps: S1: Data Processing: Collect brake disc temperature rise test data (time-temperature curves) for different products. Define a time variable (time) based on the test conditions. Organize product structural parameters related to brake disc temperature rise, such as: brake disc inner diameter, outer diameter, brake disc thickness, heat dissipation fin length, heat dissipation fin width, density, specific heat capacity, thermal conductivity, test inertia, and effective radius. Establish input variables D1-D9, and the output variable is the disc temperature value (temp) changing over time. Create a training data table. To allow the AI software to better recognize the data, it needs to be arranged in a specific order: time variable in the first column, temperature rise data in the second column, and other variables in any order. To ensure sufficient training data, it is recommended to collect data from at least 15 products and create a data table. Example: Figure 2 As shown.
[0017] S2: AI Model Training: Submit the data table obtained in the previous step to Altair's ROMAI tool to train the temperature rise prediction model.
[0018] ROMAI primarily uses neural network algorithms, and its theoretical basis is as follows: biological neurons consist of a cell body, dendrites, and axons. Dendrites are one or more processes that radiate from the cell body, resembling tree branches. Each neuron has only one axon. Neurons connect with other neurons through dendrites and axons (the connection points are called synapses), forming a biological neural network.
[0019] For a given neuron j, it may simultaneously receive many input signals Xi. Due to the different synaptic properties and strengths of biological neurons, the effects on the neuron vary, represented by weights Wji. The positive and negative values of Wji simulate the excitation and inhibition of synapses in biological neurons, and the magnitude of Wji represents the different connection strengths of the synapses. θj represents a threshold (or "bias").
[0020] The process of accumulating and integrating all input signals is equivalent to the membrane potential in biological neurons. Whether a neuron is activated depends on a certain threshold potential; that is, the neuron is only activated and fires a pulse when the total input exceeds the threshold; otherwise, it does not output a signal. For example... Figure 3 As shown, the output Indicated as net activation The function can be simply understood as follows: a neural network algorithm constructs several simple functions, stacks them according to certain weights, and then achieves the fitting of a complex function.
[0021] Model training settings in ROMAI: like Figure 4 As shown, input variables, output variables, and state variables are defined. D1-D9 are defined as input variables, the brake disc temperature rise data temp(output) is defined as output variables, and the state variables are time-related quantities. Here, the brake disc temperature rise data is defined as state variables.
[0022] Neural network training parameter settings: such as Figure 5 As shown, the Model Type is set to non-linear, using the ReLU piecewise linear function as the neural network activation function. Auto Exploration of training parameters allows romAI to automatically batch-process model training under different combinations of training parameters (activation function, number of hidden layers, number of neurons), and then post-processing compares the impact of training parameters on model accuracy. After setting parameters such as the number of hidden layers (Hidden Layers), the number of neurons per hidden layer (Neurons X Layer), and the number of training iterations (Epochs), the computation is submitted.
[0023] like Figure 6As shown, after the calculation is completed, the training results AVG RMSE values will appear under different combinations of hidden layers and number of neurons. The smaller the AVG RMSE value, the higher the relative accuracy of the model. Select the 9th training result for the next test.
[0024] S3: AI Prediction Model Test: Input new brake disc structural feature parameters (same as step S2) into the prediction model. This will generate a set of predicted temperature rise data. Compare this data with the experimental data. If the relative error is within 5%, the requirement is met; otherwise, return to steps S1 and S2: adjust the prediction data and model training algorithm until the error between the prediction result and the experimental result is within 5%. Figure 7 As shown, test data from four products and the structural parameters of the brake disc were used for testing, and the prediction accuracy meets the usage requirements.
[0025] S4: Application of AI Prediction Model for Brake Disc Temperature Rise: During the brake disc design phase, the brake disc structural parameters are input into the prediction model to generate a temperature rise curve, which guides the brake disc design.
[0026] Example 2: An electronic device, comprising a memory and a processor; The memory is used to store computer programs; The processor is configured to implement the method described in Embodiment 1 when executing the computer program.
[0027] Example 3: A computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in Embodiment 1.
[0028] Example 4: A computer program product includes a computer program that, when executed by a processor, implements the method described in Example 1.
[0029] In the above embodiments, the reference to "this embodiment" in the specification indicates that a specific feature, structure, or characteristic described in connection with the embodiment is included in at least some embodiments, but not necessarily all embodiments. Multiple appearances of "this embodiment" do not necessarily refer to the same embodiment.
[0030] In the above embodiments, although the invention has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory structures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed. The embodiments of the invention are intended to cover all such substitutions, modifications, and variations falling within the broad scope of the appended claims.
[0031] As will be understood by those skilled in the art, the computer-readable storage medium described in this embodiment allows for the implementation of all or part of the steps in the above method embodiments by computer program-related hardware. The aforementioned computer program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.
[0032] The electronic terminal provided in this embodiment includes a processor, a memory, a transceiver, and a communication interface. The memory and the communication interface are connected to the processor and the transceiver and complete communication between them. The memory is used to store computer programs, the communication interface is used to perform communication, and the processor and the transceiver are used to run the computer programs, so that the electronic terminal performs the steps of the above method.
