CN105021275B - In-vehicle sound quality Forecasting Methodology based on time domain dynamic analysis - Google Patents
In-vehicle sound quality Forecasting Methodology based on time domain dynamic analysis Download PDFInfo
- Publication number
- CN105021275B CN105021275B CN201510415624.4A CN201510415624A CN105021275B CN 105021275 B CN105021275 B CN 105021275B CN 201510415624 A CN201510415624 A CN 201510415624A CN 105021275 B CN105021275 B CN 105021275B
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- sound quality
- evaluation
- parameter
- 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.)
- Expired - Fee Related
Links
Landscapes
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
- Fittings On The Vehicle Exterior For Carrying Loads, And Devices For Holding Or Mounting Articles (AREA)
Abstract
The invention belongs to automobile vibration and noise analysis and the technical field that sound quality is predicted in control, it is related to a kind of in-vehicle sound quality Forecasting Methodology based on time domain dynamic analysis；The problem of current in-vehicle sound quality analysis and research can not describe the time domain dynamic characteristic of in-car acoustic environment when vehicle routinely travels more is overcome, is comprised the following steps：1st, voice signal sample in sound collection equipment collecting vehicle is utilized；2nd, framing windowing process is carried out to in-car voice signal sample；3rd, subjective assessment experiment is carried out to internal car noise sample；4th, the objective psychoacoustic parameter of sound quality is calculated；5th, the time domain dynamic characteristic index of objective psychoacoustic parameter is calculated；Index includes the variance of each objective psychoacoustic parameter and the extreme difference of each objective psychoacoustic parameter；6th, the objective comprehensive evaluation model of in-vehicle sound quality is established；7th, the objective comprehensive evaluation model output valve of sound quality and subjective assessment value are contrasted；8th, model is verified using test samples.
Description
Technical field
The invention belongs to automobile vibration and noise analysis and the technical field that sound quality is predicted in control, it is related to one kind and is based on
The in-vehicle sound quality Forecasting Methodology of time domain dynamic analysis.In particular, the present invention is a kind of in-car based on time domain dynamic
The in-vehicle sound quality Forecasting Methodology of specificity analysis, suitable for vehicle routinely travel when as caused by various factors unstable state in-car sound
The prediction of sound quality under sound environment.
Background technology
With the fast development of automobile industry, requirement more and more higher of the people to car comfort, in-vehicle sound quality is being commented
Role is also more and more important in valency automobile total quality.Impression of the in-vehicle sound quality quality reflection people to in-car acoustic environment,
It is the collective effect of human psychology and physiologic factor.
Existing in-vehicle sound quality research at present, most of carried out in anechoic room or semianechoic room, i.e., based on steady
State environment this it is assumed that and the operating mode of vehicle is also set as stable operating modes more in studying.In recent years, only a small number of scholars start
The in-vehicle sound quality research being related under the specific unsteady drying such as acceleration.At the same time, the in-car sound involved by these methods
The in-vehicle sound quality that the objective psychological parameter of the widely used sound quality of quality objective analysis can only generally reflect under Stable State Environment,
The time domain dynamic characteristic that in-car acoustic environment when vehicle routinely travels can not be described is applied to various factors when vehicle routinely travels
Caused by in-car unstable state sound quality research be not reported so far.And visible sound quality research at present, generally based on in-car sound
Sound is this stable premise, and selects to be analyzed and researched instead of whole acoustic environment with the single sample sound of second-time.
And during in-car actual travel, acoustic environment had both been included with driving cycle change and the continually changing vehicle engine of itself is made an uproar
Sound, tyre noise, airborne noise and body structure noise, more there are occupant's communication, multimedia and traveling periphery
The effect of other sound sources such as environment.The action time of above-mentioned sound source is from the several seconds, or even continues the traveling process of escort vehicle, because
This only obviously can not describe in-car acoustic environment comprehensively with the single sample sound of second-time.
The content of the invention
The technical problems to be solved by the invention are when overcoming that vehicle routine traveling can not be described existing for prior art
The problem of time domain dynamic characteristic of in-car acoustic environment, there is provided a kind of in-vehicle sound quality based on time domain dynamic analysis is pre-
Survey method.
Therefore, variance and extreme difference that the present invention is proposed by the use of the objective psychoacoustic parameter of sound quality refer to as dynamic characteristic
Mark, to establish in-vehicle sound quality comprehensive evaluation model, preferably to reflect the actual conditions of in-vehicle sound quality.In specific sound sample
During present treatment, the present invention propose in minute magnitude in collecting vehicle voice signal sample and on second-time carry out sound when
Domain signal framing, and " long time frame " is named as, to reflect the cumulative effect of human auditory's psychological feelingses.And propose to be applied to
The trapezoid window adding window method of long time frame processing, eliminating boundary effect the characteristic of voice signal sample can be enable abundant again simultaneously
Retain.
The present invention carries out sound quality overall merit modeling based on fuzzy synthetic appraisement method to objective evaluation result.It is fuzzy comprehensive
Close evaluation be it is a kind of by obscure boundary, be not easy the integrated evaluating method of quantitative factor quantification, because it has mathematical modeling simple
List, layer of structure are clear, are evaluated the advantages that subject evaluation value is unique, are widely used in many scientific domains.
In order to solve the above technical problems, the present invention adopts the following technical scheme that realization, it is described with reference to the drawings as follows：
A kind of in-vehicle sound quality Forecasting Methodology based on time domain dynamic analysis, comprises the following steps：
Step 1：Utilize voice signal sample in sound collection equipment collecting vehicle；
Step 2：Framing windowing process is carried out to in-car voice signal sample signal；
In collecting vehicle voice signal sample to sample sound carry out minute magnitude interception, and with second-time carry out sound when
Domain signal framing, the processing of trapezoid window adding window method is carried out to long time frame, eliminating boundary effect simultaneously, retaining voice signal sample
Characteristic；
Step 3：Subjective assessment experiment is carried out to internal car noise sample；
Step 4：Calculate the objective psychoacoustic parameter of sound quality；
The objective psychoacoustic parameter of sound quality includes：Loudness, sharpness, roughness, shake degree and speech articulation；
Step 5：Calculate the time domain dynamic characteristic index of objective psychoacoustic parameter；
The variance of the time domain dynamic characteristic index of the objective psychoacoustic parameter including each objective psychoacoustic parameter and
The extreme difference of each objective psychoacoustic parameter；
Step 6：Establish the objective comprehensive evaluation model of in-vehicle sound quality；
Step 7：To the objective comprehensive evaluation model output valve of sound quality compared with subjective assessment value, according to comparative result
Step 6 established model is adjusted, until result and actual error, within 20%, degree of error meets to require；
Step 8：Model is verified using test samples；
Model is tested using test samples, if Comparative result degree error within 20%, illustrates that model can be with
Receive, if Comparative result degree is more than 20%, return to step 6, model is readjusted, until meeting error in this result
It is required that.
Voice signal sample in collecting vehicle in step 1, is comprised the following steps that described in technical scheme：
(1) sound collection equipment cloth is placed at copilot, i.e., dummy head's or two microphone particular locations is vertical
Coordinate is seat surface center line, and abscissa is backrest surface center line, more than both intersection points at 0.7 ± 0.05m, dummy head or transaudient
Device is symmetrically arranged at the plane of symmetry of seat surface and backrest or so 0.2 ± 0.02m；
(2) road conditions and operating mode are selected：
The grade road surface of selection test vehicle traveling, and the multiple representational vehicle driving-cycles of design alternative；
(3) in-car voice signal sample collection：
The in-car voice signal sample of each operating mode is recorded, records 5~6 minutes time every time；
The specific steps of framing windowing process are carried out such as to internal car noise sample signal in step 2 described in technical scheme
Under：
(1) in-car voice signal sample framing：
It was that a frame length carries out sound time-domain signal framing with 2 seconds in collecting vehicle a length of 4 minutes during voice signal sample；
(2) window type is selected：
Select in the trapezoid window for eliminating boundary effect simultaneously and the characteristic of voice signal sample being enable to be sufficiently reserved；
(3) window shape parameter is set：
Window shape parameter is defined as：Window length is 2 seconds, and edge length is 10 milliseconds.
Subjective assessment experiment is carried out described in technical scheme to internal car noise sample in step 3 to comprise the following steps that：
(1) composition of personnel jury is chosen：
The testing crew of more than 20 be have chosen as sound quality subjective assessment main body, the ratio of men and women is in Appraising subject
1:1, between 18~70, there are the quantity of the test evaluation personnel of driving experience and the test evaluation personnel without driving experience at the age
Quantitative proportion be 1:1；
(2) jury personnel are carried out testing preceding training；
(3) sound quality evaluation index is determined：
Evaluation index is acoustic environment comfort level；
(4) evaluation method is determined：
Sample sound marking evaluation is carried out with Numerical value method.
