Targeted reconstruction method for sound field in vehicle

Abstract

The invention relates to an in-vehicle sound field targeted reconstruction method, which comprises the following steps of: dividing a sound field reconstruction target area for the internal environment of a vehicle, establishing an acoustic transfer function from a loudspeaker to each control point of the reconstruction target area, and obtaining a loudspeaker driving signal solution set by solving a sound field reconstruction equation; considering acoustic reflection of a seat back cushion and an acoustic scattering effect of a passenger head, and establishing a numerical equivalent model for calculating acoustic reflection and scattering in a sound field reconstruction target area; a sound pressure and phase compensation mechanism facing an in-vehicle high-matching reconstruction sound field is constructed by taking a binaural region of a controlled object as a targeted regulation and control position; and in combination with sound pressure and a phase compensation mechanism, carrying out secondary iterative solution through a sound field reconstruction equation by taking the incident direction of a given sound field as an undetermined solution parameter to obtain an optimal driving signal solution set so as to complete sound field targeted reconstruction for the in-vehicle environment. Compared with the prior art, acoustic reflection and scattering effect suppression and high-matching sound field reconstruction can be cooperatively realized.

Description

Technical Field

[0001] The present invention relates to the technical field of sound field reconstruction, and in particular to a method for targeted reconstruction of an in-vehicle sound field. Background Art

[0002] In recent years, car users' requirements for the acoustic environment of the driving space are no longer limited to quietness. Intelligent acoustic control that meets the personalized and diversified needs of drivers and passengers has gradually become the dominant direction of industry technology upgrades.

[0003] Reconstructing the in-car sound field allows the controlled driver and passengers to enjoy an independent acoustic environment without wearing headphones, ensuring personal acoustic privacy while alleviating physical ear fatigue. However, due to the complex acoustic cavity structure of the vehicle and the geometric boundaries of the controlled object, acoustic effects such as scattering and reflection are difficult to avoid, which can easily cause the reconstructed sound field for the in-car environment to deviate from the desired sound field based on the free field. Therefore, suppressing the acoustic reflection and scattering effects associated with in-car sound field reconstruction plays an important role in achieving high-quality in-car sound field reconstruction and its engineering implementation.

[0004] In existing research, the journal papers "Comprehensive control method for regulating the reconstruction performance of the bright area sound field and the sound energy in the dark area [J]. Acta Acoustics, 2018." and "Targeted reconstruction method of the in-vehicle sound field based on stable double conjugate gradient [J]. Journal of Tongji University (Natural Science Edition), 2024." both propose different sound field reconstruction methods for the vehicle's interior environment, aiming to meet the ever-increasing demand of drivers and passengers for an independent and diverse acoustic environment.

[0005] In addition, invention patent CN 115052225 A proposes an active control method for the sub-regional sound field in a vehicle, which can ensure the performance balance of acoustic energy contrast and sound field reconstruction error between regions in the vehicle, under the premise that the individual speaker driving signal is within the linear operating range.

[0006] However, the above solution still has the following disadvantages:

[0007] 1. While existing sound field reconstruction methods can control differences in sound pressure levels between areas, most fail to consider the impact of driver and passenger interference on the reconstructed sound field. In particular, they lack mechanisms to suppress or compensate for acoustic reflection and scattering effects.

[0008] 2. Acoustic effects such as scattering and reflection caused by the complex acoustic cavity structure of the vehicle and the geometric boundaries of the controlled object have not been fully modeled and calculated, resulting in the sound field solution model based on the free field assumption being prone to systematic deviations when facing the vehicle environment. Summary of the Invention

[0009] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide a method for targeted reconstruction of the in-vehicle sound field, which can synergistically achieve the suppression of acoustic reflection and scattering effects and high-matching reconstruction of the sound field.

[0010] The purpose of the present invention can be achieved by the following technical solution: A method for targeted reconstruction of the in-vehicle sound field, comprising the following steps:

[0011] S1. Divide the target area for sound field reconstruction based on the vehicle interior environment, construct an acoustic transfer function from the speaker to each control point in the reconstruction target area, and obtain the speaker driving signal solution set by solving the sound field reconstruction equation;

[0012] S2. Considering the acoustic reflection of the seat headrest and the acoustic scattering effect of the occupant's head, a numerical equivalent model is established to calculate the acoustic reflection and scattering within the target area of ​​the sound field reconstruction.

[0013] S3. Using the binaural area of ​​the controlled object as the target control position, a sound pressure and phase compensation mechanism is established for highly matched reconstruction of the sound field inside the vehicle.

[0014] S4. Combining the sound pressure and phase compensation mechanism, taking the incident direction of the given sound field as the undetermined solution parameter, the sound field reconstruction equation is solved through secondary iteration to obtain the optimal driving signal solution set to complete the targeted reconstruction of the sound field for the vehicle interior environment.

[0015] Furthermore, the step S1 includes the following steps:

[0016] S11. Divide the interior space of the vehicle into a number of sound field reconstruction target areas, and construct an acoustic transfer function matrix for sound field reconstruction based on the spatial relationship between the positions of the vehicle speakers and the control points of the reconstruction target areas;

[0017] S12, obtaining the linear travel range, electroacoustic transfer function, and sound pressure gain coefficient of the speaker unit, thereby determining a threshold range of the speaker driving signal;

[0018] S13. According to the acoustic requirements of each sound field reconstruction target area, the expected sound pressure level is set, and the sound field reconstruction equation is established with minimizing the regional reconstruction error as the optimization goal. The speaker driving signal solution set that meets the threshold range constraint is obtained based on the convex optimization solution method.

[0019] Furthermore, the target area for sound field reconstruction in step S11 includes a bright area and a dark area. The bright area corresponds to the audio receiving area, and the dark area corresponds to the remaining area outside the bright area. The acoustic transfer function matrix of the bright area is:

[0020]

[0021] The acoustic transfer function matrix of the dark area is:

[0022]

[0023] Among them, G b is the acoustic transfer function matrix from l loudspeakers to n control points in the bright area of ​​the reconstruction area, G d is the acoustic transfer function matrix from l loudspeakers to m control points in the dark area of ​​the reconstruction area.

[0024] Furthermore, the optimization goal in step S13 is specifically:

[0025] u(f)=min{abs{P b -E b}+abs{P d -E d}}

[0026] Among them, P b and E b are the reconstructed sound field and the expected sound field in the bright area, P d and E d They are the reconstructed sound field and the expected sound field in the dark area respectively.

[0027] Furthermore, step S2 includes the following steps:

[0028] S21. Determine the acoustic reflection boundary of the seat headrest and the acoustic scattering boundary of the occupant's head based on the spatial correspondence between the vehicle's internal structure and the target area for sound field reconstruction.

[0029] S22. Based on the scattering mechanism of the occupant's head on sound wave transmission, a rigid human head model is used to perform equivalent calculations of sound wave scattering, and a numerical calculation model of sound wave scattering for the occupant's head is constructed;

[0030] S23. Based on the reflection mechanism of the seat headrest on sound waves, the sound absorption coefficient at the headrest position is measured using an impedance tube, and a numerical calculation model for sound wave reflection facing the seat headrest is established.

[0031] S24. Calculate the acoustic reflection effect of the seat headrest and the acoustic scattering effect of the occupant's head based on the loudspeaker drive signal solution obtained in step S1 and in combination with a numerical calculation model for sound wave reflection and a numerical calculation model for sound wave scattering.

[0032] Furthermore, the specific process of step S22 is as follows:

[0033] The human head is approximated as a rigid sphere, the incident plane wave of the sound pressure amplitude p0 is converted into the composition of spherical waves of various orders, and considering the boundary condition that the vertical vibration velocity of the sphere is zero, the complex amplitude of the scattered sound waves of various orders is expressed in polar coordinates as:

[0034]

[0035] Among them, j m and are the spherical Hankel functions of the first and second kinds;

[0036] Under the far-field condition kr>>1, the equivalent simplification of the second-kind spherical Hankel function yields:

[0037]

[0038]

[0039] Where r and k are the radius and wave number of the rigid sphere, P m represents the Legendre polynomials.

[0040] Furthermore, step S3 includes the following steps:

[0041] S31, based on the three-dimensional spatial coordinate analysis of the target sound field reconstruction area, selecting a sound field control target reference position facing the binaural area of ​​the controlled object;

[0042] S32. By analyzing the frequency domain characteristics of the seat headrest reflection path and the head scattering path, we can clarify the impact of the combined acoustic effect of the two on the sound field frequency reconstruction performance.

[0043] S33. Using the sound pressure difference between the desired sound field and the reconstructed sound field that takes into account acoustic reflection and scattering effects as input, a framework for suppressing acoustic reflection and scattering effects based on the sound pressure distribution characteristics at the driver's and passenger's ears is constructed.

[0044] S34. Based on the constructed acoustic reflection and scattering effect suppression framework, the amplitude compensation coefficient and phase compensation coefficient used to correct the sound field reconstruction performance are calculated respectively.

[0045] Furthermore, the amplitude compensation coefficient is specifically:

[0046] c amp =abs{ave{∑p j}} / abs{p0}

[0047] The phase compensation coefficient is specifically:

[0048] c pha =ang{ave{∑p j}}-ang{p0}

[0049] Among them, p j is the career amplitude of the control point position on the driver's ear side.

[0050] Furthermore, step S4 includes the following steps:

[0051] S41, introducing an amplitude compensation coefficient and a phase compensation coefficient to update the sound field reconstruction equation for the target area;

[0052] S42, setting the incident direction of the given sound field as a parameter to be determined, and preprocessing the coefficient matrix of the sound field reconstruction equation;

[0053] S43, using the loudspeaker driving signal solution set without considering the acoustic reflection and scattering effects as the initial iteration point, iteratively solving the sound field reconstruction equation based on a convex optimization solution method to obtain the optimal loudspeaker driving signal solution set;

[0054] S44. Load the optimal loudspeaker driving signal solution to the corresponding loudspeaker to obtain a highly matched reconstructed sound field for the in-car environment.

[0055] Furthermore, the step S41 specifically introduces an amplitude compensation coefficient and a phase compensation coefficient to reset the desired sound field in the target area, thereby updating the sound field reconstruction equation;

[0056] The step S42 specifically performs Cholesky preprocessing on the coefficient matrix of the sound field reconstruction equation to reduce the matrix condition number and avoid the coefficient matrix ill-conditioning problem.

[0057] Compared with the prior art, the present invention has the following advantages:

[0058] The present invention first divides the sound field reconstruction target area according to the vehicle interior environment, constructs the acoustic transfer function from the loudspeaker to each control point in the reconstruction target area, solves the sound field reconstruction equation, and obtains the loudspeaker driving signal solution set; then, considering the acoustic reflection of the seat headrest and the acoustic scattering of the occupant's head, the acoustic reflection and scattering effects in the sound field reconstruction target area are calculated; then, the binaural area of ​​the controlled object is used as the target control position, and a sound pressure and phase compensation mechanism for a high-matching reconstructed sound field is constructed; finally, combined with the sound pressure and phase compensation mechanism, the incident direction of the established sound field is used as the undetermined solution parameter, and the sound field reconstruction equation is iteratively solved to obtain the optimal loudspeaker driving signal solution set, which is used to control the working state of the loudspeaker, and a high-matching reconstructed sound field for the vehicle interior environment can be obtained, thereby completing the coordinated realization of suppressing the acoustic reflection and scattering effects and the high-matching reconstructed sound field.

[0059] The present invention divides the interior space of a vehicle into several sound field reconstruction areas. Based on the spatial relationship between the positions of the on-board speakers and the control points of the reconstruction areas, an acoustic transfer function matrix for sound field reconstruction is constructed. The linear travel range, electroacoustic transfer function, and sound pressure gain coefficient of the speaker unit are then obtained through actual measurement. Based on this, the threshold range of the driving signal is determined to ensure that the subsequently solved speaker driving signal can be effectively loaded into the speaker.

[0060] The present invention combines the spatial correspondence between the vehicle's internal structure and the sound field reconstruction area to determine the acoustic reflection boundary of the seat headrest and the acoustic scattering boundary of the occupant's head. Based on the scattering mechanism of the occupant's head on sound wave transmission, a rigid human head model is used to perform equivalent calculations of sound wave scattering, constructing a numerical calculation model of sound wave scattering facing the occupant's head. Furthermore, based on the reflection mechanism of the seat headrest on sound wave transmission, the sound absorption coefficient at the headrest position is measured using an impedance tube, and a numerical calculation model of sound wave reflection facing the seat headrest is established. This fully accounts for the negative impact of the seat headrest and occupant's head on the vehicle's interior sound field reconstruction, accurately determining the acoustic reflection and scattering effects of the target reconstruction area.

[0061] The present invention selects a target reference position for sound field control in the binaural area of ​​the controlled object based on the three-dimensional spatial coordinate analysis of the target sound field reconstruction area; then, through the frequency domain characteristic analysis of the seat headrest reflection path and the head scattering path, clarifies the influence trend of the composite acoustic effect constituted by the two on the sound field frequency reconstruction performance; and then, using the sound pressure difference between the desired sound field and the reconstructed sound field considering the acoustic reflection and scattering effects as input, constructs an acoustic reflection and scattering effect suppression framework based on the sound pressure distribution characteristics on the ear side of the driver and passenger, which can obtain the amplitude compensation coefficient and phase compensation coefficient that effectively suppress the acoustic reflection and scattering effects, thereby effectively suppressing the acoustic reflection of the seat headrest and the acoustic scattering of the passenger's head.

