A noise barrier optimization method, device and medium based on noise source database

Abstract

The embodiments of this specification disclose a sound barrier optimization method, device and medium based on a noise source database, which relate to the field of sound barrier technology. The method includes: collecting multi-source noise time series data corresponding to the target road, performing noise feature extraction and noise prediction based on the multi-source noise time series data, and constructing a noise source database, wherein the noise source database includes a real-time noise feature database corresponding to the current time node and a predicted noise feature database corresponding to the future time node; performing multi-period noise propagation simulation on the noise-sensitive area of ​​the target road through the real-time noise feature database and the predicted noise feature database, determining the sensitive noise data corresponding to the multiple time nodes, and determining the multi-period constraint conditions corresponding to the sound barrier; determining the corresponding sound barrier design parameter solution set based on the multi-period constraint conditions and a preset multi-objective optimization algorithm, and determining the sound barrier optimization strategy according to the design parameter solution set.

Description

Technical Field

[0001] This specification relates to the field of sound barrier technology, and in particular to a sound barrier optimization method, equipment and medium based on a noise source database. Background Art

[0002] While the continued expansion of the expressway network improves regional connectivity, it also increases the scope and intensity of traffic noise pollution. The broadband, high-intensity noise generated by highway traffic is highly penetrating, posing a persistent health threat to sensitive areas along the route, such as residential areas, schools, and hospitals. Acoustic isolation is essential to ensure environmental compliance. Furthermore, road network expansion has become a normal part of modern transportation systems. This continuous spatial expansion not only expands the distribution range of noise sources but also, through network densification, creates spatial intersections with existing residential areas, bringing noise-sensitive areas previously far from roads into the impact radius.

[0003] The traditional sound barrier design process typically follows a linear decision-making process. First, based on short-term, fixed-point noise monitoring, current areas exceeding standards are identified. Next, noise reduction targets are set based on static environmental standards. Sound insulation structures are selected according to general material manuals, and finally, fixed-position deployment and implementation are implemented. In traditional sound barrier design, existing barrier locations are based on historical road boundaries. New ramps or parallel auxiliary roads often penetrate existing protection zones, rendering the barriers functionally ineffective and requiring their removal and reconstruction. Furthermore, structural changes in traffic volume caused by line expansions create a mismatch between the noise reduction capabilities of the original design frequency band and the actual noise spectrum characteristics, necessitating the installation of compensatory sound insulation components. Furthermore, the construction of new viaducts, tunnels, and other structures associated with road network expansion alters sound propagation paths, shrinking the original barrier's acoustic shadow area and even creating reflection focal points, creating the risk of secondary noise pollution. Traditional sound barrier design, in response to road network expansion, passively responds to road network expansion with a pattern of construction, demolition, and reconstruction, resulting in a continuous drain on public resources.

[0004] Therefore, traditional sound barrier design relies on historical road boundary settings and short-term noise data generated by traffic flow captured by short-term monitoring. When faced with planning events such as road network expansion, additional passive response operations such as demolition and reconstruction are required, which increases the loss of additional noise reduction resources. Summary of the Invention

[0005] One or more embodiments of the present specification provide a sound barrier optimization method, device, and medium based on a noise source database, which are used to solve the following technical problems: Traditional sound barrier design relies on historical road boundary settings and short-term noise data generated by traffic flow captured by short-term monitoring. When faced with planning events such as road network expansion, additional passive response operations such as demolition and reconstruction are required, which increases the loss of additional noise reduction resources.

[0006] One or more embodiments of this specification adopt the following technical solutions:

[0007] One or more embodiments of the present specification provide a sound barrier optimization method based on a noise source database, characterized in that the method includes: collecting multi-source noise time series data corresponding to the target road, performing noise feature extraction and noise prediction based on the multi-source noise time series data, and constructing a noise source database, wherein the noise source database includes a real-time noise feature database corresponding to the current time node and a predicted noise feature database corresponding to the future time node; performing multi-period noise propagation simulation on the noise-sensitive area of ​​the target road through the real-time noise feature database and the predicted noise feature database, determining the sensitive noise data corresponding to the multiple time nodes, and determining the multi-period constraint conditions corresponding to the sound barrier according to the sensitive noise data corresponding to the multiple time nodes; determining the corresponding sound barrier design parameter solution set based on the multi-period constraint conditions and a preset multi-objective optimization algorithm, and determining the sound barrier optimization strategy according to the sound barrier design parameter solution set.

[0008] One or more embodiments of this specification provide a sound barrier optimization device based on a noise source database, including:

[0009] at least one processor; and,

[0010] a memory communicatively connected to the at least one processor; wherein,

[0011] The memory stores instructions that can be executed by the at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to perform the above method.

[0012] One or more embodiments of this specification provide a non-volatile computer storage medium storing computer-executable instructions, wherein the computer-executable instructions are configured to execute the above method.