Comprising the following steps that for the objective psychoacoustic parameter of sound quality is calculated described in technical scheme in step 3：
(1) sample after framing windowing process is extracted, carries out loudness calculating：
In formula：N ' is characteristic loudness, unit sone；
ETQFor excitation corresponding with the threshold of audibility；
E is to be encouraged corresponding to acoustical signal；
E0To encourage, with reference sound intensity I0=10-12w/m2It is corresponding；
Loudness N is frequency band feature loudness sum, unit sone, and calculation formula is as follows：
(2) sample after framing windowing process is extracted, carries out sharpness calculating；
Sharpness S is calculated using Zwicker models, for its mathematical modeling based on Scale Model of Loudness, mathematical formulae is as follows：
In formula, S is sharpness, unit acum；
K is weight coefficient, and k takes 0.11；
N is total loudness value, unit sone；
N ' (z) is the characteristic loudness in z Bark domains, unit sone；
(3) sample after framing windowing process is extracted, carries out roughness calculating：
Utilize the modulating frequency f of noisemodWith the differential Δ L of excitation in individual feature bandE(z) roughness, formula are calculated
For：
In formula, R is roughness, unit asper；
fmodIt is modulating frequency；
ΔLE(z) it is the variable quantity of acoustical signal drive(r) stage, is defined as：
In formula：N′maxAnd N ' (z)min(z) maximum and minimum value of characteristic loudness are represented respectively；
(4) sample after framing windowing process is extracted, carries out the calculating of shake degree：
Zwicker shake degree computation models are：
Wherein, F represents shake degree, unit vacil；
f0Represent modulation fundamental frequency；
(5) sample after framing windowing process is extracted, carries out speech articulation AI calculating:
Speech articulation computation model is：
AI=∑ W (f) D (f)/30 (7)
In formula, W (f) is weighted coefficient；
N (f) reaches the standard grade for ambient noise, and its expression formula is UL (f)=H (f)+12dB；
LL (f) is ambient noise lower limit, and its expression formula is LL (f)=UL (f) -30dB.
The specific steps of the time domain dynamic characteristic index of objective psychoacoustic parameter are calculated described in technical scheme in step 4
It is as follows：
(1) the variance S of each objective psychoacoustic parameter2Calculate
The loudness that calculates by formula, sharpness, roughness, speech articulation, shake degree are entered according to formula of variance
Row calculates：
In formula：xiIt is each parameter with calculated value that 2 seconds are a frame；
For the average value of each parameter calculated value；
N is the number of each parameter calculated value；
(2) the extreme difference R of each objective psychoacoustic parameter is calculated
R=xmax-xmin (10)
In formula：xmaxWith 2 seconds be for each parameter a frame calculated value in maximum；
xminWith 2 seconds be for each parameter a frame calculated value in minimum value.
Comprising the following steps that for the objective comprehensive evaluation model of in-vehicle sound quality is established described in technical scheme in step 6：
(1) fuzzy overall evaluation set of factors is determined；
During the fuzzy evaluation of sound quality, the time domain dynamic characteristic for choosing each objective psychoacoustic parameter and each parameter refers to
It is denoted as evaluation parameter：
U=speech articulation, and the variance of speech articulation, loudness, the variance of loudness, roughness, the variance of roughness,
Sharpness, the variance of sharpness, shake degree, the variance of degree of shake }；
(2) alternative collection is established；
Alternative collection is set of each factor to the evaluation result of evaluation object, is a kind of vector representation：
V={ can not receive, it is impossible to receive, typically, can receive, take in good part } completely；
(3) weight is determined；
(4) membership function is determined.
Weight is determined described in technical scheme, refers to determine objective psychology using the analytic hierarchy process (AHP) that qualitative and quantitative combines
The weight of the variance of parameters,acoustic and each parameter, Judgement Matricies, detailed process are as follows accordingly：
1) level is determined to each factor of evaluation object first, similar factor is classified as one kind and dropped to a lower level, come with this
Determine the hierarchical structure of each factor.
2) objective indicator and the correlation coefficient value and judgment matrix scale of subjective evaluation index are referred to, more same level
Each factor, judgment matrix is supplemented according to the difference of importance between each factor；
3) eigenvalue and eigenvector of calculating matrix, consistency check is carried out to judgment matrix；
If 4) matrix consistency inspection is qualified, the calculating of weight vectors is carried out, is the characteristic vector of matrix after normalization.
9th, a kind of in-vehicle sound quality Forecasting Methodology based on time domain dynamic analysis according to claim 8, its
It is characterised by：
The eigenvalue and eigenvector of the calculating matrix, the specific steps of consistency check are carried out such as to judgment matrix
Under：
A, calculating matrix A eigenvalue of maximum λmax；
B, the coincident indicator C.L shown in solution formula (11)；
λmaxFor the eigenvalue of maximum of matrix A, n is exponent number.
C, Aver-age Random Consistency Index R.L. corresponding to matrix is found out
D, the Consistency Ratio C.R. shown in solution formula (12)；
If e, during C.R. ＞ 0.1, it need to rebuild or correct judgment matrix；If C.R. ＜ 0.1, it is considered that judgment matrix
Uniformity can receive；
If matrix consistency inspection is qualified, the calculating of weight vectors is carried out, is the characteristic vector of matrix after normalization, it is special
The solution procedure for levying vector is as follows：
A. matrix A is pressed into row normalization：
B. the data after each row normalization are added by row：
C. obtain and vector is normalized, obtains weight vectors：
According to step described above, it is determined that the hierarchical structure of each factor of evaluation, and then determine the weight of each level to
Amount, total weight vectors are A.
Determine that membership function step is as follows described in technical scheme：
1) membership function and index Fuzzyization processing are selected：
Discrete membership function is selected, 10 indexs are blurred；
2) membership function interval range is determined：
The interval range of psychoacoustic parameter membership function is determined with reference to the opinion that expert provides；
3) Calculation Estimation matrix：
Alternative collection and each membership of factor function with reference to determination, input membership function by the value of each factor and then obtain
The Evaluations matrix R of each factor；
4) fuzzy overall evaluation formula is calculated：
The Evaluations matrix R that each factor is calculated and weight sets A computings, obtain model of fuzzy synthetic evaluation：
B represents overall merit vector.
Compared with prior art the beneficial effects of the invention are as follows：
It is of the present invention to be ground based on the in-vehicle sound quality Forecasting Methodology of time domain dynamic analysis with traditional sound quality
Study carefully and compare, maximum difference is to have taken into full account when vehicle routinely travels the non-of as caused by various factors in-car acoustic environment
Stable situation, it is proposed that the unstable state based on the time domain dynamic characteristic index of each sound quality objective evaluation parameter and each parameter
The method for building up of sound quality comprehensive evaluation model under acoustic environment.
In addition, sound quality research different from the past selects to replace whole acoustic environment with the single sample sound of second-time
Analyzed and researched, the present invention proposes the cumulative effect that can reflect reflection human auditory's psychological feelingses, enterprising in second-time
Row sound time-domain signal framing, and it is named as " long time frame ".And trapezoid window adding window is used to long time frame, eliminating border effect
Again the characteristic of voice signal sample should can be enable to be sufficiently reserved simultaneously.
In-vehicle sound quality Forecasting Methodology of the present invention based on time domain dynamic analysis is versatile, with more reality
Border meaning, real train test is both can apply to, carry out sound quality program analysis and optimization.
In addition, the method for the invention is not limited only to in-car acoustic environment, while it is also applied for the sound product of other enclosed environments
Matter is analyzed and prediction.
Brief description of the drawings
The present invention is further illustrated below in conjunction with the accompanying drawings：
Fig. 1 is a kind of in-vehicle sound quality Forecasting Methodology main-process stream based on time domain dynamic analysis of the present invention
Figure；
Fig. 2 is the speech articulation of the in-vehicle sound quality Forecasting Methodology of the present invention based on time domain dynamic analysis
Membership function figure；
Fig. 3 is the speech articulation of the in-vehicle sound quality Forecasting Methodology of the present invention based on time domain dynamic analysis
Variance membership function figure；
Fig. 4 is the subjective index of the in-vehicle sound quality Forecasting Methodology of the present invention based on time domain dynamic analysis
Membership function figure；
Embodiment
The present invention is explained in detail below in conjunction with the accompanying drawings：
Spirit of the invention is when routinely being travelled to overcome current in-vehicle sound quality analysis and research can not describe vehicle more
The problem of time domain dynamic characteristic of in-car acoustic environment, while in order to in conventional in-vehicle sound quality optimization, it is real each time
The transformation of car sound quality will carry out cumbersome sound quality subjective assessment and the objective psychoacoustic parameter of sound quality analyzes showing for experiment
Shape, it is proposed that the unstable state acoustic environment based on the time domain dynamic characteristic index of each sound quality objective evaluation parameter and each parameter
Lower sound quality Forecasting Methodology.The present invention is solved when vehicle routinely travels by various due to introducing time domain dynamic analysis
Sound quality problem analysis in the case of the unstable state of in-car acoustic environment caused by factor.Meanwhile simplify passenger car in-car sound product
The experimental procedure of matter evaluation, establishes subjective and objective corresponding relation, and more conventional research is more of practical significance.
In in-vehicle sound quality Forecasting Methodology based on time domain dynamic analysis, equipment used and computational methods are as follows：
Described sound collection equipment can be dummy head or microphone.Wherein dummy head is simulation real head's shoulder chi
Very little manikin, ear's structure of dummy head are to simulate the internal structure of human ear, and two microphones are separately mounted to two ears
In piece, the sound field environment of auditor can be truly reduced.
Described microphone refers to the general sound transducer of in the market, as Mike.
Described computation model is the comprehensive evaluation model established using fuzzy mathematics theory：(1) determine that fuzzy synthesis is commented
Valency set of factors；(2) alternative collection is established；(3) weight is determined；(4) membership function is determined.
Described windowing process, the software for calculation Programming window type of specialty, window length and edge length can be applied.
A kind of in-vehicle sound quality Forecasting Methodology based on time domain dynamic analysis, comprises the following steps：
Step 1：Using sound collection equipment in minute magnitude voice signal sample in collecting vehicle；
Voice signal sample in collecting vehicle, is comprised the following steps that in the step 1：
(1) sound collection equipment cloth is placed at copilot, the vertical coordinate of sound collection equipment particular location is seat
Chair surface center line, abscissa are backrest surface center lines, and more than both intersection points at 0.7 ± 0.05m, dummy head or microphone are symmetrically
It is arranged at the plane of symmetry of seat surface and backrest or so 0.2 ± 0.02m；
(2) road conditions and operating mode are selected：
It is existing with reference to country《Highway technical standard》The grade road surface of selection test vehicle traveling, and design alternative is more
Individual representational vehicle driving-cycle, such as more representational operating modes of idling and at the uniform velocity 35km/h；
(3) in-car voice signal sample collection；
Using sound collection equipment, for selecting the passenger vehicle run under road conditions, the in-car sound of each operating mode is recorded
Sample of signal, 5~6 minutes time is recorded every time, ensure that enough durations divide for long time frame；
Step 2：Framing windowing process in-car voice signal sample is carried out to in-car voice signal sample signal to sound sample
This carries out framing on second-time, using long time frame trapezoid window, is eliminating boundary effect simultaneously, is retaining the spy of voice signal sample
Property；
In order to reflect the cumulative effect of the dynamic characteristic of in-car acoustic environment and human auditory's psychological feelingses, the present invention
It is proposed that voice signal sample carried out sound time-domain signal point with 2 seconds for a frame length in collecting vehicle in sound quality research in the car
Frame.And propose to be applied to the trapezoid window adding window method of long time frame processing, it can make voice signal sample again simultaneously eliminating boundary effect
This characteristic is sufficiently reserved.In design parameter selection, it is contemplated that impression integration time of the human ear for sound quality parameter is not
Should be too short, accumulated with reference to impression of the mankind to music, final to determine a length of 2s of window, edge length is the percent of window length
One, i.e., 10 milliseconds.The software for calculation Programming window type of specialty can be applied.
Framing windowing process is carried out in the step 2 to in-car sample sound signal to comprise the following steps that：
(1) in-car voice signal sample framing：
Sound time-domain signal framing was carried out for a frame length with 2 seconds to in-car voice signal sample；
(2) window type is selected：
Select in the trapezoid window for eliminating boundary effect simultaneously and the characteristic of voice signal sample being enable to be sufficiently reserved；
(3) window shape parameter is set：
Window shape parameter is defined as：Window length is 2 seconds, and edge length is 10 milliseconds.
Step 3：Voice signal sample is divided into training sample and detection sample, subjective assessment point is carried out to training sample
Analysis；
In Appraising subject quantity, the evaluation test of most of psychologic acoustics, the subjective assessment person of 20 and the above can
To draw more accurately acoustic assessment result.Based on this principle, the present invention recommends to choose 24 testing crews as sound product
Matter subjective assessment main body, the ratio of men and women is 1 in Appraising subject:1, the age, the experiment for having driving experience was commented between 18~70
The quantitative proportion of the quantity of valency personnel and the test evaluation personnel without driving experience are 1:1, Review Team's composition of personnel need to consider to the greatest extent
Different regions, occupation, sex, age, culture background and habits and customs etc. may be covered.By being familiar with field experiment ring in one's power
Border, and explain implication of experiment process and appraised index etc. training after reviewer form jury to framing adding window after
Sample sound carries out sound quality subjective assessment.Subjective assessment can choose suitable sound quality evaluation according to vehicle and crowd's needs
Index, evaluation index of the present invention are acoustic environment comfort level, and sample sound marking is carried out by 0-10 score value using Numerical value method
Evaluation.Sound quality subjective assessment of the present invention does not limit to and the method, can also use and be carried out the methods of grade scoring method.
Comprising the following steps that for subjective assessment analysis is carried out in the step 3：
(1) composition of personnel jury is chosen：
More than 20 testing crews be have chosen as sound quality subjective assessment main body, the ratio of men and women is 1 in Appraising subject:
1, between 18~70, there is the quantity of test evaluation personnel of driving experience at the age with the test evaluation personnel's without driving experience
Quantitative proportion is 1:1；
(2) jury personnel are carried out testing preceding training；
Allow jury personnel to be familiar with field experiment environment in one's power by training, and explain experiment process for jury personnel and comment
Examine the implication of index.
(3) sound quality evaluation index is determined：
Evaluation index is acoustic environment comfort level, and it describes the degree that in-car sound is disturbed to caused by passenger, upset,
Subjective assessment personnel are needed to give a mark the level of comfort that in-car acoustic environment be present.
(4) evaluation method is determined：
Using acoustic environment comfort level as index, sample sound marking evaluation is carried out using Numerical value method by jury personnel.
Step 4：Calculate the objective psychoacoustic parameter of sound quality
The objective psychoacoustic parameter of sound quality includes：Loudness, sharpness, roughness, shake degree and speech articulation；
Comprising the following steps that for the objective psychoacoustic parameter of sound quality is calculated in the step 4：
(1) sample after framing windowing process is extracted, carries out loudness calculating：
In formula：N ' is characteristic loudness, unit sone；
ETQFor excitation corresponding with the threshold of audibility；
E is to be encouraged corresponding to acoustical signal；
E0To encourage, with reference sound intensity I0=10-12w/m2It is corresponding；
Loudness N is frequency band feature loudness sum, unit sone, and calculation formula is as follows：
(2) sample after framing windowing process is extracted, carries out sharpness calculating；
Sharpness S is calculated using Zwicker models, for its mathematical modeling based on Scale Model of Loudness, mathematical formulae is as follows：
In formula, S is sharpness, unit acum；
K is weight coefficient, and k takes 0.11；
N is total loudness value, unit sone；
N ' (z) is the characteristic loudness in z Bark domains, unit sone；
(3) sample after framing windowing process is extracted, carries out roughness calculating：
Utilize the modulating frequency f of noisemodWith the differential Δ L of excitation in individual feature bandE(z) roughness, formula are calculated
For：
In formula, R is roughness, unit asper；
fmodIt is modulating frequency；
ΔLE(z) it is the variable quantity of acoustical signal drive(r) stage, is defined as：
In formula：N′maxAnd N ' (z)min(z) maximum and minimum value of characteristic loudness are represented respectively；
(4) sample after framing windowing process is extracted, carries out the calculating of shake degree：
Zwicker shake degree computation models are：
Wherein, F represents shake degree, unit vacil；
f0Represent modulation fundamental frequency (f0Take 4Hz)；fmod、ΔLE(z) definition and calculating is identical with roughness.
(5) extract the sample after windowing process to handle it, carry out speech articulation calculating:
AI indexes are obtained according to the analysis to each frequency band, speech articulation is the acoustic pressure that AI indexes depend on ambient noise
Level and frequency (1/3rd octaves), the audible speech range of human ear is in the frequency spectrum of ambient noise when people talk in the car
200~6300Hz region is showed, its computation model is
AI=∑ W (f) D (f)/30 (7)
In formula, W (f) is weighted coefficient；N (f) reaches the standard grade for ambient noise, and its expression formula is UL (f)=H (f)+12dB；LL
(f) it is ambient noise lower limit, its expression formula is LL (f)=UL (f) -30dB.Upper limit noise and meter corresponding to wherein each frequency range
Weight coefficient is as shown in table 1 below
Upper limit noise and weighted coefficient corresponding to 1 each frequency range of table
| Frequency/Hz | Upper limit noise UL (f) | Weighted coefficient W (f) |
| 200 | 64 | 1 |
| 250 | 69 | 2 |
| 315 | 71 | 3.25 |
| 400 | 73 | 4.25 |
| 500 | 75 | 4.5 |
| 630 | 75 | 5.25 |
| 800 | 75 | 6.5 |
| 1000 | 74 | 7.25 |
| 1250 | 72 | 8.5 |
| 1600 | 70 | 1.5 |
| 2000 | 67 | 11 |
| 2500 | 65 | 9.5 |
| 3150 | 63 | 9 |
| 4000 | 60 | 7.75 |
| 5000 | 56 | 6.25 |
| 6300 | 51 | 2.5 |
Step 5：Calculate the time domain dynamic characteristic index of objective psychoacoustic parameter
Time domain dynamic characteristic index includes the variance of each objective psychoacoustic parameter and the pole of each objective psychoacoustic parameter
Difference；The variance of each objective psychoacoustic parameter refers to the variance of loudness, the variance of sharpness, the variance of roughness, the side of shake degree
Difference, the variance of speech articulation；The extreme difference of each objective psychoacoustic parameter refers to the extreme difference of loudness, the extreme difference of sharpness, roughness
Extreme difference, extreme difference, the extreme difference of shake degree of speech articulation；
Comprising the following steps that for the time domain dynamic characteristic index of objective psychoacoustic parameter is calculated in the step 5：
(1) the variance S of each objective psychoacoustic parameter2Calculate
The loudness that calculates by formula, sharpness, roughness, speech articulation, shake degree are entered according to formula of variance
Row calculates：
In formula：xiIt is each parameter with calculated value that 2 seconds are a frame；
For the average value of each parameter calculated value；
N is the number of each parameter calculated value；
(2) the extreme difference R of each objective psychoacoustic parameter is calculated
R=xmax-xmin (10)
In formula：xmaxWith 2 seconds be for each parameter a frame calculated value in maximum；
xminWith 2 seconds be for each parameter a frame calculated value in minimum value.
In calculating process the software for calculation of specialty can be used to calculate.
Step 6：Establish the subjective and objective comprehensive evaluation model of in-vehicle sound quality；
Comprising the following steps that for the objective comprehensive evaluation model of in-vehicle sound quality is established in the step 6：
(1) fuzzy overall evaluation set of factors is determined；
During the fuzzy evaluation of sound quality, the time domain dynamic characteristic for choosing each objective psychoacoustic parameter and each parameter refers to
It is denoted as evaluation parameter：
U=speech articulation, and the variance of speech articulation, loudness, the variance of loudness, roughness, the variance of roughness,
Sharpness, the variance of sharpness, shake degree, the variance of degree of shake }；
(2) alternative collection is established；
Alternative collection is set of each factor to the evaluation result of evaluation object, is a kind of vector representation：
V={ can not receive, it is impossible to receive, typically, can receive, take in good part } completely；
(3) weight is determined；
Weight is determined described in above-mentioned (3) technical scheme, refers to the analytic hierarchy process (AHP) construction combined using qualitative and quantitative
Judgment matrix, determines the weight of the variance of objective psychoacoustic parameter and each parameter according to judgment matrix, and detailed process is as follows：
1) level is determined to each factor of evaluation object first, similar factor is classified as one kind and dropped to a lower level, come with this
Determine the hierarchical structure of each factor.
2) objective indicator and the correlation coefficient value and judgment matrix scale of subjective evaluation index are referred to, more same level
Each factor, judgment matrix is supplemented according to the difference of importance between each factor；
Refering to Fig. 2, by sharpness variance with exemplified by the correlation of sound comfort level, and judgment matrix scale, refering to table 2, than
Each factor of more same level, judgment matrix is supplemented according to the difference of importance between each factor, refering to shown in table 3.
The judgment matrix scale of table 2 and its meaning
| Scale value | Implication |
| 1 | AjWith AiThere is same significance level to evaluation object |
| 3 | AjCompare AiIt is somewhat important |
| 5 | AjIt is obvious important |
| 7 | AjIt is strong important |
| 9 | AjIt is extremely important |
| 2,4,6,8 | The median of adjacent two grade |
The judgment matrix of table 3
In table 3, ai jI is compared with j for expression factor, the ratio of significance level.
3) eigenvalue and eigenvector of calculating matrix, consistency check is carried out to judgment matrix；
The eigenvalue and eigenvector of above-mentioned 3) calculating matrix described in technical scheme, uniformity inspection is carried out to judgment matrix
That tests comprises the following steps that：
A, calculating matrix A eigenvalue of maximum λmax；
B, the coincident indicator C.L shown in solution formula (11)；
λmaxFor the eigenvalue of maximum of matrix A, n is exponent number；
C. the table of comparisons 4 finds out Aver-age Random Consistency Index R.L. corresponding to matrix
The mean random uniformity R.L. tables of table 4
| Exponent number | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 |
| R.L. | 0.00 | 0.00 | 0.58 | 0.90 | 1.12 | 1.24 | 1.32 | 1.41 | 1.45 | 1.49 | 1.52 | 1.54 |
D. the Consistency Ratio C.R. shown in solution formula (12)
If e, during C.R. ＞ 0.1, it need to rebuild or correct judgment matrix；If C.R. ＜ 0.1, it is considered that judgment matrix
Uniformity can receive；
If 4) matrix consistency inspection is qualified, the calculating of weight vectors is carried out.
The calculating process of the above-mentioned 4) weight vectors described in technical scheme is as follows：
A. matrix A is pressed into row normalization：
B. the data after each row normalization are added by row：
C. obtain and vector is normalized, obtains weight vectors：
According to step described above, it is determined that the hierarchical structure of each factor of evaluation, and then determine the weight of each level to
Amount, total weight vectors are A.
(4) membership function is determined；
Before collective model is established, it have selected discrete trigonometric shape membership function and be blurred 10 indexs；10
Individual index is：Speech articulation, the variance of speech articulation, loudness, the variance of loudness, roughness, the variance of roughness, sharply
Degree, the variance of sharpness, shake degree, the variance of degree of shake；
The interval range of psychoacoustic parameter membership function is determined with reference to the opinion that expert provides.It is clear with language herein
Exemplified by clear degree, the variance of speech articulation and sound comfort level.Refering to Fig. 3, Fig. 4.
Alternative collection and each membership of factor function with reference to determination, the value of each factor is inputted into membership function, and then obtained
To the Evaluations matrix R of each factor.
Determine that membership function comprises the following steps that described in above-mentioned (4) technical scheme：
1) membership function and index Fuzzyization processing are selected：
Suitable membership function is selected, by the 10 of each objective psychoacoustic parameter and the time domain dynamic characteristic index of each parameter
Individual index is blurred；
2) membership function interval range is determined：
The interval range of psychoacoustic parameter membership function is determined with reference to the opinion that expert provides；
3) Calculation Estimation matrix：
Alternative collection and each membership of factor function with reference to determination, input membership function by the value of each factor and then obtain
The Evaluations matrix R of each factor；
4) fuzzy overall evaluation formula is calculated：
The Evaluations matrix R that each factor is calculated and weight sets A computings, obtain model of fuzzy synthetic evaluation：
B represents overall merit vector.
Step 7：To the objective comprehensive evaluation model output valve of sound quality compared with subjective assessment value, according to comparative result
Step 6 established model is adjusted, until result and actual error, within 20%, degree of error meets to require.
Step 8：Model is verified using test samples；
Model is tested using new test samples, if Comparative result degree error illustrates model within 20%
It can receive, if Comparative result degree is more than 20%, return to step 6, readjust model, meet until in this result
Error requirements.
Embodiment：
For the ease of person skilled deeper into understanding this method, by the way that the method is applied into certain domestic passenger car
The specific implementation case of sound quality prediction illustrates：
1st, for enough data samples, it is test object to choose certain domestic passenger car, during sample sound signal acquisition
The normal one-level high class pavement of vehicle i.e. asphalt concrete pavement or cement concrete pavement is asked to travel, it is more with idling and 35Km/h etc.
Individual operating mode traveling, is mounted with copilot position by the method for dummy head as described above, records in-car sound under each operating mode
Sound signal sample.Each working state recording 3 times, record time are more than 5 minutes, in-car each operating mode sample interception of voice signal sample
Length is 4 minutes.
2nd, using matlab Programming adding windows.The selection of window type is trapezoid window, and a length of 2s of window, edge length is the hundred of window length
/ mono-, i.e., 10 milliseconds.To screen, be segmented, intercept after sample sound carry out windowing process.
3rd, the objective psychologic acoustics of each sample sound is directly calculated using the sound quality analysis module in LMS/test.lab
Parameter (loudness, sharpness, roughness, shake degree, speech articulation) is as shown in table 5.
The objective psychological parameter calculated value of table 5
4th, it is that variance and extreme difference calculate using the time domain dynamic characteristic index calculating of each psychoacoustic parameter of matlab progress,
As shown in table 6, table 7.
Each objective psychological parameter variance calculated value of table 6
Each objective psychological parameter extreme difference calculated value of table 7
5th, the present invention have chosen 24 testing crews as sound quality subjective assessment main body, the ratio of men and women in Appraising subject
For 1: 1, between 18~70, there are the quantity of the test evaluation personnel of driving experience and the test evaluation people without driving experience at the age
The quantitative proportion of member is 1: 1, and trained reviewer forms jury to the sample sound carry out sound after framing adding window
Quality subjective assessment, Review Team's composition of personnel need to consider to cover different regions, occupation, sex, age, culture background as far as possible
With habits and customs etc..
Subjective assessment can choose suitable sound quality evaluation index, evaluation index of the present invention according to vehicle and crowd's needs
For acoustic environment comfort level, sample sound marking evaluation is carried out by 0-10 score value using Numerical value method.It is as follows.
Before evaluation experimental, by operating personnel jury personnel are carried out with the detailed step and points for attention of test method
Explained, and answer the query of jury's proposition, during experiment, jury can hear different sample sounds successively,
Corresponding score value is filled on marking table according to its subjective sensation respectively.During experiment, after some sample sound plays,
Jury can not make to sample sound when clearly giving a mark, and the sample can be carried out again to listen to evaluation, until being satisfied with
Result.The marking value of 24 evaluation personnels is counted, the related-coefficient test between evaluation personnel is carried out using software.
Reject two minimum evaluation personnels of coefficient correlation.
