Construction, Prediction Method and System of Interior Sound Quality Satisfaction Prediction Model for Cruise Vehicles

Abstract

The present invention discloses a method and system for constructing and predicting an in-vehicle sound quality satisfaction prediction model for a cruising vehicle, including: Step 1. When the vehicle is in a cruising state, collect the sound data inside the vehicle and score the cruising noise according to the grade scoring method; Step 2. Process the collected sound data to obtain the A-weighted sound pressure level and the unweighted one-third octave spectrum. Perform a quadratic polynomial fitting on the unweighted one-third octave spectrum to obtain the quadratic term coefficient of the fitting curve, the first-order term coefficient of the fitting curve, and the fitting residual norm; Step 3. Repeat the test on the data of multiple vehicles according to the methods in Step 1 and Step 2 to obtain the subjective satisfaction scores and the in-vehicle sound quality quantification indexes of multiple sample vehicles; Step 4. For the data of the multiple sample vehicles obtained in Step 3, establish an in-vehicle sound quality satisfaction prediction model for the cruising vehicle. The model constructed by the present invention can accurately predict the in-vehicle sound quality under the cruising condition.

Description

Technical Field

[0001] The present invention belongs to the technical field of in-vehicle sound quality evaluation for cruise vehicles, and particularly relates to a construction method, prediction method and system for a satisfaction prediction model of in-vehicle sound quality during cruise. Background Art

[0002] The in-vehicle sound quality mainly has evaluation indexes such as psychoacoustic parameters such as loudness, speech intelligibility, jitter, linearity and roughness. Among them, jitter, linearity and roughness are often used to accelerate the in-vehicle sound quality indexes. Loudness and speech intelligibility can be used for the evaluation of in-vehicle sound quality during cruise, but both focus on evaluating the magnitude of sound pressure energy and are difficult to evaluate the sound balance in the cruise vehicle.

[0003] For example, patent document CN111949942A discloses an evaluation method for accelerating the linearity of in-vehicle noise. This method collects the noise data beside the driver's ear, and calculates the main order noise during accelerating driving according to the effective noise data, the engine speed step and the frequency resolution of the noise post-processing software under the condition of setting the engine speed step and the frequency resolution of the noise post-processing software, so as to obtain the main order noise engine speed curve; perform a unary linear fitting on the main order noise engine speed curve to obtain the objective parameter of the main order noise; substitute the objective parameter of the main order noise into the linear regression model to obtain the evaluation score of the accelerating in-vehicle noise linearity. This evaluation method can solve the problem that the linearity index is difficult to quantify or the quantification method is unreasonable.

[0004] Another example is that patent document CN111751119A discloses an evaluation method for the accelerating sound quality of an automobile based on the frequency characteristics of sound orders. Through the correlation study of the engine order noise components, frequency components and engine speed for the accelerating sound quality, the correlations between the total engine noise, order noise, frequency range and speed range and other indexes and the accelerating sound quality are obtained, and the recommended values of these indexes are established. The core content is to define the accelerating sound quality type through subjective evaluation, conduct objective data research on the order, frequency and speed of various types of sound quality, and summarize and define their value ranges through big data, so as to have a certain guiding significance for the development of automobile sound quality.

[0005] The above two methods respectively propose linearity quantification indexes for the accelerating in-vehicle sound quality, but are not applicable to the cruise constant speed condition. Therefore, it is necessary to develop an evaluation index for the in-vehicle sound quality during cruise; and, by combining the sound quality index with the sound pressure level, a construction method and system for a satisfaction prediction model are proposed. Summary of the Invention

[0006] The purpose of the present invention is to provide a construction method, prediction method and system for a satisfaction prediction model of in-vehicle sound quality during cruise, and the constructed model can accurately predict and evaluate the in-vehicle sound quality under cruise conditions.

[0007] In a first aspect, a method for constructing a prediction model for the satisfaction of in-vehicle sound quality of a cruising vehicle according to the present invention includes the following steps:

[0008] Step 1. When the vehicle is in a cruising state, collect the sound data inside the vehicle, and score the cruising noise according to the grading method to obtain the subjective satisfaction score of the vehicle.

[0009] Step 2. Process the collected sound data to obtain the A-weighted sound pressure level G and the unweighted one-third octave spectrum P(N) (where N represents the ordinal number corresponding to the center frequency, and the value range is 14 - 35). Perform a quadratic polynomial fitting on the unweighted one-third octave spectrum P(N), and record the fitting residual norm as D. The residual norm D characterizes the smoothness of the spectrum P(N) curve and reflects the energy ratio balance of each frequency band of the sound. Therefore, this paper proposes to use the residual norm D as the quantization index of in-vehicle sound quality during cruising.