[0013] At least one of the above technical solutions adopted in the embodiments of this specification can achieve the following beneficial effects: Through the technical solutions of the embodiments of this specification, the current road noise status is accurately captured through a real-time noise feature database. Based on the noise evolution trends derived from planning events such as road network expansion embedded in the predicted noise feature database, the traditional design that relies on short-term historical data is upgraded to a dual-track drive of current measurement and future prediction, avoiding the problem of sound barrier function lag caused by planning variables at the source; the real-time and predicted noise spectra are simulated in three dimensions to quantify the path attenuation effect of noise in sensitive areas, solving the defect of traditional methods that ignore the dynamic variation of propagation paths, such as the change of the acoustic shadow zone shape due to ramp expansion, so that the positioning of the sound barrier accurately matches the actual sound field distribution; based on sensitive noise data, current and future noise reduction constraints that meet immediate standards are simultaneously generated, establishing a joint defense mechanism to combat sound energy accumulation and material performance degradation caused by road network expansion, eliminating the need for passive modification in the middle and late stages of service; with the goals of minimizing construction costs, maximizing current noise reduction margins, and maximizing the probability of future noise reduction compliance, Pareto optimal solutions are screened from candidate solutions, achieving a one-time design that covers the entire life cycle needs, replacing the traditional resource-consuming model of retrofitting after construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are only some of the embodiments described in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without inventive work. In the drawings:

[0015] Figure 1 A flowchart of a sound barrier optimization method based on a noise source database provided in an embodiment of this specification;

[0016] Figure 2 This is a structural diagram of a sound barrier optimization device based on a noise source database provided in an embodiment of this specification. DETAILED DESCRIPTION

[0017] To help those skilled in the art better understand the technical solutions in this specification, the following will provide a clear and complete description of the technical solutions in the embodiments of this specification, in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this specification, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this specification without creative work should fall within the scope of protection of this specification.

[0018] The embodiment of this specification provides a sound barrier optimization method based on a noise source database. It should be noted that the execution entity in the embodiment of this specification can be a server or any device with data processing capabilities. Figure 1 A flow chart of a sound barrier optimization method based on a noise source database provided in an embodiment of this specification is shown as follows: Figure 1 As shown, it mainly includes the following steps:

[0019] Step S101 : collecting multi-source noise time series data corresponding to the target road, performing noise feature extraction and noise prediction based on the multi-source noise time series data, and constructing a noise source database.

[0020] The noise source database includes a real-time noise feature database corresponding to the current time node and a predicted noise feature database corresponding to the future time node;

[0021] Based on the multi-source noise time series data, noise feature extraction and noise prediction are performed to construct a noise source database, specifically including: determining current noise spectrum data, measurement timestamp and measurement location coordinates in the multi-source noise time series data, and associating the current noise spectrum data with a structured database based on the measurement timestamp and the measurement location coordinates to construct a real-time noise feature database corresponding to the current time node; obtaining a road planning factor corresponding to the target road, and performing noise prediction based on the road planning factor and the real-time noise feature database to determine a future predicted noise spectrum to construct a predicted noise feature database. Based on the road planning factor and the real-time noise feature database, noise prediction is performed to determine a future predicted noise spectrum, specifically including: obtaining lane expansion data and industrial project commissioning data in the road planning factor; determining a lane number increment based on the lane expansion data, and determining a lane number increment ratio by combining the lane number increment and a pre-acquired existing lane number; determining a predicted noise growth amount based on a pre-acquired historical traffic volume growth rate of the target road and the lane number increment ratio; determining an industrial equipment type based on the industrial project commissioning data, and calculating high-frequency noise proportion data of the industrial equipment based on the equipment type of the industrial equipment; determining a spectrum change amount based on the high-frequency noise proportion data, and superimposing the predicted noise growth amount and the high-frequency noise proportion data on the current noise spectrum corresponding to the real-time noise feature database to generate the future predicted noise spectrum.

[0022] Traditional highway noise barrier designs, relying on static data collection and rigid target setting, frequently face functional failures amidst the continuous expansion of the road network. Planning variables such as road expansion and the construction of new industrial zones inevitably alter the spatial and temporal distribution of noise. Without quantitative predictions of lane increment ratios and the spectral characteristics of heavy-duty equipment, the constructed barriers will quickly lose their noise reduction performance due to changes in traffic structure or the penetration of high-frequency noise.

[0023] In one embodiment of the present specification, first, multi-source noise time series data of the target road is collected. Fixed noise monitoring stations and vehicle-mounted mobile monitoring equipment are deployed to continuously record the original sound pressure level signals and GPS timestamps of the entire road network. At the same time, the traffic flow monitoring system can be synchronously connected to obtain real-time vehicle model composition and speed data. The above raw data with time and space labels are decomposed into structured features such as equivalent sound level values, 1 / 3 octave spectrum energy distribution from 63Hz to 8kHz, and noise event duration through a noise feature analysis engine. The measurement time and spatial coordinates corresponding to each monitoring data are then associated and stored in a real-time noise feature database with the geographic grid as the index unit, forming a real-time noise feature database for the current time node corresponding to the current road noise field.

[0024] From preset data interfaces, such as planning information published by urban planning agencies, obtain planning data such as the expansion construction data of the target road or the environmental impact assessment report of the industrial zone construction project corresponding to the target road, and extract key planning parameters. It should be noted that the planning data here needs to be industrial zones that are within a certain range of the target road and affect the spatiotemporal distribution of noise on the target road, such as within a preset radius of the target road. For lane expansion information, obtain the completion year and lane number increment of the lane expansion project. For park planning information, obtain data such as the commissioning time data and equipment model list of heavy equipment in the industrial park.

[0025] Traffic noise prediction is performed based on lane expansion information. The ratio coefficient between the existing number of lanes and the total number of lanes after the expansion is calculated. Combined with the average annual traffic growth rate curve for historical periods provided by transportation agencies, the predicted noise increase is determined based on the historical traffic growth rate and the lane increase ratio. It should be noted that this derivation process needs to account for changes in acoustic interference patterns caused by the addition of new lanes, such as the acoustic coherence effect caused by the expansion of parallel lanes.

[0026] After obtaining lane expansion construction information for the target road, the cross-sectional images contained in the lane expansion construction information are first analyzed to accurately quantify the change in lane count before and after the expansion. For example, if an existing six-lane bidirectional road (three lanes in each direction) is expanded to eight-lane bidirectional road (four lanes in each direction), the lane increment is two lanes. This incremental data must be compared with the existing lane number approval documents archived by the transportation department to ensure the accuracy of the existing baseline value. Ultimately, the lane increment ratio is calculated, which in this example is 2 / 6 ≈ 33.3%. At least five consecutive years of vehicle flow monitoring records for each vehicle type, including passenger cars, trucks, and special operation vehicles, are extracted and time series analysis is used to fit the average annual traffic flow growth rate. Different correction methods are then applied for different scenarios. For existing road widening and expansion, the growth rate is directly coupled to the lane increment ratio, which is converted into an equivalent vehicle density increase. The base component of the noise increase is then derived based on the exponential correlation between sound power level and vehicle density.

[0027] Traffic density mapping is used to couple lane addition and noise growth prediction. First, the current design capacity of the target road is determined. For example, a six-lane, two-way road has a capacity of 4,500 vehicles per hour. Based on the physical structural changes in the lane expansion drawings, such as adding two lanes to bring the total number of lanes to eight, and combined with pre-defined rules, such as the "Urban Road Engineering Design Code," the relationship between lane width and traffic efficiency is determined and converted into a capacity increase ratio. In this case, for example, the capacity is increased to 6,000 vehicles per hour, an increase of approximately 33%. This capacity increase is the theoretical vehicle density increase.

[0028] The benchmark calculation model is selected according to the lane expansion type. For the original road widening type, the lane number increment ratio directly acts on the sound power level correction. The benchmark correction amount = 10×log 10 (1+lane increment ratio), for example, the correction amount obtained by increasing the number of lanes by 33.3% is ≈1.2dB(A). If it is a new parallel double-track line, it is necessary to simultaneously refer to the lane expansion sound radiation correction coefficient in the "Highway Sound Environmental Impact Assessment Standard" and introduce the sound wave interference attenuation coefficient, which is usually -0.5dB to +0.8dB. For example, when adding lanes in the same direction, an additional 1.5dB sound interference correction value is required. The historical traffic flow growth rate is converted into an equivalent sound energy contribution through the time-flow transfer function, and the traffic flow sound energy contribution = 10×log 10 ((1+average annual growth rate) 规划年数 ), where the planning years refer to the time interval between the current time and the expected start of production. The resulting total noise increase is the sum of the baseline correction, the traffic sound energy contribution, and the interference correction term.

[0029] To analyze the evolution of industrial noise spectra, the industrial equipment noise standard database is searched based on the equipment model. The characteristic frequency band energy distribution of specific equipment is extracted, such as the high sound energy characteristics of air compressors in the 2kHz-4kHz frequency band. The change in the proportion of high-frequency energy in the total noise source of newly added equipment is calculated to obtain the spectrum change. Finally, the traffic noise growth value and the spectrum change are superimposed on the current spectrum data of the real-time noise feature database. The sound energy vector synthesis algorithm is used to generate a future noise prediction spectrum with spatial grid identification and planning time tags. For example, a grid unit is currently dominated by 63Hz low-frequency sound. Due to the addition of a new metal processing workshop, a high-frequency noise envelope will be superimposed, forming a composite spectrum prediction result, forming a predicted noise feature database containing future time nodes.

[0030] Through the above technical solution, the limitations of traditional single-time point design are broken through the dual modeling of real-time noise spectrum and future predicted spectrum. Before road expansion or new industrial zone construction, the noise growth caused by the lane increment ratio and the spectral drift trend of high-frequency noise of industrial equipment can be predicted, so that the sound barrier can be pre-embedded with redundant capabilities to cope with the deterioration of the future noise environment at the early stage of design, avoiding the waste of resources caused by frequent demolition and modification of traditional solutions in the later stage of service. For example, for the lane expansion plan five years later, the sound insulation margin corresponding to the noise increment has been reserved in the design stage, and no secondary modification is required. In addition, the traditional method only focuses on the total sound pressure level control of traffic noise, while this solution accurately maps the proportion of high-frequency noise by equipment type, so that the selection of sound barrier materials can directionally strengthen the high-frequency sound absorption structure. Road planning factors such as lane expansion coordinates, industrial layout such as heavy-duty equipment points and the location of noise-sensitive areas form spatial superposition analysis capabilities to achieve precise geographical matching.

