Vibration noise digital twin modeling method and system for large complex

Abstract

The invention discloses a vibration noise digital twinborn modeling method and system for a large complex, relates to the technical field of computer data processing, and aims at a target complex to perform sensing array deployment and twinborn reconstruction, construct a three-dimensional twinborn body and determine a vibration noise source of the target complex. Discrete aggregation is carried out in a vibration mode, a vibration equation of each aggregation class is mined, an explicit scene migration framework is introduced, modeling explicit requirements of a vibration noise scene are determined, the vibration equations are assisted, vibration noise distribution based on acoustic medium boundaries is carried out on the three-dimensional twin body, and a vibration noise model is generated; the method and the device are used for solving the technical problem of insufficient accuracy and adaptability of dynamic modeling caused by insufficient refinement degree and flexibility of vibration noise modeling for complex building scenes in the prior art. The vibration noise fine modeling and dynamic visual presentation oriented to complex building scenes are realized, and the method has the remarkable advantages of high accuracy and high adaptability.

Description

Technical Field

[0001] The present invention relates to the technical field of computer data processing, and particularly relates to a method and system for digital twin modeling of vibration and noise of a large complex. Background Art

[0002] A large complex integrates multiple functions such as transportation, commerce, and residence. Its structure is complex and functional areas are dense, and the problems of vibration and noise are becoming increasingly prominent. Especially for integrated transportation hubs of the station-city integration type, due to the dense population and numerous equipment, the complex acoustic-vibration coupling phenomenon is likely to cause problems such as structural resonance and environmental nuisance, affecting the use safety and comfort.

[0003] Existing technologies mostly use local monitoring or simplified models for analysis, which are difficult to comprehensively reflect the actual acoustic-vibration propagation characteristics. Especially in a multi-source heterogeneous interference environment, it is difficult to effectively identify the vibration and noise sources and quantify their impacts. In addition, the lack of systematic modeling means for the acoustic-vibration characteristics in complex building structures limits their applications in acoustic diagnosis, structural optimization, and intelligent operation and maintenance, etc.

[0004] Therefore, there is an urgent need for a modeling method that can integrate sensing data, structural information, and physical laws to achieve refined and flexible modeling and evaluation of vibration and noise in large complexes. Summary of the Invention

[0005] The present application provides a method and system for digital twin modeling of vibration and noise of a large complex, which are used to solve the technical problems in the prior art that the refinement degree and flexibility of vibration and noise modeling for complex building scenarios are insufficient, resulting in insufficient accuracy and adaptability of dynamic modeling.

[0006] In view of the above problems, the present application provides a method and system for digital twin modeling of vibration and noise of a large complex.

[0007] In a first aspect, the present application provides a method for digital twin modeling of vibration and noise of a large complex. The method includes: for a target complex, deploying a sensing array and performing twin reconstruction to construct a three-dimensional twin body, wherein the sensing array is communicatively connected to a modeling platform; determining the vibration and noise sources of the target complex, discretely aggregating the vibration and noise sources in terms of vibration modes, and excavating the vibration equations of each aggregation class, wherein the vibration equation is the vibration and noise diffusion attenuation relationship based on the acoustic medium under the vibration characteristics of each aggregation class; introducing an explicit scene migration framework, determining the explicit modeling requirements of the vibration and noise scene, assisting the vibration equation, performing the vibration and noise distribution based on the acoustic medium boundary on the three-dimensional twin body, generating a vibration and noise model, and establishing a mapping between the vibration and noise model and the three-dimensional twin body, wherein the vibration and noise model is a reduced-order grey-box model based on the medium framework and vibration and noise elements.

[0008] In a second aspect, the present application provides a vibration and noise digital twin modeling system for a large complex, the system comprising: a first construction unit for deploying a sensing array and performing twin reconstruction for a target complex to construct a three-dimensional twin body, wherein the sensing array is communicatively connected to a modeling platform; an aggregation and mining unit for determining the vibration and noise sources of the target complex, discretely aggregating the vibration and noise sources in a vibration mode, and mining the vibration equations of each aggregation class, wherein the vibration equation is based on the vibration and noise diffusion attenuation relationship of an acoustic medium under the vibration characteristics of each aggregation class; a second construction unit for introducing an explicit scene migration framework, determining the explicit modeling requirements of the vibration and noise scene, assisting the vibration equation, performing the vibration and noise distribution based on the acoustic medium boundary on the three-dimensional twin body, generating a vibration and noise model, and establishing a mapping between the vibration and noise model and the three-dimensional twin body, wherein the vibration and noise model is a reduced-order grey-box model based on a medium framework and vibration and noise elements.

[0009] One or more technical solutions provided in the present application have at least the following technical effects or advantages: A vibration and noise digital twin modeling method for a large complex provided in an embodiment of the present application, for a target complex, deploys a sensing array and performs twin reconstruction to construct a three-dimensional twin body, determines the vibration and noise sources of the target complex, discretely aggregates the vibration and noise sources in a vibration mode, mines the vibration equations of each aggregation class, introduces an explicit scene migration framework, determines the explicit modeling requirements of the vibration and noise scene, assists the vibration equation, performs the vibration and noise distribution based on the acoustic medium boundary on the three-dimensional twin body, generates a vibration and noise model, and establishes a mapping between the vibration and noise model and the three-dimensional twin body, wherein the vibration and noise model is a reduced-order grey-box model based on a medium framework and vibration and noise elements, and is used to solve the technical problems in the prior art that the refinement degree and flexibility of vibration and noise modeling for complex building scenes are insufficient, resulting in insufficient accuracy and adaptability of dynamic modeling. It realizes the refined modeling and dynamic visualization presentation of vibration and noise for complex building scenes, and has the significant advantages of high accuracy and high adaptability. Description of the Drawings

[0010] Figure 1 It is a schematic flow diagram of a vibration and noise digital twin modeling method for a large complex provided in the present application.

[0011] Figure 2 It is a schematic flow diagram of mining the vibration equations of each aggregation class in a vibration and noise digital twin modeling method for a large complex provided in the present application.

[0012] Figure 3 It is a schematic structural diagram of a vibration and noise digital twin modeling system for a large complex provided in the present application.

[0013] Description of reference numerals: The first construction unit 11, the aggregation mining unit 12, and the second construction unit 13. Specific implementation mode

[0014] This application provides a vibration and noise digital twin modeling method and system for large complexes, aiming to solve the technical problems in the prior art that the refinement and flexibility of vibration and noise modeling for complex building scenarios are insufficient, resulting in insufficient accuracy and adaptability of dynamic modeling.

[0015] Example 1: As Figure 1 shown, this application provides a vibration and noise digital twin modeling method for large complexes, and the method includes: S1: For the target complex, deploy a sensing array and perform twin reconstruction to construct a three-dimensional twin body, where the sensing array is communicatively connected to the modeling platform.

[0016] In the embodiments of this application, the target complex refers to large buildings, transportation hubs, activity centers, etc. that have multiple functional areas, complex structural forms, and significant vibration and noise interference. To realize the vibration and noise modeling of this complex, it is first necessary to deploy a sensing array, that is, select typical positions to install multiple types of sensor units according to the building structure layout, functional zoning, and acoustically sensitive areas. The sensor units may include accelerometers, sound pressure meters, micro-vibration probes, etc., and their function is to obtain multi-dimensional data related to vibration and noise.

[0017] Preferably, the sensing array is deployed in a distributed networking manner and is interconnected with the backend modeling platform through wireless or wired communication protocols to ensure the continuity and real-time nature of data collection. Then, combining the infrastructure data of the target complex such as building structure data, design drawings, and sensor calibration information, perform physical twin reconstruction on the target twin body, that is, based on multi-source data fusion, use means such as BIM modeling tools, laser scanning, or image reconstruction to reproduce and generate a three-dimensional visualization model reflecting the structural characteristics of the target complex as the three-dimensional twin body. Among them, the three-dimensional twin body and the target complex are in a virtual-real mirror image, and it is consistent with the actual building in geometric shape and has the ability of structural identification, and can be used as a physical carrier for subsequent vibration and noise propagation modeling and situation evolution deduction.

[0018] In the whole process, the sensing array is communicatively connected to the modeling platform. Exemplarily, it can be ensured that the data collected by the sensor can be transmitted to the modeling platform in real time by deploying communication modules or IoT middleware, etc., to support model reconstruction and dynamic update, and then realize the closed-loop operation of the twin system.

[0019] S2: Determine the vibration noise sources of the target complex, discretely aggregate the vibration noise sources with vibration modes, and mine the vibration equations of each aggregate class, wherein the vibration equations are the vibration noise diffusion attenuation relationships based on acoustic media under the vibration characteristics of each aggregate class.

[0020] In the embodiment of the present application, on the basis of constructing a three-dimensional twin, the vibration and noise sources of the target complex are further determined to realize modeling and simulation of the vibration and noise propagation laws.

[0021] Specifically, the vibration noise source refers to physical equipment or functional areas that generate mechanical vibration or sound pressure disturbance during the operation of the complex. Examples include: transportation hubs such as railways, activity areas, large-scale power generation equipment, etc.

[0022] Subsequently, the vibration noise sources are discretely aggregated according to the vibration mode, and the vibration mode refers to the typical frequency structure, vibration directionality and amplitude characteristics of the vibration noise source in the operating state. The discrete aggregation refers to clustering based on the similarity of the vibration mode among all the identified noise sources, and classifying the noise sources with similar modal characteristics into the same aggregation class, thereby reducing the modeling complexity and improving the reusability of the model. For example, multiple frequency main peaks concentrated in 50Hz-70Hz, pipeline equipment with similar period and spatial distribution can be classified into the same aggregation class, and the vibration noise sources in the same aggregation class are still discretely distributed in space, that is, based on the assembly position distribution in the target complex.

[0023] After completing the aggregation classification, the vibration equation of each aggregation class is further mined. The vibration equation is a mathematical expression that describes the propagation and attenuation law of the acoustic vibration signal in the spatial medium under a specific vibration mode, reflecting the dynamic response characteristics under different medium conditions.

[0024] Specifically, the method of data statistical analysis and mining is adopted, that is, combining the operation data of typical noise sources contained in the aggregation class, analyzing its acoustic vibration propagation situation in various transmission media (such as concrete, air), and extracting its diffusion path, attenuation rate and boundary reflection characteristics, and establishing the corresponding model expression. The vibration equation is based on the propagation parameters of the acoustic medium, and constructs the relationship between the gradual attenuation of vibration noise in three-dimensional space with the change of distance and medium, which serves as the basis for subsequent vibration noise scene simulation and model evolution.

[0025] Further, the vibration noise sources are discretely aggregated in vibration mode, and step S2 of the present application includes: Set an encoding method; traverse the vibration noise source, identify the source vibration mode and perform same-mode clustering processing to determine N clustering classes; according to the encoding method, encode and identify the N clustering classes, wherein the vibration noise sources in the same clustering class are encoded in the same way.

[0026] In the embodiments of the present application, to achieve the structured management of vibration and noise sources and the unified processing of models, an encoding method is first set to identify and classify the vibration and noise source aggregation classes identified subsequently. The encoding method refers to the predefined classification rules. Exemplarily, an encoding structure is set in combination with dimensions such as vibration mode characteristics, spatial position, and noise intensity level. This encoding structure may include fields such as a prefix identifier for the aggregation type, an infix representing the vibration mode characteristics, and a suffix identifier for geographical location information. Preferably, a unified format such as VM-XX-YY is adopted, where VM represents the vibration mode, XX represents the mode number, and YY represents the area or equipment number, to achieve the uniqueness and resolvability of the encoding.

[0027] After setting the encoding method, the vibration and noise sources are traversed, that is, all the vibration and noise source sets identified in the early stage are analyzed and processed one by one. Specifically, by calling the historical sensing data of each vibration and noise source, frequency domain analysis, modal feature extraction, and operating behavior discrimination are performed on it to identify the vibration mode of each source. Exemplarily, the vibration mode can be characterized by feature indicators such as dominant frequency, amplitude distribution, harmonic structure, and excitation directivity.

[0028] Subsequently, based on the recognition results, clustering processing of the same mode is performed, that is, clustering analysis is performed using modal similarity, and noise sources with similar vibration mode characteristics are classified into the same category, and a total of N aggregation classes are determined. Each aggregation class represents a set of vibration and noise sources with similar vibration behaviors and similar physical response mechanisms, facilitating the subsequent commonality processing of models and equation fitting.

[0029] Furthermore, according to the encoding method, the above N aggregation classes are encoded and identified, that is, each aggregation class is assigned a unique code according to the established encoding rules to achieve unified indexing in the modeling system, database, and analysis process. For example, if the water pump equipment in multiple underground pump rooms is classified as the high-frequency vibration class, it can be uniformly identified as VM-01-PB, where PB represents the Pump Basement area, for quick retrieval and batch modeling. Through the above steps, the classification and coding management of vibration and noise sources can be efficiently achieved, supporting the subsequent excavation of vibration equations and propagation simulation based on aggregation classes.

[0030] Furthermore, as Figure 2 shown, to excavate the vibration equations of each aggregation class, step S2 of the present application includes: For the first aggregation class, determine the first vibration mode; using the first vibration mode as an index, call the same-class vibration records, perform clustering based on the acoustic medium, and determine M groups of vibration records, where each group of vibration records corresponds to an acoustic medium; traverse the M groups of vibration records, excavate the spatial diffusion attenuation trend under dynamic vibration and noise, perform linear transformation and add it to the first vibration equation.

[0031] In the embodiments of the present application, for the set of vibration and noise sources that have been clustered and encoded, vibration equation mining is performed. Specifically, for the first aggregation class, the first vibration mode is determined. The first aggregation class refers to a representative set of a certain type of vibration and noise sources obtained through the aforementioned same-mode clustering process; and the first vibration mode is the vibration behavior characteristic common or dominant among the noise sources in this aggregation class, which can be composed of parameters such as the main frequency component, harmonic structure, excitation direction, and time variation trend, and serves as the basic characteristic parameter set for the noise propagation behavior of this class.

[0032] Using the first vibration mode as an index, call the vibration records of the same type, that is, use the characteristic parameters corresponding to this mode as the retrieval conditions to screen the vibration records in the vibration database or historical operation data that match its characteristics in the vibration database or historical operation data. The vibration record refers to the vibration response data collected by the sensor array during the operation of the building and stored according to time and position, including time-domain signals, frequency-domain spectra, and spatial positioning information. The above retrieval operation can adopt mechanisms such as fuzzy matching or threshold determination to ensure that the data called has sufficient similarity and representativeness.

[0033] Furthermore, cluster the vibration records of the same type of the first vibration mode based on the acoustic medium, that is, classify and aggregate the above retrieved vibration records according to the type of acoustic medium in the structural environment where they are located. The acoustic medium refers to the physical space medium through which the vibration and noise propagate, such as reinforced concrete, glass curtain wall, air, etc. Its medium properties determine the propagation rate, attenuation degree, and reflection characteristics of the acoustic vibration signal. By identifying the spatial position information and structural labels carried in the vibration records, a medium property mapping relationship is constructed, and finally the vibration records of the same type are divided into M groups, each group corresponding to an acoustic medium, thereby providing a physical stratification basis for modeling the propagation law.

[0034] Furthermore, after completing the medium grouping, traverse the M groups of vibration records to mine the spatial diffusion attenuation trend under dynamic vibration and noise, that is, analyze the multiple vibration records included in each group, extract features such as signal amplitude change, frequency attenuation, and propagation delay on different propagation paths, and form a diffusion attenuation trend for a specific medium. This model depicts the energy loss and spatial influence range when the vibration and noise propagate from the source point in a specific medium. Preferably, by fitting the attenuation curve, establishing a distance-intensity function, etc., a trend expression with continuity and deducibility is formed.

[0035] Finally, integrate the trend relationships mined under each acoustic medium and add them to the first vibration equation. Through the above steps, the constructed first vibration equation can comprehensively reflect the propagation characteristics of a specific vibration mode in different media, laying a foundation for subsequent multi-source scenario modeling.

[0036] Similarly, perform the above - mentioned step - by - step analysis for each aggregation class until the Nth vibration equation corresponding to the Nth aggregation class is determined. Integrate and regularize the first vibration equation until the Nth vibration equation as the vibration equation.

[0037] S3: Introduce an explicit scenario migration framework, determine the explicit requirements for modeling the vibration noise scenario, assist the vibration equation, perform the vibration noise distribution based on the acoustic medium boundary on the three - dimensional twin, generate a vibration noise model, and establish a mapping between the vibration noise model and the three - dimensional twin. Among them, the vibration noise model is a reduced - order gray - box model based on the medium framework and vibration noise elements.

[0038] In the embodiments of the present application, to improve the adaptability and pertinence of vibration noise modeling, an explicit scenario migration framework is further introduced to achieve dynamic adaptation and modeling optimization of the noise propagation law in complex usage environments. The explicit scenario migration framework refers to introducing a driving mechanism in the modeling process to describe specific application scenarios, define modeling boundary conditions, and guide model selection and parameter configuration. This framework can convert the actual operating scenario into a modelable problem structure through manual setting, rule matching, or knowledge - graph reasoning, thereby enhancing the generalization ability and engineering adaptability of the model in diverse scenarios.

[0039] Driven by the framework, first determine the explicit requirements for modeling the vibration noise scenario. The explicit requirements for modeling refer to the clear definition of the specific modeling objectives and constraint conditions for the acoustic - vibration problem in the current scenario to be simulated. For example, whether it is a single - excitation - source scenario, a multi - source superposition scenario, a sound - sensitivity analysis in a specific functional area, or a propagation analysis under a certain specific medium path, etc. This requirement can be extracted from the running - state sensing data, user - interaction settings, or historical case rules and used as the input premise for subsequent model construction to guide the vibration equation call and spatial boundary setting.

[0040] Furthermore, on the basis of clear modeling requirements, further assist the vibration equation to perform the vibration noise distribution based on the acoustic medium boundary on the three - dimensional twin. Specifically: taking the three - dimensional twin as the modeling carrier, combining the vibration noise sources and their propagation paths involved in the scenario, constructing the propagation boundary conditions according to the preset acoustic medium parameters (including medium type, conduction coefficient, attenuation coefficient, reflection boundary, etc.), and accordingly performing partition calculation on the twin space to simulate the spatial distribution characteristics of the vibration noise diffusion in the complex. Through the above - mentioned calculation process, a vibration noise model is generated, which is used to describe the spatial distribution trend and attenuation characteristics of the vibration noise in the multi - medium structure under the target scenario.

[0041] Among them, the vibration and noise model is a reduced-order grey-box model based on the medium framework and vibration and noise elements, that is, under the three-dimensional twin framework, only the spatial situation of the vibration and noise elements is involved. Exemplarily, the vibration and noise elements may include source intensity, spectral characteristics, periodicity, medium propagation parameters, etc.

[0042] Finally, establish the mapping between the vibration and noise model and the three-dimensional twin, that is, associate and bind each physical element in the model with the twin spatial structure, component identification, and time state, so that the vibration and noise model can be visually displayed and dynamically evolved and simulated on the three-dimensional twin, realizing the intuitive analysis and prediction and evaluation of the noise situation in the target complex. This mapping relationship also provides an indexing basis for subsequent model invocation, scenario migration, and intelligent scheduling, and constitutes one of the core links in the overall digital twin modeling system.

[0043] Furthermore, the explicit modeling requirement is a single-source scenario or a multi-distributed source superposition scenario based on the vibration and noise source.

[0044] Furthermore, to generate the vibration and noise model, step S3 of this application includes: According to the vibration and noise scenario, determine the target vibration and noise source and the coverage range, where the coverage range is located in the target complex; traverse the target vibration and noise source, match and associate the vibration equation, and frame the three-dimensional twin according to the coverage range to determine the scenario twin based on the vibration and noise scenario; based on the scenario twin, construct the vibration and noise model.

[0045] In the embodiment of this application, the explicit modeling requirement is a single-source scenario or a multi-distributed source superposition scenario based on the vibration and noise source, which specifically means that during the vibration and noise modeling process, according to the operating state and usage scenario of the target complex, determine the type of vibration and noise source configuration that needs to be focused on for the current modeling.

[0046] Among them, the single-source scenario refers to a situation where there is only a single dominant vibration and noise source in a certain time period or specific area, and its propagation path and influence range are relatively concentrated. The modeling focus is on finely depicting the acoustic and vibration diffusion situation of this single source; the multi-distributed source superposition scenario means that there are multiple vibration and noise sources with different spatial positions and obvious intensity differences within the same modeling time period, and their propagation processes may interfere, overlap, and superimpose with each other. Modeling such scenarios requires considering the characteristics of acoustic field coupling and multi-path propagation, and the complexity is significantly increased.

[0047] Further, a vibration noise model is generated, including the following key steps: First, according to the vibration noise scenario, determine the target vibration noise source and the coverage range. The target vibration noise source refers to the acoustic-vibration interference source identified as the main modeling object in the current scenario, which can be screened through system scheduling information, sensing response intensity, or usage scenario strategies. The coverage range refers to the boundary of the area that may be affected in the building space starting from the target noise source according to its acoustic-vibration propagation characteristics. It is usually estimated based on a preset sound pressure threshold or energy attenuation ratio, and the position range in the three-dimensional space is clarified. This range should be limited within the building structure of the target complex.

[0048] Subsequently, traverse the target vibration noise source and match and associate it with the vibration equation. Specifically, process each vibration noise source involved in the current scenario one by one, identify its aggregated class code, and accordingly call the corresponding vibration equation established previously to achieve rapid adaptation of physical laws and parameter reuse. This step ensures that the modeling of each noise source is based on its true physical characteristics, improving the accuracy and interpretability of the simulation results.

[0049] At the same time, according to the above coverage range, frame the three-dimensional twin body to determine the scenario twin body based on the vibration noise scenario. That is, in the three-dimensional twin body, intercept the spatial sub-structure related to the current vibration noise propagation to form a scenario twin body for modeling and simulation within the defined range. This scenario twin body has spatial structure boundaries, medium properties, and component identifiers, which is highly consistent with the actual scenario and provides a computational carrier for model execution.

[0050] Finally, based on the scenario twin body, construct a vibration noise model. That is, with the framed spatial region, the matched vibration equation, and the identified vibration noise source as the core inputs, through physical propagation calculations and model parameter loading, generate a simulation model that can describe the dynamic response characteristics of vibration noise in the target scenario.

[0051] Further, based on the scenario twin body, construct a vibration noise model. Step S3 of this application includes: Receive the distributed sensing data of the sensing array; for the scenario twin body, determine the sensing data of each target vibration noise source according to the distributed sensing data, and combine the associated vibration equation to perform diffusion attenuation deduction to determine the vibration noise model.

[0052] In the embodiment of this application, to achieve dynamic modeling of the vibration noise propagation situation in the scenario twin body, first receive the distributed sensing data of the sensing array. The sensor collects samples in real time and uploads the sampling data to the modeling platform to form continuous distributed sensing data. This data has time synchronization, spatial coverage, and type diversity, and can be used to reflect the real-time state of vibration noise in different structural regions.

[0053] Further, after obtaining the distributed sensing data, for the scenario twin, the sensing data of each target vibration noise source is determined according to the distributed sensing data. Specifically, the scenario twin is a spatial substructure extracted from the three-dimensional twin for the current acoustic-vibration modeling task. On this basis, through the mapping relationship between the sensor deployment positions and the positions of the twin components, the sensor data falling on different covered positions of the scenario twin is screened out to complete the one-to-one correspondence between each target vibration noise source and the sensing data. This process helps to clarify the actual sensing response characteristics dominated by each noise source and provides data support for subsequent physical modeling.

[0054] Further, the diffusion attenuation deduction is carried out in combination with the associated vibration equation. This vibration equation usually appears in the form of a composite expression of a spatial diffusion term and a time attenuation term. By fusing this vibration equation with the identified source sensing data and setting boundary conditions (such as the initial sound pressure value), initial parameters (such as the main frequency, wave speed), and medium property parameters (such as density, elastic modulus, etc.), the propagation process of the acoustic-vibration signal in the twin structure can be deduced and calculated, and the amplitude attenuation and frequency change trend at each spatial position can be predicted.

[0055] Finally, based on the above deduction results, a vibration noise model is determined. The vibration noise model is a modeling entity that globally quantifies and visualizes the propagation situation of all main vibration noise sources in the current scenario within their respective spatial regions. It not only reflects the dynamic acoustic-vibration response state but also provides a basis for subsequent acoustic evaluation, risk diagnosis, and governance strategies.

[0056] Further, for the diffusion attenuation deduction, step S3 of this application includes: For the scenario twin, determine the acoustic medium architecture; according to the acoustic medium architecture, starting from the sensing data based on the target vibration noise source and taking the iteration of the diffusion equation under the change of the acoustic medium as a condition, carry out the deduction of the diffusion attenuation situation to determine the vibration noise model, where the vibration noise model is a reduced-order gray-box model that covers the space with quantization particles for measuring vibration noise under the twin architecture.

[0057] In the embodiment of this application, in the process of constructing a model for a specific vibration noise scenario, first, for the scenario twin, the acoustic medium architecture needs to be determined. The scenario twin refers to a structural subset extracted from the complete three-dimensional twin for this modeling task, with clear geometric boundaries, component information, and functional zoning. On this structure basis, the acoustic medium covering the space is identified, which refers to the type of medium that the vibration noise passes through during propagation, such as different physical materials, etc.

[0058] Among them, each type of medium has a significant impact on the propagation path and attenuation rate of acoustic vibration signals due to its different density, elastic modulus, damping characteristics, etc. Therefore, it is necessary to build a complete acoustic medium framework, that is, based on the spatial distribution of components, mark the medium type and boundary conditions corresponding to each spatial unit, so as to form a spatial continuum of medium properties and provide a physical constraint basis for the subsequent diffusion of vibration noise.

[0059] Furthermore, after determining the medium structure, the sensor data based on the target vibration noise source is used as the starting point, that is, the target vibration noise source identified in the early stage is selected, and the initial sensor response data associated with it is extracted as the propagation starting point. Exemplarily, the sensor data may include parameters such as the sound pressure level at the source point, the dominant frequency, and the excitation period, which constitute the input boundary conditions of the simulation calculation and are used to drive the propagation behavior of the acoustic vibration signal in the medium structure.

[0060] Subsequently, the diffusion attenuation trend is deduced based on the iteration of the diffusion equation under the change of the acoustic medium. Specifically, as the signal propagation path progresses, the type of acoustic medium will continue to change, that is, switch from one material to another. At this time, the propagation parameters need to be dynamically adjusted according to the conduction characteristics of each medium to achieve dynamic deduction of the iterative diffusion of acoustic vibration energy in the structure, ensuring the continuity and physical rationality of the propagation trajectory.

[0061] Finally, based on the deduction results, the vibration noise model is determined. Among them, the vibration noise model is a reduced-order gray box model that covers space by measuring the quantitative display distribution of vibration noise under the twin architecture. At the same time, by introducing order reduction, the propagation dimension and the number of variables are simplified, and the model operation efficiency is improved. The quantized particles in the model are a discrete representation of the propagation of acoustic energy in space. Each particle carries sound pressure intensity, frequency information and propagation path attributes, and can achieve high-precision spatial coverage modeling in a three-dimensional twin structure. Through this model, the propagation trend, attenuation law and impact range of vibration noise in the target complex can be accurately portrayed, providing key support for acoustic assessment and governance strategies.

[0062] Further, a mapping between the vibration noise model and the three-dimensional twin is established, and step S3 of the present application includes: Construct a storage database; map and identify the modeling explicit requirements based on the vibration and noise scene and the vibration and noise model by introducing space-time codes, and store them in the storage database; based on the retrieval information, perform retrieval in the storage database and overlay display based on the three-dimensional twin.

[0063] Furthermore, the space-time code includes a timestamp code and a spatial position code located at the target complex.

[0064] In the embodiments of the present application, to achieve unified management and efficient retrieval of vibration and noise models, a storage database is first constructed. The storage database refers to a database used to centrally store various vibration and noise models and their associated information.

[0065] Furthermore, by introducing spatio-temporal codes, the explicit requirements for modeling based on vibration and noise scenarios are mapped and identified with the vibration and noise models, and then stored in the storage database. Specifically, the explicit requirements for modeling refer to the model visualization requirements corresponding to vibration and noise scenarios, such as the noise propagation process in a certain area, comparing the sound field changes under the operation of different devices, etc.; while the vibration and noise models are the mathematical expressions of the sound and vibration propagation laws in this scenario.

[0066] Over time, there may be the construction of vibration and noise models based on different time nodes and different scenario requirements. To facilitate subsequent rapid searching, spatio-temporal codes are introduced as a mapping and indexing means to ensure that the model and its applicable scenario are in one-to-one correspondence in the database and can be quickly retrieved.

[0067] Furthermore, the spatio-temporal code includes a timestamp code and a spatial position code located at the target complex. Among them, the timestamp code is used to identify the time characteristics of model generation or application, supporting dynamic tracking and historical backtracking of noise states in different time periods; the spatial position code is a spatial positioning identifier established based on the geometric structure and functional partitions of the target complex, that is, the distribution of the scenario-based vibration and noise models, for example, expressed in the form of floor numbers, component numbers, area labels, etc., to clarify the building space range corresponding to the model. The combination of the two constitutes a unique identifier to achieve precise positioning and invocation of any model.

[0068] When there is a need for scenario-based vibration and noise modeling, a vibration and noise model is constructed based on the above method steps, identified based on the spatio-temporal code, and stored in the storage database.

[0069] Based on the above structure, when receiving an external retrieval request, retrieval and invocation can be performed in the storage database through the retrieval information, and coverage display based on the three-dimensional twin body can be carried out. The retrieval information can include parameters such as the target time period, query area, model type, etc. The system quickly locks the target model based on the spatio-temporal code, and automatically maps and loads the model into the three-dimensional twin body, realizing the coverage presentation and dynamic visualization performance of the vibration and noise model with sound and vibration representation in the virtual building environment, improving the model application efficiency and user interaction experience. Through the above mechanism, a full-process data closed-loop from model generation, storage management to scenario retrieval and coverage based on the three-dimensional twin body is established, providing strong information support for the digital twin modeling of vibration and noise in large complexes.

[0070] A method for digital twin modeling of vibration and noise in a large complex provided by the present application has the following technical effects: 1. Deploy a distributed sensing array inside the target complex, combine the building structure information with the sensor spatial position information, and reconstruct a highly consistent three-dimensional digital twin. Extract the modal characteristics of the vibration noise sources, use the modal similarity clustering algorithm to achieve multi-source classification and aggregation, and set up a coding system for identification management. Effectively simplify the management dimension of the vibration noise sources, improve the reusability of the model parameters and the flexibility of system scheduling. Based on the aggregated classes and the acoustic medium types, mine and construct the diffusion attenuation vibration equations under different propagation paths, and support dynamic parameter configuration. Accurately reflect the propagation behavior of the vibration noise in heterogeneous structural media, and improve the physical credibility of the simulation. For the pain points of vibration multi-source coupling in direct sensing modeling, it can be flexibly configured and constructed according to the vibration noise modeling requirements.

[0071] 2. Establish an explicit scenario migration logic for identifying modeling requirements, guide the vibration equation call and the local mapping modeling of the three-dimensional twin. Achieve the rapid adaptation of the vibration noise model in different scenarios, and improve the generality and response efficiency of model construction. Integrate the real-time sensing data with the pre-built vibration equations, and conduct dynamic deduction of the acoustic vibration propagation trend in the scenario twin. Realize the dynamic simulation of the acoustic vibration response under the current building operation state, and improve the real-time performance and accuracy of monitoring and diagnosis.

[0072] 3. Construct a diffusion iteration model based on the change of acoustic media, introduce a quantitative particle to represent the propagation state, construct a reduced-order grey box model, reduce the model dimension and calculation load, and at the same time retain the key physical characteristics to achieve efficient modeling and rapid response. Construct a spatio-temporal code system with time stamps and spatial positions as dimensions, uniformly manage the mapping relationship between the modeling requirements and the model entities, and store and retrieve them in the database. Support the accurate call and coverage display of the model in multiple scenarios, multiple time periods, and multiple regions, and improve the maintainability and intelligent level of the system.

[0073] In summary, the refined modeling and dynamic visualization of vibration noise for complex building scenarios are realized, with the significant advantages of high accuracy and high adaptability.

[0074] Embodiment 2: Based on the same inventive concept as the vibration noise digital twin modeling method for a large complex in the foregoing embodiment, as Figure 3 shown, the present application provides a vibration noise digital twin modeling system for a large complex, and the system includes: The first construction unit 11 is used to deploy a sensing array and perform twin reconstruction for the target complex to construct a three-dimensional twin, wherein the sensing array is communicatively connected to the modeling platform; The aggregation and mining unit 12 is used to determine the vibration and noise sources of the target complex, discretely aggregate the vibration and noise sources in a vibration mode, and mine the vibration equations of each aggregation class, where the vibration equation is based on the vibration and noise diffusion attenuation relationship of the acoustic medium under the vibration characteristics of each aggregation class; The second construction unit 13 is used to introduce an explicit scene migration framework, determine the explicit modeling requirements of the vibration and noise scene, assist the vibration equation, perform the vibration and noise distribution based on the acoustic medium boundary on the three-dimensional twin, generate a vibration and noise model, and establish a mapping between the vibration and noise model and the three-dimensional twin, where the vibration and noise model is a reduced-order gray box model based on the medium framework and vibration and noise elements.

[0075] Further, the aggregation and mining unit 12 is also used to perform the following steps: set the coding method; traverse the vibration and noise sources, identify the source vibration modes and perform clustering processing of the same mode to determine N aggregation classes; according to the coding method, perform coding identification on the N aggregation classes, where the vibration and noise sources within the same aggregation class have the same code.

[0076] Further, the aggregation and mining unit 12 is also used to perform the following steps: for the first aggregation class, determine the first vibration mode; use the first vibration mode as an index to call the vibration records of the same class, perform clustering based on the acoustic medium, and determine M groups of vibration records, where each group of vibration records corresponds to an acoustic medium; traverse the M groups of vibration records, mine the spatial diffusion attenuation trend under dynamic vibration and noise, perform linear transformation and add it to the first vibration equation.

[0077] Further, the explicit modeling requirement is a single-source scene or a multi-distribution source superposition scene based on the vibration and noise source.

[0078] Further, the second construction unit 13 is also used to perform the following steps: according to the vibration and noise scene, determine the target vibration and noise sources and the coverage range, where the coverage range is located in the target complex; traverse the target vibration and noise sources, match and associate the vibration equations, frame the three-dimensional twin according to the coverage range, and determine the scene twin based on the vibration and noise scene; based on the scene twin, construct a vibration and noise model.

[0079] Further, the second construction unit 13 is also used to perform the following steps: receive the distributed sensing data of the sensing array; for the scene twin, determine the sensing data of each target vibration and noise source according to the distributed sensing data, perform diffusion attenuation deduction in combination with the associated vibration equation, and determine the vibration and noise model.

[0080] Further, the second construction unit 13 is further configured to perform the following steps: for the scenario twin, determine the acoustic medium architecture; according to the acoustic medium architecture, starting from the sensing data based on the target vibration noise source and taking the diffusion equation iteration under the acoustic medium change as a condition, perform the diffusion attenuation trend deduction to determine the vibration noise model, where the vibration noise model is a reduced-order gray-box model that covers space with quantization particles for measuring vibration noise under the twin architecture.

[0081] Further, the second construction unit 13 is further configured to perform the following steps: construct a storage database; map and identify the explicit requirements for modeling based on the vibration noise scenario and the vibration noise model by introducing space-time codes, and store them in the storage database; based on the retrieval information, retrieve and display in the storage database for the coverage based on the three-dimensional twin.

[0082] Further, the space-time code includes a time stamp code and a spatial position code located at the target complex.

[0083] Through the foregoing detailed description of a method for digital twin modeling of vibration noise in a large complex in this specification, those skilled in the art can clearly know a method and system for digital twin modeling of vibration noise in a large complex in this embodiment. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0084] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to these embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A digital twin modeling method for the vibration and noise of a large complex, characterized in that, The method comprises: For the target complex, the sensor array is deployed and the twin is reconstructed to construct a three-dimensional twin, in which the sensor array is connected to the modeling platform through communication. Determine the vibration and noise sources of the target complex, discretely aggregate the vibration and noise sources by vibration mode, and mine the vibration equation of each aggregate class, wherein the vibration equation is the vibration and noise diffusion and attenuation relationship based on the acoustic medium under the vibration characteristics of each aggregate class; An explicit scene migration framework is introduced to determine the explicit modeling requirements of the vibration noise scene, to assist the vibration equation, to perform vibration noise distribution based on the acoustic medium boundary on the three-dimensional twin, to generate a vibration noise model, and to establish a mapping between the vibration noise model and the three-dimensional twin, wherein the vibration noise model is a reduced-order gray box model based on the medium framework and vibration noise elements.

2. The digital twin modeling method for vibration and noise of a large complex as described in claim 1, characterized in that Discretely aggregating the vibration noise sources in vibration modes, including: Set the encoding method; Traversing the vibration noise sources, identifying the source vibration modes and performing same-mode clustering processing to determine N clustering classes; According to the encoding method, the N clustering classes are coded and identified, wherein the vibration noise sources in the same clustering class are coded in the same way.

3. The digital twin modeling method for vibration and noise of a large complex according to claim 2, characterized in that Mining vibration equations for each aggregation class, including: For the first aggregate class, determining a first vibration mode; Using the first vibration mode as an index, calling similar vibration records, clustering based on the acoustic medium, and determining M groups of vibration records, wherein each group of vibration records corresponds to an acoustic medium; The M groups of vibration records are traversed to explore the spatial diffusion attenuation trend under dynamic vibration noise, perform linear transformation and add it into the first vibration equation.

4. A digital twin modeling method for vibration and noise of a large complex as claimed in claim 1, characterized in that, The modeling explicit requirement is based on a single source scenario or a multi-distributed source superposition scenario of a vibration noise source.

5. A digital twin modeling method for vibration and noise of a large complex as described in claim 4, characterized in that Generate vibration and noise models, including: Determining a target vibration noise source and a coverage range according to the vibration noise scenario, wherein the coverage range is located at the target complex; Traversing the target vibration noise source, matching and associating the vibration equation, framing the three-dimensional twin according to the coverage range, and determining a scene twin based on the vibration noise scene; A vibration noise model is constructed based on the scene twin.

6. A digital twin modeling method for vibration and noise of a large complex as described in claim 5, characterized in that Based on the scene twin, a vibration noise model is constructed, including: receiving distributed sensor data from a sensor array; For the scene twin, the sensor data of each target vibration noise source is determined according to the distributed sensor data, and the diffusion attenuation is deduced in combination with the associated vibration equation to determine the vibration noise model.

7. A digital twin modeling method for vibration and noise of a large complex as described in claim 6, characterized in that, Perform diffusion attenuation simulation, including: For the scene twin, determining an acoustic medium architecture; According to the acoustic medium architecture, starting with the sensor data based on the target vibration noise source and based on the iteration of the diffusion equation under the change of the acoustic medium, the diffusion attenuation trend is deduced to determine the vibration noise model, wherein the vibration noise model is a reduced-order gray box model that uses quantitative particles to measure vibration noise for spatial coverage under the twin architecture.

8. A digital twin modeling method for vibration and noise of a large complex as claimed in claim 1, characterized in that, Establishing a mapping between the vibration noise model and the three-dimensional twin, including: Build a storage database; By introducing space-time codes, the explicit requirements for modeling based on the vibration and noise scenario are mapped and identified with the vibration and noise model, and stored in the storage database; Based on the retrieved information, retrieve and display the coverage based on the three-dimensional twin in the storage database.

9. A digital twin modeling method for vibration and noise of a large complex as described in claim 8, characterized in that, The space-time code includes a time-stamp code and a spatial position code located at the target complex.

10. A digital twin modeling system for vibration and noise of a large complex, characterized in that, Execute a vibration and noise digital twin modeling method for a large complex according to any one of claims 1-9, the system comprising: A first construction unit for deploying a sensing array and performing twin reconstruction for a target complex to construct a three-dimensional twin, wherein the sensing array is communicatively connected to a modeling platform; An aggregation mining unit for determining the vibration and noise sources of the target complex, discretely aggregating the vibration and noise sources in a vibration mode, and mining the vibration equations of each aggregation class, wherein the vibration equation is based on the vibration and noise diffusion attenuation relationship of the acoustic medium under the vibration characteristics of each aggregation class; A second construction unit for introducing an explicit scenario migration framework, determining the explicit requirements for modeling the vibration and noise scenario, assisting the vibration equation, performing the vibration and noise distribution based on the acoustic medium boundary on the three-dimensional twin, generating a vibration and noise model, and establishing a mapping between the vibration and noise model and the three-dimensional twin, wherein the vibration and noise model is a reduced-order gray-box model based on the medium framework and vibration and noise elements.

Images