Active noise control method and system for new energy automobile

Abstract

The invention discloses an active noise control method and system for a new energy automobile, belongs to the technical field of automobile noise reduction, and is used for solving the problems that an existing active noise control technology is poor in stability, influences of road conditions on cavity noise are not considered, tire noise cannot be predicted and restrained in advance according to the front road conditions, and the noise is affected. And the noise reduction effect is not ideal. The method comprises the following steps: constructing a target data set based on noise signal features and road surface point cloud data; training a noise signal prediction model through the target data set; acquiring real-time point cloud data of a road surface in front of the vehicle, and inputting the real-time point cloud data into the noise signal prediction model to obtain noise signal prediction features; according to the noise signal prediction features, noise pre-adjustment signals are generated, and vehicle noise is preliminarily suppressed; and inputting the real-time noise signal obtained after preliminary suppression into an adaptive filter, and compensating the noise pre-adjusting signal to further suppress the vehicle noise.

Description

Technical Field

[0001] The present invention relates to the technical field of automotive noise reduction, and particularly to an active noise control method and system for new energy vehicles. Background Art

[0002] Currently, most new energy vehicles are driven by electric motors. Due to the lack of engine masking, tire cavity resonance noise, road surface noise, and wind noise are more likely to reduce the ride comfort of the vehicle. Tire cavity resonance noise is generated by the resonance of the air inside the tire cavity excited by road surface unevenness, and this kind of noise belongs to low-frequency noise, mainly distributed in the range of 30 - 300 Hz. This low-frequency noise is likely to make people irritable, anxious and uncomfortable, and extremely easy to affect the driving experience of the driver. Therefore, in the field of new energy vehicles, how to suppress tire cavity resonance noise has become an important issue.

[0003] With the progress of technology, current vehicle noise control methods are mainly divided into active control technology and passive control technology. The principle of active control technology is "canceling sound with sound". By installing specific sensor and speaker systems, the vehicle can monitor and analyze the frequency and phase of the noise in real time, and then generate sound waves opposite to it to achieve the effect of reducing or eliminating the noise. Passive control technology is to use sound insulation materials or sound absorption materials at key parts of the whole vehicle in the engineering prototype stage to weaken the noise. Currently, passive control technology is widely applied, but the noise suppression effect is not ideal. And active control technology has less actual application due to poor stability and low accuracy. Moreover, the vehicle cavity noise is greatly affected by road conditions, and the noise generated by different road conditions is very different. The existing active control technology only focuses on suppressing the actual tire noise signal generated at the current moment, but does not consider the influence of road conditions on tire noise, and cannot predict in advance the tire noise that may be generated according to the road conditions ahead and generate a matching noise suppression signal in advance. Summary of the Invention

[0004] The embodiments of the present invention provide an active noise control method and system for new energy vehicles to solve the following technical problems: the existing active noise control technology has poor stability, does not consider the influence of road conditions on cavity noise, cannot predict and suppress tire noise in advance according to the road conditions ahead, and the noise reduction effect is not ideal.

[0005] The embodiments of the present invention adopt the following technical solutions:

[0006] On the one hand, the embodiments of the present invention provide an active noise control method for new energy vehicles, and the method includes: conducting a driving test on a target vehicle model, and simultaneously collecting noise signals and road point cloud data in front of the vehicle;

[0007] Extract the time-frequency domain features of the noise signal to obtain the noise signal features; and construct a target data set based on the noise signal features and the road surface point cloud data; train a noise signal prediction model through the target data set;

[0008] During the driving process of the target vehicle model, obtain the real-time road surface point cloud data in front of the vehicle, and perform quality evaluation on the real-time road surface point cloud data to obtain a road surface quality evaluation value;

[0009] When the road surface quality evaluation value is lower than the first preset threshold, input the real-time road surface point cloud data into the noise signal prediction model to obtain noise signal prediction features;

[0010] Generate a noise pre-adjustment signal according to the noise signal prediction features and perform preliminary suppression on the vehicle noise;

[0011] Input the real-time noise signal obtained after preliminary suppression into an adaptive filter to compensate the noise pre-adjustment signal, so as to further suppress the vehicle noise.