[0031] In step S102, a multi-period noise propagation simulation is performed on the noise-sensitive area of ​​the target road through the real-time noise feature database and the predicted noise feature database to determine the sensitive noise data corresponding to the multiple time nodes, so as to determine the multi-period constraint conditions corresponding to the sound barrier based on the sensitive noise data corresponding to the multiple time nodes.

[0032] As urban road networks continue to expand, the acoustic environment in noise-sensitive areas faces a variety of dynamic influences, including new buildings altering sound diffraction paths, industrial equipment introducing spectral pollution, and increased traffic leading to the accumulation of acoustic energy. Relying solely on raw spectral data from noise source databases without quantifying the physical distortion of sound waves during propagation due to building obstruction, reflection, and diffraction can lead to serious misjudgments in sound barrier design.

[0033] A multi-period noise propagation simulation is performed on the noise-sensitive area of ​​the target road using the real-time noise feature database and the predicted noise feature database to determine sensitive noise data corresponding to multiple time nodes. The simulation specifically includes: obtaining current noise spectrum data from the real-time noise feature database and future predicted noise spectrum from the predicted noise feature database; determining at least one noise-sensitive area on the target road, obtaining current building information within the noise-sensitive area, calculating diffraction path attenuation for the current noise spectrum data based on the building information, and determining current sensitive noise data; and obtaining new building information within the noise-sensitive area at the future time node, performing predicted propagation correction on the future predicted noise spectrum based on the new building information, and determining predicted sensitive noise data.

[0034] In one embodiment of this specification, sensitive areas, such as schools and hospitals, are first extracted from an urban planning platform. Three-dimensional building model data for the target noise-sensitive areas is then retrieved. This 3D building model contains the precise geometric dimensions of the existing buildings, acoustic parameters of the facade materials, such as a 0.05 sound absorption coefficient for concrete walls and a 0.03 sound absorption coefficient for glass curtain walls, and geodetic coordinate system positioning information. For the current sound field simulation, the sound source spectrum parameters from a real-time noise signature database are imported into acoustic simulation software (such as SoundPLAN or CadnaA). The road centerline is set as the elevation position of the line sound source, 1.5 meters above the ground. The diffraction path is automatically calculated based on the 3D terrain data and building coordinates. The diffraction effect of sound waves at the building edges is analyzed, such as the bending and attenuation of sound waves at the corners of hospital walls. The reflection and superposition effects of hard facades are also analyzed, such as the coherent enhancement of reflected sound from the tiled walls of a school building. For example, when simulating nighttime noise propagation in a residential area, the air absorption coefficient needs to be corrected by combining temperature and humidity gradient data from a meteorological database. The final output is a 3D noise energy distribution cloud map of the current sensitive point, with spectral characteristics.

[0035] For the predicted scenario, the 3D design model of the expansion project was loaded. BIM models of approved new construction projects within sensitive areas were obtained from authorized agencies, such as the Ministry of Housing and Urban-Rural Development. The spatial location and structural parameters, such as the proposed 30-story residential building, were extracted. Propagation path corrections were performed on the predicted noise spectrum sources. Sound barriers were inserted into the physical space of the new building, and the reconstruction of the acoustic shadow zone under their shielding effect was calculated. The scattering effect of noise on the facades of the new building was also simulated. For example, using the Future Industrial Zone Primary School as an example, the mid- and high-frequency noise from the newly added condensing tower equipment was simulated, reflecting off the metal roof of the adjacent factory building, creating a hotspot with a 12dB sound pressure increase in the playground area. All simulation results were stored in a spatial grid-coded, associative manner, forming a time-stamped dynamic map of sensitive noise.

[0036] In noise control applications, real-time and predictive noise signature databases only record the raw spectral characteristics of sound sources, such as the sound pressure level and frequency band energy distribution at the road centerline. However, noise in noise-sensitive areas depends on the physical distortion effects of the three-dimensional path from the source to the receiving point. Diffraction barriers formed by existing buildings can alter the acoustic shadow zone. Reflections from the facades of newly constructed high-rise structures can create hotspots of acoustic energy. Hidden passages such as underground utility corridors or bridge expansion joints can induce diffraction penetration of low-frequency noise. Directly using database sound source data without analyzing the propagation path results in an inability to quantify the true sound field at sensitive points. When new buildings or unusual terrain lie between a sensitive point and the sound source, changes in the propagation path can cause the predicted value to differ from the actual received value at the sensitive point. By dynamically loading existing and planned building parameters to capture path changes and using multi-period propagation simulations, the abstract sound source spectrum is transformed into the actual current and future risks of sensitive areas. This enhances the precision of sound barrier design in both path blocking and receiving point protection, effectively avoiding the retrofit losses caused by propagation blind spots associated with traditional methods.

