CN115292890A - Site soil pollutant concentration three-dimensional space prediction method based on multi-source auxiliary data development - Google Patents
Site soil pollutant concentration three-dimensional space prediction method based on multi-source auxiliary data development Download PDFInfo
- Publication number
- CN115292890A CN115292890A CN202210760419.1A CN202210760419A CN115292890A CN 115292890 A CN115292890 A CN 115292890A CN 202210760419 A CN202210760419 A CN 202210760419A CN 115292890 A CN115292890 A CN 115292890A
- Authority
- CN
- China
- Prior art keywords
- dimensional
- target area
- auxiliary data
- pollutant concentration
- grid
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000003802 soil pollutant Substances 0.000 title claims abstract description 88
- 238000000034 method Methods 0.000 title claims abstract description 70
- 238000011161 development Methods 0.000 title claims abstract description 22
- 238000009795 derivation Methods 0.000 claims abstract description 6
- 238000005527 soil sampling Methods 0.000 claims description 40
- 238000001514 detection method Methods 0.000 claims description 30
- 239000002689 soil Substances 0.000 claims description 26
- 230000005674 electromagnetic induction Effects 0.000 claims description 19
- 238000012549 training Methods 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 10
- 238000012795 verification Methods 0.000 claims description 9
- 238000010801 machine learning Methods 0.000 claims description 6
- 238000005553 drilling Methods 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 5
- 238000012706 support-vector machine Methods 0.000 claims description 5
- 238000003325 tomography Methods 0.000 claims description 5
- 238000012847 principal component analysis method Methods 0.000 claims description 3
- 239000003344 environmental pollutant Substances 0.000 abstract description 18
- 231100000719 pollutant Toxicity 0.000 abstract description 18
- 230000000704 physical effect Effects 0.000 abstract description 7
- 238000013461 design Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 6
- 238000005259 measurement Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 238000003900 soil pollution Methods 0.000 description 3
- 239000002699 waste material Substances 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000002790 cross-validation Methods 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 230000006698 induction Effects 0.000 description 2
- 244000286663 Ficus elastica Species 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011109 contamination Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000004070 electrodeposition Methods 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention relates to a field soil pollutant concentration three-dimensional space prediction method based on multi-source auxiliary data development, which is characterized in that different derivation methods are utilized to further develop field pollution multi-source auxiliary data such as functional area layout and underground physical properties, more derived auxiliary data representing pollutant concentration three-dimensional space distribution prior information are obtained, a quantitative relation model between the derived auxiliary data and pollutant concentration is further established, and three-dimensional space prediction of field soil pollutant concentration is realized. The method solves the problems that the correlation between the original auxiliary data and the soil pollutant concentration is weak and the site pollution prior information hidden in the auxiliary data is not fully developed and utilized in the prior art, and remarkably improves the three-dimensional space prediction precision of the site soil pollutant concentration.
Description
Technical Field
The invention relates to a field soil pollutant concentration three-dimensional space prediction method based on multi-source auxiliary data development, and belongs to the technical field of soil pollution prediction.
Background
The existing three-dimensional spatial distribution prediction of the field soil pollutant concentration is mainly realized by using traditional spatial interpolation methods such as three-dimensional inverse distance weighted Interpolation (IDW), three-dimensional kriging interpolation and the like, but the problems of strong smooth effect of prediction results (underestimated high concentration value, overestimated low concentration value), low precision and the like are still outstanding. Meanwhile, in the three-dimensional distribution of the soil pollution concentration of the field predicted by the existing space interpolation method, only the concentration data of the pollutants in the drilled soil is utilized, and the integration of auxiliary data of the field pollution is lacked, so that the improvement of the prediction precision is further restricted.
The field functional area layout and the multi-source auxiliary data such as underground physical properties provide a large amount of prior information about the spatial distribution of pollutant concentration, and the auxiliary data and the field soil pollutant concentration have certain spatial correlation. For example, the functional area layout can reflect the influence of site pollution sources (production process, waste stacking, etc.) on the spatial distribution of pollutant concentration; the electromagnetic data of the underground medium acquired by geophysical exploration methods such as electromagnetic induction (EM), resistivity tomography (ERT) and the like can reflect the difference of electromagnetic properties of polluted soil and uncontaminated soil and the like. In addition, compared with soil pollutant concentration data acquired by drilling sampling, the coverage density and integrity of the auxiliary data space are higher, and the data acquisition cost is lower and the data acquisition speed is higher. Therefore, the method realizes three-dimensional space prediction based on the quantitative relation between the auxiliary data and the pollutant concentration, and is one of potential technical approaches for overcoming the three-dimensional distribution limitation of the field soil pollution concentration predicted by the traditional space interpolation method. However, the following problems still exist in predicting the three-dimensional spatial distribution of the pollutant concentration by using the quantitative relationship between the auxiliary data such as the functional area layout and the underground physical properties and the pollution concentration of the field soil:
(1) The correlation between the functional area layout and the original data of the underground physical properties and the concentration of soil pollutants may be weak, so that the spatial prediction precision is reduced and even lower than that of the traditional interpolation method. For example, the field space is very complex, and ground impurities, soil characteristics, geological conditions and the like all affect the EM and ERT detection results, so that the correlation between the original auxiliary data of underground physical properties such as conductivity, resistivity and the like and the soil pollutant concentration is weakened, and the spatial prediction precision is low;
(2) Site pollution priors implied in the auxiliary data are not fully developed and utilized. For example, the functional area layout is two-dimensional discrete data, and the current method for identifying the quantitative relationship between the functional area layout and the soil pollutant concentration simply in a dummy variable mode is difficult to comprehensively reflect the influence intensity of the functional area layout on the three-dimensional spatial distribution of the pollutant concentration and the like.
Disclosure of Invention
The invention aims to solve the technical problem of providing a field soil pollutant concentration three-dimensional space prediction method based on multi-source auxiliary data development, and adopting a brand-new design strategy, so that the soil pollutant concentration three-dimensional space distribution prediction efficiency can be effectively improved.
In order to solve the technical problems, the invention adopts the following technical scheme: the invention designs a field soil pollutant concentration three-dimensional space prediction method based on multisource auxiliary data development, a soil pollutant concentration three-dimensional prediction model of a target area is obtained through steps A to D, and the soil pollutant concentration three-dimensional prediction model is applied through step i to realize the soil pollutant concentration three-dimensional distribution prediction of the underground space of the target area;
step A, acquiring soil pollutant concentration measured values of soil sampling points at preset depths at preset drilling positions in a target area, and a multi-source auxiliary data set corresponding to the target area and containing preset data types, and then entering step B;
b, preprocessing the multi-source auxiliary data set corresponding to the target area by a spatial interpolation method based on a three-dimensional grid preset in the underground space of the target area, updating the multi-source auxiliary data set corresponding to the target area, and entering the step C;
c, according to a preset three-dimensional grid of the underground space of the target area, aiming at the multi-source auxiliary data set corresponding to the target area, executing data derivation processing to obtain three-dimensional derived auxiliary data of the target area, and then entering the step D;
d, obtaining derived auxiliary data corresponding to each soil sampling point respectively based on the three-dimensional derived auxiliary data of the target area, taking the derived auxiliary data corresponding to the soil sampling points as a prediction variable and the measured soil pollutant concentration value corresponding to the soil sampling points as a target variable, and training aiming at a preset machine learning model to obtain a three-dimensional prediction model of the soil pollutant concentration of the target area;
and i, respectively aiming at each grid of the underground space of the target area, obtaining derived auxiliary data corresponding to the grid, applying a soil pollutant concentration three-dimensional prediction model to obtain a soil pollutant concentration prediction value of each grid, and further realizing three-dimensional prediction of the soil pollutant concentration of the underground space of the target area.
As a preferred technical scheme of the invention: step E, after step D is executed, step E is executed;
and E, calculating the root mean square error between the soil pollutant concentration predicted value of the soil sampling point obtained based on the three-dimensional soil pollutant concentration prediction model and the corresponding soil pollutant concentration measured value by using an independent verification method, namely obtaining the prediction precision of the three-dimensional soil pollutant concentration prediction model.
As a preferred technical scheme of the invention: and B, correspondingly setting a multi-source auxiliary data set of each preset data type in the target area in the step A, wherein the multi-source auxiliary data set comprises area ranges of various preset functional areas in the target area, conductivity two-dimensional measuring point data of the target area corresponding to each preset electromagnetic detection frequency and two-dimensional section inversion resistivity data of each preset measuring line position corresponding to the target area.
As a preferred technical scheme of the invention: and measuring the underground conductivity of the target area by an electromagnetic induction method EM to obtain conductivity two-dimensional measuring point data of the target area under the corresponding preset electromagnetic detection frequencies.
As a preferred technical scheme of the invention: and measuring the underground resistivity of the target region through a resistivity tomography ERT to obtain two-dimensional section inversion resistivity data of the target region corresponding to each preset measuring line position.
As a preferred technical scheme of the invention: in the step B, based on the conductivity two-dimensional measuring point data of the target area in the multi-source auxiliary data set corresponding to each preset electromagnetic detection frequency, carrying out interpolation processing on a two-dimensional grid in the horizontal direction of a three-dimensional grid of the underground space of the target area through a spatial interpolation method to obtain the soil conductivity two-dimensional spatial distribution of the target area corresponding to each electromagnetic detection frequency;
and performing interpolation processing on the underground space three-dimensional grid of the target area by a spatial interpolation method based on the two-dimensional section inversion resistivity data of the target area corresponding to each preset survey line position in the multi-source auxiliary data set to obtain the resistivity three-dimensional spatial distribution corresponding to the target area.
As a preferred technical scheme of the invention: the spatial interpolation method in the step B is an IDW interpolation method or a kriging interpolation method.
As a preferred technical scheme of the invention: the step C comprises the following steps C1 to C4;
step C1, based on the area range of various preset functional areas in the target area, aiming at each grid in the target area, if the grid is positioned in the preset functional area, the grid is assigned to be 1, if the grid is not positioned in the preset functional area, the grid is assigned to be 0, and the assignment of each grid in the target area is completed, namely, the three-dimensional derivative auxiliary data of the functional area type corresponding to the target area is formed; meanwhile, respectively calculating the nearest Euclidean distance from each grid to the edge of each functional area aiming at each grid in the target area, namely obtaining the nearest Euclidean distance from each grid in the target area to the edge of each functional area, and forming three-dimensional derivative auxiliary data of the distance from the functional area corresponding to the target area; then entering step C2;
step C2, obtaining the minimum value, the maximum value, the median value, the mean value and the variance of the electric conductivity of each grid of the target area in the horizontal direction respectively corresponding to each electromagnetic detection frequency based on the two-dimensional spatial distribution of the electric conductivity of each electromagnetic detection frequency corresponding to the target area, and forming two-dimensional derived auxiliary data of the electric conductivity statistic corresponding to the target area; meanwhile, based on the sliding of two-dimensional sliding windows with preset sizes in the horizontal direction of the target area, the minimum value, the maximum value, the median value, the mean value, the variance and the gradient in the two-dimensional direction of the conductivity of each grid in each two-dimensional sliding window under each electromagnetic detection frequency are obtained, and two-dimensional derivative auxiliary data of the conductivity statistics of the sliding window corresponding to the target area are formed; then entering step C3;
step C3, obtaining the minimum value, the maximum value, the median value, the mean value and the variance among the depth resistivities of each grid in the horizontal direction of the target area based on the three-dimensional spatial distribution of the resistivity corresponding to the target area, and forming two-dimensional derived auxiliary data of the resistivity statistics among the depths corresponding to the target area; meanwhile, based on the sliding of the three-dimensional sliding windows with preset sizes on the three-dimensional grids of the underground space of the target area, obtaining the minimum value, the maximum value, the median value, the mean value, the variance and the gradient in the three-dimensional direction among the grid resistivity in each three-dimensional sliding window to form three-dimensional derived auxiliary data of the resistivity statistics of the sliding window corresponding to the target area; then entering step C4;
and C4, copying the two-dimensional derived auxiliary data of the conductivity statistics, the two-dimensional derived auxiliary data of the conductivity statistics of the sliding window and the two-dimensional derived auxiliary data of the resistivity statistics among depths corresponding to the target area along the depth direction of the three-dimensional grid to obtain three-dimensional derived auxiliary data of the conductivity statistics, the three-dimensional derived auxiliary data of the conductivity statistics of the sliding window and the three-dimensional derived auxiliary data of the resistivity among depths corresponding to the target area, combining the three-dimensional derived auxiliary data of the type of the functional area, the three-dimensional derived auxiliary data of the distance of the functional area and the three-dimensional derived auxiliary data of the resistivity of the sliding window corresponding to the target area to form three-dimensional derived auxiliary data of the target area, and then entering the step D.
