Urban road dynamic noise traceability evaluation method and system based on digital twinning

Abstract

The invention discloses an urban road dynamic noise traceability evaluation method and system based on digital twinning, and the method comprises the steps: cooperatively collecting noise and road vibration signals through a non-uniform sound pressure sensor array and a vibration sensor, and dynamically compensating sound wave attenuation in combination with environment parameters; voiceprint features are extracted through wavelet packet decomposition, and dynamic parameters such as digital twinborn associated traffic flow and architectural acoustic reflection are constructed; improving a time difference positioning algorithm and fusing vibration spectrum verification to suppress environmental reflection noise; in the digital twin model, sound field distribution is predicted through sound ray tracking, health risks are evaluated based on a frequency band weighting model, and a parameterized sound insulation screen design and traffic flow regulation and control strategy is generated. The system integrates a multi-source sensing module, a digital twin engine module and an anti-interference positioning and decision optimization module, and realizes full-flow intelligentization of noise signal acquisition, feature fusion modeling, anti-interference positioning and treatment scheme generation.

Description

Technical Field

[0001] The present invention relates to the field of road noise analysis and processing, and in particular to a method and system for tracing and evaluating the source of dynamic noise on urban roads based on digital twins. Background Art

[0002] Given resource constraints, urban road noise control often relies on basic equipment and simple methods. These include the use of uniformly deployed single-type acoustic sensors, which offer low hardware cost but limited coverage, particularly in complex terrain such as curves and slopes. Data processing relies on static models, failing to account for dynamic environmental influences such as temperature, humidity, and wind speed. This results in significant deviations in sound velocity calculations, with positioning errors often exceeding two meters. Furthermore, the lack of vibration signal verification mechanisms frequently leads to the misidentification of ambient reflected noise as a valid source. Noise reduction solutions are often based on manual experience and lack scientific, quantitative evaluation, resulting in inefficient and costly control.

[0003] However, urban road noise analysis and control suffer from the following core flaws: low positioning accuracy: Traditional TDOA algorithms fail to dynamically compensate for environmental parameters (temperature, humidity, and wind speed), resulting in large positioning fluctuations due to errors in sound velocity calculation; poor anti-interference capabilities: Single acoustic signals are susceptible to multipath interference and lack cross-validation of vibration spectra, leading to a high rate of misjudgment; rigid modeling: Static models are unable to update dynamic data such as traffic flow and building materials in real time, resulting in severe simulation distortion; and crude decision-making: Sound barrier design and traffic flow control lack parametric modeling and cost-benefit quantification, relying on manual experience and poor implementability. These issues hinder the accuracy and efficiency of noise control, necessitating urgent systematic innovation. Summary of the Invention

[0004] The purpose of the present invention is to overcome one or more deficiencies of the prior art and to provide a method and system for tracing and evaluating the dynamic noise sources of urban roads based on digital twins.

[0005] The object of the present invention is achieved through the following technical solutions:

[0006] A method for tracing and evaluating the dynamic noise sources of urban roads based on digital twins includes the following steps:

[0007] Step S1. Multi-source data collaborative acquisition and dynamic compensation:

[0008] (1) The noise signal is collected through an array of sound pressure sensors deployed in a non-uniform topology, and the road vibration spectrum is obtained simultaneously through a vibration sensor;

[0009] (2) Collect environmental parameters and synchronize sensor clocks through the timing module;

[0010] (3) Dynamically adjust the sound wave attenuation compensation coefficient according to the ambient humidity;

[0011] Step S2. Noise-vibration feature fusion modeling:

[0012] (4) Perform multi-scale wavelet packet decomposition on the noise signal and extract the energy-dominant frequency band as the voiceprint feature;

[0013] (5) Constructing a digital twin of the road, correlating traffic flow and building acoustic reflection parameters;

[0014] Step S3. Anti-interference noise source location:

[0015] (6) The improved TDOA algorithm is used to calculate the preliminary coordinates of the noise source, and the environmental impact is compensated by the sound speed correction formula:

[0016] ;

[0017] in, is the linear correction term of the document to the appeal, is the nonlinear correction term of humidity to sound velocity, is the additional term of wind speed to sound speed propagation;

[0018] (7) Perform spectrum matching verification on the vibration signal and select the frequency band that is consistent with the main frequency of the noise;

[0019] Step S4. Dynamic simulation and decision optimization:

[0020] (8) Load the noise source coordinates into the digital twin model and use the ray tracing method to predict the sound field distribution in the sensitive area;

[0021] (9) Calculation of health risk index based on frequency band weighted exposure model:

[0022] ;

[0023] in, is the equivalent sound level at time period i, is the benchmark safety sound level, is the noise exposure duration in period i (hours), is the frequency band weighting coefficient;

[0024] (10) Generate noise reduction solution simulation report, including parameterized sound insulation screen design and traffic flow control strategy.