[0037] Determining the multi-period constraints corresponding to the sound barrier based on the sensitive noise data corresponding to the multiple time nodes specifically includes: obtaining the current sensitive noise data and the predicted sensitive noise data from the noise sensitive data; determining the current noise reduction constraints based on the current sensitive noise data and the mean of the current noise spectrum; obtaining the material attenuation index corresponding to the sound barrier from a preset database, and determining the future noise reduction constraints based on the material attenuation index and the predicted sensitive noise data. Determining the future noise reduction constraints based on the material attenuation index and the predicted sensitive noise data specifically includes: obtaining the future time point corresponding to the predicted sensitive noise data, and determining the corresponding cumulative attenuation parameter for the future time span based on the future time point and the material attenuation index; and determining the future noise reduction constraints based on the predicted sensitive noise data and the cumulative attenuation parameter.

[0038] The expansion of the target road caused a nonlinear increase in sound pressure levels in sensitive areas. Furthermore, acid rain corrosion caused the sound insulation performance of the materials to decline year by year, and the relocation of industrial areas changed the noise spectrum. Traditional designs, which only set a single static constraint value, failed to address the risk of exceeding the standard caused by future surges in traffic volume and failed to quantify the acoustic performance degradation of the sound barrier material.

[0039] In one embodiment of the present specification, current sensitive noise data and predicted sensitive noise data are first obtained from the noise-sensitive data. The equivalent sound pressure level (Leq) and 1 / 3 octave spectrum data corresponding to each sensitive area are then obtained using the current sensitive noise data. A data cleaning engine is used to eliminate periods of interference from extreme weather events (such as heavy rain and strong winds) and calculate the statistical mean of noise for the current 30 consecutive days. This mean is then compared with the statutory limit. For example, the measured daytime mean at a sensitive point in a residential area is 56.3 dB, compared to the 60 dB limit for a Class II acoustic environment functional zone. The current noise reduction requirement is initially set at 3.7 dB. The noise impact weight of surrounding road construction is then verified. If the municipal planning announcement indicates that an elevated ramp will be built in six months, an expected incremental margin of 2.1 dB is added, ultimately generating a current noise reduction constraint label with a calibration value of 5.8 dB.

[0040] To construct future constraints, the first step is to access the database of the National Building Materials Testing Center to retrieve durability data for sound barrier panels that match the target region's climate. For coastal projects, a standard attenuation curve for galvanized steel sheets in salt spray corrosion environments is extracted, automatically mapping the material's service life to its performance decay rate. For example, the sound insulation value in year N = initial value - annual decay rate × N. The second step is to identify the legally mandated timeframes for the commissioning of key industrial equipment. For example, a chemical plant's air compressor unit is scheduled to begin operations in October 2029. When the time span falls within a full year, the full year span from the current date to the commissioning date is calculated and substituted into the decay rate parameter to generate a cumulative decay coefficient. For example, if the air compressor is commissioned five years from now and the sheet metal has an annual decay rate of 0.75%, the cumulative decay coefficient is 5 × 0.75% = 3.75%. For non-full-year time spans, the fractional fraction across years is converted proportionally to obtain the final cumulative decay coefficient.

[0041] The third step is to load the predicted sensitive noise spectrum generated by the propagation simulation into the acoustic simulation platform, including the energy values ​​in the characteristic frequency bands of industrial equipment. Pre-compensation for material performance degradation is performed for the target sensitive point data, such as the predicted value of 68.7dB for residential buildings at the factory boundary. The cumulative attenuation coefficient is converted into an equivalent sound pressure loss value by multiplying the predicted value of the target sensitive area by the cumulative attenuation coefficient. For example, assuming a cumulative attenuation coefficient of 5.04%, the calculation is 68.7dB × 5.04% ≈ 3.46dB. This loss value is then added to the regional standard limit to form the basic target. Assuming the regional standard limit is 55, the corresponding basic target is 55 + 3.46 = 58.46dB. The key correction step is to add an impulse noise correction factor of 1.2dB when the predicted value contains high-frequency components of industrial equipment, such as when the 2kHz frequency band of the reactor accounts for more than 35%. The final result is a future noise reduction constraint value of 59.66dB with frequency band annotation.

[0042] Through multi-period constraint modeling, the current constraint value determines the immediate noise reduction demand, and the future constraint value is embedded in the predicted noise spectrum growth, material cumulative attenuation rate and industrial spectrum drift. The static single-point constraint is expanded into a multi-parameter time-varying function that integrates material durability attenuation, noise energy growth and spectrum characteristic variation. This enables the sound barrier design to accurately respond to the high-frequency noise peak impact during the operation period of industrial equipment in the future time period, effectively solving the problem that the sound barrier function in the middle and late stages of service cannot meet the noise reduction demand.

