CN109993459B - Method for identifying water bursting source of complex multi-aquifer mine - Google Patents
Method for identifying water bursting source of complex multi-aquifer mine Download PDFInfo
- Publication number
- CN109993459B CN109993459B CN201910298332.5A CN201910298332A CN109993459B CN 109993459 B CN109993459 B CN 109993459B CN 201910298332 A CN201910298332 A CN 201910298332A CN 109993459 B CN109993459 B CN 109993459B
- Authority
- CN
- China
- Prior art keywords
- water
- identification
- sample
- aquifer
- index
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2132—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on discrimination criteria, e.g. discriminant analysis
- G06F18/21322—Rendering the within-class scatter matrix non-singular
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2431—Multiple classes
-
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Mining
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2132—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on discrimination criteria, e.g. discriminant analysis
- G06F18/21322—Rendering the within-class scatter matrix non-singular
- G06F18/21324—Rendering the within-class scatter matrix non-singular involving projections, e.g. Fisherface techniques
-
- 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
Abstract
The invention discloses a method for identifying a water inrush source of a complex multi-aquifer mine, which comprises the following steps: s1: establishing a water source database of a mine according to water chemistry data of a known aquifer water sample; s2: data after data inspection and abnormal value processing are used as modeling samples; s3: determining an identification index and a threshold value thereof, and judging the effectiveness of the identification index in the primary identification model by adopting back substitution test; s4: establishing a comprehensive-step recognition method model according to the effective recognition index and a Fisher recognition method; s5: and determining the identification indexes of the water sample to be judged, and sequentially judging through a comprehensive-step identification method model to identify the type of the water source. The method has the advantages that a characteristic ion contrast method, an ion proportion coefficient method, a Fisher identification method and the like are comprehensively adopted, different methods are adopted for identifying different aquifer water sources, the method is simple and then complex, and the mine water source types with complex hydrogeological conditions and more water-filled aquifers are gradually judged.
Description
Technical Field
The invention relates to the field of water source identification, in particular to a method for identifying a water inrush source of a complex multi-aquifer mine.
Background
China is a country which takes coal as a main energy source, and the coal deposit hydrogeological conditions of China are complex and various, and the mine water disaster situation is severe, so that the China is an important factor influencing the mine safety production. Once water damage occurs in the mine, huge economic loss can be caused, and casualties can be caused. The mine water source identification is an important basic work for mine water disaster prevention and control, and can provide a basis for water prevention and control measures, post-disaster rescue and the like. And the selection of a proper water source identification method is the key of mine water source identification.
At present, the main methods for identifying mine water sources include an underground water chemistry method, a water level dynamic observation method, an isotope method, a water temperature analysis method and the like. Among them, the water chemistry method is widely used due to the abundant basic data and strong versatility. A plurality of methods for identifying the water source of the mine water inrush by using a water chemistry method are available, and the water source identification is realized mainly by combining a relevant mathematical model on the basis of water chemistry data. Such as using Fisher recognition models, Bayes recognition models, distance recognition models, neural network models, extension recognition models, cluster analysis, SVM models, and combination models that perform principal component analysis before recognition.
The single method can achieve better recognition effect under the conditions of simpler mine hydrogeological conditions and less water-filled water-bearing stratum types. However, when hydrogeological conditions are complex and water-filled aquifers are numerous, the identification effectiveness is greatly reduced, and the requirement of mine safety production on a water source identification model cannot be met.
Accordingly, there is an urgent need for a method for identifying a complicated multi-aquifer mine water inrush source, which can realize the quick and effective identification of the type of the complicated multi-aquifer water inrush source.
Disclosure of Invention
The invention aims to solve the technical problem of providing a complex multi-aquifer mine water inrush source identification method capable of realizing quick and effective identification of complex multi-aquifer water inrush source types, which is suitable for identifying mine water inrush sources with complex hydrogeological conditions and more water-filled aquifers.
The invention adopts the following technical scheme to solve the technical problems: a method for identifying a water inrush source of a complex multi-aquifer mine comprises the following steps:
s1: establishing a water source database of a mine according to the water chemistry data of the known aquifer water sample;
s2: taking the data subjected to data detection and abnormal value processing as a modeling sample;
s3: determining an identification index and a threshold value thereof, and judging the effectiveness of the identification index in the primary identification model by adopting back-up test;
s4: establishing a comprehensive-step recognition method model according to the effective recognition index and a Fisher recognition method;
s5: and determining the identification indexes of the water sample to be judged, and sequentially judging through a comprehensive-step identification method model to identify the type of the water source.
As one of the preferable modes of the invention, the water chemistry data of the water sample is the mass concentration of various ions and the milliequivalent and ratio of each ion of the water in each aquifer of the coal mine to which the water sample belongs; the conventional ion may be Na + +K + 、 Ca 2+ 、Mg 2+ 、Cl - 、HCO3 - 、SO 4 2- Etc.; the water chemistry data is used as a database building index for building a water source database.
As one of the preferable modes of the invention, the data inspection method is anion-cation balance inspection, and the error is controlled within +/-5%; the abnormal sample screening and processing method adopts one or more combinations of index histogram, box diagram, Q-Q diagram and cluster analysis diagram to screen the abnormal sample and remove the water sample data as the modeling sample
As one of the preferable modes of the invention, the selection method of the identification index is to select the index capable of distinguishing each aquifer as the identification index according to the water chemistry data of the water samples of each aquifer, determine the threshold value of the index and identify the identification relation when identifying the water source; the standard of the identification index selection is that the content change of the identification index in the aquifer water sample is larger than the content change of other indexes; the identification relation of a certain identification means that water samples of which the identification indexes are smaller than or larger than a determined identification index threshold value are classified into a class during identification, and the identification indexes can at least distinguish the water samples of two aquifers; wherein the identification index is the mass concentration of the conventional ions and the milliequivalent ratio of the conventional ions.
As one of the preferred modes of the present invention, the mass concentration of the conventional ions and the milliequivalent ratio of the conventional ions are called the characteristic ion contrast and the ion proportionality coefficient, respectively.
As one preferable embodiment of the present invention, the specific method for judging the validity of the identification index in the preliminary identification model by using the back substitution test in step S3 includes: taking a water sample containing a known water layer type as a new sample, sequentially substituting the new sample into the established preliminary identification model, and if the identification result is consistent with the actual result, indicating that the selection of the identification index and the established preliminary identification model are effective; and if the identification results of the same aquifer water sample are mostly inconsistent with the actual identification results, reselecting a new identification index and establishing a new initial identification model.
As one of the preferable modes of the invention, the comprehensive-gradual identification method in step S4 is to comprehensively adopt a characteristic ion comparison method, an ion proportion coefficient method and a Fisher identification method, and different methods are adopted for identifying different aquifer water sources, wherein the water source types are determined step by step after being simple and complex; for a single aquifer water sample to be judged, classifying characteristic ions or water samples of which the ion proportionality coefficients are smaller than a determined identification index threshold value into a certain water source type during identification, or classifying the water samples of which the indexes are larger than the threshold value into the certain water source type, stopping identification if a result can be judged, or classifying the water samples into a new sample to be judged, selecting a new identification index and a threshold value thereof, and repeating the steps; and when the type of the water inrush source cannot be judged by using the identification index, a Fisher identification method is adopted. The method can realize the rapid identification of the type of the complex multi-aquifer mine effluent source.
As one of the preferable modes of the invention, the comprehensive-step-by-step identification method comprises the following specific steps:
step one, for a certain single aquifer water sample to be judged, judging whether the water sample is AoHu water or Duhui water by adopting a characteristic ion comparison method, if the result can be judged, stopping identification, otherwise, performing identification in the step two;
secondly, judging whether the water sample is cladding body water or coal series water by adopting an ion proportionality coefficient method;
thirdly, on the basis of the second step, if the water sample is judged to be the cladding body water, judging that the water sample is cladding body gneiss water and cladding body cold grey water by adopting a characteristic ion comparison method; and if the water sample is judged to be coal-based water, judging that the water sample is sandstone water or Taigrey water by adopting a Fisher recognition method.
And judging the single aquifer water sample to be judged into one of secondary water, overlying body gneiss water, overlying body cold grey water, sandstone water, taigrey water and Ordovician grey water through the three steps.
As one of the preferable modes of the invention, the specific operation method of the comprehensive-step-by-step identification method is as follows: in the water sample after data verification and abnormal rejection: first, when TDS>4300mg/L、Cl - >2000mg/L、 Na + +K + >When 1350mg/L, the water sample is the Ordovician limestone water, and the identification is finished; otherwise, further identifying when TDS<440mg/L、Cl - <50mg/L、Na + +K + <When the water sample is 80mg/L, the water sample is secondary water, and the identification is finished; otherwise, further identify when Gamma Cl - /γCa 2+ <5.2, the water sample is the cover-pushing body water, and is further identified when TDS<1360mg/L、Cl - <474mg/L, the water sample is the water of the coating body water sheet gneiss waterRespectively finishing; otherwise, the water sample is the push-coating body water cold grey water, and the identification is finished; when gamma Cl is present - /γCa 2+ >And 5.2, identifying the sandstone water or the taigrey water by using a Fisher identification method, wherein the water sample is coal-based water, and the identification is finished at the moment.
Compared with the prior art, the invention has the advantages that: the method has the advantages that a characteristic ion contrast method, an ion proportion coefficient method, a Fisher identification method and the like are comprehensively adopted, different methods are adopted for identifying different aquifer water sources, and the mine water source type with complicated hydrogeological conditions and more water-filled aquifers can be gradually judged after the method is simple and complicated. The method can substitute the identification indexes adopted by the identification process in the unknown water sample into the established comprehensive-gradual identification method model to identify the water source directly. Therefore, the rapidity and the accuracy of identifying the complex multi-aquifer mine water inrush source can be improved.
Drawings
FIG. 1 is a flow chart of the identification of a complicated multiple aquifer water burst source in example 1;
FIG. 2 is a histogram of abnormal value processing in the present embodiment;
FIG. 3 is a diagram of the clustering result of abnormal value processing in this embodiment;
FIG. 4 is an abnormal value processing indicator box diagram of the present embodiment;
fig. 5 is a detailed flowchart of the "comprehensive-step identification method" model of the present embodiment.
Detailed Description
The following examples are given for the detailed implementation and specific operation of the present invention, but the scope of the present invention is not limited to the following examples.
Example 1
As shown in fig. 1, the method for identifying a water inrush source of a complex multi-aquifer mine in the embodiment includes the following steps:
s1: establishing a water source database of a mine according to water chemistry data of a known aquifer water sample; the water chemical data of the water sample are all the coal mines to which the water sample belongsMass concentrations of various ions of the water of the aquifer and milliequivalents and ratios of the ions; the conventional ion may be Na + +K + 、Ca 2+ 、Mg 2+ 、Cl - 、HCO 3 - 、SO 4 2- Etc.; the water chemistry data is used as a database building index for building a water source database.
S2: data after data inspection and abnormal value processing are used as modeling samples; the data inspection method is anion-cation balance inspection, and the error is controlled within +/-5%. In the anion-cation balance test of the data test: the abnormal sample screening and processing method adopts one or more combinations of index histograms, box charts, Q-Q charts and cluster analysis charts to screen the abnormal samples and remove the water sample data as a modeling sample.
S3: determining an identification index and a threshold value thereof, and judging the effectiveness of the identification index in the primary identification model by adopting back substitution test; the selection method of the identification indexes comprises the steps of selecting indexes capable of distinguishing each aquifer from water chemical data of water samples of each aquifer according to modeling as identification indexes, determining threshold values of the indexes and identifying the identification relation of water sources; the standard of the identification index selection is that the content change of the identification index in the aquifer water sample is larger than that of other indexes; the identification relation of a certain identification means that water samples of which the identification indexes are smaller than or larger than a determined identification index threshold value are classified into a class during identification, and the identification indexes can at least distinguish the water samples of two aquifers; the identification indexes are the mass concentration of the conventional ions and the milliequivalent ratio of the conventional ions, and the mass concentration of the conventional ions and the milliequivalent ratio of the conventional ions are respectively called characteristic ion contrast and an ion proportionality coefficient. The specific method for judging the validity of the identification index in the preliminary identification model by adopting the back-substitution test in the step S3 comprises the following steps: taking a water sample containing a known water layer type as a new sample, sequentially substituting the new sample into the established preliminary identification model, and if the identification result is consistent with the actual result, indicating that the selection of the identification index and the established preliminary identification model are effective; and if the identification results of the same aquifer water sample are mostly inconsistent with the actual identification results, reselecting a new identification index and establishing a new initial identification model.
S4: establishing a comprehensive-gradual identification method model according to the effective identification index and a Fisher identification method; the comprehensive-step identification method in the step S4 is to comprehensively adopt a characteristic ion contrast method, an ion proportion coefficient method and a Fisher identification method, adopt different methods for identifying different aquifer water sources, and judge the water source types step by step after simple and complex methods; for a single aquifer water sample to be judged, classifying characteristic ions or water samples of which the ion proportionality coefficients are smaller than a determined identification index threshold value into a certain water source type during identification, or classifying the water samples of which the indexes are larger than the threshold value into the certain water source type, stopping identification if a result can be judged, or classifying the water samples into a new sample to be judged, selecting a new identification index and a threshold value thereof, and repeating the steps; and when the type of the water inrush source cannot be judged by using the identification index, a Fisher identification method is adopted. The Fisher recognition method has no specific requirement on the distribution of the population, and is a linear recognition method. It features that high-dimensional data points are projected to low-dimensional space (one-dimensional straight line) so that data points become denser to overcome "dimensional root" caused by high dimension. The principle of projection is to separate the population from the population as much as possible, then determine the recognition analysis function according to the principle of maximum inter-class distance and minimum intra-class distance, and further classify and recognize the new samples. The method can realize the rapid identification of the type of the complex multi-aquifer mine effluent water source.
The comprehensive-step-by-step identification method comprises the following specific steps:
step one, for a certain single aquifer water sample to be judged, judging whether the water sample is AoHu water or Duhui water by adopting a characteristic ion comparison method, if the result can be judged, stopping identification, otherwise, performing identification in the step two;
secondly, judging whether the water sample is cladding body water or coal series water by adopting an ion proportionality coefficient method;
thirdly, on the basis of the second step, if the water sample is judged to be the cladding body water, judging that the water sample is cladding body gneiss water and cladding body cold grey water by adopting a characteristic ion comparison method; and if the water sample is judged to be coal-based water, judging that the water sample is sandstone water or Taigrey water by adopting a Fisher recognition method.
And judging the single aquifer water sample to be judged into one of secondary water, overlying body gneiss water, overlying body cold grey water, sandstone water, taigrey water and Ordovician grey water through the three steps.
S5: and determining the identification indexes of the water sample to be judged, and sequentially judging through a comprehensive-step identification method model to identify the type of the water source.
The specific operation method of the comprehensive-step-by-step identification method comprises the following steps: in the water sample after data verification and abnormal rejection: first, when TDS>4300mg/L、Cl - >2000mg/L、Na + +K + >When 1350mg/L, the water sample is the Aohui water, and the identification is finished; otherwise, further identifying when TDS<440mg/L、Cl - <50mg/L、 Na + +K + <When the water sample is 80mg/L, the water sample is secondary water, and the identification is finished; otherwise, further identification is carried out when gamma Cl - / γCa 2+ <5.2, the water sample is the cover-pushing body water, and is further identified when TDS<1360mg/L、Cl - <474mg/L, the water sample is the water of the coating body water sheet gneiss, and the identification is finished; otherwise, the water sample is the push-coating body water cold grey water, and the identification is finished; when gamma Cl is present - /γCa 2+ >And 5.2, identifying the sandstone water or the Taigray water by using the Fisher identification method, wherein the water sample is coal-based water, and the identification is finished at the moment.
In the embodiment, a characteristic ion contrast method, an ion proportion coefficient method, a Fisher identification method and the like are comprehensively adopted, different methods are adopted for identifying water sources of different aquifers, and the mine water source types with complicated hydrogeological conditions and more water-filled aquifers can be gradually judged after the methods are simple and complex. In the embodiment, the identification indexes adopted by the identification process in the unknown water sample are directly detected, and the identification indexes can be substituted into the established comprehensive-step identification method model to identify the water source. Therefore, the rapidness and the accuracy of identifying the water bursting source of the mine with complex multi-aquifer can be improved.
Next, aquifer water sample data such as new boundary secondary water, tectorial membrane gneiss water, tectorial membrane cold grey water, sandstone water, taigrey water and the like in the new collection secondary water quality standing book are selected as databases to concretely explain the design scheme and the theoretical basis of the embodiment.
Data inspection and exception sample handling
When a single aquifer characteristic index judgment is carried out, the water chemical component of the water sample to be analyzed is required to be a typical representation of the aquifer. Therefore, before characteristic index analysis and water source identification model establishment, each aquifer water sample needs to be inspected and abnormal sample treatment is carried out.
The sample data inspection method is anion-cation balance inspection, due to Na + +K + 、Ca 2+ 、Mg 2+ 、Cl - 、 HCO 3 - 、SO 4 2- The indexes are measured values, so that the error is required to be controlled within +/-5%, namely, qualified water samples with the testing error within the range are used as basic data for further analysis.
Although the water sample is qualified through anion and cation inspection, a few water samples have certain problems in a certain link in the processes of water sample collection, horizon determination, test analysis and the like, so that the water sample cannot truly reflect the water quality characteristics of a certain aquifer, is represented as an outlier sample in the water quality analysis process, and needs to be screened and rejected. The following will illustrate the screening and processing method of abnormal samples by taking Ordovician gray water sample as an example, as shown in FIGS. 2-4.
To 17 collected Oreochromis water samples TDS and Na + +K + 、Ca 2+ 、Cl - The ions were analyzed in comparison, as shown in FIG. 2. The indexes of the three water samples marked as O-03, O-04 and O-05 are obviously different from the other 14 water samples. In order to further judge the relation between the three water samples of the Ordovician gray water No. O-04, O-05 and O-06 and the rest of the water samples, the Ordovician gray water samples are subjected to system cluster analysis. Adopting TDS and Na + +K + 、Ca 2+ 、Mg 2+ 、Cl - 、HCO 3 - 、 SO 4 2- The 7 indexes are used as variables, and the clustering result is shown in FIG. 3. As can be seen from FIG. 3, the differences among the samples except the samples numbered O-03, O-04 and O-05 are the smallest and most representative of the properties of the Aohu waterThe distance between O-03, O-04, O-05 and the rest of the samples is the largest, and the classification is completed at the end. The boxplot is a statistical graph for describing data distribution, and can be used for expressing descriptive statistics such as median, 4 quantile and extreme value of observed data and observing the distribution condition of variable values from a visual angle. 17 water samples TDS and Na of Aohui water + +K + 、Ca 2+ 、Mg 2+ 、Cl - The box diagram is shown in fig. 4. As can be seen from the figure, the 5 water samples marked with the indexes of O-3, O-4 and O-5 also show significant differences from other water samples.
Therefore, three samples O-03, O-04 and O-05 in 17 samples of the Aohu water can be regarded as abnormal samples which cannot represent the typical characteristics of the Aohu water, and are removed in the establishment of a subsequent water sample type identification model and do not participate in modeling as standard water samples.
And (3) comprehensively screening and eliminating outlier samples from the other 5 single aquifer water samples such as the second aquifer water of the new kingdom, the tectorial membrane schist water, the tectorial membrane hanwu system water, the sandstone water, the taigrey water and the like by adopting the method or other methods. And the data subjected to data inspection and abnormal sample removal is used as a modeling sample.
Identification index and threshold determination
Based on the above analysis, the dihydrate and Ordovician water are in TDS and Cl - And Na + +K + The content has typical characteristics, and can be effectively distinguished from other aquifer water samples. Therefore, the above 3 indexes were selected as the identification indexes of the dihydrate and olanzapine water, and the content ranges thereof are shown in table 1.
TABLE 1 content ranges of characteristic indexes of hydrous and olanzapine
(2) Push-coated body water and coal-series water
The types of the conventional ions and water quality of the cladding body gneiss water and the cladding body cold grey water are similar, and the types of the conventional ions and water quality of the sandstone water and the cladding body taigrey water are similar. The gunite water and the cold grey water of the coating body are firstly classified into one type and are collectively called as coating body water; sandstone water and teragrey water are classified into a group, which is collectively called coal-based water. The data of the modeling samples show that the characteristic ion proportionality coefficients (gamma Cl) of the tectorial water and the coal-series water - /γCa 2+ ) The differences were significant (table 2) and could be used as a differential marker for rebinned inferred waters and coal-based waters.
TABLE 2 range of characteristic ion proportionality coefficients for overburden push water and coal-based water
(3) Gunite water and cold grey water of push-coated body
In TDS and Cl, the water of the jacket body gneiss and the water of the jacket body Hanwu system - The content has typical characteristics and can be used as a mark for distinguishing the two types of aquifer water.
TABLE 3 Water characteristic index content Range of the coating
(4) Sandstone water and taigrey water
The sandstone water and the teragrey water have no obvious difference in characteristic example indexes and ion proportionality coefficients, and the conventional means cannot distinguish the sandstone water and the teragrey water only from the two aspects. For this purpose, TDS and Ca are selected comprehensively 2+ 、Mg 2+ 、Na + +K + 、 HCO 3 - 、SO 4 2- 、Cl - And (5) taking the 7 indexes as identification factors for classification between the sandstone water and the taigrey water, and judging the type of the water source by adopting a Fisher identification method.
The Fisher identification method has no specific requirement on the distribution of the population and is a linear identification method. It features that high-dimensional data points are projected to low-dimensional space (one-dimensional straight line) so that data points become denser to overcome "dimensional root" caused by high dimension. The principle of projection is to separate the population from the population as much as possible, then determine the recognition analysis function according to the principle of maximum inter-class distance and minimum intra-class distance, and further classify and recognize the new samples.
Identification method selection and identification step
The new method for identifying the type of the water source of the second-ore complex multi-aquifer adopts a comprehensive-step identification method.
The method comprises the following specific steps:
firstly, judging whether a certain single aquifer water sample to be judged is the Aohu water or the second aquifer water by adopting a characteristic ion comparison method, if the result can be judged, stopping the identification, otherwise, carrying out the second step of identification.
And secondly, judging whether the water sample is the cladding body water or the coal-series water by adopting an ion proportionality coefficient method.
Thirdly, on the basis of the second step, if the water sample is judged to be the cladding body water, judging that the water sample is cladding body gneiss water and cladding body cold grey water by adopting a characteristic ion comparison method; and if the water sample is judged to be coal-series water, judging that the water sample is sandstone water and Taigrey water by adopting a Fisher identification method.
And judging the single aquifer water sample to be judged into one of secondary water, overlying body gneiss water, overlying body cold grey water, sandstone water, taigrey water and Ordovician grey water through the three steps.
Establishment of 'comprehensive-step-by-step recognition' model
For a single aquifer water sample to be judged, after data inspection and abnormal sample processing, identification indexes and limit values, identification method selection and identification step determination are carried out, an established identification model and a process are shown in fig. 3.
Verification of recognition effects
Judging the effectiveness of the identification index by adopting back-substitution test, taking a water sample with a known water layer type as a new sample, sequentially substituting the new sample into the established identification model, and if the identification result is consistent with the reality, indicating that the selection of the identification index and the established 'comprehensive-gradual identification method' model are effective; and if the identification results of the same aquifer water sample are mostly inconsistent with the actual identification results, a new identification index is reselected, and a new water source identification model is established.
After the comprehensive-gradual identification method model is determined, the unknown aquifer water sample is substituted into the established model after the content of the identification index is tested, and the aquifer type of the unknown water sample can be obtained.
The above description is only exemplary of the present invention and should not be taken as limiting the invention, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (7)
1. A method for identifying a water inrush source of a complex multi-aquifer mine is characterized by comprising the following steps:
s1: establishing a water source database of a mine according to water chemistry data of a known aquifer water sample;
s2: data after data inspection and abnormal value processing are used as modeling samples;
s3: determining an identification index and a threshold value thereof, and judging the effectiveness of the identification index in the primary identification model by adopting back-up test;
s4: establishing a comprehensive-step recognition method model according to the effective recognition index and a Fisher recognition method; determining identification indexes of the water samples to be judged, and sequentially judging through a comprehensive-step identification method model to identify the types of the water sources;
the comprehensive-step-by-step identification method in the step S4 is to comprehensively adopt a characteristic ion comparison method, an ion proportion coefficient method and a Fisher identification method, adopt different methods for identifying different aquifer water sources, and gradually judge the water source types after the water sources are simple and complex; for a single aquifer water sample to be judged, classifying characteristic ions or water samples of which the ion proportionality coefficients are smaller than a determined identification index threshold value into a certain water source type during identification, or classifying the water samples of which the indexes are larger than the threshold value into the certain water source type, stopping identification if a result can be judged, or classifying the water samples into a new sample to be judged, selecting a new identification index and a threshold value thereof, and repeating the steps; when the type of the water inrush source cannot be judged by using the identification index, a Fisher identification method is adopted;
the comprehensive-step-by-step identification method comprises the following specific steps:
firstly, judging whether a single aquifer water sample to be judged is the Aohui water or the diaspore water by adopting a characteristic ion comparison method, if a result can be judged, stopping the identification, and if not, performing the second step of identification;
secondly, judging whether the water sample is cladding body water or coal series water by adopting an ion proportionality coefficient method;
thirdly, on the basis of the second step, if the water sample is judged to be the cladding body water, judging that the water sample is cladding body gneiss water and cladding body cold grey water by adopting a characteristic ion comparison method; and if the water sample is judged to be coal-based water, judging that the water sample is sandstone water or Taigrey water by adopting a Fisher recognition method.
2. The method for identifying the water source of the complicated multi-aquifer mine, according to claim 1, characterized in that the water chemical data of the water sample are the mass concentration of each ion of the aquifer water of the coal mine to which the water sample belongs, and the milliequivalent and the ratio of each ion; the water chemistry data is used as a database building index for building a water source database.
3. The method for identifying the water inrush source of the complex multi-aquifer mine according to claim 1, wherein the data inspection method is a cation and anion balance inspection, and the error is controlled within +/-5%; the abnormal sample screening and processing method adopts one or more combinations of index histograms, box charts, Q-Q charts and cluster analysis charts to screen the abnormal samples and remove the water sample data as a modeling sample.
4. The method for identifying the water source of the complicated multi-aquifer mine inrush water as claimed in claim 1, wherein the identification index is selected by selecting an index capable of distinguishing each aquifer as an identification index according to water chemistry data of water samples of each aquifer, determining a threshold value of the index and identifying the identification relation when identifying the water source; the standard of the identification index selection is that the content change of the identification index in the aquifer water sample is larger than that of other indexes; a certain identification relation is that water samples of which the identification indexes are smaller than or larger than the determined identification index threshold value are classified into one class during identification; the identification index can at least distinguish water samples of two aquifers; wherein the identification index is the mass concentration of the conventional ions and the milliequivalent ratio of the conventional ions.
5. The method as claimed in claim 4, wherein the mass concentration of the conventional ions and the milliequivalent ratio of the conventional ions are respectively called the characteristic ion contrast and the ion proportionality coefficient.
6. The method for identifying the water inrush source for the complex multi-aquifer mine according to claim 1, wherein the specific method for judging the validity of the identification index in the preliminary identification model by adopting the back substitution test in the step S3 is as follows: taking a water sample containing a known water layer type as a new sample, sequentially substituting the new sample into the established preliminary identification model, and if the identification result is consistent with the actual result, indicating that the selection of the identification index and the established preliminary identification model are effective; and if the identification results of the same aquifer water sample are mostly inconsistent with the actual identification results, reselecting a new identification index and establishing a new initial identification model.
7. The method for identifying the water inrush source of the complex multi-aquifer mine according to claim 1, wherein the specific operation method of the comprehensive-step identification method is as follows: in the water sample after data verification and abnormal rejection: first, when TDS>4300mg/L、Cl - >2000mg/L、Na + +K + >1350mg/L, the water sample is aoThe grey water is identified; otherwise, further identifying when TDS<440mg/L、Cl - <50mg/L、Na + +K + <When the water sample is 80mg/L, the water sample is secondary water, and the identification is finished; otherwise, further identify when Gamma Cl - /γCa 2+ <5.2, the water sample is the cover-pushing body water, and is further identified when TDS<1360mg/L、Cl - <474mg/L, the water sample is the water of the coating body water sheet gneiss, and the identification is finished; otherwise, the water sample is the push-coating body water cold grey water, and the identification is finished; when gamma Cl is present - /γCa 2+ >And 5.2, identifying the sandstone water or the taigrey water by using a Fisher identification method, wherein the water sample is coal-based water, and the identification is finished at the moment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910298332.5A CN109993459B (en) | 2019-04-15 | 2019-04-15 | Method for identifying water bursting source of complex multi-aquifer mine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910298332.5A CN109993459B (en) | 2019-04-15 | 2019-04-15 | Method for identifying water bursting source of complex multi-aquifer mine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109993459A CN109993459A (en) | 2019-07-09 |
| CN109993459B true CN109993459B (en) | 2022-09-23 |
Family
ID=67133385
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910298332.5A Active CN109993459B (en) | 2019-04-15 | 2019-04-15 | Method for identifying water bursting source of complex multi-aquifer mine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109993459B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110852364A (en) * | 2019-10-31 | 2020-02-28 | 中国煤炭地质总局勘查研究总院 | Method and device for identifying water source of water burst in mine and electronic equipment |
| CN111562285A (en) * | 2020-06-03 | 2020-08-21 | 安徽大学 | Mine water inrush source identification method and system based on big data and deep learning |
| CN112381117B (en) * | 2020-10-22 | 2023-10-17 | 合肥工业大学 | Coal mine water inrush source mixing proportion calculation and dynamic monitoring method based on conventional water chemistry |
| CN112508330B (en) * | 2020-10-29 | 2024-04-23 | 中煤科工集团西安研究院有限公司 | Method for distinguishing mine water sources under disturbance of mining in western mining area |
| CN112633318B (en) * | 2020-11-04 | 2023-08-11 | 中国地质大学（北京） | Water source identification method based on Java and android platforms |
| CN114264680B (en) * | 2021-11-15 | 2023-06-13 | 中煤科工集团西安研究院有限公司 | Method for predicting concentration of fluoride in mine water based on comparison method |
| CN114167021A (en) * | 2021-12-09 | 2022-03-11 | 山西启诚电子科技有限公司 | Mine water source rapid identification instrument and control method thereof |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101923084A (en) * | 2010-07-15 | 2010-12-22 | 北京华安奥特科技有限公司 | A kind of mining water source recognition methods and identification equipment |
| CN103617147A (en) * | 2013-11-27 | 2014-03-05 | 中国地质大学（武汉） | Method for identifying mine water-inrush source |
| CN104122319A (en) * | 2014-08-13 | 2014-10-29 | 北京华安奥特科技有限公司 | Method and system for identifying water source in mining area based on ion composite electrode detecting technology and spectrum analysis technology |
| CN104297308A (en) * | 2014-10-23 | 2015-01-21 | 淮南矿业(集团)有限责任公司 | Device for rapidly distinguishing water inrush source of mine |
| CN104597516A (en) * | 2015-01-19 | 2015-05-06 | 天地科技股份有限公司 | Quick distinguishing system for water bursting source of mine |
| WO2016115816A1 (en) * | 2015-01-23 | 2016-07-28 | 中国矿业大学 | Structural discrimination indexes of ordovician limestone top filling zones and determination method |
| WO2017004552A1 (en) * | 2015-07-02 | 2017-01-05 | Abramowitz Rachel | Customized personal care management based on water quality and other environmental factors |
| CN108805357A (en) * | 2018-06-13 | 2018-11-13 | 安徽理工大学 | A kind of Fisher discrimination model water bursting source prediction techniques based on PCA analyses |
| CN108876030A (en) * | 2018-06-11 | 2018-11-23 | 安徽理工大学 | A kind of water bursting source prediction technique based on Fisher discrimination model |
| WO2019019709A1 (en) * | 2017-07-24 | 2019-01-31 | 厦门快商通科技股份有限公司 | Method for detecting water leakage of tap water pipe |
-
2019
- 2019-04-15 CN CN201910298332.5A patent/CN109993459B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101923084A (en) * | 2010-07-15 | 2010-12-22 | 北京华安奥特科技有限公司 | A kind of mining water source recognition methods and identification equipment |
| CN103617147A (en) * | 2013-11-27 | 2014-03-05 | 中国地质大学（武汉） | Method for identifying mine water-inrush source |
| CN104122319A (en) * | 2014-08-13 | 2014-10-29 | 北京华安奥特科技有限公司 | Method and system for identifying water source in mining area based on ion composite electrode detecting technology and spectrum analysis technology |
| CN104297308A (en) * | 2014-10-23 | 2015-01-21 | 淮南矿业(集团)有限责任公司 | Device for rapidly distinguishing water inrush source of mine |
| CN104597516A (en) * | 2015-01-19 | 2015-05-06 | 天地科技股份有限公司 | Quick distinguishing system for water bursting source of mine |
| WO2016115816A1 (en) * | 2015-01-23 | 2016-07-28 | 中国矿业大学 | Structural discrimination indexes of ordovician limestone top filling zones and determination method |
| WO2017004552A1 (en) * | 2015-07-02 | 2017-01-05 | Abramowitz Rachel | Customized personal care management based on water quality and other environmental factors |
| WO2019019709A1 (en) * | 2017-07-24 | 2019-01-31 | 厦门快商通科技股份有限公司 | Method for detecting water leakage of tap water pipe |
| CN108876030A (en) * | 2018-06-11 | 2018-11-23 | 安徽理工大学 | A kind of water bursting source prediction technique based on Fisher discrimination model |
| CN108805357A (en) * | 2018-06-13 | 2018-11-13 | 安徽理工大学 | A kind of Fisher discrimination model water bursting source prediction techniques based on PCA analyses |
Non-Patent Citations (6)
| Title |
|---|
| Recognition model of groundwater inrush source of coal mine: a case study on Jiaozuo coal mine in China;Huang ping-hua等;《Arabian Journal of Geosciences》;20170729;全文 * |
| 基于多元统计分析的矿井突水水源Fisher识别及混合模型;黄平华等;《煤炭学报》;20110531;第36卷;全文 * |
| 基于聚类分析方法的矿井水源识别;姚洁等;《煤矿安全》;20130220(第02期);全文 * |
| 多元统计分析模型在矿井突水水源判别中的应用;刘杰刚等;《中国煤炭》;20130222(第02期);全文 * |
| 新集一矿突水水源综合判别模型研究;金洲洋;《中国优秀硕士学位论文全文数据库（工程科技I辑）》;20160815(第08期);全文 * |
| 用Fisher判别法确定矿井突水水源;陈红江等;《中南大学学报(自然科学版)》;20090826(第04期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109993459A (en) | 2019-07-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109993459B (en) | Method for identifying water bursting source of complex multi-aquifer mine | |
| CN105095238B (en) | For detecting the decision tree generation method of fraudulent trading | |
| CN106709242A (en) | Method for identifying authenticity of sewage monitoring data | |
| CN113838054B (en) | Mechanical part surface damage detection method based on artificial intelligence | |
| CN110852364A (en) | Method and device for identifying water source of water burst in mine and electronic equipment | |
| CN106529571A (en) | Multilayer image characteristics extraction modeling, image recognition method and apparatus based on neural network | |
| CN110261560A (en) | The water source recognition methods of complex hydrologic geology water bursting in mine and system | |
| CN108844941B (en) | Method for identifying and classifying different-grade phosphate ores based on Raman spectrum and PCA-HCA | |
| CN113516228B (en) | Network anomaly detection method based on deep neural network | |
| CN104951842A (en) | Novel method for predicting oil field output | |
| Kumar et al. | Application of chemometric techniques in the assessment of groundwater pollution in a suburban area of Chennai city, India | |
| CN110619418A (en) | Multi-feature water quality prediction method based on mixed model combination algorithm | |
| CN112926893A (en) | Horizontal well profile control effect evaluation method based on fuzzy comprehensive evaluation and hierarchical analysis | |
| Prentice | Quantitative birch (Betula L.) pollen separation by analysis of size frequency data | |
| CN105653788A (en) | Mining area water source recognition method, mining area water source recognition device and mining area water source recognition system | |
| CN108931815A (en) | A kind of hierarchical identification method of lithology | |
| CN106528665A (en) | Method and system for searching testing document of AOI equipment | |
| CN110222981B (en) | Reservoir classification evaluation method based on parameter secondary selection | |
| CN113570652A (en) | Quantitative analysis method for sandstone reservoir mineral intercrystalline pores based on SEM image | |
| CN116823047A (en) | Lake eutrophication evaluation method based on Monte Carlo-variable fuzzy set | |
| CN111881974B (en) | Water inrush source identification method based on Piper-PCA-FCL discrimination model | |
| CN109779622B (en) | Method and device for characterizing low-efficiency water injection zone of oil reservoir in ultrahigh water cut period | |
| Congestri et al. | Evaluating biomass of Baltic filamentous cyanobacteria by image analysis | |
| CN110826762A (en) | Mine water disaster evolution trend prediction method suitable for complex groundwater environment | |
| CN105824785A (en) | Rapid abnormal point detection method based on penalized regression |
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 |