Noise monitoring point site selection method and system based on noise map

Abstract

The invention discloses a noise monitoring point site selection method and system based on a noise map, and relates to the technical field of environmental noise monitoring, and the method comprises the steps: constructing a three-dimensional topology network based on a three-dimensional noise intensity matrix, calculating an optimal sound propagation path through employing a quantum heuristic topology optimization algorithm, and obtaining an optimal sound propagation path; the weights of nodes in the three-dimensional topology network are optimized in combination with quantum annealing, and an acoustic topology characteristic spectrum is generated; extracting multi-scale noise propagation features through a convolutional neural network according to the acoustic topological feature map, constructing a gridding space demand matrix in combination with a monitoring equipment parameter library and coverage constraint conditions, and generating an optimal noise monitoring network site selection scheme according to the gridding space demand matrix; according to the invention, a quantum heuristic topology optimization algorithm is combined with quantum annealing to optimize the node weight of the three-dimensional topology network, and intelligent optimization and dynamic adjustment of the sound propagation path are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental noise monitoring, and in particular to a noise monitoring point location selection method and system based on a noise map. Background Art

[0002] Noise monitoring point site selection technology plays an important role in urban environmental noise control. At present, the traditional noise monitoring point site selection method is mainly based on empirical rules or simplified acoustic models, such as the use of equally spaced grid points or static site selection strategies based on two-dimensional noise maps. In recent years, with the development of BIM technology and GIS, some studies have begun to try to combine building geometry and sound propagation characteristics for three-dimensional noise modeling, and use heuristic algorithms to optimize the layout of monitoring points. However, the existing methods still have obvious deficiencies in the dynamic optimization of sound propagation paths, the extraction of multi-scale noise features, and the precise matching of monitoring equipment and sound field characteristics, resulting in problems such as uneven coverage of the monitoring network and low resource utilization.

[0003] The main defects of the existing technology are: on the one hand, the traditional site selection method is difficult to accurately reflect the complex impact of the building environment on sound propagation, especially the modeling accuracy of acoustic phenomena such as reflection and diffraction is insufficient; on the other hand, most optimization algorithms only consider the noise characteristics of a single scale, and lack the coordinated analysis of near-field direct sound, mid-field reflected waves and far-field attenuation gradients, resulting in the inability of monitoring points to fully capture the spatial heterogeneity of noise propagation. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a noise monitoring point location selection method based on noise map to solve the problems of insufficient accuracy of building environment sound propagation modeling and incomplete capture of noise spatial heterogeneity in the prior art.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a noise monitoring point site selection method based on a noise map, which includes obtaining sound source monitoring data and a building BIM model of a target area, and performing three-dimensional grid division on the spatial range of the target area to generate a three-dimensional noise intensity matrix; based on the three-dimensional noise intensity matrix, constructing a three-dimensional topological network, using a quantum-inspired topological optimization algorithm to calculate the optimal path of sound propagation, and combining quantum annealing to optimize the weights of nodes in the three-dimensional topological network to generate an acoustic topological feature map; according to the acoustic topological feature map, multi-scale noise propagation characteristics are extracted through a convolutional neural network, and a gridded space demand matrix is ​​constructed in combination with a monitoring equipment parameter library and coverage constraints, and an optimal noise monitoring network site selection plan is generated according to the gridded space demand matrix; based on the optimal noise monitoring network site selection plan, a feasibility assessment is performed through coverage verification, and the optimal noise monitoring network site selection plan is dynamically adjusted according to the assessment results.

[0007] As a preferred solution of the noise monitoring point location selection method based on noise map of the present invention, wherein: the acquisition process of the building BIM model is to generate a triangular mesh model through Poisson reconstruction; Use the NURBS algorithm to fit the building's exterior surface and generate a parametric surface model. Use the IFC geometry parsing engine to identify the geometric features of the building structure and generate a preliminary building BIM model. The acoustic parameters in the associated material library are input into the preliminary building BIM model to obtain the building BIM model.

[0008] As a preferred solution of the noise monitoring point location selection method based on the noise map of the present invention, the steps of generating a three-dimensional noise intensity matrix are as follows: According to the boundary parameters of the three-dimensional calculation domain, a Cartesian grid is established. Through ray collision detection, the grid lines in the Cartesian grid are adjusted to offset the outer surface of the building BIM model to generate a building-fitting grid. According to the building fitting grid and sound source monitoring data, the point sound source is associated to the nearest grid center point through the nearest neighbor interpolation algorithm, and the global sound pressure level distribution is established through the spatial interpolation method to generate a three-dimensional noise intensity matrix.

[0009] As a preferred solution of the noise monitoring point location selection method based on the noise map of the present invention, the steps of constructing a three-dimensional topological network are as follows: The grid unit coordinates and sound pressure level data in the noise intensity matrix are read through the COO format parser, and a three-dimensional grid point set with sound pressure level annotations is generated; According to the three-dimensional grid point set, the 26-neighborhood rule is used to connect the nodes in the three-dimensional grid point set to form an initial topological connection relationship diagram; The Dijkstra shortest path algorithm is used to calculate the gradient weights of the sound pressure levels of adjacent nodes in the initial topological connection graph, and the building obstacle edges are marked as inaccessible to generate a three-dimensional topological network.

[0010] As a preferred solution of the noise monitoring point location selection method based on the noise map described in the present invention, the quantum-inspired topological optimization algorithm is used to calculate the optimal path of sound propagation, and the weights of nodes in the three-dimensional topological network are optimized by quantum annealing to generate an acoustic topological feature map, and the steps are as follows: Calculate the gradient weights of the sound pressure levels of adjacent nodes in the initial topological connection relationship graph; According to the deterministic mapping rules, the nodes in the topological network are mapped to quantum bits and a quantum bit string is formed; Initialize all quantum bits to a uniform superposition state through Hadamard gate transformation and calculate the optimal path for sound propagation; The weights of nodes in the three-dimensional topological network are iteratively optimized through the quantum annealing algorithm, and combined with the optimal path of sound propagation to generate an acoustic topological feature map.

[0011] As a preferred solution of the noise monitoring point site selection method based on noise map described in the present invention, the extraction of multi-scale noise propagation characteristics through convolutional neural network refers to extracting the near-field characteristics of the sound source through small-scale convolution kernel, extracting the reflected wave characteristics through medium-scale void convolution, and extracting the sound wave diffraction attenuation characteristics of the building edge through separable convolution kernel.