[0043] Step S103, based on the multi-period constraint conditions and the preset multi-objective optimization algorithm, determine the corresponding sound barrier design parameter solution set, and determine the sound barrier optimization strategy according to the sound barrier design parameter solution set.

[0044] Based on the multi-period constraint conditions and the preset multi-objective optimization algorithm, the corresponding sound barrier design parameter solution set is determined, specifically including: according to the constructed noise source database, the hot zone of the target road is identified to determine the sound barrier candidate solution set, wherein the sound barrier candidate solution set includes at least one candidate solution, and the candidate solution includes a candidate position interval, length and height combination; through the multi-objective optimization algorithm and the multi-period constraint conditions, with the minimization of construction cost, maximization of current noise reduction margin, and maximization of future noise reduction compliance probability as the optimization goals, the corresponding sound barrier design parameter solution set is determined in the sound barrier candidate solution set. Based on the constructed noise source database, hot spots are identified on the target road to determine a set of candidate sound barrier solutions, specifically including: obtaining a real-time noise feature database and a predicted noise feature database corresponding to future time nodes in the constructed noise source database; based on the current noise spectrum data in the real-time noise feature database, screening at least one road section exceeding the standard that is greater than a preset minimum threshold value for complaint noise, and determining the corresponding matching future predicted noise spectrum in the predicted noise feature database based on the measured position coordinates of the exceeding road section; determining the predicted noise increment of the exceeding road section through the matching future predicted noise and the current noise spectrum data, and determining a hot spot section in the exceeding road section based on the predicted noise increment; according to the pre-acquired sound barrier design rules, based on the section length corresponding to the hot spot section, performing parameter combination on the position interval, sound barrier length and sound barrier height of the sound barrier setting to determine the set of candidate sound barrier solutions.

[0045] Against the backdrop of the continuous densification of highway networks, traditional sound barrier designs lack precise positioning capabilities, and their uniform layout across the entire area leads to serious waste of resources. Experience-based selection of hotspots easily misses future expansion risk points, making it difficult to cope with the spatial drift of noise hotspots under the dynamic evolution of road network topology.

[0046] In one embodiment of this specification, statistics are first collected from historical noise complaint data to determine the minimum threshold for complaint noise. For example, the statistical lower limit for persistent complaint hotspots in residential areas is 61 dB (daytime), which is configured as the preset minimum threshold for complaint noise. Current noise spectrum data for the entire road network from a real-time noise signature database is loaded into a GIS (Geographic Information System). A spatial overlay algorithm is used to identify all road sections where the average daily equivalent sound pressure level has exceeded the threshold for 30 consecutive days, creating a vector layer of these sections, organized by pile number intervals. Spatial matching is performed within the predicted noise signature database. Based on the spatial coordinate range of the exceeding sections, such as longitude 118.25°-118.31° and latitude 32.18°-32.22°, future predicted noise spectrum data for the corresponding geographic grid is extracted. The current noise spectrum sound pressure level at the point is subtracted from the future predicted sound pressure level in the future predicted noise spectrum data to generate a noise increment distribution map labeled with pile numbers. This noise increment distribution map includes the predicted noise increment for the exceeding road sections.

[0047] Based on the predicted noise increment, hotspot sections are identified within the marked road sections. Sections where the predicted noise increment exceeds a preset increment threshold are designated as hotspot sections. For example, the increment threshold can be set to 3dB. Based on the previously acquired sound barrier design rules and the length of the road section corresponding to the hotspot section, a parameter combination of the sound barrier location interval, sound barrier length, and sound barrier height is calculated to determine the set of candidate sound barrier solutions.

[0048] The candidate solution set generated by the hotspot identification phase was imported into the city's multi-objective decision-making platform, which integrates a construction cost quota database, an acoustic performance simulation engine, and a Pareto optimization algorithm. Using this multi-objective optimization algorithm and multi-period constraints, the corresponding sound barrier design parameter solution set was determined from the candidate sound barrier solutions, with the optimization objectives of minimizing construction cost, maximizing current noise reduction margin, and maximizing the probability of meeting future noise reduction standards. Based on this solution set, the sound barrier optimization strategy was determined. During the initialization phase, the parameter set (pile number range / barrier type / length / height combination) for each candidate solution was retrieved for third-order objective modeling. Construction cost modeling began with the latest material price list for the corresponding area in the construction cost quota database. For example, a 3.5m micro-perforated barrier costs ¥5,800 / linear meter. Special terrain surcharges, such as bridge section construction costs × 1.8 and tunnel entrance hoisting costs × 2.2, were then added to generate a total cost budget. Next, a noise reduction performance simulation was conducted. The proposed 3D model and the current sensitive noise spectrum were loaded into SoundPLAN software. The ISO 9613-2 standard algorithm was used to calculate the noise reduction compliance rate at sensitive points. For example, the simulated noise reduction at the K13+200 residential area was 6.3 dB. Material attenuation parameters, such as a 5-year attenuation rate of 3.75%, were loaded, along with the predicted sensitive noise spectrum. The probability of achieving the noise reduction standard after material aging was simulated using the same model. Multi-period constraints were then introduced, including current and future constraint values. The core constraint rule was set: current noise reduction ≥ (current constraint value - current value) × 1.05 (safety factor), and future compliance probability ≥ 95%.

