Method and device for predicting drainage volume of controlled aquifer in mining area
By constructing a BP neural network model, combining historical data and water chemistry analysis, the problem of difficult to predict and control the aquifer drying amount in mine drying operations is solved, and accurate prediction of the aquifer drying amount in the next cycle of the mine area is achieved. Scientific and reasonable groundwater management.
Patent Information
- Application Number
- CN202410661697.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-27
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-05-27
AI Technical Summary
In mine dry drainage operations, due to insufficient geological structure exploration accuracy and hydraulic connections caused by years of mining, it is difficult to accurately predict the dry drainage amount of aquifers during the next mining cycle, which affects the storage and sustainability of groundwater resources.
By obtaining historical hydrological measurement data and mining reports in the mining area, a prediction model for dry sparse amount of mining area control aquifers is constructed based on the BP neural network. Using historical data and predetermined values, combining water chemical analysis and meteorological data, inputting planned coal mining volume and expected sparse amounts are carried out. Dry quantity, optimize the model to quantitatively predict the dryness of the controlled aquifer in the next cycle.
It provides a scientific and effective method to accurately predict the drying amount of aquifer in the mining area, help the groundwater management department to make more scientific and reasonable management decisions, and improve the accuracy and reliability of the drying amount prediction.
Smart Images

Figure CN118551557B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of mine drainage water control, and more specifically, to a method and device for predicting the drainage amount of a controlled aquifer in a mining area. Background Art
[0002] Drainage water, as the water discharged during coal mining to ensure production, greatly affects the storage and sustainability of groundwater resources in the mining area. In mine drainage operations, in theory, the discharged water is only the filling of the coal-bearing aquifer. However, due to the existence of a hydraulic channel between the coal-bearing aquifer and the controlled aquifer that affects the ecological environment of the mining area, the drainage behavior often has a great impact on the controlled aquifer.
[0003] In groundwater management, the drainage volume of the mine in the next period is generally estimated through the expected mining plan of the mine in the next period. However, many mining areas have insufficient geological structure exploration accuracy and less data. The geological structure of the mining area is complex and due to years of mining operations, there are channels such as fracture structures between the various aquifers, which cause hydraulic connections. This makes it difficult to predict the drainage volume of the controlled aquifer in the next mining period. Therefore, it is necessary to invent a method and device for predicting the drainage volume of the controlled aquifer in a mining area, which can use the local historical mining volume and drainage behavior to quantitatively predict the drainage volume of the controlled aquifer in the next period. Summary of the invention
[0004] One purpose of the present invention is to provide a method and device for predicting the drainage volume of a controlled aquifer in a mining area, which can use the local historical mining volume and drainage behavior to quantitatively predict the drainage volume of the controlled aquifer in the next cycle, so that administrative personnel can more scientifically and reasonably predict the drainage volume of the controlled aquifer in the mining area they manage.
[0005] In order to achieve these purposes and other advantages according to the present invention, according to one aspect of the present invention, the present invention provides a method and device for predicting the amount of drainage of aquifers under control in a mining area, comprising the following steps:
[0006] S1. Obtain historical hydrological measured data and mining reports of the mining area over the years. The historical hydrological measured data of the mining area include the drainage water volume of each mine in the mining area over the years, the hydrochemical analysis report over the years, and the precipitation report of the mining area over the years. The mining report includes the mining volume of each mine in the mining area over the years, the tunnel conditions of the mining area, and the mining plan for the next control cycle;
[0007] S2. Obtain the mine drainage volume, mine control aquifer drainage volume, and coal mining volume of each mine in the mining area over the years, and summarize them into the total mine drainage volume, mine control aquifer drainage volume, and mine coal mining volume over the years, and use the mine area's meteorological data over the years to obtain the mine area's water recharge volume over the years;
[0008] S3. Using the mining plan of each mine in the mining area in the next control cycle, the planned coal mining volume and the expected drainage volume of the mining area are obtained, and the expected water replenishment volume of the mining area is obtained by using the meteorological data of the mining area;
[0009] S4. Use the total drainage volume of the mine area, the water replenishment volume of the mine area, the coal mining volume of the mine area, and the drainage volume of the controlled aquifer in the mine area in previous years as the sample set data to build a prediction model for the drainage volume of the controlled aquifer in the mine area. Input the planned coal mining volume of the mine area, the expected drainage volume of the mine area, and the expected water replenishment volume of the mine area to obtain the predicted value of the drainage volume of the controlled aquifer in the mine area.
[0010] Preferably, the method of obtaining the annual mining area water supply volume using the annual meteorological data of the mining area comprises the following steps:
[0011] A1: Obtain geological information of the mining area: According to the geological report, the mining area is divided into various types of recharge unit areas, the area of each area is confirmed and the corresponding infiltration coefficient is determined;
[0012] A2: Determine the precipitation replenishment information of the mining area: Obtain the annual precipitation of the mining area based on the precipitation report records of the meteorological stations around the mining area over the years;
[0013] A3: Calculate the annual water recharge for each geological region within the mining area and summarize it. The calculation formula for each geological region is as follows:
[0014] Qa=10 -1 ·F·A·P
[0015] Where Qa is the regional recharge water volume, m 3 / a;
[0016] F—recharge area, m 2 ;
[0017] A—infiltration coefficient;
[0018] P—average annual precipitation, mm / a.
[0019] Preferably, obtaining the drainage amount of the mining area controlled aquifer includes the following steps:
[0020] B1: Carry out drainage measurement and water sampling of each mine in the mining area: Measure the drainage water of each mine and conduct water sampling regularly, collect water samples of mine drainage water, coal-bearing aquifers and controlled aquifers periodically, and send samples for testing and storage of results;
[0021] B2: Determine the source of drainage water in each mine: The water source is determined based on the hydrochemical characteristics of the drainage water and the hydrochemical characteristics of each aquifer, and is analyzed using the hydrochemical piper three-line diagram and isotope method. The hydrochemical piper three-line diagram is based on Na + , K + , Ca2+ Mg 2+ , Cl - 、SO4 2- 、HCO3 - Ions are used as discriminant indicators to obtain the drainage volume of the mine-controlled aquifer in each mine;
[0022] B3: Summarize the drainage volume of each mine-controlled aquifer in the mining area to obtain the drainage volume of the controlled aquifer in the mining area.
[0023] Preferably, the step S4 comprises:
[0024] S41, using a BP neural network including an input layer, a hidden layer and an output layer, wherein the input layer of the BP neural network model includes a plurality of input neuron nodes, and the hidden layer is provided with a plurality of hidden neuron nodes, and each of the input neuron nodes is respectively connected to each of the hidden neuron nodes; the hidden neuron nodes are all connected to the output layer, and the total drainage volume of the mining area, the water replenishment volume of the mining area, and the coal mining volume of the mining area are used as input values of the input neuron nodes, and the output result is compared with the drainage volume of the corresponding mining area controlled aquifer, and the BP neural network model is optimized according to the PSO algorithm to obtain a prediction model for the drainage volume of the mining area controlled aquifer;
[0025] S42. Obtain the planned coal production volume, expected water replenishment volume and expected drainage volume of the mining area for the next cycle, and input them into the optimized mining area controlled aquifer drainage volume prediction model to obtain the predicted value of the mining area controlled aquifer drainage volume for the next cycle.
[0026] Preferably, the method for optimizing the BP neural network comprises:
[0027] S411, taking the total drainage volume of the mining area, the water supply volume of the mining area, and the coal mining volume of the mining area as the influencing factors, the total number of parameter types as the number of neuron nodes m, and the drainage volume of the mining area controlled aquifer as the only output value, that is, the number of nodes in the output layer c=1, and the number of nodes in the hidden layer c1 is , a is a random constant between 1 and 10;
[0028] S412, normalize the sample set data, and its mathematical expression is:
[0029]
[0030] in, represents the sample data of the influencing factors, , are the minimum and maximum values in the sample data, respectively. is the influencing factor data after dimensionless processing;
[0031] S413, initializing the mapping relationship between the population particles and the BP neural network weights and thresholds, including particle dimensions, initial velocity, population size, learning factor, and inertia weight;
[0032] S414, input the normalized input variables and output variables into the BP neural network model, calculate the fitness function value of the particle, and obtain the historical optimal fitness and global fitness of the particle. The fitness function value of the particle is the mean square error of the calculation result, and its function expression is:
[0033]
[0034] in, Represented as the predicted value of the i-th sample, is the true value of the i-th sample, and n is the total number of calculation results of the neural network;
[0035] S415, iteratively calculating the particle fitness, and updating the historical optimal fitness and the global fitness according to a preset update condition, until a preset iteration end condition is met;
[0036] S416. Update the weights and thresholds of the preset BP neural network model to obtain an optimized BP neural network model for calculating the drainage amount of the controlled aquifer in the mining area.
[0037] Preferably, the method for predicting the drainage volume of a controlled aquifer in a mining area is characterized in that the expected drainage volume of the mining area is obtained by summarizing the expected drainage volumes of the mines in the mining area, and the expected drainage volume of the mines is calculated using the hydrogeological parameters obtained by the pumping test, and the calculation formula is as follows:
[0038] ;
[0039] R
[0040] R0= R +r0
[0041] Where Q is the predicted drainage volume, m 3 ;
[0042] K—permeability coefficient;
[0043] S—water level drop, m;
[0044] M—aquifer thickness, m;
[0045] R0—reference influence radius of the prediction area, m;
[0046] r0—the converted radius of the prediction area, m;
[0047] R—influence radius, m.
[0048] A mining area controlled aquifer drainage prediction device, applied to the above-mentioned mining area controlled aquifer drainage prediction method, comprising: a plurality of mine data acquisition modules, a mining area sample set acquisition module and a controlled aquifer drainage prediction module;
[0049] The mine data acquisition module is arranged in each mine in the mining area, and can obtain the mine drainage volume and coal mining volume of the mine in the cycle, and obtain the drainage volume of the mine-controlled aquifer in the current cycle by means of water chemical analysis. The mine data acquisition module can record the planned mining volume of the mine in the next cycle and obtain the corresponding expected drainage volume of the mine;
[0050] The mining area sample set acquisition module can collect precipitation records from meteorological stations around the mining area over the years, and calculate the mining area water replenishment volume over the years based on the distribution of different rock formations in the mining area. The sample set acquisition module can summarize the drainage data of each mine over the years, thereby obtaining the total drainage volume of the mining area over the years and the drainage volume of the controlled aquifer in the mining area over the years. The sample set acquisition module can obtain the planned coal mining volume of each mine in the mining area and the expected drainage volume of the mine, and obtain the coal mining volume of the mining area over the years, the planned coal mining volume of the mining area in the next cycle, and the expected drainage volume of the mining area;
[0051] The controlled aquifer depletion prediction module is used to optimize the BP neural network model, and the expected depletion volume of the mining area, the expected water replenishment volume of the mining area, and the planned coal mining volume of the mining area are input into the optimized mining area controlled aquifer depletion prediction model to obtain the predicted value of the mining area controlled aquifer depletion volume.
