Mine water filling source modeling analysis method

Abstract

The invention provides a mine water filling source modeling analysis method, and relates to the technical field of water source analysis. 2) constructing a single-index measurement function and a machine selection index weight; (3) carrying out EW-UVA model effect test and analysis; finally, the result of identifying the five test sample water filling sources by using the EW-UCA model is completely consistent with the actual condition, so that the application of the EW-UCA model to mine water source identification is scientific and feasible; the method comprises the following steps: optimizing an unascertained clustering method, comprehensively considering the importance of hydrochemical indexes on water source recognition, selecting seven hydrochemical indexes such as Na + + K +, Mg < 2 + >, Ca < 2 + >, HCO3, CI <->, SO2-4 and TH as discrimination indexes for water source recognition, and establishing an EW-UCA model for mine water filling source recognition by taking 22 groups of water sample data as training samples; the EW-UCA water source identification model is used for identifying and inspecting five groups of test sample water samples, and is compared and analyzed with the actual situation, and the result shows that the EW-UCA water source identification model has higher accuracy and stronger applicability.

Description

Technical Field

[0001] The present invention relates to the technical field of water source analysis, and in particular to a mine filling water source modeling and analysis method. Background Art

[0002] During the production process of coal mines, mining projects expose and destroy aquifers, impermeable layers and water-conducting faults on the one hand, and cause the movement of surrounding rock strata and surface collapse on the other hand, which will lead to the entry of groundwater or surface water. This kind of water that enters the shaft, tunnel and working face during the development and mining of mines is called mine water. There are many sources of mine water. They enter the mine mining project through various means such as infiltration, dripping, pouring, inflow, gushing and bursting. The phenomenon is called mine water filling or mine water inrush. The replenishment water source for mine water filling is called mine water filling source. The passage for mine water to enter the mine is called water filling channel.

[0003] How to reasonably and effectively prevent and control mine water hazards is a major problem that needs to be solved urgently in mines. The premise of mine water hazard prevention and control is to timely and accurately identify the source of mine water filling. Commonly used criteria or methods include water chemical indicators, trace elements, isotopes, dynamic changes in water levels, etc. Practice has shown that the essential characteristics of groundwater depend on its chemical composition. Using water chemical indicators to identify water sources has the advantages of convenience, reliability, economy and high efficiency. In recent years, methods based on water chemical indicators to identify water sources have developed rapidly. In addition, some mathematical and statistical methods such as Fisher discriminant method, distance discriminant analysis method, support vector machine method, extension theory, Bayes method, etc. have been successively used in the identification of water sources and have achieved certain results. Practice has shown that although the above methods have their own advantages, they all have certain limitations in application, so a modeling and analysis method for mine water filling sources is designed.

[0004] Some existing mathematical and statistical methods for identifying water sources based on water chemical indicators of mine filling have low accuracy, poor applicability and certain limitations. Summary of the invention

[0005] The disclosed embodiments relate to a mine water source modeling and analysis method to solve the problems of low accuracy, poor applicability and certain limitations of some existing mathematical and statistical methods for identifying water sources using water chemical indicators of mine water.

