CN114970302A - Regional underground water condition prediction method based on underground water monitoring system - Google Patents
Regional underground water condition prediction method based on underground water monitoring system Download PDFInfo
- Publication number
- CN114970302A CN114970302A CN202210500900.7A CN202210500900A CN114970302A CN 114970302 A CN114970302 A CN 114970302A CN 202210500900 A CN202210500900 A CN 202210500900A CN 114970302 A CN114970302 A CN 114970302A
- Authority
- CN
- China
- Prior art keywords
- prediction
- observation
- prediction result
- underground water
- water
- 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.)
- Granted
Links
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 176
- 238000012544 monitoring process Methods 0.000 title claims abstract description 44
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000005086 pumping Methods 0.000 claims abstract description 71
- 239000003673 groundwater Substances 0.000 claims description 96
- 238000009826 distribution Methods 0.000 claims description 54
- 230000006870 function Effects 0.000 claims description 48
- 238000004364 calculation method Methods 0.000 claims description 18
- 239000002245 particle Substances 0.000 claims description 18
- 238000009499 grossing Methods 0.000 claims description 17
- 230000009466 transformation Effects 0.000 claims description 17
- 238000004422 calculation algorithm Methods 0.000 claims description 14
- 238000000342 Monte Carlo simulation Methods 0.000 claims description 11
- 238000012417 linear regression Methods 0.000 claims description 10
- 238000004458 analytical method Methods 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 9
- 230000008878 coupling Effects 0.000 claims description 6
- 238000010168 coupling process Methods 0.000 claims description 6
- 238000005859 coupling reaction Methods 0.000 claims description 6
- 238000004088 simulation Methods 0.000 claims description 6
- 238000001514 detection method Methods 0.000 claims description 5
- 238000007619 statistical method Methods 0.000 claims description 5
- 230000010354 integration Effects 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 4
- 238000005457 optimization Methods 0.000 claims description 4
- 238000010835 comparative analysis Methods 0.000 claims description 3
- 238000003708 edge detection Methods 0.000 claims description 3
- 238000001556 precipitation Methods 0.000 claims description 3
- 239000002349 well water Substances 0.000 claims description 3
- 235000020681 well water Nutrition 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 2
- 238000012216 screening Methods 0.000 claims description 2
- 230000001131 transforming effect Effects 0.000 claims description 2
- 229910052731 fluorine Inorganic materials 0.000 claims 1
- 125000001153 fluoro group Chemical group F* 0.000 claims 1
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000009977 dual effect Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 238000013398 bayesian method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000002262 irrigation Effects 0.000 description 1
- 238000003973 irrigation Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000002040 relaxant effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000003911 water pollution 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
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
- G06Q50/06—Energy or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A20/00—Water conservation; Efficient water supply; Efficient water use
- Y02A20/152—Water filtration
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Water Supply & Treatment (AREA)
- Tourism & Hospitality (AREA)
- Public Health (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Human Resources & Organizations (AREA)
- Artificial Intelligence (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention provides a regional underground water regime prediction method based on an underground water monitoring system, which comprises the following steps: the method has the advantages that through observation data in the underground water observation system, real-time monitoring data are combined, a pumping quantity prediction function and a pumping quantity-underground water level prediction function of each observation well are established based on Support Vector Regression (SVR), and the underground water level is predicted in real time, so that the problems that the traditional underground water regime estimation modeling is long in time consumption, needs a large number of hydrogeological parameters, is limited by factors such as complex geological conditions, uncertainty of water regime prediction parameters, mathematical modeling burden and the like, does not form a precise regional underground water regime quantitative prediction tool, cannot refine accurate prediction of underground water regimes of all parts of a region, and cannot well serve underground water management of city, town and country regions are solved.
Description
Technical Field
The invention belongs to the technical field of computer machine learning, and particularly relates to a regional underground water regime prediction method based on an underground water monitoring system.
Background
Under the influence of climate change and uneven water resource distribution in most areas of China, the application of water resources is more severe, and due to complex hydrogeological conditions, the hydrological data of related underground water is deficient. Because the groundwater has the characteristics of stable water quantity, good water quality and the like, the groundwater is one of important water sources for agricultural irrigation, enterprise production and urban resident life, and particularly in arid and semiarid regions with water shortage on the ground surface, the groundwater is often used as a local main water source. Underground water resources play a crucial role in regional water supply, and if the underground water resources are not properly managed, excessive use of the underground water resources easily causes the problems of subsidence, collapse and the like of stratums, causes water resource attenuation in local areas, even causes the problem of underground water pollution, and brings serious environmental problems. In 2020, the underground water level of the whole North China plain tends to decrease due to less rainfall in recent years. Therefore, it is necessary to enhance the reasonable use of underground water resources, implement scientific and effective supervision, protect the safety of underground water, and promote the sustainable development and utilization of underground water resources.
The traditional estimation of the groundwater conditions is to calculate the available water amount of a groundwater system by various hydraulic tests and numerical modes, the modeling is long in time consumption and needs a large amount of hydrogeological parameters, the modeling is limited by factors such as complex geological conditions, uncertainty of the water condition prediction parameters, mathematical modeling burden and the like, a precise regional groundwater condition quantitative prediction tool is not formed, accurate prediction of groundwater conditions of all parts of a region cannot be refined, regional water supply management cannot be well served, especially, the water supply management cannot be well served for the region, water supply is limited and water supply time is allocated to the whole region according to a single early warning parameter model, inconvenience is caused to lives of residents, and even enterprise production or engineering schedule is influenced.
Disclosure of Invention
Therefore, the technical problem to be solved by the present invention is to provide a regional groundwater regime prediction method based on a groundwater monitoring system, which can solve the problems that the traditional groundwater regime estimation modeling takes long time and needs a large amount of hydrogeological parameters, is limited by factors such as complex geological conditions, uncertainty of the water regime prediction parameters, mathematical modeling burden and the like, does not form a precise regional groundwater regime quantitative prediction tool, cannot refine accurate prediction of groundwater regimes of all parts of a region, and cannot well serve groundwater management of urban, town and rural regions.
In order to solve the above problems, the present invention provides a regional groundwater regime prediction method based on a groundwater monitoring system, the method comprising the steps of:
s1: establishing a water pumping amount prediction function and a water pumping amount-underground water level prediction function of each observation well by combining the observation data in the underground water observation system with the real-time monitoring data and based on Support Vector Regression (SVR), and predicting the underground water level in real time;
s2: calculating the relation between pumping and water level leakage and drop of a pumping well and a corresponding observation well by adopting a Markov-Monte Carlo method, simultaneously calculating a plurality of underground water regimen parameters by utilizing observation data monitored by an observation network in real time, obtaining particles to approximate posterior probability distribution by carrying out iterative screening on samples based on probability distribution established by a Bayesian theory, and simultaneously calculating water level prediction results of the observation wells and water pumping amount prediction results of the observation wells to avoid coupling among the prediction parameters;
s3: according to the coordinates and distribution conditions of the observation wells, a prediction area covering all the observation wells is defined, the prediction area covers the whole monitoring network area, grid division is carried out on the whole prediction area, and n multiplied by n grid points are generated;
s4: constructing a Thiessen polygon of a monitoring network area, and transforming a corresponding coordinate set U by combining Euclidean distance e Set of coordinates U incorporated into observation well m In the middle, a two-dimensional array U is formed m+e And then merging the vertex coordinate set U of the Thiessen polygon V Forming a coordinate point set U of a two-dimensional array representing set observation well coordinates, Euclidean distance transformation coordinates and Thiessen polygon vertex coordinates m+e+V ；
Simulating a prediction area covering the whole monitoring network according to the Thiessen polygon constructed by the observation area by utilizing the water level prediction result of the observation well, assigning values to the generated n multiplied by n grid points, and setting the Thiessen polygon V of the observation well i in the Thiessen polygon i Assigning values to the regional grid points, and assigning the values to the underground water level prediction result of the observation well i and the water pumping amount prediction result of the observation well V i The area grid points generate a preliminary prediction result of the water pumping amount of the area and simultaneously generate a preliminary prediction result of the groundwater level of the area;
s5: constructing a convex polygon point set U containing all scattered points through observation well coordinates, Euclidean distance transformation coordinates and Thiessen polygon vertex coordinates m+e+V Dividing the convex polygon into triangles which are not intersected with each other, adjusting adjacent triangles to obtain weighted triangulation, calculating by interpolation values of the triangles, correcting according to preliminary prediction results corresponding to various points in a coordinate point set, and correcting generated preliminary prediction results of groundwater level-water pumping amount area distribution in real time;
s6: and smoothing the corrected groundwater level-pumping amount distribution region water regime prediction result by adopting a Gaussian smoothing function to obtain a final prediction result, further correcting the prediction result by adopting gradient dual-threshold detection, performing statistical analysis on the groundwater regime prediction result by adopting a support vector regression algorithm and the groundwater regime prediction result by adopting a Markov-Monte Carlo method algorithm, creating an array of storage results according to the accuracy of comparison residual statistical distribution comparison analysis prediction results, and determining a storage optimization prediction model as a stable groundwater level prediction system model.
Optionally, S1 specifically includes the following steps:
s11: establishing a training sample set corresponding to different observation well coordinates according to a time sequence
Historical observation data representing water pumping quantity, and establishing a training sample set corresponding to different observation well coordinates according to a time sequence
Historical observation data representing groundwater levels;
s12: training model functions in sample space under the following constraints
y i E { -1, +1}, relaxation variable ξ i Not less than 0, relaxation variable, η i ≧ 0, i ═ 0,1,2 … t, i denoting the ith time point; t represents the total number of observation time points; the loss function is:
under the following constraint
Obtaining a model function f for predicting the groundwater level 2 (x a ，x b ) Wherein x is a Representing the water pumping amount; x is the number of b Indicating groundwater level; namely, the water pumping amount and the groundwater level at the future moment are predicted in real time;
s13: observe data of water pumping quantity
And historical observation data of groundwater level
And performing linear regression, constructing various pumping quantity-underground water level linear regression models, calculating to obtain residual errors, performing comparative analysis, comparing the prediction result of the linear regression model function with the underground water level result predicted by the support vector regression, and analyzing the prediction precision to obtain an optimized model.
Optionally, in step S2, the water pumping-water level leakage relation between the water pumping well and the corresponding observation well is calculated based on the markov-monte carlo method, so as to avoid coupling between the prediction parameters, which specifically includes the following steps:
s21: establishing a probability model for continuously calculating the water regime parameters based on a Bayesian theory, wherein the water regime observation historical data can simultaneously calculate the probability distribution of a plurality of water regime parameters including the water pumping amount and the groundwater level;
setting t 0 Time and t 0 Each observation well before the moment and the collected water regime observation data are X t Posterior probability P (x) a ,x b ,t|X t ) Can be expressed as:
wherein x is a Representing the water pumping amount; x is the number of b Indicating groundwater level; t is t 0 Predicting a starting time; t is t
Is a predicted time;
s22: simplified model phi for taking samples t (x a ，x b ，t|X t ) Fit to the q distribution, q (-) is the prior distribution P (phi) t |φ t-1 ) I denotes the ith particle corresponding to the q distribution and is denoted as
t represents the time, i is 1, …, N cyclesLoop iteration;
the posterior probability distribution can be approximated as:
wherein, the first weight coefficient v of the ith particle at the time t t i The total number can be normalized to 1,
n is the total number of particles, and delta represents equal proportional probability;
based on updated observation data X t And participating in the calculation of the water regime parameters, and re-normalizing the weights of all the particles into the following weight:
wherein, L (X) t |φ t )＝L(X t |x a ，x b And t) as a second weight coefficient, calculating the pumping-water level leakage and drop relation between the pumping well and the corresponding observation well through a plurality of particles with weights, and constructing a regional underground water condition prediction model.
Optionally, S3 specifically includes the following steps:
s31: defining a prediction area covering all observation wells according to the coordinates and distribution conditions of the observation wells, wherein the prediction area covers the whole monitoring network area, and performing grid division and establishment of an observation well coordinate array on the prediction area based on real-time monitoring data provided by a monitoring network;
s32: creating a two-dimensional array of grid point region centers to have horizontal and vertical coordinates (x) according to the grids divided by the prediction region range n ，y n ) Centered n × n region U n ＝{(x 1 ,y 1 ),(x 2 ,y 2 ),…,(x n-1 ,y n-1 ),(x n ,y n ) Initializing all points in the array as a unit matrix function, wherein l and h are grids adjacent to the x direction and the y direction respectivelySpacing between dots, x n ＝l×n，y n ＝h×n；
S33: constructing a two-dimensional array U corresponding to the position coordinates of the observation well m ＝{(x 1 ,y 1 ),(x 2 ,y 2 ),…,(x m-1 ,y m-1 ),(x m ,y m )}，(x m ，y m ) Is U m And predicting the water regime corresponding to the m-th observation well.
Optionally, S4 specifically includes the following steps:
s41: constructing a Thiessen polygon of the monitoring net area: constructing a Thiessen polygon for the prediction area according to the coordinates of the N observation wells, and generating the Thiessen polygon by taking the adjacent observation wells as the perpendicular bisector, namely V N A drawing;
v constructed according to observation area by using observation well water level prediction result N Simulating a prediction area covering the whole monitoring network, assigning values to the generated n multiplied by n grid points, and setting a Thiessen polygon V of an observation well i in the Thiessen polygons i Assigning values to the regional grid points, and assigning the values to the underground water level prediction result of the observation well i and the water pumping amount prediction result of the observation well V i Area grid points;
s42: with D f (x i ，y i ) Representing coordinates (x) i ，y i ) Is calculated as a Euclidean distance transform of f (x) j ，y j ) Representing coordinates (x) j ，y j ) (ii) a consumption function of (x) i ,y i )∈U n ，(x j ,y j )∈U n ，D f (x i ,y i )＝min(((x i ,y i )-(x j ,y j )) 2 +f(x j ,y j ) Go through n × n region U n ＝{(x 1 ,y 1 ),(x 2 ,y 2 ),…,(x n-1 ,y n-1 ),(x n ,y n ) And calculating to obtain a coordinate set U corresponding to Euclidean distance transformation e Integration into observation well coordinate set U m In the middle, a two-dimensional array U is formed m+e And then combining the coordinate sets U of the vertices of the Thiessen polygon V Forming a two-dimensional array representation setCoordinate set U corresponding to coordinate and Euclidean distance transformation of observation well e Coordinate point set U of coordinates and Thiessen polygon vertex coordinates m+e+V 。
Optionally, S5 specifically includes the following steps:
s51: corresponding the generated two-dimensional array to the position coordinates of the observation well and the Euclidean distance transformation corresponding coordinate set U e Two-dimensional array U of coordinates and Thiessen polygon vertex coordinates m+e+V Each coordinate point corresponds to U m+e+V The corresponding water regime prediction result of the observation well in (1), U m+e+V Each coordinate point corresponds to a time series groundwater regime prediction result, which is measured by U m+e+V The points in the triangulation network are used as end points to form a closed line section of the triangulation network;
S52：U m+e+V the points in the triangle form a triangulation network convex polygon as end points, and the triangulation network convex polygon is positioned at a grid point U n The point set of the middle envelope is U k ＝{(x 1 ,y 1 ),(x 2 ,y 2 ),…,(x k-1 ,y k-1 ),(x k ,y k ) Is weighted by ω for any two dot shapes ij Dividing the convex polygon into a plurality of mutually disjoint triangles to generate a weighted triangulation network;
s53: the structure comprises U m+e+V Triangulating all coordinate point regions, recursively obtaining triangulation of the triangulation network, performing triangle interpolation calculation on the triangulation network, and calculating U for each coordinate point set in the triangulation network k And further correcting the corresponding time series water pumping quantity prediction result and the groundwater level prediction result, and optimizing and updating the generated preliminary prediction result.
Optionally, S6 specifically includes the following steps:
S61：U n ＝{(x 1 ,y 1 ),(x 2 ,y 2 ),…,(x n-1 ,y n-1 ),(x n ,y n ) Center (x) of grid points in the region i ,y i ) The gaussian kernel function is constructed as follows:
wherein, sigma is the standard deviation of Gaussian distribution in the Gaussian kernel generator, and represents the discrete degree of the prediction result data;
s62: performing Gaussian smoothing on the corrected water pumping amount prediction result and the groundwater level prediction result, calculating the horizontal gradient and the longitudinal gradient of the prediction result by adopting a sobel function, setting double thresholds by adopting a canny function for edge detection, performing double threshold detection by utilizing a weak gradient value, further correcting the prediction result of the whole area, and obtaining a final prediction result;
s63: based on the actual measurement data, generating a regional groundwater level distribution simulation result, smoothing the generated prediction result by adopting a Gaussian smoothing function, further correcting to obtain a final prediction result, and simulating regional distribution on a groundwater regimen prediction result by adopting an SVR regression algorithm and a groundwater regimen prediction result based on Markov-Monte Carlo algorithm;
s64: and comparing results predicted based on different algorithms, carrying out statistical analysis on the prediction accuracy and the calculation efficiency, adding the information of air temperature, rainfall or/and snowfall and earth surface settlement during monitoring, analyzing the influence of different conditions on the final prediction result, and storing the regional groundwater regime prediction model with robustness according to the analysis of the accuracy and the calculation efficiency.
Advantageous effects
The regional underground water regime prediction method based on the underground water monitoring system can realize prediction of the whole underground water regime in the whole monitoring region, is not influenced by the positions of observation wells and the distribution condition of the observation wells, can predict only by acquiring observation data from a plurality of limited observation wells and pumping wells (groups) which are discretely distributed in the monitoring region no matter whether the observation data are sufficient or not, can perform comparison analysis of precision and calculation efficiency aiming at a plurality of different prediction methods by the existing observation data, refine the underground water level distribution of each part of the prediction region, and reflect the underground water regime of each part of the prediction region in real time. In addition, the method can realize the prediction of the groundwater condition by establishing the pumping capacity-groundwater level leakage relation function through establishing the pumping capacity prediction function, can also reduce the coupling between the pumping capacity and the groundwater level, and simultaneously establishes the pumping capacity-groundwater level leakage relation prediction function based on the Markov-Monte Carlo method. And according to the precision analysis of the actual case prediction result, storing a region underground water regime prediction model with robustness. The problems that the traditional underground water regimen estimation modeling is long in time consumption, needs a large number of hydrogeological parameters, is limited by complex geological conditions, uncertainty of water regimen prediction parameters, mathematical modeling burden and other factors, does not form a precise regional underground water regimen quantitative prediction tool, cannot refine accurate prediction of underground water conditions of all parts of a region, cannot well serve regional water supply management, and particularly, water supply is stopped, water supply is limited, water supply amount is limited and water supply time is allocated to the whole region according to a single early warning parameter model, inconvenience is brought to lives of residents, and even enterprise production or engineering schedule is influenced are solved.
Drawings
FIG. 1 is a flow chart of regional groundwater regime prediction based on a groundwater monitoring system according to an embodiment of the invention;
FIG. 2 is a schematic diagram of a region simulation distribution of groundwater level observation data according to an embodiment of the invention;
Detailed Description
According to an embodiment of the present invention, a regional groundwater condition prediction method based on a groundwater monitoring system, please refer to fig. 1, includes the following steps:
s1: collecting underground water level observation information collected by water level observation equipment of each observation well (group) according to a time sequence, collecting water pumping amount observation information near the observation wells, collecting air temperature and rainfall (snowfall) information, and collecting and sorting extracted observation data according to coordinates of each observation well.
The method comprises the steps of constructing a water pumping amount prediction function and a water pumping amount-underground water level prediction function through a Support Vector Regression (SVR) model by utilizing existing observation data, and predicting the future underground water level.
S1 specifically includes the following steps:
s11: establishing a training sample set corresponding to different observation well coordinates according to a time sequence
Historical observation data representing water pumping quantity, and establishing a training sample set corresponding to different observation well coordinates according to a time sequence
Historical observation data representing groundwater levels.
S12: training a model function in a sample space, determining a width epsilon given to a position of a decision surface, determining mu as a normal vector of the decision surface, determining the direction of the decision surface, and introducing a relaxation variable xi i And η i Under the following constraint
y i Epsilon { -1, +1} relaxation variable ξ i Not less than 0, relaxation variable η i ≧ 0, i ═ 0,1,2 … t, i denotes the ith time point, t denotes the total number of observed time points, the loss function is as follows:
wherein C is>A constant of 0, ε gives the position of the decision surface, μ represents the normal vector of the decision surface, ξ i And η i Obtaining a precipitation prediction model function f for relaxing variables and meeting an optimization target 1 (x a ，t i ) Wherein t is i Denotes time, x a The water pumping amount is expressed under the following constraint conditions
Obtaining a model function f for predicting the groundwater level 2 (x a ，x b )，x a Indicates the amount of pumping, x b And the underground water level is expressed, so that the pumping quantity and the underground water level at the future moment can be predicted in real time.
S13: observe data of water pumping quantity
Historical observation data of groundwater level
And performing linear regression, constructing various pumping quantity-underground water level linear regression models (such as least square, polynomial regression and logarithmic linear regression), and calculating residual errors for comparative analysis. And comparing the prediction result of the linear regression model function with the underground water level result predicted by the SVR in the S12, and analyzing the prediction accuracy.
S2: and (3) estimating the pumping-water level leakage and drop relation between the pumping well (group) and the corresponding observation well based on the Markov-Monte Carlo method, and constructing a regional underground water situation prediction model.
S2 specifically includes the following steps:
s21: based on Bayesian theory (Bayesian method), a probability model for continuous calculation of water regime parameters is established, and probability distribution of a plurality of water regime parameters including water pumping quantity and groundwater level can be estimated simultaneously from water regime observation historical data. Let t 0 Time and t0 the water regime observation data collected by each observation well and nearby before the moment is X t Posterior probability P (x) a ，x b ，t|X t ) Can be expressed as follows:
wherein x is a Indicates the amount of pumping, x b Indicates groundwater level, t 0 And predicting the starting time, wherein t is the predicted time.
S22: with a large number of weighted particles to approximate the posterior distribution, the estimate can be calculated iteratively:
simplified model phi for taking samples t (x a ，x b ，t|X t ) Conforming to the q-distribution, q (-) is generally defined in simplified form as the prior distribution P (φ) t |φ t-1 ) I denotes the ith particle corresponding to the q distribution and is denoted as
t denotes the time instant, starting the loop iteration with i equal to 1, …, N. The posterior probability distribution can be approximated as follows:
wherein, the weight coefficient v of the ith particle at the time t t i The total number of which can be normalized to 1,
n is the total number of particles and δ represents the equal proportional probability.
Observation data X updated with each step t Participating in the estimation of the regimen parameters, all particle weights are re-normalized to
Wherein, L (X) t |φ t )＝L(X t |x a ，x b And t) as a new weight coefficient, estimating the pumping-water level leakage relation of the pumping well (group) and the corresponding observation well through a large number of particles with weights, and constructing a regional underground water condition prediction model.
S3: the method comprises the steps of defining a prediction area covering all observation wells (groups) according to coordinates and distribution conditions of the observation wells, covering the whole monitoring network area by the prediction area, carrying out grid division on the area, providing initial conditions for subsequent initial prediction of a Thiessen polygon combined with Euclidean distance transformation, interpolation calculation and simulation smoothing of the area, dividing the area into n multiplied by n grids, setting transverse and longitudinal distances l and h of the grids, and uniformly initializing the areas by a unit matrix function.
S3 specifically includes the following steps:
s31: a prediction area covering all observation wells (groups) is defined according to the coordinates and distribution conditions of the observation wells, the prediction area covers the whole monitoring network area, and the prediction area is subjected to grid division and the establishment of an observation well coordinate array by using real-time monitoring data provided by the monitoring network as completely as possible;
s32: creating a two-dimensional array of grid point region centers to have horizontal and vertical coordinates (x) according to the grids divided by the prediction region range n ，y n ) Centered n × n region U n ＝{(x 1 ,y 1 ),(x 2 ,y 2 ),…,(x n-1 ,y n-1 ),(x n ,y n ) Initializing all points in the array as an identity matrix function, wherein l and h are the distances between grid points adjacent to each other in the x direction and the y direction respectively, and x is n ＝l×n，y n ＝h×n；
S33: constructing a two-dimensional array U corresponding to the position coordinates of the observation well m ＝{(x 1 ,y 1 ),(x 2 ,y 2 ),…,(x m-1 ,y m-1 ),(x m ,y m )}，(x m ，y m ) Is U m Predicting a water regime prediction result corresponding to the mth observation well;
through the steps, the division of the grids of the prediction region and the establishment of the array of the water situation prediction conditions of the observation well are completed, conditions are provided for subsequent preliminary prediction, correction calculation and smooth calculation, and even if the monitoring networks are distributed sparsely and lack of sufficient observation data, the rainfall and the groundwater level at any position in the prediction region can still be predicted.
S4: and simulating a preliminary prediction result for the prediction area: constructing Voronoi polygons with adjacent observation wells as midperpendicular lines, N observation wells, i.e. V N Dividing the observation well into N Thiessen polygons by taking the coordinates of the observation well as the center, adding one observation well, and generating N +1 Thiessen polygons, namely V N+1 Graph, corresponding coordinate set U combined with Euclidean distance transform e Integration into observation well coordinate set U m In the middle, a new two-dimensional array U is formed m+e And then merging the vertex coordinate set U of the Thiessen polygon V Forming a two-dimensional array representing coordinates of the set observation well, coordinates of Euclidean distance transformation and the vertex coordinates of the Thiessen polygonTarget set of coordinate points U m+e+V So as to simulate the whole prediction area and obtain the preliminary prediction result of the whole area;
and simulating the whole prediction area to obtain a preliminary prediction result of the groundwater regime distribution of the whole area.
S4 specifically includes the following steps:
s41: constructing a Thiessen polygon of the monitoring net area: constructing a Thiessen polygon for the prediction area according to the coordinates of the N observation wells, and generating the Thiessen polygon by taking the adjacent observation wells as a perpendicular bisector, namely V N Figure (a). If an observation well is newly added, a new V is generated N A drawing;
v constructed according to observation area by using observation well water level prediction result N Simulating a prediction area covering the whole monitoring network, assigning values to n multiplied by n grid points generated in S3, and enabling a Thiessen polygon V of an observation well i in the Thiessen polygons i The regional grid points are assigned to the underground water level prediction result of the observation well i, and meanwhile, the water pumping amount prediction result nearby the observation well is assigned to the underground water level prediction result V i The grid points of the region(s),
s42: with D f (x i ，y i ) Representing coordinates (x) i ，y i ) Is calculated as a Euclidean distance transform of f (x) j ，y j ) Representing coordinates (x) j ，y j ) (ii) a consumption function of (x) i ,y i )∈U n ，(x j ,y j )∈U n ，D f (x i ,y i )＝min(((x i ,y i )-(x j ,y j )) 2 +f(x j ,y j ) Go through n × n region U n ＝{(x 1 ,y 1 ),(x 2 ,y 2 ),…,(x n-1 ,y n-1 ),(x n ,y n ) Calculating a coordinate set U corresponding to Euclidean distance transformation e Set of coordinates U incorporated into observation well m In the middle, a new two-dimensional array U is formed m+e And then combining the coordinate sets U of the vertices of the Thiessen polygon V Forming a two-dimensional array representing coordinates of a set observation well, Euclidean distance transformed coordinates, and coordinates of vertices of a Thiessen polygonCoordinate point set U m+e+V 。
The prediction results of the groundwater conditions in the steps S1 and S2 are assigned according to different Thiessen polygons in the Thiessen polygons constructed in the observation area, the Thiessen polygons are assigned as the observation water level or the prediction water level of the observation well by taking the coordinates of the observation well as the center, and the coordinate set U corresponding to the Euclidean distance transformation is combined e Form a new two-dimensional array U m+e And then merging the vertex coordinate set U of the Thiessen polygon V Forming a coordinate point set U of a two-dimensional array representing coordinates of a set observation well, coordinates corresponding to Euclidean distance transformation and vertex coordinates of a Thiessen polygon m+e+V So as to simulate the whole prediction area and obtain the preliminary prediction result of the whole area;
s5: all coordinate points U are constructed through observation well coordinates, Euclidean distance transformation corresponding coordinates and Thiessen polygon vertex coordinates m+e+V Adjusting adjacent triangles, correcting the preliminary prediction result in real time through the weighted triangulation of the maximum convex hull point set, forming two triangles with common sides by observation wells to synthesize a polygon, enclosing all coordinate points, correcting the corresponding water regimen prediction result of the maximum convex polygon point set point by point according to the prediction result of each observation well, and correcting the preliminary prediction result generated in the step S4 in real time;
s5 specifically includes the following steps:
s51: the two-dimensional array U generated in the step S4 corresponds to the position coordinate of the observation well, the coordinate corresponding to Euclidean distance transformation and the vertex coordinate of the Thiessen polygon m+e+V Each coordinate point corresponds to U m+e+V One of the steps S1 and S2 is the prediction result of the corresponding water regime of the observation well, U m+e+V Each coordinate point corresponds to a time series groundwater regime prediction result, which is measured by U m+e+V The points in the triangulation network are used as end points to form a closed line section of the triangulation network;
S52：U m+e+V the points in the triangle form a triangulation network convex polygon as end points, and the triangulation network convex polygon is positioned at a grid point U n The point set of the middle envelope is U k ＝{(x 1 ,y 1 ),(x 2 ,y 2 ),…,(x k-1 ,y k-1 ),(x k ,y k ) Is weighted by ω for any two dot shapes ij And dividing the convex polygon into a plurality of non-intersected triangles to generate the weighted triangulation network.
S53: the structure comprises U m+e+V Dividing triangles of all coordinate point areas, recursively obtaining weight function values corresponding to the triangular net, performing triangle interpolation value calculation on the triangular net, and performing U-shaped interpolation value calculation on each coordinate point set in the triangular net k And further correcting the corresponding time series water pumping amount prediction result and the groundwater level prediction result, and optimizing and updating the preliminary prediction result generated in the step S4.
S6: smoothing the prediction result of the step S5 by adopting a Gaussian smoothing function to obtain a final prediction result; and (4) performing statistical analysis on the groundwater regime prediction result obtained by adopting different regression algorithms in the step S1 and the groundwater regime prediction result obtained by the Markov-Monte Carlo method in the step S2, and analyzing the accuracy of the prediction result.
S61：U n ＝{(x 1 ,y 1 ),(x 2 ,y 2 ),…,(x n-1 ,y n-1 ),(x n ,y n ) Center (x) of grid points in the region i ,y i ) Constructing a Gaussian kernel function
Sigma is the standard deviation of Gaussian distribution in the Gaussian kernel generator and represents the dispersion degree of the prediction result data;
s62: and S5, performing Gaussian smoothing on the corrected pumping quantity prediction result and the groundwater level prediction result, calculating the horizontal and longitudinal gradients of the prediction result by adopting a sobel function, setting dual thresholds by adopting a canny function to perform edge detection, performing dual threshold detection by utilizing a weak gradient value, further correcting the prediction result of the whole area, and obtaining a final prediction result.
S63: according to the simulation results of the water level distribution of the regional groundwater generated in the steps S4 to S6 based on the measured data, smoothing the prediction result generated in the step S5 by using a Gaussian smoothing function, further correcting the smoothing result to obtain a final prediction result, and simulating the regional distribution of the groundwater regimen prediction results of the step S1 by using various different regression algorithms such as support vector regression and the like and the groundwater regimen prediction results based on the Markov-Monte Carlo method in the step S2 by using the steps S4 to S6.
S64: different groundwater regimen prediction algorithms can be loaded in the step S1 and the step S2, results predicted by the different algorithms are compared, the prediction accuracy, the calculation efficiency and the like are analyzed in a statistical mode, information such as air temperature, rainfall and/or snowfall, surface subsidence and the like during monitoring is added, the influence of the detail conditions on the final prediction result is analyzed, and a regional groundwater regimen prediction model with robustness is stored according to the analysis of the accuracy and the calculation efficiency.
As shown in FIG. 2, a numerical simulation process of groundwater level distribution of a certain city construction area and a certain day groundwater level observation data is implemented through steps S4-S6, groundwater level distribution in the area within the range of 150m × 40m is obtained, circle points represent positions of observation wells, colors represent heights of water levels of the observation wells from low to high (-5-25 m) from light to deep, groundwater level distribution interpolated in a triangulation network is obtained through steps S4 and S5, and finally regional groundwater level distribution smoothed through step S6 is obtained.
The method is different from the traditional underground water regime numerical prediction model or machine learning method which adopts single observation wells in discrete distribution for prediction, estimates the underground water regime of the whole area according to an empirical formula, and under the condition that certain areas lack observation data, the estimation has great error on the water regime distribution condition of the whole area, so that the underground water regime prediction is not accurate enough. The invention can realize the prediction of the whole groundwater regimen in the whole monitoring area, is not influenced by the position of the observation well and the distribution condition of the observation well, can predict whether the observation data is sufficient or not by only obtaining the observation data from a plurality of limited observation wells and pumping wells (groups) which are discretely distributed in the monitoring area, and can simultaneously carry out the comparison analysis of precision and calculation efficiency aiming at a plurality of different prediction methods by the existing observation data, refine the groundwater water level distribution of each part of the prediction area, and reflect the groundwater regimen of each part of the prediction area in real time. In addition, the method can realize the prediction of the groundwater condition by establishing the pumping capacity-groundwater level leakage relation function through establishing the pumping capacity prediction function, can also reduce the coupling between the pumping capacity and the groundwater level, and simultaneously establishes the pumping capacity-groundwater level leakage relation prediction function based on the Markov-Monte Carlo method. And according to the precision analysis of the actual case prediction result, storing a region underground water regime prediction model with robustness.
It is readily understood by a person skilled in the art that the advantageous ways described above can be freely combined, superimposed without conflict.
Claims (7)
1. A regional underground water condition prediction method based on an underground water monitoring system is characterized by comprising the following steps:
s1: establishing a water pumping amount prediction function and a water pumping amount-underground water level prediction function of each observation well through observation data in an underground water observation system in combination with real-time monitoring data based on support vector regression, and predicting the underground water level in real time;
s2: calculating the pumping-water level leakage and drop relation between a pumping well and a corresponding observation well by adopting a Markov-Monte Carlo method, simultaneously calculating a plurality of underground water regimen parameters by utilizing observation data monitored by an observation network in real time, obtaining particles to approximate posterior probability distribution by carrying out iterative screening on a sample based on probability distribution established by a Bayesian theory, and simultaneously calculating water level prediction results of the observation wells and water pumping amount prediction results of the observation wells to avoid coupling among the prediction parameters;
s3: according to the coordinates and distribution conditions of the observation wells, a prediction area covering all the observation wells is defined, the prediction area covers the whole monitoring network area, and the whole prediction area is subjected to grid division to generate n multiplied by n grid points;
s4: constructing a Thiessen polygon of a monitoring network area, and transforming a corresponding coordinate set U by combining Euclidean distance e Integration into observation well coordinate set U m In the middle, a two-dimensional array U is formed m+e And then merging the vertex coordinate set U of the Thiessen polygon V Forming a coordinate point set U of a two-dimensional array representing set observation well coordinates, Euclidean distance transformation coordinates and Thiessen polygon vertex coordinates m+e+V ；
Simulating a prediction area covering the whole monitoring network according to the Thiessen polygon constructed by the observation area by utilizing the water level prediction result of the observation well, assigning values to the generated n multiplied by n grid points, and setting the Thiessen polygon V of the observation well i in the Thiessen polygon i Assigning values to the regional grid points, and assigning the values to the underground water level prediction result of the observation well i and the water pumping amount prediction result of the observation well V i The area grid points generate a preliminary prediction result of the water pumping amount of the area and simultaneously generate a preliminary prediction result of the groundwater level of the area;
s5: constructing a convex polygon point set U containing all scattered points through observation well coordinates, Euclidean distance transformation intersection point coordinates and Thiessen polygon vertex coordinates m+e+V Dividing the convex polygon into triangles which are not intersected with each other, adjusting adjacent triangles to obtain weighted triangulation, calculating by interpolation values of the triangles, correcting according to preliminary prediction results corresponding to various points in a coordinate point set, and correcting generated preliminary prediction results of groundwater level-water pumping amount area distribution in real time;
s6: and smoothing the corrected groundwater level-pumping amount distribution region water regime prediction result by adopting a Gaussian smoothing function to obtain a final prediction result, further correcting the prediction result by adopting gradient dual-threshold detection, performing statistical analysis on the groundwater regime prediction result by adopting a support vector regression algorithm and the groundwater regime prediction result by adopting a Markov-Monte Carlo method algorithm, creating an array of stored results according to the accuracy of comparison residual statistical distribution comparison analysis prediction results, and storing an optimized prediction model as a stable groundwater level prediction system model.
2. A regional groundwater situation prediction method based on a groundwater monitoring system as claimed in claim 1, wherein S1 specifically comprises the following steps:
s11: establishing a training sample set corresponding to different observation well coordinates according to a time sequence
Historical observation data representing water pumping quantity, and establishing a training sample set corresponding to different observation well coordinates according to a time sequence
Historical observation data representing groundwater levels;
s12: training model functions in sample space under the following constraints
y i E { -1, +1}, relaxation variable ξ i Not less than 0, relaxation variable η i ≧ 0, i ═ 0,1,2 … t, i denoting the ith time point; t represents the total number of observation time points; the loss function is:
under the following constraint
Obtaining a model function f for predicting the groundwater level 2 (x a ，x b ) Wherein, in the step (A),x a representing the water pumping amount; x is the number of b Indicating groundwater level; namely, the water pumping amount and the groundwater level at the future moment are predicted in real time;
s13: observe data of water pumping quantity
And historical observation data of groundwater level
And performing linear regression, constructing various pumping quantity-underground water level linear regression models, calculating to obtain residual errors, performing comparative analysis, comparing the prediction result of the linear regression model function with the underground water level result predicted by support vector regression, analyzing the prediction precision, and optimizing the prediction model.
3. The method according to claim 1, wherein in step S2, the method for calculating the pumping-water level leakage relationship between the pumping well and the corresponding observation well based on the markov-monte carlo method to avoid the coupling between the prediction parameters includes the following steps:
s21: establishing a probability model for continuously calculating the water regime parameters based on a Bayesian theory, wherein the water regime observation historical data can simultaneously calculate the probability distribution of a plurality of water regime parameters including the water pumping amount and the groundwater level;
setting t 0 Time and t 0 Each observation well before the moment and the collected water regime observation data are X t Posterior probability P (x) a ,x b ,t|X t ) Can be expressed as:
wherein x is a Representing the water pumping amount; x is a radical of a fluorine atom b Indicating groundwater level; t is t 0 Predicting a starting time; t is the predicted time;
s22: simplified model phi for taking samples t (x a ，x b ，t|X t ) Conforming to the q distribution, q (-) being the prior distribution P (phi) t |φ t-1 ) I denotes the ith particle corresponding to the q distribution, denoted by
t represents the time, and is iterated by i being 1, …, N loop;
the posterior probability distribution can be approximated as:
wherein, the first weight coefficient of the ith particle at the time t
The total number can be normalized to 1,
n is the total number of particles; δ represents the equal proportional probability;
based on updated observation data X t And participating in the calculation of the water regime parameters, and re-normalizing the weights of all the particles into the following weight:
wherein, L (X) t |φ t )＝L(X t |x a ，x b And t) as a second weight coefficient, calculating the pumping-water level leakage and drop relation between the pumping well and the corresponding observation well through a plurality of particles with weights, and constructing a regional underground water condition prediction model.
4. The method according to claim 1, wherein S3 specifically comprises the steps of:
s31: defining a prediction area covering all observation wells according to the coordinates and distribution conditions of the observation wells, wherein the prediction area covers the whole monitoring network area, and performing grid division and establishment of an observation well coordinate array on the prediction area based on real-time monitoring data provided by a monitoring network;
s32: creating a two-dimensional array of grid point region centers to have horizontal and vertical coordinates (x) according to the grids divided by the prediction region range n ，y n ) Centered n × n region U n ＝{(x 1 ,y 1 ),(x 2 ,y 2 ),…,(x n-1 ,y n-1 ),(x n ,y n ) Initializing all points in the array as an identity matrix function, wherein l and h are the distances between grid points adjacent to each other in the x direction and the y direction respectively, and x is n ＝l×n，y n ＝h×n；
S33: constructing a two-dimensional array U corresponding to the position coordinates of the observation well m ＝{(x 1 ,y 1 ),(x 2 ,y 2 ),…,(x m-1 ,y m-1 ),(x m ,y m )}，(x m ，y m ) Is U m And predicting the water regime corresponding to the m-th observation well.
5. The method according to claim 1, wherein S4 specifically comprises the steps of:
s41: constructing a Thiessen polygon of the monitoring net area: constructing a Thiessen polygon for the prediction area according to the coordinates of the N observation wells, and generating the Thiessen polygon by taking the adjacent observation wells as the perpendicular bisector, namely V N A drawing;
v constructed according to observation area by using observation well water level prediction result N Simulating a prediction area covering the whole monitoring network, assigning values to the generated n multiplied by n grid points, and setting a Thiessen polygon V of an observation well i in the Thiessen polygons i Assigning values to the regional grid points, and assigning the values to the underground water level prediction result of the observation well i and the water pumping amount prediction result of the observation well V i Area grid points;
s42: with D f (x i ，y i ) Representing coordinates (x) i ，y i ) Is calculated as a Euclidean distance transform of f (x) j ，y j ) Representing coordinates (x) j ，y j ) Is consumed byNumber (x) i ,y i )∈U n ，(x j ,y j )∈U n ，D f (x i ,y i )＝min(((x i ,y i )-(x j ,y j )) 2 +f(x j ,y j ) Go through n × n region U n ＝{(x 1 ,y 1 ),(x 2 ,y 2 ),…,(x n-1 ,y n-1 ),(x n ,y n ) And calculating to obtain a coordinate set U corresponding to Euclidean distance transformation e Integration into observation well coordinate set U m In the middle, a two-dimensional array U is formed m+e And then combining the coordinate sets U of the vertices of the Thiessen polygon V Forming a coordinate point set U of a two-dimensional array representing set observation well coordinates, Euclidean distance transformation coordinates and Thiessen polygon vertex coordinates m+e+V 。
6. The method according to claim 1, wherein S5 specifically comprises the steps of:
s51: the generated two-dimensional array is corresponding to the two-dimensional array U of the position coordinate of the observation well, the Euclidean distance transformation coordinate and the vertex coordinate of the Thiessen polygon m+e+V Each coordinate point corresponds to U m+e+V The corresponding water regime prediction result of the observation well in (1), U m+e+V Each coordinate point corresponds to a time series groundwater regime prediction result, which is measured by U m+e+V The points in the triangulation network are used as end points to form a closed line section of the triangulation network;
S52：U m+e+V the points in the triangle form a triangulation network convex polygon as end points, and the triangulation network convex polygon is positioned at a grid point U n The point set of the middle envelope is U k ＝{(x 1 ,y 1 ),(x 2 ,y 2 ),…,(x k-1 ,y k-1 ),(x k ,y k ) Is weighted by ω for any two dot shapes ij Dividing the convex polygon into a plurality of mutually disjoint triangles to generate a weighted triangulation network;
s53: the structure comprises U m+e+V Triangulating all coordinate point regions, recursively obtaining triangulation of a triangulation network, performing triangle interpolation calculation on the triangulation network,for each coordinate point set U in the triangular net k And further correcting the corresponding time series water pumping quantity prediction result and the groundwater level prediction result, and optimizing and updating the generated preliminary prediction result.
7. The method according to claim 1, wherein S6 specifically comprises the steps of:
S61：U n ＝{(x 1 ,y 1 ),(x 2 ,y 2 ),…,(x n-1 ,y n-1 ),(x n ,y n ) Center (x) of grid points in the region i ,y i ) The gaussian kernel function is constructed as follows:
wherein, sigma is the standard deviation of Gaussian distribution in the Gaussian kernel generator, and represents the discrete degree of the prediction result data;
s62: performing Gaussian smoothing on the corrected water pumping amount prediction result and the groundwater level prediction result, calculating the horizontal gradient and the longitudinal gradient of the prediction result by adopting a sobel function, setting double thresholds by adopting a canny function for edge detection, performing double threshold detection by utilizing a weak gradient value, further correcting the prediction result of the whole area, and obtaining a final prediction result;
s63: based on the actual measurement data, generating a regional groundwater level distribution simulation result, smoothing the generated prediction result by adopting a Gaussian smoothing function, further correcting to obtain a final prediction result, and simulating regional distribution on a groundwater regimen prediction result by adopting an SVR regression algorithm and a groundwater regimen prediction result based on Markov-Monte Carlo algorithm;
s64: and comparing results predicted based on different algorithms, carrying out statistical analysis on the prediction accuracy and the calculation efficiency, adding the information of air temperature, rainfall or/and snowfall and earth surface settlement during monitoring, analyzing the influence of different conditions on the final prediction result, and storing the regional groundwater regime prediction model with robustness according to the analysis of the accuracy and the calculation efficiency.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210500900.7A CN114970302B (en) | 2022-05-10 | 2022-05-10 | Regional groundwater condition prediction method based on groundwater monitoring system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210500900.7A CN114970302B (en) | 2022-05-10 | 2022-05-10 | Regional groundwater condition prediction method based on groundwater monitoring system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114970302A true CN114970302A (en) | 2022-08-30 |
| CN114970302B CN114970302B (en) | 2024-03-15 |
Family
ID=82981874
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210500900.7A Active CN114970302B (en) | 2022-05-10 | 2022-05-10 | Regional groundwater condition prediction method based on groundwater monitoring system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114970302B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115687883A (en) * | 2022-11-18 | 2023-02-03 | 水利部信息中心 | Shallow groundwater storage variable calculation method based on kilometer grids |
| CN117390471A (en) * | 2023-12-12 | 2024-01-12 | 山东省地质矿产勘查开发局第八地质大队(山东省第八地质矿产勘查院) | Hydrogeological data management method and system |
| CN117969789A (en) * | 2024-03-29 | 2024-05-03 | 山东省地矿工程勘察院(山东省地质矿产勘查开发局八〇一水文地质工程地质大队) | Water regime prediction method and monitoring system for underground water resource |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019076078A1 (en) * | 2017-10-16 | 2019-04-25 | 中国环境科学研究院 | Multi-objective optimization method for groundwater pollution monitoring network |
| CN110580328A (en) * | 2019-09-11 | 2019-12-17 | 江苏省地质工程勘察院 | Method for repairing underground water level monitoring value loss |
| CN114254561A (en) * | 2021-12-17 | 2022-03-29 | 奥格科技股份有限公司 | Waterlogging prediction method, waterlogging prediction system and storage medium |
-
2022
- 2022-05-10 CN CN202210500900.7A patent/CN114970302B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019076078A1 (en) * | 2017-10-16 | 2019-04-25 | 中国环境科学研究院 | Multi-objective optimization method for groundwater pollution monitoring network |
| CN110580328A (en) * | 2019-09-11 | 2019-12-17 | 江苏省地质工程勘察院 | Method for repairing underground water level monitoring value loss |
| CN114254561A (en) * | 2021-12-17 | 2022-03-29 | 奥格科技股份有限公司 | Waterlogging prediction method, waterlogging prediction system and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 刘长生;: "基于支持向量机理论的地下水水源地动态预测模型", 勘察科学技术, no. 01, 28 February 2010 (2010-02-28), pages 22 - 25 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115687883A (en) * | 2022-11-18 | 2023-02-03 | 水利部信息中心 | Shallow groundwater storage variable calculation method based on kilometer grids |
| CN115687883B (en) * | 2022-11-18 | 2024-04-30 | 水利部信息中心 | Shallow groundwater storage variable calculation method based on kilometer grid |
| CN117390471A (en) * | 2023-12-12 | 2024-01-12 | 山东省地质矿产勘查开发局第八地质大队(山东省第八地质矿产勘查院) | Hydrogeological data management method and system |
| CN117390471B (en) * | 2023-12-12 | 2024-03-26 | 山东省地质矿产勘查开发局第八地质大队(山东省第八地质矿产勘查院) | Hydrogeological data management method and system |
| CN117969789A (en) * | 2024-03-29 | 2024-05-03 | 山东省地矿工程勘察院(山东省地质矿产勘查开发局八〇一水文地质工程地质大队) | Water regime prediction method and monitoring system for underground water resource |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114970302B (en) | 2024-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114970302B (en) | Regional groundwater condition prediction method based on groundwater monitoring system | |
| Hancock et al. | Spatial interpolation of large climate data sets using bivariate thin plate smoothing splines | |
| Hickey | Slope angle and slope length solutions for GIS | |
| CN112446559B (en) | Large-range ground subsidence space-time prediction method and system based on deep learning | |
| CN114997541B (en) | Urban waterlogging prediction method and early warning platform based on digital twin technology | |
| CN114254561A (en) | Waterlogging prediction method, waterlogging prediction system and storage medium | |
| CN107119657B (en) | A kind of view-based access control model measurement pit retaining monitoring method | |
| CN113610301B (en) | Prediction algorithm for representing ground settlement time sequence evolution | |
| CN113128055B (en) | Distributed hydrological model space calibration method based on runoff yield coefficient | |
| CN114881323A (en) | Foundation pit dewatering area underground water level prediction and updating method based on deep neural network | |
| CN115182398B (en) | Groundwater level and earth surface subsidence prediction method for earthquake early warning area | |
| CN113051638B (en) | Building height optimal configuration method and device | |
| CN112381334B (en) | High-steep slope deformation trend prediction method based on multi-factor fuzzy time sequence | |
| Hu et al. | Dynamically Optimized Unstructured Grid (DOUG) for Analog Ensemble of numerical weather predictions using evolutionary algorithms | |
| CN115641175A (en) | Medium-and-long-term power transaction assistant decision-making determination method and device for new energy power station | |
| CN117236668B (en) | Regional water resource allocation analysis decision method and system based on supply and consumption | |
| CN113849943A (en) | Water supply network node water demand amount checking method coupled with pressure prior information | |
| CN110414084A (en) | A kind of semivariable function model solution method based on tri-level programming | |
| Lobbrecht et al. | Machine learning in real-time control of water systems | |
| Kumar et al. | GIUH based Clark and Nash models for runoff estimation for an ungauged basin and their uncertainty analysis | |
| Willems | Stochastic generation of spatial rainfall for urban drainage areas | |
| Mikołajewski et al. | Development of cluster analysis methodology for identification of model rainfall hyetographs and its application at an urban precipitation field scale | |
| CN115620171A (en) | Forest storage amount evaluation method, device, equipment and storage medium | |
| CN114971097B (en) | Soil moisture content data reconstruction method and prediction method | |
| CN118428277B (en) | Open channel water surface line simulation method and system based on source code SWMM model |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |