CN114357750A - Goaf water filling state evaluation method - Google Patents
Goaf water filling state evaluation method Download PDFInfo
- Publication number
- CN114357750A CN114357750A CN202111607063.XA CN202111607063A CN114357750A CN 114357750 A CN114357750 A CN 114357750A CN 202111607063 A CN202111607063 A CN 202111607063A CN 114357750 A CN114357750 A CN 114357750A
- Authority
- CN
- China
- Prior art keywords
- goaf
- water filling
- water
- data
- filling degree
- 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 144
- 238000011156 evaluation Methods 0.000 title claims abstract description 41
- 238000000034 method Methods 0.000 claims abstract description 39
- 238000005065 mining Methods 0.000 claims abstract description 36
- 238000005516 engineering process Methods 0.000 claims abstract description 22
- 238000012544 monitoring process Methods 0.000 claims abstract description 22
- 238000004088 simulation Methods 0.000 claims abstract description 14
- 238000012360 testing method Methods 0.000 claims abstract description 9
- 238000004364 calculation method Methods 0.000 claims description 34
- 238000012549 training Methods 0.000 claims description 30
- 230000035699 permeability Effects 0.000 claims description 19
- 239000013598 vector Substances 0.000 claims description 18
- 238000009826 distribution Methods 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000013461 design Methods 0.000 claims description 4
- 238000007418 data mining Methods 0.000 claims description 3
- 238000002474 experimental method Methods 0.000 claims description 3
- 238000011835 investigation Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 abstract description 5
- 230000006870 function Effects 0.000 description 8
- 238000004422 calculation algorithm Methods 0.000 description 7
- 238000012706 support-vector machine Methods 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 239000011435 rock Substances 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005553 drilling Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000009412 basement excavation Methods 0.000 description 1
- 238000002939 conjugate gradient method Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000009533 lab test Methods 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 230000001502 supplementing effect Effects 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/08—Fluids
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Geometry (AREA)
- Economics (AREA)
- Quality & Reliability (AREA)
- Computer Graphics (AREA)
- Tourism & Hospitality (AREA)
- Game Theory and Decision Science (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a goaf water filling state evaluation method, belongs to the technical field of mining safety evaluation, and discloses a goaf water filling state evaluation method based on numerical simulation and increment subset SVR technology. The invention establishes a stope numerical model of the goaf containing water filling by defining the water filling degree as an important index of the goaf water filling degree, establishing a priori database consisting of the water filling degree, the water inflow and various hydrological parameters through numerical simulation in the forms of remote sensing interpretation, geological survey and geophysical prospecting, indoor test and field monitoring, realizing dynamic and efficient goaf water filling state evaluation through an increment subset SVR technology, and providing reference for the safety production of mine enterprises. The invention has the advantages that: the method solves the difficult problem of evaluating the water filling degree of the goaf under the complex geological condition, and remarkably reduces the problem of overhigh cost caused by exploring the water filling degree of the goaf by a dynamic object.
Description
Technical Field
The invention belongs to the technical field of mining safety evaluation, and particularly relates to a goaf water filling state evaluation method based on numerical simulation and increment subset SVR (support vector regression) technology
Background
With the development of strip mine mining, the conditions of slope expanding deep excavation mining, river and lake water level change and the like are met. The influence of the change of hydrological conditions on the side wall and the deep goaf can be gradually shown, and the water filling degree of the goaf influences the production progress and the safety of life and property. The change in the goaf water filling level affects the entire seepage field, as the hydraulic path decreases, the increase in hydraulic gradient increases the seepage flow rate, directly related to the water inflow of the stope, with the possible consequence that the pit water inflow is dramatically increased, and further affects the stability of the rock-soil mass including the side slope and the goaf roof. At present, the assessment of the water filling state of the goaf is very difficult, and the cost is too high if the water filling degree of the goaf is frequently detected by means of geophysical prospecting and the like. Therefore, the evaluation of the water filling state of the goaf becomes an important safety issue to be solved urgently for mine enterprises.
The development of modern numerical calculation technology and information science makes it possible to realize the goaf water filling state evaluation under complex conditions. On one hand, numerical calculation simulation can obtain the water inflow under complex hydrological conditions and water filling states at lower cost, and further form a database containing multiple hydrological parameters under different working conditions of a stope of a water-filled goaf; on the other hand, on the basis of establishing a database by using numerical simulation, the time sequence of the monitored hydrological parameters can be identified by using an increment subset SVR technology so as to evaluate the water filling state of the goaf; more importantly, the early warning can be carried out on the abnormal data, and the occurrence of water burst disasters in a stope is avoided.
Disclosure of Invention
In order to solve the problems in the prior art, the invention aims to provide a goaf water filling state evaluation method, which is based on numerical simulation and an increment subset SVR technology, can realize evaluation of the goaf water filling state according to hydrographic parameter data monitored on site, can greatly save the cost of real-time geophysical prospecting for detecting the goaf water filling state, and simultaneously provides reference for safe production of mines.
The purpose of the invention is realized by the following technical scheme:
the invention discloses a goaf water filling state evaluation method, which is characterized by comprising the following steps:
s1, collecting data
Acquiring topographic data, geological data of a stratum and distribution data of a goaf by various geological exploration means, and acquiring the position, scale and form of the goaf and water filling degree data of the goaf by a geophysical prospecting means;
the multiple geological exploration means comprise remote sensing interpretation, geological exploration and field geophysical prospecting; obtaining data mining area topographic contour line data according to remote sensing interpretation, obtaining geological data of strata according to geological exploration, wherein the geological data comprises the quantity, the type, the lithology and the spatial distribution of the strata, and obtaining position, scale, form and goaf water filling degree data of a goaf according to field geophysical exploration; the calculation formula of the water filling degree of the goaf is as follows:
S＝Vw/V (1)
wherein, VwRepresents the volume m of water filled in the goaf3And V represents the volume m of the gob3And the value range of the water filling degree S of the goaf is 0-1, and the water filling degree S of the goaf is used for representing the water filling degree of the goaf.
S2, establishing a three-dimensional numerical model of the mining area
Establishing a three-dimensional numerical model of the mining area by using Patran preprocessing software according to the data acquired by S1;
the method is characterized in that a three-dimensional numerical model of a mining area is established by using Patran pretreatment software, and the specific process is as follows: establishing a ground surface of a mining area according to contour line data obtained by remote sensing interpretation, establishing a geologic body according to geological survey profile data, and establishing a goaf according to on-site geophysical prospecting data so as to obtain a three-dimensional numerical model of the mining area;
s3, obtaining permeability parameters and hydrological parameters through experiments, and verifying the three-dimensional numerical model of the mining area by using the permeability parameters and the hydrological parameters
Obtaining permeability parameters and hydrological parameters through laboratory and field tests; calculating seepage fields of different water filling degrees and under hydrological boundaries of the goaf according to actual hydrological conditions and the established three-dimensional numerical model of the mining area, and then verifying the correctness of numerical calculation of the three-dimensional numerical model of the mining area by using permeability parameters and hydrological parameters;
the correctness of numerical calculation of the three-dimensional numerical model of the mining area is verified by using the permeability parameters and the hydrological parameters, the water filling degree of the goaf participates in calculation in the form of a water head boundary in seepage calculation analysis, the on-site actual measurement water inflow is used as a correctness judgment standard, and the relative error between the calculated water inflow and the on-site actual measurement water inflow is set to be not more than 5 percent as the correctness.
S4, establishing a priori database
On the basis of verifying the correctness of numerical calculation of a three-dimensional numerical model of a mining area, establishing a prior database comprising the water filling degree, the water inflow amount and each hydrological parameter of the goaf by using a numerical simulation method;
the method is characterized in that a priori database of the goaf water filling degree, the water inflow amount and each hydrological parameter is established by using a numerical simulation method, and the specific process is as follows: on the basis of the correctness verification of a three-dimensional numerical model of a mining area, the variation range of each hydrological parameter influencing the seepage of a stope of a water-filled goaf is obtained through field investigation, each factor is divided into 5-7 levels, a multi-factor and multi-level test scheme is established by applying an orthogonal and uniform design method, and numerical calculation is carried out on different working conditions in a test scheme table to form a prior database with hydrological evaluation vectors including water inflow and each hydrological parameter and goaf water filling degree.
S5, evaluating the goaf water filling degree by utilizing the increment subset SVR technology
Determining the effective scale of the SVR sample by adopting the prior data of the prior database, then adding new monitoring data into the sample by adopting a dynamic increment method, selecting a subset sample of the effective scale from the sample sorted latest in time for training, and finally evaluating the filling degree of the goaf according to hydrologic monitoring data, including hydrologic evaluation vectors formed by hydrologic parameters such as water inflow, surface runoff water level and flow rate, underground seepage water head and flow rate and the like.
The incremental subset SVR technology is used for determining the scale of an effective subset by adopting a priori database, namely, predicting data following the subset by continuously enlarging the scale of the subset, determining the effective subset when a predicted value meets the requirement, then continuously inputting hydrologic evaluation vectors obtained by monitoring into a training set, selecting the latest sample of the scale of the effective subset for training, and realizing the evaluation of the water filling degree in the calculation taking the water filling degree of the goaf as an output value.
When the water filling degree of the goaf is evaluated by using the increment subset SVR technology, two more outstanding problems are faced: firstly, with the increase of monitoring information, the number of samples of the prior database is gradually increased, so that the training efficiency is reduced; secondly, the monitoring information has higher correlation with time, namely the correlation degree of the latest information is higher. To solve the two problems, the invention adopts the effective subset to reduce the training cost of the sample, and simultaneously adopts an increment method to ensure that the database is kept to be dynamically updated. The selected incremental subset SVR technology firstly sorts according to the time attribute of the data, secondly evaluates the latest monitoring data in a mode of continuously increasing the size of the subset, and determines the effective size of the subset by taking the evaluation value as a standard when reaching the expected requirement; and adding the latest monitoring data into the sample, and finally, effectively evaluating the water filling degree of the goaf by utilizing an increment subset SVR technology according to the monitoring hydrological parameters.
Compared with the prior art, the invention has the advantages that:
1. the method combines the modern numerical simulation technology, the increment subset SVR technology and the field monitoring technology to form a new method for realizing the goaf water filling degree evaluation by using the numerical simulation technology to establish a prior database, using the field monitoring data as input information and using the increment subset SVR technology.
2. The method can solve the difficult problem of evaluating the water filling degree of the goaf under the complex geological condition, remarkably reduce the problem of overhigh cost caused by exploring the water filling degree of the goaf by the dynamic object, and simultaneously provide reference for the safety production of mine enterprises.
Drawings
FIG. 1 is a schematic block diagram of an implementation flow of the present invention;
Detailed Description
The invention will be further described with reference to the following detailed description and drawings:
as shown in fig. 1, the goaf water-filling state evaluation method of the present invention is characterized by comprising the following steps:
s1, collecting data
Acquiring topographic data, geological data of a stratum and distribution data of a goaf by various geological exploration means, and acquiring the position, scale and form of the goaf and water filling degree data of the goaf by a geophysical prospecting means;
the multiple geological exploration means comprise remote sensing interpretation, geological exploration and field geophysical prospecting; obtaining data mining area topographic contour line data according to remote sensing interpretation, obtaining geological data of strata according to geological exploration, wherein the geological data comprises the quantity, the type, the lithology and the spatial distribution of the strata, and obtaining position, scale, form and goaf water filling degree data of a goaf according to field geophysical exploration; the calculation formula of the water filling degree of the goaf is as follows:
S＝Vw/V (1)
wherein, VwRepresents the volume m of water filled in the goaf3And V represents the volume m of the gob3And the value range of the water filling degree S of the goaf is 0-1, and the water filling degree S of the goaf is used for representing the water filling degree of the goaf.
S2, establishing a three-dimensional numerical model of the mining area
Establishing a three-dimensional numerical model of the mining area by using Patran preprocessing software according to the data acquired by S1;
the method is characterized in that a three-dimensional numerical model of a mining area is established by using Patran pretreatment software, and the specific process is as follows: firstly, digitizing the obtained topographic data by methods such as telemetering interpretation and the like to obtain topographic contour lines of a mining area; secondly, acquiring geological data of the stratum by means of drilling and geophysical prospecting, wherein the geological data comprises the quantity, the type, the lithology and the spatial distribution of the stratum, and further forming a plurality of representative geological section maps, wherein the positions, the scales, the shapes, the water filling degree and the like of the goaf are mainly determined by means of geophysical prospecting and the like; and integrating the terrain contour lines, geological profile data and goaf data by using Pantran modeling software, forming a surface according to the line, and establishing a complete three-dimensional numerical model of the mining area, which comprises the terrain, the stratum and the goaf distribution, from the surface forming order, so as to lay a foundation for further numerical simulation.
S3, obtaining permeability parameters and hydrological parameters through experiments, and verifying the three-dimensional numerical model of the mining area by using the permeability parameters and the hydrological parameters
Firstly, obtaining permeability parameters and hydrological parameters in a laboratory and a field test; and then calculating seepage fields of different water filling degrees and under hydrological boundaries of the goaf according to the actual hydrological conditions and the established three-dimensional numerical model of the mining area, and finally verifying the correctness of numerical calculation of the three-dimensional numerical model of the mining area by using the permeability parameters and the hydrological parameters. In the seepage calculation analysis, the water filling degree of the goaf participates in calculation in the form of a water head boundary, the actually measured water inflow on site is taken as a correctness judgment standard, and the relative error between the calculated water inflow and the actually measured water inflow on site is set to be not more than 5 percent as a correctness.
The specific process is as follows:
firstly, obtaining the permeability parameter (unit: m/s) of the rock mass in a laboratory test, correcting the permeability parameter (unit: m/s) by combining the statistical data of site drilling and joint to determine the permeability parameter (unit: m/s) of each geological rock mass, and assigning the determined permeability parameter (unit: m/s) to the geological rock mass.
And then, applying hydrological boundaries such as water filling degree, water distribution boundary, water separation boundary, surface runoff, underground seepage and the like of the goaf obtained by site survey of the stope to the three-dimensional numerical model, wherein the water level in the goaf under different water filling degree conditions needs to be determined by combining the form of the explored goaf, and the hydrological conditions in the goaf are calculated by seepage.
And finally, calculating the seepage fields of different water filling degrees and different hydrological boundaries of the goaf according to the actual hydrological conditions, the established mining area numerical model and the permeability parameters of each rock stratum. And obtaining the water inflow amount in the pit under the conditions of different goaf water filling degrees through numerical calculation, and comparing the water inflow amount with the measured data to verify the correctness of calculation.
S4, establishing a priori database
On the basis of verifying the correctness of numerical calculation of a three-dimensional numerical model of a mining area, establishing a prior database comprising the water filling degree, the water inflow amount and each hydrological parameter of the goaf by using a numerical simulation method;
adding the water filling degree of the goaf into a multi-water reference list as a main influence factor, and then performing a series of numerical calculations by using a three-dimensional numerical model. And establishing a prior database of hydrological parameters including the water filling degree, the water inflow amount, the water diversion boundary, the water-resisting boundary, the surface runoff, the underground seepage and the like, wherein the prior database is an information basis for further evaluating the water filling degree of the goaf.
The specific process is as follows:
each hydrological parameter affecting seepage was investigated to determine its reasonable range of variation. Dividing the variation range of each hydrological parameter into 5-7 levels, and carrying out numerical calculation on the working conditions of different complex hydrological parameter combinations to form a prior database containing the water filling degree and the water inflow amount. In order to avoid large amount of calculation of a multi-factor and multi-level working condition full scheme, an orthogonal and uniform design method is applied to establish a multi-factor and multi-level test design scheme, and the aim is to obtain more comprehensive information with lower calculation cost. And calculating under the condition of multi-water parameter change to form a prior database containing the water filling degree and the water inflow amount. It is important to emphasize in this section that there is a monotonic relationship between the water filling level of a particular gob and the water level in the gob, which should be considered as an important boundary condition.
S5, evaluating the goaf water filling degree by utilizing the increment subset SVR technology
Determining the effective scale of the SVR sample by adopting the prior data of the prior database, then adding new monitoring data into the sample by adopting a dynamic increment method, selecting a subset sample of the effective scale from the sample sorted latest in time for training, and finally evaluating the filling degree of the goaf according to hydrologic monitoring data, including hydrologic evaluation vectors formed by hydrologic parameters such as water inflow, surface runoff water level and flow rate, underground seepage water head and flow rate and the like.
The incremental subset SVR technology is used for determining the scale of an effective subset by adopting a priori database, namely, predicting data following the subset in a mode of continuously enlarging the scale of the subset, determining the effective subset when a predicted value meets the requirement, then continuously inputting hydrologic evaluation vectors obtained by monitoring into a training set, selecting the latest sample of the scale of the effective subset for training, and realizing the evaluation of the water filling degree in the calculation taking the water filling degree of the goaf as an output value.
The SVR technique refers to a regression-type Support Vector Machine (SVM) method, which is derived from a Support Vector Machine (SVM) method, but instead of seeking an optimal classification plane to separate two types of samples, an optimal classification plane is sought to minimize the error of all training samples from the optimal classification plane.
In the established prior database, the SVR technology is applied to predict the data following the subset in a mode of continuously expanding the subset, and the scale of the effective subset is determined by taking the predicted value reaching the precision requirement as a standard. And continuously supplementing and updating hydrological evaluation vectors obtained by on-site monitoring into a training set, selecting the latest sample of the effective subset scale to perform dynamic training of the incremental subset SVR, and obtaining the evaluation of the water filling degree in the calculation with the water filling degree as an output item. And (3) carrying out emergency investigation under the condition that the data is abnormal (for example, the water filling degree value is far beyond the reasonable range of 0-1), finding out the reason of the data abnormality to avoid hidden dangers, and further providing reference and reference for safety production planning.
The specific process is as follows:
training set sample pair containing l training samples { (x)i,yi) 1,2, … l, which is the sum ofIn xi(xi∈Rd) Is the input column vector for the ith training sample,
is the corresponding output value.
Setting the linear regression function established in the high-dimensional feature space as follows:
f(x)＝wΦ(x)+b (2)
where Φ (x) is a non-linear mapping function.
Introducing a linear insensitive loss function epsilon, and a relaxation variable xii，ξiThen, find the mathematical problem of w, b to translate into:
wherein C is a penalty factor, the larger C is the penalty larger for the sample with the training error larger than epsilon, epsilon specifies the error requirement of the regression function, and the smaller epsilon is the error of the regression function.
And (3) introducing a Largrange function when solving the step (3), and converting into a dual form:
wherein, K (x)i,xj)＝Φ(xi)Φ(xj) Is a kernel function, αiAnd alphaiAs the optimal solution, there are:
wherein N isnsvFor the number of support vectors, the regression function is:
in which only part of the parameter alphai-αi *Not zero, for sample xiI.e. the support vector in question.
The solving problem of the support vector machine is finally converted into a Quadratic Programming (QP) problem with constraints, and when the training samples are few, the traditional Newton method, the conjugate gradient method and the interior point method can be adopted for solving. However, when the number of training samples is large, the complexity of the conventional algorithm increases sharply, and a large amount of memory resources are occupied. In order to reduce the complexity of the algorithm and improve the efficiency of the algorithm, some strategies need to be adopted. For example, the Chunking algorithm decomposes the original optimization problem into a series of QP subsets with smaller sizes, trains each subset sample in turn to obtain the support vector of each subset sample, removes the non-support vector, and completes the training by solving each subset in turn. The Osuna algorithm divides training samples into a working sample set B and a non-working sample set N, the scale of the working sample set B is kept unchanged during solving, and after solving the QP problem of the B set, the samples in the B set are replaced by the samples in the N set, so that the termination condition is achieved. The SMO algorithm can be regarded as a special case of the Osuna algorithm, that is, the scale of the working sample set B is fixed to 2, only the QP problem of 2 training samples is solved each time, and the optimal solution can be obtained by an analytical method without iteration. However, the above methods are all completed off-line, and for the information acquired on-line in real time, an incremental learning method is needed, in which training samples are added one by one, and the training is modified and adjusted by using partial results related to the newly added training samples, which is suitable for obtaining dynamic models.
When the SVR technology is applied to evaluate the water filling degree of the water-filled goaf, three main problems are faced: (a) hydrogeological conditions such as water inflow monitored on site are dynamically changed, so an incremental SVR method should be selected for evaluation. (b) The hydrological parameter sequence obtained by monitoring has correlation with time, namely the latest monitoring information has higher reference value for water filling degree evaluation, so that the training with the latest information can obtain better expected results. (c) With the increase of the monitoring sequence, the calculation amount is greatly increased, and if the samples with poor correlation are removed, the calculation amount can be reduced by adopting the effective subsets in the sequence for training.
In view of the above problems faced in goaf water filling degree evaluation, when the SVR technology is applied for evaluation, firstly, the scale of the effective subset is determined by adopting the prior database, namely, the data following the subset is predicted by continuously enlarging the size of the subset, and when the predicted value meets the requirement, the scale of the effective subset is determined. And then continuously inputting the hydrological evaluation vector obtained by monitoring into a training set, selecting the latest sample of the effective subset scale for training, and realizing the evaluation of the water filling degree in the calculation taking the water filling degree as an output value.
Claims (6)
1. A goaf water filling state evaluation method is characterized by comprising the following steps:
s1, collecting data
Acquiring topographic data, geological data of a stratum and distribution data of a goaf by various geological exploration means, and acquiring the position, scale and form of the goaf and water filling degree data of the goaf by a geophysical prospecting means;
s2, establishing a three-dimensional numerical model of the mining area
Establishing a three-dimensional numerical model of the mining area by using Patran preprocessing software according to the data acquired by S1;
s3, obtaining permeability parameters and hydrological parameters through experiments, and verifying the three-dimensional numerical model of the mining area by using the permeability parameters and the hydrological parameters
Obtaining permeability parameters and hydrological parameters through laboratory and field tests; calculating seepage fields of different water filling degrees and under hydrological boundaries of the goaf according to actual hydrological conditions and the established three-dimensional numerical model of the mining area, and then verifying the correctness of numerical calculation of the three-dimensional numerical model of the mining area by using permeability parameters and hydrological parameters;
s4, establishing a priori database
On the basis of the verification of the numerical calculation correctness of the three-dimensional numerical model of the mining area, establishing a hydrological evaluation vector formed by hydrological parameters such as water inflow, surface runoff water level and flow velocity, an underground seepage head and flow velocity and the like, and forming a prior database comprising the hydrological evaluation vector and the goaf water filling degree by using a numerical simulation method;
s5, evaluating the goaf water filling degree by utilizing the increment subset SVR technology
Determining the effective scale of the SVR sample by adopting the prior data of the prior database, then adding new monitoring data into the sample by adopting a dynamic increment method, selecting a subset sample of the effective scale from the sample sorted latest in time for training, and finally evaluating the filling degree of the goaf according to hydrologic monitoring data, including hydrologic evaluation vectors formed by hydrologic parameters such as water inflow, surface runoff water level and flow rate, underground seepage water head and flow rate and the like.
2. The method of claim 1, wherein in S1, the plurality of geological exploration means includes telemetry, geological exploration and field geophysical exploration; obtaining data mining area topographic contour line data according to remote sensing interpretation, obtaining geological data of strata according to geological exploration, wherein the geological data comprises the quantity, the type, the lithology and the spatial distribution of the strata, and obtaining position, scale, form and goaf water filling degree data of a goaf according to field geophysical exploration; the calculation formula of the water filling degree of the goaf is as follows:
S＝Vw/V (1)
wherein, VwRepresents the volume m of water filled in the goaf3And V represents the volume m of the gob3And the value range of the water filling degree S of the goaf is 0-1, and the water filling degree S of the goaf is used for representing the water filling degree of the goaf.
3. The method for evaluating the water filling state of a goaf according to claim 1, wherein in S2, the three-dimensional numerical model of the mining area is established by using a patrran preprocessing software, and the specific process is as follows: and establishing a ground surface of the mining area according to contour line data obtained by remote sensing interpretation, establishing a geologic body according to geological survey profile data, and establishing a goaf according to on-site geophysical prospecting data so as to obtain a three-dimensional numerical model of the mining area.
4. The goaf water filling state evaluation method according to claim 1, wherein in S3, the permeability parameters and hydrological parameters are used to verify the correctness of numerical calculation of the three-dimensional numerical model of the mining area, in the seepage calculation analysis, the goaf water filling degree participates in the calculation in the form of a waterhead boundary, the actually measured water inflow on site is used as a correctness judgment standard, and the relative error between the calculated water inflow and the actually measured water inflow on site is set to be less than or equal to 5% as a correctness.
5. The goaf water filling state evaluation method according to claim 1, wherein in S4, the numerical simulation method is used to establish a priori database of goaf water filling degree, water inflow and hydrological parameters, and the specific process is as follows: on the basis of the correctness verification of a three-dimensional numerical model of a mining area, the variation range of each hydrological parameter influencing the seepage of a stope of a water-filled goaf is obtained through field investigation, each factor is divided into 5-7 levels, a multi-factor and multi-level test scheme is established by applying an orthogonal and uniform design method, and numerical calculation is carried out on different working conditions in a test scheme table to form a prior database with water inflow, hydrological parameter evaluation vectors of each hydrological parameter and goaf water filling degree.
6. The goaf water-filling state evaluation method according to claim 1, wherein in S5, the incremental subset SVR technique is used to determine the size of the effective subset by using the priori database, that is, predict the data following the subset by continuously enlarging the size of the subset, determine the effective subset when the predicted value meets the requirement, continuously input the hydrologic evaluation vector obtained by monitoring into the training set, select the latest sample of the size of the effective subset for training, and realize the evaluation of the water-filling degree in the calculation using the goaf water-filling degree as the output value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111607063.XA CN114357750B (en) | 2021-12-24 | 2021-12-24 | Goaf water filling state assessment method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111607063.XA CN114357750B (en) | 2021-12-24 | 2021-12-24 | Goaf water filling state assessment method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114357750A true CN114357750A (en) | 2022-04-15 |
| CN114357750B CN114357750B (en) | 2024-09-17 |
Family
ID=81100457
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111607063.XA Active CN114357750B (en) | 2021-12-24 | 2021-12-24 | Goaf water filling state assessment method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114357750B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114861271A (en) * | 2022-04-29 | 2022-08-05 | 中铁科学研究院有限公司 | Method for analyzing disease causes of goaf in railway tunnel construction |
| CN115221687A (en) * | 2022-06-22 | 2022-10-21 | 中国水利水电科学研究院 | Numerical simulation method for influence of coal mining on river runoff |
| CN116520451A (en) * | 2023-04-24 | 2023-08-01 | 四川阳光上元能源技术有限公司 | Underground goaf detection method based on quantum detection technology |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190354873A1 (en) * | 2018-02-16 | 2019-11-21 | Lucas Pescarmona | Analysis system and hydrology management for basin rivers |
| CN110851991A (en) * | 2019-11-18 | 2020-02-28 | 核工业二〇八大队 | Underground water flow numerical simulation method |
| CN111415038A (en) * | 2020-03-19 | 2020-07-14 | 中煤科工集团西安研究院有限公司 | Multi-working-face goaf water inflow refining-with-semen prediction method |
| CN113221228A (en) * | 2021-06-04 | 2021-08-06 | 中国电建集团成都勘测设计研究院有限公司 | Hydropower station underground cave group surrounding rock mechanical parameter inversion method |
-
2021
- 2021-12-24 CN CN202111607063.XA patent/CN114357750B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190354873A1 (en) * | 2018-02-16 | 2019-11-21 | Lucas Pescarmona | Analysis system and hydrology management for basin rivers |
| CN110851991A (en) * | 2019-11-18 | 2020-02-28 | 核工业二〇八大队 | Underground water flow numerical simulation method |
| CN111415038A (en) * | 2020-03-19 | 2020-07-14 | 中煤科工集团西安研究院有限公司 | Multi-working-face goaf water inflow refining-with-semen prediction method |
| CN113221228A (en) * | 2021-06-04 | 2021-08-06 | 中国电建集团成都勘测设计研究院有限公司 | Hydropower station underground cave group surrounding rock mechanical parameter inversion method |
Non-Patent Citations (2)
| Title |
|---|
| XUAN LIAO ET AL.: "Predictive Analytics and Statistical Learning for Waterflooding Operations in Reservoir Simulations", 《2019 18TH IEEE INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND APPLICATIONS (ICMLA)》, 17 February 2020 (2020-02-17) * |
| 王永增 等: "某露天铁矿隐伏采空区充水量监测模拟研究", 《现代矿业》, no. 10, 31 October 2021 (2021-10-31) * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114861271A (en) * | 2022-04-29 | 2022-08-05 | 中铁科学研究院有限公司 | Method for analyzing disease causes of goaf in railway tunnel construction |
| CN115221687A (en) * | 2022-06-22 | 2022-10-21 | 中国水利水电科学研究院 | Numerical simulation method for influence of coal mining on river runoff |
| CN116520451A (en) * | 2023-04-24 | 2023-08-01 | 四川阳光上元能源技术有限公司 | Underground goaf detection method based on quantum detection technology |
| CN116520451B (en) * | 2023-04-24 | 2024-03-12 | 四川阳光上元能源技术有限公司 | Underground goaf detection method based on quantum detection technology |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114357750B (en) | 2024-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114357750B (en) | Goaf water filling state assessment method | |
| Cooley et al. | Nonlinear‐regression groundwater flow modeling of a deep regional aquifer system | |
| RU2002122397A (en) | Comprehensive reservoir optimization | |
| CN109799540B (en) | Volcanic rock type uranium deposit magnetic susceptibility inversion method based on geological information constraint | |
| CN113570226A (en) | Method for evaluating occurrence probability grade of tunnel water inrush disaster in fault fracture zone | |
| Zhao et al. | Unfavorable geology recognition in front of shallow tunnel face using machine learning | |
| CN114066084B (en) | Method and system for predicting phase permeation curve based on machine learning | |
| CN104899928A (en) | Three-directional geology modeling method based on sparse borehole points | |
| CN105467438A (en) | Three-modulus-based shale ground stress three-dimensional seismic characterization method | |
| Wang et al. | Application of fuzzy analytic hierarchy process in sandstone aquifer water yield property evaluation | |
| CN117711140A (en) | Tunnel water bursting disaster timing early warning method and system based on multi-source data fusion | |
| CN112948924A (en) | Near unconsolidated formation mining water flowing fractured zone height determination method based on overlying strata structure | |
| Hu et al. | Calculation of average reservoir pore pressure based on surface displacement using image-to-image convolutional neural network model | |
| CN110705168A (en) | Simulation method of structural stress field | |
| CN110988997A (en) | Hydrocarbon source rock three-dimensional space distribution quantitative prediction technology based on machine learning | |
| CN118273768B (en) | Coal mine water disaster holographic natural source mode early warning method and system based on GIS base | |
| CN115877447A (en) | Reservoir prediction method for seismic restraint three-dimensional geological modeling under straight-flat combined well pattern condition | |
| CN117251802B (en) | Heterogeneous reservoir parameter prediction method and system based on transfer learning | |
| Sarkheil et al. | The fracture network modeling in naturally fractured reservoirs using artificial neural network based on image loges and core measurements | |
| Li et al. | Study on potential groundwater yield zone in sandstone aquifer based on a dual dynamic variable weight model: A case study in Shuangma Coal Mine of Ordos Basin | |
| Yang et al. | Optimizing and accelerating history matching progress of numerical reservoir simulation by using material balance analysis | |
| CN114066271A (en) | Tunnel water inrush disaster monitoring and management system | |
| CN116011268A (en) | Quantitative description method of dominant seepage channel | |
| CN111274736A (en) | Water flowing fractured zone prediction method based on supervised learning neural network algorithm | |
| CN118430692B (en) | Three-dimensional prospecting prediction method, system, equipment and medium |
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 |