[0049] The NSGA-II algorithm performs multi-objective problem solving. First, a population is initialized, randomly generating 200 parameter combinations (with a range of ±10% between the starting and ending points of the pile numbers and a ±0.3m height fluctuation). Cost accounting and dual-period noise reduction simulations are performed on each individual solution in parallel. Solutions that violate hard constraints, such as those with a future probability of compliance less than 90% or those with cost exceeding the budget, are eliminated. The remaining solutions are then non-dominated and sorted based on the objectives of lowest cost, largest current margin, and highest future probability. The top 20% of these solutions advance to the next generation of evolutionary iterations. After multiple generations of evolution, a Pareto-optimal frontier solution set is generated. For example, Solution A: pile numbers K12+080-570, 3.5m micro-perforated barrier, cost ¥1.42 million, current noise reduction margin of 7.2dB, and future compliance probability of 96%; Solution B: pile numbers K12+100-550, 3.2m composite barrier, cost ¥1.31 million, current margin of 6.8dB, and future compliance probability of 95%.

[0050] Through the above technical solutions, dynamic prediction and intelligent decision-making enable full-chain optimization of the sound barrier project, from planning to implementation, transforming traditional static empirical design into a data-driven approach. A hotspot identification mechanism based on predicted noise increments accurately identifies future noise-exceeding risk points, focusing noise reduction resources on high-threat sections of road, avoiding the ineffective investment caused by traditional uniform, all-encompassing layouts. The coordinated optimization of multi-period constraints and material attenuation models ensures that the sound barrier continues to respond to the dual pressures of traffic growth and industrial relocation throughout its service life. The multi-solution comparison capability provided by the Pareto solution set enables designers to quickly generate alternatives in the event of sudden budget fluctuations, significantly improving project resilience.

[0051] Through the technical solutions of the embodiments of this specification, the current road noise status is accurately captured through a real-time noise feature database. Based on the noise evolution trends derived from planning events such as road network expansion embedded in the predicted noise feature database, the traditional design that relies on short-term historical data is upgraded to a dual-track drive of current measurement and future prediction, avoiding the problem of sound barrier function lag caused by planning variables at the source; the real-time and predicted noise spectra are simulated in three dimensions to quantify the path attenuation effect of noise in sensitive areas, solving the defect of traditional methods that ignore the dynamic variation of propagation paths, such as the change of the acoustic shadow zone shape due to ramp expansion, so that the positioning of the sound barrier accurately matches the actual sound field distribution; based on sensitive noise data, current and future noise reduction constraint values ​​that meet immediate standards are simultaneously generated, establishing a joint defense mechanism to combat sound energy accumulation and material performance degradation caused by road network expansion, eliminating the need for passive modification in the middle and late stages of service; with the goals of minimizing construction costs, maximizing current noise reduction margins, and maximizing the probability of future noise reduction compliance, the Pareto optimal solution is screened from candidate solutions, achieving a one-time design covering the entire life cycle needs, replacing the traditional resource dissipation model of build and then modify.

[0052] The embodiment of this specification also provides a sound barrier optimization device based on a noise source database, such as Figure 2 As shown, the device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the above method.

[0053] The embodiments of this specification also provide a non-volatile computer storage medium storing computer executable instructions, wherein the computer executable instructions are configured to execute the above method.

[0054] The various embodiments in this specification are described in a progressive manner. Similar portions between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from the other embodiments. In particular, the device, apparatus, and non-volatile computer storage medium embodiments are generally similar to the method embodiments, so their descriptions are relatively simplified. For relevant details, refer to the descriptions of the method embodiments.

[0055] The foregoing description is merely one or more embodiments of this specification and is not intended to limit this specification. It will be apparent to those skilled in the art that various modifications and variations may be made to one or more embodiments of this specification. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of one or more embodiments of this specification are intended to be within the scope of the claims of this specification.

Claims

1. A noise barrier optimization method based on a noise source database, characterized in that: The method comprises: Collecting multi-source noise time series data corresponding to the target road, performing noise feature extraction and noise prediction based on the multi-source noise time series data, and constructing a noise source database, wherein the noise source database includes a real-time noise feature database corresponding to a current time node and a predicted noise feature database corresponding to a future time node; Performing a multi-period noise propagation simulation on the noise-sensitive area of ​​the target road using the real-time noise feature database and the predicted noise feature database to determine sensitive noise data corresponding to multiple time nodes, and determining multi-period constraint conditions corresponding to the sound barrier based on the sensitive noise data corresponding to the multiple time nodes; Determine a corresponding sound barrier design parameter solution set based on the multi-period constraint conditions and a preset multi-objective optimization algorithm, and determine a sound barrier optimization strategy based on the sound barrier design parameter solution set; Based on the multi-source noise time series data, noise feature extraction and noise prediction are performed to build a noise source database, specifically including: Determining current noise spectrum data, a measurement timestamp, and a measurement location coordinate in the multi-source noise time series data, and associating the current noise spectrum data with a structured database based on the measurement timestamp and the measurement location coordinate to construct a real-time noise feature database corresponding to the current time node; Obtaining a road planning factor corresponding to the target road, performing noise prediction based on the road planning factor and the real-time noise feature database, determining a future predicted noise spectrum, and constructing a predicted noise feature database; Based on the road planning factors and the real-time noise feature database, noise prediction is performed to determine a future predicted noise spectrum, specifically including: Obtaining lane expansion data and industrial project commissioning data from the road planning factors; Determining a lane number increment based on the lane expansion data, and determining a lane number increment ratio by comparing the lane number increment with a pre-acquired number of existing lanes; Determining a predicted noise growth amount based on a previously acquired historical traffic volume growth rate of the target road and the lane number increment ratio; Determine the type of industrial equipment according to the industrial project commissioning data, and calculate the high-frequency noise proportion data of the industrial equipment based on the type of the industrial equipment; The spectrum change is determined by the high-frequency noise proportion data, and the predicted noise growth and the high-frequency noise proportion data are superimposed on the current noise spectrum corresponding to the real-time noise feature database to generate the future predicted noise spectrum.

2. The noise barrier optimization method based on the noise source database according to claim 1, characterized in that: Using the real-time noise feature database and the predicted noise feature database, a multi-period noise propagation simulation is performed on the noise-sensitive area of ​​the target road to determine the sensitive noise data corresponding to multiple time nodes, specifically including: Acquire current noise spectrum data in the real-time noise feature database and future predicted noise spectrum in the predicted noise feature database; determining at least one noise-sensitive area in the target road, obtaining current building information in the noise-sensitive area, calculating diffraction path attenuation for the current noise spectrum data based on the building information, and determining current sensitive noise data; New building information of the noise-sensitive area within the future time node is obtained, and based on the new building information, a predicted propagation correction is performed on the future predicted noise spectrum to determine predicted sensitive noise data.

3. The noise barrier optimization method based on the noise source database according to claim 2, characterized in that: Determine the multi-period constraint conditions corresponding to the sound barrier based on the sensitive noise data corresponding to the multiple time nodes, specifically including: Acquiring current sensitive noise data and predicted sensitive noise data from the noise sensitive data; determining a current noise reduction constraint condition according to a mean value of the current sensitive noise data and the current noise spectrum data; In a preset database, the material attenuation index corresponding to the sound barrier is obtained to determine future noise reduction constraints based on the material attenuation index and the predicted sensitive noise data.

4. The noise barrier optimization method based on the noise source database according to claim 3 is characterized in that: Based on the material attenuation index and the predicted sensitive noise data, future noise reduction constraints are determined, specifically including: Obtaining a future time point corresponding to the predicted sensitive noise data, and determining a corresponding cumulative attenuation parameter in the future time span based on the future time point and the material attenuation index; Based on the predicted sensitive noise data and the accumulated attenuation parameter, future noise reduction constraints are determined.

5. The noise barrier optimization method based on the noise source database according to claim 1, characterized in that: Based on the multi-period constraints and the preset multi-objective optimization algorithm, the corresponding sound barrier design parameter solution set is determined, specifically including: Based on the noise source database constructed, hotspot identification is performed on the target road to determine a set of sound barrier candidate solutions, wherein the set of sound barrier candidate solutions includes at least one candidate solution, and the candidate solution includes a candidate position interval, length, and height combination; Through the multi-objective optimization algorithm and the multi-period constraints, with the optimization goals of minimizing construction costs, maximizing current noise reduction margins, and maximizing the probability of meeting future noise reduction standards, the corresponding sound barrier design parameter solution set is determined in the sound barrier candidate solution set.

6. The noise barrier optimization method based on the noise source database according to claim 5, characterized in that: Based on the noise source database constructed, hotspot identification is performed on the target road to determine a set of candidate sound barrier solutions, specifically including: Acquire a real-time noise feature database and a predicted noise feature database corresponding to future time nodes in the constructed noise source database; Based on current noise spectrum data in the real-time noise signature database, screening at least one road section exceeding the standard with a noise level greater than a preset minimum complaint noise threshold, and determining a corresponding matching future predicted noise spectrum in the predicted noise signature database based on the measured location coordinates of the road section exceeding the standard; Determining a predicted noise increment of the road section exceeding the standard by matching the future predicted noise with the current noise spectrum data, and determining a hotspot section in the road section exceeding the standard based on the predicted noise increment; According to the pre-acquired sound barrier design rules and based on the section length corresponding to the hot zone section, the position interval, length and height of the sound barrier are parameterized to determine the set of candidate sound barrier solutions.

7. A sound barrier optimization device based on a noise source database, characterized in that: The device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 6.

8. A non-volatile computer storage medium storing computer executable instructions, characterized in that: The computer executable instructions are configured to execute the method according to any one of claims 1 to 6.

Images