Vehicle road noise processing method, device and equipment

Abstract

The embodiment of the invention provides a vehicle road noise processing method, device and equipment, and the method comprises the steps: building a road noise prediction model based on the acceleration data and road noise data of vehicle devices, so as to determine the correlation coefficient between the acceleration data and road noise data of each vehicle device; determining a key vehicle device among the vehicle devices based on the correlation coefficient, the acceleration of the key vehicle device being determined by a plurality of design parameters; key design parameters influencing the acceleration of the key vehicle device are determined; and performing iterative processing on the key design parameters based on an intelligent optimization algorithm and the road noise prediction model until the optimized key design parameters meet preset requirements. The accelerated speed of a key vehicle device serves as an intermediate variable, key design parameters influencing the vehicle road noise can be efficiently determined, and effective control over the road noise can be achieved in combination with an intelligent optimization algorithm and a road noise prediction model.

Description

Technical Field

[0001] The present application relates to the field of vehicle control, and specifically to a method, device and equipment for vehicle road noise processing. Background Art

[0002] With the development of vehicle electrification, without the masking of engine noise, road noise has become increasingly prominent in new energy vehicles and has become a top problem complained by users. The existing road noise optimization technologies have the following problems: The optimization method of real vehicle tuning is relatively lagging, which cannot meet the characteristics of short development cycle and limited verification methods, with low efficiency and low accuracy. Since road noise is the result of the comprehensive action of multiple variables, there is a strong cross-effect between variables, and the vehicle response is very complex within the feasible region, with multiple peaks and valleys. Therefore, it is difficult for traditional simulation optimization algorithms to find the global optimal solution. Summary of the Invention

[0003] In view of this, the present application provides a method, device and equipment for vehicle road noise processing, which is conducive to solving the problem that road noise is difficult to effectively reduce during vehicle driving in the prior art.

[0004] In a first aspect, an embodiment of the present application provides a method for vehicle road noise processing, including: Constructing a road noise prediction model based on the acceleration data and road noise data of vehicle components to determine the correlation coefficient between the acceleration data of each vehicle component and the road noise data; Determining key vehicle components among the vehicle components based on the correlation coefficient, where the acceleration of the key vehicle components is determined by multiple design parameters; Determining the key design parameters that affect the acceleration of the key vehicle components; Performing iterative processing on the key design parameters based on an intelligent optimization algorithm and the road noise prediction model until the optimized key design parameters meet the preset requirements.

[0005] In an optional embodiment, the constructing a road noise prediction model based on the acceleration data and road noise data of vehicle components to determine the correlation coefficient between the acceleration data of each vehicle component and the road noise data includes: Constructing a long short-term memory network (LSTM) model with the acceleration data of the vehicle components as input data and the road noise data as output data; Determining the Pearson correlation coefficient between the acceleration data of the vehicle components and the road noise data based on the LSTM model, and the Pearson correlation coefficient is the correlation coefficient.

[0006] In an optional embodiment, determining key vehicle components among the vehicle components based on the correlation coefficient includes: Determine the key vehicle components corresponding to the Pearson correlation coefficients exceeding the first threshold as the key vehicle components.

[0007] In an alternative embodiment, the determining the key design parameters affecting the acceleration of the key vehicle components includes: Construct a vehicle road noise model corresponding to the key vehicle components; Perform an Operational Deflection Shape (ODS) analysis on the vehicle road noise model to determine the design parameters affecting the acceleration of the key vehicle components; Perform a modal contribution analysis on the design parameters to determine the key design parameters affecting the acceleration of the key vehicle components.

[0008] In an alternative embodiment, the iteratively processing the key design parameters based on the intelligent optimization algorithm and the road noise prediction model until the optimized key design parameters meet the preset requirements includes: Optimize the key design parameters based on the Particle Swarm Optimization (PSO) algorithm and determine the optimized acceleration data of each key vehicle component; Input the optimized acceleration data into the road noise prediction model to output the optimized road noise data; When the optimized road noise data meets the preset standard, determine that the key design parameters meet the preset requirements; when the optimized road noise data does not meet the preset standard, perform the next optimization process on the key design parameters.