[0033] In this embodiment, the memory may include random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device.
[0034] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.
[0035] This invention can be used in a wide range of general-purpose or special-purpose computing system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices, etc.
[0036] This invention can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This invention can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.
[0037] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.
Claims
1. An AI prediction method for the temperature rise curve of a floating caliper type automotive brake disc during testing, characterized in that: Includes the following steps: S1: Data processing: Collect brake disc temperature rise test data and structural design parameters for different products, define time parameters, input variables and output variables, and establish a training data table; S2: Input the training data table into the AI prediction model to train the brake disc temperature rise curve prediction model; S3: Test the prediction model using the new brake disc structural characteristic parameter variables. If the error between the test results and the experimental data meets the conditions, the training is completed; otherwise, adjust the prediction model parameters and retrain. S4: During the brake disc design phase, input the brake disc structural parameters into the prediction model to generate a temperature rise curve, which guides the brake disc design.
2. The AI prediction method for the test temperature rise curve of a floating caliper type automotive brake disc according to claim 1, characterized in that: In step S1, brake disc temperature rise test data of different products are collected. Combined with the test conditions, time variables are defined, and the product structural parameters related to brake disc temperature rise are sorted out, including brake disc inner diameter, outer diameter, brake disc thickness, heat dissipation fin length, heat dissipation fin width, density, specific heat capacity, thermal conductivity, test inertia and effective radius. Input variables are established, and the disc temperature value that changes with time is used as the output variable to establish a training data table.
3. The AI prediction method for the test temperature rise curve of a floating caliper type automotive brake disc according to claim 1, characterized in that: The training data table is arranged in a certain order, with time variable as the first column, temperature rise data as the second column, and other variables not restricted in sorting.
4. The AI prediction method for the test temperature rise curve of a floating caliper type automotive brake disc according to claim 1, characterized in that: In step S2, Altair's ROMAI tool is used as the basic AI model. ROMAI uses neural network algorithms as its theoretical basis. Biological neurons consist of a cell body, dendrites, and axons. Dendrites are one or more protrusions that radiate from the cell body, resembling tree branches. Each neuron has only one axon. Neurons are connected to other neurons through dendrites and axons to form a biological neural network. For a given neuron j, it may simultaneously receive many input signals Xi. Because biological neurons have different synaptic properties and synaptic strengths, the effects on the neuron vary. This is represented by weights W. ji This indicates that its positive and negative values simulate the excitation and inhibition of synapses in biological neurons, and its magnitude represents different connection strengths at the synapse, θ. j This represents a threshold; The summation and integration of all input signals is equivalent to the membrane potential in biological neurons. Whether a neuron is activated depends on a certain threshold potential. That is, the neuron is activated and fires a pulse only when the total input exceeds the threshold; otherwise, it will not output a signal. Output Indicates net activation The neural network algorithm constructs several simple functions, stacks them according to certain weights, and then fits the complex function.
5. The AI prediction method for the test temperature rise curve of a floating caliper type automotive brake disc according to claim 1, characterized in that: The model training settings in the ROMAI tool include: Define input variables, output variables, and state variables. Define input variables D1-D9. Define the brake disc temperature rise data temp(output) as the output variable. The state variable is a time-related quantity. Here, the brake disc temperature rise data is defined as the state variable. The model type is set to non-linear, and the ReLU piecewise linear function is used as the activation function of the neural network. The training parameters are explored using Auto Exploration, allowing ROMAI to automatically batch-process and perform model training under different combinations of training parameters. The post-processing focuses on comparing the impact of training parameters on model accuracy. The number of hidden layers (HiddenLayers) and the number of neurons per layer in the Neurons X Layer are set, and the number of training iterations (Epochs) is used before submission. After the calculation is completed, the training results are displayed under different combinations of hidden layers and number of neurons. The smaller the AVG RMSE value, the higher the relative accuracy of the model. The training result with the smallest AVG RMSE value is selected for the next test.
6. The AI prediction method for the test temperature rise curve of a floating caliper type automotive brake disc according to claim 1, characterized in that: In step S3, testing the prediction model includes: inputting new brake disc structural feature parameter variables into the prediction model, generating a set of predicted temperature rise data, comparing the data with the experimental data, and if the relative error is within 5%, the requirement is met; otherwise, return to step S1 and adjust the prediction data and model training algorithm until the error between the prediction result and the experimental result is within 5%.
7. An electronic device, characterized in that, Including memory and processor; The memory is used to store computer programs; The processor is configured to, when executing the computer program, implement the AI prediction method for the test temperature rise curve of a floating caliper type automotive brake disc as described in any one of claims 1-6.
8. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the AI prediction method for the test temperature rise curve of a floating caliper type automotive brake disc as described in any one of claims 1-6.
9. A computer program product, characterized in that: The invention includes a computer program that, when executed by a processor, implements the AI prediction method for the test temperature rise curve of a floating caliper automotive brake disc as described in any one of claims 1-6.