[0062] The present invention takes into account that the coefficient matrices in the sound field reconstruction equation are mostly large-scale complex linear matrices, so the matrix condition number can be reduced through coefficient matrix preprocessing to improve the coefficient matrix morbidity problem; and considering that the driver and passengers are relatively slow to judge the directionality of sound due to the cramped and narrow vehicle sound cavity structure, the design uses the incident direction of the established sound field as the undetermined solution parameter to load it into the sound field reconstruction equation for iterative solution; in addition, the driving signal when the acoustic reflection and scattering effects are not considered is used as the initial iteration point to ensure the feasibility of the solution, thereby ensuring that the optimal speaker driving signal solution set is obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 Schematic diagram of the method flow of the present invention;

[0064] Figure 2 Schematic diagram of the application process of the embodiment;

[0065] Figure 3 The results of the speaker linear actuation stroke test in the embodiment;

[0066] Figure 4a : The test results of the sound pressure amplitude difference of the linear electroacoustic transfer function of the loudspeaker in the embodiment;

[0067] Figure 4b The test results of the sound pressure phase difference of the linear electroacoustic transfer function of the loudspeaker in the embodiment;

[0068] Figure 5 The test results of the loudspeaker sound pressure gain coefficient in the embodiment;

[0069] Figure 6 The test results of the sound absorption coefficient at the vehicle seat headrest position in the embodiment;

[0070] Figure 7 This is the result of reconstructing the sound field before the acoustic reflection and scattering effects are suppressed in the embodiment;

[0071] Figure 8 This is the result of reconstructing the sound field after the acoustic reflection and scattering effects are suppressed in the embodiment;

[0072] Figure 9 This is the verification result of reconstructing the sound field after the acoustic reflection and scattering effects in the embodiment are suppressed. DETAILED DESCRIPTION

[0073] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0074] Example

[0075] like Figure 1 As shown, a method for targeted reconstruction of the in-vehicle sound field includes the following steps:

[0076] S1. Divide the target area for sound field reconstruction into the vehicle interior environment, construct an acoustic transfer function from the speaker to each control point in the reconstruction target area, solve the sound field reconstruction equation, and obtain the speaker driving signal solution set;

[0077] S2. Considering the acoustic reflection from the seat headrest and the acoustic scattering from the occupant's head, a numerical equivalent model is established to calculate the acoustic reflection and scattering effects within the target area for sound field reconstruction.

[0078] S3. Using the binaural area of ​​the controlled object as the target control position, a sound pressure and phase compensation mechanism is constructed for high-matching reconstruction of the sound field;

[0079] S4. Combining the sound pressure and phase compensation mechanisms, and taking the incident direction of the given sound field as the undetermined solution parameter, the sound field reconstruction equation is iteratively solved to obtain the optimal loudspeaker drive signal solution set, which is used to control the working state of the loudspeaker and obtain a highly matched reconstructed sound field for the in-vehicle environment.

[0080] The specific process of step S1 is as follows:

[0081] Several sound field reconstruction areas are divided inside the vehicle. The reconstruction area is usually divided into longitudinal and transverse planes enveloping the human head. The audio receiving area is defined as the acoustic bright area, and the remaining area is defined as the acoustic dark area.

[0082] The sound source array used for sound field reconstruction usually consists of several loudspeakers. Without considering the directivity, it can be equivalent to a monopole sound source. Its radiated sound pressure in a free field environment can be expressed as:

[0083]

[0084] Where f is the spatial frequency, p0 is the sound pressure amplitude, c is the sound speed, and x s is the sound source position.

[0085] The sound pressure distribution in the reconstructed target area can be expressed as the mathematical product of the sound source driving signal and the acoustic transfer function. The sound pressure distribution can be expressed as:

[0086] p r (x,f) m×1 =G(x,f) m×n u(x,f) n×1

[0087] where G is the acoustic transfer function that depends on the target area and the location of the sound source array.

[0088] The acoustic transfer functions of the bright area and dark area can be defined and expressed in matrix form as follows:

[0089]

[0090] Among them, G b is the acoustic transfer function matrix from l loudspeakers to n control points in the bright area of ​​the reconstruction area, G d is the acoustic transfer function matrix from l loudspeakers to m control points in the dark area of ​​the reconstruction area.

[0091] Acoustic energy contrast is a basic indicator for measuring the effectiveness of sound field targeted reconstruction. It is defined as the ratio of the average acoustic energy density between the bright area and the dark area. A higher acoustic energy contrast indicates a more significant difference in sound pressure levels between different target areas. It is expressed as follows:

[0092]

[0093] For linear travel and electroacoustic transfer function, a sine wave signal generator with a constant frequency is applied to the speaker, and a monitoring microphone is used to collect sound pressure signals at a constant position. The voltage amplitude of the signal generator is increased at equal intervals to observe the change in sound pressure at a fixed position.