[0012] As a preferred solution of the noise monitoring point location selection method based on the noise map of the present invention, the steps of combining the monitoring equipment parameter library and the coverage constraint condition to construct a gridded space demand matrix are as follows: Through the multi-scale sound field decomposition algorithm, the spatial distribution characteristics of multi-scale noise propagation characteristics are identified; The monitoring equipment is divided into high-frequency, medium-frequency and low-frequency equipment by frequency band division, and the monitoring equipment parameter library is defined according to the monitoring equipment in different frequency bands; Through the spatial grid discretization algorithm, the multi-scale noise propagation characteristics are converted into grid coverage indicators; According to the gridded coverage index, the coverage constraint conditions are generated by Kriging interpolation method; Based on the coverage constraints, the device deployment candidate set is converted into a gridded spatial demand matrix through ArcGIS.

[0013] As a preferred solution of the noise monitoring point location selection method based on the noise map of the present invention, wherein: the generation of the optimal noise monitoring network location selection plan refers to calculating the spatial layout score of the monitoring point by a nonlinear coverage contribution algorithm; Based on the spatial layout scores of monitoring points, the spatial layout set of monitoring points is iteratively optimized through genetic algorithm to generate the optimal noise monitoring network site selection plan.

[0014] As a preferred solution of the noise monitoring point site selection method based on the noise map of the present invention, wherein: the site selection scheme based on the optimal noise monitoring network is feasibility evaluated through coverage verification, and the optimal noise monitoring network site selection scheme is dynamically adjusted according to the evaluation results, the steps are as follows: Based on the coverage data in the equipment parameter library, the GIS spatial analysis method is used to analyze the coverage of each monitoring device on the grid unit, and the coverage score of the optimal noise monitoring network site selection plan is calculated through a weighted penalty function; A feasibility analysis was conducted based on the coverage score of the optimal noise monitoring network site selection scheme, and the spatial layout set of monitoring points was iteratively optimized using the particle swarm optimization algorithm.

[0015] In a second aspect, the present invention provides a noise monitoring point location selection system based on a noise map, comprising a matrix generation module, a map generation module, a location selection scheme generation module and an evaluation module; The matrix generation module is used to obtain the sound source monitoring data and building BIM model of the target area, and divide the spatial range of the target area into three-dimensional grids to generate a three-dimensional noise intensity matrix; The spectrum generation module is used to construct a three-dimensional topological network based on the three-dimensional noise intensity matrix, use a quantum-inspired topological optimization algorithm to calculate the optimal path for sound propagation, and combine quantum annealing to optimize the weights of nodes in the three-dimensional topological network to generate an acoustic topological feature spectrum; The site selection plan generation module is used to extract multi-scale noise propagation characteristics through convolutional neural networks based on the acoustic topology feature map, and to build a gridded space demand matrix based on the monitoring equipment parameter library and coverage constraints. Based on the gridded space demand matrix, the optimal noise monitoring network site selection plan is generated; The evaluation module is used to conduct feasibility evaluation based on the optimal noise monitoring network site selection plan through coverage verification, and dynamically adjust the optimal noise monitoring network site selection plan according to the evaluation results.

[0016] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the method for selecting a noise monitoring point based on a noise map as described in the first aspect of the present invention is implemented.

[0017] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the method for selecting a noise monitoring point location based on a noise map as described in the first aspect of the present invention is implemented.

[0018] The beneficial effects of the present invention are as follows: by combining a quantum-inspired topology optimization algorithm with quantum annealing to optimize the weights of three-dimensional topological network nodes, intelligent optimization and dynamic adjustment of sound propagation paths are achieved, effectively capturing the complex influence of the building environment on sound wave reflection and diffraction, so that the generated acoustic topological feature map can accurately characterize the sound field propagation characteristics; at the same time, through the convolutional neural network multi-scale feature extraction technology, small-scale convolution kernels, medium-scale void convolutions and separable convolution kernels are used to collaboratively analyze the near-field direct sound, reflected wave and diffraction attenuation characteristics, thereby achieving a refined analysis of the spatial heterogeneity of noise propagation. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0020] Figure 1 This is a flow chart of the noise monitoring point location selection method based on the noise map in Example 1.

[0021] Figure 2 Schematic diagram of the noise monitoring point location selection system based on the noise map in Example 1.

[0022] Figure 3 This is a schematic diagram of the three-dimensional topological network construction process in Example 1.

[0023] Figure 4 This is a flowchart of the execution of the quantum heuristic optimization and annealing algorithm in Example 1. DETAILED DESCRIPTION

[0024] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0025] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0026] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0027] Example 1, reference Figure 1~Figure 4 This embodiment provides a noise monitoring point location selection method based on a noise map, comprising the following steps: Step 1: Obtain the sound source monitoring data and building BIM model of the target area, divide the spatial range of the target area into three-dimensional grids, and generate a three-dimensional noise intensity matrix; The sound source monitoring data includes the time history record of sound pressure level, spectrum characteristics, spatial location of sound source and environmental parameters.

[0028] It should be noted that the sound pressure level time history record is collected by a first-level sound level meter that complies with the IEC 61672-1 standard, and the time series waveform data is recorded using the A / C / Z weighting mode; the spectrum characteristics are obtained by a real-time spectrum analyzer, and the frequency domain energy distribution is calculated based on 1 / 3 octave or narrow-band Fourier transform; the spatial position of the sound source is measured by a multi-channel microphone array, and the three-dimensional coordinates and propagation direction angle are determined by combining the beamforming algorithm and time difference positioning technology; environmental parameters are synchronously collected by an integrated temperature and humidity sensor and anemometer, and background noise data is measured and recorded by a sound level meter when the sound source is silent. All data are marked with metadata such as sensor model, calibration date, and sampling configuration.

[0029] The high-density point cloud data of the target building is obtained by using a ground-based 3D laser scanner. Based on the high-density point cloud data, a statistical outlier removal algorithm is used to remove noise points, and a triangular mesh model is generated through Poisson reconstruction. Furthermore, high-density point cloud data of the target building is obtained through the scanning resolution of the ground 3D laser scanner and the layout plan of the scanning station. After the high-density point cloud data is imported into the point cloud processing software, multi-site cloud alignment is first performed, and the iterative nearest point algorithm is used to achieve millimeter-level precision alignment; the high-density point cloud data after alignment adopts a statistical outlier removal algorithm based on k-nearest neighbors, and the neighborhood radius and standard deviation multiple thresholds are set to identify and remove discrete noise points and flying points; the denoised high-density point cloud data is processed by the Poisson surface reconstruction algorithm, and the octree depth and sampling weight parameters are set to generate a triangular mesh model with a continuous topological structure and retaining the geometric characteristics of the building. The vertex normal vectors of the triangular mesh model are obtained by local plane fitting calculation.

[0030] Based on the triangular mesh model, the NURBS algorithm is used to fit the building's exterior surface to generate a parametric surface model. The geometric features of the building structure are identified through the IFC geometry parsing engine to generate a preliminary building BIM model (Building Information Model). Furthermore, based on the triangular mesh model, the NURBS algorithm is first used for surface fitting. By setting the number of control points (no less than 16 per square meter) and the node vector parameters in the triangular mesh model, the discrete triangular facets are converted into continuous NURBS surface patches. The least squares method is used to optimize the geometric deviation between the surface and the original mesh during the fitting process (controlled within ±5mm). After the generated parametric surface model is imported into the BIM software environment, the IFC geometry parsing engine automatically recognizes the geometric features in the NURBS surface: for the plane area (curvature < 0.001 / m), it is directly converted to IfcWallStandardCase entity; for the double curvature surface (such as the dome), it is converted to IfcCurtainWall and the NURBS parametric definition is retained; for the intersection area of ​​the structural components, the IfcConnectionGeometry connection relationship is generated through spatial Boolean operations. All the identified geometric features are assigned entity type attributes (such as IfcWall / IfcSlab) according to the IFC4.3 standard, and the material partition information of the original triangular mesh model is inherited. The resulting preliminary building BIM model contains a complete building component classification tree (including LOD350 level of geometric detail) and a property set that complies with the ISO 16739 specification.

[0031] Through the transfer matrix method, the acoustic parameters in the associated material library are obtained and input into the preliminary building BIM model to obtain the building BIM model.

[0032] Furthermore, the propagation characteristics of sound waves in multi-layer media are calculated by the transfer matrix method. Based on the sound pressure level attenuation curve measured in the reverberation chamber and the normal incidence sound absorption coefficient measured by the impedance tube, the acoustic parameters such as dynamic flow resistivity, porosity and tortuosity of the material are inverted. After the acoustic parameters are matched and verified with the standard material records in the material library, they are input into the preliminary building BIM model in the form of attribute sets. The specific operations are as follows: add acoustic parameter attributes in the IfcMaterial resource definition, configure the acoustic transmission characteristics of the multi-layer structure in IfcMaterialLayerSetUsage, and bind the acoustic parameters to the building components in the BIM model through the IfcRelAssociatesMaterial relationship entity. The final building BIM model contains a complete definition of acoustic material properties. The acoustic parameters of each component are marked with the measurement error range (such as ±5%), and meet the attribute extension specifications in the ISO 16739-1 standard.

[0033] It should be noted that the acoustic parameters in the associated material library refer to the inherent properties such as sound absorption coefficient and sound insulation value assigned to the building components (such as walls and glass) in the building BIM model.

[0034] Based on the sound source monitoring data of the building BIM model, define the spatial scope and generate the three-dimensional calculation domain boundary parameters; Furthermore, based on the geometric data of the exterior wall components of the building BIM model and the coordinate data of the monitoring points in the sound source monitoring data, the following steps are performed to generate the three-dimensional calculation domain boundary parameters: first, the vertex coordinates of all the exterior wall components in the building BIM model are extracted to calculate the building outer contour polygon; then, combined with the spatial position coordinates of each monitoring point in the sound source monitoring data, the horizontal boundary range is determined by expanding 50 meters outward based on the building outer contour polygon; in the vertical direction, the vertical boundary range is taken from the bottom elevation of the building BIM model to the highest elevation plus 20 meters; finally, the three-dimensional calculation domain boundary parameters including the minimum and maximum values ​​of the X-axis, the minimum and maximum values ​​of the Y-axis, and the minimum and maximum values ​​of the Z-axis are output.

[0035] According to the boundary parameters of the three-dimensional calculation domain, a Cartesian grid is established. Through ray collision detection, the grid lines in the Cartesian grid are adjusted to offset the outer surface of the building BIM model to generate a building-fitting grid. Furthermore, according to the maximum and minimum values ​​of the X / Y / Z axes in the boundary parameters of the three-dimensional calculation domain, an initial Cartesian grid is established, and the grid line spacing is evenly distributed according to the preset resolution. The outer surface geometric data of the building BIM model is converted into a set of triangular facets, and ray collision detection is performed on each Cartesian grid line: a detection ray parallel to the coordinate axis is emitted from the grid line endpoint, the intersection with the triangular facet of the building BIM model is calculated, and the grid line endpoint is adjusted to 0.3 meters away from the nearest intersection. Repeat the above process until all grid lines are detected, and finally generate a building fit grid that maintains an offset distance of 0.3 meters from the outer surface of the building BIM model. The grid line is automatically disconnected in the internal area of ​​the building BIM model to form a cavity. In the processing process, the bounding box acceleration algorithm is used to optimize the efficiency of collision detection to ensure that the building fit grid strictly follows the range constraints of the boundary parameters of the three-dimensional calculation domain.

[0036] According to the building fitting grid and sound source monitoring data, the point sound source is associated to the nearest grid center point through the nearest neighbor interpolation algorithm, and the global sound pressure level distribution is established through the spatial interpolation method to generate a three-dimensional noise intensity matrix.

[0037] Furthermore, a spatial index structure of the center point of each unit of the building-fitting grid is established, and the nearest neighbor interpolation algorithm is used to accelerate the nearest neighbor search; the coordinates of the point sound source in the sound source monitoring data are matched with the center point of the grid unit, and the nearest grid unit corresponding to each point sound source is determined by Euclidean distance calculation, and the sound pressure level value is recorded. After completing the sound source matching, the grid units that are not directly associated with the sound source are processed based on the Kriging spatial interpolation method: a 50-meter range and a 0.2 nugget value parameter are set; the Kriging equation group is solved to obtain the weight coefficient of each monitoring point, and the sound pressure level estimate of the unsampled grid unit is calculated. The final output is a three-dimensional noise intensity matrix containing the sound pressure level data of all grid units. Each element in the matrix corresponds to the A-weighted sound pressure level value at a specific location of the building-fitting grid, and the data storage uses the COO sparse format to record non-zero elements. The processing process is strictly based on the original measurement values ​​of the sound source monitoring data and the geometric structure of the building-fitting grid to ensure that the spatial distribution characteristics of the three-dimensional noise intensity matrix are consistent with the actual situation.

[0038] Step 2: Based on the three-dimensional noise intensity matrix, a three-dimensional topological network is constructed, and the optimal path of sound propagation is calculated using a quantum-inspired topological optimization algorithm. The weights of the nodes in the three-dimensional topological network are optimized by combining quantum annealing to generate an acoustic topological feature map. The grid unit coordinates and sound pressure level data in the noise intensity matrix are read through the COO format parser, and a three-dimensional grid point set with sound pressure level annotations is generated; Furthermore, the COO format parser reads the non-zero elements of the noise intensity matrix row by row, extracts the row index of each non-zero element as the horizontal position coordinate of the grid cell, the column index as the vertical position coordinate of the grid cell, and the numerical value as the sound pressure level data; the horizontal position coordinate, the vertical position coordinate and the preset height reference value are combined to form a spatial position point, and the sound pressure level data is associated with the corresponding spatial position point as an attribute annotation, and finally outputs a three-dimensional grid point set consisting of spatial position points and sound pressure level annotations.

[0039] According to the three-dimensional grid point set, the 26-neighborhood rule is used to connect the nodes in the three-dimensional grid point set to form an initial topological connection relationship diagram; Furthermore, based on the spatial coordinate data of the three-dimensional grid point set, the nearest neighbor node is searched for in 26 directions, such as front and back, left and right, and up and down, for each node in the three-dimensional grid point set within the spatial cube. When the distance between two nodes is less than or equal to the grid unit benchmark distance, a bidirectional connection relationship is established, and the starting and ending coordinates of the connecting line segment are recorded, and finally an initial topological connection relationship diagram consisting of all valid connection relationships is generated. The connection process strictly follows the 26-neighborhood rule, and only the existing node coordinates and sound pressure level data in the three-dimensional grid point set are used for calculation, without adding additional attributes or constraints.

[0040] The Dijkstra shortest path algorithm is used to calculate the gradient weights of the sound pressure levels of adjacent nodes in the initial topological connection relationship graph, and the building obstacle edges are marked as inaccessible to generate a three-dimensional topological network. Furthermore, in the initial topological connection relationship diagram, the nodes of the three-dimensional grid point set are taken as vertices, and the absolute values ​​of the sound pressure level differences between adjacent nodes are calculated as edge weights; the edges corresponding to the locations of building obstacles are identified; the Dijkstra shortest path algorithm (Dijkstra algorithm) takes each vertex as the starting point, and iteratively calculates the shortest path to all other vertices, and the path length is obtained by accumulating the edge weights; and finally outputs a three-dimensional topological network containing vertex coordinates, sound pressure level data, edge connection relationships and shortest path information.

[0041] It should be noted that a building obstacle edge refers to an edge that connects two nodes in a three-dimensional topological network and passes through the physical components of the building BIM model. The identification process of the building obstacle edge is as follows: first, the geometric data of the triangular facets of all physical components such as IfcWall and IfcSlab in the building BIM model are extracted to construct an axial bounding box acceleration structure; then, the intersection detection between the line segment and the triangular facet is performed on each edge in the topological network. When the number of intersections between the straight line segment of the edge and the triangular facet of the building BIM model is an odd number, it is determined to be a building obstacle edge. The identification process uses the separating axis theorem for precise collision detection to ensure that the result is completely consistent with the geometric boundary of the building BIM model.

[0042] Calculate the gradient weight of the sound pressure level of adjacent nodes in the initial topological connection relationship graph. The expression is: ; in, For Node and nodes The adjacent sound pressure weights, For Node The A-weighted sound pressure level value, For Node The A-weighted sound pressure level value, Is a node The three-dimensional coordinate vector of Is a node The three-dimensional coordinate vector of is a very small constant that prevents zero distance, Represents a single node index in a three-dimensional topological network, j is The index of the adjacent node.

[0043] It should be noted that the process of calculating the adjacent node sound pressure level gradient weight is as follows: for each pair of adjacent nodes in the initial topological connection relationship graph, first obtain the node The three-dimensional coordinate vector of and nodes The three-dimensional coordinate vector of , calculate the Euclidean distance between the two ; Read nodes at the same time A-weighted sound pressure level and nodes A-weighted sound pressure level , find the absolute difference ; The ratio of the sound pressure level difference to the distance is used as the basic weight, and then the minimum constant ϵ to prevent zero distance is superimposed, and finally the node is obtained and nodes The adjacent sound pressure weights .

[0044] After the nodes of the three-dimensional topological network are numbered in order of spatial coordinates, a deterministic mapping rule is established for a one-to-one correspondence between the node index and the quantum bit register, and each node in the three-dimensional topological network is mapped to a quantum bit according to the deterministic mapping rule to form a quantum bit string; Furthermore, the nodes of the three-dimensional topological network are primarily sorted according to the horizontal component of the spatial position, the vertical component is secondarily sorted, and the height component is finally sorted, so as to establish a strictly ordered node numbering sequence; the node numbering range is determined according to the capacity of the quantum bit register, and a linear mapping relationship between the node index and the quantum bit register position is established, in which the node number directly corresponds to the physical position index of the quantum bit register; according to the sorted node numbering order, each three-dimensional topological network node is mapped to a quantum bit in the quantum bit register in turn, and the correspondence between the node attribute and the quantum bit state is maintained; and finally a quantum bit string is formed.

[0045] Based on the quantum bit string, all quantum bits are initialized to a uniform superposition state through Hadamard gate transformation, and the sound propagation path score is calculated. The expression is: ; in, is the sound propagation path score, is the distance penalty coefficient (default 0.1), which is used to balance the weight of sound pressure level and path length, and is used to identify and traverse the path The specific location point on It is a candidate path from the starting point to the end point, consisting of a series of continuously connected nodes in a three-dimensional topological network; It should be noted that based on the quantum bit string, the Hadamard gate transformation is first applied to all quantum bits to make the quantum bits in a uniform superposition state of state 0 and state 1; then all possible path combinations represented by the quantum bit string are traversed, and for each candidate path P, all nodes on the path are calculated. A-weighted sound pressure level The sum of the adjacent nodes in the path and The Euclidean distance between ; Multiply the sum of the sound pressure levels and the total path length by the distance penalty factor Add the results of The quantum state amplitude corresponding to the optimal path is enhanced by the quantum amplitude amplification algorithm, and finally the measurement is obtained. The shortest sound propagation path is used as the optimal path for sound propagation.

[0046] A-weighted sound pressure level of nodes based on three-dimensional topology network and adjacent sound pressure weights , generating weighted network structure data through adjacency list data structure conversion; It should be noted that each node in the three-dimensional topological network is stored as a vertex element of the adjacency list, and the vertex attribute records the A-weighted sound pressure level value of the node ; Traverse all valid connection edges in the initial topological connection graph, and for each pair of adjacent nodes in the adjacency table and Create edge records, and the edge attributes store the adjacent sound pressure weights ; For the connection relationship corresponding to the building obstacle edge, the weight of the connection relationship corresponding to the building obstacle edge is marked as infinity in the adjacency table; the weighted network structure data finally generated contains a vertex set, an edge set and the corresponding sound pressure level value and weight parameter, which completely retains the geometric connection relationship and acoustic characteristics of the three-dimensional topological network.

[0047] Initialize quantum annealing parameters based on weighted network structure data; Furthermore, the number of quantum bits required for the quantum annealing algorithm is determined based on the number of vertices and connection relationships in the weighted network structure data, and the adjacent sound pressure weights in the weighted network structure data are used to calculate the number of quantum bits required for the quantum annealing algorithm. and building obstacle edge markers, construct the Ising model Hamiltonian that describes the sound propagation path optimization problem, where the coupling term coefficient is given by Determine, the local field term coefficient is determined by the vertex sound pressure level value Determine; for example, set the initial temperature of quantum annealing to the maximum value in the weighted network structure data 10 times the value, the termination temperature is the minimum The annealing time is set to 20 microseconds according to the characteristics of the quantum processor. This step realizes the accurate mapping of the sound propagation path optimization problem to the quantum annealing calculation model, providing standardized input parameters for subsequent quantum annealing calculations.

[0048] According to the initialized quantum annealing parameters, the weights of the nodes in the three-dimensional topological network are iteratively optimized through the quantum annealing algorithm, and combined with the optimal path of sound propagation, an acoustic topological feature map is generated.

[0049] Furthermore, based on the weighted network structure data, we first determine the computing scale required for quantum annealing, and match the number of quantum bits with the scale of network vertices. Then, we construct a physical model that describes the characteristics of sound propagation, in which the network connection relationship and acoustic parameters are directly converted into model parameters. We set the temperature parameters of the annealing process, with the initial value determined by the maximum connection strength of the network and the termination value determined by the minimum connection strength. We iteratively optimize the network node parameters through the quantum annealing process, and adjust the connection weights while maintaining the original network topology. Finally, we fuse the optimization results with the sound propagation path data to form a topological feature map that reflects the sound field distribution and propagation characteristics.

[0050] Step 3: Based on the acoustic topological feature map, the multi-scale noise propagation characteristics are extracted through the convolutional neural network, and the gridded space demand matrix is ​​constructed in combination with the monitoring equipment parameter library and the coverage constraint conditions. Based on the gridded space demand matrix, the optimal noise monitoring network site selection plan is generated; The multi-scale noise propagation characteristics include the near-field characteristics of the sound source, the reflection wave characteristics, and the sound wave diffraction attenuation characteristics. The extraction process is as follows: Based on the acoustic topological feature map, the near-field characteristics of the sound source are extracted through small-scale convolution kernels, the reflected wave characteristics are extracted through medium-scale hole convolution, and the sound wave diffraction attenuation characteristics of the building edge are extracted through separable convolution kernels. Furthermore, in the process of processing the acoustic topological feature map, the map is first scanned densely using a 3×3 small-scale convolution kernel, and the sliding calculation is performed in the near-field area of ​​the sound source with a step size of 0.5 meters to extract the local detail features of the rapid change of the sound pressure level; then a 5×5 medium-scale hole convolution kernel with an expansion rate of 2 is used to perform sparse sampling in the reflection area at intervals of 1 meter to capture the periodic intensity distribution pattern formed by the reflected wave; at the same time, a deep separable convolution kernel is applied to perform directional convolution operations along the edge contour of the building, where a 3×1 horizontal convolution kernel detects lateral diffraction attenuation and a 1×3 vertical convolution kernel detects longitudinal diffraction attenuation. The three convolution operations use the same input data but are performed independently. The small-scale convolution kernel focuses on the 0-5 meter near-field area, the medium-scale hole convolution covers the 5-20 meter reflection area, and the separable convolution kernel specifically processes data within 0.3 meters of the building edge. All convolution results maintain the same spatial reference system and coordinate accuracy as the original acoustic topological feature map. The final output is a three-channel feature map containing near-field details, reflection characteristics, and diffraction attenuation, which fully retains the multi-scale physical characteristics of sound field propagation.

[0051] Through the multi-scale sound field decomposition algorithm, the spatial distribution characteristics of multi-scale noise propagation characteristics are identified; It should be noted that during the execution of the multi-scale sound field decomposition algorithm, the acoustic topological feature map is sent to three parallel processing channels for feature extraction. The first channel uses a Gaussian difference filter group to process the near field area, and extracts the radiation characteristics of the point sound source through Gaussian kernel operations of different scales; the second channel uses a multi-directional Gabor filter group to scan the mid-field area, and uses a direction-sensitive filter kernel to capture the propagation mode of the reflected wave; the third channel processes the far field area based on the signal time-frequency analysis method, and separates the environmental noise component through non-stationary signal decomposition technology. After the processing results of each channel are spatially aligned and feature fused, a feature distribution map containing three types of propagation mechanisms, namely direct sound, reflected sound and environmental noise, is generated, which fully characterizes the spatial distribution law of the propagation characteristics of the sound field at different scales.

[0052] Based on the spatial distribution characteristics of multi-scale noise propagation features, the monitoring equipment is divided into high-frequency, medium-frequency and low-frequency equipment through frequency band division, and a monitoring equipment parameter library containing monitoring equipment in different frequency bands is defined; Furthermore, based on the spatial distribution characteristics of multi-scale noise propagation features, the monitoring equipment is first divided into three categories: high-frequency monitoring equipment, medium-frequency monitoring equipment and low-frequency monitoring equipment according to the distribution range of frequency components in the sound field decomposition results; the configuration parameters of high-frequency monitoring equipment focus on the ability to capture rapidly changing near-field sound pressure, the parameters of medium-frequency monitoring equipment emphasize the steady-state measurement accuracy of the reflected sound field, and the parameters of low-frequency monitoring equipment optimize the long-term environmental noise monitoring performance; the technical specifications, frequency response range, sensitivity indicators and other parameters of the three types of monitoring equipment are structured and stored to form a monitoring equipment parameter library containing monitoring equipment in different frequency bands.

[0053] Based on the spatial distribution characteristics of multi-scale noise propagation features, the near-field sound source concentration area, the reflected wave interference area and the far-field attenuation gradient area are marked through feature intensity clustering analysis; Furthermore, in the analysis of the spatial distribution characteristics of multi-scale noise propagation characteristics, the three groups of data, namely, the near-field characteristics of the sound source, the reflected wave characteristics, and the sound wave diffraction attenuation characteristics, are first standardized, and the spectral clustering algorithm is used to divide the spatial area into three categories: the first type of area is dominated by the near-field characteristics of the sound source with high sound pressure level gradient, marked as the near-field sound source concentration area; the second type of area is dominated by the reflected wave characteristics with periodic intensity changes, marked as the reflected wave interference area; the third type of area is marked by the continuously attenuated sound wave diffraction attenuation characteristics, marked as the far-field attenuation gradient area. This step realizes the accurate division of noise propagation areas based on multi-physical characteristics.

[0054] According to the near-field sound source concentration area, reflected wave interference area and far-field attenuation gradient area, the multi-scale noise propagation characteristics are converted into grid coverage indicators through the spatial grid discretization algorithm; Furthermore, based on the spatial distribution characteristics of the near-field sound source concentration area, the reflected wave interference area, and the far-field attenuation gradient area, the monitoring area is discretized into regular grid units using an adaptive grid division method. Each grid unit is assigned a corresponding feature weight value according to the characteristic intensity ratio of the near-field sound source concentration area, the reflected wave interference area, and the far-field attenuation gradient area, and the noise coverage index value of the regular grid unit is obtained through weighted fusion calculation.

[0055] According to the gridded coverage index, the coverage constraints are generated by the Kriging interpolation method.

[0056] Furthermore, based on the gridded coverage index matrix, the ordinary Kriging interpolation method is used to calculate the semivariogram of the coverage index values ​​between grid cells, and the spherical model is fitted to determine the range, sill value and nugget effect parameters; then the Kriging equation group is established, and the coverage index values ​​of the known grid cells are used to solve the optimal linear unbiased estimate of the point to be estimated; finally, the coverage constraint conditions are generated.

[0057] The coverage constraints that should be explained include the quantitative requirements for the coverage of noise propagation characteristics at each location in the monitoring area, which are used to guide the optimal deployment of monitoring equipment. For example, at the boundary of an industrial plant, the noise coverage index is required to be no less than 0.8 to ensure complete monitoring of the main noise sources.

[0058] Based on the monitoring equipment parameter library, high-frequency, medium-frequency and low-frequency devices are matched to the near-field sound source concentration area, the reflected wave interference area and the far-field attenuation gradient area respectively to generate a candidate set of equipment deployment; Furthermore, according to the technical specifications of high-frequency monitoring equipment, medium-frequency monitoring equipment and low-frequency monitoring equipment in the monitoring equipment parameter library, the high-frequency monitoring equipment is deployed to the near-field sound source concentration area to capture the rapidly changing near-field noise characteristics, the medium-frequency monitoring equipment is arranged in the reflected wave interference area to monitor the steady-state noise formed by the reflected wave, and the low-frequency monitoring equipment is arranged in the far-field attenuation gradient area to collect long-distance propagation environmental noise; by traversing all possible installation positions within the three types of characteristic areas, different configuration schemes of high-frequency monitoring equipment, medium-frequency monitoring equipment and low-frequency monitoring equipment are combined to form a device deployment candidate set containing complete information such as equipment type, installation location and technical parameters.

[0059] Based on the coverage constraint, the equipment deployment candidate set is converted into a gridded spatial demand matrix through ArcGIS; Furthermore, in the ArcGIS (Geographic Information System) platform, the spatial location coordinates of the equipment deployment candidate set and the coverage constraint conditions are superimposed and analyzed, and the location of each candidate device is mapped to the corresponding grid unit; based on the technical parameters of the high-frequency monitoring equipment, medium-frequency monitoring equipment, and low-frequency monitoring equipment in the equipment deployment candidate set, the minimum number of equipment configurations required for each grid unit to meet the coverage constraint conditions is calculated; through spatial connection and attribute calculation, a gridded spatial demand matrix containing grid unit ID, coordinate range, equipment type requirements, and coverage compliance is generated. The matrix data maintains the same spatial reference as the original coverage constraint conditions, accurately reflecting the specific needs of each region for monitoring equipment.

[0060] Based on the gridded space demand matrix, the spatial layout score of the monitoring points is calculated using a nonlinear coverage contribution algorithm. , the expression is: ; in, is the gridded space requirement matrix, Represents the gridded space requirement matrix The index variable of the grid cell in , Indicates The demand intensity of each grid unit, Indicates The spatial coordinates of the center of the grid cell, Indicates The detection level of each monitoring device, is the cost penalty coefficient (default 0.3), The spatial layout of monitoring points. It is The location coordinates of each monitoring device; Furthermore, in calculating the spatial layout score of monitoring points When traversing the gridded space demand matrix Each grid cell in , get the grid unit required strength and center coordinates ; For the spatial layout set of monitoring points Each monitoring device in , calculate the position coordinates With the grid cell center The square of the Euclidean distance, combined with the detection level of the monitoring equipment and the minimum constant , get the device For grid cells The contribution of all monitoring devices to the same grid unit is summed up, normalized by the hyperbolic tangent function, and then multiplied by the grid unit demand intensity. ; After accumulating the calculation results of all grid cells, subtract the spatial layout set of monitoring points Total cost and cost penalty coefficient The product of , finally get the layout score The calculation process strictly uses the gridded space requirement matrix and monitoring point spatial layout set The original data of the layout is used to ensure that the scoring results accurately reflect the coverage effect and economy of the layout plan.

[0061] Scoring based on spatial layout of monitoring points ,The spatial layout set of monitoring points is iteratively optimized through genetic algorithms to generate the optimal noise monitoring network site selection plan.

[0062] Furthermore, the genetic algorithm optimization process first initializes a population containing multiple monitoring point spatial layout sets, and each individual represents a possible equipment distribution plan. During the iteration process, the monitoring point spatial layout score corresponding to each monitoring point spatial layout set is calculated. Based on the scoring results, the equipment location information in different monitoring point spatial layout sets is exchanged through a single-point crossover operation, and the Gaussian mutation operation is used to randomly perturb some equipment coordinates. After a preset number of selection, crossover and mutation operations, the monitoring point spatial layout set with the highest monitoring point spatial layout score is output as the optimal noise monitoring network site selection plan. The optimization process relies entirely on the mathematical definition of the monitoring point spatial layout score and the standard genetic operator to ensure that the final site selection plan achieves the optimal balance between coverage performance and deployment cost.

[0063] Step 4: Based on the optimal noise monitoring network site selection plan, conduct a feasibility assessment through coverage verification, and dynamically adjust the optimal noise monitoring network site selection plan according to the evaluation results.

[0064] Based on the coverage capacity data in the equipment parameter library, the GIS spatial analysis method is used to analyze the coverage of each monitoring device on the grid unit, and the coverage score of the optimal noise monitoring network site selection scheme is calculated through the weighted penalty function. The expression is: ; in, is the coverage score of the optimal noise monitoring network site selection plan, Indicates The weight coefficient of the grid cell, Indicates The coverage standard of grid cells is It is The effective coverage radius of each monitoring device; Furthermore, in calculating the coverage score of the optimal noise monitoring network site selection plan When traversing the gridded space demand matrix Each grid cell in , get the grid cell weight coefficient and coverage criteria ; For the spatial layout set of monitoring points Each monitoring device in , determine the position coordinates With the grid cell center Is the Euclidean distance within the effective coverage radius of the device? Within the range, the coverage is recorded through the indicator function; the total coverage of all monitoring devices on the grid units Calculate the coverage gap value of the grid cell in the gridded space demand matrix , when the result is negative, take zero; square the gap value of each grid unit and multiply it by the weight coefficient , the calculation results of all grid cells are accumulated to get the final coverage score ; The calculation process strictly uses the gridded space requirement matrix and monitoring point spatial layout set The original parameters of the layout are used to ensure that the scoring results accurately reflect the actual coverage gap of the layout plan.

[0065] Based on historical monitoring data, the feasibility threshold H1 is defined; Furthermore, when defining the feasibility threshold H1, first collect all valid noise monitoring records in the historical monitoring data, and extract the sound pressure level measurement value sequence of each monitoring point; perform percentile analysis on the historical monitoring data, for example, select the 95th percentile value as the benchmark reference value; combine the environmental noise standard limit and the technical parameters of the monitoring equipment, and increase the safety margin on the basis of the benchmark reference value. In the example, 3dB is used as the typical safety margin value, and finally determine the specific value of the feasibility threshold H1.

[0066] when When ≥H1, the current optimal noise monitoring network site selection scheme is considered feasible; when When When the current optimal noise monitoring network site selection scheme is feasible, the spatial layout set of monitoring points is iteratively optimized through the particle swarm optimization algorithm until it meets the requirements. ≥H1 or the maximum number of iterations is reached.

[0067] Furthermore, during the execution of the particle swarm optimization algorithm, a particle swarm containing multiple monitoring point spatial layout sets is first initialized. In the example, the number of particles is set to 50, and each particle represents a possible monitoring point spatial layout set.​ The initial position and velocity are randomly assigned. The coverage score of each particle is calculated at each iteration. , for example when When ≥H1, the optimization is terminated immediately; otherwise, the particle speed and position are updated according to the optimal position of the individual particle and the optimal position of the group. In the example, the inertia weight decreases linearly from 0.9 to 0.4. The iteration process continues until the coverage score is When the feasibility threshold H1 is reached or the preset maximum number of iterations (set to 200 in this example) is reached, the final output satisfies The optimal spatial layout set of monitoring points ≥H1.

[0068] This embodiment also provides a noise monitoring point location selection system based on a noise map, comprising: a matrix generation module, a map generation module, a location selection scheme generation module and an evaluation module; The matrix generation module is used to obtain the sound source monitoring data and building BIM model of the target area, and divide the spatial range of the target area into three-dimensional grids to generate a three-dimensional noise intensity matrix; The spectrum generation module is used to construct a three-dimensional topological network based on the three-dimensional noise intensity matrix, use a quantum-inspired topological optimization algorithm to calculate the optimal path of sound propagation, and combine quantum annealing to optimize the weights of nodes in the three-dimensional topological network to generate an acoustic topological feature spectrum; The site selection plan generation module is used to extract multi-scale noise propagation characteristics through convolutional neural networks based on the acoustic topology feature map, and to build a gridded space demand matrix based on the monitoring equipment parameter library and coverage constraints. Based on the gridded space demand matrix, the optimal noise monitoring network site selection plan is generated; The evaluation module is used to conduct feasibility evaluation based on the optimal noise monitoring network site selection plan through coverage verification, and dynamically adjust the optimal noise monitoring network site selection plan according to the evaluation results.

[0069] This embodiment also provides a computer device, which is suitable for the method of noise monitoring point site selection based on noise map, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the method of noise monitoring point site selection based on noise map as proposed in the above embodiment.

[0070] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0071] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for selecting a noise monitoring point based on a noise map as proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic storage device, a flash memory, a magnetic disk or an optical disk.

[0072] In summary, the present invention realizes intelligent optimization and dynamic adjustment of sound propagation paths by combining a quantum-inspired topology optimization algorithm with quantum annealing to optimize the weights of three-dimensional topological network nodes, effectively capturing the complex influence of the building environment on sound wave reflection and diffraction, so that the generated acoustic topological feature map can accurately characterize the propagation characteristics of the sound field; at the same time, through the convolutional neural network multi-scale feature extraction technology, small-scale convolution kernels, medium-scale hole convolutions and separable convolution kernels are used to collaboratively analyze the near-field direct sound, reflected wave and diffraction attenuation characteristics, thereby realizing a refined analysis of the spatial heterogeneity of noise propagation.

[0073] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A noise monitoring point location selection method based on a noise map, characterized in that: include, Acquire the sound source monitoring data and building BIM model of the target area, and perform three-dimensional grid division on the spatial range of the target area to generate a three-dimensional noise intensity matrix. The sound source monitoring data includes the time history record of the sound pressure level, the spectrum characteristics, the spatial position of the sound source and the environmental parameters. Based on the three-dimensional noise intensity matrix, a three-dimensional topological network is constructed, and the optimal path of sound propagation is calculated using a quantum-inspired topological optimization algorithm. The weights of the nodes in the three-dimensional topological network are optimized by combining quantum annealing to generate an acoustic topological feature map. According to the acoustic topological feature map, the multi-scale noise propagation characteristics are extracted through the convolutional neural network, and the grid space demand matrix is ​​constructed by combining the monitoring equipment parameter library and the coverage constraint conditions. According to the grid space demand matrix, the optimal noise monitoring network site selection plan is generated; Based on the optimal noise monitoring network site selection plan, a feasibility assessment is carried out through coverage verification, and the optimal noise monitoring network site selection plan is dynamically adjusted according to the evaluation results.

2. The noise monitoring point location selection method based on noise map according to claim 1 is characterized in that: The acquisition process of the building BIM model is to generate a triangular mesh model through Poisson reconstruction; Use the NURBS algorithm to fit the building's exterior surface and generate a parametric surface model. Use the IFC geometry parsing engine to identify the geometric features of the building structure and generate a preliminary building BIM model. The acoustic parameters in the associated material library are input into the preliminary building BIM model to obtain the building BIM model.

3. The noise monitoring point location selection method based on noise map according to claim 2 is characterized in that: The steps of generating a three-dimensional noise intensity matrix are as follows: According to the boundary parameters of the three-dimensional calculation domain, a Cartesian grid is established. Through ray collision detection, the grid lines in the Cartesian grid are adjusted to offset the outer surface of the building BIM model to generate a building-fitting grid. According to the building fitting grid and sound source monitoring data, the point sound source is associated to the nearest grid center point through the nearest neighbor interpolation algorithm, and the global sound pressure level distribution is established through the spatial interpolation method to generate a three-dimensional noise intensity matrix.

4. The noise monitoring point location selection method based on noise map according to claim 1 is characterized in that: The steps of constructing a three-dimensional topological network are as follows: The grid unit coordinates and sound pressure level data in the noise intensity matrix are read through the COO format parser, and a three-dimensional grid point set with sound pressure level annotations is generated; According to the three-dimensional grid point set, the 26-neighborhood rule is used to connect the nodes in the three-dimensional grid point set to form an initial topological connection relationship diagram; The Dijkstra shortest path algorithm is used to calculate the gradient weights of the sound pressure levels of adjacent nodes in the initial topological connection graph, and the building obstacle edges are marked as inaccessible to generate a three-dimensional topological network.

5. The noise monitoring point location selection method based on noise map according to claim 4 is characterized in that: The quantum-inspired topology optimization algorithm is used to calculate the optimal path for sound propagation, and the weights of nodes in the three-dimensional topological network are optimized by quantum annealing to generate an acoustic topological feature map. The steps are as follows: Calculate the gradient weights of the sound pressure levels of adjacent nodes in the initial topological connection relationship graph; According to the deterministic mapping rules, the nodes in the topological network are mapped to quantum bits and a quantum bit string is formed; Initialize all quantum bits to a uniform superposition state through Hadamard gate transformation and calculate the optimal path for sound propagation; The weights of nodes in the three-dimensional topological network are iteratively optimized through the quantum annealing algorithm, and combined with the optimal path of sound propagation to generate an acoustic topological feature map.

6. The noise monitoring point location selection method based on noise map according to claim 1 is characterized in that: The extraction of multi-scale noise propagation features through a convolutional neural network refers to extracting near-field features of the sound source through a small-scale convolution kernel, extracting reflected wave features through a medium-scale hole convolution, and extracting sound wave diffraction attenuation features at the edge of the building through a separable convolution kernel.

7. The noise monitoring point location selection method based on noise map according to claim 1 is characterized in that: The steps of combining the monitoring equipment parameter library and the coverage constraint conditions to construct a gridded space demand matrix are as follows: Through the multi-scale sound field decomposition algorithm, the spatial distribution characteristics of multi-scale noise propagation characteristics are identified; The monitoring equipment is divided into high-frequency, medium-frequency and low-frequency equipment by frequency band division, and the monitoring equipment parameter library is defined according to the monitoring equipment in different frequency bands; Through the spatial grid discretization algorithm, the multi-scale noise propagation characteristics are converted into grid coverage indicators; According to the gridded coverage index, the coverage constraint conditions are generated by Kriging interpolation method; Based on the coverage constraints, the device deployment candidate set is converted into a gridded spatial demand matrix through ArcGIS.

8. The noise monitoring point location selection method based on noise map according to claim 7 is characterized in that: Generating the optimal noise monitoring network site selection plan refers to calculating the spatial layout score of the monitoring points through a nonlinear coverage contribution algorithm; Based on the spatial layout scores of monitoring points, the spatial layout set of monitoring points is iteratively optimized through genetic algorithm to generate the optimal noise monitoring network site selection plan.

9. The noise monitoring point location selection method based on noise map according to claim 1, characterized in that: The optimal noise monitoring network site selection scheme is based on the feasibility assessment through coverage verification, and the optimal noise monitoring network site selection scheme is dynamically adjusted according to the assessment results. The steps are as follows: Based on the coverage data in the equipment parameter library, the GIS spatial analysis method is used to analyze the coverage of each monitoring device on the grid unit, and the coverage score of the optimal noise monitoring network site selection plan is calculated through a weighted penalty function; A feasibility analysis was conducted based on the coverage score of the optimal noise monitoring network site selection scheme, and the spatial layout set of monitoring points was iteratively optimized using the particle swarm optimization algorithm.

10. A noise monitoring point location selection system based on a noise map, based on the noise monitoring point location selection method based on a noise map according to any one of claims 1 to 9, characterized in that: Including, matrix generation module, map generation module, site selection plan generation module and evaluation module; The matrix generation module is used to obtain the sound source monitoring data and building BIM model of the target area, and divide the spatial range of the target area into three-dimensional grids to generate a three-dimensional noise intensity matrix; The spectrum generation module is used to construct a three-dimensional topological network based on the three-dimensional noise intensity matrix, use a quantum-inspired topological optimization algorithm to calculate the optimal path of sound propagation, and combine quantum annealing to optimize the weights of nodes in the three-dimensional topological network to generate an acoustic topological feature spectrum; The site selection plan generation module is used to extract multi-scale noise propagation characteristics through convolutional neural networks based on the acoustic topology feature map, and to build a gridded space demand matrix based on the monitoring equipment parameter library and coverage constraints. Based on the gridded space demand matrix, the optimal noise monitoring network site selection plan is generated; The evaluation module is used to conduct feasibility evaluation based on the optimal noise monitoring network site selection plan through coverage verification, and dynamically adjust the optimal noise monitoring network site selection plan according to the evaluation results.

Images