As a preferred technical scheme of the invention: the step D comprises the following steps D1 to D3;
step D1, standardizing three-dimensional derivative auxiliary data of a target area, obtaining each main component data in the derivative auxiliary data by using a main component analysis method, and then entering step D2;
step D2, acquiring data of each soil sampling point corresponding to each main component data type respectively based on the three-dimensional derivative auxiliary data of the target area to form derivative auxiliary data corresponding to each soil sampling point respectively, and then entering step D3;
and D3, training aiming at a regression model of a support vector machine based on the soil sampling points by taking the derived auxiliary data corresponding to the soil sampling points as prediction variables and the measured value of the concentration of the soil pollutants corresponding to the soil sampling points as target variables to obtain a three-dimensional prediction model of the concentration of the soil pollutants in the target area.
Compared with the prior art, the field soil pollutant concentration three-dimensional space prediction method based on multi-source auxiliary data development has the following technical effects:
the invention designs a field soil pollutant concentration three-dimensional space prediction method based on multi-source auxiliary data development, aiming at field pollution multi-source auxiliary data such as functional area layout, underground physical properties and the like, further development is carried out by utilizing different derivation methods, more derived auxiliary data representing soil pollutant concentration three-dimensional space distribution prior information are obtained, a quantitative relation model between the derived auxiliary data and pollutant concentration is further established by utilizing machine learning, high-precision three-dimensional space prediction of the field soil pollutant concentration is realized, the problems that in the prior art, the correlation between original auxiliary data and the soil pollutant concentration is weak, and field pollution prior information hidden in the auxiliary data is not fully developed and utilized are solved, the pollutant concentration three-dimensional space prediction precision is improved, and the problems that the traditional interpolation method is strong in smoothing effect and low in prediction precision are solved.
Drawings
FIG. 1 is a flow chart of an implementation of a field soil pollutant concentration three-dimensional space prediction method based on multi-source auxiliary data development according to the invention;
FIG. 2 is a plot of a functional zone layout and a plot of a borehole sampling point profile for a field according to an embodiment of the present invention;
FIG. 3 is a three-dimensional spatial interpolation plot of ERT resistivity for a field according to an embodiment of the present invention;
FIG. 4 is an example of functional area layout three-dimensional derivative assistance data (left) and an example of resistivity three-dimensional derivative assistance data (right);
fig. 5 is a comparison of the three-dimensional spatial distribution of soil Zn concentration at a field according to an embodiment of the present invention (left) and conventional IDW interpolation (right).
Detailed Description
The following description will explain embodiments of the present invention in further detail with reference to the accompanying drawings.
The invention designs a field soil pollutant concentration three-dimensional space prediction method based on multi-source auxiliary data development, as shown in figure 1, in practical application, specifically executing the steps A to D to obtain a soil pollutant concentration three-dimensional prediction model of a target area.
Step A, obtaining soil pollutant concentration measured values of soil sampling points at preset depths at preset drilling positions in a target area, and a multi-source auxiliary data set corresponding to the target area and containing preset data types, and then entering step B.
In application, the specific design of each preset data type contained in the multi-source auxiliary data set comprises the area range of each preset functional area in a target area, the conductivity two-dimensional measuring point data of the target area corresponding to each preset electromagnetic detection frequency, and the two-dimensional section inversion resistivity data of each preset measuring line position corresponding to the target area.
In practical application, the type and the boundary of each functional area of the field are interpreted by utilizing a high-resolution remote sensing image and combining field reconnaissance to obtain the area range of each preset functional area; the underground conductivity of a target area is measured through an electromagnetic induction method EM, conductivity two-dimensional measuring point data of the target area under corresponding preset electromagnetic detection frequencies are obtained, a primary magnetic field is generated on the ground surface through current, a secondary induction magnetic field is generated in field soil with uneven media through the primary magnetic field, and the conductivity information of the underground media can be obtained through inversion by receiving the secondary induction magnetic field through the ground surface. The conductivity information of different probing frequencies reflects the conductivity of the underground medium in different probing depths, and the probing depth is deeper when the probing frequency is lower. The electromagnetic induction method EM is widely applied to pollution site investigation, and the detection cost is extremely low, so that the target site is uniformly covered by the electromagnetic induction method EM two-dimensional measuring point arrangement as far as possible.
And with respect to two-dimensional section inversion resistivity data, the specific design in application measures the underground resistivity of a target area through a resistivity tomography ERT to obtain two-dimensional section inversion resistivity data of the target area corresponding to preset measuring line positions, the resistivity is usually changed when the concentration of soil pollutants changes, the resistivity tomography ERT method is to introduce direct current into the ground, then observe the power supply current intensity and the potential difference between measuring electrodes to calculate apparent resistivity, and record the information of geographic coordinates (x, y), electrode intervals, electrode devices and the like of the electrodes during measurement, so that the two-dimensional section resistivity data of different ERT measuring line positions are obtained through inversion. Two-dimensional section resistivity data after inversion of different ERT measuring lines needs to be associated with geographic coordinates (x, y) and inversion depth z of each electrode, so that three-dimensional scattered point data of the resistivity is obtained. ERT detection is also less costly, and therefore, the wire trace placement should also cover the target site as uniformly as possible.
And B, preprocessing the multi-source auxiliary data set corresponding to the target area by a spatial interpolation method based on a three-dimensional grid preset in the underground space of the target area, updating the multi-source auxiliary data set corresponding to the target area, and then entering the step C.