[0052] A computer device, characterized in that it includes a memory and a processor, wherein the memory stores computer instructions, and the processor executes the above-mentioned method for predicting the drainage amount of aquifers under mining area control by executing the computer instructions.
[0053] The present invention has at least the following beneficial effects:
[0054] It provides a scientific and effective method for predicting the drainage volume of the controlled aquifer. The logic is clear and reasonable. It uses a neural network model to accurately predict the drainage volume of the controlled aquifer in the mining area in the next cycle based on historical data and multiple predetermined values, which is convenient for the groundwater management department to manage.
[0055] Other advantages, objectives and features of the present invention will be embodied in part through the following description, and in part will be understood by those skilled in the art through study and practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0056] Figure 1 It is a flow chart of a method for predicting the drainage amount of aquifers under control in a mining area in a technical solution of the present invention;
[0057] Figure 2 A distribution map of mines in the over-exploited area of the present invention;
[0058] Figure 3 The historical data table of each mine in the over-exploited area of the present invention;
[0059] Figure 4 A comparison chart of the prediction results of drainage water from the controlled aquifer in the over-exploited area of the present invention with the actual data based on the year as the time node;
[0060] Figure 5 This is a schematic diagram of the water supply area of zone H of the present invention;
[0061] Figure 6 The precipitation statistics table of the weather stations around the H area described in the present invention;
[0062] Figure 7 The distribution map of the coal mine containing aquiclude of the ZF coal mine of the present invention;
[0063] Figure 8 The ZF coal mine water sample collection location and the corresponding hydrogeochemical and isotope analysis results statistical table of the present invention;
[0064] Fig. 9 This is the isotope relationship diagram of each water source of the ZF coal mine described in the present invention;
[0065] Fig.10 It is the three-line diagram of the ZF coal mine water sample piper described in the present invention. DETAILED DESCRIPTION
[0066] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can implement the invention with reference to the description.
[0067] It should be understood that terms such as “having”, “including” and “comprising” used herein do not exclude the existence or addition of one or more other elements or combinations thereof.
[0068] It should be noted that the experimental methods described in the following embodiments are conventional methods unless otherwise specified, and the reagents and materials can be obtained from commercial sources unless otherwise specified; in the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "set" should be understood in a broad sense, for example, they can be fixedly connected, set, or detachably connected, set, or connected and set in one piece. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood in specific circumstances. The orientation or position relationship indicated by the terms "lateral", "longitudinal", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc. is based on the orientation or position relationship shown in the accompanying drawings, which is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention.
[0069] like Figure 1-10 As shown, the present invention provides a method for predicting the drainage amount of aquifers under control in a mining area, comprising the following steps:
[0070] S1. Obtain historical hydrological measured data and mining reports of the mining area over the years. The historical hydrological measured data of the mining area include the drainage water volume of each mine in the mining area over the years, the hydrochemical analysis reports of the years, and the precipitation reports of the mining area over the years. The mining report includes the coal mining volume of each mine in the mining area over the years, the conditions of the mining area tunnels, and the mining plan for the next control cycle. Specifically, the drainage water is the gushing water discharged from the inside and outside of the tunnels during normal mining and tunnel excavation of the mine. The mine drainage volume is the total amount of water discharged from the inside and outside of the mine tunnels in a certain year. The hydrochemical analysis report includes the source report of the drainage water obtained by sampling at different positions of the mine. The precipitation report of the mining area includes the precipitation data of the current year recorded by the meteorological stations and precipitation monitoring stations around the mining area. The mine mining report needs to reflect the coal mining volume of the mine in that year and the main mined coal seams and the rock strata in which they are located. The mining plan for the next control cycle needs to determine the coal seams planned to be mined in the next year.
[0071] S2. Obtain the mine drainage volume, mine control aquifer drainage volume, and coal mining volume of each mine in the mining area over the years, and summarize them into the total mine drainage volume, mine control aquifer drainage volume, and mine coal mining volume over the years. Use the meteorological data of the mining area over the years to obtain the mine area recharge water volume over the years. Specifically, the mine control aquifer drainage volume is the water volume in the mine drainage that comes from the aquifer restricted by the groundwater management department, and the mine area recharge water volume is the additional water volume obtained by various geological aquifers in the mining area through natural rainfall;
[0072] S3. Utilize the mining plan of each mine in the mining area for the next control cycle to obtain the planned coal mining volume and the expected drainage volume of the mining area. Utilize the meteorological data of the mining area to obtain the expected water replenishment volume of the mining area. Specifically, the expected drainage volume of the mining area can be obtained by summarizing the expected drainage volume of each mine under the mining area. The mined coal seams and mining volume of the mine in the next control cycle can be calculated by analytical method, analogy method, water-rich coefficient method and other methods to calculate the mine water inflow in the cycle, which can be regarded as the expected drainage volume of the mine.
[0073] S4. Use the total drainage volume of the mining area, the water replenishment volume of the mining area, the coal mining volume of the mining area, and the drainage volume of the controlled aquifer in the mining area in previous years as the sample set data to build a mining area controlled aquifer drainage volume prediction model, and input the planned coal mining volume, the expected drainage volume of the mining area, and the expected water replenishment volume of the mining area to obtain the predicted value of the drainage volume of the controlled aquifer in the mining area.
[0074] In this technical solution, the controlled aquifer in the mining area is a geological aquifer used by the people and enterprises in the mining area for their lives and production. In theory, the drainage water in the mining area will not affect the controlled aquifer. However, due to the influence of underground geological activities and human mining, there is a hydraulic channel between the controlled aquifer and the directly drained coal-bearing aquifer. The mine drainage behavior also causes drainage of the controlled aquifer, and the drainage of the controlled aquifer is difficult to plan and estimate in advance through conventional means.
[0075] In this technical solution, since there are many influencing factors in the mining area, and it is difficult to establish an accurate linear relationship between the factors, especially it is impossible to simply obtain the predicted value of the drainage volume of the controlled aquifer in the mining area, the prediction model of the drainage volume of the controlled aquifer in the mining area is built based on the BP neural network model. The BP neural network is a multi-layer feedforward network, and the three-layer neural network with a hidden layer is relatively more commonly used. Studies have shown that any three-layer BP neural network can approximate any nonlinear continuous function and can achieve any accuracy. Since the controlled aquifer in the mining area generally has a hydraulic connection with the coal-bearing aquifer directly drained by the mining enterprise, the drainage volume of the controlled aquifer in the mining area is closely related to the coal mining volume, the total drainage volume and the water replenishment volume of the mining area. However, it is difficult to find and determine the precise hydraulic channel, and it is difficult to formulate an accurate linear formula to directly infer the drainage behavior of the mining area for the controlled aquifer under this mining plan according to the mining plan.
[0076] In the present technical solution, the sample set of the BP neural network is the total drainage volume of the mining area, the water replenishment volume of the mining area, the coal mining volume of the mining area, and the drainage volume of the controlled aquifer in the mining area. If the data volume of a certain mining area is too small due to incomplete data collection, the data obtained from the mining area with similar geographical and hydrological conditions can be used to supplement the sample set. If the geographical and hydrological conditions of the mining area from which the data is called are significantly different from those of the current mining area, it is necessary to increase the factor and correct the data in other ways.
[0077] In this technical solution, the total amount of mine drainage, the amount of mine water replenishment, and the amount of coal mining in the mine area over the years are used as influencing factors as the input of the BP neural network, and the amount of mine-controlled aquifer drainage is used as the output of the BP neural network. If the predicted output of the BP neural network is significantly different from the actual output, the output signal of the neural network begins to pass through the hidden layers from the output layer in the direction of reducing the error, and gradually corrects the connection weights within the BP neural network until the output of the neural network meets the expected output or the gap is not large, thereby obtaining a mine-controlled aquifer drainage prediction model that meets the prediction needs.
[0078] In another technical solution, the method of obtaining the annual mining area water supply volume using the annual meteorological data of the mining area includes the following steps:
[0079] A1: Obtain geological information of the mining area: According to the geological report, the mining area is divided into various types of recharge unit areas, the area of each area is confirmed and the corresponding infiltration coefficient is determined. Specifically, the recharge unit area needs to be determined based on the geological structure characteristics and watershed area of the unit area;
[0080] A2: Determine the precipitation replenishment information of the mining area: According to the precipitation report records of the meteorological stations around the mining area over the years, the annual precipitation of the mining area can be obtained. Specifically, the above-mentioned meteorological stations can be divided on the map using Thiessen polygons to obtain the multi-year rainfall data of each area within the calculation area;
[0081] A3: Calculate the annual water recharge volume of each geological region within the mining area and summarize it. The calculation formula for a separate geological region is as follows:
[0082] Qa=10 -1 ·F·A·P
[0083] Where Qa is the regional recharge water volume, m 3 / a;
[0084] F—recharge area, m 2 ;
[0085] A—infiltration coefficient;
[0086] P—average annual precipitation, mm / a.
[0087] In another technical solution, obtaining the drainage amount of the mining area controlled aquifer includes the following steps:
[0088] B1: Carry out drainage measurement and water sampling of each mine in the mining area: Measure the drainage water of each mine and conduct water sampling regularly, collect water samples of mine drainage water, coal-bearing aquifers, and controlled aquifers periodically, and send samples for testing and storage of results;
[0089] B2: Determine the source of drainage water in each mine: The water source is determined based on the hydrochemical characteristics of the drainage water and the hydrochemical characteristics of each aquifer, and is analyzed using the hydrochemical piper three-line diagram and isotope method. The hydrochemical piper three-line diagram is based on Na + , K + , Ca 2+ Mg 2+ , Cl - 、SO4 2- 、HCO3 - Ions are used as discriminant indicators to obtain the drainage volume of the mine-controlled aquifer in each mine;
[0090] B3: Summarize the drainage volume of each mine-controlled aquifer in the mining area to obtain the drainage volume of the controlled aquifer in the mining area.
[0091] In another technical solution, step S4 includes:
[0092] S41, using a BP neural network including an input layer, a hidden layer and an output layer, wherein the input layer of the BP neural network model includes a plurality of input neuron nodes, and the hidden layer is provided with a plurality of hidden neuron nodes, and each of the input neuron nodes is respectively connected to each of the hidden neuron nodes; the hidden neuron nodes are all connected to the output layer, and the total drainage volume of the mining area, the water replenishment volume of the mining area, and the coal mining volume of the mining area are used as input values of the input neuron nodes, and the output result is compared with the drainage volume of the corresponding mining area controlled aquifer, and the BP neural network model is optimized according to the PSO algorithm to obtain a prediction model for the drainage volume of the mining area controlled aquifer;
[0093] S42. Obtain the planned coal production volume, expected water replenishment volume and expected drainage volume of the mining area for the next cycle, and input them into the optimized mining area controlled aquifer drainage volume prediction model to obtain the predicted value of the mining area controlled aquifer drainage volume for the next cycle.
