Vehicle noise reduction method and device, vehicle and storage medium
By acquiring vehicle noise and equipment data, and utilizing predictive models and control modules to achieve dynamic noise reduction, the problem that static noise reduction methods cannot adapt to changes in equipment operating conditions and electromagnetic environment is solved, thus realizing real-time perception and efficient noise reduction.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHERY AUTOMOBILE CO LTD
- Filing Date
- 2026-01-12
- Publication Date
- 2026-04-17
AI Technical Summary
Existing technologies cannot perceive the dynamic coupling relationship between equipment operating conditions and the electromagnetic environment in real time, making it difficult to cope with the nonlinear changes in complex electromagnetic noise. Static noise reduction solutions cannot adapt to changes in equipment load and environmental fluctuations.
By acquiring vehicle noise data and target equipment operating data, processing them into multi-dimensional feature vectors, inputting them into a noise prediction model, determining noise reduction conditions, and controlling the preset noise reduction device and target equipment to make adjustments, dynamic noise reduction is achieved.
It can adapt to complex electromagnetic environments without human intervention, balancing noise reduction and equipment operating efficiency, and achieves real-time perception and dynamic adjustment of electromagnetic noise.
Smart Images
Figure CN121884756A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of vehicle technology, and in particular to a method, apparatus, vehicle, and storage medium for noise reduction in vehicles. Background Technology
[0002] With the increasing density and intelligence of vehicles, electromagnetic interference (EMI) is becoming a more prominent issue affecting equipment operational stability and surrounding electronic systems. Current EMI management methods primarily rely on static absorbing material placement or manual setting of equipment parameters.
[0003] However, the relevant technologies have the following problems: (1) they cannot perceive the dynamic coupling noise of environmental changes and equipment operation status in real time; (2) manual adjustment is inefficient and it is difficult to cope with the nonlinear noise characteristics of complex electromagnetic environments; (3) static noise reduction schemes cannot adapt to dynamic scenarios such as equipment load changes and temperature and humidity fluctuations. Summary of the Invention
[0004] This application provides a noise reduction method, device, vehicle, and storage medium for vehicles to solve the problems that static noise reduction methods cannot perceive the dynamic coupling relationship between equipment operating conditions and electromagnetic environment in real time and are difficult to cope with complex nonlinear noise changes. This application can adapt to complex electromagnetic environments without human intervention and takes into account both noise reduction effect and equipment operating efficiency.
[0005] The first aspect of this application provides a noise reduction method for a vehicle, comprising the following steps: Acquire noise data from the vehicle and operational data from the target equipment; The noise data and the running data are processed to obtain a multi-dimensional feature vector. The multi-dimensional feature vector is then input into a preset noise prediction model to obtain noise distribution information. Based on the noise distribution information, it is determined whether the vehicle meets the preset noise reduction conditions. If the vehicle meets the preset noise reduction conditions, the noise distribution information is input into the preset noise reduction model to obtain the adjustment instructions for the preset noise reduction device and the adjustment instructions for the target device. The preset noise reduction device is controlled to adjust based on the adjustment instructions, and the target device is controlled to adjust based on the adjustment instructions.
[0006] Optionally, in some embodiments, the noise distribution information includes the predicted noise amplitude at each frequency point, and the preset noise reduction condition is that the predicted noise amplitude at any frequency point exceeds the corresponding noise threshold.
[0007] Optionally, in some embodiments, before inputting the multidimensional feature vector into the preset noise prediction model to obtain the noise distribution information, the following steps are included: Obtain historical feature vectors and construct a dataset based on the historical feature vectors; Based on a preset partitioning ratio, the dataset is divided into a training set, a validation set, and a test set. Construct a target neural network by inputting the training set into the target neural network for training to obtain initial model parameters; Based on the initial model parameters, the validation set is input into the target neural network for performance evaluation, and the initial model parameters are adjusted according to the performance evaluation results until the joint loss function of the validation set converges to obtain the optimal model parameters. Based on the optimal model parameters, the test set is input into the target neural network for model testing, and when the test results meet the preset requirements, the preset noise prediction model is obtained.
[0008] Optionally, in some embodiments, the state space of the preset noise reduction model includes at least one of the noise distribution information and the operating parameters of the target device; The motion space includes at least one of the following: the pose adjustment amount of the preset noise reduction device and the operating parameter adjustment amount of the target device.
[0009] Optionally, in some embodiments, after controlling the preset noise reduction device to adjust based on the adjustment command and controlling the target device to adjust based on the adjustment command, the process includes: New noise data is acquired, and the new noise data is processed to obtain a noise spectrum diagram, which is then visualized.
[0010] A second aspect of this application provides a noise reduction device for a vehicle, comprising: The acquisition module is used to acquire noise data from the vehicle and operational data from the target equipment. The judgment module is used to process the noise data and the running data to obtain a multi-dimensional feature vector, input the multi-dimensional feature vector into a preset noise prediction model to obtain noise distribution information, and judge whether the vehicle meets the preset noise reduction conditions based on the noise distribution information. The control module is used to input the noise distribution information into a preset noise reduction model to obtain adjustment instructions for a preset noise reduction device and adjustment instructions for the target device when the vehicle meets the preset noise reduction conditions, and to control the preset noise reduction device to adjust based on the adjustment instructions and control the target device to adjust based on the adjustment instructions.
[0011] Optionally, in some embodiments, the noise distribution information includes the predicted noise amplitude at each frequency point, and the preset noise reduction condition is that the predicted noise amplitude at any frequency point exceeds the corresponding noise threshold.
[0012] Optionally, in some embodiments, before inputting the multidimensional feature vector into the preset noise prediction model to obtain the noise distribution information, the judgment module includes: An acquisition unit is used to acquire historical feature vectors and construct a dataset based on the historical feature vectors; A partitioning unit is used to divide the dataset into a training set, a validation set, and a test set based on a preset partitioning ratio. A training unit is used to construct a target neural network, and to obtain initial model parameters by inputting the training set into the target neural network for training. An optimization unit is used to input the validation set into the target neural network for performance evaluation based on the initial model parameters, and adjust the initial model parameters according to the performance evaluation results until the joint loss function of the validation set converges to obtain the optimal model parameters. The testing unit is used to input the test set into the target neural network based on the optimal model parameters to test the model, and obtain the preset noise prediction model when the test results meet the preset requirements.
[0013] Optionally, in some embodiments, the state space of the preset noise reduction model includes at least one of the noise distribution information and the operating parameters of the target device; The motion space includes at least one of the following: the pose adjustment amount of the preset noise reduction device and the operating parameter adjustment amount of the target device.
[0014] Optionally, in some embodiments, after controlling the preset noise reduction device to adjust based on the adjustment command and controlling the target device to adjust based on the adjustment command, the control module includes: A visualization unit is used to acquire new noise data, process the new noise data to obtain a noise spectrum, and visualize the noise spectrum.
[0015] A third aspect of this application provides a vehicle, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the noise reduction method for the vehicle as described in the above embodiments.
[0016] A fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement the noise reduction method for a vehicle as described in the above embodiments.
[0017] Therefore, by acquiring vehicle noise data and target equipment operating data, and processing the noise data and operating data to obtain multi-dimensional feature vectors, the multi-dimensional feature vectors are input into a preset noise prediction model to obtain noise distribution information. Based on the noise distribution information, it is determined whether the vehicle meets the preset noise reduction conditions. If the vehicle meets the preset noise reduction conditions, the noise distribution information is input into the preset noise reduction model to obtain adjustment instructions for the preset noise reduction device and adjustment instructions for the target equipment. The preset noise reduction device is controlled to adjust based on the adjustment instructions, and the target equipment is also controlled to adjust based on the adjustment instructions. This solves the problem that static noise reduction methods cannot perceive the dynamic coupling relationship between equipment operating conditions and the electromagnetic environment in real time and are difficult to cope with complex noise changes. This application can adapt to complex electromagnetic environments without manual intervention, balancing noise reduction effect and equipment operating efficiency.
[0018] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description
[0019] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a schematic diagram of a noise reduction system according to an embodiment of this application; Figure 2 This is a flowchart of a vehicle noise reduction method provided according to an embodiment of this application; Figure 3 This is a schematic diagram of a vehicle noise reduction method according to an embodiment of this application; Figure 4 This is a block diagram of a vehicle noise reduction device provided according to an embodiment of this application; Figure 5 This is a structural schematic diagram of a vehicle provided according to an embodiment of this application. Detailed Implementation
[0020] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.
[0021] The following description, with reference to the accompanying drawings, describes a vehicle noise reduction method, apparatus, vehicle, and storage medium according to embodiments of this application. Addressing the problems mentioned in the background art regarding the inability of static noise reduction methods to perceive the dynamic coupling relationship between equipment operating conditions and the electromagnetic environment in real time, and their difficulty in coping with complex noise changes, this application provides a vehicle noise reduction method. In this method, noise data of the vehicle and operating data of the target equipment are acquired, and the noise data and operating data are processed to obtain a multi-dimensional feature vector. The multi-dimensional feature vector is input into a preset noise prediction model to obtain noise distribution information. Based on the noise distribution information, it is determined whether the vehicle meets preset noise reduction conditions. If the vehicle meets the preset noise reduction conditions, the noise distribution information is input into the preset noise reduction model to obtain a preset adjustment command for the noise reduction device and an adjustment command for the target equipment. The preset noise reduction device is controlled to adjust based on the adjustment command, and the target equipment is controlled to adjust based on the adjustment command. Therefore, the problems of static noise reduction methods being unable to perceive the dynamic coupling relationship between equipment operating conditions and the electromagnetic environment in real time, and their difficulty in coping with complex noise changes, are solved. This application can adapt to complex electromagnetic environments without manual intervention, balancing noise reduction effect and equipment operating efficiency.
[0022] Before introducing the vehicle noise reduction method of the embodiments of this application, let's first introduce a noise reduction system used in the embodiments of this application, such as... Figure 1 As shown, the noise reduction system includes: a sensor module, a data processing module, a machine learning module, and an execution module.
[0023] The sensor module is used to collect electromagnetic noise signals and operating status parameters of the target device and transmit the data through V2X (Vehicle-to-Everything). It includes an electromagnetic noise sensor, a device operating parameter sensor, and a V2X communication unit.
[0024] Among them, electromagnetic noise sensors (such as loop antennas and electric field probes) are used to collect electromagnetic radiation signals in the 0.15MHz-6GHz frequency band in real time, with a resolution ≤1dBμV / m; equipment operating parameter sensors (vibration sensors and current / voltage transmitters) are used to collect operating status data such as equipment speed, load current, and vehicle body vibration; and V2X communication units (4G / 5G modules and WiFi / Bluetooth modules) are used to encrypt and upload multi-source data to edge computing nodes or cloud servers. It should be noted that the sensors can be deployed by attaching vibration sensors to the cabin floor area and installing electromagnetic noise sensors at the four corners and the center area.
[0025] The data processing module is used to preprocess the data to generate a dataset containing noise characteristics and equipment operating conditions, including: an edge computing unit and a cloud server.
[0026] The edge computing unit is used to preprocess the original signal, including the time-frequency domain transformation of the noise signal (FFT), normalization of the running parameters, and to generate a dataset containing features such as noise amplitude, frequency distribution, and device load. The cloud server is used to store historical data and build an electromagnetic noise feature library, supporting multi-device data correlation analysis.
[0027] The machine learning module is used to generate dynamic noise reduction instructions based on the dataset, including a noise prediction model and a noise reduction strategy generator.
[0028] The noise prediction model, based on LSTM (Long Short-Term Memory) or Convolutional Neural Network (CNN), takes historical noise data and equipment operating parameters as input to predict the noise distribution trend under the current operating conditions. The noise reduction strategy generator uses reinforcement learning (RL) algorithms to optimize noise amplitude to be lower than the industry standard, and outputs absorbing material layout adjustment schemes (such as the rotation angle and coverage position of absorbing materials (ferrite sheets)) or equipment parameter adjustment commands (such as the carrier frequency offset of frequency converter equipment and the duty cycle PWM (Pulse Width Modulation) of power modules).
[0029] The execution module is used to adjust the layout of the absorbing material or the operating parameters of the equipment according to the noise reduction command, including: a dynamic adjustment unit for the absorbing material and a parameter adjustment unit for the equipment.
[0030] The microwave absorbing material dynamic adjustment unit includes movable microwave absorbing sheets (such as ferrite sheets or graphene microwave absorbing coatings) and their driving mechanisms (servo motors or shape memory alloys), which adjust the coverage area, angle, or thickness of the microwave absorbing material according to instructions; the equipment parameter adjustment unit connects to the target equipment controller via an industrial bus to modify operating parameters (such as motor drive frequency and switching power supply pulse width) in real time.
[0031] In addition, the noise reduction system may also include a human-computer interaction module, which includes a visual interface and a threshold setting module.
[0032] The visualization interface is used to display noise spectrum, equipment operating status, and noise reduction strategy execution effect in real time; the threshold setting module is used to support users to customize noise warning thresholds and noise reduction priorities (such as safety priority mode and energy efficiency priority mode).
[0033] Figure 2 This is a schematic flowchart of a vehicle noise reduction method provided in an embodiment of this application.
[0034] like Figure 2 As shown, the noise reduction method for this vehicle includes the following steps: In step S101, noise data of the vehicle and operating data of the target equipment are acquired.
[0035] The noise data consists of electromagnetic radiation signals in the 0.15MHz–6GHz frequency band collected by the vehicle-mounted electromagnetic noise sensor, reflecting the frequency domain distribution of the vehicle's electromagnetic interference intensity. The target devices are active electronic / electrical devices in the vehicle that may generate or be affected by electromagnetic interference, such as motor controllers, DC-DC converters, on-board chargers, and inverters. The operating data of the target devices consists of real-time operating parameters collected by current / voltage transmitters, vibration sensors, speed encoders, etc., including load current, bus voltage, motor speed, PWM switching frequency, and mechanical vibration intensity.
[0036] Specifically, in this embodiment, the sensor module synchronously collects electromagnetic noise signals and equipment operating parameters at a preset sampling rate. Preferably, the preset sampling rate is greater than or equal to 10kHz, and the collected data is uploaded to the data processing module through the V2X communication unit.
[0037] In step S102, noise data and running data are processed to obtain multidimensional feature vectors. The multidimensional feature vectors are then input into a preset noise prediction model to obtain noise distribution information. Based on the noise distribution information, it is determined whether the vehicle meets the preset noise reduction conditions.
[0038] Furthermore, in some embodiments, the noise distribution information includes the predicted noise amplitude at each frequency point, and the preset noise reduction condition is that the predicted noise amplitude at any frequency point exceeds the corresponding noise threshold.
[0039] Among them, the noise distribution information is the predicted noise amplitude of each frequency point output by the preset noise prediction module, and the noise threshold is the maximum allowable noise amplitude limit of each frequency point.
[0040] Specifically, the edge computing unit performs a Fast Fourier Transform (FFT) on the synchronously acquired noise data to extract frequency domain features and fuses them with the running data to generate a multi-dimensional feature vector. This vector is input into a preset noise prediction model and outputs noise distribution information, i.e., the predicted noise amplitude at each frequency point. The system determines whether the predicted noise amplitude at any frequency point exceeds its corresponding noise threshold. If so, it determines that the vehicle meets the preset noise reduction conditions and needs to trigger the generation of a noise reduction strategy.
[0041] In actual implementation, the embodiments of this application perform time-frequency analysis on the noise signal, extract features such as peak frequency, harmonic components, and noise energy, and construct a multi-dimensional feature vector with parameters such as equipment load and temperature. The pre-built noise reduction prediction model is used to predict the noise spectrum under different loads. When the predicted noise amplitude exceeds the noise threshold, the noise reduction strategy is triggered, which improves the accuracy and timeliness of noise reduction triggering.
[0042] Optionally, in some embodiments, before inputting the multidimensional feature vector into a preset noise prediction model to obtain noise distribution information, the method includes: acquiring historical feature vectors and constructing a dataset based on the historical feature vectors; dividing the dataset into a training set, a validation set, and a test set based on a preset partitioning ratio; constructing a target neural network, inputting the training set into the target neural network for training to obtain initial model parameters; based on the initial model parameters, inputting the validation set into the target neural network for performance evaluation, and adjusting the initial model parameters according to the performance evaluation results until the joint loss function of the validation set converges to obtain optimal model parameters; based on the optimal model parameters, inputting the test set into the target neural network for model testing, and obtaining a preset noise prediction model when the test results meet preset requirements.
[0043] Among them, the historical feature vector is a multi-dimensional feature vector generated by synchronously collecting historical noise data of vehicles and historical operating data of target equipment, and then performing time-frequency analysis and fusion processing; the target neural network is a machine learning network structure used to build a noise prediction model, including LSTM or CNN.
[0044] Specifically, before inputting real-time multidimensional feature vectors into the preset noise prediction model, the system first generates historical feature vectors based on historically collected noise data and equipment operation data, and constructs a dataset containing feature vectors and corresponding noise distribution labels. Then, the dataset is divided into training, validation, and test sets according to a preset ratio. Next, a target neural network (such as LSTM or CNN) is constructed, and the network is trained using the training set to obtain initial model parameters. The validation set is then input into the network to calculate the joint loss function and evaluate performance, iteratively adjusting the model parameters until the validation loss converges. Finally, the converged model is tested using the test set. If the test results meet the preset accuracy or error requirements, the model is determined as the preset noise prediction model for subsequent online prediction.
[0045] In step S103, if the vehicle meets the preset noise reduction conditions, the noise distribution information is input to the preset noise reduction model to obtain the adjustment instructions of the preset noise reduction device and the adjustment instructions of the target device. The preset noise reduction device is controlled to adjust based on the adjustment instructions, and the target device is controlled to adjust based on the adjustment instructions.
[0046] Furthermore, in some embodiments, the state space of the preset noise reduction model includes at least one of noise distribution information and operating parameters of the target device; the action space includes at least one of the pose adjustment amount of the preset noise reduction device and the operating parameter adjustment amount of the target device.
[0047] The preset noise reduction device is a reconfigurable electromagnetic absorbing structure, including a movable absorbing plate and its driving mechanism; the adjustment command is used to control the operation of the preset noise reduction device, such as the rotation angle or displacement of the absorbing plate; the adjustment command is used to control the operating parameters of the target equipment, such as the carrier frequency offset of the frequency converter and the PWM duty cycle of the power module.
[0048] Specifically, when the system determines that the vehicle meets the preset noise reduction conditions, it sends the noise distribution information and the operating parameters of the target equipment together as the state input to the preset noise reduction model. This preset noise reduction model can be trained using reinforcement learning. Its state space includes the current noise characteristics and equipment parameters, while its action space includes the adjustment range of the absorbing material and the adjustment step size of the equipment parameters. The reward function is defined as the weighted value of the noise amplitude decrease rate and energy consumption change. Based on its state space, the preset noise reduction model performs policy reasoning and outputs specific adjustment and control commands from its action space. Subsequently, the system controls the preset noise reduction device to change its spatial layout according to the adjustment commands (e.g., rotating the absorbing sheet), while simultaneously controlling the target equipment to modify its operating parameters according to the adjustment commands (e.g., adjusting the PWM duty cycle), achieving coordinated noise reduction.
[0049] Optionally, in some embodiments, after controlling the preset noise reduction device to adjust based on the adjustment command and controlling the target device to adjust based on the adjustment command, the method includes: acquiring new noise data, processing the new noise data to obtain a noise spectrum diagram, and visualizing the noise spectrum diagram.
[0050] The new noise data consists of electromagnetic radiation signals in the 0.15MHz–6GHz frequency band, which were re-collected by the electromagnetic noise sensor after the preset noise reduction device and target equipment were adjusted. The noise spectrum diagram is a graph generated after performing a fast Fourier transform on the new noise data, with the horizontal axis representing frequency and the vertical axis representing the noise amplitude at each frequency point.
[0051] Specifically, after the preset noise reduction device completes the pose adjustment based on the adjustment command and the target device completes the parameter modification based on the adjustment command, the system restarts the electromagnetic noise sensor to collect new noise data after the noise reduction is performed; the edge computing unit performs a fast Fourier transform on the data, extracts the noise amplitude at each frequency point, and generates a noise spectrum; subsequently, the spectrum is visualized through the vehicle human-machine interface (such as the central control screen or diagnostic terminal), so that users or testers can intuitively view the spectrum changes before and after noise reduction and the current electromagnetic noise level.
[0052] In summary, as shown in Figure 3, this embodiment of the application synchronously collects electromagnetic noise signals and equipment operating parameters through a sensor module, transmits them to a data processing module via the Internet of Things, and performs time-frequency analysis and feature extraction on the collected signals to construct a multi-dimensional feature vector including noise amplitude, frequency, equipment load, etc. Then, a machine learning algorithm is used to train a noise prediction model and a noise reduction strategy generator to generate an optimal adjustment strategy. Finally, the execution module automatically rotates the absorbing sheet to the direction of strongest noise radiation according to the optimal adjustment strategy, reduces the load current output, and adjusts the vehicle body posture to reduce component noise and mechanical vibration coupling noise. Thus, by integrating electromagnetic noise signals and equipment operating parameters, a comprehensive monitoring system encompassing environmental characteristics and equipment operating conditions is constructed; the noise reduction strategy is autonomously optimized through machine learning algorithms, adapting to complex electromagnetic environments without manual intervention; and the combination of dynamic adjustment of absorbing materials and optimization of equipment parameters balances noise reduction effectiveness and equipment operating efficiency.
[0053] The vehicle noise reduction method proposed in this application involves acquiring vehicle noise data and target equipment operating data, processing the noise data and operating data to obtain a multi-dimensional feature vector, inputting the multi-dimensional feature vector into a preset noise prediction model to obtain noise distribution information, and determining whether the vehicle meets preset noise reduction conditions based on the noise distribution information. If the vehicle meets the preset noise reduction conditions, the noise distribution information is input into the preset noise reduction model to obtain adjustment instructions for the preset noise reduction device and adjustment instructions for the target equipment. The preset noise reduction device is controlled to adjust based on the adjustment instructions, and the target equipment is also controlled to adjust based on the adjustment instructions. This solves the problem that static noise reduction methods cannot perceive the dynamic coupling relationship between equipment operating conditions and the electromagnetic environment in real time and are difficult to cope with complex noise changes. This application can adapt to complex electromagnetic environments without manual intervention, balancing noise reduction effect and equipment operating efficiency.
[0054] Next, the noise reduction device for a vehicle according to an embodiment of this application is described with reference to the accompanying drawings.
[0055] Figure 4 This is a block diagram of a vehicle noise reduction device according to an embodiment of this application.
[0056] like Figure 4 As shown, the noise reduction device 10 of the vehicle includes: an acquisition module 100, a judgment module 200, and a control module 300.
[0057] The acquisition module 100 is used to acquire noise data of the vehicle and operating data of the target equipment.
[0058] The judgment module 200 is used to process noise data and running data to obtain multi-dimensional feature vectors, input the multi-dimensional feature vectors into a preset noise prediction model to obtain noise distribution information, and judge whether the vehicle meets the preset noise reduction conditions based on the noise distribution information.
[0059] The control module 300 is used to input noise distribution information into a preset noise reduction model to obtain adjustment instructions for the preset noise reduction device and the target device when the vehicle meets the preset noise reduction conditions, and to control the preset noise reduction device to adjust based on the adjustment instructions and control the target device to adjust based on the adjustment instructions.
[0060] Optionally, in some embodiments, the noise distribution information includes the predicted noise amplitude at each frequency point, and the preset noise reduction condition is that the predicted noise amplitude at any frequency point exceeds the corresponding noise threshold.
[0061] Optionally, in some embodiments, before inputting the multidimensional feature vector into a preset noise prediction model to obtain noise distribution information, the judgment module 200 includes: an acquisition unit, a partitioning unit, a training unit, an optimization unit, and a testing unit.
[0062] The acquisition unit is used to acquire historical feature vectors and construct a dataset based on the historical feature vectors.
[0063] The partitioning unit is used to divide the dataset into training, validation, and test sets based on a preset partitioning ratio.
[0064] The training unit is used to build the target neural network. It is obtained by inputting the training set into the target neural network for training and obtaining the initial model parameters.
[0065] The optimization unit is used to input the validation set into the target neural network for performance evaluation based on the initial model parameters, and adjust the initial model parameters according to the performance evaluation results until the joint loss function of the validation set converges to obtain the optimal model parameters.
[0066] The testing unit is used to input the test set into the target neural network based on the optimal model parameters to test the model, and obtain the preset noise prediction model when the test results meet the preset requirements.
[0067] Optionally, in some embodiments, the state space of the preset noise reduction model includes at least one of noise distribution information and operating parameters of the target device; the action space includes at least one of the pose adjustment amount of the preset noise reduction device and the operating parameter adjustment amount of the target device.
[0068] Optionally, in some embodiments, after controlling the preset noise reduction device to adjust based on the adjustment command and controlling the target device to adjust based on the adjustment command, the control module includes: a visualization unit.
[0069] The visualization unit is used to acquire new noise data, process the new noise data to obtain a noise spectrum, and visualize the noise spectrum.
[0070] It should be noted that the explanation of the above-described vehicle noise reduction method embodiment also applies to the vehicle noise reduction device of this embodiment, and will not be repeated here.
[0071] The vehicle noise reduction device proposed in this application acquires vehicle noise data and target equipment operating data, processes the noise data and operating data to obtain a multi-dimensional feature vector, inputs the multi-dimensional feature vector into a preset noise prediction model to obtain noise distribution information, and determines whether the vehicle meets preset noise reduction conditions based on the noise distribution information. If the vehicle meets the preset noise reduction conditions, the noise distribution information is input into the preset noise reduction model to obtain a preset adjustment command for the noise reduction device and an adjustment command for the target equipment. The preset noise reduction device is controlled to adjust based on the adjustment command, and the target equipment is controlled to adjust based on the adjustment command. This solves the problem that static noise reduction methods cannot perceive the dynamic coupling relationship between equipment operating conditions and the electromagnetic environment in real time and are difficult to cope with complex noise changes. This application can adapt to complex electromagnetic environments without manual intervention, balancing noise reduction effect and equipment operating efficiency.
[0072] Figure 5 A schematic diagram of the structure of a vehicle provided in an embodiment of this application. The vehicle may include: The memory 501, the processor 502, and the computer program stored on the memory 501 and capable of running on the processor 502.
[0073] When processor 502 executes the program, it implements the vehicle noise reduction method provided in the above embodiments.
[0074] Furthermore, the vehicle also includes: Communication interface 503 is used for communication between memory 501 and processor 502.
[0075] The memory 501 is used to store computer programs that can run on the processor 502.
[0076] The memory 501 may include high-speed RAM (Random Access Memory) memory, and may also include non-volatile memory, such as at least one disk storage.
[0077] If the memory 501, processor 502, and communication interface 503 are implemented independently, then the communication interface 503, memory 501, and processor 502 can be interconnected via a bus to complete communication between them. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 5 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.
[0078] Optionally, in a specific implementation, if the memory 501, processor 502, and communication interface 503 are integrated on a single chip, then the memory 501, processor 502, and communication interface 503 can communicate with each other through an internal interface.
[0079] The processor 502 may be a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of this application.
[0080] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described vehicle noise reduction method.
[0081] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.
[0082] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.
[0083] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.
[0084] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (FPGAs), field-programmable gate arrays (FPGAs), etc.
[0085] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it includes one or a combination of the steps of the method embodiments.
[0086] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.
Claims
1. A method for reducing noise in a vehicle, characterized in that, Includes the following steps: Acquire noise data from the vehicle and operational data from the target equipment; The noise data and the running data are processed to obtain a multi-dimensional feature vector. The multi-dimensional feature vector is then input into a preset noise prediction model to obtain noise distribution information. Based on the noise distribution information, it is determined whether the vehicle meets the preset noise reduction conditions. If the vehicle meets the preset noise reduction conditions, the noise distribution information is input into the preset noise reduction model to obtain the adjustment instructions for the preset noise reduction device and the adjustment instructions for the target device. The preset noise reduction device is controlled to adjust based on the adjustment instructions, and the target device is controlled to adjust based on the adjustment instructions.
2. The method according to claim 1, characterized in that, The noise distribution information includes the predicted noise amplitude at each frequency point, and the preset noise reduction condition is that the predicted noise amplitude at any frequency point exceeds the corresponding noise threshold.
3. The method according to claim 1, characterized in that, Before inputting the multidimensional feature vector into the preset noise prediction model to obtain the noise distribution information, the process includes: Obtain historical feature vectors and construct a dataset based on the historical feature vectors; Based on a preset partitioning ratio, the dataset is divided into a training set, a validation set, and a test set. Construct a target neural network by inputting the training set into the target neural network for training to obtain initial model parameters; Based on the initial model parameters, the validation set is input into the target neural network for performance evaluation, and the initial model parameters are adjusted according to the performance evaluation results until the joint loss function of the validation set converges to obtain the optimal model parameters. Based on the optimal model parameters, the test set is input into the target neural network for model testing, and when the test results meet the preset requirements, the preset noise prediction model is obtained.
4. The method according to claim 1, characterized in that, The state space of the preset noise reduction model includes at least one of the noise distribution information and the operating parameters of the target device; The motion space includes at least one of the following: the pose adjustment amount of the preset noise reduction device and the operating parameter adjustment amount of the target device.
5. The method according to claim 1, characterized in that, After controlling the preset noise reduction device to adjust based on the adjustment command and controlling the target device to adjust based on the adjustment command, the process includes: New noise data is acquired, and the new noise data is processed to obtain a noise spectrum diagram, which is then visualized.
6. A noise reduction device for a vehicle, characterized in that, include: The acquisition module is used to acquire noise data from the vehicle and operational data from the target equipment. The judgment module is used to process the noise data and the running data to obtain a multi-dimensional feature vector, input the multi-dimensional feature vector into a preset noise prediction model to obtain noise distribution information, and judge whether the vehicle meets the preset noise reduction conditions based on the noise distribution information. The control module is used to input the noise distribution information into a preset noise reduction model to obtain adjustment instructions for a preset noise reduction device and adjustment instructions for the target device when the vehicle meets the preset noise reduction conditions, and to control the preset noise reduction device to adjust based on the adjustment instructions and control the target device to adjust based on the adjustment instructions.
7. The apparatus according to claim 6, characterized in that, The noise distribution information includes the predicted noise amplitude at each frequency point, and the preset noise reduction condition is that the predicted noise amplitude at any frequency point exceeds the corresponding noise threshold.
8. The apparatus according to claim 6, characterized in that, Before inputting the multidimensional feature vector into the preset noise prediction model to obtain the noise distribution information, the judgment module includes: An acquisition unit is used to acquire historical feature vectors and construct a dataset based on the historical feature vectors; A partitioning unit is used to divide the dataset into a training set, a validation set, and a test set based on a preset partitioning ratio. A training unit is used to construct a target neural network, and to obtain initial model parameters by inputting the training set into the target neural network for training. An optimization unit is used to input the validation set into the target neural network for performance evaluation based on the initial model parameters, and adjust the initial model parameters according to the performance evaluation results until the joint loss function of the validation set converges to obtain the optimal model parameters. The testing unit is used to input the test set into the target neural network based on the optimal model parameters to test the model, and obtain the preset noise prediction model when the test results meet the preset requirements.
9. A vehicle, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the noise reduction method for a vehicle as described in any one of claims 1-5.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the noise reduction method for the vehicle as described in any one of claims 1-5.