In the specific implementation of step B, based on the conductivity two-dimensional measurement point data of the target region in the multi-source auxiliary data set corresponding to each preset electromagnetic detection frequency, interpolation processing is performed on the two-dimensional grid in the horizontal direction of the three-dimensional grid in the underground space of the target region by using a spatial interpolation method such as an IDW interpolation method or a kriging interpolation method, so as to obtain the soil conductivity two-dimensional spatial distribution of the target region corresponding to each electromagnetic detection frequency.
And performing interpolation processing on the underground space three-dimensional grid of the target area by a spatial interpolation method such as an IDW (inverse discrete wavelet) interpolation method or a Kriging interpolation method based on the two-dimensional section inversion resistivity data of the preset line positions corresponding to the target area in the multi-source auxiliary data set to obtain the three-dimensional spatial distribution of the resistivity corresponding to the target area.
And C, performing data derivation processing aiming at the multi-source auxiliary data set corresponding to the target area according to a three-dimensional grid preset in the underground space of the target area to obtain three-dimensional derived auxiliary data of the target area, and entering the step D.
In a specific practical application, the step C includes the following steps C1 to C4.
Step C1, based on the area range of various preset functional areas in the target area, aiming at each grid in the target area, if the grid is positioned in the preset functional area, the grid is assigned to be 1, if the grid is not positioned in the preset functional area, the grid is assigned to be 0, and the assignment of each grid in the target area is completed, namely, the three-dimensional derivative auxiliary data of the functional area type corresponding to the target area is formed; meanwhile, respectively calculating the nearest Euclidean distance from each grid to the edge of each functional area aiming at each grid in the target area, namely obtaining the nearest Euclidean distance from each grid in the target area to the edge of each functional area, and forming three-dimensional derivative auxiliary data of the distance from the functional area corresponding to the target area; the different functional zone types and the distance from the functional zone edge reflect the influence strength of potential pollution sources on the field soil pollutant concentration, for example, a waste residue stacking zone and a production workshop, the soil characteristic pollutant concentration is often higher, and the pollutant concentration is generally higher when the functional zone is closer to the waste residue stacking zone and the production workshop, and then the step C2 is carried out.
Step C2, acquiring the minimum value, the maximum value, the median value, the mean value and the variance among the conductivities of the grids in the horizontal direction of the target area under the electromagnetic detection frequencies respectively based on the two-dimensional spatial distribution of the conductivities of the target area under the electromagnetic detection frequencies respectively, and forming two-dimensional derived auxiliary data of the conductivity statistics corresponding to the target area; meanwhile, based on the sliding of two-dimensional sliding windows with preset sizes in the horizontal direction of the target area, the minimum value, the maximum value, the median value, the mean value, the variance and the gradient in the two-dimensional direction of the conductivity of each grid in each two-dimensional sliding window under each electromagnetic detection frequency are obtained, and two-dimensional derivative auxiliary data of the conductivity statistics of the sliding window corresponding to the target area are formed; then step C3 is entered.
Step C3, obtaining the minimum value, the maximum value, the median value, the mean value and the variance among the depth resistivities of each grid in the horizontal direction of the target area based on the three-dimensional spatial distribution of the resistivity corresponding to the target area, and forming two-dimensional derived auxiliary data of the resistivity statistics among the depths corresponding to the target area; meanwhile, based on the sliding of the three-dimensional sliding windows with preset sizes on the three-dimensional grids of the underground space of the target area, the minimum value, the maximum value, the median value, the mean value, the variance and the gradient in the three-dimensional direction of the grid resistivity in each three-dimensional sliding window are obtained, and three-dimensional derived auxiliary data of the resistivity statistics of the sliding window corresponding to the target area are formed; then step C4 is entered.
And C4, copying the two-dimensional derived auxiliary data of the conductivity statistics, the two-dimensional derived auxiliary data of the conductivity statistics of the sliding window and the two-dimensional derived auxiliary data of the resistivity statistics among depths corresponding to the target area along the depth direction of the three-dimensional grid to obtain three-dimensional derived auxiliary data of the conductivity statistics, the three-dimensional derived auxiliary data of the conductivity statistics of the sliding window and the three-dimensional derived auxiliary data of the resistivity among depths corresponding to the target area, combining the three-dimensional derived auxiliary data of the type of the functional area, the three-dimensional derived auxiliary data of the distance of the functional area and the three-dimensional derived auxiliary data of the resistivity of the sliding window corresponding to the target area to form three-dimensional derived auxiliary data of the target area, and then entering the step D.
And C, further mining the information which is hidden in the two-dimensional and three-dimensional auxiliary data and has stronger correlation with the spatial distribution of the pollutant concentration so as to establish a spatial prediction model by using the information and improve the prediction precision of the pollutant concentration.
And D, acquiring derived auxiliary data corresponding to each soil sampling point respectively based on the three-dimensional derived auxiliary data of the target area, training a preset machine learning model by taking the derived auxiliary data corresponding to the soil sampling points as a prediction variable and the measured soil pollutant concentration value corresponding to the soil sampling points as a target variable, acquiring a three-dimensional prediction model of the soil pollutant concentration of the target area, and entering the step E.
In practical applications, the step D specifically performs the following steps D1 to D3.
Step D1, because the machine learning prediction model is easily influenced by the magnitude of the prediction variable, before the pollutant concentration three-dimensional prediction model is constructed, firstly, the derived auxiliary data needs to be standardized so as to eliminate the magnitude difference of different derived auxiliary data, therefore, the three-dimensional derived auxiliary data of the target area is standardized, each principal component data in the derived auxiliary data is obtained by using a principal component analysis method, and then the step D2 is carried out.
And D2, acquiring data of each soil sample point corresponding to each main component data type respectively based on the three-dimensional derivative auxiliary data of the target area to form derivative auxiliary data corresponding to each soil sample point respectively, and then entering the step D3.
And D3, training aiming at a regression model of a support vector machine based on the soil sampling points by taking the derived auxiliary data corresponding to the soil sampling points as prediction variables and the measured value of the concentration of the soil pollutants corresponding to the soil sampling points as target variables to obtain a three-dimensional prediction model of the concentration of the soil pollutants in the target area.
The support vector machine regression can map low-dimensional data to a high-dimensional feature space, prediction is achieved by searching for hyperplanes meeting the regression, and when the model is built, the optimal values of model parameters are determined by cross validation of training set data.
And E, calculating the root mean square error between the soil pollutant concentration predicted value of the soil sampling point obtained based on the three-dimensional soil pollutant concentration prediction model and the corresponding soil pollutant concentration measured value by using an independent verification method, namely obtaining the prediction precision of the three-dimensional soil pollutant concentration prediction model.
And (e) based on the acquisition of the three-dimensional prediction model of the soil pollutant concentration of the target area, further realizing the three-dimensional distribution prediction of the soil pollutant concentration of the underground space of the target area by applying the three-dimensional prediction model of the soil pollutant concentration through the step i.
And i, respectively aiming at each grid of the underground space of the target area, obtaining derived auxiliary data corresponding to the grid, applying a soil pollutant concentration three-dimensional prediction model to obtain a soil pollutant concentration prediction value of each grid, and further realizing three-dimensional prediction of the soil pollutant concentration of the underground space of the target area.
In practical application, the design method takes a certain rubber plant site as an object example, the three-dimensional spatial distribution of the Zn concentration of the soil is predicted by adopting the method, the implementation flow is shown in figure 1, and the specific implementation mode is described in detail as follows:
step A, obtaining soil pollutant concentration measured values of soil sampling points at preset depths at preset drilling positions in a target area, and a multi-source auxiliary data set corresponding to the target area and containing preset data types, and then entering step B.
(1.1) sampling boreholes in an example field, collecting 282 soil samples with different depths, recording three-dimensional coordinates (x, y, z) of soil sampling points, performing laboratory analysis, and measuring the Zn concentration of the soil;
(1.2) interpreting and identifying the type and the boundary of each functional area by utilizing the high-resolution second remote sensing image in the range of the example site and combining with site survey of the site to obtain a site functional area layout as shown in figure 2;
(1.3) collecting two-dimensional measuring point data of the underground conductivity of the site by using a multi-frequency electromagnetic detector, wherein 5 detection frequencies are adopted in the EM measurement, namely 474Hz, 1625Hz, 5475Hz, 18575Hz and 63025Hz respectively, and the two-dimensional measuring point data of the EM conductivity of the 5 detection frequencies are obtained respectively;
(1.4) measuring the underground apparent resistivity of the field by using a high-density resistivity meter, arranging 14 ERT measuring lines on the field, wherein the measuring lines are shortest 130m, longest 150m and 2m in electrode spacing, measuring the apparent resistivity by using a Wenner-Schlumberger electrode device, recording geographic coordinates (x, y) of the electrode positions, and inverting the apparent resistivity data of each ERT measuring line by using AGI Earth imager software to obtain the resistivity data of 14 two-dimensional sections.
And B, preprocessing the multi-source auxiliary data set corresponding to the target area by a spatial interpolation method based on a three-dimensional grid preset in the underground space of the target area, updating the multi-source auxiliary data set corresponding to the target area, and then entering the step C.
(2.1) according to the range of the horizontal and vertical directions of the field, setting the grid resolution to be 2m multiplied by 1m (length multiplied by width multiplied by depth), discretizing the underground space of the field into a three-dimensional grid, and then performing auxiliary data preprocessing, derivative development and soil Zn concentration three-dimensional space prediction on the basis of the three-dimensional grid;
(2.2) with the two-dimensional grid in the horizontal direction of the three-dimensional grid as a grid point position to be interpolated, respectively interpolating the conductivity two-dimensional measurement point data of 5 detection frequencies into two-dimensional grid data by IDW interpolation to obtain 5 conductivity two-dimensional grid data serving as two-dimensional auxiliary data; the three-dimensional grid is used as a grid point position to be interpolated, and the resistivity data of 14 two-dimensional sections are interpolated into 1 three-dimensional grid data by utilizing three-dimensional kriging interpolation, as shown in figure 3, and used as three-dimensional auxiliary data.
And C, according to a preset three-dimensional grid in the underground space of the target area, executing data derivation processing aiming at the multi-source auxiliary data set corresponding to the target area to obtain three-dimensional derived auxiliary data of the target area, and then entering the step D.
(3.1) for a field functional area layout, respectively converting 10 functional area types into 0-1 binary discrete data, and further respectively assigning values to grids in a three-dimensional grid, namely for a certain functional area type, when grid points are within the underground range of the functional area of the type, assigning the values to be 1, otherwise, assigning the values to be 0, traversing 10 functional area types, and obtaining 10 functional area type three-dimensional derivative auxiliary data; secondly, respectively calculating the nearest Euclidean distances from grid points in the three-dimensional grid to the edges of 10 functional area polygons to obtain 10 functional area distance three-dimensional derivative auxiliary data, and displaying the three-dimensional distribution of Euclidean distances between the grids in the three-dimensional grid and a gelling workshop as shown in figure 4 (left);
(3.2) for the EM conductivity two-dimensional spatial distribution data of 5 detection frequencies, firstly, calculating the minimum value, the maximum value, the median value, the mean value and the variance among the conductivities of different detection frequencies to obtain 5 conductivity two-dimensional derivative auxiliary data; secondly, setting 3 two-dimensional sliding windows with the sizes of 3 × 3, 9 × 9 and 11 × 11 respectively, and calculating the minimum value, the maximum value, the median value, the mean value, the variance and the gradients in the x and y directions of the conductivity data in different sliding windows respectively for the EM conductivity two-dimensional space distribution data of each detection frequency to obtain 105 EM conductivity two-dimensional derivative auxiliary data;
(3.3) for the three-dimensional spatial distribution data of the ERT resistivity, firstly, extracting two-dimensional distribution data of the resistivity at different depths, and calculating the minimum value, the maximum value, the median value, the mean value and the variance among the resistivity at different depths to obtain 5 pieces of two-dimensional derived auxiliary data of the resistivity; secondly, setting 3 three-dimensional sliding windows with the sizes of 3 × 3 × 3, 9 × 9 × 3 and 11 × 11 × 3 respectively, further calculating the minimum value, the maximum value, the median value, the mean value, the variance and the gradients in the x, y and z directions of the resistivity data in the sliding windows under different sliding windows, and obtaining 24 three-dimensional derived auxiliary data of the resistivity sliding window statistics in total, wherein, as shown in fig. 4 (right), the three-dimensional spatial distribution of the maximum value of the resistivity sliding window 11 × 11 × 3 is displayed;
and (3.4) respectively and sequentially copying 110 EM conductivity two-dimensional derivative auxiliary data and 5 resistivity two-dimensional derivative auxiliary data calculated according to the depth layers of the three-dimensional grid, and respectively expanding the data to a three-dimensional space so as to keep consistent with the defined underground three-dimensional grid.
Through the above steps, 159 pieces of three-dimensional derived auxiliary data (i.e., 159 pieces of three-dimensional grid data) are obtained in the present embodiment, wherein 20 pieces of functional region layout derived auxiliary data, 110 pieces of EM conductivity derived auxiliary data, and 29 pieces of ERT resistivity derived auxiliary data are obtained.
And D, acquiring derived auxiliary data corresponding to each soil sampling point based on the three-dimensional derived auxiliary data of the target area, taking the derived auxiliary data corresponding to the soil sampling points as prediction variables and the measured value of the concentration of the soil pollutants corresponding to the soil sampling points as target variables, and training aiming at a preset machine learning model to acquire a three-dimensional prediction model of the concentration of the soil pollutants of the target area.
(4.1) firstly, respectively standardizing the derived auxiliary data (namely 159 three-dimensional grid data) so as to eliminate the order difference among different variables, then, carrying out dimensionality reduction on the derived auxiliary data by using a principal component analysis method to obtain 14 principal components with characteristic values larger than or equal to 1 in total, wherein the cumulative contribution rate of variance is 96%;
(4.2) extracting main component scores of derivative auxiliary data corresponding to 282 soil sampling point positions, constructing a soil sampling point data set by using Zn concentration data and 14 main components, and randomly dividing the soil sampling point data set into a training set and a verification set according to the proportion of 8;
(4.3) based on the soil sampling point training set, with the principal component score of the derived auxiliary data as a prediction variable and the Zn concentration of the soil as a target variable, establishing a three-dimensional spatial prediction model of the Zn concentration of the soil by using regression of a support vector machine, wherein in the model construction process, the optimal parameters are set through a 5-fold cross validation result of the training set data, and the method specifically comprises the following steps: the kernel function is set to a gaussian radial basis function, the penalty factor C is set to 100, and the insensitive loss function gamma is set to 0.1.
And E, calculating the root mean square error between the soil pollutant concentration predicted value of the soil sampling point obtained based on the soil pollutant concentration three-dimensional prediction model and the corresponding soil pollutant concentration measured value by applying an independent verification method, namely obtaining the prediction precision of the soil pollutant concentration three-dimensional prediction model.
(5.1) predicting the grid positions in the defined three-dimensional grid one by using the soil Zn concentration three-dimensional prediction model constructed in the fourth step to obtain a three-dimensional space distribution map of the site soil Zn concentration, as shown in FIG. 5 (left);
and (5.2) predicting the Zn concentration of the positions of the soil sampling points in the verification set by using the soil Zn concentration three-dimensional prediction model constructed in the fourth step, and calculating the Root Mean Square Error (RMSE) between the predicted value and the measured value of the Zn concentration of the positions of the soil sampling points in the verification set for evaluating the spatial prediction precision.
In this embodiment, the correlation coefficient between the original data of the three-dimensional distribution of resistivity (fig. 3) and the Zn concentration of soil is only 0.29, and the correlation coefficients between the maximum derived auxiliary data of the sliding window of resistivity (fig. 4 right) and the variance derived auxiliary data of the sliding window of resistivity (fig. 11 × 3 × 3) obtained after the development of the auxiliary data and the Zn concentration of soil are respectively increased to 0.5 and 0.51; in addition, the correlation coefficient between the resistivity z-direction (i.e., vertical direction) gradient-derived auxiliary data and the Zn concentration is increased to 0.31. The correlation between the derived auxiliary data and the soil contamination concentration is significantly improved compared to the original auxiliary data.
In the embodiment, the RMSE of the soil Zn concentration three-dimensional prediction independent verification is 175.01 mg/kg, while the RMSE of the traditional IDW space interpolation method is as high as 242.33 mg/kg by adopting the same independent verification data set, and compared with the traditional IDW interpolation method, the prediction error of the method disclosed by the invention is reduced by about 28%. In addition, the Zn concentration predicted by the method (on the left of figure 5) and the traditional IDW interpolation method (on the right of figure 5) is 841.00 mg/kg and 861.00 mg/kg respectively at the maximum value, and 3.15 mg/kg and 42.6 mg/kg at the minimum value respectively, which proves that the method can obviously reduce the effect of 'undervalue overestimation' of the pollutant concentration, thereby reducing the smoothing effect and better depicting the three-dimensional space distribution detail characteristics of the pollutant concentration.
The embodiment can prove that the correlation between the field pollution auxiliary data such as the functional area layout and the underground physical property and the like and the pollutant concentration can be improved by further developing the field pollution auxiliary data, and the field pollution prior information hidden in the auxiliary data is fully utilized, so that the three-dimensional space prediction precision of the field soil pollutant concentration is improved, and the smooth effect of a prediction result is reduced.
The embodiments of the present invention have been described in detail with reference to the drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the gist of the present invention.
Claims (9)
1. A field soil pollutant concentration three-dimensional space prediction method based on multi-source auxiliary data development is characterized by comprising the following steps: obtaining a three-dimensional prediction model of the soil pollutant concentration of the target area through the steps A to D, and applying the three-dimensional prediction model of the soil pollutant concentration to realize three-dimensional distribution prediction of the soil pollutant concentration of the underground space of the target area through the step i;
step A, acquiring soil pollutant concentration measured values of soil sampling points at preset depths at preset drilling positions in a target area, and a multi-source auxiliary data set corresponding to the target area and containing preset data types, and then entering step B;
b, preprocessing a multi-source auxiliary data set corresponding to the target area by a spatial interpolation method based on a three-dimensional grid preset in the underground space of the target area, updating the multi-source auxiliary data set corresponding to the target area, and entering the step C;
c, according to a preset three-dimensional grid of the underground space of the target area, aiming at the multi-source auxiliary data set corresponding to the target area, executing data derivation processing to obtain three-dimensional derived auxiliary data of the target area, and then entering the step D;
step D, obtaining derived auxiliary data corresponding to each soil sampling point respectively based on the three-dimensional derived auxiliary data of the target area, taking the derived auxiliary data corresponding to the soil sampling points as a prediction variable and the measured value of the concentration of the soil pollutants corresponding to the soil sampling points as a target variable, and training aiming at a preset machine learning model to obtain a three-dimensional prediction model of the concentration of the soil pollutants of the target area;
and i, respectively aiming at each grid of the underground space of the target area, obtaining derived auxiliary data corresponding to the grid, applying a soil pollutant concentration three-dimensional prediction model to obtain a soil pollutant concentration prediction value of each grid, and further realizing three-dimensional prediction of the soil pollutant concentration of the underground space of the target area.
2. The field soil pollutant concentration three-dimensional space prediction method based on multisource auxiliary data development as claimed in claim 1, characterized in that: step E is also included, after step D is executed, step E is entered;
and E, calculating the root mean square error between the soil pollutant concentration predicted value of the soil sampling point obtained based on the three-dimensional soil pollutant concentration prediction model and the corresponding soil pollutant concentration measured value by using an independent verification method, namely obtaining the prediction precision of the three-dimensional soil pollutant concentration prediction model.
3. The field soil pollutant concentration three-dimensional space prediction method based on multisource auxiliary data development as claimed in claim 1, characterized in that: and B, correspondingly setting a multi-source auxiliary data set of each preset data type in the target area in the step A, wherein the multi-source auxiliary data set comprises area ranges of various preset functional areas in the target area, conductivity two-dimensional measuring point data of the target area corresponding to each preset electromagnetic detection frequency and two-dimensional section inversion resistivity data of each preset measuring line position corresponding to the target area.
4. The field soil pollutant concentration three-dimensional space prediction method based on multisource auxiliary data development of claim 3, characterized in that: and measuring the underground conductivity of the target area by an electromagnetic induction method EM to obtain conductivity two-dimensional measuring point data of the target area under the corresponding preset electromagnetic detection frequencies.
5. The field soil pollutant concentration three-dimensional space prediction method based on multisource auxiliary data development as claimed in claim 3, characterized in that: and measuring the underground resistivity of the target region through a resistivity tomography ERT to obtain two-dimensional section inversion resistivity data of the target region corresponding to each preset measuring line position.
6. The field soil pollutant concentration three-dimensional space prediction method based on multisource auxiliary data development as claimed in claim 1, wherein: in the step B, based on conductivity two-dimensional measuring point data of the target area in the multi-source auxiliary data set corresponding to each preset electromagnetic detection frequency, interpolation processing is carried out on a two-dimensional grid in the horizontal direction of a three-dimensional grid of the underground space of the target area through a spatial interpolation method, and soil conductivity two-dimensional spatial distribution of the target area corresponding to each electromagnetic detection frequency is obtained;
and performing interpolation processing on the underground space three-dimensional grid of the target area by a spatial interpolation method based on the resistivity data of the two-dimensional section at each preset survey line position corresponding to the target area in the multi-source auxiliary data set to obtain the resistivity three-dimensional spatial distribution corresponding to the target area.
7. The field soil pollutant concentration three-dimensional space prediction method based on multisource auxiliary data development as claimed in claim 6, characterized in that: the spatial interpolation method in the step B is an IDW interpolation method or a kriging interpolation method.
8. The field soil pollutant concentration three-dimensional space prediction method based on multisource auxiliary data development as claimed in claim 1, wherein: the step C comprises the following steps C1 to C4;
step C1, based on the area range of various preset functional areas in the target area, aiming at each grid in the target area, if the grid is positioned in the preset functional area, the grid is assigned to be 1, if the grid is not positioned in the preset functional area, the grid is assigned to be 0, and the assignment of each grid in the target area is completed, namely, the three-dimensional derived auxiliary data of the functional area type corresponding to the target area is formed; meanwhile, respectively calculating the nearest Euclidean distance from each grid to the edge of each functional area aiming at each grid in the target area, namely obtaining the nearest Euclidean distance from each grid in the target area to the edge of each functional area, and forming three-dimensional derivative auxiliary data of the distance from the functional area corresponding to the target area; then entering step C2;
step C2, acquiring the minimum value, the maximum value, the median value, the mean value and the variance among the conductivities of the grids in the horizontal direction of the target area under the electromagnetic detection frequencies respectively based on the two-dimensional spatial distribution of the conductivities of the target area under the electromagnetic detection frequencies respectively, and forming two-dimensional derived auxiliary data of the conductivity statistics corresponding to the target area; meanwhile, based on the sliding of two-dimensional sliding windows with preset sizes in the horizontal direction of the target area, the minimum value, the maximum value, the median value, the mean value, the variance and the gradient in the two-dimensional direction of the conductivity of each grid in each two-dimensional sliding window under each electromagnetic detection frequency are obtained, and two-dimensional derivative auxiliary data of the conductivity statistics of the sliding window corresponding to the target area are formed; then entering step C3;
step C3, obtaining the minimum value, the maximum value, the median value, the mean value and the variance among the depth resistivities of each grid in the horizontal direction of the target area based on the three-dimensional spatial distribution of the resistivity corresponding to the target area, and forming two-dimensional derived auxiliary data of the resistivity statistics among the depths corresponding to the target area; meanwhile, based on the sliding of the three-dimensional sliding windows with preset sizes on the three-dimensional grids of the underground space of the target area, the minimum value, the maximum value, the median value, the mean value, the variance and the gradient in the three-dimensional direction of the grid resistivity in each three-dimensional sliding window are obtained, and three-dimensional derived auxiliary data of the resistivity statistics of the sliding window corresponding to the target area are formed; then entering step C4;
and C4, copying the two-dimensional derived auxiliary data of the conductivity statistics, the two-dimensional derived auxiliary data of the conductivity statistics of the sliding window and the two-dimensional derived auxiliary data of the resistivity statistics among depths corresponding to the target area along the depth direction of the three-dimensional grid to obtain three-dimensional derived auxiliary data of the conductivity statistics, the three-dimensional derived auxiliary data of the conductivity statistics of the sliding window and the three-dimensional derived auxiliary data of the resistivity among depths corresponding to the target area, combining the three-dimensional derived auxiliary data of the type of the functional area, the three-dimensional derived auxiliary data of the distance of the functional area and the three-dimensional derived auxiliary data of the resistivity of the sliding window corresponding to the target area to form three-dimensional derived auxiliary data of the target area, and then entering the step D.
9. The field soil pollutant concentration three-dimensional space prediction method based on multisource auxiliary data development as claimed in claim 1, wherein: the step D comprises the following steps D1 to D3;
step D1, standardizing three-dimensional derived auxiliary data of a target area, obtaining each principal component data in the derived auxiliary data by using a principal component analysis method, and then entering step D2;
step D2, acquiring data of each soil sampling point corresponding to each main component data type respectively based on the three-dimensional derivative auxiliary data of the target area to form derivative auxiliary data corresponding to each soil sampling point respectively, and then entering step D3;
and D3, training a support vector machine regression model based on the soil sampling points by taking the derived auxiliary data corresponding to the soil sampling points as a prediction variable and the measured value of the soil pollutant concentration corresponding to the soil sampling points as a target variable to obtain a three-dimensional prediction model of the soil pollutant concentration in the target area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210760419.1A CN115292890A (en) | 2022-06-29 | 2022-06-29 | Site soil pollutant concentration three-dimensional space prediction method based on multi-source auxiliary data development |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210760419.1A CN115292890A (en) | 2022-06-29 | 2022-06-29 | Site soil pollutant concentration three-dimensional space prediction method based on multi-source auxiliary data development |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115292890A true CN115292890A (en) | 2022-11-04 |
Family
ID=83823170
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210760419.1A Pending CN115292890A (en) | 2022-06-29 | 2022-06-29 | Site soil pollutant concentration three-dimensional space prediction method based on multi-source auxiliary data development |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115292890A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116167285A (en) * | 2023-02-27 | 2023-05-26 | 北京市生态环境保护科学研究院 | Organic pollutant migration prediction method and device and electronic equipment |
| CN116773781A (en) * | 2023-08-18 | 2023-09-19 | 北京建工环境修复股份有限公司 | Pollution analysis method, system and medium for perfluorinated compounds in soil |
-
2022
- 2022-06-29 CN CN202210760419.1A patent/CN115292890A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116167285A (en) * | 2023-02-27 | 2023-05-26 | 北京市生态环境保护科学研究院 | Organic pollutant migration prediction method and device and electronic equipment |
| CN116773781A (en) * | 2023-08-18 | 2023-09-19 | 北京建工环境修复股份有限公司 | Pollution analysis method, system and medium for perfluorinated compounds in soil |
| CN116773781B (en) * | 2023-08-18 | 2023-12-05 | 北京建工环境修复股份有限公司 | Pollution analysis method, system and medium for perfluorinated compounds in soil |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Linde et al. | Geological realism in hydrogeological and geophysical inverse modeling: A review | |
| Auken et al. | A review of airborne electromagnetic methods with focus on geotechnical and hydrological applications from 2007 to 2017 | |
| Fernández-Martínez et al. | Particle swarm optimization applied to solving and appraising the streaming-potential inverse problem | |
| He et al. | Transition probability‐based stochastic geological modeling using airborne geophysical data and borehole data | |
| Tronicke et al. | Crosshole traveltime tomography using particle swarm optimization: A near-surface field example | |
| Fienen et al. | An interactive Bayesian geostatistical inverse protocol for hydraulic tomography | |
| CN115292890A (en) | Site soil pollutant concentration three-dimensional space prediction method based on multi-source auxiliary data development | |
| MX2007000380A (en) | Computer-based method for while-drilling modeling and visualization of layered subterranean earth formations. | |
| Wang et al. | Anisotropic singularity: A novel way to characterize controlling effects of geological processes on mineralization | |
| Allroggen et al. | Attribute-based analysis of time-lapse ground-penetrating radar data | |
| Jaumann et al. | Soil hydraulic material properties and layered architecture from time-lapse GPR | |
| CN110286416B (en) | Fast two-dimensional density inversion method based on physical property function | |
| CN110133644B (en) | Ground penetrating radar three-dimensional forward modeling method based on interpolation scale function method | |
| Hubbard et al. | Estimation of soil classes and their relationship to grapevine vigor in a Bordeaux vineyard: advancing the practical joint use of electromagnetic induction (EMI) and NDVI datasets for precision viticulture | |
| Bortolozo et al. | VES/TEM 1D joint inversion by using Controlled Random Search (CRS) algorithm | |
| CN117274520A (en) | Mapping method for geological exploration | |
| CN105372631A (en) | Polarizing radar inversion method based on genetic-partial least square algorithm, and application of polarizing radar inversion method | |
| Jaumann et al. | Effect of unrepresented model errors on estimated soil hydraulic material properties | |
| Pang et al. | 3D multi-scale resistivity inversion method applied in the tunnel face to borehole observations for tunnel-ahead prospecting | |
| CN115857028A (en) | Time-space quantitative expression method of hydrogeological parameters based on non-contact observation method | |
| CN115293024A (en) | Site soil pollutant concentration three-dimensional spatial interpolation method integrating multi-source cooperative variables | |
| CN115420778A (en) | Method and system for quickly investigating and positioning shallow soil pollution | |
| Christensen | Interpretation attributes derived from airborne electromagnetic inversion models using the continuous wavelet transform | |
| Stephan et al. | Adding realistic noise models to synthetic ground‐penetrating radar data | |
| CN113406707A (en) | Magnetotelluric multi-scale and multi-time-period detection method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |