A sound barrier form intelligent generation method, system and readable medium for residential behavior noise control

Abstract

This invention discloses a method, system, and readable medium for intelligent generation of sound barrier shapes for residential area behavioral noise control. The method includes: (1) acquiring behavioral noise data and spatial data of several residential areas, identifying typical residential area types through clustering, and constructing corresponding centroid cases accordingly; (2) constructing a sound barrier shape spectrum library based on folded geometry and parametric design; (3) performing acoustic simulation on each sound barrier shape in the sound barrier shape spectrum library constructed in step (2), and selecting sound barrier shapes that meet the set requirements for acoustic evaluation indicators as candidate sound barrier shapes; (4) placing the candidate sound barrier shapes obtained in step (3) into the centroid cases constructed in step (1) for sound field simulation, and selecting the optimal sound barrier shape that is compatible with each residential area type. This invention can realize the quantitative, automated, and precise intelligent generation of sound barrier shapes, and achieve precise adaptation of sound barrier shapes to real sound fields.

Description

Technical Field

[0001] This invention relates to the field of noise control technology, and in particular to a method, system, and readable medium for intelligent generation of sound barrier shapes for residential behavioral noise control. Background Technology

[0002] Noise from activities such as square dancing and open-air karaoke in residential areas is predominantly mid-to-high frequency, with a wide propagation range and weak vertical attenuation, easily causing disturbances to residents throughout the entire building. This is a core challenge in current residential noise management. Sound barriers are a direct and effective method for controlling this type of noise in residential areas, and their shape is a key factor determining the noise reduction effect. However, traditional technologies have the following significant drawbacks:

[0003] (1) Standardized form, poor noise reduction: Existing sound barrier technology is mainly designed for road traffic noise, and most of them are flat plate structures. Their form and size are determined based on the road scene. Their acoustic structure is optimized for low frequency noise. Their scattering, sound absorption and noise reduction control effects on mid-to-high frequency behavioral noise are limited. They cannot adapt to the mid-to-high frequency propagation characteristics of behavioral noise, and are difficult to adapt to the complex spatial layout and landscape integration needs of residential areas. Their scattering and sound absorption capabilities are insufficient, and their noise reduction efficiency is low.

[0004] (2) Detached from real residential scenarios: Existing design methods rely on experience for form design, lack a systematic generation logic based on typical noise exposure scenarios, and do not combine the spatial layout of residential areas, sound source distribution, and sound reception characteristics. The form design of sound barriers is singular and does not match the sound field scenario of residential areas. The scene adaptability is weak, the ability to specifically reduce mid-to-high frequency behavioral noise is weak, and the form design is disconnected from real residential scenarios. It cannot solve the differentiated problems such as strong reverberation in central concentrated sound source residential areas and prominent direct sound in edge point sound source residential areas.

[0005] (3) Lack of systematic intelligent generation process: There is no closed-loop method of "scene recognition - morphological design - acoustic verification - scene adaptation", the morphological design has no quantitative basis, and the replicability and engineering adaptability are poor.

[0006] Therefore, there is an urgent need for an intelligent method for generating sound barrier shapes that is oriented towards behavioral noise, based on typical residential scenarios, and integrates acoustic simulation, so as to achieve accurate, scenario-based, and intelligent generation of sound barrier shapes. Summary of the Invention

[0007] Purpose of the invention: To address the above-mentioned shortcomings, this invention proposes a method, system, and readable medium for intelligent generation of sound barrier shapes for residential behavioral noise control. This invention solves the technical problems of traditional sound barriers having a single shape, weak targeted noise reduction capability for mid-to-high frequency behavioral noise, a disconnect between shape design and real residential scenarios, and a lack of systematic intelligent generation methods. It achieves acoustic driving, scene adaptation, and intelligent generation of sound barrier shapes.

[0008] Technical solution: This invention provides a method for intelligent generation of sound barrier shapes for residential area behavioral noise control, characterized by comprising:

[0009] (1) Obtain behavioral noise data and spatial data of several residential areas, identify typical residential area types through clustering, and construct corresponding centroid cases for each type;

[0010] (2) Based on folded geometry and parametric design, construct the prototype of the sound barrier form, including linear folded form, spatial cavity form and multi-dimensional curved surface form. Among them: the linear folded form is composed of continuous folded plates, the spatial cavity form forms a closed or semi-closed cavity on the basis of the folded structure, and the multi-dimensional curved surface form adopts continuous curved surface generated by NURBS curve or fractal geometry.

[0011] Based on this, a sound barrier morphology database was constructed;

[0012] (3) Perform acoustic simulation on each sound barrier morphology in the sound barrier morphology library constructed in step (2), and select the sound barrier morphology that meets the set requirements for acoustic evaluation index as candidate sound barrier morphology.

[0013] (4) Place the candidate sound barrier shapes obtained in step (3) into the centroid case constructed in step (1) for sound field simulation, and select the optimal sound barrier shape that is suitable for each residential area type.

[0014] Specifically, in step (1), based on the noise complaint platform, several residential cases of unresolved behavioral noise complaints in the target city are obtained, and a multi-channel acoustic analyzer is used to continuously collect the sound pressure level, 1 / 3 octave band spectrum data and propagation path data of the behavioral noise of the residential cases, thereby obtaining the noise complaint data of the corresponding residential cases.

[0015] The residential area spatial data includes building attribute data, noise source locations, distribution of greenery and hard paving, and distribution of sound receiving points; among which, the building attribute data includes building height, number of floors, building plan outline, building spacing and layout type, which are obtained from publicly available real estate transaction and geographic information platforms.

[0016] Specifically, in step (1), the spatial data of several residential areas are used as high-dimensional feature vectors, and K-means clustering is used to obtain two typical residential area types that are most significantly affected by behavioral noise and have universal representativeness: central concentrated sound source residential area and edge point sound source residential area.

[0017] The cluster centers of all residential areas in various types of residential areas are obtained as centroids. The residential area whose spatial characteristics are closest to the corresponding centroid is selected as the centroid case. Its building attributes, noise source location, sound receiving point distribution and boundary conditions are extracted to construct a standardized three-dimensional acoustic benchmark model.

[0018] Specifically, in step (2), the constructed sound barrier morphology prototype is parameterized and encoded, and a set of morphological variables are defined, including: the folding angle α of the folded plate, the radius of curvature R of the spatial cavity, the depth D of the spatial cavity, the opening ratio σ of the spatial cavity, and the cross-sectional profile function F(x), to complete the parameterized modeling;

[0019] The range and step size of each morphological variable are set, and all morphological combinations are generated by traversing to obtain the morphological spectrum of each type of sound barrier morphological prototype, thereby obtaining the sound barrier morphological spectrum library. Each morphology is defined by a unique parameter vector (α, R, D, σ, F(x)).

[0020] Specifically, in step (3), the mid-to-high frequency band of 500Hz-4kHz dominated by behavioral noise is used as the target frequency band. Each sound barrier morphology in the sound barrier morphology library constructed in step (2) is imported into an acoustic simulation platform based on the sound ray tracing method or the finite difference time-domain method for acoustic simulation.

[0021] Specifically, in step (3), the acoustic evaluation indicators of each sound barrier morphology in the sound barrier morphology library include scattering coefficient δ, sound energy attenuation ΔE, and insertion loss IL.

[0022] The acoustic evaluation indicators of each sound barrier form are normalized, and a performance threshold for acoustic evaluation is set. The requirement is that all acoustic evaluation indicators of the sound barrier form are higher than the set performance threshold, thereby selecting candidate sound barrier forms.

[0023] More specifically, in step (3), after screening by performance thresholds, a further selection process can be performed, as follows:

[0024] The comprehensive performance score of each candidate sound barrier form is calculated. The comprehensive performance score is obtained by the equal weighted average of each acoustic evaluation index. The forms are sorted from high to low, and the top N candidate sound barrier forms are selected as the final candidate sound barrier forms.

[0025] Specifically, in step (4), the noise reduction difference of each candidate sound barrier form in the centroid case corresponding to different types of residential areas is calculated according to the sound field simulation. The candidate sound barrier forms that satisfy the noise reduction difference ΔL being greater than the set sound pressure level change threshold are obtained. The residential areas applicable to the candidate sound barrier forms are summarized and the optimal sound barrier form that is suitable for each residential area type is selected accordingly.

[0026] More specifically, the noise reduction difference ΔL ij The calculation is as follows:

[0027]

[0028] Among them, L0 ij L1 represents the sound pressure level at the i-th receiving point and in the j-th frequency band under unbarrier conditions. ij The sound pressure levels at the corresponding sound receiving points and frequency bands after the sound barrier is installed are obtained from the sound pressure level distribution of each floor before and after the sound barrier is installed, calculated in the sound field simulation.

[0029] This invention also provides a sound barrier shape intelligent generation system that applies the aforementioned sound barrier shape intelligent generation method for residential area behavioral noise control, comprising:

[0030] The case construction module is used to acquire behavioral noise data and spatial data of several residential areas, identify typical residential area types through clustering, and construct corresponding centroid cases for each type.

[0031] The genealogy library construction module is used to construct sound barrier morphological prototypes based on folded geometry and parametric design, including linear folded morphologies, spatial cavity morphologies, and multi-dimensional curved surface morphologies, and to construct a sound barrier morphological genealogy library accordingly.

[0032] The acoustic simulation module is used to perform acoustic simulation on each sound barrier morphology in the constructed sound barrier morphology library and select sound barrier morphologies that meet the set requirements for acoustic evaluation indicators as candidate sound barrier morphologies.

[0033] The sound field simulation module is used to place candidate sound barrier shapes into centroid cases constructed by the case construction module for sound field simulation, and select the optimal sound barrier shape that is suitable for each residential area type.

[0034] The present invention also provides a readable medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned method for intelligent generation of sound barrier morphology for residential behavioral noise control.

[0035] Beneficial effects: This invention can realize the quantitative, automated, and precise intelligent generation of sound barrier shapes, achieve precise matching between sound barrier shapes and real sound fields, and differentiate the solutions to problems such as reverberation accumulation in centrally concentrated sound source residential areas and direct sound intrusion in edge point sound source residential areas. It can be widely applied to urban residential areas, renovation of old communities, and optimization of the sound environment in public spaces. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in this invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely embodiments of this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1This is a flowchart of the intelligent generation method for sound barrier morphology based on residential behavioral noise control according to the present invention.

[0038] Figure 2 These are plan and three-dimensional schematic diagrams of two typical residential areas that are most significantly affected by behavioral noise and are generally representative in this invention; the two diagrams on the left are plan and three-dimensional schematic diagrams of a centrally concentrated sound source residential area, and the two diagrams on the right are plan and three-dimensional schematic diagrams of an edge-point sound source residential area.

[0039] Figure 3 These are example diagrams of 36 examples of the sound barrier morphology system library in this invention; wherein, the left column is an example diagram of the linear folding morphology system, the middle column is an example diagram of the spatial cavity morphology system, and the right column is an example diagram of the multidimensional curved surface morphology system.

[0040] Figure 4 This is a schematic diagram of acoustic particle simulation and acoustic energy distribution of the knife-shaped folded sound barrier morphology in this invention; wherein, the left figure is a planar schematic diagram and the right figure is a three-dimensional schematic diagram.

[0041] Figure 5 This is a schematic diagram of acoustic particle simulation and acoustic energy distribution of the box-shaped folded sound barrier in this invention; the left figure is a planar schematic diagram and the right figure is a three-dimensional schematic diagram.

[0042] Figure 6 This is a schematic diagram of acoustic particle simulation and acoustic energy distribution of the multi-groove folded sound barrier morphology in this invention; wherein, the left figure is a planar schematic diagram and the right figure is a three-dimensional schematic diagram.

[0043] Figure 7 The top 10 candidate sound barrier designs were selected based on their performance; 6 of them performed better in centrally located residential areas with concentrated sound sources, and 4 performed better in peripheral, point-source residential areas.

[0044] Figure 8 The optimal sound barrier form incorporates a sound field simulation of a residential area with a centrally concentrated sound source, and compares the noise reduction in the target frequency band with that of no sound barrier and traditional sound barriers.

[0045] Figure 9 The simulation of the sound field in a residential area with edge point sound sources is used to determine the optimal sound barrier shape, and the comparison analysis of the noise reduction in the target frequency band with no sound barrier and traditional sound barriers is presented. Detailed Implementation

[0046] To make the objectives, technical solutions and advantages of the present invention clearer, the present application will be further described in detail below with reference to specific embodiments and accompanying drawings.

[0047] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of the present invention should have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0048] This invention provides a method for intelligent generation of sound barrier shapes for residential behavioral noise control, such as... Figure 1 As shown, it includes:

[0049] (1) Obtain behavioral noise data and spatial data of several residential areas, identify typical residential area types through clustering, and construct corresponding centroid cases accordingly;

[0050] (2) Based on folded geometry and parametric design, construct the prototype of the sound barrier form, including linear folded form, spatial cavity form and multi-dimensional curved surface form. Among them: the linear folded form is composed of continuous folded plates, the spatial cavity form forms a closed or semi-closed cavity on the basis of the folded structure, and the multi-dimensional curved surface form adopts continuous curved surface generated by NURBS curve or fractal geometry.

