CN-122020226-A - Pollution source grid space tracing method

CN122020226ACN 122020226 ACN122020226 ACN 122020226ACN-122020226-A

Abstract

The invention relates to the technical field of pollution source positioning, and particularly discloses a pollution source grid space tracing method. The method comprises the steps of obtaining three-dimensional wind speed vector data and pollutant concentration distribution data, calculating partial derivative construction tensor, extracting a main strain direction, determining grid parameters based on a characteristic value ratio, aligning grid long axis to adjust unit sizes, generating dynamic deformation grid layout through interpolation, calculating vector included angles, clustering and marking wind direction mutation area boundaries, extracting boundary points to perform spline fitting, reconstructing grid main shafts to construct a transmission path grid, calculating gradients based on concentration distribution, tracking a transmission path, identifying a concentration peak area, and obtaining a pollution source space positioning result. According to the invention, by combining wind speed change and pollutant concentration information and combining dynamic grid adjustment and wind direction mutation clustering, the tracking precision and spatial resolution of a transmission path are improved, grid division errors are overcome, stable identification of pollutant migration rules is enhanced, and the accuracy and reliability of pollution source positioning are improved.

Inventors

DING YIFAN
LU LUNHUI
XIE XIANGXIANG
ZHANG YUANYUAN
YANG PEI

Assignees

中国长江三峡集团有限公司
中国科学院重庆绿色智能技术研究院

Dates

Publication Date: 20260512
Application Date: 20260204

Claims (10)

