A method and apparatus for evaluating sound insulation of a building

Abstract

The application relates to the technical field of building acoustics evaluation, and particularly relates to a building sound insulation effect evaluation method and device, which comprises the following steps: collecting design drawings of a building and acoustic attribute parameters of used materials, building a three-dimensional geometric model based on BIM technology according to the design drawings, and constructing an initial building acoustic digital twin according to the three-dimensional geometric model and the acoustic attribute parameters; synchronously collecting objective data and subjective quantitative data in real time; comparing the real-time collected data with initial simulation data of the initial building acoustic digital twin and a preset standard frequency spectrum library, and positioning sound insulation weak points; feeding back the real-time collected data to the initial building acoustic digital twin, and calibrating the initial building acoustic digital twin; constructing multiple sound insulation optimization schemes in the calibrated building acoustic digital twin based on the positioned sound insulation weak points, and performing virtual effect testing to generate a comprehensive evaluation report. The application is helpful to improve sound insulation accuracy and efficiency.

Description

Technical Field

[0001] This application relates to the field of architectural acoustics assessment technology, and in particular to a method and apparatus for assessing the sound insulation effect of a building. Background Technology

[0002] In the field of building engineering, sound insulation performance is one of the core indicators affecting the comfort and functionality of buildings, directly impacting the physical and mental health and work / study efficiency of residents and users. As people's demands for quality of life increase, the need for evaluating building sound insulation is becoming increasingly urgent, especially in acoustically sensitive environments such as residences, offices, hospitals, and schools. Accurate sound insulation assessment has become a crucial step in building acceptance, renovation, and optimization.

[0003] However, existing methods for evaluating the sound insulation effect of buildings have many shortcomings: traditional evaluations rely heavily on manual inspection and experience-based judgment, lack systematic model support, and the simulation results deviate significantly from the actual acoustic state of the building, making it difficult to fully reflect the weak points in sound insulation; they cannot accurately pinpoint specific spatial locations and defect types, and mostly remain at the regional level, which brings difficulties to subsequent optimization and renovation; optimization schemes lack virtual verification, and rely heavily on on-site trial and error, which is costly, time-consuming, and inefficient.

[0004] Therefore, there is an urgent need for a building sound insulation performance evaluation method that is both accurate and efficient, in order to overcome the shortcomings of existing technologies. Summary of the Invention

[0005] Therefore, it is necessary to provide a method for evaluating the sound insulation effect of buildings that can improve the accuracy and efficiency of sound insulation, addressing the aforementioned technical problems.

[0006] In a first aspect, this application provides a method for evaluating the sound insulation effect of a building, the method comprising:

[0007] Collect the building's design drawings and the acoustic property parameters of the materials used. Based on the design drawings, build a three-dimensional geometric model using BIM technology. Based on the three-dimensional geometric model and the acoustic property parameters, construct an initial building acoustic digital twin.

[0008] Simultaneously collect objective data and subjective quantitative data in real time. The objective data includes acoustic signal data, structural vibration data, and environmental parameter data. The subjective quantitative data includes noise annoyance quantification scores, speech intelligibility interference scores, and scene annotation information.

[0009] The real-time collected data is compared with the initial simulation data of the initial building acoustic digital twin and the preset standard spectrum library to locate the weak points in sound insulation;

[0010] The real-time acquired data is fed back to the initial building acoustic digital twin for calibration.

[0011] Based on the location of weak points in sound insulation, multiple sound insulation optimization schemes are constructed in a calibrated digital twin of building acoustics, and virtual effect tests are conducted to generate a comprehensive evaluation report.

[0012] In one embodiment, comparing the real-time acquired data with the initial simulation data of the initial building acoustic digital twin and a preset standard spectrum library to locate weak points in sound insulation includes:

[0013] The real-time collected data is compared with the initial simulation data of the initial building acoustic digital twin and the preset standard spectrum library on a frequency band basis to identify abnormal frequency bands where the sound insulation is lower than the simulation value and the standard threshold.

[0014] Extract the time-domain and frequency-domain features of the abnormal frequency band and match them with the corresponding typical defect features in a preset standard spectrum library;

[0015] Based on the extracted abnormal frequency band features, the main noise source types are identified through a fusion model of support vector machine and random forest.

[0016] Based on the abnormal frequency band, the typical defect characteristics of the matching, and the main noise source type, the weak points in sound insulation are located.

[0017] In one embodiment, locating the weak point in sound insulation based on the abnormal frequency band, the typical defect characteristics of the matching signal, and the type of the main noise source includes:

[0018] Based on the abnormal frequency band and the typical defect characteristics of the matching, the microphone array is controlled to scan the corresponding building area, and the sound signal of the abnormal frequency band and the vibration data of the building components in the corresponding area are collected simultaneously.

[0019] Based on the collected acoustic signals and vibration data, the signal phase is calibrated and the acoustic propagation path of the abnormal frequency band is focused using beamforming technology. Combined with the arrival time difference of signals collected by multi-node microphones, the spatial range of acoustic leakage is initially determined.

[0020] Based on the propagation characteristics of the main noise source types, the initially identified spatial range of sound energy leakage is precisely corrected to pinpoint the specific spatial location of the weak point in sound insulation.

[0021] In one embodiment, prior to the synchronous real-time acquisition of objective data and subjective quantitative data, the method further includes:

[0022] Based on the simulation results of the initial building acoustic digital twin, the optimal deployment scheme of the sensor and microphone array is determined;

[0023] According to the optimal deployment scheme, sensors and microphone arrays are deployed at key building nodes to collect both objective and subjective quantitative data.