[0025] Furthermore, in step S1:

[0026] The deployment density of the sound pressure sensor array in the curved area is higher than that in the straight section;

[0027] The vibration sensor is embedded in the roadbed in an asymmetric layout, forming spatial coordination with the sound pressure sensor.

[0028] Furthermore, in step S3(6), improving the TDOA algorithm includes:

[0029] Perform time series smoothing on the positioning results through Kalman filter;

[0030] When the vibration spectrum matching fails, the beamforming algorithm is triggered to suppress the environmental reflection noise.

[0031] Furthermore, in step S4(8), the ray tracing method includes:

[0032] Dynamically load the reflection coefficient according to the building surface material;

[0033] The attenuation value of the diffracted sound wave is calculated according to the following formula:

[0034] ;

[0035] in, is the wave number, is the diffraction coefficient ( ), 、 is the distance from the sound source to the obstacle and from the obstacle to the receiving point (m).

[0036] Furthermore, the parameterized sound barrier design in step S4(10) includes:

[0037] Enter the height and sound absorption coefficient range to calculate the insertion loss:

[0038] ;

[0039] in, is the insertion loss, is the sound power when no sound insulation screen is set, is the sound power after the sound insulation screen is installed;

[0040] Calculate the scoring function:

[0041] ;

[0042] in, 、 is the weight coefficient, For implementation costs, is the noise reduction amount; the recommended solution is output according to the calculation result of the scoring function.

[0043] A digital twin-based urban road dynamic noise source tracing and evaluation system includes:

[0044] Multi-source sensing unit: consists of a non-uniform topology sound pressure array, a vibration sensor, and an environmental sensor;

[0045] Digital twin engine: used to perform the modeling and dynamic update of step S2 of claim 1;

[0046] Anti-interference positioning unit: integrates improved TDOA algorithm and vibration spectrum matching verification logic;

[0047] Decision optimization unit: used to generate a simulation report of step S4 described in claim 1.

[0048] Furthermore, in the multi-source sensing unit:

[0049] The surface of the sound pressure sensor is coated with a hydrophobic and anti-noise coating; the vibration sensor adopts a three-axis MEMS structure and is deployed in the mechanical coupling area between the roadbed and the road surface.

[0050] Furthermore, the digital twin engine includes:

[0051] A sound field dynamic injection interface for real-time synchronization of sensor data; a diffraction attenuation calculation module for executing the sound ray tracing method described in claim 4.

[0052] Furthermore, the anti-interference positioning unit includes:

[0053] An environmental compensation calculator, used to run the sound speed correction formula described in claim 1; an abnormality verification module, which triggers the drone movement review when the positioning deviation exceeds the threshold.

[0054] Furthermore, the decision optimization unit includes: a parameterized sound insulation screen simulator for performing the insertion loss calculation described in claim 5; a traffic flow control simulator for predicting the noise-congestion rate correlation curve and outputting an optimization strategy.

[0055] The beneficial effects of the present invention are:

[0056] (1) Through the deployment of non-uniform sensor arrays, dynamic compensation of environmental parameters, and verification of vibration signal spectrum, the accuracy of noise source positioning in complex urban road scenarios (such as curves and viaducts) is significantly improved, effectively solving the positioning deviation problem caused by environmental interference in traditional methods, and achieving high-precision noise source positioning;

[0057] (2) Design of parametric noise reduction schemes based on digital twin models (such as optimization of sound insulation screen height and sound absorption coefficient, and traffic flow control strategies), combined with cost-benefit quantitative evaluation, to achieve scientific allocation of noise reduction resources, reduce project implementation costs and resource waste, and contribute to low-cost and efficient governance;

[0058] (3) Through real-time sound field simulation, dynamic health risk assessment and automatic comparison of multiple options, the noise control decision-making cycle is greatly shortened, and the focus shifts from relying on manual experience to data-driven precision management, thereby improving the scientific nature and response speed of urban noise management and providing intelligent decision-making support. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 A flowchart of the steps of the urban road dynamic noise source tracing and evaluation method based on digital twins provided in the embodiment;

[0060] Figure 2 This is the structural diagram of the urban road dynamic noise source tracing and evaluation system based on digital twin. DETAILED DESCRIPTION

[0061] The following will clearly and completely describe the technical solutions of the present invention in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work shall fall within the scope of protection of the present invention.

[0062] See Figure 1 , provides a method for tracing and evaluating the dynamic noise sources of urban roads based on digital twins, which includes the following steps:

[0063] Step S1. Multi-source data collaborative acquisition and dynamic compensation:

[0064] (1) The noise signal is collected through an array of sound pressure sensors deployed in a non-uniform topology, and the road vibration spectrum is obtained simultaneously through a vibration sensor;

[0065] (2) Collect environmental parameters and synchronize sensor clocks through the timing module;

[0066] (3) Dynamically adjust the sound wave attenuation compensation coefficient according to the ambient humidity;

[0067] Step S2. Noise-vibration feature fusion modeling:

[0068] (4) Perform multi-scale wavelet packet decomposition on the noise signal and extract the energy-dominant frequency band as the voiceprint feature;

[0069] (5) Constructing a digital twin of the road, correlating traffic flow and building acoustic reflection parameters;

[0070] Step S3. Anti-interference noise source location:

[0071] (6) The improved TDOA algorithm is used to calculate the preliminary coordinates of the noise source, and the environmental impact is compensated by the sound speed correction formula:

[0072] ;

[0073] in, is the linear correction term of the document to the appeal, is the nonlinear correction term of humidity to sound velocity, is the additional term of wind speed to sound speed propagation;

[0074] (7) Perform spectrum matching verification on the vibration signal and select the frequency band that is consistent with the main frequency of the noise;

[0075] Step S4. Dynamic simulation and decision optimization:

[0076] (8) Load the noise source coordinates into the digital twin model and use the ray tracing method to predict the sound field distribution in the sensitive area;

[0077] (9) Calculation of health risk index based on frequency band weighted exposure model:

[0078] ;

[0079] in, is the equivalent sound level at time period i, is the benchmark safety sound level, is the noise exposure duration in period i (hours), is the frequency band weighting coefficient;

[0080] (10) Generate noise reduction solution simulation report, including parameterized sound insulation screen design and traffic flow control strategy.

[0081] In step S1:

[0082] The deployment density of the sound pressure sensor array in the curved area is higher than that in the straight section;

[0083] The vibration sensor is embedded in the roadbed in an asymmetric layout, forming spatial coordination with the sound pressure sensor.

[0084] Improving the TDOA algorithm in step S3(6) includes:

[0085] Perform time series smoothing on the positioning results through Kalman filter;

[0086] When the vibration spectrum matching fails, the beamforming algorithm is triggered to suppress the environmental reflection noise.

[0087] In step S4(8), the ray tracing method includes:

[0088] Dynamically load the reflection coefficient according to the building surface material;

[0089] The attenuation value of the diffracted sound wave is calculated according to the following formula:

[0090] ;

[0091] in, is the wave number, is the diffraction coefficient ( ), 、 is the distance from the sound source to the obstacle and from the obstacle to the receiving point (m).

[0092] The parameterized sound barrier design in step S4 (10) includes:

[0093] Enter the height and sound absorption coefficient range to calculate the insertion loss:

[0094] ;

[0095] in, is the insertion loss, is the sound power when no sound insulation screen is set, is the sound power after the sound insulation screen is installed;

[0096] Calculate the scoring function:

[0097] ;

[0098] in, 、 is the weight coefficient, For implementation costs, is the noise reduction amount; the recommended solution is output according to the calculation result of the scoring function.

[0099] See Figure 2 , an urban road dynamic noise source tracing evaluation system based on digital twins is constructed based on the urban road dynamic noise source tracing evaluation method based on digital twins. The system includes:

[0100] Multi-source sensing unit: consists of a non-uniform topology sound pressure array, a vibration sensor, and an environmental sensor;

[0101] Digital twin engine: used to perform the modeling and dynamic update of step S2 of claim 1;

[0102] Anti-interference positioning unit: integrates improved TDOA algorithm and vibration spectrum matching verification logic;

[0103] Decision optimization unit: used to generate a simulation report of step S4 described in claim 1.

[0104] In the multi-source sensing unit:

[0105] The surface of the sound pressure sensor is coated with a hydrophobic and anti-noise coating; the vibration sensor adopts a three-axis MEMS structure and is deployed in the mechanical coupling area between the roadbed and the road surface.

[0106] The digital twin engine includes:

[0107] A sound field dynamic injection interface for real-time synchronization of sensor data; a diffraction attenuation calculation module for executing the sound ray tracing method described in claim 4.

[0108] The anti-interference positioning unit includes:

[0109] An environmental compensation calculator, used to run the sound speed correction formula described in claim 1; an abnormality verification module, which triggers the drone movement review when the positioning deviation exceeds the threshold.

[0110] The decision optimization unit includes: a parameterized sound insulation screen simulator for performing the insertion loss calculation described in claim 5; a traffic flow control simulator for predicting the noise-congestion rate correlation curve and outputting an optimization strategy.

[0111] Example 1

[0112] Parameter Table

[0113] parameter Meaning Data Source 331.4m / s Speed ​​of sound in standard dry air (0°C) Sound speed reference value in international standard ISO 9613-1 0.6T Linear correction term for temperature to sound velocity The relationship between the speed of sound and temperature is: 331.4 + 0.6, v = 331.4 + 0.6T (T is Celsius) 0.0124h Nonlinear correction term of humidity to sound velocity Based on experimental data fitting (humidity h is in percentage), chapter 2.3 of the Handbook of Acoustic Environment <![CDATA[0.17v wi nd ]]> The additional term of wind speed on the propagation of sound speed <![CDATA[Hydrodynamic model of acoustic wave propagation in a wind field (v wind is the wind speed, unit: m / s)]]> k Wave number (k=2πf / c) f is the noise frequency, c is the speed of sound, derived from the basic definition of wave theory D Diffraction coefficient Diffraction field calculation model based on UTD (Uniform Geometric Theory of Diffraction), Chapter 5 of the document "Computational Acoustics" <![CDATA[d1、d2]]> Distance from sound source to obstacle, and from obstacle to receiving point Geometric acoustic path calculation, obtained by interpolating 3D coordinates in the digital twin model <![CDATA[W without ]]> Sound power without sound insulation screen Obtained through actual measurement of sound pressure sensor array or simulation calculation of digital twin model <![CDATA[W with ]]> Sound power after installing the sound insulation screen Simulation results of acoustic performance of soundproof screens based on the boundary element method (BEM) <![CDATA[L i ]]> Equivalent sound level at time period i (dB) Obtained through real-time measurement of sound pressure sensor or prediction of digital twin model <![CDATA[L base ]]> Benchmark safety sound level (e.g. 70dB) Refer to the WHO's "Community Noise Guidelines" for nighttime noise limits in residential areas. <![CDATA[T i ]]> Noise exposure duration in period i (hours) Duration of noise events recorded by the system <![CDATA[W i ]]> Frequency band weighting coefficient Based on A-weighting curve and low-frequency enhancement rules (for noise <200Hz)

[0114] A city's main road is 5 kilometers long, with eight lanes in both directions and an average daily traffic volume of 150,000 vehicles. It passes through commercial areas, residential areas, and elevated bridges, and has diverse noise sources (vehicle engines, tire friction, and building reflections). Traditional monitoring systems have the following problems:

[0115] Uniform sensor deployment leads to blind spots in curved and elevated areas, with large positioning errors of 3 meters or more. Temperature and humidity fluctuations (humidity 60% to 90%, wind speed 2 to 8 m / s) lead to deviations in sound speed calculations, and environmental interference is severe. Noise reduction solutions rely on manual experience and lack quantitative evaluation, resulting in high implementation costs, unstable results, and delayed decision-making.

[0116] Step S1: Multi-source data collaborative acquisition and dynamic compensation:

[0117] The sensor topology design and deployment are as follows:

[0118] Sound pressure sensor array: Straight road sections: main node spacing is 5m, auxiliary node spacing is 2m, double height layers (0.5m, 1.5m), and a total of 200 nodes are deployed;

[0119] Curves and viaducts: Nodes are densely packed to 1m, with a waterproof and shockproof design (IP67 rating), and a total of 80 nodes are deployed.

[0120] The Beidou timing module ensures that the time stamp synchronization error of each node is ≤0.2ms and synchronizes the timing.

[0121] Three-axis MEMS vibration sensors are embedded in the roadbed (50 in total), with a sampling frequency of 15kHz, and are spatially coordinated with the sound pressure sensors (spacing ≤ 2m).

[0122] Deploy 10 high-precision thermometers and hygrometers (±1%RH) and ultrasonic anemometers (±0.3m / s) environmental sensors to collect data in real time.

[0123] Data preprocessing and dynamic compensation:

[0124] Environmental parameter collection: temperature T = 32 ° C, humidity h = 85%, wind speed v_wind = 6 m / s; sound speed correction calculation:

[0125] ;

[0126] In comparison and analysis, the sound speed is 343m / s when not compensated, resulting in an increase of about 2.8 meters in positioning error (the traditional method has an error of 3.5 meters).

[0127] When humidity is >80%, the attenuation coefficient of 0.15dB / m is enabled to reduce signal distortion and perform attenuation compensation.

[0128] Step S2: Noise-vibration feature fusion modeling:

[0129] Noise signal processing:

[0130] Through multi-scale wavelet packet decomposition, the noise signal is decomposed into 6 layers to extract the frequency band with the top 15% of energy (main frequency range 200-600Hz);

[0131] Through the voiceprint feature library, vehicle engine noise: 200-300Hz (peak 250Hz); tire friction noise: 300-600Hz (peak 450Hz); building reflection noise: high frequency band (>800Hz).

[0132] The candidate noise source types were screened by feature matching correlation coefficient (≥0.7).

[0133] Based on BIM data, the road curvature, viaduct height (25m), and building outline are loaded for three-dimensional geometric modeling and digital twin model construction.

[0134] Real-time traffic flow data (vehicle density 1.2 vehicles / second); material acoustic parameters: asphalt pavement sound absorption coefficient α = 0.3, glass curtain wall reflection coefficient β = 0.4, concrete guardrail β = 0.7; noise field initialization: inject an initial sound pressure level distribution map (baseline noise 65dB) and associate relevant dynamic parameters.

[0135] Step S3: Anti-interference noise source location and verification:

[0136] Using the TDOA algorithm and time difference calculation, the arrival time difference of the signals between the sound pressure array nodes is Δt=0.0015s (example value);

[0137] Direction angle calculation:

[0138] ;

[0139] Combining multi-node data, the coordinates of the noise source are calculated as (X=102.3m, Y=45.7m) and coordinate positioning is performed.

[0140] FFT analysis: Extract the vibration signal frequency band 100-500Hz, calculate the power spectral density (PSD) peak frequency f_v=248Hz, and verify it through the vibration signal spectrum.

[0141] Matching judgment: A deviation of Δf=2Hz (≤5Hz) from the noise main frequency of 250Hz is determined to be a valid noise source; Anti-interference processing: Kalman filtering smoothes the positioning trajectory and suppresses instantaneous jumps (coordinate fluctuation ≤0.5m); when matching fails, beamforming is used to suppress signals in non-target directions (improving the signal-to-noise ratio by 12dB).

[0142] Step S4: Dynamic simulation and noise reduction decision optimization:

[0143] Sound field propagation simulation

[0144] Ray tracing method:

[0145] Load the diffraction attenuation formula:

[0146] ;

[0147] Among them, the wave number , diffraction coefficient D = 0.5 (cylindrical obstacle);

[0148] Predict equivalent sound levels in sensitive areas (residential areas) (Exceeds the standard by 5dB).

[0149] Health Risk Assessment:

[0150] ;

[0151] Threshold setting: Dose>200 triggers an early warning and requires priority treatment.

[0152] Noise reduction scheme generation and scoring:

[0153] Parametric design of sound insulation screen: Input parameters: height H=3m, sound absorption coefficient α=0.6;

[0154] Insertion loss calculation:

[0155] ;

[0156] The cost estimate includes material costs of 100,000 yuan, construction costs of 50,000 yuan, and a total cost of 150,000 yuan.

[0157] If the truck restriction ratio η = 25%, the noise reduction ΔL = 4.0dB and the congestion rate increase ΔC = 8% are predicted;

[0158] Scoring function calculation:

[0159] ;

[0160] In the comparison scheme, H=4m (Score=3.52), and the H=3m sound insulation screen is recommended.

[0161] System execution process:

[0162] Sound pressure, vibration, and environmental data are transmitted to the cloud in real time through a multi-source sensing unit. The digital twin engine updates the noise field distribution every 5 minutes and dynamically adjusts the reflection coefficient. The anti-interference positioning unit triggers a drone review when an abnormal deviation exceeds 2 meters (coordinate verification error ≤ 0.3m). The decision optimization unit automatically generates a PDF report and pushes it to the management platform.

[0163] The comparison of effects and data analysis is shown in Table 1;

[0164] Table 1

[0165]

[0166] Sensor anti-interference design includes:

[0167] Hydrophobic coating: When humidity is >80%, the contact angle of the sensor surface is >150°, preventing water film from affecting sound wave reception; wind noise filtering: An embedded FIR filter (cutoff frequency 100Hz) suppresses low-frequency noise when wind speed is >5m / s.

[0168] Dynamic update of the digital twin model: Data injection frequency: synchronize traffic flow and weather data every 30 seconds; self-learning mechanism: when the diffraction attenuation prediction error is greater than 10%, the diffraction coefficient D is automatically adjusted (iteration step size 0.05).

[0169] Scoring function weight optimization:

[0170] (Here is the noise reduction effect weight) Every increase of 0.1, the Score increases by 0.8-1.2;

[0171] (Here is the cost weight) Too high will lead to the selection of inefficient and cheap solutions;

[0172] Different weight combinations are set according to the dynamic weight adjustment governance stage (emergency period / normal period).

[0173] Through the deployment of non-uniform sensor arrays, dynamic compensation of environmental parameters, and verification of vibration signal spectra, the accuracy of noise source positioning in complex urban road scenarios (such as curves and viaducts) is significantly improved, effectively solving the positioning deviation problem caused by environmental interference in traditional methods; based on the design of parameterized noise reduction solutions of digital twin models (such as optimization of sound insulation screen height and sound absorption coefficient, and traffic flow control strategy), combined with cost-benefit quantitative evaluation, scientific allocation of noise reduction resources is achieved, reducing project implementation costs and resource waste; through real-time sound field simulation, dynamic health risk assessment, and automatic comparison of multiple solutions, the noise control decision-making cycle is greatly shortened, shifting from reliance on manual experience to data-driven precision control, and improving the scientific nature and responsiveness of urban noise management.

[0174] This example fully demonstrates the entire process from multi-source data acquisition, dynamic environmental compensation, noise source location, to intelligent decision-making. Through detailed formula calculations and data analysis, it verifies the practical effects of each technical feature in the claims. Compared to traditional methods, this invention improves positioning accuracy and reduces noise reduction costs in complex urban road scenarios. It also provides scientific and quantitative decision-making support, setting a new technical benchmark for smart city noise management.

[0175] The foregoing description is merely a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein and should not be construed as excluding other embodiments. Rather, the present invention can be used in various other combinations, modifications, and environments and can be modified within the scope of the concept described herein through the above teachings or techniques or knowledge in the relevant field. Modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention are intended to be protected by the appended claims.

Claims

1. A method for tracing and evaluating the source of dynamic noise on urban roads based on digital twins, characterized in that: The following steps are involved: Step S1. Multi-source data collaborative acquisition and dynamic compensation: (1) The noise signal is collected through an array of sound pressure sensors deployed in a non-uniform topology, and the road vibration spectrum is obtained simultaneously through a vibration sensor; (2) Collect environmental parameters and synchronize sensor clocks through the timing module; (3) Dynamically adjust the sound wave attenuation compensation coefficient according to the ambient humidity; Step S2. Noise-vibration feature fusion modeling: (4) Perform multi-scale wavelet packet decomposition on the noise signal and extract the energy-dominant frequency band as the voiceprint feature; (5) Constructing a digital twin of the road, correlating traffic flow and building acoustic reflection parameters; Step S3. Anti-interference noise source location: (6) The improved TDOA algorithm is used to calculate the preliminary coordinates of the noise source, and the environmental impact is compensated by the sound speed correction formula: ； in, is the linear correction term of the document to the appeal, is the nonlinear correction term of humidity to sound velocity, is the additional term of wind speed to sound speed propagation; (7) Perform spectrum matching verification on the vibration signal and select the frequency band that is consistent with the main frequency of the noise; Step S4. Dynamic simulation and decision optimization: (8) Load the noise source coordinates into the digital twin model and use the ray tracing method to predict the sound field distribution in the sensitive area; (9) Calculation of health risk index based on frequency band weighted exposure model: ； in, is the equivalent sound level at time period i, is the benchmark safety sound level, is the noise exposure duration in period i, is the frequency band weighting coefficient; (10) Generate noise reduction solution simulation report, including parameterized sound insulation screen design and traffic flow control strategy.

2. The urban road dynamic noise source tracing and evaluation method based on digital twin according to claim 1 is characterized in that: In the step S1: The deployment density of the sound pressure sensor array in the curved area is higher than that in the straight section; The vibration sensor is embedded in the roadbed in an asymmetric layout, forming spatial coordination with the sound pressure sensor.

3. The urban road dynamic noise source tracing and evaluation method based on digital twin according to claim 1 is characterized in that: The improved TDOA algorithm in step S3(6) includes: Perform time series smoothing on the positioning results through Kalman filter; When the vibration spectrum matching fails, the beamforming algorithm is triggered to suppress the environmental reflection noise.

4. The urban road dynamic noise source tracing and evaluation method based on digital twin according to claim 1 is characterized in that: The ray tracing method in step S4(8) includes: Dynamically load the reflection coefficient according to the building surface material; The attenuation value of the diffracted sound wave is calculated according to the following formula: ； in, is the wave number, is the diffraction coefficient, 、 The distance from the sound source to the obstacle and from the obstacle to the receiving point.

5. The urban road dynamic noise source tracing and evaluation method based on digital twin according to claim 1 is characterized in that: The parameterized sound insulation screen design in step S4 (10) includes: Enter the height and sound absorption coefficient range to calculate the insertion loss: ； in, is the insertion loss, is the sound power when no sound insulation screen is set, is the sound power after the sound insulation screen is installed; Calculate the scoring function: ； in, 、 is the weight coefficient, For implementation costs, is the noise reduction amount; the recommended solution is output according to the calculation result of the scoring function.

6. A digital twin-based urban road dynamic noise source tracing and evaluation system, characterized by: include: Multi-source sensing unit: consists of a non-uniform topology sound pressure array, a vibration sensor, and an environmental sensor; Digital twin engine: used to perform the modeling and dynamic update of step S2 of claim 1; Anti-interference positioning unit: integrates improved TDOA algorithm and vibration spectrum matching verification logic; Decision optimization unit: used to generate a simulation report of step S4 described in claim 1.

7. The urban road dynamic noise source tracing and evaluation system based on digital twin according to claim 6 is characterized in that: In the multi-source sensing unit: the surface of the sound pressure sensor is coated with a hydrophobic and noise-resistant coating; the vibration sensor adopts a three-axis MEMS structure and is deployed in the mechanical coupling area between the roadbed and the road surface.

8. The urban road dynamic noise source tracing and evaluation system based on digital twin according to claim 6 is characterized in that: The digital twin engine includes: a sound field dynamic injection interface for real-time synchronization of sensor data; and a diffraction attenuation calculation module for executing the sound ray tracing method described in claim 4.

9. The urban road dynamic noise source tracing and evaluation system based on digital twin according to claim 6 is characterized in that: The anti-interference positioning unit includes: an environmental compensation calculator for running the sound speed correction formula described in claim 1; and an abnormality verification module for triggering a drone movement review when the positioning deviation exceeds a threshold.

10. The urban road dynamic noise source tracing and evaluation system based on digital twin according to claim 6 is characterized in that: The decision optimization unit includes: a parameterized sound insulation screen simulator, which is used to perform the insertion loss calculation described in claim 5; and a traffic flow control simulator, which predicts the noise-congestion rate correlation curve and outputs an optimization strategy.

Images