[0094] For the sound pressure gain coefficient, a sine wave signal generator excitation signal of a certain constant frequency is loaded to the speaker from the computer and the signal generator respectively, and the sound pressure gain coefficient under the same amplitude loading is calculated;

[0095] Define a plane acoustic wave with a constant amplitude as the reconstruction target of the target area, which can be expressed as:

[0096]

[0097] With the goal of minimizing the sound field reconstruction error in the reconstruction area, the speaker driving signal that satisfies the threshold constraint is solved based on the convex optimization solution method, and the reconstructed sound field that meets the established requirements can be obtained. It can be expressed through the mathematical relationship as follows:

[0098] u(f)=min{abs{P b -E b}+abs{P d -E d}}

[0099] Where, P b and E b are the reconstructed sound field and the expected sound field in the bright area, P d and E d They are the reconstructed sound field and the expected sound field in the dark area respectively.

[0100] The specific process of step S2 is:

[0101] Combining the spatial correspondence between the vehicle's internal structure and the sound field reconstruction area, the acoustic reflection boundary of the seat head and the acoustic scattering boundary of the occupant's head are determined respectively.

[0102] The scattering effect is related to the size of the obstacle and the length of the sound wave. Specifically, when the obstacle is close to or larger than the wavelength, the scattering effect is significant. Given the complexity of calculating the effect of sound scattering on obstacles of varying shapes, the human head is typically approximated as a rigid sphere when analyzing acoustic scattering.

[0103] Convert the incident plane wave with a sound pressure amplitude of p0 into the composition of spherical waves of various orders, and consider the boundary condition that the vertical vibration velocity of the sphere is zero. Then, the complex amplitude of the scattered sound waves of various orders can be expressed in polar coordinates as follows:

[0104]

[0105] Among them, r0 and k are the radius and wave number of the rigid sphere, j m and is the first kind of spherical Hankel function and the second kind of spherical Hankel function, P m represents the Legendre polynomials.

[0106] Under the far-field condition kr>>1, the second-kind spherical Hankel function can be equivalently simplified, and the above formula is equivalent to:

[0107]

[0108] Based on the reflection effect of the seat headrest structure on sound wave transmission, a standardized impedance tube test is used to measure the sound absorption coefficient of the headrest contact area;

[0109] Based on the physical characteristics of acoustic wave reflection in the seat headrest structure, a numerical simulation model of its acoustic reflection process was constructed. The back-propagating plane wave theory was used to model the reflected sound field as equivalent to the back-propagating plane wave fluctuation form.

[0110] A coupled numerical model of the seat headrest reflection acoustic characteristics and the occupant head scattering acoustic characteristics is constructed to calculate the impact of the seat headrest acoustic reflection effect and the occupant head acoustic scattering effect on the sound field reconstruction performance.

[0111] The specific process of step S3 is:

[0112] Based on the three-dimensional spatial coordinate analysis of the target sound field reconstruction area, the target reference position of the sound field control facing the binaural area of ​​the controlled object is selected;

[0113] By analyzing the frequency domain characteristics of the seat headrest reflection path and the head scattering path, we can clarify the impact of the composite acoustic effect of the two on the sound field frequency reconstruction performance.

[0114] Taking the sound pressure difference between the desired sound field and the reconstructed sound field that takes into account acoustic reflection and scattering effects as input, a framework for suppressing acoustic reflection and scattering effects based on the sound pressure distribution characteristics at the driver's and passenger's ears is constructed.

[0115] Among them, the reconstructed sound field amplitude compensation coefficient used to suppress acoustic reflection and scattering effects is:

[0116] c amp =abs{ave{∑p j}} / abs{p0}

[0117] The phase compensation coefficient of the reconstructed sound field used to suppress acoustic reflection and scattering effects is:

[0118] c pha =ang{ave{∑p j}}-ang{p0}

[0119] The specific process of step S4 is:

[0120] Amplitude compensation coefficient and phase compensation coefficient are introduced to reset the expected sound field of the target area, and the sound field reconstruction equation for the target area is updated;

[0121] Considering that the coefficient matrices in the sound field reconstruction equations are mostly large-scale complex linear matrices, coefficient matrix preprocessing is introduced to reduce the matrix condition number and improve the ill-conditioned problem of the coefficient matrix;

[0122] Due to the narrow and confined vehicle acoustic cavity structure, the driver and passengers are slow to judge the directionality of sound. Therefore, the incident direction of the given sound field is used as the undetermined parameter and loaded into the sound field reconstruction equation for an iterative solution.

[0123] The acoustic field equation is solved twice based on the convex optimization method, and the driving signal without considering the acoustic reflection and scattering effects is used as the initial iteration point to ensure the feasibility of the solution.

[0124] The optimal driving signal solution is loaded onto the corresponding speaker to calculate the suppression effect of the acoustic reflection of the seat headrest and the acoustic scattering of the occupant's head.