[0024] In one embodiment, the location-based weak points in sound insulation are identified by constructing multiple sound insulation optimization schemes in a calibrated digital twin of building acoustics and conducting virtual effect tests to generate a comprehensive evaluation report, including:

[0025] Based on the identified weak points in sound insulation, at least three differentiated sound insulation optimization schemes were developed.

[0026] Import each sound insulation optimization scheme into the calibrated building acoustic digital twin, simulate the application effect of each sound insulation optimization scheme in the actual complex acoustic environment, and output the objective sound insulation index and abnormal frequency band improvement range corresponding to each sound insulation optimization scheme.

[0027] Based on the subjective quantitative data and the preset correlation model, predict the reduction in user noise annoyance, the improvement in speech intelligibility, and the soundscape satisfaction corresponding to each sound insulation optimization scheme.

[0028] A comprehensive evaluation report is generated based on the reduction in user noise annoyance, the improvement in speech clarity, the soundscape satisfaction, the objective sound insulation index, and the improvement in abnormal frequency bands.

[0029] In one embodiment, calibrating the initial architectural acoustic digital twin includes:

[0030] The real-time collected objective data is compared with the initial simulation data of the initial building acoustic digital twin in a dimension-by-dimensional manner to identify deviations;

[0031] Based on the aforementioned deviation term, the acoustic attribute parameters in the initial building acoustic digital twin are corrected using the least squares method, and the correction results are compensated and calibrated in conjunction with the environmental parameter data.

[0032] By integrating the structural vibration data with the sound energy propagation path analysis results, the sound transmission attenuation coefficient of the structural components in the initial building acoustic digital twin is adjusted to ensure that the sound transmission mechanism of the digital twin is consistent with the actual state of the building.

[0033] In one embodiment, the method for evaluating the sound insulation effect of a building further includes:

[0034] The data from each stage of the entire assessment process are structured and archived to form a full life-cycle archive of building acoustics, providing a reusable reference sample for subsequent sound insulation effect assessments of similar buildings.

[0035] Based on the long-term accumulated architectural acoustics full life cycle archive, the architectural acoustics digital twin is continuously iterated and updated.

[0036] Secondly, this application also provides a device for evaluating the sound insulation effect of a building. The device includes:

[0037] The digital twin construction module is used to collect the design drawings of the building and the acoustic property parameters of the materials used. Based on the design drawings, a three-dimensional geometric model is built using BIM technology, and an initial building acoustic digital twin is constructed based on the three-dimensional geometric model and the acoustic property parameters.

[0038] The data acquisition module is used to synchronously and in real time acquire objective data and subjective quantitative data. The objective data includes acoustic signal data, structural vibration data, and environmental parameter data. The subjective quantitative data includes noise annoyance quantification scores, speech intelligibility interference scores, and scene annotation information.

[0039] The weak point location module is used to compare the real-time collected data with the initial simulation data of the initial building acoustic digital twin and the preset standard spectrum library to locate the weak points in sound insulation.

[0040] The digital twin calibration module is used to feed back the real-time acquired data to the initial building acoustic digital twin and calibrate the initial building acoustic digital twin.

[0041] The assessment report generation module is used to construct multiple sound insulation optimization schemes and conduct virtual effect tests in a calibrated digital twin of building acoustics based on the location of weak points in sound insulation, and generate a comprehensive assessment report.

[0042] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0043] Collect the building's design drawings and the acoustic property parameters of the materials used. Based on the design drawings, build a three-dimensional geometric model using BIM technology. Based on the three-dimensional geometric model and the acoustic property parameters, construct an initial building acoustic digital twin.

[0044] Simultaneously collect objective data and subjective quantitative data in real time. The objective data includes acoustic signal data, structural vibration data, and environmental parameter data. The subjective quantitative data includes noise annoyance quantification scores, speech intelligibility interference scores, and scene annotation information.

[0045] The real-time collected data is compared with the initial simulation data of the initial building acoustic digital twin and the preset standard spectrum library to locate the weak points in sound insulation;

[0046] The real-time acquired data is fed back to the initial building acoustic digital twin for calibration.

[0047] Based on the location of weak points in sound insulation, multiple sound insulation optimization schemes are constructed in a calibrated digital twin of building acoustics, and virtual effect tests are conducted to generate a comprehensive evaluation report.

[0048] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:

[0049] Collect the building's design drawings and the acoustic property parameters of the materials used. Based on the design drawings, build a three-dimensional geometric model using BIM technology. Based on the three-dimensional geometric model and the acoustic property parameters, construct an initial building acoustic digital twin.

[0050] Simultaneously collect objective data and subjective quantitative data in real time. The objective data includes acoustic signal data, structural vibration data, and environmental parameter data. The subjective quantitative data includes noise annoyance quantification scores, speech intelligibility interference scores, and scene annotation information.

[0051] The real-time collected data is compared with the initial simulation data of the initial building acoustic digital twin and the preset standard spectrum library to locate the weak points in sound insulation;

[0052] The real-time acquired data is fed back to the initial building acoustic digital twin for calibration.

[0053] Based on the location of weak points in sound insulation, multiple sound insulation optimization schemes are constructed in a calibrated digital twin of building acoustics, and virtual effect tests are conducted to generate a comprehensive evaluation report.