[0010] Step 3. Repeat the tests on the data of multiple vehicles according to the methods of Step 1 and Step 2 to obtain the subjective satisfaction scores, sound pressure levels, and in-vehicle sound quality indexes of multiple sample vehicles.

[0011] Step 4. For the multi-sample vehicle data obtained in Step 3, establish an expression of the subjective satisfaction score F with respect to the sound pressure level G, the quadratic term coefficient A of the fitting curve, the first-order term coefficient B of the fitting curve, and the fitting residual norm D, which is the prediction model for the satisfaction of in-vehicle sound quality during cruising.

[0012] Optionally, in Step 1, a microphone sensor is arranged on the right side of the headrest of the passenger seat to obtain the sound data at the position of the right ear when a person is sitting.

[0013] Optionally, in Step 1, the test road surface is selected as an asphalt road surface without water accumulation, with a complete and uniform paving, a wind speed less than 5 m / s. The test vehicle keeps moving straight at a preset speed, and the sound data for a preset time is tested; multiple groups of data are repeatedly tested, and each group of tests ensures the same direction and the same starting point.

[0014] Optionally, in Step 2, the tested sound data is screened for consistency, and the screened data is then post-processed to obtain the A-weighted sound pressure level G and the unweighted one-third octave spectrum P(N).

[0015] Optionally, in Step 4, the prediction model for the satisfaction of in-vehicle sound quality during cruising is:

[0016] F = mA + nB + pD + qG + z

[0017] Wherein, F is the subjective satisfaction score, G is the A-weighted sound pressure level, A is the quadratic term coefficient of the fitting curve, B is the linear term coefficient of the fitting curve, D is the fitting residual norm, and m, n, p, q, and z are all constant coefficients.

[0018] In a second aspect, a system for constructing a prediction model for in-vehicle sound quality satisfaction of a cruising vehicle according to the present invention includes a memory and a controller. A computer-readable program is stored in the memory. When the computer-readable program is called by the controller, it can execute the steps of the method for constructing a prediction model for in-vehicle sound quality satisfaction of a cruising vehicle as described in the present invention.

[0019] In a third aspect, a method for predicting in-vehicle sound quality satisfaction of a cruising vehicle according to the present invention includes the following steps:

[0020] Step 1: When the vehicle is in a cruising state, collect the in-vehicle sound data, process the collected sound data to obtain the A-weighted sound pressure level G and the unweighted one-third octave spectrum P(N), and perform a quadratic polynomial fitting on the unweighted one-third octave spectrum P(N) to obtain the quadratic term coefficient A of the fitting curve, the linear term coefficient B of the fitting curve, and the fitting residual norm D.

[0021] Step 2: Use the A-weighted sound pressure level G, the quadratic term coefficient A of the fitting curve, the linear term coefficient B of the fitting curve, and the fitting residual norm D as the inputs of the prediction model for in-vehicle sound quality satisfaction of a cruising vehicle, and the prediction model for in-vehicle sound quality satisfaction of a cruising vehicle outputs the subjective satisfaction score F.

[0022] Wherein, the prediction model for in-vehicle sound quality satisfaction of a cruising vehicle is constructed by using the method for constructing a prediction model for in-vehicle sound quality satisfaction of a cruising vehicle as described in the present invention.

[0023] In a fourth aspect, a system for predicting in-vehicle sound quality satisfaction of a cruising vehicle according to the present invention includes:

[0024] A data acquisition and processing module, which is used for collecting in-vehicle sound data when the vehicle is in a cruising state, processing the collected sound data to obtain the A-weighted sound pressure level G and the unweighted one-third octave spectrum P(N), and performing a quadratic polynomial fitting on the unweighted one-third octave spectrum P(N) to obtain the quadratic term coefficient A of the fitting curve, the linear term coefficient B of the fitting curve, and the fitting residual norm D.

[0025] The subjective satisfaction score prediction module stores a prediction model for the in-vehicle sound quality satisfaction of cruising vehicles, which is used to receive the A-weighted sound pressure level G, the quadratic term coefficient A of the fitting curve, the linear term coefficient B of the fitting curve, and the fitting residual norm D output by the data acquisition and processing module, and takes the A-weighted sound pressure level G, the quadratic term coefficient A of the fitting curve, the linear term coefficient B of the fitting curve, and the fitting residual norm D as the inputs of the prediction model for the in-vehicle sound quality satisfaction of cruising vehicles. The prediction model for the in-vehicle sound quality satisfaction of cruising vehicles outputs the subjective satisfaction score F. The subjective satisfaction score prediction module is connected to the data acquisition and processing module;

[0026] Among them, the prediction model for the in-vehicle sound quality satisfaction of cruising vehicles is constructed by using the construction method of the prediction model for the in-vehicle sound quality satisfaction of cruising vehicles as described in the present invention.

[0027] The present invention has the following advantages: The cruising sound quality index proposed by the present invention can more accurately evaluate the sound balance, and the established prediction model for the in-vehicle sound quality satisfaction of cruising vehicles can more accurately predict and evaluate the satisfaction degree of people with the in-vehicle sound under cruising conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required in the embodiments. It should be understood that the following drawings only show some embodiments of the present invention, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0029] Figure 1 It is a flowchart of the construction method of the prediction model for the in-vehicle sound quality satisfaction of cruising vehicles in this embodiment;

[0030] Figure 2 It is a flowchart of the prediction method for the in-vehicle sound quality satisfaction of cruising vehicles in this embodiment;

[0031] Figure 3 It is a principle block diagram of the prediction system for the in-vehicle sound quality satisfaction of cruising vehicles in this embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] The following will describe the present invention in detail with reference to the drawings.

[0033] As Figure 1 shown, in this embodiment, a construction method of a prediction model for the in-vehicle sound quality satisfaction of cruising vehicles includes the following steps:

[0034] Step 1. Arrange a microphone sensor on the right side of the occupant seat headrest to obtain the sound data at the position of the right ear when a person is sitting. Connect the test equipment, set the sound acquisition bandwidth to 25600 Hz and the frequency resolution to 2 Hz. When the vehicle is in the cruise driving state, collect the sound data inside the vehicle, and score the cruise noise according to the grade scoring method to obtain the subjective satisfaction score F of the vehicle.

[0035] In this embodiment, the test road surface is selected as an asphalt road surface without water accumulation, the road surface is completely and evenly paved, the wind speed is less than 5 m / s, and the test vehicle keeps driving straight at a preset vehicle speed to collect the sound data for a preset time; repeat the test for multiple groups of data (such as 3 groups), and ensure that each group of tests is in the same direction and at the same starting point. When the overall difference between two groups of data is 0.2 dB, it is regarded as valid data.

[0036] Step 2. Conduct consistency screening on the collected sound data, and then post-process the screened data to obtain the A-weighted sound pressure level G and the unweighted one-third octave spectrum P(N). Use the least squares method to perform quadratic polynomial fitting on the unweighted one-third octave spectrum P(N) to obtain Formula 1, and determine the fitted value P ‘ (N) and the values of parameters A, B, and C.

[0037] P‘(N) = AN 2 + BN + C (Formula 1)

[0038] Among them, N represents the ordinal number corresponding to the center frequency, taking values from 14 to 35 (corresponding to the center frequencies of 25 - 3150 Hz), and A, B, and C are all constants fitted according to Formula 1.

[0039] The calculation method of the fitting residual norm D is as follows:

[0040]

[0041] Take the fitting residual norm D as the evaluation standard for measuring the smoothness of the curve, and use it as the quantization index for the balance of the in-cruise vehicle sound quality.

[0042] Step 3. Repeat the test for the data of multiple vehicles (at least 10 vehicles or more) according to the methods of Step 1 and Step 2 to obtain the subjective satisfaction scores F and the quantization indexes of the in-cruise vehicle sound quality of multiple sample vehicles.

[0043] Step 4. For the data of multiple sample vehicles obtained in Step 3, establish an expression of the subjective satisfaction score F with respect to the A-weighted sound pressure level G, the quadratic term coefficient A of the fitting curve, the first-order term coefficient B of the fitting curve, and the fitting residual norm D, which is the prediction model for the satisfaction of the in-cruise vehicle sound quality.

[0044] In this embodiment, the prediction model for the satisfaction of the in-cruise vehicle sound quality is:

[0045] F = mA + nB + pD + qG + z (Equation 3)

[0046] Wherein, F is the subjective satisfaction score, G is the A-weighted sound pressure level, A is the quadratic term coefficient of the fitting curve, B is the first-order term coefficient of the fitting curve, D is the fitting residual norm, and m, n, p, q, and z are all constant coefficients obtained by fitting according to Equation 3.

[0047] In this embodiment, a system for constructing a prediction model of in-vehicle sound quality satisfaction of a cruise vehicle includes a memory and a controller. The memory stores a computer-readable program. When the computer-readable program is called by the controller, it can execute the steps of the method for constructing a prediction model of in-vehicle sound quality satisfaction of a cruise vehicle as described in this embodiment.

[0048] As Figure 2 shown, in this embodiment, a method for predicting in-vehicle sound quality satisfaction of a cruise vehicle includes the following steps:

[0049] Step 1: When the vehicle is in a cruise driving state, collect the in-vehicle sound data, process the collected sound data to obtain the A-weighted sound pressure level G and the unweighted one-third octave spectrum P(N), and perform a quadratic polynomial fitting on the unweighted one-third octave spectrum P(N) to obtain the quadratic term coefficient A of the fitting curve, the first-order term coefficient B of the fitting curve, and the fitting residual norm D;

[0050] Step 2: Take the A-weighted sound pressure level G, the quadratic term coefficient A of the fitting curve, the first-order term coefficient B of the fitting curve, and the fitting residual norm D as the inputs of the prediction model of in-vehicle sound quality satisfaction of a cruise vehicle (F = mA + nB + pD + qG + z), and the prediction model of in-vehicle sound quality satisfaction of a cruise vehicle outputs the subjective satisfaction score F;

[0051] Wherein, the prediction model of in-vehicle sound quality satisfaction of a cruise vehicle is constructed by using the method for constructing a prediction model of in-vehicle sound quality satisfaction of a cruise vehicle as described in this embodiment.

[0052] As Figure 3 shown, in this embodiment, a prediction system for in-vehicle sound quality satisfaction of a cruise vehicle includes:

[0053] A data acquisition and processing module, which is used to collect in-vehicle sound data when the vehicle is in a cruise driving state, process the collected sound data to obtain the A-weighted sound pressure level G and the unweighted one-third octave spectrum P(N), and perform a quadratic polynomial fitting on the unweighted one-third octave spectrum P(N) to obtain the quadratic term coefficient A of the fitting curve, the first-order term coefficient B of the fitting curve, and the fitting residual norm D;

[0054] Subjective satisfaction score prediction module, which stores a prediction model for the in-vehicle sound quality satisfaction of the cruise vehicle, is used to receive the A-weighted sound pressure level G, the quadratic term coefficient A of the fitting curve, the first-order term coefficient B of the fitting curve, and the fitting residual norm D output by the data acquisition and processing module, and uses the A-weighted sound pressure level G, the quadratic term coefficient A of the fitting curve, the first-order term coefficient B of the fitting curve, and the fitting residual norm D as the input of the prediction model for the in-vehicle sound quality satisfaction of the cruise vehicle. The prediction model for the in-vehicle sound quality satisfaction of the cruise vehicle outputs the subjective satisfaction score F; this subjective satisfaction score prediction module is connected to the data acquisition and processing module.

[0055] Among them, the prediction model for the in-vehicle sound quality satisfaction of the cruise vehicle is constructed by using the construction method of the prediction model for the in-vehicle sound quality satisfaction of the cruise vehicle as described in this embodiment.

[0056] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications made without departing from the spirit and principle of the present invention shall be equivalent replacement methods and are all included in the protection scope of the present invention.

Claims

1. A method for constructing a prediction model of the satisfaction degree of the in-vehicle sound quality of a cruise vehicle, characterized in that, Including the following steps: Step 1. When the vehicle is in the cruise state, collect the sound data inside the vehicle, and score the cruise noise according to the grading scoring method to obtain the subjective satisfaction score F of the vehicle. Step 2. Process the collected sound data to obtain the A-weighted sound pressure level G and the unweighted one-third octave spectrum P(N). Perform a quadratic polynomial fitting on the unweighted one-third octave spectrum P(N) to obtain the quadratic term coefficient A of the fitting curve, the first-order term coefficient B of the fitting curve, and the fitting residual norm D, and use the residual norm D as the quantization index of the in-vehicle sound quality during cruise. Step 3. Repeat the test on the data of multiple vehicles according to the methods in Step 1 and Step 2 to obtain the subjective satisfaction scores F and the quantization indexes of the in-vehicle sound quality during cruise of multiple sample vehicles. Step 4. For the data of the multiple sample vehicles obtained in Step 3, establish an expression of the subjective satisfaction score F with respect to the A-weighted sound pressure level G, the quadratic term coefficient A of the fitting curve, the first-order term coefficient B of the fitting curve, and the fitting residual norm D, which is the prediction model of the in-vehicle sound quality satisfaction during cruise. The prediction model of the in-vehicle sound quality satisfaction during cruise is: F = mA + nB + pD + qG + z Where, F is the subjective satisfaction score, G is the A-weighted sound pressure level, A is the quadratic term coefficient of the fitting curve, B is the first-order term coefficient of the fitting curve, D is the fitting residual norm, and m, n, p, q, and z are all constant coefficients.

2. The method for constructing a prediction model of the satisfaction degree of the in-vehicle sound quality of a cruise vehicle according to claim 1, characterized in that: In Step 1, a microphone sensor is arranged on the right side of the headrest of the occupant seat to obtain the sound data at the position of the right ear when a person is sitting.

3. The method for constructing a prediction model of the satisfaction degree of the in-vehicle sound quality of a cruise vehicle according to claim 1, characterized in that: In Step 1, the test road surface is selected as an asphalt road surface without water accumulation, the road surface is completely and evenly paved, the wind speed is less than the preset wind speed, the test vehicle keeps driving straight at the preset vehicle speed, and the sound data for the preset time is tested; multiple groups of data are repeatedly tested, and each group of tests ensures the same direction and the same starting point.

4. The method for constructing a prediction model of the satisfaction degree of the in-vehicle sound quality of a cruise vehicle according to claim 1, characterized in that: In Step 2, the consistency screening is performed on the tested sound data, and the screened data is then post-processed to obtain the A-weighted sound pressure level G and the unweighted one-third octave spectrum P(N).

5. A system for constructing a prediction model of the satisfaction degree of the in-vehicle sound quality of a cruise vehicle, characterized in that: It includes a memory and a controller. The memory stores a computer-readable program. When the computer-readable program is called by the controller, it can execute the steps of the method for constructing the prediction model of the in-vehicle sound quality satisfaction during cruise as described in any one of Claims 1 to 4.

6. A method for predicting the satisfaction degree of the in-vehicle sound quality of a cruise vehicle, characterized in that, Including the following steps: Step 1. When the vehicle is in the cruise state, collect the sound data inside the vehicle, process the collected sound data to obtain the A-weighted sound pressure level G and the unweighted one-third octave spectrum P(N), and perform a quadratic polynomial fitting on the unweighted one-third octave spectrum P(N) to obtain the quadratic term coefficient A of the fitting curve, the first-order term coefficient B of the fitting curve, and the fitting residual norm D. Step 2. Use the A-weighted sound pressure level G, the quadratic term coefficient A of the fitting curve, the first-order term coefficient B of the fitting curve, and the fitting residual norm D as the inputs of the prediction model of the in-vehicle sound quality satisfaction during cruise, and the prediction model of the in-vehicle sound quality satisfaction during cruise outputs the subjective satisfaction score F. Among them, the in-vehicle sound quality satisfaction prediction model of the cruising vehicle is constructed by using the construction method of the in-vehicle sound quality satisfaction prediction model described in any one of claims 1 to 4.

7. A system for predicting the satisfaction degree of the in-vehicle sound quality of a cruise vehicle, characterized in that, It includes: A data acquisition and processing module, which is used to collect in-vehicle sound data when the vehicle is in a cruising state, process the collected sound data to obtain the A-weighted sound pressure level G and the unweighted one-third octave spectrum P(N), perform a quadratic polynomial fitting on the unweighted one-third octave spectrum P(N) to obtain the quadratic term coefficient A of the fitting curve, the first-order term coefficient B of the fitting curve, and the fitting residual norm D; A subjective satisfaction score prediction module, which stores the in-vehicle sound quality satisfaction prediction model of the cruising vehicle, is used to receive the A-weighted sound pressure level G, the quadratic term coefficient A of the fitting curve, the first-order term coefficient B of the fitting curve, and the fitting residual norm D output by the data acquisition and processing module, uses the A-weighted sound pressure level G, the quadratic term coefficient A of the fitting curve, the first-order term coefficient B of the fitting curve, and the fitting residual norm D as the input of the in-vehicle sound quality satisfaction prediction model of the cruising vehicle, and the in-vehicle sound quality satisfaction prediction model outputs the subjective satisfaction score F; this subjective satisfaction score prediction module is connected to the data acquisition and processing module; Among them, the in-vehicle sound quality satisfaction prediction model of the cruising vehicle is constructed by using the construction method of the in-vehicle sound quality satisfaction prediction model described in any one of claims 1 to 4.

Images