The Pearson correlation coefficient between the subjective marking value of 24 evaluation personnels is calculated, it is as shown in table 8 below：
The coefficient correlation of table 8
| 1 | 2 | 3 | 4 | 5 | 6 | … | 20 | 21 | 22 | 23 | 24 | |
| 1 | 1.00 | 0.78 | 0.68 | 0.87 | 0.73 | 0.71 | … | 0.62 | 0.63 | 0.40 | 0.73 | 0.62 |
| 2 | 0.78 | 1.00 | 0.83 | 0.74 | 0.68 | 0.70 | … | 0.69 | 0.74 | 0.58 | 0.62 | 0.61 |
| 3 | 0.68 | 0.83 | 1.00 | 0.75 | 0.64 | 0.77 | … | 0.70 | 0.73 | 0.45 | 0.69 | 0.75 |
| 4 | 0.87 | 0.74 | 0.75 | 1.00 | 0.87 | 0.76 | … | 0.87 | 0.79 | 0.51 | 0.61 | 0.64 |
| 5 | 0.73 | 0.68 | 0.64 | 0.87 | 1.00 | 0.81 | … | 0.61 | 0.77 | 0.42 | 0.63 | 0.73 |
| 6 | 0.71 | 0.70 | 0.77 | 0.76 | 0.81 | 1.00 | … | 0.76 | 0.85 | 0.40 | 0.65 | 0.83 |
| … | … | … | … | … | … | … | … | … | … | … | … | … |
| 20 | 0.62 | 0.69 | 0.70 | 0.87 | 0.61 | 0.76 | … | 1.00 | 0.59 | 0.36 | 0.61 | 0.77 |
| 21 | 0.63 | 0.74 | 0.73 | 0.79 | 0.77 | 0.85 | … | 0.59 | 1.00 | 0.24 | 0.72 | 0.78 |
| 22 | 0.40 | 0.58 | 0.45 | 0.51 | 0.42 | 0.40 | … | 0.36 | 0.24 | 1.00 | 0.56 | 0.71 |
| 23 | 0.73 | 0.62 | 0.69 | 0.61 | 0.63 | 0.65 | … | 0.61 | 0.72 | 0.56 | 1.00 | 0.70 |
| 24 | 0.62 | 0.61 | 0.75 | 0.64 | 0.73 | 0.83 | … | 0.77 | 0.78 | 0.71 | 0.70 | 1.00 |
It can be seen that the coefficient correlation between No. 22 subjective assessment personnel and other subjective assessment personnel is universal relatively
It is low, so the evaluation result of this two subjective assessment personnel is rejected.
6th, the objective synthesis of in-vehicle sound quality is carried out in the in-vehicle sound quality Forecasting Methodology based on time domain dynamic analysis to comment
Valency model is established；
(1) fuzzy overall evaluation set of factors is determined
In the fuzzy evaluation theory of sound quality, factor is the evaluation parameter chosen, and is expressed as
U=speech articulation, and the variance of speech articulation, loudness, the variance of loudness, roughness, the variance of roughness,
Sharpness, the variance of sharpness, shake degree, the variance of degree of shake }
(2) alternative collection is established
Alternative collection is set of each factor to the evaluation result of evaluation object, be a kind of vector representation, i.e. V={ completely not
It can receive, it is impossible to receive, typically, can receive, take in good part }.
(3) weight is determined
The emphasis for determining appraisement system weight is Judgement Matricies, and the present invention refers to car in-vehicle sound quality fuzzy synthesis
Evaluation model method for building up, and please have abundant automobile sound quality, acoustics experience and for many years the expert of driving age come structure of advising
Make judgment matrix.
The correlation coefficient value with reference to objective indicator and subjective evaluation index is proposed in the determination of weight vectors and judges square
The method that battle array scale is combined.With reference to objective indicator and the correlation coefficient value and judgment matrix scale of subjective evaluation index, compare
Each factor of same level, the difference of importance supplements judgment matrix between each factor.
In objective evaluation index, psychoacoustic parameter and the variance of each psychoacoustic parameter and the pole of psychoacoustic parameter
Factor of the difference as the first level.And the parameter such as speech articulation, loudness in psychoacoustic parameter as the second level because
Element, the variance and extreme difference of each parameter also serve as the factor of the second dimension.It is in-car using analytic hierarchy process (AHP) construction following table microbus
Each rank judgment matrix of objective evaluation index of masking sound environment sound quality.
In the judgment matrix construction of passenger area and being defined below for weight.
The dimension judgment matrix of table 9 first
| Psychoacoustic parameter | Parameters,acoustic variance | Parameters,acoustic extreme difference | |
| Psychoacoustic parameter | 1 | 1.5 | 2 |
| Parameters,acoustic variance | 2/3 | 1 | 1.5 |
| Parameters,acoustic extreme difference | 1/2 | 2/3 | 1 |
It can be calculated, λmax=3.0015, C.I.=0.0007, R.I.=0.58, C.R.=0.0013, therefore this sentences
The uniformity of disconnected matrix is qualified.
Table 10 " psychoacoustic parameter " judgment matrix
| Speech articulation | Loudness | Sharpness | Roughness | Shake degree | |
| Speech articulation | 1 | 1 | 2 | 3 | 4 |
| Loudness | 1 | 1 | 2 | 3 | 5 |
| Sharpness | 1/2 | 1/2 | 1 | 2 | 2 |
| Roughness | 1/3 | 1/3 | 1/2 | 1 | 2 |
| Shake degree | 1/4 | 1/5 | 1/2 | 1/2 | 1 |
It can be calculated, λmax=5.0380, C.I.=0.0095, R.I.=0.90, C.R.=0.0106, therefore this sentences
The uniformity of disconnected matrix is qualified.
Table 11 " each parameter variance " judgment matrix
It can be calculated, λmax=4.0407, C.I.=0.0136, R.I.=0.90, C.R.=0.015, therefore this judges
The uniformity of matrix is qualified.
Table 12 " each parameter extreme difference " judgment matrix
It can be calculated, λmax=5.1745, C.I.=0.0436, R.I.=0.90, C.R.=0.0485, therefore this sentences
The uniformity of disconnected matrix is qualified.
The weight of above-mentioned two-stage factor is calculated, in tabulated below.
The objective evaluation index factor weights at different levels of table 13
I.e. (0.1366,0.1427,0.0729,0.0472,0.0304,0.0967,0.0967,0.0536,0.0226,
0.0501,0.0867,0.0867,0.0428,0.0110,0.0225)
(4) membership function is determined
The membership function such as figure of each sound quality subjective evaluation parameter have selected Triangleshape grade of membership function, functional form
For discrete type function.The speech articulation and speech articulation variance and the degree of membership letter of subjective evaluation index only enumerated herein
Number, refering to Fig. 3, Fig. 4.
It is determined that during membership function, a program is write with matlab, by the subjective and objective parameter and calculated value of acoustic environment
Input program calculates the membership function of each parameter.
Only by taking speech articulation index membership function as an example, it is expressed as below：
Alternative collection and each membership of factor function with reference to determination, the value of each factor is inputted into membership function, and then obtained
To the Evaluations matrix R of each factor.
The Evaluations matrix R that each factor is calculated and weight sets A computings, obtain fuzzy overall evaluation.
The sound quality result that objective evaluation model calculates is as follows.
The objective comprehensive evaluation result of the sound quality of table 14
The sound quality result that subjective assessment calculates is as follows.
The subjective evaluation result of table 15
7th, to the objective comprehensive evaluation model output valve of sound quality compared with subjective assessment value, according to comparative result to step
Rapid 6 established models are adjusted, until result and actual error, within 20%, degree of error meets to require.
8th, the inspection of comprehensive evaluation model
The inspection of sound quality model is carried out to model using new test samples, the experiment in rear 12 kinds of acoustic environments is carried out
Calculate and prediction counts, as a result such as following table.
The model testing result of table 16
| Additional acoustic environment | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 |
| The acceptable degree of prediction | 5.16 | 5.06 | 4.67 | 4.95 | 3.77 | 4.47 | 4.83 | 4.53 | 4.44 | 3.96 | 3.54 | 3.12 |
| Actual acceptable degree | 5.73 | 4.39 | 4.11 | 4.26 | 3.58 | 4.17 | 5.44 | 5.02 | 5.21 | 4.39 | 3.59 | 3.27 |
| Error % | 11 | 13.2 | 12 | 13.9 | 5 | 6.7 | 12.6 | 10.8 | 17.3 | 10.9 | 1.1 | 4.8 |
The sound quality Comprehensive Model established herein and acceptable degree forecast model are can be seen that by model testing result
By control errors within 15%, objective in-vehicle sound quality can be preferably predicted.
Claims (1)
1. a kind of in-vehicle sound quality Forecasting Methodology based on time domain dynamic analysis, it is characterised in that comprise the following steps：
Step 1：Utilize voice signal sample in sound collection equipment collecting vehicle；
Step 2：Framing windowing process is carried out to in-car voice signal sample signal；
Voice signal sample carries out the interception of minute magnitude to sample sound in collecting vehicle, and carries out sound time domain letter with second-time
Number framing, the processing of trapezoid window adding window method is carried out to long time frame, eliminating boundary effect simultaneously, retaining the spy of voice signal sample
Property；
Step 3：Subjective assessment experiment is carried out to internal car noise sample；
Step 4：Calculate the objective psychoacoustic parameter of sound quality；
The objective psychoacoustic parameter of sound quality includes：Loudness, sharpness, roughness, shake degree and speech articulation；
Step 5：Calculate the time domain dynamic characteristic index of objective psychoacoustic parameter；
The time domain dynamic characteristic index of the objective psychoacoustic parameter includes the variance of each objective psychoacoustic parameter and each visitor
See the extreme difference of psychoacoustic parameter；
Step 6：Establish the objective comprehensive evaluation model of in-vehicle sound quality；
Step 7：To the objective comprehensive evaluation model output valve of sound quality compared with subjective assessment value, according to comparative result to step
Rapid 6 established models are adjusted, until result and actual error, within 20%, degree of error meets to require；
Step 8：Model is verified using test samples；
Model is tested using test samples, if Comparative result degree error within 20%, illustrates that model can receive,
If Comparative result degree is more than 20%, return to step 6, model is readjusted, until meeting error requirements in this result；
Voice signal sample in collecting vehicle, is comprised the following steps that in the step 1：
(1) sound collection equipment cloth is placed at copilot, i.e. the vertical coordinate of dummy head or two microphone particular locations
It is seat surface center line, abscissa is backrest surface center line, more than both intersection points at 0.7 ± 0.05m, dummy head or microphone pair
It is arranged at the plane of symmetry of seat surface and backrest or so 0.2 ± 0.02m with claiming；
(2) road conditions and operating mode are selected：
The grade road surface of selection test vehicle traveling, and the multiple representational vehicle driving-cycles of design alternative；
(3) in-car voice signal sample collection：
The in-car voice signal sample of each operating mode is recorded, records 5~6 minutes time every time；
Framing windowing process is carried out in the step 2 to internal car noise sample signal to comprise the following steps that：
(1) in-car voice signal sample framing：
It was that a frame length carries out sound time-domain signal framing with 2 seconds in collecting vehicle a length of 4 minutes during voice signal sample；
(2) window type is selected：
Select in the trapezoid window for eliminating boundary effect simultaneously and the characteristic of voice signal sample being enable to be sufficiently reserved；
(3) window shape parameter is set：
Window shape parameter is defined as：Window length is 2 seconds, and edge length is 10 milliseconds；
Subjective assessment experiment is carried out in the step 3 to internal car noise sample to comprise the following steps that：
(1) composition of personnel jury is chosen：
The testing crew of more than 20 be have chosen as sound quality subjective assessment main body, the ratio of men and women is 1 in Appraising subject:1,
Between 18~70, there are the quantity and the number of the test evaluation personnel without driving experience of the test evaluation personnel of driving experience at age
Amount ratio is 1:1；
(2) jury personnel are carried out testing preceding training；
(3) sound quality evaluation index is determined：
Evaluation index is acoustic environment comfort level；
(4) evaluation method is determined：
Sample sound marking evaluation is carried out with Numerical value method；
Comprising the following steps that for the objective psychoacoustic parameter of sound quality is calculated in the step 3：
(1) sample after framing windowing process is extracted, carries out loudness calculating：
N '=0.08 (ETQ/E0)0.28[(0.5+E/ETQ)0.23-1] (1)
In formula：N ' is characteristic loudness, unit sone；
ETQFor excitation corresponding with the threshold of audibility；
E is to be encouraged corresponding to acoustical signal；
E0To encourage, with reference sound intensity I0=10-12w/m2It is corresponding；
Loudness N is frequency band feature loudness sum, unit sone, and calculation formula is as follows：
<mrow>
<mi>N</mi>
<mo>=</mo>
<msubsup>
<mo>&Integral;</mo>
<mn>0</mn>
<mrow>
<mn>24</mn>
<mi>B</mi>
<mi>a</mi>
<mi>r</mi>
<mi>k</mi>
</mrow>
</msubsup>
<msup>
<mi>N</mi>
<mo>&prime;</mo>
</msup>
<mi>d</mi>
<mi>z</mi>
<mrow>
<mo>(</mo>
<mi>s</mi>
<mi>o</mi>
<mi>n</mi>
<mi>e</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>2</mn>
<mo>)</mo>
</mrow>
</mrow>
(2) sample after framing windowing process is extracted, carries out sharpness calculating；
Sharpness S is calculated using Zwicker models, for its mathematical modeling based on Scale Model of Loudness, mathematical formulae is as follows：
<mrow>
<mi>S</mi>
<mo>=</mo>
<mi>k</mi>
<mfrac>
<mrow>
<munderover>
<mo>&Integral;</mo>
<mn>0</mn>
<mrow>
<mn>24</mn>
<mi>B</mi>
<mi>a</mi>
<mi>r</mi>
<mi>k</mi>
</mrow>
</munderover>
<msup>
<mi>N</mi>
<mo>&prime;</mo>
</msup>
<mrow>
<mo>(</mo>
<mi>z</mi>
<mo>)</mo>
</mrow>
<mi>z</mi>
<mi>g</mi>
<mrow>
<mo>(</mo>
<mi>z</mi>
<mo>)</mo>
</mrow>
<mi>d</mi>
<mi>z</mi>
</mrow>
<mi>N</mi>
</mfrac>
<mrow>
<mo>(</mo>
<mi>a</mi>
<mi>c</mi>
<mi>u</mi>
<mi>m</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>3</mn>
<mo>)</mo>
</mrow>
</mrow>
In formula, S is sharpness, unit acum；
K is weight coefficient, and k takes 0.11；
N is total loudness value, unit sone；
N ' (z) is the characteristic loudness in z Bark domains, unit sone；
(3) sample after framing windowing process is extracted, carries out roughness calculating：
Utilize the modulating frequency f of noisemodWith the differential Δ L of excitation in individual feature bandE(z) roughness is calculated, formula is：
<mrow>
<mi>R</mi>
<mo>=</mo>
<mn>0.3</mn>
<msub>
<mi>f</mi>
<mi>mod</mi>
</msub>
<msubsup>
<mo>&Integral;</mo>
<mn>0</mn>
<mrow>
<mn>24</mn>
<mi>B</mi>
<mi>a</mi>
<mi>r</mi>
<mi>k</mi>
</mrow>
</msubsup>
<msub>
<mi>&Delta;L</mi>
<mi>E</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>z</mi>
<mo>)</mo>
</mrow>
<mi>d</mi>
<mi>z</mi>
<mrow>
<mo>(</mo>
<mi>a</mi>
<mi>s</mi>
<mi>p</mi>
<mi>e</mi>
<mi>r</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>4</mn>
<mo>)</mo>
</mrow>
</mrow>
In formula, R is roughness, unit asper；
fmodIt is modulating frequency；
ΔLE(z) it is the variable quantity of acoustical signal drive(r) stage, is defined as：
<mrow>
<msub>
<mi>&Delta;L</mi>
<mi>E</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>z</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mn>20</mn>
<msub>
<mi>log</mi>
<mn>10</mn>
</msub>
<mrow>
<mo>(</mo>
<mfrac>
<mrow>
<msubsup>
<mi>N</mi>
<mrow>
<mi>m</mi>
<mi>a</mi>
<mi>x</mi>
</mrow>
<mo>&prime;</mo>
</msubsup>
<mrow>
<mo>(</mo>
<mi>z</mi>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msubsup>
<mi>N</mi>
<mi>min</mi>
<mo>&prime;</mo>
</msubsup>
<mrow>
<mo>(</mo>
<mi>z</mi>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>5</mn>
<mo>)</mo>
</mrow>
</mrow>
In formula：N′maxAnd N ' (z)min(z) maximum and minimum value of characteristic loudness are represented respectively；
(4) sample after framing windowing process is extracted, carries out the calculating of shake degree：
Zwicker shake degree computation models are：
<mrow>
<mi>F</mi>
<mo>=</mo>
<mn>0.008</mn>
<mfrac>
<mrow>
<munderover>
<mo>&Integral;</mo>
<mn>0</mn>
<mrow>
<mn>24</mn>
<mi>B</mi>
<mi>a</mi>
<mi>r</mi>
<mi>k</mi>
</mrow>
</munderover>
<msub>
<mi>&Delta;L</mi>
<mi>E</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>z</mi>
<mo>)</mo>
</mrow>
<mi>d</mi>
<mi>z</mi>
</mrow>
<mrow>
<mo>(</mo>
<msub>
<mi>f</mi>
<mi>mod</mi>
</msub>
<mo>/</mo>
<msub>
<mi>f</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
<mo>+</mo>
<mo>(</mo>
<msub>
<mi>f</mi>
<mn>0</mn>
</msub>
<mo>/</mo>
<msub>
<mi>f</mi>
<mi>mod</mi>
</msub>
<mo>)</mo>
</mrow>
</mfrac>
<mrow>
<mo>(</mo>
<mi>v</mi>
<mi>a</mi>
<mi>c</mi>
<mi>i</mi>
<mi>l</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>6</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein, F represents shake degree, unit vacil；
f0Represent modulation fundamental frequency；
(5) sample after framing windowing process is extracted, carries out speech articulation AI calculating:
Speech articulation computation model is：
AI=∑ W (f) D (f)/30 (7)
<mrow>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mrow>
<mi>N</mi>
<mrow>
<mo>(</mo>
<mi>f</mi>
<mo>)</mo>
</mrow>
<mo>></mo>
<mi>U</mi>
<mi>L</mi>
<mrow>
<mo>(</mo>
<mi>f</mi>
<mo>)</mo>
</mrow>
<mo>,</mo>
<mi>D</mi>
<mrow>
<mo>(</mo>
<mi>f</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mi>L</mi>
<mi>L</mi>
<mrow>
<mo>(</mo>
<mi>f</mi>
<mo>)</mo>
</mrow>
<mo><</mo>
<mi>N</mi>
<mrow>
<mo>(</mo>
<mi>f</mi>
<mo>)</mo>
</mrow>
<mo><</mo>
<mi>U</mi>
<mi>L</mi>
<mrow>
<mo>(</mo>
<mi>f</mi>
<mo>)</mo>
</mrow>
<mo>,</mo>
<mi>D</mi>
<mrow>
<mo>(</mo>
<mi>f</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mi>U</mi>
<mi>L</mi>
<mrow>
<mo>(</mo>
<mi>f</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mi>N</mi>
<mrow>
<mo>(</mo>
<mi>f</mi>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mi>N</mi>
<mrow>
<mo>(</mo>
<mi>f</mi>
<mo>)</mo>
</mrow>
<mo><</mo>
<mi>L</mi>
<mi>L</mi>
<mrow>
<mo>(</mo>
<mi>f</mi>
<mo>)</mo>
</mrow>
<mo>,</mo>
<mi>D</mi>
<mrow>
<mo>(</mo>
<mi>f</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mn>30</mn>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>8</mn>
<mo>)</mo>
</mrow>
</mrow>
In formula, W (f) is weighted coefficient；
N (f) reaches the standard grade for ambient noise, and its expression formula is UL (f)=H (f)+12dB；
LL (f) is ambient noise lower limit, and its expression formula is LL (f)=UL (f) -30dB；
Comprising the following steps that for the time domain dynamic characteristic index of objective psychoacoustic parameter is calculated in the step 4：
(1) the variance S of each objective psychoacoustic parameter2Calculate
The loudness that calculates by formula, sharpness, roughness, speech articulation, shake degree are counted according to formula of variance
Calculate：
<mrow>
<msup>
<mi>S</mi>
<mn>2</mn>
</msup>
<mo>=</mo>
<mfrac>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<mo>-</mo>
<mover>
<mi>x</mi>
<mo>&OverBar;</mo>
</mover>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
<mrow>
<mi>n</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</mfrac>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>9</mn>
<mo>)</mo>
</mrow>
</mrow>
In formula：xiIt is each parameter with calculated value that 2 seconds are a frame；
For the average value of each parameter calculated value；
N is the number of each parameter calculated value；
(2) the extreme difference R of each objective psychoacoustic parameter is calculated
R=xmax-xmin (10)
In formula：xmaxWith 2 seconds be for each parameter a frame calculated value in maximum；
xminWith 2 seconds be for each parameter a frame calculated value in minimum value；
Comprising the following steps that for the objective comprehensive evaluation model of in-vehicle sound quality is established in the step 6：
(1) fuzzy overall evaluation set of factors is determined；
During the fuzzy evaluation of sound quality, the time domain dynamic characteristic index for choosing each objective psychoacoustic parameter and each parameter is made
For evaluation parameter：
U=speech articulation, and the variance of speech articulation, loudness, the variance of loudness, roughness, the variance of roughness, sharply
Degree, the variance of sharpness, shake degree, the variance of degree of shake }；
(2) alternative collection is established；
Alternative collection is set of each factor to the evaluation result of evaluation object, is a kind of vector representation：
V={ can not receive, it is impossible to receive, typically, can receive, take in good part } completely；
(3) weight is determined；
(4) membership function is determined；
The determination weight, refer to determine objective psychoacoustic parameter and each ginseng using the analytic hierarchy process (AHP) that qualitative and quantitative combines
The weight of several variances, Judgement Matricies, detailed process are as follows accordingly：
1) level is determined to each factor of evaluation object first, similar factor is classified as one kind and dropped to a lower level, determined with this
The hierarchical structure of each factor；
2) with reference to the correlation coefficient value and judgment matrix scale of objective indicator and subjective evaluation index, more same level it is each because
Element, judgment matrix is supplemented according to the difference of importance between each factor；
3) eigenvalue and eigenvector of calculating matrix, consistency check is carried out to judgment matrix；
If 4) matrix consistency inspection is qualified, the calculating of weight vectors is carried out, is the characteristic vector of matrix after normalization；
The eigenvalue and eigenvector of the calculating matrix, consistency check is carried out to judgment matrix and comprised the following steps that：
A, calculating matrix A eigenvalue of maximum λmax；
B, the coincident indicator C.L shown in solution formula (11)；
<mrow>
<mi>C</mi>
<mo>.</mo>
<mi>L</mi>
<mo>=</mo>
<mfrac>
<mrow>
<msub>
<mi>&lambda;</mi>
<mi>max</mi>
</msub>
<mo>-</mo>
<mi>n</mi>
</mrow>
<mrow>
<mi>n</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</mfrac>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>11</mn>
<mo>)</mo>
</mrow>
</mrow>
λmaxFor the eigenvalue of maximum of matrix A, n is exponent number；
C, Aver-age Random Consistency Index R.L. corresponding to matrix is found out
D, the Consistency Ratio C.R. shown in solution formula (12)；
<mrow>
<mi>C</mi>
<mo>.</mo>
<mi>R</mi>
<mo>.</mo>
<mo>=</mo>
<mfrac>
<mrow>
<mi>C</mi>
<mo>.</mo>
<mi>L</mi>
</mrow>
<mrow>
<mi>R</mi>
<mo>.</mo>
<mi>L</mi>
</mrow>
</mfrac>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>12</mn>
<mo>)</mo>
</mrow>
</mrow>
If e, during C.R. ＞ 0.1, it need to rebuild or correct judgment matrix；If C.R. ＜ 0.1, it is considered that the one of judgment matrix
Cause property can receive；
If matrix consistency inspection is qualified, carry out the calculating of weight vectors, be the characteristic vector of matrix after normalization, feature to
The solution procedure of amount is as follows：
A. matrix A is pressed into row normalization：
<mrow>
<msub>
<mi>b</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<mo>=</mo>
<mfrac>
<msub>
<mi>a</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msub>
<mi>a</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
</mrow>
</mfrac>
<mo>,</mo>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
<mo>,</mo>
<mn>2</mn>
<mo>,</mo>
<mo>...</mo>
<mo>,</mo>
<mi>n</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>13</mn>
<mo>)</mo>
</mrow>
</mrow>
B. the data after each row normalization are added by row：
<mrow>
<msub>
<mi>W</mi>
<mi>i</mi>
</msub>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msub>
<mi>b</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>14</mn>
<mo>)</mo>
</mrow>
</mrow>
C. obtain and vector is normalized, obtains weight vectors：
<mrow>
<msub>
<mover>
<mi>W</mi>
<mo>&OverBar;</mo>
</mover>
<mi>i</mi>
</msub>
<mo>=</mo>
<mfrac>
<msub>
<mi>W</mi>
<mi>i</mi>
</msub>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msub>
<mi>b</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
</mrow>
</mfrac>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>15</mn>
<mo>)</mo>
</mrow>
</mrow>
According to step described above, it is determined that the hierarchical structure of each factor of evaluation, and then the weight vectors of each level are determined, always
Weight vectors are A；
The determination membership function step is as follows：
1) membership function and index Fuzzyization processing are selected：
Discrete membership function is selected, 10 indexs are blurred；
2) membership function interval range is determined：
The interval range of psychoacoustic parameter membership function is determined with reference to the opinion that expert provides；
3) Calculation Estimation matrix：
With reference to alternative collection and each membership of factor function of determination, by the value of each factor input membership function so that obtain it is each because
The Evaluations matrix R of element；
4) fuzzy overall evaluation formula is calculated：
The Evaluations matrix R that each factor is calculated and weight sets A computings, obtain model of fuzzy synthetic evaluation：
B represents overall merit vector.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510415624.4A CN105021275B (en) | 2015-07-15 | 2015-07-15 | In-vehicle sound quality Forecasting Methodology based on time domain dynamic analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510415624.4A CN105021275B (en) | 2015-07-15 | 2015-07-15 | In-vehicle sound quality Forecasting Methodology based on time domain dynamic analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105021275A CN105021275A (en) | 2015-11-04 |
| CN105021275B true CN105021275B (en) | 2018-01-19 |
Family
ID=54411414
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510415624.4A Expired - Fee Related CN105021275B (en) | 2015-07-15 | 2015-07-15 | In-vehicle sound quality Forecasting Methodology based on time domain dynamic analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105021275B (en) |
Families Citing this family (51)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107123427B (en) * | 2016-02-21 | 2020-04-28 | 珠海格力电器股份有限公司 | Method and device for determining noise sound quality |
| CN106384598A (en) * | 2016-08-18 | 2017-02-08 | 海信（山东）空调有限公司 | Noise quality determination method and device |
| CN108335704A (en) * | 2017-01-19 | 2018-07-27 | 晨星半导体股份有限公司 | Vagitus detection circuit and relevant detection method |
| CN106933146B (en) * | 2017-03-14 | 2023-04-14 | 吉林大学 | Pedestrian warning sound control method for electric car |
| CN108663115A (en) * | 2017-03-31 | 2018-10-16 | 华晨汽车集团控股有限公司 | A kind of car inside idle noise objective quantification evaluation method |
| CN107084851A (en) * | 2017-04-18 | 2017-08-22 | 常州大学 | Method based on psychoacoustic parameter in statistics Energy Flow Analysis prediction high-speed train |
| CN107238502B (en) * | 2017-05-25 | 2020-09-08 | 西南交通大学 | Method for evaluating transmission abnormal sound of vehicle hydraulic shock absorber |
| US10520356B2 (en) * | 2018-01-05 | 2019-12-31 | Center For Integrated Smart Sensors Foundation | Apparatus, method and monitoring system for measuring noise between floors |
| CN108357445A (en) * | 2018-03-20 | 2018-08-03 | 吉林大学 | Car masking sound quality self-adapting control system and method |