[0054] In summary, this application includes the following beneficial technical effects:

[0055] By constructing an initial digital acoustic twin of the building using BIM technology and combining it with real-time data acquisition, the initial digital acoustic twin is calibrated to ensure a high degree of consistency between the simulation results and the actual acoustic state of the building. Simultaneous real-time acquisition of objective and subjective quantitative data ensures both the scientific rigor of the assessment and aligns with actual user experience, avoiding biases caused by single-dimensional evaluation. Comparing the real-time data with the initial simulation data of the initial digital acoustic twin and a preset standard spectrum library identifies weak points in sound insulation and accurately identifies abnormal frequency bands where sound insulation fails to meet standards, avoiding the vague positioning problems of traditional assessments. Multiple sound insulation optimization schemes are constructed within the calibrated digital acoustic twin, and virtual effect tests are conducted. This allows for the prediction of the sound insulation effect and the extent of improvement in subjective experience without on-site trial and error. The generated comprehensive evaluation report provides a scientific basis for sound insulation renovation decisions, improving the efficiency and effectiveness of sound insulation renovation. Attached Figure Description

[0056] Figure 1 This is a flowchart illustrating a method for evaluating the sound insulation effect of a building in one embodiment;

[0057] Figure 2 This is a flowchart illustrating a method for evaluating the sound insulation effect of a building in another embodiment;

[0058] Figure 3 This is a structural block diagram of a building sound insulation effect evaluation device in one embodiment. Detailed Implementation

[0059] This invention provides a method and apparatus for evaluating the sound insulation effect of buildings.

[0060] The embodiments of the present invention will now be described in more detail with reference to the accompanying drawings. While some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the accompanying drawings and embodiments are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0061] In the description of the embodiments disclosed in this invention, the term "comprising" and similar terms should be understood as open-ended inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc., may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0062] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the building sound insulation effect evaluation method in this invention includes:

[0063] S100 collects the building's design drawings and the acoustic properties of the materials used. Based on the design drawings, it builds a three-dimensional geometric model using BIM technology, and constructs an initial digital twin of the building's acoustics based on the three-dimensional geometric model and the acoustic properties.

[0064] Specifically, the process begins by collecting the building's design drawings and the acoustic properties of the materials used (such as sound insulation, damping coefficient, and transmission loss) to ensure the data covers the core acoustic information of the building structure and materials. Based on the design drawings, the parametric modeling advantages of BIM technology are utilized to build a 1:1 matching 3D geometric model of the building, accurately reproducing the building's spatial distribution, component dimensions, node connection methods, and material distribution. Subsequently, this 3D geometric model is combined with the material acoustic property parameters to construct an initial digital twin of the building's acoustics. This twin can initially simulate the acoustic propagation patterns and sound insulation performance of the building under ideal operating conditions, providing a basic model support for subsequent evaluation.

[0065] In this embodiment, an initial digital twin of building acoustics is constructed using a three-dimensional geometric model and acoustic property parameters, which enables the digital replication of the building's acoustic state and breaks through the limitations of traditional assessments that lack precise model support. Through the deep integration of the three-dimensional geometric model and acoustic property parameters, the sound propagation law under ideal building conditions can be simulated in advance, providing clear targets for subsequent data collection and weak point location, and avoiding blind assessment.

[0066] S200 synchronously collects objective data and subjective quantitative data in real time.

[0067] Specifically, after the initial digital twin of the building acoustics is constructed, a synchronous real-time data acquisition process is initiated. The acquired data is divided into two categories: objective data and subjective quantitative data, ensuring that the evaluation takes into account both physical indicators and the actual user experience. Objective data is acquired through deployed multimodal sensors and microphone arrays, covering acoustic signal data across the entire 20Hz-20kHz frequency band (such as sound pressure level, audio waveform, and spectral distribution in each band), structural vibration data (vibration frequency, amplitude, and acceleration) of building components (walls, floors, beams, and manhole walls), and environmental parameter data (temperature, humidity, air pressure, and wind speed). All objective data is associated with a unique timestamp and spatial coordinates to ensure the spatiotemporal consistency and traceability of the data. Subjective quantitative data is collected through various methods such as standardized evaluation questionnaires and online feedback. Differentiated evaluation dimensions are designed for different usage scenarios (such as sleep, office, and study), including a quantitative score for noise annoyance, a score for speech intelligibility interference, and scene annotation information (such as the time of noise occurrence, duration, and range of impact), comprehensively capturing the user's actual acoustic experience feedback.

[0068] The S300 compares real-time collected data with the initial simulation data of the initial building acoustic digital twin and the preset standard spectrum library to locate weak points in sound insulation.

[0069] Specifically, the actual collected objective data and subjective quantitative data are compared and analyzed in multiple dimensions with the initial simulation data of the initial building acoustic digital twin and the preset standard spectrum. The preset standard spectrum covers industry standards in the field of building acoustics and measured data of typical building sound insulation performance, including standard sound insulation spectra and typical defect feature maps corresponding to different building types, different components, and different noise sources. By comparing frequency bands one by one, abnormal frequency bands with sound insulation values ​​lower than the simulated value and standard threshold are identified. The time domain (peak value, effective value, pulse width) and frequency domain (center frequency, spectral centroid, harmonic components) features of abnormal frequency bands are extracted and matched with typical defect features in the preset standard spectrum library (such as resonance, sound leakage through gaps, lateral sound transmission, etc.). Combined with machine learning algorithms, the main noise source types (such as traffic noise, equipment noise, social noise) are identified, and finally, the specific spatial location, defect type, and impact range of the weak point in sound insulation are accurately located.

[0070] In this embodiment, by comparing real-time collected data with the initial simulation data of the initial building acoustic digital twin and the frequency band of the preset standard spectrum library, abnormal frequency bands with substandard sound insulation can be accurately identified, avoiding the limitations of the overall fuzzy judgment in traditional assessments.

[0071] The S400 feeds back the real-time collected data to the initial building acoustic digital twin for calibration.

[0072] Specifically, to improve the simulation accuracy of the digital twin, real-time collected data is fed back to the initial architectural acoustic digital twin for calibration and optimization. Specifically, by comparing the measured data with the simulated data dimension by dimension, deviations in the digital twin (such as material acoustic property deviations, structural sound transmission coefficient deviations, environmental adaptation deviations, etc.) are identified. The least squares method is used to correct the material acoustic property parameters in the digital twin, and the correction results are compensated and calibrated in combination with environmental parameter data. At the same time, structural vibration data and sound energy propagation path analysis results are integrated to adjust the sound transmission attenuation coefficient of structural components and the lateral sound transmission calculation weights to ensure that the sound transmission mechanism of the digital twin is consistent with the building's time state. After calibration, the error between the simulation results of the digital twin and the measured data can be controlled within a preset range.

[0073] In this embodiment, calibrating the initial building acoustic digital twin can solve the deviation problem between the initial building acoustic digital twin and the actual working conditions, and provide reliable model support for the virtual testing of subsequent optimization schemes.

[0074] S500, based on the location of weak points in sound insulation, constructs multiple sound insulation optimization schemes in a calibrated digital twin of building acoustics and conducts virtual effect tests to generate a comprehensive evaluation report.

[0075] Specifically, based on the identified weak points and root causes of sound insulation defects, at least three differentiated sound insulation optimization schemes are constructed in the calibrated digital twin of building acoustics. The scheme types include any combination of material replacement (such as using high sound insulation materials), structural sealing (such as filling gaps and strengthening joint sealing), sound barrier addition (such as setting up sound barriers in the noise propagation path), damping layer superposition (such as adding damping layers on the surface of components), and soundscape adaptation (such as optimizing the spatial sound environment through acoustic design). The parameters of each optimization scheme (such as material acoustic properties, structural dimensions, installation location, construction process, etc.) are imported into the calibrated digital twin to simulate the application effect of each scheme in a real complex acoustic environment. The objective sound insulation index and the improvement of abnormal frequency bands corresponding to each scheme are output. Combined with subjective quantitative data and preset correlation models, the user noise annoyance reduction value, speech intelligibility improvement value and soundscape satisfaction corresponding to each optimization scheme are predicted. Based on the above test data, a quantitative comparison is carried out from multiple dimensions such as sound insulation effect, cost budget, construction difficulty, environmental protection, service life and maintenance cost. A comprehensive evaluation report is generated, which includes scheme priority ranking, implementation suggestions, long-term performance prediction and targeted maintenance schemes, providing a scientific and feasible decision-making basis for building sound insulation optimization.

[0076] In one embodiment, such as Figure 2 As shown, S300 includes:

[0077] S310 compares the real-time acquired data with the initial simulation data of the initial building acoustic digital twin and the preset standard spectrum library band by band to identify abnormal frequency bands where the sound insulation is lower than the simulation value and the standard threshold.

[0078] S320 extracts the time-domain and frequency-domain features of abnormal frequency bands and matches them with the corresponding typical defect features in the preset standard spectrum library;

[0079] S330, based on the extracted abnormal frequency band features, identifies the main noise source types through a fusion model of support vector machine and random forest;

[0080] S340 locates weak points in sound insulation based on abnormal frequency bands, typical defect characteristics of matching, and the main noise source types.

[0081] Specifically, the real-time acquired data is compared and analyzed band by band with the initial simulation data of the initial building acoustic digital twin and the preset standard spectrum library. During the comparison process, the sound insulation-related parameters of each frequency band are compared one by one according to the preset frequency band division standard (such as dividing the sound frequency band into multiple frequency bands according to 1 / 3 octave or octave). For example, the sound pressure level, transmission loss and other parameters of each frequency band acquired in real time are compared with the corresponding frequency band parameters obtained by the initial building acoustic digital twin simulation. At the same time, they are compared with the standard sound insulation threshold of the corresponding building type and corresponding component in the preset standard spectrum library in that frequency band to screen out abnormal frequency bands. Subsequently, time-domain and frequency-domain features of the abnormal frequency bands were extracted. These features were then matched with typical defect features in a pre-defined standard spectrum. The pre-defined standard spectrum library stores feature parameter templates corresponding to common defects in architectural acoustics (such as component resonance, gap leakage, coincidence effect, lateral sound transmission, and material aging). By calculating the similarity between the measured features and the template features, the defect type of the weak point in sound insulation was preliminarily determined. Next, the extracted abnormal frequency band features were input into a support vector machine and random forest fusion model. Through the model's classification and identification, the main noise source types were output. Finally, the results were... By combining the abnormal frequency band, typical defect characteristics of the matching, and the main noise source types, and combining the three-dimensional geometric model of the initial building acoustic digital twin, spatial positioning analysis is carried out. For example, if the abnormal frequency band is low frequency, the matching defect is structural sound transmission, and the main noise source is equipment noise, then the focus is on analyzing the structural components around the equipment (such as the floor slab and walls of the equipment room); if the abnormal frequency band is mid-high frequency, the matching defect is sound leakage through gaps, and the main noise source is traffic noise, then the focus is on investigating areas with concentrated gaps, such as door and window joints and pipe shaft penetrations on the building exterior walls, and finally accurately locating the specific spatial location of the weak sound insulation point.

[0082] In one embodiment, locating weak points in sound insulation based on abnormal frequency bands, typical defect characteristics of matching, and the type of main noise source includes:

[0083] Based on the anomalous frequency band and the typical defect characteristics of the matching, the microphone array is controlled to scan the corresponding building area, and the acoustic signals of the anomalous frequency band and the vibration data of the building components in the corresponding area are collected simultaneously. According to the collected acoustic signals and vibration data, the signal phase is calibrated and the sound energy propagation path of the anomalous frequency band is focused through beamforming technology. Combined with the arrival time difference of the signals collected by the multi-node microphones, the spatial range of sound energy leakage is initially locked. According to the propagation characteristics of the main noise source types, the spatial range of sound energy leakage initially locked is precisely corrected to locate the specific spatial location of the weak sound insulation point.

[0084] Specifically, based on the abnormal frequency band and the typical defect characteristics matched, the target scanning area is determined. For example, if the abnormal frequency band is 125Hz-250Hz and the defect characteristic is sound leakage from pipe wells, then the target scanning area is all pipe wells in the building and their surrounding walls and floors. The microphone array is controlled to perform a full-coverage scan of this target area. Simultaneously, a structural vibration sensor is linked to synchronously collect the acoustic signals of the abnormal frequency band and the vibration data of the corresponding building components. Then, the collected acoustic signals and vibration data are processed to focus the sound energy propagation path. Specifically, beamforming technology is used to perform phase calibration and signal superposition on the abnormal frequency band acoustic signals collected by the microphone array. Algorithms are used to suppress interference signals generated by scattering and reflection, and to enhance the sound energy signal in the target direction, thereby accurately focusing on the abnormal frequency band. The propagation path of sound energy in the normal frequency band is calculated, and the time difference of arrival (TDOA) of the same signal collected by multiple microphones is calculated. Combined with the known layout coordinates and sound velocity of the microphone array, the approximate spatial range of sound energy leakage is deduced by the principle of triangulation, and the candidate areas of weak points in sound insulation are initially identified. Finally, the spatial range is precisely corrected based on the propagation characteristics of the main noise source types. The propagation characteristics of different types of noise sources are significantly different. For example, traffic noise is mainly low-frequency airborne sound with strong diffraction ability and is easily transmitted through weak points such as door and window gaps and pipe wells; equipment noise is mainly structural sound transmission, with vibration transmitted through components such as beams, columns, and floor slabs, and the sound energy is concentrated in the components around the equipment; social noise is mainly mid-to-high frequency airborne sound with a relatively linear propagation path.

[0085] In one embodiment, before synchronously collecting objective data and subjective quantitative data in real time, the method further includes:

[0086] Based on the simulation results of the initial building acoustic digital twin, the optimal deployment scheme of the sensor and microphone array is determined. According to the optimal deployment scheme, the sensor and microphone array are deployed at key nodes of the building to achieve the collection of objective data and subjective quantitative data.

[0087] Specifically, based on the simulation results of the initial building acoustic data twin, the distribution pattern of building acoustics is analyzed to determine the optimal deployment scheme of sensor and microphone arrays, clarifying the deployment location, density and angle; then, according to the scheme, sensor and microphone arrays are deployed at key nodes of the building to ensure coverage of key parts such as potentially weak areas and densely populated areas, so as to achieve comprehensive and accurate collection of objective data and subjective quantitative data, and avoid data omission or redundancy due to unreasonable deployment.

[0088] In one embodiment, based on the located weak points in sound insulation, multiple sound insulation optimization schemes are constructed in a calibrated digital twin of building acoustics, and virtual effect tests are conducted to generate a comprehensive evaluation report, including:

[0089] Based on the identified weak points in sound insulation, at least three differentiated sound insulation optimization schemes are constructed. Each sound insulation optimization scheme is imported into a calibrated digital twin of building acoustics to simulate the application effect of each scheme in a real complex acoustic environment. The objective sound insulation index and the improvement range of abnormal frequency bands corresponding to each sound insulation optimization scheme are output. Based on subjective quantitative data and a preset correlation model, the reduction value of user noise annoyance, the improvement range of speech intelligibility, and the soundscape satisfaction corresponding to each sound insulation optimization scheme are predicted. A comprehensive evaluation report is generated based on the reduction value of user noise annoyance, the improvement range of speech intelligibility, the soundscape satisfaction, the objective sound insulation index, and the improvement range of abnormal frequency bands.

[0090] Specifically, firstly, based on the identified weak points in sound insulation, at least three differentiated sound insulation optimization schemes are constructed. The scheme design must address the type, location, and direction of influence of the weak points to ensure the scheme's relevance and diversity. For example, for weak points with sound leakage through gaps, a sealing material filling scheme (using high-elasticity sealant to fill the gaps) and a joint reinforcement scheme (adding a sealing gasket + sealant for double sealing) can be designed. Next, each sound insulation optimization scheme is imported into a calibrated digital twin of building acoustics to simulate the application effect of each scheme in a real, complex acoustic environment, including sound insulation performance under different noise source intensities and environmental conditions. During the testing process, objective sound insulation indicators corresponding to each scheme are output, including weighted sound insulation, sound insulation in each frequency band, and impact sound insulation. The improvement margin for abnormal frequency bands is calculated, i.e., the difference between the sound insulation of abnormal frequency bands after implementation and before implementation, clarifying the improvement effect of the solution on weak points. Then, the virtual test results (objective sound insulation indicators, improvement margin for abnormal frequency bands, etc.) of each sound insulation optimization solution are input into a preset correlation model. Combined with subjective quantitative data, and through the model's built-in mapping rules and scene adaptation coefficients, the user experience improvement indicators corresponding to each solution are calculated, including the reduction in user noise annoyance, the improvement in speech intelligibility, and soundscape satisfaction. The preset correlation model is trained with a large amount of sample data and has established a mapping relationship between objective sound insulation indicators (such as sound insulation and frequency band improvement margin) and subjective evaluation results (noise annoyance, speech intelligibility, and soundscape satisfaction). The comprehensive evaluation report is generated by combining the reduction in user noise annoyance, the improvement in speech intelligibility, soundscape satisfaction, objective sound insulation indicators, and improvement margin for abnormal frequency bands of each solution, along with additional evaluation dimensions such as cost budget, construction difficulty, environmental friendliness, service life, and maintenance costs.

[0091] In one embodiment, calibrating an initial architectural acoustic digital twin includes:

[0092] The objective data collected in real time is compared with the initial simulation data of the initial building acoustic digital twin in each dimension to identify deviations. Based on the deviations, the acoustic attribute parameters in the initial building acoustic digital twin are corrected using the least squares method, and the correction results are compensated and calibrated in combination with environmental parameter data. The sound transmission attenuation coefficient of the structural components in the initial building acoustic digital twin is adjusted by fusing structural vibration data and sound energy propagation path analysis results to ensure that the sound transmission mechanism of the digital twin is consistent with the actual state of the building.

[0093] Specifically, real-time collected objective data is compared dimension-by-dimensionally with the initial simulation data of the initial building acoustic digital twin. Comparison dimensions include sound pressure level, sound insulation, vibration frequency, and amplitude across various frequency bands. By comparing the differences between measured and simulated data, deviations in the digital twin are identified. These deviations include: material acoustic property deviations (i.e., the preset material sound insulation, sound absorption coefficient, and other parameters in the digital twin do not match the measured data); structural sound transmission parameter deviations (i.e., the sound attenuation coefficient and lateral sound transmission calculation weights of structural components in the digital twin are inconsistent with the actual sound transmission mechanism); and environmental adaptation deviations (i.e., the digital twin does not fully consider the influence of environmental parameters such as temperature, humidity, and air pressure on acoustic performance, leading to a deviation between the simulation results and the measured data). Then, the least squares method is used to correct the identified material acoustic property deviations, minimizing the error between the corrected simulation results and the measured data, while simultaneously compensating and calibrating using environmental parameter data. Finally, based on the sound energy propagation path analysis results (obtained through beamforming technology and TDOA calculations), the role of structural components in the sound energy propagation process is clarified. For example, some beams are key carriers of lateral sound transmission, and their sound attenuation coefficient directly affects the sound insulation effect of adjacent spaces. Combining structural vibration data, the correlation between the vibration propagation law of components and sound energy transmission is analyzed. If the vibration frequency of a component is consistent with the frequency of an abnormal frequency band sound signal, and the vibration phase leads the sound signal, it indicates that the component is the core carrier of structural sound transmission, and its sound attenuation coefficient needs to be appropriately reduced (i.e., increasing the sound energy transmission loss in the component). Conversely, if the component has no significant impact on sound energy propagation, its sound attenuation coefficient is maintained or finely adjusted. Through the above adjustments, the sound transmission mechanism of the digital twin is ensured to be consistent with the actual state of the building, providing support for subsequent virtual testing of optimization schemes.

[0094] In one embodiment, the method for evaluating the sound insulation effect of a building further includes:

[0095] The data from each stage of the entire assessment process are structured and archived to form a building acoustics lifecycle archive, providing a reusable reference sample for subsequent sound insulation effect assessments of similar buildings; based on the long-term accumulated building acoustics lifecycle archive, the building acoustics digital twin is continuously iterated and updated.

[0096] Specifically, the data from each stage of the entire assessment process is first structured and archived. This archived data covers core information throughout the assessment process, including basic building data (design drawings, material acoustic properties), equipment deployment data (deployment schemes, installation locations, and equipment parameters for sensors and microphone arrays), collected data (real-time objective and subjective quantitative data), analyzed data (abnormal frequency band identification results, defect feature matching results, noise source identification results, and data on weak points in sound insulation), model data (initial building acoustic digital twin parameters, calibration records, and optimization scheme parameters), test data (objective indicators of virtual effect testing and subjective experience prediction results), and report data. This data is then categorized and organized, establishing standardized data formats and classification indexes (indexed by building type, assessment stage, data type, etc.), forming a complete building acoustics lifecycle archive. This archive uses a traceable and searchable storage format, providing reusable reference samples for subsequent sound insulation effect assessments of similar buildings. Then, based on this long-term accumulated building acoustics lifecycle archive, the building acoustics digital twin is continuously iterated and updated.

[0097] In one embodiment, such as Figure 3 As shown, a building sound insulation effect evaluation device is provided, including: a digital twin construction module 10, a data acquisition module 20, a weak point location module 30, a digital twin calibration module 40, and an evaluation report generation module 50, wherein:

[0098] The digital twin construction module 10 is used to collect the building's design drawings and the acoustic property parameters of the materials used. Based on the design drawings, a three-dimensional geometric model is built using BIM technology, and an initial building acoustic digital twin is constructed based on the three-dimensional geometric model and the acoustic property parameters.

[0099] The data acquisition module 20 is used to synchronously and in real time acquire objective data and subjective quantitative data. The objective data includes acoustic signal data, structural vibration data and environmental parameter data, while the subjective quantitative data includes noise annoyance quantification scores, speech intelligibility interference scores and scene annotation information.

[0100] The weak point location module 30 is used to compare the real-time acquired data with the initial simulation data of the initial building acoustic digital twin and the preset standard spectrum library to locate the weak points in sound insulation.

[0101] The digital twin calibration module 40 is used to feed back real-time acquired data to the initial building acoustic digital twin and calibrate the initial building acoustic digital twin.

[0102] The assessment report generation module 50 is used to construct multiple sound insulation optimization schemes and conduct virtual effect tests in a calibrated digital twin of building acoustics based on the location of weak points in sound insulation, and generate a comprehensive assessment report.

[0103] In one embodiment, the weak point location module 30 is further configured to compare the real-time acquired data with the initial simulation data of the initial building acoustic digital twin and the preset standard spectrum library on a frequency band basis to identify abnormal frequency bands where the sound insulation is lower than the simulation value and the standard threshold; extract the time domain and frequency domain features of the abnormal frequency bands and match them with the corresponding typical defect features in the preset standard spectrum library; based on the extracted abnormal frequency band features, identify the main noise source types through a support vector machine and random forest fusion model; and locate the sound insulation weak points according to the abnormal frequency bands, the matched typical defect features, and the main noise source types.

[0104] In one embodiment, the weak point location module 30 is further configured to control the microphone array to scan the corresponding building area based on the abnormal frequency band and the matching typical defect characteristics, and simultaneously collect the sound signal of the abnormal frequency band and the vibration data of the building components in the corresponding area; based on the collected sound signal and vibration data, the signal phase is calibrated by beamforming technology, the sound energy propagation path of the abnormal frequency band is focused, and the spatial range of sound energy leakage is initially locked by combining the arrival time difference of the signals collected by the multi-node microphones; based on the propagation characteristics of the main noise source type, the spatial range of sound energy leakage is precisely corrected to locate the specific spatial location of the weak point in sound insulation.

[0105] In one embodiment, the building sound insulation effect evaluation device further includes an equipment deployment planning module, which is used to determine the optimal deployment scheme of sensors and microphone arrays based on the simulation results of the initial building acoustic digital twin; and to deploy sensors and microphone arrays at key nodes of the building according to the optimal deployment scheme, so as to realize the collection of objective data and subjective quantitative data.

[0106] In one embodiment, the evaluation report generation module 50 is further configured to construct at least three differentiated sound insulation optimization schemes based on the location of weak sound insulation points; import each sound insulation optimization scheme into a calibrated building acoustic digital twin to simulate the application effect of each sound insulation optimization scheme in a real complex acoustic environment, and output the objective sound insulation index and abnormal frequency band improvement range corresponding to each sound insulation optimization scheme; predict the user noise annoyance reduction value, speech intelligibility improvement range, and soundscape satisfaction corresponding to each sound insulation optimization scheme based on subjective quantitative data and a preset correlation model; and generate a comprehensive evaluation report based on the user noise annoyance reduction value, speech intelligibility improvement range, soundscape satisfaction, objective sound insulation index, and abnormal frequency band improvement range.

[0107] In one embodiment, the digital twin calibration module 40 is further used to compare the real-time collected objective data with the initial simulation data of the initial building acoustic digital twin dimension by dimension to identify deviation items; based on the deviation items, the acoustic attribute parameters in the initial building acoustic digital twin are corrected by the least squares method, and the correction results are compensated and calibrated by combining environmental parameter data; the structural vibration data and the sound energy propagation path analysis results are fused to adjust the sound transmission attenuation coefficient of the structural components in the initial building acoustic digital twin to ensure that the sound transmission mechanism of the digital twin is consistent with the actual state of the building.

[0108] In one embodiment, the building sound insulation effect assessment device further includes a data update module, which is used to structure and archive data at each stage of the entire assessment process to form a building acoustic life cycle archive, providing a reusable reference sample for subsequent similar building sound insulation effect assessments; and continuously iterates and updates the building acoustic digital twin based on the long-term accumulated building acoustic life cycle archive.

[0109] In one embodiment, this application discloses a computer device including a memory and a processor. The memory stores a computer program that can run on the processor. When the processor loads the computer program, it executes a method for evaluating the sound insulation effect of a building as described in the above embodiment.

[0110] In one embodiment, this application discloses a computer-readable storage medium storing a computer program, wherein when the computer program is loaded by a processor, it executes a method for evaluating the sound insulation effect of a building as described in the above embodiment.

[0111] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method of evaluating sound insulation effect of a building, characterized by, The method comprises the following steps: Collecting design drawings of a building and acoustic property parameters of materials used in the building, building a three-dimensional geometric model based on the BIM technology according to the design drawings, and constructing an initial building acoustic digital twin based on the three-dimensional geometric model and the acoustic property parameters; Synchronously collecting objective data and subjective quantitative data in real time, wherein the objective data comprises acoustic signal data, structural vibration data and environmental parameter data, and the subjective quantitative data comprises noise annoyance quantitative scores, language intelligibility interference degree scores and scene annotation information; Comparing the real-time collected objective data with initial simulation data of the initial building acoustic digital twin and a preset standard spectrum library to locate sound insulation weak points; Feeding back the real-time collected data to the initial building acoustic digital twin to calibrate the initial building acoustic digital twin; Based on the located sound insulation weak points, constructing multiple sound insulation optimization schemes in the calibrated building acoustic digital twin and performing virtual effect testing to generate a comprehensive evaluation report.

2. A method of assessing the soundproofing effect of a building according to claim 1, wherein, The comparison of the real-time collected objective data with the initial simulation data of the initial building acoustic digital twin and the preset standard spectrum library to locate sound insulation weak points comprises: Comparing the real-time collected objective data with the initial simulation data of the initial building acoustic digital twin and the preset standard spectrum library in each frequency band to identify abnormal frequency bands with sound insulation values lower than simulation values and standard thresholds; Extracting time domain and frequency domain features of the abnormal frequency bands and matching corresponding typical defect features in the preset standard spectrum library; Based on the extracted abnormal frequency band features, identifying main noise source types through a support vector machine and a random forest fusion model; Locating sound insulation weak points according to the abnormal frequency bands, the matched typical defect features and the main noise source types.

3. A method of assessing the soundproofing effect of a building according to claim 2, wherein, The locating of sound insulation weak points according to the abnormal frequency bands, the matched typical defect features and the main noise source types comprises: Based on the abnormal frequency bands and the matched typical defect features, controlling a microphone array to scan corresponding building areas and synchronously collecting acoustic signals of the abnormal frequency bands and vibration data of building components in the corresponding areas; According to the collected acoustic signals and vibration data, calibrating signal phases through a beamforming technology, focusing acoustic energy propagation paths of the abnormal frequency bands, and combining time difference of arrival of signals collected by multiple node microphones to preliminarily lock spatial ranges of acoustic energy leakage; According to propagation characteristics of the main noise source types, accurately correcting the preliminarily locked spatial ranges of acoustic energy leakage to locate specific spatial positions of sound insulation weak points.

4. The method of claim 1, wherein, Before the synchronous collection of objective data and subjective quantitative data in real time, the method further comprises the following steps: Based on simulation results of the initial building acoustic digital twin, determining an optimal deployment scheme of sensors and a microphone array; According to the optimal deployment scheme, deploying sensors and a microphone array at key nodes of the building to realize collection of objective data and subjective quantitative data.

5. The method of claim 1, wherein, The construction of multiple sound insulation optimization schemes in the calibrated building acoustic digital twin based on the located sound insulation weak points and the virtual effect testing to generate a comprehensive evaluation report comprise: Based on the positioning of the sound insulation weak points, at least three differentiated sound insulation optimization schemes are constructed; Each sound insulation optimization scheme is introduced into the calibrated building acoustics digital twin to simulate the application effect of each sound insulation optimization scheme in the actual complex sound environment, and the corresponding objective sound insulation indicators and abnormal frequency band improvement amplitude of each sound insulation optimization scheme are output; According to the subjective quantitative data and the pre-set correlation model, the user noise annoyance reduction value, the language intelligibility improvement amplitude and the soundscape satisfaction corresponding to each sound insulation optimization scheme are predicted; According to the user noise annoyance reduction value, the language intelligibility improvement amplitude, the soundscape satisfaction, the objective sound insulation indicators and the abnormal frequency band improvement amplitude, a comprehensive evaluation report is generated.

6. The method of claim 1, wherein, The calibration of the initial building acoustics digital twin includes: The real-time collected objective data is compared with the initial simulation data of the initial building acoustics digital twin dimension by dimension to identify the deviation items; Based on the deviation items, the acoustic property parameters in the initial building acoustics digital twin are corrected by the least square method, and the correction result is compensated and calibrated in combination with the environmental parameter data; The structure vibration data and the sound energy propagation path analysis result are fused to adjust the sound transmission attenuation coefficient of the structural member in the initial building acoustics digital twin, so as to ensure that the sound transmission mechanism of the digital twin is consistent with the actual state of the building.

7. The method of claim 1, wherein, It also includes: Structural archiving of the data in each stage of the whole process is performed to form a building acoustics whole life cycle archive, which provides reusable reference samples for subsequent similar building sound insulation effect evaluation; Based on the long-term accumulated building acoustics whole life cycle archive, the building acoustics digital twin is continuously iteratively updated.

8. A building sound insulation effect evaluation device characterized by comprising: It includes: A digital twin construction module is used to collect the design drawings of the building and the acoustic property parameters of the materials used, to build a three-dimensional geometric model based on BIM technology according to the design drawings, and to construct an initial building acoustics digital twin according to the three-dimensional geometric model and the acoustic property parameters; A data acquisition module is used to synchronously collect objective data and subjective quantitative data in real time, wherein the objective data includes acoustic signal data, structure vibration data and environmental parameter data, and the subjective quantitative data includes noise annoyance quantitative score, language intelligibility interference degree score and scene annotation information; A weak point positioning module is used to compare the real-time collected objective data with the initial simulation data of the initial building acoustics digital twin and the pre-set standard spectrum library to locate the sound insulation weak points; A digital twin calibration module is used to feed back the real-time collected data to the initial building acoustics digital twin to calibrate the initial building acoustics digital twin; An evaluation report generation module is used to construct multiple sound insulation optimization schemes in the calibrated building acoustics digital twin based on the positioning of the sound insulation weak points and perform virtual effect testing to generate a comprehensive evaluation report. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The processor executes the computer program to realize the steps of the method of any one of claims 1 to 7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the method of any one of claims 1 to 7.

Images