[0094] In another technical solution, the method for optimizing the BP neural network includes:
[0095] S411, taking the total drainage volume of the mining area, the water supply volume of the mining area, and the coal mining volume of the mining area as the influencing factors, the total number of parameter types as the number of neuron nodes m, and the drainage volume of the mining area controlled aquifer as the only output value, that is, the number of nodes in the output layer c=1, and the number of nodes in the hidden layer c1 is , a is a random constant between 1 and 10;
[0096] S412, normalize the sample set data, and its mathematical expression is:
[0097]
[0098] in, represents the sample data of the influencing factors, , are the minimum and maximum values in the sample data, respectively. is the influencing factor data after dimensionless processing;
[0099] S413, initializing the mapping relationship between the population particles and the BP neural network weights and thresholds, including particle dimensions, initial velocity, population size, learning factor, and inertia weight;
[0100] S414, input the normalized input variables and output variables into the BP neural network model, calculate the fitness function value of the particle, and obtain the historical optimal fitness and global fitness of the particle. The fitness function value of the particle is the mean square error of the calculation result, and its function expression is:
[0101]
[0102] in, Represented as the predicted value of the i-th sample, is the true value of the i-th sample, and n is the total number of calculation results of the neural network;
[0103] S415, iteratively calculating the particle fitness, and updating the historical optimal fitness and the global fitness according to a preset update condition, until a preset iteration end condition is met;
[0104] S416. Update the weights and thresholds of the preset BP neural network model to obtain an optimized BP neural network model for calculating the drainage amount of the aquifer under control in the current mining area.
[0105] In another technical solution, the method for predicting the drainage volume of a controlled aquifer in a mining area is characterized in that the expected drainage volume of the mining area is obtained by summarizing the expected drainage volumes of each mine in the mining area, and the expected drainage volume of the mine is calculated using the hydrogeological parameters obtained by the pumping test, and the calculation formula is as follows:
[0106] ;
[0107] R
[0108] R0= R +r0
[0109] Where Q is the regional predicted drainage volume, m 3 ;
[0110] K—permeability coefficient;
[0111] S—water level drop, m;
[0112] M—aquifer thickness, m;
[0113] R0—reference influence radius of the prediction area, m;
[0114] r0—the converted radius of the prediction area, m;
[0115] R—influence radius, m.
[0116] The following is a specific on-site implementation example:
[0117] Taking the over-exploited mining area in District H of EEDS City as an example, there are many coal mining companies concentrated in the over-exploited area, and the level of basic hydrogeological research is relatively low. With the development and utilization of coal resources in the region in recent years, the recharge and drainage geogeological conditions in the area have changed significantly. In recent years, it has been found that the over-exploited area has extracted water for production and life of the people and enterprises in the area through drainage. The water level of the Ordovician limestone karst aquifer where the local people draw water has dropped significantly. However, since the over-exploited area plays an important role in coal mining in District H, although many coal mining companies in the mining area have drained the controlled aquifers, they still need to maintain production. The relevant water conservancy management departments need to accurately predict the drainage volume of the controlled aquifers in the over-exploited area in the next management cycle, so as to carry out subsequent groundwater resource management and allocation.
[0118] S1. Obtain the historical hydrological measured data and mining reports of the six coal mining enterprises in the over-exploited area, the coal mining volume and the total drainage water volume of the six coal mining enterprises over the years. Figure 3 As mentioned above, based on the total drainage volume of the mines over the years and the water chemical analysis report of the year, the drainage volume of the controlled aquifers of each mine in the year is calculated. Taking ZF Coal Mine in 2023 as an example, the distribution of the aquicludes in the coal mine of ZF Coal Mine is as follows: Figure 7 As shown in the figure, ZF Coal Mine is an underground coal mine, mainly mining 8 coal, 9 coal and 16 coal. The planned coal mining volume of ZF Coal Mine in 2023 is 1.2 million tons, and the actual coal mining volume is 1.0168 million tons. The total drainage water discharge and mine drainage water volume caused by ZF Coal Mine mining are 952,500 m³. ZF Coal Mine mainly mines 8 coal, 9 coal and 16 coal in 2023. Therefore, the coal-bearing aquifers directly filled with water are the 8 coal roof sandstone aquifer and the 16 coal roof sandstone aquifer. The water sample collection location of ZF Coal Mine is as follows: Figure 8 The water chemistry and isotope analysis results of ZF coal mine are shown in Figure 8~Figure 10 As shown, the result is that the water volume of the controlled aquifer accounts for 15.59% of the drainage volume of the ZF coal mine, and the drainage volume of the controlled aquifer of the ZF coal mine in 2023 is 148,500 m³;
[0119] The planned coal mining volume of ZF Coal Mine in 2024 is 1.2 million tons. The main coal seam mined in 2024 is 16 coal seam, which is a mineable coal seam in the sandstone aquifer of Taiyuan Formation. ZF Coal Mine has conducted drilling tests in Taiyuan Formation through preliminary exploration drilling. The average parameters are K=0.0451 m / d, M=106.05m, =1.12*(375+285) / 4=184.8m, Taiyuan aquifer water level elevation +1036m, 16# coal seam floor elevation is 888.91m, thus the water level drop is determined as:
[0120] S = 1036-888.91 = 147.09 (m);
[0121] ;
[0122] R0= R +r0=497.17(m);
[0123] Substituting the above parameters into the formula , we get the expected drainage volume of ZF Coal Mine in 2024, Q=2859.73m 3 / d=1,032,360 m³ / a.
[0124] S2: Analyze the six enterprises in the over-exploited area one by one in each year and summarize the data. It can be seen that the total drainage volume of the mining area in the over-exploited area in 2023 is 2.6844 million m³, the coal mining volume in the mining area is 3.794 million tons, and the drainage volume of the controlled aquifer in the mining area is 314,800 m³;
[0125] The recharge unit area of the H zone where the overexploitation area is located is divided into soluble rock outcrop recharge area and non-soluble rock cover recharge area. The total recharge area of the calculated area is 271.03 km 2 , of which the recharge area of soluble rock outcrop area is 72.57km 2 The non-soluble rock recharge area is 198.46km 2 According to the geological report, the local soluble rock precipitation infiltration coefficient is 0.16, and the non-soluble rock precipitation infiltration coefficient is 0.052. Taking the 2023 monitoring period unit as an example, the precipitation value of the year with four series of long-term meteorological stations around the H area is used. On this basis, the above meteorological stations are divided into regions using the Thiessen polygon to obtain the annual average rainfall P in each region within the calculation area. The formula Q a =10 -1 ·F·a·P Calculate the periodic regional water capacity of each area within the calculation area and sum them up. It is found that the water recharge volume of area H in 2023 is 8.775 million m³. Considering the proportion of drainage volume and mining volume of over-exploited area in area H, it is obtained that the water recharge volume of over-exploited mining area in 2023 is 2.2045 million m³.
[0126] S3: Summarizing the mining plans of the six coal mining companies in the over-exploited area in 2024, it is learned that the planned coal mining volume of the over-exploited area in 2024 is 3 million tons, the expected drainage volume of the mining area is 2.8965 million m³, and the expected water replenishment volume of the mining area is 2.1045 million m³.
[0127] S4: Since 2010, all mines in the over-exploited area have only started to carry out data-controlled water sampling work on a regular basis, and have formed systematic reports on water chemistry. A total of 30 sets of data on the total amount of mine drainage, mine water replenishment, and mine coal mining from 2010 to 2023 from the over-exploited area and similar mining areas with their own systematic water sampling reports are obtained to establish a training set. A total of 13 sets of data from 2010 to 2022 in the over-exploited area are used as the test set. The total number of parameter types is used as the number of neuron nodes m, that is, the number of nodes in the input layer is m=3. For mining areas with more detailed water level monitoring data and mining data, factors such as the annual increase in the length of the mine tunnel, the change in the water volume of the coal-bearing aquifer in the mining area, and the increase in the old kiln water in the mining area can also be used as additional influencing factors. The drainage volume of the controlled aquifer in the mining area is used as the only output value, that is, the number of nodes in the output layer is c=1, and the number of nodes in the hidden layer is c1. , a is a random constant between 1 and 10. The final value of c1 needs to be tried one by one through the algorithm to obtain a c1 value with a smaller mean square error MSE. In this embodiment, c1=4;
[0128] Normalize the total drainage volume of the mining area, the water replenishment volume of the mining area, the coal mining volume of the mining area, the drainage volume of the controlled aquifer of the mining area, the planned coal mining volume of the mining area, the expected drainage volume of the mining area, and the expected water replenishment volume of the mining area, and optimize the mapping relationship between the population particles and the BP neural network weights and thresholds, including particle dimensions, initial velocity, population size, learning factor, and inertia weight;
[0129] The normalized input variables and output variables are input into the BP neural network model, and the fitness function value of the particle is calculated to obtain the historical optimal fitness and global fitness of the particle. The fitness function value of the particle is the mean square error of the calculation result, and its function expression is:
[0130]
[0131] in, Represented as the predicted value of the i-th sample, is the true value of the i-th sample, n is the total number of calculation results of the neural network, the particle fitness is iteratively calculated, the historical optimal fitness and the global fitness are updated according to the preset update conditions, until the preset iteration end conditions are met, the weights and thresholds of the preset BP neural network model are updated, and the optimized BP neural network model for calculating the current predicted coal production in the mining area is obtained.
[0132] Comparison of forecast results and actual data in overexploited areas from 2010 to 2022 Figure 4 It can be seen that the curve of the constructed neural network model is smooth and has good stability. The MSE value of the model meets the use requirements. Since no water chemical analysis was carried out in the mines in the mining area before 2010, the drainage volume of the controlled aquifer in the mining area can only be obtained by combining the water level of the controlled aquifer in the monitoring well recorded in that year with the water consumption of local enterprises and the masses in that year. The approximate drainage data of the controlled aquifer is obtained, and its value is approximately equal to the predicted value of the drainage volume of the controlled aquifer obtained by this model based on the relevant data of each year before 2010.
[0133] Taking the 2023 forecast data as an example, the planned coal production of the over-exploited area at the beginning of 2023 is 3.8 million tons, the expected drainage volume of the mine area is 2.6 million m³, and the expected water replenishment volume of the mine area is 2.0005 million m³. By inputting these three parameters into the model, it is known that the predicted drainage volume of the controlled aquifer in the mine area in 2023 is 155,800 m³. Compared with the actual measured value of 148,500 m³ of drainage volume of the controlled aquifer in the over-exploited area in 2023, the actual value error is within 10%. Except for the 2023 data, the relevant data formulated at the beginning of the year from 2010 to 2022 are substituted, and the errors between the results and the actual values are all within an acceptable range.
[0134] The total drainage volume of the over-exploited mining area in 2024 obtained in step S1, the expected drainage volume of the mining area, the planned coal mining volume of the mining area, and the expected water replenishment volume of the mining area are input into the calculation model to obtain the predicted value of the drainage volume of the controlled aquifer in the over-exploited mining area in 2024.
[0135] A mining area controlled aquifer drainage prediction device, applied to the above-mentioned mining area controlled aquifer drainage prediction method, comprising: a plurality of mine data acquisition modules, a mining area sample set acquisition module and a controlled aquifer drainage prediction module;
[0136] The mine data acquisition module is arranged in each mine in the mining area, and can obtain the mine drainage volume and coal mining volume of the mine in the cycle, and obtain the drainage volume of the mine-controlled aquifer in the current cycle by means of water chemical analysis. The mine data acquisition module can record the planned mining volume of the mine in the next cycle and obtain the corresponding expected drainage volume of the mine;
[0137] The mining area sample set acquisition module can collect precipitation records from meteorological stations around the mining area over the years, and calculate the mining area water replenishment volume over the years based on the distribution of different rock formations in the mining area. The sample set acquisition module can summarize the drainage data of each mine over the years, thereby obtaining the total drainage volume of the mining area over the years and the drainage volume of the controlled aquifer in the mining area over the years. The sample set acquisition module can obtain the planned coal mining volume of each mine in the mining area and the expected drainage volume of the mine, and obtain the coal mining volume of the mining area over the years, the planned coal mining volume of the mining area in the next cycle, and the expected drainage volume of the mining area;
[0138] The controlled aquifer depletion prediction module is used to optimize the BP neural network model, and the expected depletion volume of the mining area, the expected water replenishment volume of the mining area, and the planned coal mining volume of the mining area are input into the optimized mining area controlled aquifer depletion prediction model to obtain the predicted value of the mining area controlled aquifer depletion volume.
[0139] A computer device, characterized in that it includes a memory and a processor, wherein the memory stores computer instructions, and the processor executes the above-mentioned method for predicting the dewatering amount of aquifers under mining area management by executing the computer instructions. The computer device can be any terminal device including a mobile phone, a laptop computer, a desktop computer, a tablet computer, a PDA (Personal Digital Assistant), a POS (Point of Sales), a vehicle-mounted computer, etc.
[0140] Computer instructions are stored in a readable storage medium, such as a computer floppy disk, USB flash drive, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc., including a number of instructions for enabling a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.
[0141] It should be noted that, although the above describes the various steps in a specific order, it does not mean that the various steps must be performed in the above specific order. In fact, some of these steps can be performed concurrently or even in a different order, as long as the required functions can be achieved. The number of devices and processing scales described here are used to simplify the description of the present invention, and the application, modification and variation of the present invention are obvious to those skilled in the art.
[0142] Although the embodiments of the present invention have been disclosed as above, they are not limited to the applications listed in the specification and the implementation modes, and they can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and the illustrations shown and described herein.
Claims
1. A method for predicting the amount of drainage of aquifers under control in a mining area, characterized in that: The following steps are involved: S1. Obtain historical hydrological measured data and mining reports of the mining area over the years. The historical hydrological measured data of the mining area include the drainage water volume of each mine in the mining area over the years, the hydrochemical analysis report over the years, and the precipitation report of the mining area over the years. The mining report includes the coal mining volume of each mine in the mining area over the years, the tunnel conditions of the mining area, and the mining plan for the next control cycle; S2. Obtain the mine drainage volume, mine control aquifer drainage volume, and coal mining volume of each mine in the mining area over the years, and summarize them into the total mine drainage volume, mine control aquifer drainage volume, and mine coal mining volume over the years, and use the mine area's meteorological data over the years to obtain the mine area's water recharge volume over the years; S3. Using the mining plan of each mine in the mining area in the next control cycle, the planned coal mining volume and the expected drainage volume of the mining area are obtained, and the expected water replenishment volume of the mining area is obtained by using the meteorological data of the mining area; S4. Use the total drainage volume of the mine area, the water replenishment volume of the mine area, the coal mining volume of the mine area, and the drainage volume of the controlled aquifer in the mine area in previous years as sample set data to construct a prediction model for the drainage volume of the controlled aquifer in the mine area, and input the planned coal mining volume of the mine area, the expected drainage volume of the mine area, and the expected water replenishment volume of the mine area to obtain the predicted value of the drainage volume of the controlled aquifer in the mine area, wherein a BP neural network including an input layer, a hidden layer and an output layer is adopted, the input layer of the BP neural network model includes a plurality of input neuron nodes, the hidden layer is provided with a plurality of hidden neuron nodes, and each of the input neuron nodes is respectively connected to each of the hidden neuron nodes. The hidden neuron nodes are all connected with the output layer, and the total drainage volume of the mining area, the water replenishment volume of the mining area, and the coal production volume of the mining area are used as input values of the input neuron nodes. The output result is compared with the corresponding drainage volume of the mining area controlled aquifer, and the BP neural network model is optimized according to the PSO algorithm to obtain a mining area controlled aquifer drainage volume prediction model, obtain the mining area's planned coal production volume, the expected water replenishment volume and the expected drainage volume of the mining area in the next cycle, and input them into the optimized mining area controlled aquifer drainage volume prediction model to obtain the predicted value of the mining area controlled aquifer drainage volume in the next cycle.
2. The method for predicting the amount of drainage of aquifers under control in a mining area according to claim 1, characterized in that: The method of using the meteorological data of the mining area over the years to obtain the mining area water supply over the years includes the following steps: A1: Obtain geological information of the mining area: According to the geological report, the mining area is divided into various types of recharge unit areas, the area of each area is confirmed and the corresponding infiltration coefficient is determined; A2: Determine the precipitation replenishment information of the mining area: Obtain the annual precipitation of the mining area based on the precipitation report records of the meteorological stations around the mining area over the years; A3: Calculate the annual water recharge volume for each geological region within the mining area and summarize it. The calculation formula for a separate geological region is as follows: Q a =10 -1 ·F·A·P Where Q a is the regional recharge water volume, m 3 / a; F—recharge area, m 2 ; A—infiltration coefficient; P—average annual precipitation, mm / a.
3. The method for predicting the amount of drainage of aquifers under control in a mining area according to claim 1, characterized in that: The acquisition of the drainage volume of the mining area controlled aquifer comprises the following steps: B1: Carry out drainage measurement and water sampling of each mine in the mining area: Measure the drainage water of each mine and conduct water sampling regularly, collect water samples of mine drainage water, coal-bearing aquifers and controlled aquifers periodically, and send samples for testing and storage of results; B2: Determine the source of drainage water in each mine: The water source is determined based on the hydrochemical characteristics of the drainage water and the hydrochemical characteristics of each aquifer, and is analyzed using the hydrochemical piper three-line diagram and isotope method. The hydrochemical piper three-line diagram is based on Na + , K + , Ca 2+ Mg 2+ , Cl - 、SO4 2- 、HCO3 - Ions are used as discriminant indicators to obtain the drainage volume of the mine-controlled aquifer in each mine; B3: Summarize the drainage volume of each mine-controlled aquifer in the mining area to obtain the drainage volume of the controlled aquifer in the mining area.
4. The method for predicting the amount of drainage of aquifers under control in a mining area according to claim 1, characterized in that: The method for optimizing the BP neural network comprises: S411, taking the total drainage volume of the mining area, the water supply volume of the mining area, and the coal mining volume of the mining area as the influencing factors, the total number of parameter types as the number of neuron nodes m, and the drainage volume of the mining area controlled aquifer as the only output value, that is, the number of nodes in the output layer c=1, and the number of nodes in the hidden layer c1 is , a is a random constant between 1 and 10; S412, normalize the sample set data, and its mathematical expression is: in, represents the sample data of the influencing factors, are the minimum and maximum values in the sample data, respectively. is the influencing factor data after dimensionless processing; S413, initializing the mapping relationship between the population particles and the BP neural network weights and thresholds, including particle dimensions, initial velocity, population size, learning factor, and inertia weight; S414, input the normalized input variables and output variables into the BP neural network model, calculate the fitness function value of the particle, and obtain the historical optimal fitness and global fitness of the particle. The fitness function value of the particle is the mean square error of the calculation result, and its function expression is: in, Represented as the predicted value of the i-th sample, is the true value of the i-th sample, and n is the total number of calculation results of the neural network; S415, iteratively calculating the particle fitness, and updating the historical optimal fitness and the global fitness according to a preset update condition, until a preset iteration end condition is met; S416. Update the weights and thresholds of the BP neural network model to obtain an optimized BP neural network model for calculating the drainage amount of the controlled aquifer in the mining area.
5. The method for predicting the amount of drainage of aquifers under control in a mining area according to claim 1, characterized in that: The expected drainage volume of the mining area is obtained by summarizing the expected drainage volume of each mine in the mining area. The expected drainage volume of the mine is calculated using the hydrogeological parameters obtained from the pumping test. The calculation formula is as follows: ; R R0= R +r0 Where Q is the regional predicted drainage volume, m 3 ; K—permeability coefficient; S—water level drop, m; M—aquifer thickness, m; R0—reference influence radius of the prediction area, m; r0—the converted radius of the prediction area, m; R—influence radius, m.
6. A device for predicting the amount of drainage from a controlled aquifer in a mining area, applied to the method for predicting the amount of drainage from a controlled aquifer in a mining area according to any one of claims 1 to 5, characterized in that: include: Multiple mine data collection modules, mining area sample set acquisition module and control aquifer dewatering prediction module; The mine data acquisition module is arranged in each mine in the mining area, and can obtain the mine drainage volume and coal mining volume of the mine in the cycle, and obtain the drainage volume of the mine-controlled aquifer in the current cycle by means of water chemical analysis. The mine data acquisition module can record the planned mining volume of the mine in the next cycle and obtain the corresponding expected drainage volume of the mine; The mining area sample set acquisition module can collect precipitation records from meteorological stations around the mining area over the years, and calculate the mining area water replenishment volume over the years based on the distribution of different rock formations in the mining area. The sample set acquisition module can summarize the drainage data of each mine over the years, thereby obtaining the total drainage volume of the mining area over the years and the drainage volume of the controlled aquifer in the mining area over the years. The sample set acquisition module can obtain the planned coal mining volume of each mine in the mining area and the expected drainage volume of the mine, and obtain the coal mining volume of the mining area over the years, the planned coal mining volume of the mining area in the next cycle, and the expected drainage volume of the mining area; The controlled aquifer depletion prediction module is used to optimize the BP neural network model, and the expected depletion volume of the mining area, the expected water replenishment volume of the mining area, and the planned coal mining volume of the mining area are input into the optimized mining area controlled aquifer depletion prediction model to obtain the predicted value of the mining area controlled aquifer depletion volume.
7. A computer device, characterized in that: It comprises a memory and a processor, wherein the memory stores computer instructions, and the processor executes the method for predicting the dewatering amount of aquifers under control in a mining area according to any one of claims 1 to 5 by executing the computer instructions.
Citation Information
Patent Citations
Water networking system
CN105323321A
Mine water inflow prediction method and system based on coal mine big data
CN113742995A
Flooding prediction system and method for environmental facilities
KR102343736B1