[0051] Based on this, a sound barrier morphology database was constructed;

[0052] (3) Perform acoustic simulation on each sound barrier morphology in the sound barrier morphology library constructed in step (2), and select the sound barrier morphology that meets the set requirements as candidate sound barrier morphologies.

[0053] (4) The candidate sound barrier shapes obtained in step (3) are respectively placed into the centroid cases constructed in step (1) for sound field simulation, and the optimal sound barrier shape that is suitable for each residential area type is selected.

[0054] In this invention, in step (1), several residential cases with unresolved behavioral noise complaints in the target city can be obtained based on the noise complaint platform (such as 48 cases obtained in this project). The sound pressure level, 1 / 3 octave band spectrum data and propagation path data of the behavioral noise are continuously collected using a multi-channel acoustic analyzer, thereby obtaining the noise complaint data of the corresponding residential cases. Specifically, the collection period covers the peak period of behavioral noise in the residential area, such as evenings or holidays.

[0055] In this invention, in step (1), the spatial data of the corresponding residential area includes building attribute data, noise source location, distribution of greening and hard paving, and distribution of sound receiving points; wherein, the building attribute data includes building height, number of floors, building plan outline, building spacing and layout type (such as row or enclosed), which can be obtained from publicly available real estate transaction and geographic information platforms (such as Lianjia, Anjuke, map information software, etc.).

[0056] In this invention, in step (1), typical residential area types are identified through clustering, as follows:

[0057] Using spatial data from several residential areas as high-dimensional feature vectors, K-means clustering was employed to identify two typical residential area types that are most significantly affected by behavioral noise and are generally representative: centrally concentrated noise source areas and edge-point noise source areas; among them, such as Figure 2 , 3 As shown, the characteristics of a centrally concentrated sound source residential area are that there is a central activity space such as a central square inside the residential area, forming an internal concentrated sound source with strong sound field reverberation and significant sound gathering effect on the facade of high-rise buildings; the characteristics of an edge point sound source residential area are that there are boundary point sound sources such as entrances and exits, street-facing activity areas, etc. at the boundary of the residential area, with direct sound intensity and prominent sound diffraction effect.

[0058] In this invention, in step (1), for each type of residential area, the cluster center of all residential areas in that type of residential area can be obtained as the centroid, and the residential area whose spatial characteristics are closest to the centroid among all residential areas in that type of residential area can be selected as the centroid case. Its building attributes, noise source location, sound receiving point distribution and boundary conditions are extracted, and a standardized three-dimensional acoustic benchmark model is constructed. The building boundary contour, sound source characteristics and sound receiving point distribution are fixed as a unified benchmark environment for subsequent sound barrier morphology verification.

[0059] In this invention, reference is made to Figure 3 Step (2) specifically involves:

[0060] (21) Based on the folded geometric topology design, three types of sound barrier prototypes with different physical acoustic functions are constructed: linear folded form, spatial cavity form and multidimensional curved surface form; among them: the linear folded form is composed of continuous folded plates with a periodic triangle or trapezoidal cross section, and its function is directional reflection and sound wave guidance; the spatial cavity form forms a closed or semi-closed cavity on the basis of the folded structure, and its function is resonant sound absorption and sound energy attenuation; the multidimensional curved surface form adopts a continuous curved surface generated by NURBS curves or fractal geometry, and its function is broadband diffusion and reverberation suppression.

[0061] (22) The three types of sound barrier morphological prototypes are parametrically encoded, and a set of continuous and distinct morphological variables are defined, including: the bending angle α of the folded plate, the radius of curvature R of the spatial cavity, the depth D of the spatial cavity, the opening ratio σ of the spatial cavity and the cross-sectional profile function F(x), to complete the parametric modeling;

[0062] (23) Set the value range and step size of each morphological variable, and automatically generate all morphological combinations through parametric modeling tools. This will give you the morphological spectrum of each type of sound barrier prototype, which can be referenced. Figure 3 This leads to a continuous, traversable, and standardized library of sound barrier morphologies, where each morphology is defined by a unique parameter vector (α, R, D, σ, F(x)) to cover the acoustic functional requirements of the entire scenario.