1. The method for tracing the pollution source grid space is characterized by comprising the following steps: S1, acquiring three-dimensional wind speed vector data and pollutant concentration distribution data, calculating wind speed component deviation by adopting a finite difference method, constructing a wind speed change rate tensor, performing eigenvalue decomposition on the tensor to extract a main strain direction vector, and calculating a main and secondary eigenvalue ratio to obtain wind field adaptability grid parameters; S2, aligning the long axis direction of the grid based on the wind field adaptive grid parameters, adjusting the size of the cells, calculating the vector angle difference of the main strain direction of the adjacent cells, and executing the interpolation operation of the main axis direction by adopting linear interpolation when the angle threshold value is exceeded, so as to generate a dynamic deformation grid layout; s3, calling a main strain direction vector in the dynamic deformation grid layout, calculating an included angle value of the vector of the adjacent positions point by point, marking the wind direction mutation points when the included angle value exceeds a mutation threshold value, and spatially grouping the mutation points by adopting a K-means clustering algorithm to form a wind direction mutation region boundary; And S4, calling the boundary of the wind direction mutation area to extract coordinates of boundary points at two sides, fitting a buffer zone boundary curve by adopting a cubic spline interpolation function, reconstructing the direction of a grid main shaft in the dynamic deformation grid layout, and constructing a pollutant transmission path tracking grid.
2. The method of claim 1, wherein the wind field adaptive grid parameters include a principal axis direction, a cell scale ratio, and a directional consistency coefficient, the dynamically deformed grid layout includes a grid orientation, a local continuity, and a spatial resolution, the wind direction abrupt region boundaries include a boundary range, a boundary morphology, and a spatial partition, and the contaminant transmission path tracking grid includes a path ductility, a boundary constraint, and a directional continuity.
3. The method for tracing the grid space of the pollution source according to claim 1, wherein the specific steps of S1 are as follows: S101, acquiring three-dimensional wind speed vector data and pollutant concentration distribution data, performing partial derivative calculation on wind speed components by adopting a finite difference method based on wind speed component values adjacent to grid nodes, constructing a change rate matrix at the grid nodes, and extracting partial derivative values of the wind speed components based on the matrix to obtain wind speed component partial derivative values; S102, calling the wind speed component partial derivative values, constructing a wind speed change rate tensor in a combined mode, decomposing the characteristic values of grid nodes based on the tensor, extracting main strain direction vectors, analyzing strain direction information of each grid node through matrix operation, and generating a main strain direction vector group; and S103, calculating the ratio of the main characteristic value to the secondary characteristic value based on the main strain direction vector group, and carrying out weighted comparison according to the characteristic value ratio and the pollutant concentration distribution data of each grid node to generate the wind field adaptive grid parameters.
4. A method for tracing a pollution source grid space according to claim 3, wherein the specific steps of S2 are as follows: S201, aligning the grid long axis direction based on the wind field adaptability grid parameters, obtaining boundary vectors of grid cells, comparing the wind field main directions, adjusting the included angle between the grid long axis and the wind field main directions, and correcting the length-width ratio value of the grid cells to obtain cell orientation angle values; s202, calling the unit orientation angle value, calculating the angle difference between main strain direction vectors of adjacent units, judging the angle difference according to a set angle threshold value, recording the units with the angle difference exceeding the threshold value, and extracting the angle difference value to obtain the angle difference value of the adjacent units; And S203, performing direction interpolation operation in the main axis direction by adopting linear interpolation according to the angle difference values of the adjacent cells, and adjusting the arrangement sequence of the cells by combining the coordinate positions of the grid cells to obtain the dynamic deformation grid layout.
5. The method for tracing a pollution source grid space according to claim 4, wherein the specific steps of S3 are as follows: S301, calling a main strain direction vector in the dynamic deformation grid layout, calculating an included angle value of the vector of the adjacent position point by point, comparing the calculated angle with a set mutation threshold value, and when the angle exceeds the threshold value, marking the angle as a mutation point, and generating an angle mutation marking value; s302, extracting marked mutation point coordinate information according to the angle mutation marking value, setting an initial clustering center by adopting a K-means clustering algorithm, performing iterative computation according to the coordinate information of the mutation points in the space, and adjusting class division according to deviation from the clustering center to generate mutation point clustering coefficients; s303, according to the mutation point clustering coefficient, carrying out boundary fitting operation on coordinates of mutation points in the same category through a space fitting algorithm, dividing the space boundary of the mutation points, and forming a wind direction mutation region boundary.
6. The method for tracing a pollution source grid space according to claim 5, wherein the angle mutation marking value is marking data carried out at a corresponding position when a vector included angle of adjacent positions exceeds a preset mutation threshold; the mutation point clustering coefficient refers to data generated after mutation points are classified by a clustering algorithm; the abrupt change threshold is a preset value for judging whether the vector included angles of adjacent positions form abrupt change.
7. The method for tracing a pollution source grid space according to claim 5, wherein the specific steps of S4 are as follows: S401, calling the boundary of the wind direction mutation area to extract boundary point coordinates at two sides, detecting the extracted boundary point coordinates according to a sequence order, calculating space distance values of adjacent points, summarizing and averaging to generate a boundary point distance value; s402, calling the boundary point interval value, fitting boundary point coordinates on two sides by adopting a cubic spline interpolation function, calculating tangential angles of interpolation nodes and calculating curvature, and generating a curvature coefficient of a buffer zone; S403, reconstructing the grid main axis direction in the dynamic deformation grid layout according to the curvature coefficient of the buffer zone, registering with the grid cell boundary according to the geometric distribution characteristics of the fitting curve, calculating the offset angle value of the main axis direction relative to the curve direction in the cell, and establishing a pollutant transmission path tracking grid.
8. The method for tracing the space of the pollution source grid according to claim 7, wherein the distance value between boundary points refers to a value obtained by calculating the space distance between adjacent boundary points on the boundary of the wind direction mutation area and averaging; The buffer band curvature coefficient refers to a curvature value calculated based on tangential angle change after boundary point coordinates are fitted through a cubic spline interpolation function.
9. The method of grid space tracing of a pollution source of claim 1, further comprising the step of S5: s5, calling the pollutant transmission path tracking grid to distribute pollutant concentration values to unit center points, calculating unit concentration gradient values along a main strain direction vector, determining a pollutant transmission path through gradient reverse tracking, and identifying a concentration peak point at the starting end of the path to obtain a pollution source space positioning result; the pollution source spatial positioning result comprises a peak value coordinate, a spatial position and source point distribution.
10. The method for tracing a pollution source grid space according to claim 9, wherein the specific step of S5 is: S501, distributing a pollutant concentration value to a unit center point position based on the pollutant transmission path tracking grid, analyzing multi-unit concentration differences by combining space coordinate parameters, and generating a unit concentration distribution value according to concentration distribution conditions of concentration difference calculation units among units in a grid area; S502, calling the concentration distribution value of the unit, comparing the included angle relation between the concentration difference value of the adjacent unit and the direction vector in a differential mode according to the space distance between the main strain direction vector and the coordinates of the adjacent unit, calculating the concentration gradient of multiple units, and obtaining a unit concentration gradient sequence; s503, according to the unit concentration gradient sequence, invoking concentration gradient values of adjacent units to carry out gradient reverse tracking, judging the transmission direction of path nodes in space coordinates step by step, locking the starting end of the path and positioning coordinates of peak points, and obtaining a pollution source space positioning result.

Description

Pollution source grid space tracing method Technical Field The invention relates to the technical field of pollution source positioning, in particular to a pollution source grid space tracing method. Background The method mainly researches related methods for identifying and tracking a pollution source in environmental monitoring and pollution control, acquires concentration information of pollutants in an environmental medium through means of air quality monitoring, water pollution detection, soil sample analysis and the like, analyzes distribution and diffusion conditions of the pollutants by combining geographic space information to determine the spatial position and emission characteristics of the pollution source, and is widely applied to pollution treatment work of atmospheric environment, water environment and soil environment in terms of pollutant monitoring data acquisition, spatial distribution modeling, diffusion path inference, pollution source region locking and the like. The traditional pollution source grid space tracing method is to collect and compare monitoring point data in a set monitoring area according to a fixed grid division mode, judge the pollution source position through the space difference of pollutant concentration measurement results in grid units, and complete the method by adopting the modes of monitoring point layout sampling, concentration value grid-by-grid comparison, space superposition analysis based on geographic coordinates and the like. In addition, a pollution source inversion method based on reverse track analysis or an accompanying model also exists, a pollution source prediction model is established by simulating a pollutant reverse diffusion path through a meteorological field to calculate the position of a source region and an empirical tracing method based on statistical correlation or a machine learning model, and historical monitoring data and meteorological features are utilized. However, the conventional method has the problems of large calculation amount, strong dependence on meteorological input, incapability of dynamically responding to wind field change and the like. Particularly, under the condition of complex wind speed gradient or obvious local disturbance, the traditional inversion or empirical model is difficult to accurately describe the real transmission path of the pollutants, so that the positioning accuracy of the pollution sources is insufficient. The space tracing method of the fixed grid cannot effectively adjust grid division when the wind field changes suddenly or in a gradient mode, so that the wind field change cannot be accurately reflected. The limitation is particularly prominent when the wind speed direction changes obviously or the topography is complex, and particularly in complex environments such as urban multi-air channels or mountainous areas, the fixed grid cannot adapt to air flow turning and local disturbance, so that calculation deviation of pollutant diffusion paths is caused, the positioning accuracy of pollution sources is affected, and the uncertainty of environmental treatment decisions is increased. In the prior art, the grid self-adaptive mechanism aiming at the wind field space nonlinear change characteristic is less researched, a static or semi-dynamic grid division mode is mostly adopted, the concentration distribution is updated only in the time dimension, and the real-time response to the wind speed tensor characteristic change is lacking. Especially in the pollutant tracing process, the grid structure cannot be directionally adjusted by combining the wind field main strain direction, so that the tracing precision is insufficient and the path continuity is poor. In addition, the traditional method often adopts empirical partition or fixed threshold division in the wind direction mutation area, so that the spatial continuity of the wind field structure is ignored, and the mutation boundary cannot be accurately identified. Disclosure of Invention In order to solve the problem that grid division cannot be effectively adjusted when a wind field changes suddenly or in a gradient manner by a space tracing method of a fixed grid, so that the wind field change cannot be accurately reflected. The limitation is particularly prominent when the wind speed direction changes obviously or the topography is complex, and particularly under the complex environments such as urban multi-air channels or mountainous areas, the fixed grid cannot adapt to air flow turning and local disturbance, so that the calculation deviation of pollutant diffusion paths is caused, the positioning accuracy of pollution sources is influenced, and the uncertainty of environmental treatment decisions is increased. In order to achieve the purpose, the invention adopts the following technical scheme that the pollution source grid space tracing method comprises the following steps: S1, acquiring three-dimensional wind speed vector data and pollutant con