[0009] In an alternative embodiment, before constructing the road noise prediction model based on the acceleration data and road noise data of the vehicle components, the method further includes: During the vehicle test, acquire the acceleration data of the vehicle components based on acceleration sensors, and the acceleration sensors are arranged at each vehicle component; Acquire the road noise data based on microphones inside the vehicle.

[0010] In an alternative embodiment, constructing a Long Short-Term Memory (LSTM) model with the acceleration data of the vehicle components as input data and the road noise data as output data includes: Input the acceleration data of the vehicle components into the LSTM model to enable the LSTM model to output prediction data; Determine a loss function based on the prediction data and the road noise data; Adjust the weight parameters of each layer of the LSTM based on the loss function.

[0011] In a second aspect, an embodiment of the present application provides a vehicle road noise processing device, including: A building module for building a road noise prediction model based on the acceleration data and road noise data of vehicle components to determine the correlation coefficient between the acceleration data of each vehicle component and the road noise data; A first determination module for determining critical vehicle components among the vehicle components based on the correlation coefficient, where the acceleration of the critical vehicle components is determined by multiple design parameters; A second determination module for determining the critical design parameters that affect the acceleration of the critical vehicle components; A processing module for iteratively processing the critical design parameters based on an intelligent optimization algorithm and the road noise prediction model until the optimized critical design parameters meet the preset requirements.

[0012] In a third aspect, an embodiment of the present application provides an electronic device, including a memory for storing computer program instructions and a processor for executing the program instructions. When the computer program instructions are executed by the processor, the electronic device is triggered to execute the method according to any one of the first aspects described above.

[0013] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, where the computer-readable storage medium includes a stored program. When the program runs, it controls the device where the computer-readable storage medium is located to execute the method according to any one of the first aspects described above.

[0014] In a fifth aspect, an embodiment of the present application provides a computer program product, where the computer program product includes executable instructions. When the executable instructions are executed on a computer, the computer is caused to execute the method according to any one of the first aspects described above.

[0015] Adopting the solution provided by the embodiment of the present application, a road noise prediction model is built based on the acceleration data and road noise data of vehicle components to determine the correlation coefficient between the acceleration data of each vehicle component and the road noise data; critical vehicle components are determined among the vehicle components based on the correlation coefficient, and the acceleration of the critical vehicle components is determined by multiple design parameters; the critical design parameters that affect the acceleration of the critical vehicle components are determined; the critical design parameters are iteratively processed based on an intelligent optimization algorithm and the road noise prediction model until the optimized critical design parameters meet the preset requirements. By using the acceleration of the critical vehicle components as an intermediate variable, the critical design parameters that affect vehicle road noise can be efficiently determined, and the effective control of road noise can be achieved by combining the intelligent optimization algorithm and the road noise prediction model. Description of the Drawings

[0016] To more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0017] Figure 1 It is a schematic flowchart of a vehicle road noise processing method provided by an embodiment of the present application; Figure 2 It is an example schematic diagram of a vehicle road noise processing method provided by an embodiment of the present application; Figure 3 It is an example schematic diagram of another vehicle road noise processing method provided by an embodiment of the present application; Figure 4 It is an example schematic diagram of another vehicle road noise processing method provided by an embodiment of the present application; Figure 5 It is an example schematic diagram of another vehicle road noise processing method provided by an embodiment of the present application; Figure 6 It is a schematic flowchart of another vehicle road noise processing method provided by an embodiment of the present application; Figure 7 It is an example schematic diagram of another vehicle road noise processing method provided by an embodiment of the present application; Figure 8 It is a schematic structural diagram of a vehicle road noise processing device provided by an embodiment of the present application; Figure 9 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0018] To better understand the technical solutions of the present application, the following will describe the embodiments of the present application in detail with reference to the drawings.

[0019] It should be clear that the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0020] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. The singular forms "a", "the" and "said" used in the embodiments of the present application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise.

[0021] It should be understood that the term "and / or" used herein is merely a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. Additionally, the character " / " in this text generally indicates that the associated objects before and after are in an "or" relationship.

[0022] The existing road noise control methods have at least the following problems: (1) The optimization method of real vehicle tuning is relatively lagging, unable to meet the short development cycle. The verification methods and means are limited, and the verification methods may change the original structural state or be contrary to the actual optimization direction, requiring experienced experts to make judgments, with low efficiency and inaccurate accuracy. (2) Using the traditional vehicle road noise simulation method requires a prototype vehicle to perform road spectrum load inverse calculation, or requires a multi-body dynamics vehicle model and virtual road surface to calculate the load, with poor accuracy and large errors, resulting in inaccurate subsequent road noise simulation. Due to the complexity of vehicle road noise, the traditional road noise optimization method cannot determine the quantitative relationship between road noise and design parameters. (3) Since road noise is the result of the combined action of multiple design parameters, there is a strong cross-effect between parameters, and the vehicle response is very complex within the feasible region, with multiple peaks and valleys. Therefore, it is difficult for traditional simulation optimization algorithms to find the global optimal solution.

[0023] In view of the above problems, the embodiments of the present application provide a vehicle road noise processing method, using the acceleration data of key vehicle components as an intermediate value, and establishing connections with road noise data and key design parameters respectively above and below, effectively improving the performance and efficiency of road noise optimization.

[0024] Figure 1 It is a schematic flowchart of a vehicle road noise processing method provided by the embodiments of the present application. This method can be applied to a processing device. Optionally, the processing device can include electronic devices with logical processing functions such as a computer, a server, etc. As Figure 1 shown, this method can include: Step 101, constructing a road noise prediction model based on the acceleration data and road noise data of vehicle components to determine the correlation coefficient between the acceleration data of each vehicle component and the road noise data.

[0025] Step 102, determining key vehicle components among vehicle components based on the correlation coefficient. The acceleration of the key vehicle components is determined by multiple design parameters.

[0026] Step 103, determining the key design parameters that affect the acceleration of the key vehicle components.

[0027] Step 104, performing iterative processing on the key design parameters based on the intelligent optimization algorithm and the road noise prediction model until the optimized key design parameters meet the preset requirements.

[0028] During the vehicle driving process, vehicle components at certain positions generate noise due to vibration, which affects the driver's experience. The acceleration data of vehicle components can be used to characterize information such as the vibration direction, magnitude, and pattern of vehicle components, and the road noise data can be used to characterize the magnitude of the noise that can be heard inside the vehicle. The distribution of vehicle components that can generate noise is irregular. Referring to Figure 2 , the positions marked with numbers can represent vehicle components, and noise may be generated due to vibration in its chassis and body.

[0029] In an alternative embodiment, the acceleration data and road noise data of vehicle components can be obtained through on-road vehicle tests. Use an engineering prototype vehicle to drive on a standard rough road surface and collect the acceleration data and road noise data of vehicle components in real time. The test steps are as follows: (1) Check that the vehicle status meets the requirements. (2) Paste acceleration sensors at key positions on the selected chassis and body. (3) Arrange microphones at the ear positions of the driver and rear passengers inside the vehicle. (4) Connect the test equipment, acceleration sensors, and microphones with wires. (5) Open the test software at the test equipment and debug it to the normal state. (6) Drive the vehicle to a dedicated standard rough road surface and drive at a constant speed of 50 km / h, and at the same time start the test software for testing. (7) Data processing and output.

[0030] In an alternative embodiment, the acceleration data and road noise data of vehicle components can also be obtained by using a vehicle multi-body dynamics model and a virtual road surface through a dynamic simulation method. The specific steps include: (1) Build a multi-body dynamics model. (2) Establish acceleration output at key positions on the chassis and body. (3) Model debugging. (4) Conduct virtual test simulations. (5) Extract the required acceleration data at key positions on the chassis and body and check.

[0031] In step 101, the processing device can construct a road noise prediction model based on the obtained acceleration data and road noise data to determine the correlation coefficient between the acceleration data and road noise data of each vehicle component. Among them, the road noise prediction model can be constructed based on a Long Short-Term Memory (LSTM) neural network.

[0032] The method of obtaining data through simulation can be used in the early stage of project development, and the method of obtaining data through on-road vehicle tests can be used in the prototype vehicle tuning stage. The vehicle road noise processing method of this application is applicable to the entire process of road noise development. LSTM is a special architecture of Recurrent Neural Network (RNN), which is specifically designed to handle and predict problems based on time-series data. Compared with traditional RNNs, LSTM performs better in capturing long-term dependencies. Through the Forget Gate, Input Gate, and Output Gate, it can selectively remember or forget information, effectively solving the problem of vanishing gradients or exploding gradients in long time-series data and improving the accuracy of long time-series data prediction.

[0033] During the model construction process, the input signal is the acceleration data of key vehicle components on the chassis and body during vehicle driving, and the output of the model is road noise data, that is, the sound pressure level of the noise generated during vehicle driving. By using the LSTM model to learn and model the input data, the complex non-linear relationship between the acceleration data and the road noise data can be captured. The specific modeling process is as follows: 1. Data preprocessing.

[0034] (1) Data cleaning: Detect and process outliers in the data. Use statistical methods (such as the 3σ principle, box plots, etc.) to detect and remove outliers in the acceleration data. For the removed data points, interpolation methods (such as linear interpolation, polynomial interpolation) are used to fill them to maintain the continuity of the data. Smooth the acceleration data through low-pass filtering technology to remove high-frequency noise components in the signal.

[0035] (2) Feature extraction: Extract time-domain features from the acceleration data: mean, variance, standard deviation, peak value, peak-to-peak value, energy, zero-crossing rate, etc. These features can reflect the basic statistical characteristics and waveform features of the signal. Convert the acceleration data from the time domain to the frequency domain through the Fast Fourier Transform (FFT) to extract frequency-domain features, including spectral density, main frequency, frequency center, frequency bandwidth, etc. These features help to capture the frequency components and distribution characteristics in the signal.

[0036] (3) Data normalization: To eliminate the influence of different dimensions and numerical ranges on model training, normalize the extracted features and scale the feature values to the range of [0, 1].

[0037] 2. Data serialization (1) Determine the sequence length: The length of the data sequence is determined according to the sampling frequency of the test equipment. Usually, the test sampling frequency is 2048Hz, so the data sequence length can be determined as 2048.

[0038] (2)Slicing process: After determining the sequence length, slice the processed acceleration data according to this length. The sliding window method starts from a starting point, intercepts a section of data as the first sequence according to the sequence length, then slides backward by one or more data points, and intercepts the second section of data as the second sequence, and so on, until the entire acceleration data is traversed. This step can generate more sequence data while keeping there being overlapping parts between adjacent sequences.

[0039] 3. Model construction In the embodiment of this application, when constructing a road noise prediction model based on the LSTM neural network, the TensorFlow deep learning framework is adopted. Among them, the input layer is responsible for receiving the serialized acceleration data. According to the results of data preprocessing, the size of the input layer should match the feature dimension of each column of data. For example, if the acceleration data at each time step is represented as a feature vector, the size of the input layer should be the dimension of this feature vector; the hidden layer consists of multiple LSTM units, and these units are responsible for capturing the long-term dependencies in the time series data. Stack multiple LSTM layers as needed to increase the depth and complexity of the model. In each LSTM layer, the number of hidden units (also known as the number of LSTM units or the size of the hidden state) needs to be specified. At the same time, a Dropout layer is added after the LSTM layer to prevent overfitting and improve the generalization ability of the model; the output layer is responsible for outputting the predicted road noise data. The output layer of the model is a fully connected layer, and its size matches the dimension of the predicted road noise pressure signal (for example, if only the sound pressure level value at the driver's right ear is predicted, the size of the output layer is 1; if the sound pressure level values at the driver's right ear and the rear passenger compartment are predicted at the same time, the size of the output layer is 2), and the output layer uses a linear activation function.

[0040] 4. Model training: Optimize the model parameters through the backpropagation algorithm to enable the model to accurately predict the road noise pressure signal. During the training process, the mean squared error (MSE) is used as the loss function, and Adam is used as the optimizer.

[0041] 5. Model verification and evaluation: After training is completed, in order to ensure the accuracy of the model, the performance of the model is evaluated using the validation set, and the evaluation index is the coefficient of determination (R²). When R²≥90%, the model accuracy meets the requirements. When R²<90%, the model needs to be parameter-adjusted and trained again until the accuracy meets the requirements.

[0042] After the LSTM training is completed, the Pearson correlation coefficient between the acceleration data and road noise data of each vehicle component can be calculated. Among them, the Pearson correlation technique can be used as the correlation coefficient in step 101. The main variables affecting road noise can be determined through the Pearson correlation coefficient. The larger the Pearson coefficient, the greater the impact of the acceleration of the vehicle component on road noise. The smaller the Pearson coefficient, the smaller the impact of the acceleration of the vehicle component on road noise.

[0043] In step 102, based on the correlation coefficient (i.e., the Pearson correlation coefficient), the processing device can determine the key vehicle components among multiple vehicle components. The vibration of each vehicle component may affect the magnitude of road noise. Based on the correlation coefficient, the tester can focus on the key vehicle components that have a greater impact on road noise to improve the efficiency of road noise control. Optionally, the processing device can determine the vehicle components corresponding to the Pearson correlation coefficient exceeding the first threshold as the key vehicle components.

[0044] The acceleration of each key vehicle component is determined by multiple design parameters, such as elastic modulus, density, and Poisson's ratio. By adjusting the design parameters, the acceleration (i.e., the degree of vibration) of the key vehicle component during vehicle driving can be changed, thereby reducing the magnitude of road noise.

[0045] Since there are many design parameters, in step 103, the processing device can first determine the key design parameters that have a greater impact on the acceleration of the key vehicle component. Specifically, the process of determining the key design parameters may include: (1) Establish a vehicle road noise CAE model: This CAE model includes a chassis suspension system, a power transmission system, an interior body system (trimBody), and an in-vehicle acoustic cavity model, etc. Refer to Figure 3 , different systems can be displayed in different colors. To ensure the accuracy of the CAE model, it is necessary to compare it with the test results to ensure that the simulated road noise curve and the test curve are consistent in trend. At the same time, it is necessary to ensure that the frequency band where the peak of the problem point is located is the same. Refer to Figure 4 , if the difference between the CAE simulation result and the test result is too large, it is necessary to recheck the connection relationship and material properties of the CAE model until the accuracy meets the requirements.

[0046] (2) Vehicle Operational Deflection Shape (ODS) analysis: Vehicle ODS analysis is an advanced CAE analysis technology, mainly used to determine and visualize the vibration mode of vehicle components under given operating conditions. By analyzing the vibration response of vehicle components under specific operating conditions, the dynamic characteristics of vehicle components are identified, so as to determine the design parameters that affect the vibration of vehicle components.

[0047] (3)Modal contribution analysis: Through modal contribution analysis, it is possible to find out which order of mode makes the greatest contribution to the problem peak and whether it is a positive or negative contribution. By performing modal contribution analysis on various design parameters of vehicle components, the key design parameters that have a greater impact on the vibration of vehicle components can be determined.

[0048] In step 104, the processing device performs iterative optimization processing on the key design parameters based on the intelligent optimization algorithm and the road noise prediction model until the optimized key design parameters meet the preset requirements. During vehicle driving, the key design parameters affect the vibration conditions of key vehicle components, and the vibration conditions of key vehicle components in turn affect the road noise level. Therefore, when the optimized road noise data meets the preset standard, the processing device can determine that the key design parameters meet the preset requirements; when the optimized road noise data does not meet the preset standard, the processing device needs to perform the next optimization processing on the key design parameters.

[0049] During the iterative optimization process, the processing device takes the key design parameters as the optimization variables and determines the feasible region range of the optimization variables according to the design boundaries. The optimization objective is the acceleration of key vehicle components. Using an intelligent optimization algorithm (particle swarm optimization algorithm), a global search is performed within the feasible region of the optimization variables to find the global optimal solution. The mathematical principle of this optimization algorithm is as follows: Among them, v is the velocity of the particle, and x is the position of the particle. The subscript j represents the j-th dimension of the particle, the subscript i represents the i-th particle, t represents the current iteration number, c1 and c2 are both acceleration constants, usually taking values in the interval (0, 2), and r1 and r2 are two independent random numbers with values in the range [0, 1].

[0050] The specific optimization steps may include: (1)Initialization: Set the key design parameters as particles, randomly generate the initial position and velocity of the particles within the value range, and initialize the best individual position and the best group position. The initial position and velocity are randomly generated within the upper and lower limits of the search space.

[0051] (2)Evaluation: Calculate the fitness value of each particle, that is, the objective function value. The individual best position (Pbest) is the individual with the optimal fitness value among the initialized positions, and the group best position (Gbest) is the individual with the optimal fitness value among all particles.

[0052] (3)Update velocity and position: Update the velocity and position of the particles according to the individual best position and the group best position, as well as some weights and random factors. The velocity update formula usually includes an inertial part, a self-awareness part, and a social awareness part, and the position update is based on the new velocity.

[0053] (4) Update the best position: Compare the current position of each particle with its historical best position, and update the individual best position and the global best position.

[0054] The processing device sets a maximum number of iterations (e.g., 100 times). After each optimization is completed, it checks whether the key design parameters meet the requirements. When the maximum number of iterations is reached or the requirements are met, the iteration stops; otherwise, the next optimization is performed. Specifically, after each optimization of the key design parameters, the processing device inputs the corresponding optimized acceleration data into the road noise prediction model to output the optimized road noise data, and then determines whether the road noise level meets the preset standard. Refer to Figure 5 , the road noise data predicted by the LSTM is very close to the actual measured road noise data. By outputting the road noise data through the LSTM model in the iterative optimization, the prediction accuracy of the road noise data can be improved.

[0055] In the embodiments of the present application, with the key vehicle components as the connection, the key design parameters affecting the vehicle road noise can be accurately identified. Through the cooperation of the intelligent optimization algorithm and the LSTM model, a global search is performed from the feasible region of the key design parameters, which can avoid falling into the numerical oscillation of the local optimal solution, effectively identify the global optimal solution, and can significantly reduce the road noise peak.

[0056] Figure 6 FIG. is a schematic flowchart of another vehicle road noise processing method provided by the embodiments of the present application. As Figure 6 shown, the method may include: Step 601, obtain road noise data and acceleration data of vehicle components.

[0057] In the initial stage of the project, the acceleration data and road noise data of vehicle components can be obtained through the simulation model. In the real vehicle tuning stage, the acceleration and road noise data of vehicle components can be obtained through real vehicle tests.

[0058] Step 602, establish an LSTM road noise prediction model.

[0059] Train the LSTM road noise prediction model with the acceleration data of vehicle components as the input data and the road noise data as the output data.

[0060] Step 603, determine the correlation coefficient based on the road noise prediction model.

[0061] After the LSTM road noise prediction model is trained, it outputs the Pearson correlation coefficient, and then determines the influence degree of the vibration of each vehicle component on the road noise.

[0062] Step 604, determine the key vehicle components.

[0063] Based on the Pearson correlation coefficient, screen out the key vehicle components that have a greater impact on the road noise among the vehicle components.

[0064] Step 605, perform ODS analysis and modal contribution analysis on the vehicle road noise model.

[0065] Build a CAE model, perform ODS analysis on the CAE model to determine the design parameters that affect the vibration modes of key vehicle components. Perform modal contribution analysis on the CAE model to determine the degree of influence of each design parameter on the vibration modes of key vehicle components.

[0066] Step 606, determine the key design parameters.

[0067] Based on the modal contribution, screen out the key design parameters that have a greater impact on the vibration of key vehicle components among the design parameters.

[0068] Step 607, optimize the key design parameters based on the particle swarm optimization algorithm.

[0069] Taking the acceleration of the key vehicle components as the optimization target, optimize the key design parameters based on the particle swarm optimization algorithm.

[0070] Step 608, check whether the preset requirements are met or the maximum number of iterations is reached. If so, return to Step 607; otherwise, proceed to Step 609.

[0071] Output the optimized acceleration data to the road noise prediction model to output the optimized road noise data, and then determine whether the road noise generated by the vehicle meets the expected standard. If the expected standard is met, end the optimization process; otherwise, continue the next iteration of optimization.

[0072] Step 609, end the optimization process.

[0073] Refer to Figure 7 , the black curve is the original road noise data, and the gray interval is the road noise data after optimization processing, and its peak value has a significant decrease.

[0074] In the embodiment of the present application, by using the acceleration of the key vehicle components as an intermediate variable, the key design parameters that affect the vehicle road noise can be efficiently determined, and the effective control of the road noise can be achieved by combining the intelligent optimization algorithm and the road noise prediction model. Figure 8 It is a schematic structural diagram of a vehicle road noise processing device provided by an embodiment of the present application. As Figure 8 shown, the device may include: A construction module 810, configured to build a road noise prediction model based on the acceleration data and road noise data of vehicle components to determine the correlation coefficient between the acceleration data of each vehicle component and the road noise data.

[0075] A first determination module 820, configured to determine key vehicle components among the vehicle components based on the correlation coefficient, and the acceleration of the key vehicle components is determined by multiple design parameters.

[0076] A second determination module 830, configured to determine key design parameters that affect the acceleration of the key vehicle device.

[0077] A processing module 840, configured to perform iterative processing on the key design parameters based on an intelligent optimization algorithm and the road noise prediction model until the optimized key design parameters meet preset requirements.

[0078] Corresponding to the above embodiments, the present application further provides an electronic device. Figure 9 FIG. is a schematic structural diagram of an electronic device provided by an embodiment of the present application. The electronic device 900 may include: a processor 901, a memory 902, and a communication unit 903. These components communicate through one or more buses. Those skilled in the art can understand that the structure of the electronic device shown in the figure does not constitute a limitation on the embodiments of the present application. It can be a bus structure, a star structure, and may also include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0079] Among them, the communication unit 903 is configured to establish a communication channel, so that the electronic device can communicate with other devices. Receive user data sent by other devices or send user data to other devices.

[0080] The processor 901 is the control center of the electronic device, connecting various parts of the entire electronic device through various interfaces and lines. By running or executing software programs, instructions, and / or modules stored in the memory 902, and calling data stored in the memory, to perform various functions of the electronic device and / or process data. The processor may be composed of an integrated circuit (IC). For example, it may be composed of a single packaged IC, or may be composed of multiple packaged ICs with the same or different functions connected. For example, the processor 901 may only include a central processing unit (CPU). In the embodiment of the present application, the CPU may be a single operation core or may include multiple operation cores.

[0081] The memory 902 is used to store execution instructions of the processor 901. The memory 902 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, a magnetic disk, or an optical disc.

[0082] When the execution instructions in the memory 902 are executed by the processor 901, the electronic device 900 is enabled to execute some or all of the steps in the above embodiments.

[0083] In a specific implementation, the present application further provides a computer storage medium. The computer storage medium may store a program, and when the program is executed, it may include some or all of the steps in the embodiments of the vehicle road noise processing method provided by the present application. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), or the like.

[0084] In a specific implementation, the present application further provides a computer program product. The computer program product includes executable instructions, and when the executable instructions are executed on a computer, the computer is enabled to execute some or all of the steps in the embodiments of the vehicle road noise processing method provided by the present application.

[0085] The embodiments of the present application further provide a non-transitory computer-readable storage medium. The non-transitory computer-readable storage medium stores computer instructions, and the computer instructions cause the computer to execute the vehicle road noise processing method provided by the embodiments of the present application.

[0086] The above non-transitory computer-readable storage medium may be any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the computer-readable storage medium include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (hereinafter referred to as: ROM), an erasable programmable read-only memory (hereinafter referred to as: EPROM), or a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, the computer-readable storage medium may be any tangible medium that contains or stores a program, and the program may be used by or in combination with an instruction execution system, apparatus, or device.

[0087] A computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. Such a propagated data signal may take various forms, including - but not limited to - electromagnetic signals, optical signals, or any suitable combination of the foregoing. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device.