[0063] like Figure 3 As shown, the linear folding class includes two subclasses: straight-line and curved, each with 8 forms, for a total of 16 forms. Both subclasses of linear folding include accordion folds (vertical), accordion folds (vertical) variant 1, accordion folds (vertical) variant 2, accordion folds (horizontal), oblique accordion folds, incremental accordion folds, knife-shaped folds, and incremental knife-shaped folds; one is linear and the other is curved. The spatial cavity class includes two subclasses: straight-line and curved, each with 5 forms, for a total of 10 forms. Both subclasses of spatial cavity folding include... The sound barrier morphology library includes box-shaped folds, vertical folds, non-parallel vertical folds, non-parallel vertical fold variant 1, and cut folds, with one being linear and the other curved. The multidimensional curved surface category includes two subcategories: linear and curved, each with 5 forms, for a total of 10 forms. Both subcategories of multidimensional curved surfaces include multiple V-shaped folds, multiple groove and V-shaped folds, multiple groove folds, square grid folds, and triangular grid folds, with one being linear and the other curved. In summary, in the specific embodiments of this invention, the constructed sound barrier morphology library contains a total of 36 sound barrier morphologies.

[0064] In this invention, in step (3), for each sound barrier morphology in the sound barrier morphology library constructed in step (2), the morphology is imported into an acoustic simulation platform based on the ray tracing method or the finite-difference time-domain method, such as the open-source simulation tool PachydermAcoustics, to conduct microscale acoustic particle simulations, such as... Figure 4 , 5 As shown in Figure 6, the acoustic energy distribution of the acoustic particles at different times can be obtained from different keyframe screenshot indices.

[0065] In this invention, in step (3), sound barrier shapes that meet the set requirements are selected as candidate shapes, as follows:

[0066] Using the mid-to-high frequency band dominated by behavioral noise (ranging from 500Hz to 4kHz) as the target frequency band, acoustic evaluation indicators for each sound barrier morphology in the sound barrier morphology library were calculated, including scattering coefficient δ, sound energy attenuation ΔE, and insertion loss IL. Among them, the scattering coefficient δ was obtained by performing hemispherical energy sampling in front of the sound barrier morphology and is a dimensionless indicator for measuring the uniformity of reflected sound, used to evaluate the diffusion capability of the sound barrier morphology; the sound energy attenuation ΔE was calculated by the decibel difference between incident energy and outgoing energy and used to evaluate the cavity sound absorption performance of the sound barrier morphology; the insertion loss IL is the sound pressure level difference before and after the sound barrier morphology is installed and used to evaluate the overall noise reduction function.

[0067] The acoustic evaluation indicators of each sound barrier form were normalized, and performance thresholds for acoustic evaluation were set. Sound barrier forms in which all acoustic evaluation indicators were higher than the corresponding performance thresholds were selected as candidate sound barrier forms.

[0068] Specifically, the acoustic evaluation indicators for each sound barrier form were normalized, as follows:

[0069] ;

[0070] Among them, f n f0 is a normalized acoustic evaluation index for a certain sound barrier shape, namely the aforementioned scattering coefficient δ, sound energy attenuation ΔE, and insertion loss IL. min f max These represent the extreme values ​​of the acoustic evaluation indices corresponding to all sound barrier morphologies in the sound barrier morphology database.

[0071] Therefore, performance thresholds for each acoustic evaluation index are set according to engineering requirements, and sound barrier forms in which each acoustic evaluation index, namely the aforementioned three acoustic evaluation indices, is not lower than the corresponding performance threshold are retained as preliminary candidate sound barrier forms to ensure that diffusion, sound absorption and overall noise reduction functions are all achieved.

[0072] In this embodiment, the performance threshold of the normalized values ​​of the three acoustic evaluation indicators is set to 0.5. Then, the sound barrier forms with all three acoustic evaluation indicators not lower than 0.5 are retained as preliminary candidate sound barrier forms. In this embodiment, a total of 26 preliminary candidate sound barrier forms are selected.

[0073] Due to the large number of preliminary candidate forms, in order to control the computational cost of subsequent full-domain sound field simulation, a further selection process is adopted. Specifically, the comprehensive performance score of each preliminary candidate sound barrier form is calculated. The comprehensive performance score is obtained by the equal weighted average of each acoustic evaluation index. They are sorted from high to low, and the top N (e.g., N=10) preliminary candidate sound barrier forms are selected as the final candidate sound barrier forms.

[0074] In this invention, in step (4), the final candidate sound barrier shapes obtained in step (3) (10 in this embodiment) are respectively placed into the corresponding centroid cases constructed in step (1), and imported into a sound field simulation platform based on ISO 9613 standard or geometric acoustics method (such as CadnaA) to carry out the sound field simulation of the entire residential area, and calculate the sound pressure level distribution, insertion loss, noise reduction amount and sound shadow area coverage of each floor before and after the installation of the sound barrier.

[0075] By analyzing simulation results, the noise reduction performance of each candidate sound barrier form in centroid cases corresponding to different types of residential areas was analyzed. Candidate sound barrier forms that met the condition of a noise reduction difference ΔL greater than a set sound pressure level change threshold were identified. These candidate sound barrier forms were then categorized to determine their applicable scenarios, i.e., the corresponding residential area types. Specifically, the noise reduction difference ΔL calculated for each candidate sound barrier form in its corresponding centroid case was used as the criterion. Referring to the sound pressure level change threshold perceptible to the human ear (e.g., 3dB), sound barrier forms in each type of residential area with ΔL exceeding this threshold were statistically analyzed.

[0076] For example, such as Figure 7 As shown, after conducting full-domain sound field simulations on 10 candidate sound barrier shapes in a centrally located sound source residential area, 6 of the candidate sound barrier shapes had a ΔL exceeding 3dB. These were the incremental accordion-shaped folding shape, the multiple V-shaped folding shape, the non-parallel vertical folding shape, the multiple groove-shaped folding shape, the square grid folding shape, and the triangular grid folding shape. These 6 candidate sound barrier shapes performed better in this type of residential area (solving problems such as strong reverberation, sound concentration, and superposition of reflected sound). Among them, the multi-dimensional curved surface sound barrier shape had the best noise reduction effect. In the edge point-source residential area, 4 candidate sound barrier shapes had a ΔL exceeding 3dB. These were the knife-shaped folding shape, the incremental knife-shaped folding shape, the box-shaped folding shape, and the triangular grid folding shape. These 4 candidate sound barrier shapes performed better in this type of residential area (solving problems such as direct sound intensity, prominent diffraction, and long propagation distance). Among them, the linear folding sound barrier shape had the best noise reduction effect.

[0077] The noise reduction effect can be measured using the noise reduction difference, specifically, by calculating the noise reduction difference ΔL. ij as follows:

[0078]

[0079] Among them, L0 ij L1 represents the sound pressure level at the i-th receiving point and in the j-th frequency band under unbarrier conditions. ij The sound pressure level at the corresponding sound receiving point and frequency band after the sound barrier is installed can be obtained from the sound pressure level distribution of each floor before and after the sound barrier is installed, which was calculated above.

[0080] The scene adaptability is determined based on the noise reduction effect, and the optimal sound barrier shape adapted to each residential area type is output to verify the superiority of the generated shape in terms of noise reduction in the target frequency band.

[0081] Example 1:

[0082] In this embodiment, it can be referred to Figure 8The centroid case type is a centrally located residential area with concentrated sound sources. The sound field simulation was performed by placing the unbarriered state, the traditional vertical sound barrier, and the optimal sound barrier form (multi-dimensional curved surface) selected in this embodiment into the corresponding centroid case. The results show that within a 50-meter radius of the central space, such as the central square, the average sound pressure level on each floor is reduced by 4.5 dB(A) compared to the unbarriered state and by 2.9 dB(A) compared to the traditional vertical sound barrier. This verifies that the sound barrier form selected in this embodiment has the best noise reduction effect in this type of residential area.

[0083] Example 2:

[0084] In this embodiment, it can be referred to Figure 9 The centroid case type is an edge-point sound source residential area. The state without a sound barrier, the traditional upright sound barrier, and the optimal sound barrier form (linear folding type) selected in this embodiment were placed into their respective centroid cases for full-domain sound field simulation. The results show that within a 50-meter radius of the central space, such as the central square, the average sound pressure level on each floor is reduced by 10 dB(A) compared to the state without a sound barrier, and by 4 dB(A) compared to the traditional upright sound barrier. This verifies that the sound barrier form selected in this embodiment has the best noise reduction effect in this type of residential area.

[0085] This invention also provides a sound barrier morphology intelligent generation system for residential behavioral noise control, comprising:

[0086] The case construction module is used to acquire behavioral noise data and spatial data of several residential areas, identify typical residential area types through clustering, and construct corresponding centroid cases for each type as standardized three-dimensional acoustic benchmark models.

[0087] The genealogy library construction module is used to construct sound barrier morphological prototypes based on folded geometry and parametric design, including linear folded morphologies, spatial cavity morphologies, and multi-dimensional curved surface morphologies, and to construct a sound barrier morphological genealogy library accordingly.

[0088] The acoustic simulation module is used to perform acoustic simulation on each sound barrier morphology in the constructed sound barrier morphology library and select sound barrier morphologies that meet the set requirements for acoustic evaluation indicators as candidate sound barrier morphologies.

[0089] The sound field simulation module is used to place candidate sound barrier shapes into centroid cases constructed by the case construction module for sound field simulation, and select the optimal sound barrier shape that is suitable for each residential area type.

[0090] The invention also includes a shape output module for outputting the optimal sound barrier shape for scene adaptation.

[0091] The present invention also provides a readable medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned method for intelligent generation of sound barrier morphology for residential behavioral noise control.

[0092] This invention takes the acoustic characteristics of behavioral noise as the core and residential area type as the benchmark. Based on the actual measured data of residential areas affected by behavioral noise, it clusters and extracts two typical residential area types and constructs a centroid case model. Based on folded geometry, it establishes an acoustic functional morphology library to form a parameterizable and optimizable morphology generation basis. It adopts a combination of microscopic acoustic particle simulation and residential area-scale sound field simulation to quantitatively screen the optimal noise reduction morphology adapted to residential area type and construct a full-process method of "scene recognition - morphology design - acoustic verification - scene adaptation". This invention achieves automated, precise, and intelligent generation of sound barrier morphology. The quantification is controllable, reusable, and scalable, enabling precise adaptation of sound barrier morphology to the real sound field. It addresses issues such as reverberation accumulation in centrally located sound source residential areas and direct sound intrusion in edge-located point sound source residential areas in a differentiated manner. The morphology spectrum of this invention is scalable and can adapt to residential areas of different sizes and sound source types. It is purely morphological topology optimization, without material or process constraints, and adaptable to various engineering scenarios. It has strong versatility and provides decision support for the refined management of residential sound environment. It can be widely applied to urban residential areas, renovation of old communities, and optimization of sound environment in public spaces.

[0093] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of the invention (including the claims) is limited to these examples; within the framework of the invention, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of the invention as described above, which are not provided in the details for the sake of brevity.

[0094] The embodiments of this invention are intended to cover all such substitutions, modifications, and variations falling within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of this invention should be included within the protection scope of this invention.

Claims

1. A method for intelligently generating sound barrier shapes for residential behavioral noise control, characterized in that, include: (1) Obtain behavioral noise data and spatial data of several residential areas, identify typical residential area types through clustering, and construct corresponding centroid cases for each type; (2) Based on folded geometry and parametric design, construct the prototype of the sound barrier form, including linear folded form, spatial cavity form and multi-dimensional curved surface form. Among them: the linear folded form is composed of continuous folded plates, the spatial cavity form forms a closed or semi-closed cavity on the basis of the folded structure, and the multi-dimensional curved surface form adopts continuous curved surface generated by NURBS curve or fractal geometry. Based on this, a sound barrier morphology database was constructed; (3) Perform acoustic simulation on each sound barrier morphology in the sound barrier morphology library constructed in step (2), and select the sound barrier morphology that meets the set requirements for acoustic evaluation index as candidate sound barrier morphology. (4) Place the candidate sound barrier shapes obtained in step (3) into the centroid case constructed in step (1) for sound field simulation, and select the optimal sound barrier shape that is suitable for each residential area type.

2. The method for intelligently generating the shape of a sound barrier according to claim 1, characterized in that, In step (1), several residential cases with unresolved behavioral noise complaints in the target city are obtained based on the noise complaint platform. A multi-channel acoustic analyzer is used to continuously collect the sound pressure level, 1 / 3 octave band spectrum data and propagation path data of the behavioral noise of the residential cases, thereby obtaining the noise complaint data of the corresponding residential cases. The residential area spatial data includes building attribute data, noise source locations, distribution of greenery and hard paving, and distribution of sound receiving points; among which, the building attribute data includes building height, number of floors, building plan outline, building spacing and layout type, which are obtained from publicly available real estate transaction and geographic information platforms.

3. The method for intelligently generating the shape of a sound barrier according to claim 1, characterized in that, In step (1), the spatial data of several residential areas are used as high-dimensional feature vectors. K-means clustering is used to obtain two typical residential area types that are most significantly affected by behavioral noise and have universal representativeness: central concentrated sound source residential area and edge point sound source residential area. The cluster centers of all residential areas in various types of residential areas are obtained as centroids. The residential area whose spatial characteristics are closest to the corresponding centroid is selected as the centroid case. Its building attributes, noise source location, sound receiving point distribution and boundary conditions are extracted to construct a standardized three-dimensional acoustic benchmark model.

4. The method for intelligently generating the shape of a sound barrier according to claim 1, characterized in that, In step (2), the constructed sound barrier morphology prototype is parametrically encoded, and a set of morphological variables are defined, including: the folding angle α of the folded plate, the radius of curvature R of the spatial cavity, the depth D of the spatial cavity, the opening ratio σ of the spatial cavity, and the cross-sectional profile function F(x), to complete the parametric modeling; The range and step size of each morphological variable are set, and all morphological combinations are generated by traversing to obtain the morphological spectrum of each type of sound barrier morphological prototype, thereby obtaining the sound barrier morphological spectrum library. Each morphology is defined by a unique parameter vector (α, R, D, σ, F(x)).

5. The method for intelligently generating the shape of a sound barrier according to claim 1, characterized in that, In step (3), the mid-to-high frequency band of 500Hz-4kHz dominated by behavioral noise is used as the target frequency band. Each sound barrier morphology in the sound barrier morphology library constructed in step (2) is imported into an acoustic simulation platform based on the ray tracing method or the finite difference time-domain method for acoustic simulation.

6. The method for intelligently generating the shape of a sound barrier according to claim 1, characterized in that, In step (3), the acoustic evaluation indicators of each sound barrier morphology in the sound barrier morphology library include scattering coefficient δ, sound energy attenuation ΔE and insertion loss IL. The acoustic evaluation indicators of each sound barrier form are normalized, and a performance threshold for acoustic evaluation is set. The requirement is that all acoustic evaluation indicators of the sound barrier form are higher than the set performance threshold, thereby selecting candidate sound barrier forms.

7. The method for intelligently generating the shape of a sound barrier according to claim 6, characterized in that, In step (3), after screening by performance threshold, further selection can be performed, as follows: The comprehensive performance score of each candidate sound barrier form is calculated. The comprehensive performance score is obtained by the equal weighted average of each acoustic evaluation index. The forms are sorted from high to low, and the top N candidate sound barrier forms are selected as the final candidate sound barrier forms.

8. The method for intelligently generating the shape of a sound barrier according to claim 1, characterized in that, In step (4), the noise reduction difference of each candidate sound barrier form in the centroid case corresponding to different types of residential areas is calculated according to the sound field simulation. The candidate sound barrier forms that satisfy the noise reduction difference ΔL being greater than the set sound pressure level change threshold are obtained. The types of residential areas to which the candidate sound barrier forms are applicable are summarized, and the optimal sound barrier form that is compatible with each type of residential area is selected accordingly.

9. The method for intelligently generating the shape of a sound barrier according to claim 8, characterized in that, The noise reduction difference ΔL ij The calculation is as follows: ； Among them, L0 ij L1 represents the sound pressure level at the i-th receiving point and in the j-th frequency band under unbarrier conditions. ij The sound pressure levels at the corresponding sound receiving points and frequency bands after the sound barrier is installed are obtained from the sound pressure level distribution of each floor before and after the sound barrier is installed, calculated in the sound field simulation.

10. A sound barrier shape intelligent generation system applying the sound barrier shape intelligent generation method for residential behavioral noise control according to any one of claims 1-9, characterized in that, include: The case construction module is used to acquire behavioral noise data and spatial data of several residential areas, identify typical residential area types through clustering, and construct corresponding centroid cases for each type. The genealogy library construction module is used to construct sound barrier morphological prototypes based on folded geometry and parametric design, including linear folded morphologies, spatial cavity morphologies, and multi-dimensional curved surface morphologies, and to construct a sound barrier morphological genealogy library accordingly. The acoustic simulation module is used to perform acoustic simulation on each sound barrier morphology in the constructed sound barrier morphology library and select sound barrier morphologies that meet the set requirements for acoustic evaluation indicators as candidate sound barrier morphologies. The sound field simulation module is used to place candidate sound barrier shapes into centroid cases constructed by the case construction module for sound field simulation, and select the optimal sound barrier shape that is suitable for each residential area type.

Images