[0012] In a feasible implementation manner, conduct a driving test on the target vehicle model, and simultaneously collect noise signals and the road surface point cloud data in front of the vehicle, specifically including:

[0013] Install acceleration sensors at several positions of the target vehicle model, and install error microphones at the target noise reduction point positions;

[0014] Conduct a vibration excitation test when the target vehicle model is in a stationary state, and obtain the vibration acceleration signal at the installation position through the acceleration sensor, and obtain the sound signal at the target noise reduction point position through the error microphone;

[0015] Conduct coherence analysis on the vibration acceleration signals collected at each position to determine a preset number of recommended installation positions of the acceleration sensors, and remove the acceleration sensors installed at non-recommended installation positions;

[0016] Conduct vehicle driving tests at different driving speeds on road surfaces of different qualities, and simultaneously collect driving vibration acceleration signals through the acceleration sensors installed at the recommended installation positions, obtain driving sound signals at the target noise reduction point positions through the error microphones, and obtain the road surface point cloud data within a preset distance in front of the vehicle through the lidar installed at the front of the vehicle;

[0017] Use the driving vibration acceleration signal as the reference signal and the driving sound signal as the error signal to form the noise signal.

[0018] In a feasible implementation, coherence analysis is performed on the vibration acceleration signals collected at each position to determine a preset number of recommended installation positions for the acceleration sensors, specifically including:

[0019] Calculate the constant coherence function values of the vibration acceleration signals collected at each position and the corresponding sound signals, and screen out the target acceleration sensors with constant coherence function values greater than the second preset threshold;

[0020] Based on the preset number, perform permutations and combinations on the target acceleration sensors, and calculate the re-coherence function values of the vibration acceleration signals in each combination;

[0021] Determine the installation positions of the target acceleration sensors corresponding to the vibration acceleration signals in the combination with the largest re-coherence function value as the recommended installation positions.

[0022] In a feasible implementation, time-frequency domain feature extraction is performed on the noise signal to obtain noise signal features; and based on the noise signal features and the road surface point cloud data, a target data set is constructed, specifically including:

[0023] Based on a preset feature extraction algorithm, perform time-frequency domain feature extraction on the noise signal to obtain corresponding noise signal features;

[0024] Based on the vehicle driving speed, calculate the delay time between the noise signal collected at the same road surface position and the road surface point cloud data;

[0025] Based on the delay time, one-to-one correspondence is performed between the noise signal features collected at the same road surface position and the road surface point cloud data to obtain a number of data groups, constituting the target data set.

[0026] In a feasible implementation, based on a preset feature extraction algorithm, time-frequency domain feature extraction is performed on the noise signal to obtain corresponding noise signal features, specifically including:

[0027] Perform continuous wavelet transform on the reference signal and the error signal in the noise signal respectively to obtain the time-frequency distribution matrix of each signal;

[0028] Perform singular value decomposition on the time-frequency distribution matrix to obtain the singular feature values corresponding to each signal;

[0029] Based on the singular feature values, calculate the between-class scatter and within-class scatter of each type of signal, and determine the separability threshold according to the between-class scatter and within-class scatter;

[0030] Based on the separability threshold, evaluate and screen the singular feature values of each signal to obtain a singular feature subset, constituting the noise signal features.

[0031] In a feasible implementation, before training the noise signal prediction model using the target data set, the method further includes:

[0032] Add a causal convolutional layer and a dilated convolutional layer between the input layer and the output layer of the convolutional neural network. Each convolutional layer is connected by a ReLU activation function, and at the output end of the last convolutional layer, it is added to the original input through a skip connection to form a residual link, constituting a local feature extraction module. Among them, the causal convolutional layer is used to extract the causal relationship between input features, and the dilated convolutional layer is used to increase the receptive field of the convolutional kernel to capture the long-range dependence relationship between features;

[0033] Connect the output end of the local feature extraction module to the input end of the Transformer encoder to construct the noise signal prediction model.

[0034] In a feasible implementation, during the driving process of the target vehicle model, obtain the real-time point cloud data of the road surface in front of the vehicle, and perform a quality assessment on the real-time point cloud data of the road surface to obtain a road surface quality assessment value, specifically including:

[0035] During the actual driving process of the target vehicle model, obtain the real-time point cloud data of the road surface within a preset distance in front of the vehicle through a lidar installed in the front of the vehicle;

[0036] Based on the real-time point cloud data of the road surface, determine the quality parameters of the road surface within a preset distance in front. Among them, the quality parameters at least include any one or more of the following: road surface material, road surface flatness, road surface roughness, road surface damage condition, and road surface anti-skid performance;

[0037] Perform weight assignment on the quality parameters and perform weighted calculation to obtain the road surface quality assessment value.

[0038] In a feasible implementation, generate a noise pre-adjustment signal based on the noise signal prediction feature and perform preliminary suppression on the vehicle noise, specifically including:

[0039] Extract the frequency information and phase information of the predicted noise signal according to the noise signal prediction feature;

[0040] Based on the frequency information and phase information, generate a noise pre-adjustment signal with the same frequency and opposite phase as the predicted noise signal;

[0041] Based on the current driving speed of the vehicle, calculate the delay time required for the vehicle to travel from the current position to the road surface within a preset distance in front;

[0042] After the elapse of the delay time, the noise pre-adjustment signal is played through the in-vehicle speaker system to perform preliminary suppression on the vehicle noise.

[0043] In a feasible implementation manner, the real-time noise signal obtained after preliminary suppression is input into an adaptive filter to compensate the noise pre-adjustment signal for further suppressing the vehicle noise, specifically including:

[0044] Obtain the real-time noise signal after preliminary suppression through a noise signal acquisition device installed in the target vehicle model;

[0045] Input the real-time noise signal into an LMS adaptive filter, output a control signal with the opposite phase, and superimpose it with the noise pre-adjustment signal and play it through the in-vehicle speaker system to further suppress the vehicle noise.

[0046] On the other hand, the embodiment of the present invention also provides a noise active control system for a new energy vehicle, and the system includes:

[0047] A model training module, configured to conduct a driving test on the target vehicle model, and simultaneously collect noise signals and road surface point cloud data in front of the vehicle; extract time-frequency domain features of the noise signals to obtain noise signal features; and construct a target data set based on the noise signal features and the road surface point cloud data; train a noise signal prediction model through the target data set;

[0048] A noise pre-adjustment module, configured to obtain real-time road surface point cloud data in front of the vehicle during the driving of the target vehicle model, and perform quality evaluation on the real-time road surface point cloud data to obtain a road surface quality evaluation value; in the case where the road surface quality evaluation value is lower than a first preset threshold, input the real-time road surface point cloud data into the noise signal prediction model to obtain noise signal prediction features; generate a noise pre-adjustment signal according to the noise signal prediction features and perform preliminary suppression on the vehicle noise;

[0049] A noise real-time suppression module, configured to input the real-time noise signal obtained after preliminary suppression into an adaptive filter to compensate the noise pre-adjustment signal for further suppressing the vehicle noise.

[0050] Compared with the prior art, the noise active control method and system for a new energy vehicle provided by the embodiment of the present invention have the following beneficial effects:

[0051] Through driving tests on the target vehicle model, the present invention obtains road condition data and noise signals, and through a self-designed convolutional neural network model, learns the correlation between the road condition data and the noise signals, so as to predict in advance the characteristics of the noise signals that may be generated after the vehicle travels to a certain position according to the road conditions ahead during the vehicle driving process, and thus generate suppression signals in advance, which can make up for the short delay caused by the operation of the existing noise active control algorithm, avoid the situation of signal suppression lag, and can perform different noise suppression according to different road conditions, improve the noise suppression effect, and make the noise suppression feeling smoother.

[0052] After the advance suppression, the present invention further suppresses the real-time noise signal after the preliminary suppression through an adaptive filter, which can further improve the noise suppression effect on the basis of the preliminary suppression, and can reduce the noise decibel to a smaller range. According to the experimental verification results, the algorithm proposed by the present invention has a better noise suppression effect than the existing active control algorithm. Description of the Drawings

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments recorded in the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to these drawings. In the drawings:

[0054] Figure 1 It is a flowchart of a method for active noise control of a new energy vehicle provided by an embodiment of the present invention;

[0055] Figure 2 It is a schematic structural diagram of an active noise control system for a new energy vehicle provided by an embodiment of the present invention. Detailed Embodiments

[0056] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0057] The embodiment of the present invention provides a method for active noise control of a new energy vehicle, as Figure 1 shown, the method for active noise control of a new energy vehicle specifically includes steps S101-S106:

[0058] S101. Conduct a driving test on the target vehicle model, and simultaneously collect the noise signal and the road surface point cloud data in front of the vehicle.

[0059] Specifically, in the vehicle model R & D stage, first conduct a vibration excitation test on the target vehicle model in a stationary state. The test process is as follows:

[0060] Install acceleration sensors at several positions of the target vehicle model, and install error microphones at the target noise reduction point positions. The target noise reduction points can be set at the positions of the driver's ears. Then arrange the vibration exciter at the positions of the target vehicle model where low-frequency noise is likely to be generated, such as at the wheel hubs, engine, chassis, etc., and conduct a vibration excitation test.

[0061] Then obtain the vibration acceleration signals at the installation positions through the acceleration sensors, and obtain the sound signals at the target noise reduction point positions through the error microphones. Then conduct a coherence analysis on the vibration acceleration signals collected at each position to determine a preset number of recommended installation positions of the acceleration sensors, and remove the acceleration sensors installed at non-recommended installation positions.

[0062] As a feasible implementation method, the specific process of obtaining the recommended installation positions is as follows: First, calculate the constant coherence function values of the vibration acceleration collected at each position and the corresponding sound signals, and screen out the target acceleration sensors with constant coherence function values greater than the second preset threshold. Then, based on the preset number, perform permutations and combinations on the target acceleration sensors, and calculate the re-coherence function values of the vibration acceleration signals in each combination. Determine the installation positions of the target acceleration sensors corresponding to the vibration acceleration signals in the combination with the largest re-coherence function value as the recommended installation positions.

[0063] In one embodiment, if 20 acceleration sensors are arranged on the target vehicle model, then through the constant coherence function, calculate the constant coherence function values of the vibration acceleration signals collected by the 20 acceleration sensors and the sound signals collected by the microphone respectively, and pre-select 15 installation positions with constant coherence function values greater than a specific threshold. Then, according to the actual requirements, set the number of acceleration sensors to be installed in the final vehicle model (i.e., the preset number) to 4. Then, for the selected 15 installation positions, Conduct combinations to obtain 1365 combinations. Calculate the re-coherence functions of these combinations respectively, and determine the 4 positions in the combination with the largest value as the recommended installation positions.

[0064] Further, after the static test, vehicle driving tests are carried out on roads with different qualities at different driving speeds. Meanwhile, the driving vibration acceleration signals are collected through the acceleration sensors installed at the recommended installation positions, the driving sound signals at the target noise reduction point positions are obtained through the error microphones, and the road point cloud data within a preset distance in front of the vehicle is obtained through the lidar installed at the front of the vehicle. The driving vibration acceleration signals are used as the reference signals, and the driving sound signals are used as the error signals to form the noise signals.

[0065] S102. Extract the time-frequency domain features of the noise signals to obtain the noise signal features; and construct a target data set based on the noise signal features and the road point cloud data.

[0066] Specifically, based on a preset feature extraction algorithm, the time-frequency domain features of the noise signals are extracted to obtain the corresponding noise signal features.

[0067] As a feasible implementation manner, the preset feature extraction algorithm is as follows: perform continuous wavelet transform on the reference signal and the error signal in the noise signal respectively to obtain the time-frequency distribution matrix of each signal. Then perform singular value decomposition on the time-frequency distribution matrix to obtain the singular eigenvalues corresponding to each signal. Further, based on the singular eigenvalues, calculate the between-class scatter and the within-class scatter of each type of signal, and calculate the separability threshold according to the between-class scatter and the within-class scatter. Screen the singular eigenvalues greater than the separability threshold to obtain the singular feature subset, which constitutes the noise signal features.

[0068] Further, based on the vehicle driving speed, calculate the delay time between the noise signals collected at the same road position and the road point cloud data. Based on this delay time, the noise signal features collected at the same road position are corresponded one by one with the road point cloud data to obtain a number of data groups, which constitute the target data set.

[0069] In one embodiment, if according to the vehicle driving speed, it is calculated that the time required for the vehicle to drive from the current position to the lidar scanning position is 1 second, then the noise signal features collected at the current moment are corresponded with the road point cloud data collected 1 second ago to form a data group, which is stored in the target data set.

[0070] S103. Train the noise signal prediction model through the target data set.

[0071] Specifically, a causal convolutional layer and a dilated convolutional layer are added between the input layer and the output layer of the convolutional neural network. Each convolutional layer is connected by a ReLU activation function, and at the output end of the last convolutional layer, a skip connection is used to add it to the original input to form a residual link, thus constituting a local feature extraction module. Among them, the causal convolutional layer is used to extract the causal relationship between input features, and the dilated convolutional layer is used to increase the receptive field of the convolutional kernel and capture the long-range dependence relationship between features.

[0072] Furthermore, the output end of the local feature extraction module is connected to the input end of the Transformer encoder to construct a noise signal prediction model. Then, through the target data set, the above noise signal prediction model is trained until the model converges.

[0073] The present invention adds a causal convolutional layer and a dilated convolutional layer on the basis of the convolutional neural network. The convolutional layer captures the local features in the input sequence to generate a feature map, and then through the non-linear activation function between the convolutional layers, multiple convolutional layers are stacked to enhance the model complexity. The causal convolutional layer is a unidirectional structure, and each time step only considers the current and past information, avoiding the leakage of future information and ensuring that the model conforms to the causal relationship of real data. The dilated convolutional layer can increase the receptive field of the convolutional kernel and improve the ability of the model to capture long-range dependence relationships, while improving its computational efficiency without deepening the network hierarchy.

[0074] S104. During the driving process of the target vehicle model, obtain the real-time road surface point cloud data in front of the vehicle, and perform a quality assessment on the real-time road surface point cloud data to obtain a road surface quality assessment value.

[0075] Specifically, during the actual driving process of the target vehicle model, the real-time road surface point cloud data within a preset distance in front of the vehicle is obtained through a lidar installed in the front of the vehicle.

[0076] Based on the real-time road surface point cloud data, analyze the quality parameters of the road surface within the preset distance in front; among them, the quality parameters at least include any one or more of the following: road surface material, road surface flatness, road surface roughness, road surface damage condition, and road surface anti-skid performance.

[0077] Furthermore, weight assignment is performed on the quality parameters, and weighted calculation is carried out to obtain a road surface quality assessment value.

[0078] S105. In the case where the road surface quality assessment value is lower than the first preset threshold, input the real-time road surface point cloud data into the noise signal prediction model to obtain noise signal prediction features; according to the noise signal prediction features, generate a noise pre-adjustment signal and preliminarily suppress the vehicle noise.

[0079] Specifically, when the road surface quality evaluation value is lower than a specific threshold, it is preliminarily determined that the road surface ahead may cause large noise fluctuations when the vehicle passes by. Therefore, at this time, the real-time point cloud data of the road surface is input into the noise signal prediction model to obtain the noise signal prediction features.

[0080] Further, according to the noise signal prediction features, the frequency information and phase information of the predicted noise signal are extracted. Based on the frequency information and phase information, a noise pre-adjustment signal with the same frequency and opposite phase as the predicted noise signal is generated.

[0081] Further, based on the current driving speed of the vehicle, the delay time required for the vehicle to travel from the current position to the road surface within a preset distance ahead is calculated; after the delay time, the noise pre-adjustment signal is played through the in-vehicle speaker system to cancel the current noise signal, so as to preliminarily suppress the vehicle noise.

[0082] S106. Input the real-time noise signal obtained after preliminary suppression into the adaptive filter to compensate the noise pre-adjustment signal, so as to further suppress the vehicle noise.

[0083] Specifically, the real-time noise signal after preliminary suppression is obtained through the noise signal acquisition device installed in the target vehicle model.

[0084] Further, the real-time noise signal is input into the LMS adaptive filter, and a control signal with the opposite phase is output, which is superimposed on the noise pre-adjustment signal and played through the in-vehicle speaker system to further suppress the vehicle noise.

[0085] As a feasible implementation manner, after suppressing the current noise signal through the noise pre-adjustment signal, due to various reasons such as the prediction accuracy of the model, the suppression effect may not reach the ideal effect. Therefore, the present invention combines the existing adaptive filter to perform real-time analysis on the noise signal after preliminary suppression again, and generates a cancellation signal again to be superimposed on the currently played noise pre-adjustment signal, which can further suppress the noise signal on the basis of pre-adjustment and greatly reduce the noise decibel value in the driver's ear.

[0086] In addition, the embodiment of the present invention also provides a noise active control system for a new energy vehicle, as Figure 2 shown. The noise active control system 200 of the new energy vehicle specifically includes:

[0087] A model training module 210, configured to perform a driving test on the target vehicle model, and simultaneously collect the noise signal and the road surface point cloud data in front of the vehicle; extract the time-frequency domain features of the noise signal to obtain the noise signal features; and construct a target data set based on the noise signal features and the road surface point cloud data; train the noise signal prediction model through the target data set;

[0088] A noise pre - adjustment module 220, configured to obtain real - time road surface point cloud data in front of the vehicle during the driving of the target vehicle model, perform quality assessment on the real - time road surface point cloud data to obtain a road surface quality assessment value; in the case where the road surface quality assessment value is lower than a first preset threshold, input the real - time road surface point cloud data into the noise signal prediction model to obtain noise signal prediction features; generate a noise pre - adjustment signal according to the noise signal prediction features and perform preliminary suppression on the vehicle noise;

[0089] A noise real - time suppression module 230, configured to input the real - time noise signal obtained after preliminary suppression into an adaptive filter to compensate the noise pre - adjustment signal, so as to further suppress the vehicle noise.

[0090] Each embodiment in the present invention is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the embodiments of the device, equipment, and non - volatile computer storage medium, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiments.

[0091] The above - described specific embodiments of the present invention have been described. Additionally, the processes depicted in the drawings do not necessarily require the specific order or consecutive order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0092] The above are only the embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, various changes and modifications can be made to the embodiments of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the embodiments of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for active noise control of a new energy vehicle, characterized in that, The method includes: Conduct a driving test on the target vehicle model while collecting noise signals and road surface point cloud data in front of the vehicle; Extract time-frequency domain features from the noise signals to obtain noise signal features; and based on the noise signal features and the road surface point cloud data, construct a target data set; through the target data set, train a noise signal prediction model; During the driving process of the target vehicle model, obtain real-time road surface point cloud data in front of the vehicle, and conduct a quality assessment on the real-time road surface point cloud data to obtain a road surface quality assessment value; In the case where the road surface quality assessment value is lower than a first preset threshold, input the real-time road surface point cloud data into the noise signal prediction model to obtain noise signal prediction features; Generate a noise pre-regulation signal according to the noise signal prediction features and preliminarily suppress the vehicle noise; Input the real-time noise signal obtained after preliminary suppression into an adaptive filter to compensate the noise pre-regulation signal, so as to further suppress the vehicle noise.

2. The noise active control method for a new energy vehicle according to claim 1, characterized in that, Conduct a driving test on the target vehicle model while collecting noise signals and road surface point cloud data in front of the vehicle, specifically including: Install acceleration sensors at several positions of the target vehicle model, and install an error microphone at the target noise reduction point position; Conduct a vibration excitation test when the target vehicle model is in a stationary state, and obtain the vibration acceleration signal at the installation position through the acceleration sensor, and obtain the sound signal at the target noise reduction point position through the error microphone; Conduct a coherence analysis on the vibration acceleration signals collected at each position to determine a preset number of recommended installation positions of the acceleration sensors, and remove the acceleration sensors installed at non-recommended installation positions; Conduct a vehicle driving test on road surfaces of different qualities at different driving speeds. At the same time, collect driving vibration acceleration signals through the acceleration sensors installed at the recommended installation positions, obtain driving sound signals at the target noise reduction point position through the error microphone, and obtain road surface point cloud data within a preset distance in front of the vehicle through a lidar installed at the front of the vehicle; Use the driving vibration acceleration signal as a reference signal and the driving sound signal as an error signal to form the noise signal.

3. A noise active control method for a new energy vehicle according to claim 2, characterized in that, Conduct a coherence analysis on the vibration acceleration signals collected at each position to determine a preset number of recommended installation positions of the acceleration sensors, specifically including: Calculate the constant coherence function values of the vibration acceleration signals collected at each position and the corresponding sound signals, and screen out the target acceleration sensors with constant coherence function values greater than a second preset threshold; Based on the preset number, perform permutation and combination on the target acceleration sensors, and calculate the re-coherence function values of the vibration acceleration signals in each combination; Determine the installation positions of the target acceleration sensors corresponding to the vibration acceleration signals in the combination with the largest re-coherence function value as the recommended installation positions.

4. A method for active noise control of a new energy vehicle according to claim 1, characterized in that, Extract time-frequency domain features from the noise signals to obtain noise signal features; And based on the noise signal features and the road surface point cloud data, construct a target data set, specifically including: Based on a preset feature extraction algorithm, perform time-frequency domain feature extraction on the noise signal to obtain corresponding noise signal features; Based on the vehicle driving speed, calculate the delay time between the noise signal collected at the same road surface position and the road surface point cloud data; Based on the delay time, one-to-one correspond the noise signal features collected at the same road surface position with the road surface point cloud data to obtain a number of data groups, which constitute the target data set.

5. The noise active control method for a new energy vehicle according to claim 4, wherein Based on a preset feature extraction algorithm, perform time-frequency domain feature extraction on the noise signal to obtain corresponding noise signal features, specifically including: Perform continuous wavelet transform on the reference signal and the error signal in the noise signal respectively to obtain the time-frequency distribution matrix of each signal; Perform singular value decomposition on the time-frequency distribution matrix to obtain the singular eigenvalues corresponding to each signal; Based on the singular eigenvalues, calculate the between-class scatter and within-class scatter of each type of signal, and determine the separability threshold according to the between-class scatter and within-class scatter; Based on the separability threshold, evaluate and screen the singular eigenvalues of each signal to obtain a singular feature subset, which constitutes the noise signal features.

6. The noise active control method for a new energy vehicle according to claim 1, characterized in that, Before training the noise signal prediction model through the target data set, the method further includes: Add a causal convolutional layer and a dilated convolutional layer between the input layer and the output layer of the convolutional neural network. Each convolutional layer is connected by a ReLU activation function, and the output end of the last convolutional layer is added to the original input through a skip connection to form a residual link, constituting a local feature extraction module; wherein, the causal convolutional layer is used to extract the causal relationship between input features, and the dilated convolutional layer is used to increase the receptive field of the convolutional kernel and capture the long-range dependence relationship between features; Connect the output end of the local feature extraction module to the input end of the Transformer encoder to construct the noise signal prediction model.

7. A noise active control method for a new energy vehicle according to claim 1, characterized in that, During the driving process of the target vehicle model, obtain the real-time road surface point cloud data in front of the vehicle and perform quality evaluation on the real-time road surface point cloud data to obtain a road surface quality evaluation value, specifically including: During the actual driving process of the target vehicle model, obtain the real-time road surface point cloud data within a preset distance in front of the vehicle through the lidar installed at the front of the vehicle; Based on the real-time road surface point cloud data, determine the quality parameters of the road surface within a preset distance in front; wherein, the quality parameters at least include any one or more of the following: road surface material, road surface flatness, road surface roughness, road surface damage condition, and road surface anti-skid performance; Perform weight assignment on the quality parameters and perform weighted calculation to obtain the road surface quality evaluation value.

8. A noise active control method for a new energy vehicle according to claim 1, characterized in that, Generate a noise pre-adjustment signal according to the noise signal prediction features and perform preliminary suppression on the vehicle noise, specifically including: According to the noise signal prediction features, extract the frequency information and phase information of the predicted noise signal; Based on the frequency information and phase information, generate a noise pre-adjustment signal with the same frequency and opposite phase as the predicted noise signal; Based on the current driving speed of the vehicle, calculate the delay time required for the vehicle to travel from the current position to the road surface within a preset distance in front; After the delay time has elapsed, the noise pre-adjustment signal is played through the in-vehicle speaker system to preliminarily suppress vehicle noise.

9. The noise active control method for a new energy vehicle according to claim 1, characterized in that, The real-time noise signal obtained after preliminary suppression is input into an adaptive filter to compensate the noise pre-adjustment signal for further suppression of vehicle noise, specifically including: Obtain the real-time noise signal after preliminary suppression through the noise signal acquisition device installed in the target vehicle model; Input the real-time noise signal into the LMS adaptive filter, output a control signal with the opposite phase, and superimpose it with the noise pre-adjustment signal and play it through the in-vehicle speaker system to further suppress vehicle noise.

10. An active noise control system for a new energy vehicle, characterized in that, The system includes: A model training module, configured to conduct a driving test on the target vehicle model, and simultaneously collect noise signals and road surface point cloud data in front of the vehicle; extract time-frequency domain features of the noise signals to obtain noise signal features; and based on the noise signal features and the road surface point cloud data, construct a target data set; train a noise signal prediction model through the target data set; A noise pre-adjustment module, configured to obtain the real-time road surface point cloud data in front of the vehicle during the driving of the target vehicle model, and perform a quality assessment on the real-time road surface point cloud data to obtain a road surface quality assessment value; in the case where the road surface quality assessment value is lower than a first preset threshold, input the real-time road surface point cloud data into the noise signal prediction model to obtain noise signal prediction features; generate a noise pre-adjustment signal according to the noise signal prediction features and preliminarily suppress vehicle noise; A noise real-time suppression module, configured to input the real-time noise signal obtained after preliminary suppression into an adaptive filter to compensate the noise pre-adjustment signal for further suppression of vehicle noise.

Images