CN115222085A - Noise analysis optimization method based on noise evaluation, terminal and storage medium - Google Patents

Noise analysis optimization method based on noise evaluation, terminal and storage medium Download PDF

Info

Publication number
CN115222085A
CN115222085A CN202111007706.7A CN202111007706A CN115222085A CN 115222085 A CN115222085 A CN 115222085A CN 202111007706 A CN202111007706 A CN 202111007706A CN 115222085 A CN115222085 A CN 115222085A
Authority
CN
China
Prior art keywords
noise
evaluation data
subjective
principal component
objective
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
Application number
CN202111007706.7A
Other languages
Chinese (zh)
Inventor
黄剑锋
王常伟
张志达
许静超
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangzhou Automobile Group Co Ltd
Original Assignee
Guangzhou Automobile Group Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guangzhou Automobile Group Co Ltd filed Critical Guangzhou Automobile Group Co Ltd
Priority to CN202111007706.7A priority Critical patent/CN115222085A/en
Publication of CN115222085A publication Critical patent/CN115222085A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Mathematical Physics (AREA)
  • Development Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Operations Research (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Tourism & Hospitality (AREA)
  • Educational Administration (AREA)
  • Mathematical Analysis (AREA)
  • Quality & Reliability (AREA)
  • Data Mining & Analysis (AREA)
  • Game Theory and Decision Science (AREA)
  • Computational Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • Algebra (AREA)
  • Computing Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)

Abstract

The invention discloses a noise analysis optimization method based on noise evaluation, a terminal and a storage medium. The noise analysis optimization method based on noise evaluation comprises the following steps: acquiring subjective evaluation data and objective evaluation data aiming at target noise; establishing an association model of objective evaluation data and subjective evaluation data; and calculating the change rule of the subjective evaluation data in the target noise frequency spectrum according to the correlation model so as to determine the target noise frequency band to be optimized. By the mode, the subjective noise evaluation parameters can be obtained according to the objective evaluation data, the client sensitive frequency band in the target noise can be obtained, and the noise optimization direction can be indicated.