[0088] The program code contained on a computer-readable medium may be transmitted using any appropriate medium, including - but not limited to - wireless, wire, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0089] Those skilled in the art can clearly understand that the technology in the embodiments of this application can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solutions in the embodiments of this application, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disc, etc., and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0090] For the same or similar parts among the various embodiments in this specification, reference can be made to each other. In particular, for the apparatus embodiments and the terminal embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the descriptions in the method embodiments.

Claims

1. A vehicle road noise processing method, characterized in that: include: Building a road noise prediction model based on the acceleration data of vehicle components and the road noise data to determine a correlation coefficient between the acceleration data of each vehicle component and the road noise data; determining a key vehicle component among the vehicle components based on the correlation coefficient, wherein the acceleration of the key vehicle component is determined by a plurality of design parameters; Determining key design parameters that affect the acceleration of the key vehicle components; The key design parameters are iteratively processed based on the intelligent optimization algorithm and the road noise prediction model until the optimized key design parameters meet the preset requirements.

2. The method according to claim 1, characterized in that The road noise prediction model is constructed based on the acceleration data and road noise data of the vehicle components to determine the correlation coefficient between the acceleration data of each vehicle component and the road noise data, including: Using the acceleration data of the vehicle device as input data and the road noise data as output data, a long short-term memory network LSTM model is constructed; A Pearson correlation coefficient between the acceleration data of the vehicle device and the road noise data is determined based on the LSTM model, and the Pearson correlation coefficient is the association coefficient.

3. The method according to claim 2, characterized in that Determining a key vehicle component among the vehicle components based on the correlation coefficient includes: The vehicle component corresponding to the Pearson correlation coefficient exceeding the first threshold is determined as the key vehicle component.

4. The method according to claim 1, characterized in that: The determining of key design parameters affecting the acceleration of the key vehicle components includes: Constructing a whole vehicle road noise model corresponding to the key vehicle components; Performing an operating deformation (ODS) analysis on the vehicle road noise model to determine the design parameters that affect the acceleration of the key vehicle components; A modal contribution analysis is performed on the design parameters to determine key design parameters that affect the acceleration of the key vehicle components.

5. The method according to claim 1, characterized in that The iterative processing of the key design parameters based on the intelligent optimization algorithm and the road noise prediction model until the optimized key design parameters meet the preset requirements includes: Optimizing the key design parameters based on a particle swarm optimization algorithm and determining the optimized acceleration data of each key vehicle component; Inputting the optimized acceleration data into the road noise prediction model to output optimized road noise data; When the optimized road noise data meets the preset standard, it is determined that the key design parameter meets the preset requirement. When the optimized road noise data does not meet the preset standard, the key design parameter is optimized for the next time.

6. The method according to claim 1, characterized in that Before constructing the road noise prediction model based on the acceleration data and road noise data of the vehicle device, the method further includes: During the vehicle testing process, the acceleration data of the vehicle components are acquired based on the acceleration sensor, wherein the acceleration sensor is arranged at each vehicle component; The road noise data is acquired based on a microphone inside the vehicle.

7. The method according to claim 2, characterized in that The method uses the acceleration data of the vehicle device as input data and the road noise data as output data to construct a long short-term memory network LSTM model, including: Inputting the acceleration data of the vehicle device into the LSTM model so that the LSTM model outputs prediction data; determining a loss function based on the predicted data and the road noise data; The weight parameters of each layer of the LSTM are adjusted based on the loss function.

8. A vehicle road noise processing device, characterized in that: include: A construction module, for constructing a road noise prediction model based on the acceleration data of vehicle components and the road noise data, so as to determine a correlation coefficient between the acceleration data of each vehicle component and the road noise data; A first determination module, configured to determine a key vehicle component among the vehicle components based on the correlation coefficient, wherein the acceleration of the key vehicle component is determined by a plurality of design parameters; A second determination module, used to determine key design parameters that affect the acceleration of the key vehicle component; The processing module is used to iteratively process the key design parameters based on the intelligent optimization algorithm and the road noise prediction model until the optimized key design parameters meet the preset requirements.

9. An electronic device, characterized in that: The electronic device comprises a memory for storing computer program instructions and a processor for executing the program instructions, wherein when the computer program instructions are executed by the processor, the electronic device executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method according to any one of claims 1 to 7.

Images