CN112161815A - Vehicle road noise subjective evaluation value prediction method - Google Patents
Vehicle road noise subjective evaluation value prediction method Download PDFInfo
- Publication number
- CN112161815A CN112161815A CN202010927866.2A CN202010927866A CN112161815A CN 112161815 A CN112161815 A CN 112161815A CN 202010927866 A CN202010927866 A CN 202010927866A CN 112161815 A CN112161815 A CN 112161815A
- Authority
- CN
- China
- Prior art keywords
- road noise
- vehicle
- data
- neural network
- subjective evaluation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M17/00—Testing of vehicles
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Abstract
The invention discloses a vehicle road noise subjective evaluation value prediction method, which comprises the following steps: 1) acquiring vehicle road noise test data of each sample and a corresponding subjective evaluation value as training sample data; 2) determining 1/s octave frequency spectrum data of the road noise of each sample vehicle according to the road noise test data of each sample vehicle; 3) establishing a BP neural network model; 4) setting an output error limit value of the BP neural network, training the BP neural network, and obtaining a neural network model for predicting a subjective evaluation value after training; 5) collecting road noise data of a vehicle to be evaluated; 6) determining 1/s octave frequency spectrum data of the road noise of the vehicle to be evaluated; 7) and obtaining the subjective evaluation predicted value of the road noise of the vehicle to be evaluated by using the trained neural network model. The method carries out neural network training by converting the road noise data into the vehicle road noise 1/s octave frequency spectrum data, and can predict the vehicle road noise subjective evaluation value through the road noise data.
Description
Technical Field
The invention relates to a vehicle road noise research technology, in particular to a vehicle road noise subjective evaluation value prediction method.
Background
The evaluation of the vehicle road noise level generally adopts objective evaluation and subjective evaluation. Objective evaluation, namely acquiring and analyzing vehicle road noise data to obtain road noise frequency spectrum data; and subjective evaluation, namely subjective evaluation of the vehicle road noise level by a professional. The objective evaluation plays an important role in the aspects of vehicle road noise data comparison, problem analysis and the like; subjective evaluation is indispensable, the quality of the vehicle road noise level is finally evaluated by a user, and the evaluation of the user is subjective evaluation. Therefore, the objective evaluation and the subjective evaluation complement each other, and the objective evaluation and the subjective evaluation are not indispensable.
The objective evaluation is road noise frequency spectrum data; the subjective evaluation is the subjective perception score of an evaluator on the noise in the vehicle, the road noise spectrum data is the objective embodiment of the noise in the vehicle, and the subjective evaluation is also the evaluation on the noise in the vehicle corresponding to the road noise spectrum. Therefore, there is a certain relationship (mapping) between the objective evaluation data and the subjective evaluation data, which is a complicated nonlinear relationship and is difficult to obtain.
Due to the subjectivity of the subjective evaluation of the vehicle road noise, the subjective evaluation results of the same vehicle under the same working condition are different from person to person, the difference is large, and the accuracy of the subjective evaluation results is not high.
Disclosure of Invention
The invention aims to solve the technical problem of providing a vehicle road noise subjective evaluation value prediction method aiming at the defects in the prior art.
The technical scheme adopted by the invention for solving the technical problems is as follows: a vehicle road noise subjective evaluation value prediction method comprises the following steps:
1) acquiring vehicle road noise test data of each sample and a corresponding subjective evaluation value as training sample data;
2) determining 1/s octave frequency spectrum data of the road noise of each sample vehicle according to the road noise test data of each sample vehicle;
3) establishing a BP neural network model, wherein the neural network model takes 1/s octave frequency spectrum data of each sample vehicle road noise as input and takes a subjective evaluation value as output;
4) setting an output error limit value of the BP neural network, training the BP neural network, and obtaining a neural network model for predicting a subjective evaluation value after training;
5) collecting road noise data of a vehicle to be evaluated;
6) determining 1/s octave frequency spectrum data of the road noise of the vehicle to be evaluated; s is an integer and can generally take on values of 3,6,12,24,48, 96. Such as 1/3 octaves refers to dividing each band of an octave into 3 subbands.
7) And obtaining the subjective evaluation predicted value of the road noise of the vehicle to be evaluated by using the trained neural network model.
According to the scheme, the vehicle road noise test data in the step 1) is the vehicle interior noise data of each vehicle under the same vehicle speed, the same working condition, the same road surface and the same test condition.
According to the scheme, the 1/s octave frequency spectrum data of the road noise of each sample vehicle is determined according to the road noise test data of each sample vehicle in the step 2), and the method specifically comprises the following steps:
2.1) applying a window function to the vehicle road noise test data, and carrying out FFT analysis to obtain a sound pressure value under each frequency;
2.2) converting the data in 2.1) into sound pressure levels, the calculation formula is as follows:
wherein p isiThe sound pressure value p at the i-th frequency obtained in step 2.1)0The reference sound pressure is used; emSPL1 being an energy correction factor for an applied window functioniIs the sound pressure level at the ith frequency;
2.3) converting the sound pressure level into a weighted sound pressure level, wherein the calculation formula is as follows:
SPL2i＝SPL1i-Li
wherein L isiFor the attenuation value of weighted characteristic curve (such as A weight, C weight, linear weight) at ith frequency, SPL2iA weighted sound pressure level at the ith frequency;
2.4) converting the weighted sound pressure level into a 1/s octave weighted sound pressure level, wherein the calculation formula is as follows:
wherein f isljLower limit frequency, f, of the jth frequency band of 1/s octaveujThe upper limit frequency of the jth frequency band of 1/s octave, f (i) the frequency corresponding to the ith data in the first step, SPL3jSound pressure level of the jth frequency band which is 1/s octave;
2.5) calculating the sound pressure levels of all frequency bands of the 1/s octave, namely obtaining the frequency spectrum data of the 1/s octave of the vehicle road noise.
According to the scheme, the BP neural network model in the step 3) is composed of three layers, wherein the first layer is an input layer, the second layer is a hidden layer, and the third layer is an output layer.
According to the scheme, the number n of nodes of an input layer in the BP neural network model in the step 3) is determined according to the selected 1/s octave and the considered frequency range, the number p of nodes of an output layer is 1, the number m of nodes of a hidden layer is determined according to the number n of nodes of the input layer, and the node is estimated according to the following formula:
m＝[log2n]。
according to the scheme, the neural network model in the step 3) takes 1/s octave spectrum data of each sample vehicle road noise as input, and data is normalized and processed into a number between 0 and 1 before input.
The invention has the following beneficial effects:
the method for predicting the subjective evaluation value of the vehicle road noise can be used for predicting the subjective evaluation value of the vehicle road noise by testing the vehicle road noise data, the subjective evaluation value is high in accuracy, and the method is greatly convenient for the subjective evaluation work of the vehicle road noise.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a flow chart of a method of an embodiment of the present invention;
FIG. 2 is a schematic structural diagram of a BP neural network according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a neural network activation function sigmoid according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
As shown in fig. 1, the present invention provides a method for predicting a vehicle road noise subjective evaluation value by testing vehicle road noise data.
Referring to the method flow charts and the BP neural network structure diagram in fig. 1 and 2, the algorithm of the invention is divided into two flows: a training process and a prediction process. In the training process, vehicle road noise test data and subjective evaluation data of a certain sample amount are obtained, a BP neural network model is established, the output error limit value of the BP neural network is set, calculated vehicle road noise frequency spectrum data are used as the input of the BP neural network, the vehicle road noise subjective evaluation data are used as the output of the BP neural network, and the BP neural network is trained. If the error of the training result does not meet the set error limit value, resetting the output error limit value of the BP neural network, and re-training the BP neural network. And if the error of the training result meets the set error limit value, finishing the training of the BP neural network. In the prediction process, road noise data of a certain vehicle is tested, frequency spectrum data of the vehicle is obtained through calculation and is input into the trained BP neural network, and then the subjective evaluation value of the certain vehicle can be obtained.
In the invention, the training process comprises the following steps:
the method comprises the following steps: and acquiring road noise test data of each vehicle.
The vehicle road noise test data is the vehicle interior noise data of each vehicle under the same vehicle speed, the same working condition, the same road surface and the same test condition.
The vehicle road noise test data sample size is large enough, and generally needs to be larger than 30. The samples should be broad and representative. The sample size cannot be too large, otherwise overfitting can occur, and the generalization capability of the BP neural network is poor.
Step two: calculating 1/s octave frequency spectrum data of the road noise of each vehicle, wherein the calculation process is as follows:
firstly, applying a window function, such as a Hanning window, to vehicle road noise test data, and performing FFT analysis;
secondly, converting the data in the first step into sound pressure levels, wherein a calculation formula is as follows:
wherein p isiThe sound pressure value (unit: Pa) under the frequency i obtained in the step one; p is a radical of0As reference sound pressure, p0＝2*10- 5Pa；EmThe coefficients are modified for the energy of the applied window function. SPL1iIs the sound pressure level at frequency i in dB.
Thirdly, converting the data in the second step into weighted sound pressure level, wherein the calculation formula is as follows:
SPL2i＝SPL1i-Li
wherein L isiThe attenuation value of the weighting characteristic curve at the frequency i is obtained. SPL2iIf weighting is carried out by A, the weighting is marked as dBA; if the weighting of C is adopted, the weight is marked as dBC; and so on.
Fourthly, the method comprises the following steps: converting the data in the third step into 1/s octave (such as 1/1 octaves, 1/3 octaves and the like) weighting sound pressure level, wherein the calculation formula is as follows:
wherein f isljLower limit frequency, f, of the jth frequency band of 1/s octaveujThe upper limit frequency of the jth frequency band of 1/s octave, and f (i) is the frequency corresponding to the ith data in the first step. SPL3jIs the sound pressure level of the 1/s octave jth frequency band.
Calculating the sound pressure level of all frequency bands to obtain the frequency spectrum data of the 1/s octave of the vehicle road noise.
And step three, acquiring the subjective evaluation value of the road noise of each vehicle.
The road noise subjective evaluation value of each vehicle is the subjective evaluation value of the road noise level of an evaluator on the same vehicle speed, the same working condition and the same road surface. The subjective evaluation value is required to be authoritative, and the subjective evaluation of the road noise level of each vehicle can be carried out by adopting an authoritative expert; the evaluation value of the road noise level of each vehicle by a plurality of professionals can be averaged to obtain the evaluation value.
Step four: and establishing a BP neural network model.
The BP neural network system structure adopted by the invention is composed of three layers, wherein the first layer is an input layer, n nodes are provided in total, and n input parameters are correspondingly represented. The second layer is a hidden layer, and has m nodes, and is determined in a self-adaptive mode by the training process of the network. The third layer is an output layer, p nodes are provided in total, and the output is determined by the response actually needed by the system.
The mathematical model of the network is:
inputting a vector: x ═ x1,x2,...,xn)T
Intermediate layer vector: y ═ y1,y2,...,ym)T
Outputting a vector: o ═ o (o)1,o2,...,op)T
The functional relationship between the input vector and the output vector is:
o＝F2(H*F1(W*x+b1)+b2) (1)
in formula 1, x and o are input vectors and output vectors, and the expression of the intermediate layer vector y is:
y＝W*x+b1 (2)
in formula 1, W and H are weight coefficient matrices between the input layer and the intermediate layer (hidden layer) and between the intermediate layer and the output layer, respectively.
In the formula (1), b1 and b2 are bias matrices between the input layer and the intermediate layer (hidden layer) and between the intermediate layer and the output layer, respectively.
In the formula (1), F1(z) and F2(z) are activation functions, and sigmoid functions are generally adopted:
the function graph is shown in fig. 3.
In the invention, the number n of nodes of the input layer is determined according to the selected 1/s octave and the considered frequency range, and the number p of nodes of the output layer is 1. The number m of hidden layer nodes is estimated by the following formula:
m＝[log2n]
the n parameters of the input signal are road noise spectrum data. Taking 1/3 octave spectrum data as an example, considering road noise data in the range of 20-5000 Hz, sound pressure level data corresponding to 20, 25, 31.5, 40, 50, 63, 80, 100, 125, 160, 200, 250, 315, 400, 500, 630, 800, 1000, 1250, 1600, 2000, 2500, 3150, 4000, 5000Hz respectively as a group of input. Table 1 is an example of a set of input data where the first column is the center frequency and the second column is the sound pressure level corresponding to the center frequency. In this example, n is 25, i.e., 25 center frequencies. And combining the sound pressure level data of the vehicles together to form the BP neural network input matrix.
TABLE 1
Before data is input into the BP artificial neural network, the data needs to be normalized into a number between 0 and 1.
Specifically, the sound pressure level L at the center frequency is normalized as follows
Wherein L ismaxAnd LminMaximum and minimum values of sound pressure level, x, at the center frequency, respectivelyiIs the sound pressure level coefficient at the center frequency.
And the parameter of the output layer is a subjective evaluation value of the vehicle road noise.
Step five: and setting an output error limit value of the BP neural network.
The error limit is set in relation to the number of samples. The larger the sample size, the larger the error limit. When the number of samples is constant, a smaller error limit, such as 0.001 or 0.005, is set first, and the BP neural network is trained. And if the training result does not meet the requirement of the error limit value, gradually increasing the error limit value until the requirement is met.
Step six: and training the BP neural network.
After the BP neural network node model is established, the training of the BP neural network can be carried out. Setting a connection weight W between an input node i and a hidden layer node j according to the obtained training sampleijConnection weight H between hidden layer node j and output layer node kjkBias value b of hidden layer node j1jBias value b of output layer node k2kThe 4 matrices are all initialized with random numbers between-1 and 1.
Based on input vector x, matrix W, bias matrix b1The intermediate layer (hidden layer) vector y is calculated using equation 2.
Based on the intermediate layer vector y, the matrix H, the offset matrix b2Using formulae 3 and 3And 1, calculating to obtain an output vector o.
And calculating the deviation between the output vector o and the subjective evaluation value of the vehicle road noise. Based on this deviation, W is continuously corrected using an error back propagation algorithm (BP algorithm)ijAnd HjkUntil the system error is less than or equal to the set output error limit value, the training process of the BP neural network is completed.
It is supplemented here to explain how the data needs to be normalized to a number between 0 and 1 before entering the BP artificial neural network. Fig. 3 is a diagram of a neural network activation function sigmoid. As can be seen from fig. 3, when the absolute value of the independent variable X is large (e.g., > 3), the dependent variable Y is in the saturation range, and the gradient (derivative) thereof is small and close to 0. According to the neural network BP algorithm, W at this timeijAnd HjkThe training process of the neural network is lengthened and even the training fails due to the fact that timely and effective correction cannot be obtained. When limiting the argument X to [0,1 ]]In the range of (1), the dependent variable Y is in a linear section, and the gradient (derivative) thereof is large, in which case W isijAnd HjkCan be corrected timely and effectively, and the neural network training process is easier.
It will be understood that modifications and variations can be made by persons skilled in the art in light of the above teachings and all such modifications and variations are intended to be included within the scope of the invention as defined in the appended claims.
Claims (6)
1. A method for predicting a subjective evaluation value of vehicle road noise is characterized by comprising the following steps:
1) acquiring vehicle road noise test data of each sample and a corresponding subjective evaluation value as training sample data;
2) determining 1/s octave frequency spectrum data of the road noise of each sample vehicle according to the road noise test data of each sample vehicle;
3) establishing a BP neural network model, wherein the neural network model takes 1/s octave frequency spectrum data of each sample vehicle road noise as input and takes a subjective evaluation value as output;
4) setting an output error limit value of the BP neural network, training the BP neural network, and obtaining a neural network model for predicting a subjective evaluation value after training;
5) collecting road noise data of a vehicle to be evaluated;
6) determining 1/s octave frequency spectrum data of the road noise of the vehicle to be evaluated;
7) and obtaining the subjective evaluation predicted value of the road noise of the vehicle to be evaluated by using the trained neural network model.
2. The method for predicting the subjective evaluation value of the vehicle road noise according to claim 1, wherein the vehicle road noise test data in the step 1) is vehicle interior noise data of each vehicle under the same vehicle speed, the same working condition, the same road surface and the same test condition.
3. The method for predicting the subjective evaluation value of the vehicle road noise according to claim 1, wherein in the step 2), the 1/s octave spectrum data of each sample vehicle road noise is determined according to each sample vehicle road noise test data, and specifically, the method comprises the following steps:
2.1) applying a window function to the vehicle road noise test data, and carrying out FFT analysis to obtain a sound pressure value under each frequency;
2.2) converting the data in 2.1) into sound pressure levels, the calculation formula is as follows:
wherein p isiThe sound pressure value p at the i-th frequency obtained in step 2.1)0The reference sound pressure is used; emSPL1 being an energy correction factor for an applied window functioniIs the sound pressure level at the ith frequency;
2.3) converting the sound pressure level into a weighted sound pressure level, wherein the calculation formula is as follows:
SPL2i＝SPL1i-Li
wherein L isiTo weight the attenuation of the characteristic at the ith frequency, SPL2iA weighted sound pressure level at the ith frequency;
2.4) converting the weighted sound pressure level into a 1/s octave weighted sound pressure level, wherein the calculation formula is as follows:
wherein f isljLower limit frequency, f, of the jth frequency band of 1/s octaveujThe upper limit frequency of the jth frequency band of 1/s octave, f (i) the frequency corresponding to the ith data in the first step, SPL3jSound pressure level of the jth frequency band which is 1/s octave;
2.5) calculating the sound pressure levels of all frequency bands of the 1/s octave, namely obtaining the frequency spectrum data of the 1/s octave of the vehicle road noise.
4. The method for predicting the subjective evaluation value of the vehicle road noise according to claim 1, wherein the BP neural network model in the step 3) is composed of three layers, a first layer is an input layer, a second layer is a hidden layer, and a third layer is an output layer.
5. The method for predicting the subjective evaluation value of the vehicle road noise according to claim 4, wherein the number n of nodes of the input layer in the BP neural network model in step 3) is determined according to the selected 1/s octave and the considered frequency range, the number p of nodes of the output layer is 1, the number m of nodes of the hidden layer is determined according to the number n of nodes of the input layer, and is estimated according to the following formula:
m＝[log2n]。
6. the method for predicting the subjective evaluation value of the vehicle road noise according to claim 1, wherein the neural network model in step 3) takes the 1/s octave spectrum data of each sample vehicle road noise as input, and performs data normalization processing to a number between 0 and 1 before the input.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010927866.2A CN112161815A (en) | 2020-09-07 | 2020-09-07 | Vehicle road noise subjective evaluation value prediction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010927866.2A CN112161815A (en) | 2020-09-07 | 2020-09-07 | Vehicle road noise subjective evaluation value prediction method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112161815A true CN112161815A (en) | 2021-01-01 |
Family
ID=73858077
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010927866.2A Pending CN112161815A (en) | 2020-09-07 | 2020-09-07 | Vehicle road noise subjective evaluation value prediction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112161815A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112992182A (en) * | 2021-02-10 | 2021-06-18 | 东风汽车集团股份有限公司 | Vehicle wind noise level testing system and testing method thereof |
| CN113702071A (en) * | 2021-09-18 | 2021-11-26 | 燕山大学 | NVH evaluation result prediction method under idle working condition |
| CN113886974A (en) * | 2021-10-28 | 2022-01-04 | 重庆长安汽车股份有限公司 | Method for predicting sound path noise of in-vehicle structure |
| CN114544194A (en) * | 2022-01-25 | 2022-05-27 | 东风汽车集团股份有限公司 | Vehicle road noise evaluation method based on spectrum analysis |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN2566372Y (en) * | 2002-12-12 | 2003-08-13 | 谌德荣 | Noise signal processing chip |
| CN103471709A (en) * | 2013-09-17 | 2013-12-25 | 吉林大学 | Method for predicting noise quality of noise inside passenger vehicle |
| CN206095421U (en) * | 2016-10-11 | 2017-04-12 | 云南环绿环境检测技术有限公司 | Many functional type noise meter |
| CN109324120A (en) * | 2018-11-30 | 2019-02-12 | 重庆长安汽车股份有限公司 | A kind of speed changer sound transmission loss test method and test macro |
| CN110928861A (en) * | 2018-09-18 | 2020-03-27 | 上汽通用汽车有限公司 | Auxiliary analysis and evaluation method and system for vehicle road noise |
-
2020
- 2020-09-07 CN CN202010927866.2A patent/CN112161815A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN2566372Y (en) * | 2002-12-12 | 2003-08-13 | 谌德荣 | Noise signal processing chip |
| CN103471709A (en) * | 2013-09-17 | 2013-12-25 | 吉林大学 | Method for predicting noise quality of noise inside passenger vehicle |
| CN206095421U (en) * | 2016-10-11 | 2017-04-12 | 云南环绿环境检测技术有限公司 | Many functional type noise meter |
| CN110928861A (en) * | 2018-09-18 | 2020-03-27 | 上汽通用汽车有限公司 | Auxiliary analysis and evaluation method and system for vehicle road noise |
| CN109324120A (en) * | 2018-11-30 | 2019-02-12 | 重庆长安汽车股份有限公司 | A kind of speed changer sound transmission loss test method and test macro |
Non-Patent Citations (3)
| Title |
|---|
| 刘小民 等: ""苍鹰翼尾缘结构的单元仿生叶片降噪机理研究"", 《西安交通大学学报》 * |
| 吴鹏 等: ""基于BP神经网络的电梯噪声评价方法"", 《数字技术与应用》 * |
| 崔海洋: ""车内稳态噪声声品质主观评价预测模型研究"", 《汽车工业 工程科技II辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112992182A (en) * | 2021-02-10 | 2021-06-18 | 东风汽车集团股份有限公司 | Vehicle wind noise level testing system and testing method thereof |
| CN113702071A (en) * | 2021-09-18 | 2021-11-26 | 燕山大学 | NVH evaluation result prediction method under idle working condition |
| CN113886974A (en) * | 2021-10-28 | 2022-01-04 | 重庆长安汽车股份有限公司 | Method for predicting sound path noise of in-vehicle structure |
| CN113886974B (en) * | 2021-10-28 | 2024-01-02 | 重庆长安汽车股份有限公司 | Method for predicting sound path noise of in-vehicle structure |
| CN114544194A (en) * | 2022-01-25 | 2022-05-27 | 东风汽车集团股份有限公司 | Vehicle road noise evaluation method based on spectrum analysis |
| CN114544194B (en) * | 2022-01-25 | 2023-06-23 | 东风汽车集团股份有限公司 | Vehicle road noise evaluation method based on spectrum analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112161815A (en) | Vehicle road noise subjective evaluation value prediction method | |
| CN107358966B (en) | No-reference speech quality objective assessment method based on deep learning speech enhancement | |
| CN108764520B (en) | Water quality parameter prediction method based on multilayer cyclic neural network and D-S evidence theory | |
| CN108447495B (en) | Deep learning voice enhancement method based on comprehensive feature set | |
| CN111257934B (en) | Seismic oscillation peak acceleration prediction method based on second-order neuron deep neural network | |
| CN109543763A (en) | A kind of Raman spectrum analysis method based on convolutional neural networks | |
| CN106656357B (en) | Power frequency communication channel state evaluation system and method | |
| CN111931983B (en) | Precipitation prediction method and system | |
| CN108734266A (en) | Compression method and device, terminal, the storage medium of deep neural network model | |
| CN108734287A (en) | Compression method and device, terminal, the storage medium of deep neural network model | |
| CN108734268A (en) | Compression method and device, terminal, the storage medium of deep neural network model | |
| CN111060316A (en) | Rolling bearing state monitoring method and system based on convolutional neural network model | |
| CN112562698B (en) | Power equipment defect diagnosis method based on fusion of sound source information and thermal imaging characteristics | |
| CN108630228A (en) | Sound quality recognition methods, device, system and vehicle | |
| KR101314043B1 (en) | Method for sound quality analysis of axle gear whine sound and apparatus thereof | |
| CN112885375A (en) | Global signal-to-noise ratio estimation method based on auditory filter bank and convolutional neural network | |
| CN112086100A (en) | Quantization error entropy based urban noise identification method of multilayer random neural network | |
| CN108538301B (en) | Intelligent digital musical instrument based on neural network audio technology | |
| CN115169218A (en) | Gear vibration noise estimation method based on deep belief network | |
| Duangpummet et al. | A robust method for blindly estimating speech transmission index using convolutional neural network with temporal amplitude envelope | |
| Bader Eddin et al. | A sound insulation prediction model for floor structures in wooden buildings using neural networks approach | |
| Li et al. | A neural network model for speech intelligibility quantification | |
| CN111105127A (en) | Modular product design evaluation method based on data driving | |
| CN114464159A (en) | Vocoder voice synthesis method based on half-flow model | |
| CN112992182A (en) | Vehicle wind noise level testing system and testing method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210101 |