[0006] In a first aspect, the present disclosure provides a mine water source modeling and analysis method, comprising the following steps: 1). Selection of discrimination indicators; At present, the common water sources for filling water in coal mines are: surface water, Cenozoic loose aquifer, Permian sandstone fissure aquifer, Carboniferous karst fissure aquifer, Ordovician karst fissure aquifer, old pond water, etc. Among them, Permian sandstone water, Carboniferous limestone water, Ordovician limestone water conservancy old pond water are generally the direct water sources for filling water in mines. Studies have shown that different groundwaters or even the same groundwater have different hydrological environments and recharge, runoff and discharge conditions in different locations, and the chemical compositions they contain are naturally different. It can be seen that it is obviously unrealistic to use all the water chemical indicators of the water source to identify the water source. Combining the research results of predecessors, comprehensively considering the importance of various indicators and the validity of the data, Na is selected. + +K + Mg 2+ , Ca 2 + , HCO ¯ 3. CI - 、SO 2- 4. TH (total hardness) and other 7 indicators are used as the identification indicators of water sources. The hydrological environment of groundwater at different depths and locations is different. Therefore, in order to improve the accuracy of water source identification, sufficient and representative water source data should be selected for research; 1.1). Determination of training samples and classification criteria for discriminant indicators; 27 groups of water samples were collected, 22 of which were used as training samples and the remaining 5 were used as test samples. Cluster mean calculation was performed on different distant category data based on the training sample data, and the sample means of 7 discriminant indicators under 3 water source classification modes were obtained. Thus, the classification standards of each discriminant indicator of mine water source were established. 2). Construct a single indicator measurement function and randomly select indicator weights; 2.1) According to the classification standard of mine water source identification index, establish Na + +K + Mg 2+ , Ca 2+ , HCO ¯ 3. CI - 、SO 2- 4. Single index measurement function of TH and other 7 discriminant indexes, and calculate Na respectively + +K + Mg 2+ , Ca 2+ , HCO ¯ 3. CI - 、SO 2- 4. The weight of TH, and the weight vector α of the discriminant index is obtained; 3).EW-UVA model effect test and analysis; The remaining five groups of water samples were used as test samples, and the EW-UCA model was used to identify their water sources and the identification results were compared with the actual situation. 3.1). Take the test sample 1 as an example to identify and verify its water source. The steps are as follows: + +K + The concentration is 19.24 mg / L, based on Na + +K + The single index measurement function of the discriminant index is used to calculate its unascertained measurement relative to the three types of water sources: coal-based water, Ordovician ash water and Taihu ash water, which is (0.6604, 03396, 0). Similarly, the unascertained measurements of the six indicators of the water sample relative to the three types of water sources can be calculated respectively. Thus, the single index measurement matrix of test sample 1 can be obtained. According to the single index measurement matrix of sample 1, the multi-index comprehensive measurement vector of sample 1 is calculated, and k0=1 is obtained. Obviously, test sample 1 belongs to Class B, that is, the water source is coal-based water, which is completely consistent with the actual situation. Similarly, the multi-index comprehensive measurement vectors and water source categories of the other four test samples can be obtained in turn. Finally, it was concluded that the results of using the EW-UCA model to identify the water sources of the five test samples were completely consistent with the actual situation. It can be seen that it is scientific and feasible to use the EW-UCA model for mine water source identification.

[0007] The present invention provides a mine water source modeling and analysis method, which has the following beneficial effects: The present invention optimizes the unascertained clustering method, comprehensively considers the importance of water chemical indicators to water source identification, and selects Na + +K + Mg 2+ , Ca 2+ , HCO ¯ 3. CI - 、SO 2- 4. Seven water chemical indicators such as TH were used as discriminant indicators for water source identification. With 22 groups of water sample data as training samples, the EW-UCA model for mine water source identification was established. Then, the model was used to identify and test 5 groups of test sample water samples, and a comparative analysis was made with the actual situation. The results showed that the EW-UCA water source identification model has high accuracy and strong applicability, and can be promoted and applied in engineering practice. DETAILED DESCRIPTION

[0008] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "install", "set", "connect", "fix", "screw" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined, for ordinary technicians in the field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. . Example

[0011] The present invention proposes a mine water source modeling and analysis method, comprising the following steps: 1). Selection of discrimination indicators; At present, the common water sources for filling water in coal mines are: surface water, Cenozoic loose aquifer, Permian sandstone fissure aquifer, Carboniferous karst fissure aquifer, Ordovician karst fissure aquifer, old pond water, etc. Among them, Permian sandstone water, Carboniferous limestone water, Ordovician limestone water conservancy old pond water are generally the direct water sources for filling water in mines. Studies have shown that different groundwaters or even the same groundwater have different hydrological environments and recharge, runoff and discharge conditions in different locations, and the chemical compositions they contain are naturally different. It can be seen that it is obviously unrealistic to use all the water chemical indicators of the water source to identify the water source. Combining the research results of predecessors, comprehensively considering the importance of various indicators and the validity of the data, Na is selected. + +K + Mg 2+ , Ca 2 + , HCO ¯ 3. CI - 、SO 2- 4. TH (total hardness) and other 7 indicators are used as the identification indicators of water sources. The hydrological environment of groundwater at different depths and locations is different. Therefore, in order to improve the accuracy of water source identification, sufficient and representative water source data should be selected for research; 1.1). Determination of training samples and classification criteria for discriminant indicators; 27 groups of water samples were collected, 22 of which were used as training samples and the remaining 5 were used as test samples. Cluster mean calculation was performed on different distant category data based on the training sample data, and the sample means of 7 discriminant indicators under 3 water source classification modes were obtained. Thus, the classification standards of each discriminant indicator of mine water source were established. 2. Construct a single indicator measurement function and randomly select indicator weights; 2.1) According to the classification standard of mine water source identification index, establish Na + +K + Mg 2+ , Ca 2+ , HCO ¯ 3. CI - 、SO 2- 4. Single index measurement function of TH and other 7 discriminant indexes, and calculate Na respectively + +K + Mg 2+ , Ca 2+ , HCO ¯ 3. CI - 、SO 2- 4. The weight of TH, and the weight vector α of the discriminant index is obtained; 3).EW-UVA model effect test and analysis; The remaining five groups of water samples were used as test samples, and the EW-UCA model was used to identify their water sources and the identification results were compared with the actual situation. 3.1). Take the test sample 1 as an example to identify and verify its water source. The steps are as follows: + +K + The concentration is 19.24 mg / L, based on Na + +K + The single index measurement function of the discriminant index is used to calculate its unascertained measurement relative to the three types of water sources: coal-based water, Ordovician ash water and Taihu ash water, which is (0.6604, 03396, 0). Similarly, the unascertained measurements of the six indicators of the water sample relative to the three types of water sources can be calculated respectively. Thus, the single index measurement matrix of test sample 1 can be obtained. According to the single index measurement matrix of sample 1, the multi-index comprehensive measurement vector of sample 1 is calculated, and k0=1 is obtained. Obviously, test sample 1 belongs to Class B, that is, the water source is coal-based water, which is completely consistent with the actual situation. Similarly, the multi-index comprehensive measurement vectors and water source categories of the other four test samples can be obtained in turn. Finally, it was concluded that the results of using the EW-UCA model to identify the water sources of the five test samples were completely consistent with the actual situation. It can be seen that it is scientific and feasible to use the EW-UCA model for mine water source identification.

[0012] The above are only specific implementation methods of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present disclosure, which should be covered by the protection scope of the present disclosure.

Claims

1. A mine water source modeling and analysis method, characterized by: The following steps are involved: 1). Selection of discrimination indicators; At present, the common water sources for filling water in coal mines are: surface water, Cenozoic loose aquifer, Permian sandstone fissure aquifer, Carboniferous karst fissure aquifer, Ordovician karst fissure aquifer, old pond water, etc. Among them, Permian sandstone water, Carboniferous limestone water, Ordovician limestone water conservancy old pond water are generally the direct water sources for filling water in mines. Studies have shown that different groundwaters or even the same groundwater have different hydrological environments and recharge, runoff and discharge conditions in different locations, and the chemical compositions they contain are naturally different. It can be seen that it is obviously unrealistic to use all the water chemical indicators of the water source to identify the water source. Combining the research results of predecessors, comprehensively considering the importance of various indicators and the validity of the data, Na is selected. + +K + Mg 2+ , Ca 2+ , HCO ¯ 3. CI - 、SO 2- 4. TH (total hardness) and other 7 indicators are used as the identification indicators of water sources. The hydrological environment of groundwater at different depths and locations is different. Therefore, in order to improve the accuracy of water source identification, sufficient and representative water source data should be selected for research; 1.1). Determination of training samples and classification criteria for discriminant indicators; 27 groups of water samples were collected, 22 of which were used as training samples and the remaining 5 were used as test samples. Cluster mean calculation was performed on different distant category data based on the training sample data, and the sample means of 7 discriminant indicators under 3 water source classification modes were obtained. Thus, the classification standards of each discriminant indicator of mine water source were established. 2). Construct a single indicator measurement function and randomly select indicator weights; 2.1) According to the classification standard of mine water source identification index, establish Na + +K + Mg 2+ , Ca 2+ , HCO ¯ 3. CI - 、SO 2- 4. Single index measurement function of TH and other 7 discriminant indexes, and calculate Na respectively + +K + Mg 2+ , Ca 2+ , HCO ¯ 3. CI - 、SO 2- 4. The weight of TH, and the weight vector α of the discriminant index is obtained; 3).EW-UVA model effect test and analysis; The remaining five groups of water samples were used as test samples, and the EW-UCA model was used to identify their water sources and the identification results were compared with the actual situation. 3.1). Take the test sample 1 as an example to identify and verify its water source. The steps are as follows: + +K + The concentration is 19.24 mg / L, based on Na + +K + The single index measurement function of the discriminant index is used to calculate its unascertained measurement relative to the three types of water sources: coal-based water, Ordovician ash water and Taihu ash water, which is (0.6604, 03396, 0). Similarly, the unascertained measurements of the six indicators of the water sample relative to the three types of water sources can be calculated respectively. Thus, the single index measurement matrix of test sample 1 can be obtained. According to the single index measurement matrix of sample 1, the multi-index comprehensive measurement vector of sample 1 is calculated, and k0=1 is obtained. Obviously, test sample 1 belongs to Class B, that is, the water source is coal-based water, which is completely consistent with the actual situation. Similarly, the multi-index comprehensive measurement vectors and water source categories of the other four test samples can be obtained in turn. Finally, it was concluded that the results of using the EW-UCA model to identify the water sources of the five test samples were completely consistent with the actual situation. It can be seen that it is scientific and feasible to use the EW-UCA model for mine water source identification.