[0125] This embodiment applies the above solution, such as Figure 2 As shown, the main contents include:

[0126] Step 1: Divide the target area for sound field reconstruction based on the vehicle interior environment and construct the acoustic transfer function from the speaker to the control point in the target area;

[0127] The vehicle driving position is defined as the bright sound field and the passenger seat is defined as the dark sound field. This setting simulates the situation where the driver expects to receive navigation information while the passenger in the passenger seat is not disturbed during vehicle driving.

[0128] The acoustic transfer function of the bright zone is defined and expressed in matrix form as follows:

[0129]

[0130] Among them, G b is the acoustic transfer function matrix from l loudspeakers to n control points in the bright area of ​​the reconstruction area.

[0131] The acoustic transfer function of the dark zone is defined and expressed in matrix form as follows:

[0132]

[0133] Among them, G d is the acoustic transfer function matrix from l loudspeakers to m control points in the dark area of ​​the reconstruction area.

[0134] In this embodiment, the expected sound pressure level in the bright area is set to 73.98 dB and the reconstruction frequency range is set to 0.2-3 kHz. Such settings can cover the frequency band where human voice navigation is located.

[0135] Step 2: Measure the speaker's linear travel range, electroacoustic transfer function, and sound pressure gain coefficient to determine the driving signal constraint threshold based on the algorithm solution;

[0136] A certain type of monitoring speaker is used to simulate a point sound source. Its frequency response performance in the free field is 75-20kHz (±2.5dB). Set a 1kHz standard sine wave as the signal generator excitation signal, and gradually increase the voltage at 0.5V intervals to observe the sound pressure changes at a fixed position. The results are as follows: Figure 3 shown.

[0137] In the voltage range of 0.5-7V, the speaker maintains a linearly changing actuation stroke (the excitation voltage and sound pressure show a linearly changing actuation stroke), and the sound pressure amplitude difference and phase difference at different sound pressure positions show good numerical stability, as shown in Figure 2. Figure 4a and Figure 4b shown.

[0138] The driving signal constraint threshold for sound field reconstruction is determined by the sound pressure gain coefficient of the loudspeaker's driving signal and the voltage signal, which is given by Figure 5 It can be seen that the sound pressure gain coefficient is approximately 6.8 1 / V, and the corresponding driving signal amplitude constraint is ±0.74.

[0139] Step 3: Obtain the loudspeaker driving signal solution set by solving the sound field reconstruction equation;

[0140] Taking the minimum sound field reconstruction error at the driver and co-driver positions as the optimization goal, the reconstructed sound field that meets the established requirements can be obtained. Its mathematical expression is:

[0141] u(f)=min{abs{P b -E b}+abs{P d -E d}}

[0142] Step 4: Use an impedance tube to measure the full-band sound absorption coefficient of the vehicle seat headrest;

[0143] For the sound absorption coefficient of vehicle seats, sponge was used as the filling material, and two types of polyvinyl chloride artificial leather with the same material but different patterns on the vehicle seats were selected as test objects.

[0144] The test simulates two different contact states between leather and sponge: unconstrained fit and constrained tension. The sound absorption coefficient measurement results are as follows: Figure 6 shown.

[0145] Step 5: Analyze the impact of acoustic reflection and scattering effects on the reconstructed sound field;

[0146] The acoustic reflection and scattering effects have an irregular and nonlinear effect on the reconstructed sound field, which may lower the sound pressure level of the bright area sound field and also have a superimposed effect on the bright area sound field, such as Figure 7 shown.

[0147] At 0.5kHz, the reconstructed sound field is shifted slightly compared to the free-field sound field, but the shift is not significant because the wavelength of the low-frequency reflected sound waves is longer and the scattering effect is mainly concentrated in front of the occupant's face.

[0148] As the frequency increases, the superposition of reflection and scattering will complicate the negative acoustic effects. Not only is there an obvious scattering effect at the ear side, but the reflected waves from the seat also make the reconstructed sound field more chaotic.

[0149] Step 6: Construct a targeted compensation framework based on the sound pressure amplitude and phase at the driver's and passenger's ears;

[0150] This solution only considers the direction of sound wave incidence perpendicular to the driver's face. Since the scattered and reflected sound fields formed by other sound wave incidence directions are not within the target area of ​​sound field reconstruction or have a slight impact, sound pressure / phase compensation coefficients are constructed to make the sound field on the ear side of the controlled object close to the established sound pressure.

[0151] Step 7: Introduce the amplitude compensation coefficient and the phase compensation coefficient to reset the expected sound field of the target area, and update the sound field reconstruction equation for the target area;

[0152] Step 8: Considering that the coefficient matrix in the sound field reconstruction equation is mostly a large-scale complex linear matrix, the matrix condition number can be reduced to improve the ill-conditioned problem of the coefficient matrix through coefficient matrix preprocessing;

[0153] Step 9: Load the optimal driving signal solution set to the corresponding speaker and calculate the suppression effect of the acoustic reflection of the seat headrest and the acoustic scattering of the occupant's head, such as Figure 8 shown.

[0154] Step 10: Build a sound field reconstruction test platform to verify the effect of suppressing acoustic reflection and scattering effects.

[0155] In this embodiment, the sound field reconstruction verification platform consists of two parts: a speaker drive signal driving unit and a microphone sound pressure receiving unit. The speaker drive signal driving unit consists of a computer, a multi-channel audio processor, an AD / DA expansion port and an active speaker. Figure 9As shown in the figure, thanks to the compensation mechanism for the negative acoustic effects, the sound pressure level in the bright area can be close to the set target, especially at the extreme positions where the reconstructed sound field deviates.

[0156] In summary, this solution addresses the acoustic reflections from seat backrests and acoustic scattering from occupant heads that accompany the reconstruction of the vehicle's interior sound field. By establishing a numerical equivalent model to evaluate the acoustic reflection and scattering effects within the sound field reconstruction region, this solution proposes a targeted in-vehicle sound field reconstruction method based on a sound pressure and phase compensation mechanism designed to achieve a highly matched reconstructed sound field, ensuring that the reconstructed sound field is close to the intended sound field. The main technical issues addressed include: 1) setting the desired sound pressure level based on the acoustic requirements of each sound field reconstruction region, solving the sound field reconstruction with the optimization goal of minimizing the regional reconstruction error, establishing a numerical equivalent model to evaluate the acoustic reflection and scattering effects within the sound field reconstruction region, and constructing a sound pressure and phase compensation mechanism for a highly matched reconstructed sound field; 2) analyzing the frequency domain characteristics of the seat backrest reflection path and the head scattering path, a framework for suppressing acoustic reflection and scattering effects based on the sound pressure distribution characteristics at the driver's and occupant's ear is constructed. By introducing amplitude and phase compensation coefficients, the sound field equation is solved quadratically using a convex optimization solution method. Applying this solution in practice can achieve the coordinated realization of suppressing acoustic reflection and scattering effects and highly matching the reconstructed sound field. It can fully consider and improve the negative impact of seat headrests and occupant heads on the reconstruction of the in-vehicle sound field, laying an engineering foundation for the highly matching installation of sound field reconstruction in actual vehicle cabins.

Claims

1. A method for targeted reconstruction of the in-vehicle sound field, characterized in that: The following steps are involved: S1. Divide the target area for sound field reconstruction based on the vehicle interior environment, construct an acoustic transfer function from the speaker to each control point in the reconstruction target area, and obtain the speaker driving signal solution set by solving the sound field reconstruction equation; S2. Considering the acoustic reflection of the seat headrest and the acoustic scattering effect of the occupant's head, a numerical equivalent model is established to calculate the acoustic reflection and scattering within the target area of ​​the sound field reconstruction. S3. Using the binaural area of ​​the controlled object as the target control position, a sound pressure and phase compensation mechanism is established for highly matched reconstruction of the sound field inside the vehicle. S4. Combining the sound pressure and phase compensation mechanism, taking the incident direction of the given sound field as the undetermined solution parameter, the sound field reconstruction equation is solved through secondary iteration to obtain the optimal driving signal solution set to complete the targeted reconstruction of the sound field for the vehicle interior environment.

2. The method for targeted reconstruction of the in-vehicle sound field according to claim 1, characterized in that: The step S1 comprises the following steps: S11. Divide the interior space of the vehicle into a number of sound field reconstruction target areas, and construct an acoustic transfer function matrix for sound field reconstruction based on the spatial relationship between the positions of the vehicle speakers and the control points of the reconstruction target areas; S12, obtaining the linear travel range, electroacoustic transfer function, and sound pressure gain coefficient of the speaker unit, thereby determining a threshold range of the speaker driving signal; S13. According to the acoustic requirements of each sound field reconstruction target area, the expected sound pressure level is set, and the sound field reconstruction equation is established with minimizing the regional reconstruction error as the optimization goal. The speaker driving signal solution set that meets the threshold range constraint is obtained based on the convex optimization solution method.

3. The method for targeted reconstruction of the in-vehicle sound field according to claim 2, characterized in that: The target area for sound field reconstruction in step S11 includes a bright area and a dark area. The bright area corresponds to the audio receiving area, and the dark area corresponds to the remaining area outside the bright area. The acoustic transfer function matrix of the bright area is: The acoustic transfer function matrix of the dark area is: Among them, G b is the acoustic transfer function matrix from l loudspeakers to n control points in the bright area of ​​the reconstruction area, G d is the acoustic transfer function matrix from l loudspeakers to m control points in the dark area of ​​the reconstruction area.

4. The method for targeted reconstruction of the in-vehicle sound field according to claim 3, characterized in that: The optimization goal in step S13 is specifically: u(f)=min{abs{P b -E b }+abs{P d -E d }} Among them, P b and E b are the reconstructed sound field and the expected sound field in the bright area, P d and E d They are the reconstructed sound field and the expected sound field in the dark area respectively.

5. The method for targeted reconstruction of the in-vehicle sound field according to claim 1, characterized in that: The step S2 comprises the following steps: S21. Determine the acoustic reflection boundary of the seat headrest and the acoustic scattering boundary of the occupant's head based on the spatial correspondence between the vehicle's internal structure and the target area for sound field reconstruction. S22. Based on the scattering mechanism of the occupant's head on sound wave transmission, a rigid human head model is used to perform equivalent calculations of sound wave scattering, and a numerical calculation model of sound wave scattering for the occupant's head is constructed; S23. Based on the reflection mechanism of the seat headrest on sound waves, the sound absorption coefficient at the headrest position is measured using an impedance tube, and a numerical calculation model for sound wave reflection facing the seat headrest is established. S24. Calculate the acoustic reflection effect of the seat headrest and the acoustic scattering effect of the occupant's head based on the loudspeaker drive signal solution obtained in step S1 and in combination with a numerical calculation model for sound wave reflection and a numerical calculation model for sound wave scattering.

6. The method for targeted reconstruction of the in-vehicle sound field according to claim 5, characterized in that: The specific process of step S22 is as follows: The human head is approximated as a rigid sphere, the incident plane wave of the sound pressure amplitude p0 is converted into the composition of spherical waves of various orders, and considering the boundary condition that the vertical vibration velocity of the sphere is zero, the complex amplitude of the scattered sound waves of various orders is expressed in polar coordinates as: Among them, j m and are the spherical Hankel functions of the first and second kinds; Under the far-field condition kr>>1, the equivalent simplification of the second-kind spherical Hankel function yields: Where r and k are the radius and wave number of the rigid sphere, P m represents the Legendre polynomials.

7. The method for targeted reconstruction of the in-vehicle sound field according to claim 6, characterized in that: The step S3 comprises the following steps: S31, based on the three-dimensional spatial coordinate analysis of the target sound field reconstruction area, selecting a sound field control target reference position facing the binaural area of ​​the controlled object; S32. By analyzing the frequency domain characteristics of the seat headrest reflection path and the head scattering path, we can clarify the impact of the combined acoustic effect of the two on the sound field frequency reconstruction performance. S33. Using the sound pressure difference between the desired sound field and the reconstructed sound field that takes into account acoustic reflection and scattering effects as input, a framework for suppressing acoustic reflection and scattering effects based on the sound pressure distribution characteristics at the driver's and passenger's ears is constructed. S34. Based on the constructed acoustic reflection and scattering effect suppression framework, the amplitude compensation coefficient and phase compensation coefficient used to correct the sound field reconstruction performance are calculated respectively.

8. The method for targeted reconstruction of the in-vehicle sound field according to claim 7, characterized in that: The amplitude compensation coefficient is specifically: c amp =abs{ave{∑p j }} / abs{p0} The phase compensation coefficient is specifically: c pha =the{ave{∑p j }}-the{p0} Among them, p j is the career amplitude of the control point position on the driver's ear side.

9. The method for targeted reconstruction of the in-vehicle sound field according to claim 7, characterized in that: The step S4 comprises the following steps: S41, introducing an amplitude compensation coefficient and a phase compensation coefficient to update the sound field reconstruction equation for the target area; S42, setting the incident direction of the given sound field as a parameter to be determined, and preprocessing the coefficient matrix of the sound field reconstruction equation; S43, using the loudspeaker driving signal solution set without considering the acoustic reflection and scattering effects as the initial iteration point, iteratively solving the sound field reconstruction equation based on a convex optimization solution method to obtain the optimal loudspeaker driving signal solution set; S44. Load the optimal loudspeaker driving signal solution to the corresponding loudspeaker to obtain a highly matched reconstructed sound field for the in-car environment.

10. The method for targeted reconstruction of the in-vehicle sound field according to claim 9, characterized in that: The step S41 specifically introduces an amplitude compensation coefficient and a phase compensation coefficient to reset the desired sound field of the target area, thereby updating the sound field reconstruction equation; The step S42 specifically performs Cholesky preprocessing on the coefficient matrix of the sound field reconstruction equation to reduce the matrix condition number and avoid the coefficient matrix ill-conditioning problem.

Images