Description

Noise analysis optimization method based on noise evaluation, terminal and storage medium
Technical Field
The present invention relates to the field of noise control technologies, and in particular, to a noise analysis optimization method based on noise evaluation, a terminal, and a storage medium.
Background
In the prior art, a single classical sound quality parameter (such as roughness, prominence ratio, scheduling, definition, sound pressure level and the like) cannot comprehensively reflect a noise subjective evaluation result. However, the current comprehensive sound quality indexes, such as comfort level P, unpleasantness UP, sound quality preference SQ, etc., only involve a few sound quality parameters, and cannot completely characterize the characteristics of noise. Moreover, the relation between the noise spectrum characteristics and subjective evaluation is not disclosed in a single sound quality parameter and a comprehensive sound quality index, and the quality of the noise can only be roughly measured, but the requirement of improvement cannot be specified, and the method has no guiding significance to engineering development.
Disclosure of Invention
The invention aims to provide a noise analysis optimization method based on noise evaluation, a terminal and a storage medium, which can obtain subjective noise evaluation parameters according to objective evaluation data, obtain client sensitive frequency bands in target noise and indicate the noise optimization direction.
In order to solve the above technical problem, the present application provides a noise analysis optimization method based on noise evaluation, including:
acquiring subjective evaluation data and objective evaluation data aiming at target noise;
establishing a correlation model of the objective evaluation data and the subjective evaluation data;
and calculating a change rule of the subjective evaluation data in the target noise frequency spectrum according to the correlation model so as to determine a target noise frequency band to be optimized.
Wherein the establishing of the association model of the objective evaluation data and the subjective evaluation data includes:
preprocessing the objective evaluation data;
extracting principal component parameters of the objective evaluation data;
and establishing a correlation model of the principal component parameters and the subjective evaluation data.
The objective evaluation data comprises a plurality of sound quality parameters, and the preprocessing of the objective evaluation data comprises:
calculating a basic mean value and a basic standard deviation of the sound quality parameters according to subjective evaluation data and objective evaluation data;
adjusting a standard deviation standardization formula of the sound quality parameter according to the basic mean value and the basic standard deviation;
and normalizing the sound quality parameters according to the adjusted standard deviation normalization formula.
Wherein the standard deviation standardized formula after adjustment is as follows:
Figure BDA0003237611200000021
in the formula, i is a target noise serial number; j is the sound quality parameter serial number; z is a linear or branched member i A normalized value of a jth acoustic quality parameter for an ith target noise; x is a radical of a fluorine atom i A value of a jth acoustic quality parameter for an ith target noise; AX is the base mean value of the jth sound quality parameter; AS is the base standard deviation of the jth sound quality parameter.
Wherein, the extracting of principal component parameters in the objective data includes:
establishing a normalized variable matrix of the acoustic quality parameter
Figure BDA0003237611200000022
Calculating a correlation coefficient matrix of the normalized variable matrix Z
Figure BDA0003237611200000023
Calculating characteristic equation | R-lambada I of correlation coefficient matrix R j |＝0，I j As a unit matrix, a characteristic root λ is obtained j And corresponding feature vector u j ；
Sorting the feature roots according to sizes, and calculating the ratio of the sum of the first m feature roots to the sum of the feature roots
Figure BDA0003237611200000031
If the ratio r is larger than the preset ratio, extracting the first m eigenvectors to form a matrix as follows:
Figure BDA0003237611200000032
and calculating to obtain a principal component matrix Y according to the standardized variable matrix Z and the eigenvector matrix U as follows:
Figure BDA0003237611200000033
in the formula, the principal component matrix Y includes m principal component parameters.
Wherein the establishing of the association model of the principal component parameters and the subjective evaluation data comprises:
inputting the principal component parameters and the subjective evaluation data into a neural network model, wherein the principal component parameters are independent variables, and the subjective evaluation data are dependent variables;
and outputting to obtain a correlation model of the principal component parameters and the subjective evaluation data.
Wherein, the calculating the change rule of the subjective evaluation data in the target noise frequency spectrum according to the correlation model to determine the target noise frequency band to be optimized includes:
acquiring frequency spectrum data of target noise;
dividing the frequency spectrum data into a plurality of noise frequency bands;
adjusting objective evaluation data corresponding to the noise frequency band, and calculating a change rule of corresponding subjective evaluation data according to the correlation model;
and determining the target noise frequency band to be optimized according to the change rule of the subjective evaluation data in each frequency band.
The adjusting of the objective evaluation data corresponding to the noise frequency band includes:
scaling the noise of each noise frequency band;
and calculating corresponding objective evaluation data according to the scaling parameters.
The present application further provides a terminal comprising at least one processor and at least one memory coupled to the at least one processor and storing instructions for execution by the at least one processor, which when executed by the at least one processor, cause the terminal to perform a noise analysis optimization method based on noise evaluation as described above.
The present application further provides a storage medium having computer program instructions stored thereon; the computer program instructions, when executed by a processor, implement a noise analysis optimization method based on noise evaluation as described above.
The noise analysis optimization method based on noise evaluation, the terminal and the storage medium comprise the following steps: acquiring subjective evaluation data and objective evaluation data aiming at target noise; establishing an association model of objective evaluation data and subjective evaluation data; and calculating the change rule of the subjective evaluation data in the target noise frequency spectrum according to the correlation model so as to determine the target noise frequency band to be optimized. By the mode, the subjective noise evaluation parameters can be obtained according to the objective evaluation data, the client sensitive frequency band in the target noise can be obtained, and the noise optimization direction can be indicated.
The foregoing description is only an overview of the technical solutions of the present application, and in order to make the technical means of the present application more clearly understood, the present application may be implemented in accordance with the content of the description, and in order to make the above and other objects, features, and advantages of the present application more clearly understood, the following preferred embodiments are described in detail with reference to the accompanying drawings.
Drawings
Fig. 1 is a schematic flow chart of a noise analysis optimization method based on noise evaluation according to an embodiment of the present invention;
FIG. 2 is a comparison of actual subjective evaluation data and predicted subjective evaluation data according to an embodiment of the invention;
fig. 3 is a line graph showing a change rule of subjective evaluation data in a road noise spectrum collected by a sample car 1 according to an embodiment of the present invention;
fig. 4 is a line graph illustrating a change rule of subjective evaluation data in a road noise spectrum collected by the sample car 2 according to the embodiment of the present invention;
fig. 5 is a line graph illustrating a variation rule of subjective evaluation data in a brownian random noise spectrum according to an embodiment of the present invention;
fig. 6 is a specific flowchart schematic diagram of a noise analysis optimization method based on noise evaluation according to an embodiment of the present invention.
Detailed Description
The following embodiments are provided to illustrate the present disclosure, and other advantages and effects will be apparent to those skilled in the art from the disclosure.
In the following description, reference is made to the accompanying drawings that describe several embodiments of the application. It is to be understood that other embodiments may be utilized and that mechanical, structural, electrical, and operational changes may be made without departing from the spirit and scope of the present application. The following detailed description is not to be taken in a limiting sense, and the scope of embodiments of the present application is defined only by the claims of the issued patent. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.
Although the terms first, second, etc. may be used in some instances to describe various elements in the present embodiments, these elements should not be limited by these terms. These terms are only used to distinguish one element from another.
Furthermore, as used in this embodiment, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context indicates otherwise. It will be further understood that the terms "comprises," "comprising," "includes" and/or "including," when used in this specification, specify the presence of stated features, steps, operations, elements, components, items, species, and/or groups, but do not preclude the presence, or addition of one or more other features, steps, operations, elements, components, species, and/or groups thereof. The terms "or" and/or "as used herein are to be construed as inclusive or meaning any one or any combination. Thus, "a, B or C" or "a, B and/or C" means "any of the following: a; b; c; a and B; a and C; b and C; A. b and C ". An exception to this definition will occur only when a combination of elements, functions, steps or operations are inherently mutually exclusive in some way.
Fig. 1 is a flowchart illustrating a noise analysis optimization method based on noise evaluation according to a first embodiment of the present invention. As shown in fig. 1, a noise analysis optimization method based on noise evaluation provided in an embodiment of the present invention includes:
step 201: acquiring subjective evaluation data and objective evaluation data aiming at target noise;
step 202: establishing an association model of objective evaluation data and subjective evaluation data;
step 203: and calculating the change rule of the subjective evaluation data in the target noise frequency spectrum according to the correlation model so as to determine the target noise frequency band to be optimized.
In this embodiment, the target noise may be road noise collected in the target vehicle, or white noise or brownian random noise. When objective evaluation data for target noise is acquired, a constant speed road noise test can be performed in a test field or a social road, and the in-vehicle noise of a target vehicle is recorded as objective evaluation data. When subjective evaluation data are obtained, as expert evaluation in the prior art cannot well reflect the requirements of target clients, the target population can be organized for subjective evaluation, and road noise subjective scores of different vehicles can be obtained. Specifically, a target client is organized to conduct real vehicle noise evaluation in a test field or conduct noise evaluation in a sound quality playback room, and the subjective score of the target client on each vehicle is obtained and used as subjective evaluation data. By determining the evaluator taking the target client as the subjective evaluation of the road noise, the obtained subjective evaluation data can more accurately reveal the road noise pain point of the target client.
In one embodiment, establishing a correlation model between objective evaluation data and subjective evaluation data includes:
preprocessing objective evaluation data;
extracting principal component parameters of objective evaluation data;
and establishing a correlation model of the principal component parameters and the subjective evaluation data.
It should be noted that objective evaluation data for road noise is only digital, and has no guiding significance for road noise optimization. The embodiment can be used for guiding the optimization development of the road noise by establishing the relation between objective evaluation data and subjective evaluation data. The correlation model can not only obtain subjective evaluation data according to the objective evaluation data, but also indicate which frequency band of the noise has the greatest influence on the subjective evaluation, thereby indicating the optimization direction.
In one embodiment, the objective evaluation data includes a plurality of sound quality parameters, and the preprocessing the objective evaluation data includes:
calculating a basic mean value and a basic standard deviation of the sound quality parameters according to the subjective evaluation data and the objective evaluation data;
adjusting a standard deviation standardization formula of the sound quality parameters according to the basic mean value and the basic standard deviation;
and standardizing the sound quality parameters according to the adjusted standard deviation standardization formula.
The preprocessing of the objective evaluation data includes data normalization of the noise quality parameter of the road noise. The sound quality parameters in this embodiment include: roughnesss, loudness-Stevens-6, loudness-Zwicker-diffuse, tone-to-noise-Ratio, promotion-Ratio, tonality-DIN45681, sharpness-diffuse, sharpness-free, sharpness-Aures-diffuse, sharpness-DIN45692-diffuse, NR, NC, NCB, AI, open-AI, SIL3, PSIL, SPL, FS. The values of the twenty acoustic quality parameters are represented by x, and for example, the parameters of the sample vehicle 01 are respectively: x is the number of 0101 、x 0102 、x 0103 、x 0104 、x 0105 、x 0106 、x 0107 、x 0108 、x 0109 、x 0110 、x 0111 、x 0112 、x 0113 、x 0114 、x 0115 、x 0116 、x 0117 、x 0118 、x 0119 、x 0120 . The parameters of the sample car 02 are: x is a radical of a fluorine atom 0201 、x 0202 、x 0203 、…、x 0118 、x 0119 、x 0120 . And the rest vehicles are analogized in the same way.
The 20 parameters were then normalized using a modified standard deviation normalization method. The classical standard deviation normalization method is as follows:
Figure BDA0003237611200000071
in the formula, x i Representing the ith song sample value;
Figure BDA0003237611200000072
represents the sample mean; s represents the sample standard deviation; z is a radical of i Indicating the normalized value of the ith sample.
As can be seen from the above equation, the normalized value of the acoustic quality parameter varies. When the same sound quality parameter is placed in different sample banks, the sample mean and the standard deviation may change, and thus the normalized value changes. For example, the Roughness roughnesss of car a is 0.25. He placed in 5 carts to calculate a normalized value, which may be 0.9. The normalized value was calculated in a sample bank of 100 trucks, and the result may become 0.8. This can lead to a series of difficulties in subsequent calculations and a degree of uncertainty in the final result. Therefore, this embodiment proposes the concept of the base mean AX and the base standard deviation AS to replace the sample mean x and the standard deviation s in the formula. For vehicle A, noise data with subjective score distribution of 5-8 are collected, and then the mean and standard deviation are calculated. The above-mentioned mean value was used AS a base mean value AX, and 1.5 times the above-mentioned standard deviation was used AS a base standard deviation AS. After AX and AS are determined, they are no longer updated with the sample size, and for each trolley parameter, the normalized value is calculated directly AS follows.
The normalized formula of the standard deviation after adjustment is as follows:
Figure BDA0003237611200000081
in the formula, i is a target noise serial number; j is the sound quality parameter serial number; z i Normalized value of jth sound quality parameter for ith target noise; x is a radical of a fluorine atom i A value of a jth acoustic quality parameter for an ith target noise; AX is the base mean value of the jth sound quality parameter; AS is the base standard deviation of the jth acoustic quality parameter.
It should be illustrated that the 20 sound quality parameters of the sample car 1 are normalized as follows:
Figure BDA0003237611200000082
thus, an improved standard deviation standardization method is introduced into the principal component analysis, so that the calculated principal component numerical value does not change along with the sample amount, and a stable result is obtained.
In one embodiment, extracting principal component parameters from the objective data comprises:
establishing a normalized variable matrix of acoustic quality parameters
Figure BDA0003237611200000083
Calculating a correlation coefficient matrix of the normalized variable matrix Z
Figure BDA0003237611200000084
Calculating characteristic equation | R-lambada I of correlation coefficient matrix R j |＝0，I j For the identity matrix, the characteristic root λ is obtained j And corresponding feature vector u j ；
Sorting the feature roots according to sizes, and calculating the ratio of the sum of the first m feature roots to the sum of the feature roots
Figure BDA0003237611200000091
If the ratio r is larger than the preset ratio, extracting the first m eigenvectors to form a matrix as follows:
Figure BDA0003237611200000092
and calculating to obtain a principal component matrix Y according to the standardized variable matrix Z and the eigenvector matrix U as follows:
Figure BDA0003237611200000093
in the formula, the principal component matrix Y includes m principal component parameters.
It should be noted that the objective evaluation data includes a plurality of sound quality parameters, the number of parameters that can be selected in the actual calculation process is small, the coupling among the parameters is severe, and it is difficult to establish the correlation model. Therefore, in the embodiment, the principal component analysis method is used, dimension reduction is performed on the basis of furthest retaining feature information of each dimension of noise, the number of parameters is not limited, and the principal components are independent of each other, so that the algorithm overhead of the model can be reduced. The model accuracy is improved while the sample size requirement is reduced.
Specifically, in the present embodiment, 20 sound quality parameters of 6 dollies are selected and respectively normalized to obtain the following normalized variable matrix Z.
Figure BDA0003237611200000094
Then, a characteristic equation of the correlation coefficient matrix R, | R- λ I, is calculated 20 I =0, wherein I 20 For the identity matrix, 20 feature roots and corresponding feature vectors can be obtained, and the feature roots are sorted according to the size to obtain lambda 1 、λ 2 、…、λ 20 . The corresponding feature vector is u 1 、u 2 、…、u 20 . For the jth feature vector, there is u j ＝[u 1j u 2j … u 20j ] T 。
Further, the ratio of the first m principal component parameters, i.e., the cumulative contribution ratio, is calculated
Figure BDA0003237611200000101
If the ratio r is greater than the preset ratio, the first m principal component parameters can be extracted. The preset ratio may be 80-90%. In this embodiment, when m is 4, the cumulative contribution rate reaches 85%, that is, the first four principal component parameters may represent most information of 20 parameters.
Next, the first four principal component parameters are calculated. The matrix of 4 eigenvectors is as follows:
Figure BDA0003237611200000102
and (3) multiplying the normalized variable matrix Z by the eigenvector matrix U to obtain a principal component matrix:
Figure BDA0003237611200000103
in the present embodiment, 4 principal component parameters y1, y2, y3, y4 are retained by principal component analysis. Since the present embodiment has 6 sample cars, there are 6 data per principal component parameter. In the case of 100 trolleys, this would become a 4 by 100 matrix.
The present embodiment introduces principal component analysis before the subjective and objective correlation model. And the principal components are utilized to perform dimension reduction decoupling, so that the sample size requirement for establishing the model is reduced, and the model precision is improved. Due to the mutual independence of the principal components and the retention of the representativeness of the original data information, the calculation overhead is reduced, and the model precision is improved. Due to the capability of dimension reduction and decoupling, the quantity of sound quality parameters is not limited, the workload of parameter screening in the early stage is reduced, and the representativeness of later-stage analysis can be ensured. The main components are sorted according to the variance and the contribution amount, and indexes can be selected according to needs to reduce the interference of non-key factors.
In one embodiment, the establishing a correlation model of the principal component parameters and the subjective evaluation data comprises:
inputting principal component parameters and subjective evaluation data into a neural network model, wherein the principal component parameters are independent variables, and the subjective evaluation data are dependent variables;
and outputting to obtain a correlation model of the principal component parameters and the subjective evaluation data.
When a correlation model of the principal component parameters and the subjective evaluation data is established, the four principal component vectors are input to the neural network model as the independent variables of the neural network. And (3) taking the road noise subjective scores (6, 6.25, 7, 6.25, 6.75 and 6.5) of the target client for 6 trolleys as dependent variables, inputting the dependent variables into a neural network, and training to obtain the subjective and objective correlation model.
The subjective-objective correlation model may be used to predict subjective assessment data for other vehicles. And inputting the principal component vector of the sample car into the subjective and objective correlation model to obtain the subjective score of the sample car. After a subjective and objective correlation model is established by the subjective evaluation scores and objective sound quality parameters of the 6 sample vehicles in the embodiment, the ratio of the subjective score to the actual score of the 6 sample vehicles obtained through prediction is shown in fig. 2. As can be seen from the figure, the predicted subjective score of the correlation model is consistent with the actual subjective score, R 2 The value reached 0.94. In actual implementation, R 2 The value should be greater than 0.85.
In one embodiment, calculating a change rule of subjective evaluation data in a target noise frequency spectrum according to a correlation model to determine a target noise frequency band to be optimized includes:
acquiring frequency spectrum data of target noise;
dividing the frequency spectrum data into a plurality of noise frequency bands;
adjusting objective evaluation data corresponding to the noise frequency band, and calculating a change rule of corresponding subjective evaluation data according to the correlation model;
and determining a target noise frequency band to be optimized according to the change rule of the subjective evaluation data in each frequency band.
In one embodiment, adjusting the objective evaluation data corresponding to the noise frequency band includes:
scaling the noise of each noise frequency band;
and calculating corresponding objective evaluation data according to the scaling parameters.
In this embodiment, the increase/decrease operation is performed on the spectrum data of the noise for each frequency band, and the sensitivity of each frequency band to the subjective evaluation is analyzed. Specifically, the road noise is divided into a plurality of frequency bands, noise scaling processing is performed on a certain frequency band, for example, 54dB is originally corresponding to 80Hz, and after 6dB is respectively increased or decreased, three corresponding sets of objective evaluation data are obtained through calculation. And then, three groups of objective evaluation data are brought into the subjective and objective association model, and the subjective scores of the three groups of objective evaluation data can be predicted to be 7, 6.98 and 7.01 respectively. Due to the fact that the subjective evaluation data are small in difference, the target client is not sensitive to 80Hz, namely 80Hz road noise is not easy to cause complaints of the target client. And analyzing each noise frequency band one by one according to the scheme to obtain the sensitivity of the target customer to each frequency band, and further extracting the frequency band with high sensitivity to the target population.
As shown in fig. 3 to 5, after increasing or decreasing each frequency spectrum of the sample car 1, sample car 2, and brown random noise, subjective evaluation prediction is performed, and it can be concluded that: the target client is sensitive to 100 Hz-315 Hz, especially 125Hz and 250Hz. Sensitivity to 40Hz and 50Hz is common. Is less sensitive to the frequency range above 400 Hz. In the embodiment, by analyzing the influence of each frequency band on the subjective evaluation of the road noise, the influence of the spectral characteristics of the road noise on the subjective evaluation is no longer a black box, the frequency band which is most sensitive to a target client is grasped, an acoustic quality frequency domain target is established, and the engineering development can be actually guided.
The noise analysis optimization method based on noise evaluation of the embodiment includes: acquiring subjective evaluation data and objective evaluation data aiming at target noise; establishing an association model of objective evaluation data and subjective evaluation data; and calculating the change rule of the subjective evaluation data in the target noise frequency spectrum according to the correlation model so as to determine the target noise frequency band to be optimized. By the mode, the subjective noise evaluation parameters can be obtained according to objective evaluation data, the customer sensitive frequency band in the target noise can be obtained, and the noise optimization direction can be indicated.
Fig. 6 is a specific flowchart schematic diagram of a noise analysis optimization method based on noise evaluation according to an embodiment of the present invention. As shown in fig. 6, the present embodiment first acquires objective evaluation data, for example, tests objective evaluation data based on road noise. Meanwhile, subjective evaluation data are obtained, and the representative population of the target client can be organized to carry out subjective evaluation in an acoustic quality evaluation room or on a runway. And then, standardizing the objective evaluation data, and extracting principal component parameters in the objective evaluation data to retain original information to the maximum extent, reduce the number of parameters, eliminate redundant data noise and reduce the calculation overhead of an algorithm. And then, establishing an objective association model based on the principal component parameters and the subjective evaluation data, and training to obtain the objective association model by taking the principal component parameters and the subjective evaluation data as input. And predicting the sound quality score through the correlation model, and predicting subjective scores of different noises. And finally, analyzing the influence of the noise of each frequency band on the sound quality, scaling the noise of each frequency band, and analyzing the influence of the noise of each frequency band on the sound quality. And establishing a frequency domain sound quality target, establishing a frequency domain sound quality target of the road noise according to the analysis, mastering the key frequency band of interest of a target client, and guiding the development of the road noise.
According to the embodiment of the application, through a solidified sound quality target making process, a target customer is definitely taken as an evaluator of the road noise subjective evaluation, the evaluator directly gives subjective scoring, the ranking of the quality of the road noise is determined by the evaluator, and the development target is really realized by taking the customer as the center. The acoustic quality target can be used to direct path noise development efforts. The method and the device can obtain subjective evaluation data according to objective evaluation data, can indicate which frequency band of noise has the greatest influence on subjective evaluation, and indicate the optimization direction. Meanwhile, the embodiment reduces the sample size required for establishing the subjective and objective correlation model, namely, reduces the required vehicle and test cost, and achieves the purposes of cost reduction and efficiency improvement.
Embodiments of the present invention also provide a terminal including at least one processor and at least one memory coupled to the at least one processor and storing instructions for execution by the at least one processor, the instructions, when executed by the at least one processor, causing the terminal to perform the noise analysis optimization method based on noise evaluation of the above embodiments.
The embodiment of the application also provides a computer storage medium, wherein the computer storage medium is stored with computer program instructions; the computer program instructions, when executed by the processor, implement a noise analysis optimization method based on noise evaluation as described above.
The above embodiments are merely illustrative of the principles and utilities of the present application and are not intended to limit the application. Any person skilled in the art can modify or change the above-described embodiments without departing from the spirit and scope of the present application. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical concepts disclosed in the present application shall be covered by the claims of the present application.

Claims (10)

1. A noise analysis optimization method based on noise evaluation is characterized by comprising the following steps:
acquiring subjective evaluation data and objective evaluation data aiming at target noise;
establishing a correlation model of the objective evaluation data and the subjective evaluation data;
and calculating a change rule of the subjective evaluation data in the target noise frequency spectrum according to the correlation model so as to determine a target noise frequency band to be optimized.
2. The noise analysis optimization method based on noise evaluation according to claim 1, wherein the establishing of the correlation model between the objective evaluation data and the subjective evaluation data includes:
preprocessing the objective evaluation data;
extracting principal component parameters of the objective evaluation data;
and establishing a correlation model of the principal component parameters and the subjective evaluation data.
3. The noise analysis optimization method based on noise evaluation according to claim 2, wherein the objective evaluation data includes a plurality of sound quality parameters, and the preprocessing the objective evaluation data includes:
calculating a basic mean value and a basic standard deviation of the sound quality parameters according to subjective evaluation data and objective evaluation data;
adjusting a standard deviation standardization formula of the sound quality parameter according to the basic mean value and the basic standard deviation;
and normalizing the sound quality parameters according to the adjusted standard deviation normalization formula.
4. The noise analysis optimization method based on noise evaluation according to claim 3, wherein the adjusted standard deviation normalization formula is:
Figure FDA0003237611190000011
in the formula, i is a target noise serial number; j is the sound quality parameter serial number; z i A normalized value of a jth acoustic quality parameter for an ith target noise; x is the number of i A value of a jth sound quality parameter for an ith target noise; AX is the base mean value of the jth sound quality parameter; AS is the base standard deviation of the jth sound quality parameter.
5. The noise analysis optimization method based on noise evaluation according to claim 4, wherein the extracting principal component parameters from the objective data comprises:
establishing a normalized variable matrix of the acoustic quality parameter
Figure FDA0003237611190000021
Calculating a correlation coefficient matrix of the normalized variable matrix Z
Figure FDA0003237611190000022
Calculating characteristic equation | R-lambada I of correlation coefficient matrix R j |＝0，I j As a unit matrix, a characteristic root λ is obtained j And corresponding feature vector u j ；
Sorting the feature roots according to sizes, and calculating the sum of the first m feature roots in the sum of the feature rootsRatio of
Figure FDA0003237611190000023
If the ratio r is larger than the preset ratio, extracting the first m characteristic vectors to form a matrix as follows:
Figure FDA0003237611190000024
and calculating to obtain a principal component matrix Y according to the standardized variable matrix Z and the eigenvector matrix U as follows:
Figure FDA0003237611190000025
in the formula, the principal component matrix Y includes m principal component parameters.
6. The method according to claim 2, wherein the establishing of the correlation model between the principal component parameters and the subjective evaluation data comprises:
inputting the principal component parameters and the subjective evaluation data into a neural network model, wherein the principal component parameters are independent variables, and the subjective evaluation data are dependent variables;
and outputting to obtain a correlation model of the principal component parameters and the subjective evaluation data.
7. The noise analysis optimization method based on noise evaluation according to claim 1, wherein the calculating a variation rule of the subjective evaluation data in the target noise spectrum according to the correlation model to determine a target noise frequency band to be optimized includes:
acquiring frequency spectrum data of target noise;
dividing the frequency spectrum data into a plurality of noise frequency bands;
adjusting objective evaluation data corresponding to the noise frequency band, and calculating a change rule of corresponding subjective evaluation data according to the correlation model;
and determining the target noise frequency band to be optimized according to the change rule of the subjective evaluation data in each frequency band.
8. The noise analysis optimization method based on noise evaluation according to claim 7, wherein the adjusting the objective evaluation data corresponding to the noise frequency band comprises:
scaling the noise of each noise frequency band;
and calculating corresponding objective evaluation data according to the scaling parameters.
9. A terminal comprising at least one processor and at least one memory coupled to the at least one processor and storing instructions for execution by the at least one processor, which when executed by the at least one processor, cause the terminal to perform the noise analysis optimization method based on noise evaluation of any of claims 1 to 8.
10. A storage medium having computer program instructions stored thereon; the computer program instructions, when executed by a processor, implement a noise analysis optimization method based on noise evaluation according to any one of claims 1 to 8.
CN202111007706.7A 2021-08-30 2021-08-30 Noise analysis optimization method based on noise evaluation, terminal and storage medium Pending CN115222085A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111007706.7A CN115222085A (en) 2021-08-30 2021-08-30 Noise analysis optimization method based on noise evaluation, terminal and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111007706.7A CN115222085A (en) 2021-08-30 2021-08-30 Noise analysis optimization method based on noise evaluation, terminal and storage medium

Publications (1)

Publication Number Publication Date
CN115222085A true CN115222085A (en) 2022-10-21

Family

ID=83606219

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111007706.7A Pending CN115222085A (en) 2021-08-30 2021-08-30 Noise analysis optimization method based on noise evaluation, terminal and storage medium

Country Status (1)

Country Link
CN (1) CN115222085A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116429245A (en) * 2023-06-13 2023-07-14 江铃汽车股份有限公司 Method and system for testing noise of wiper motor

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116429245A (en) * 2023-06-13 2023-07-14 江铃汽车股份有限公司 Method and system for testing noise of wiper motor
CN116429245B (en) * 2023-06-13 2023-09-01 江铃汽车股份有限公司 Method and system for testing noise of wiper motor

Similar Documents

Publication Publication Date Title
CN108241901B (en) Transformer early warning evaluation method and device based on prediction data
CN109902441B (en) KANO model principle-based vehicle performance index weight distribution system and method
CN111294812B (en) Resource capacity-expansion planning method and system
CN110852476A (en) Passenger flow prediction method and device, computer equipment and storage medium
Hayden et al. Statistical methods to develop rating models
CN111985751B (en) Human-computer chat experience assessment system
CN109636467A (en) A kind of comprehensive estimation method and system of the internet digital asset of brand
CN115222085A (en) Noise analysis optimization method based on noise evaluation, terminal and storage medium
CN110852600A (en) Method for evaluating dynamic risk of market subject
CN113935535A (en) Principal component analysis method for medium-and-long-term prediction model
CN114583767B (en) Data-driven wind power plant frequency modulation response characteristic modeling method and system
CN109977977B (en) Method for identifying potential user and corresponding device
Mishra et al. SBLMD–ANN–MOPSO-based hybrid approach for determining optimum parameter in CNC milling
CN113052633A (en) Vehicle residual value evaluation method, device, equipment and medium
CN113283673A (en) Model performance attenuation evaluation method, model training method and device
CN115307721A (en) Method, device and equipment for evaluating quality of automobile acceleration sound and storage medium
CN113313397A (en) Sound quality satisfaction degree grading and limit value determining method
CN112232555A (en) Power customer arrearage early warning method based on grey system theory
CN113486448A (en) Method for evaluating transmission squeal based on masking effect
CN113011748A (en) Recommendation effect evaluation method and device, electronic equipment and readable storage medium
CN112348685A (en) Credit scoring method, device, equipment and storage medium
CN112116139A (en) Power demand prediction method and system
US20200034735A1 (en) System for generating topic inference information of lyrics
CN114528763B (en) Amorphous alloy three-dimensional wound core transformer noise processing method and system
CN113487080B (en) Wind speed dynamic scene generation method, system and terminal based on wind speed classification

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