| CN108827452B (en) * | 2018-03-27 | 2020-07-31 | 江苏大学 | Noise evaluation method of permanent magnet synchronous hub motor |
| CN109030014B (en) * | 2018-06-06 | 2020-11-27 | 上海汽车集团股份有限公司 | Prediction method for subjective scoring of noise in vehicle during vehicle acceleration |
| CN108896274B (en) * | 2018-06-14 | 2019-12-27 | 大连理工大学 | Distributed optical fiber strain demodulation method based on subset window length optimization algorithm |
| CN110751959A (en) * | 2018-07-24 | 2020-02-04 | 上汽通用五菱汽车股份有限公司 | Method for evaluating noise discomfort degree of automobile |
| CN109241838A (en) * | 2018-07-27 | 2019-01-18 | 重庆理工大学 | Speed changer quality evaluating method based on psychologic acoustics objective parameter |
| CN109141623B (en) * | 2018-08-14 | 2020-11-20 | 中车青岛四方机车车辆股份有限公司 | Method and device for evaluating sound quality in train |
| CN109186750A (en) * | 2018-08-24 | 2019-01-11 | 重庆长安汽车股份有限公司 | A kind of inside gadget to vehicle every sound absorption harmony qualitative effects prediction technique |
| CN109668626A (en) * | 2018-12-25 | 2019-04-23 | 东莞材料基因高等理工研究院 | A kind of sound quality evaluation method based on human-computer interaction interface |
| CN109919511B (en) * | 2019-03-20 | 2021-02-26 | 西南交通大学 | Existing railway line shape evaluation method and system |
| CN110296847B (en) * | 2019-04-17 | 2021-06-01 | 开沃新能源汽车集团有限公司 | Noise testing method for new energy passenger car |
| CN110110437B (en) * | 2019-05-07 | 2023-08-29 | 中汽研（天津）汽车工程研究院有限公司 | Automobile high-frequency noise prediction method based on related interval uncertainty theory |
| CN110208002A (en) * | 2019-05-28 | 2019-09-06 | 上汽通用五菱汽车股份有限公司 | A kind of sound quality subjective assessment system based on LAN backbone |
| CN110210153B (en) * | 2019-06-10 | 2022-09-13 | 西北工业大学 | Single crystal blade impact test measurement data correction method and correction system |
| CN110285982B (en) * | 2019-07-31 | 2021-02-05 | 重庆长安汽车股份有限公司 | Method and device for evaluating relative change of sound quality of automobile or engine and controller |
| CN110503937B (en) * | 2019-07-31 | 2023-03-31 | 江苏大学 | Sound quality optimization system and method |
| CN110737206A (en) * | 2019-09-16 | 2020-01-31 | 五邑大学 | method, device and storage medium for quantifying driving feeling based on vehicle body dynamics |
| CN110737970A (en) * | 2019-09-24 | 2020-01-31 | 天津大学 | engine acceleration sound quality evaluation method |
| CN110648087B (en) * | 2019-11-15 | 2024-06-25 | 吉林大学 | Method and system for evaluating driving performance of sound quality in vehicle |
| CN110926592B (en) * | 2019-11-15 | 2021-11-02 | 四川大学 | Blasting vibration comfort evaluation method |
| CN111598395A (en) * | 2020-04-16 | 2020-08-28 | 天津大学 | Engine sound quality comprehensive evaluation method |
| CN113758720B (en) * | 2020-06-01 | 2023-11-24 | 陕西汽车集团股份有限公司 | Truck driving jitter detection method |
| CN111982274B (en) * | 2020-07-22 | 2022-09-09 | 江苏大学 | Motor noise source identification method and system based on sound quality contribution coefficient |
| CN112067117B (en) * | 2020-08-14 | 2022-10-28 | 中国第一汽车股份有限公司 | Method for evaluating automobile wind noise performance |
| CN112097894A (en) * | 2020-08-17 | 2020-12-18 | 浙江工业大学 | Method for detecting radiation noise qualification of horizontal driver of automobile seat |
| CN112215469A (en) * | 2020-09-15 | 2021-01-12 | 中国第一汽车股份有限公司 | Sound design method for automobile key switch |
| CN112163189A (en) * | 2020-11-03 | 2021-01-01 | 北京机电工程研究所 | Equipment fault diagnosis method based on test sequence probability matrix |
| CN112665705B (en) * | 2020-11-27 | 2022-10-21 | 浙江大学 | Distributed hearing test method |
| CN112932489A (en) * | 2020-12-30 | 2021-06-11 | 广东电网有限责任公司电力科学研究院 | Transformer substation noise subjective annoyance degree evaluation model establishing method and model establishing system |
| CN112884286B (en) * | 2021-01-22 | 2024-03-15 | 中国航空工业集团公司北京长城航空测控技术研究所 | Comprehensive evaluation method for ground test scheme of aviation equipment |
| CN115112383A (en) * | 2021-03-17 | 2022-09-27 | 华晨宝马汽车有限公司 | Evaluation system and method for vehicle noise performance |
| CN113091875B (en) * | 2021-04-20 | 2024-06-25 | 中国汽车工程研究院股份有限公司 | System and method for detecting and evaluating abnormal sound in vehicle |
| CN113125000B (en) * | 2021-04-20 | 2022-11-29 | 中国汽车工程研究院股份有限公司 | Abnormal sound grade judging method for in-vehicle air conditioning system |
| CN113435116B (en) * | 2021-06-24 | 2022-07-05 | 武汉理工大学 | Sound quality self-adaptive design method and device based on virtual driving stand |
| CN113782053B (en) * | 2021-09-04 | 2023-09-22 | 天津大学 | Automatic monitoring method for urban sound landscape quality worthy of protection |
| CN113702071B (en) * | 2021-09-18 | 2022-08-16 | 燕山大学 | NVH evaluation result prediction method under idle working condition |
| CN114046999A (en) * | 2021-09-28 | 2022-02-15 | 上海汽车制动系统有限公司 | Psychoacoustic analysis method for electronic brake |
| CN114912781A (en) * | 2022-04-29 | 2022-08-16 | 中国第一汽车股份有限公司 | Vehicle door sound quality subjective evaluation method, device and equipment based on weight-counting analysis |
| CN115444381B (en) * | 2022-09-19 | 2024-07-02 | 吉林大学 | Method and system for evaluating sound quality in vehicle based on physiological signals |
| CN115796705B (en) * | 2023-01-31 | 2023-05-12 | 中汽研新能源汽车检验中心(天津)有限公司 | New energy automobile electric drive assembly NVH grading evaluation method |
| CN116629655B (en) * | 2023-03-22 | 2024-04-16 | 哈尔滨工业大学 | Comprehensive evaluation method for non-native language classroom sound environment of colleges and universities based on learning efficiency |
| CN117990204B (en) * | 2024-04-03 | 2024-06-25 | 深圳市美格信测控技术有限公司 | Motor noise evaluation method and system under multiple working conditions of electric automobile |
| CN117990378B (en) * | 2024-04-07 | 2024-07-19 | 潍柴动力股份有限公司 | Engine sound quality evaluation method and device, electronic equipment and storage medium |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100580783B1 (en) * | 2004-05-03 | 2006-05-16 | 삼성전자주식회사 | Method and apparatus for measuring the speech quality according to measuring mode |
| CN101672690A (en) * | 2009-09-27 | 2010-03-17 | 吉林大学 | Method for objectively and quantifiably evaluating noise fret degree in vehicle based on auditory model |
| CN104102803B (en) * | 2013-04-07 | 2017-02-08 | 上海工程技术大学 | Vehicle noise sound quality roughness information processing method |
| CN103471709A (en) * | 2013-09-17 | 2013-12-25 | 吉林大学 | Method for predicting noise quality of noise inside passenger vehicle |
-
2015
- 2015-07-15 CN CN201510415624.4A patent/CN105021275B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN105021275A (en) | 2015-11-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105021275B (en) | In-vehicle sound quality Forecasting Methodology based on time domain dynamic analysis | |
| CN103471709A (en) | Method for predicting noise quality of noise inside passenger vehicle | |
| Wang et al. | A sound quality model for objective synthesis evaluation of vehicle interior noise based on artificial neural network | |
| CN110688712A (en) | Evaluation index for objective annoyance degree of automobile wind vibration noise sound quality and calculation method thereof | |
| CN102324229B (en) | Method and system for detecting abnormal use of voice input equipment | |
| CN106933146A (en) | Electrocar pedestrian's caution sound method for designing, caution sound control system and method | |
| CN108491999A (en) | A kind of objective quantification method to electric vehicle steady-state noise subjective assessment | |
| Jin et al. | Evaluation and modeling of automotive transmission whine noise quality based on MFCC and CNN | |
| CN101672690A (en) | Method for objectively and quantifiably evaluating noise fret degree in vehicle based on auditory model | |
| Liu et al. | Strategy and implementing techniques for the sound quality target of car interior noise during acceleration | |
| Miśkiewicz et al. | Psychoacoustics in the automotive industry | |
| Liang et al. | The evaluation of vehicle interior impact noise inducing by speed bumps based on multi-features combination and support vector machine | |
| Huang et al. | Subjective discomfort model of the micro commercial vehicle vibration over different road conditions | |
| Qian et al. | Evaluation and optimization of sound quality in high-speed trains | |
| Xie et al. | Research and development of sound quality in portable testing and evaluation system based on self-adaptive neural network | |
| CN110751959A (en) | Method for evaluating noise discomfort degree of automobile | |
| CN112131662A (en) | Passenger car wind noise subjective evaluation objective quantification method | |
| Feng et al. | Sound feature space effects on the performance of annoyance evaluation model based on support vector machine | |
| de la Prida et al. | Methodology for the subjective evaluation of airborne sound insulation through 2-AC and Thurstonian models | |
| Harriet et al. | Auralisation of an urban soundscape | |
| Dromey et al. | Bidirectional interference between simulated driving and speaking | |
| CN206671813U (en) | Pure electric or hybrid pedestrian caution sound control system | |
| CN107832967A (en) | A kind of sound scape degrees of coordination dynamic evaluation method suitable for bamboo grove space | |
| Lee et al. | Annoyance modeling using personal and situational variables for construction site noise in urban areas | |
| Wang et al. | Developing multivariate models for predicting the levels of multi-dimensional critical perceptions due to metro noise inside buildings |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180119 Termination date: 20210715 |