CN106703883B - A kind of personalized method for determining Water Inrush From Working-faces danger classes - Google Patents

Abstract

The invention discloses a method for individually determining the danger level of water inrush from the floor of a coal mining face, belonging to the field of coal safety mining. This method comprehensively considers the factors such as the water pressure of the coal mining face floor, the thickness of the water-resisting layer, the failure phenomenon of the coal seam floor pressure, the lithology combination of the rock strata in the water-resisting section, and the development of the original conduction of groundwater under pressure on the floor, and establishes the relationship between these influencing factors and the outburst. The implicit function dependence of the water inrush hazard level is analyzed through interpolation to establish a comprehensive evaluation index system for evaluating the water inrush hazard level of the floor. This method comprehensively considers various factors that affect the threat of water inrush from the coal mining floor, and establishes the functional dependence relationship between each influencing factor and the water inrush coefficient. The problem of functional dependence of water coefficient provides a new method for calculating water inrush coefficient.

Description

A personalized method for determining the risk level of floor water inrush in coal mining face

technical field

The invention belongs to the technical field of coal mine safety mining, and in particular relates to a method for individually determining the risk level of water inrush from the floor of a coal mining working face.

Background technique

At present, the water inrush coefficient method is mainly used to evaluate the risk of floor water inrush in coal mine safety mining. The formula for calculating the water inrush coefficient T=p/M was proposed by the Hungarian engineer Wegflens on the concept of water-repelling thickness (T=M/p) during the 1964 Jiaozuo Hydrogeological Conference in my country. In the formula, p is the water content of the coal seam floor layer hydraulic pressure, and M is the thickness of the coal seam floor water-resisting layer. After the formula was put forward, based on the floor water inrush cases and basic data in Jiaozuo, Fengfeng, Jingxing, Handan, Feicheng, Zibo and other Dashui mining areas, it was calculated that T≤0.06MPa/m is safe, and 0.06MPa/m<T ≤MPa/m is relatively safe, and 0.1MPa/m<T is dangerous. Water inrush coefficient reflects the most basic law of groundwater seepage in hydrogeology.

In a rather long historical period, the water inrush coefficient played an important role in guiding the safe mining of coal mines. However, as the shallow coal seam resources are exhausted, coal mining gradually shifts to the deep, and the hydraulic pressure of the Austrian ash water gradually increases. In actual mining, the water inrush coefficient is far greater than 0.1MPa/m in the "Regulations on Water Prevention and Control in Coal Mine". In the later period, although some people put forward various calculation formulas for the water inrush coefficient considering the damage degree of the coal seam floor, the lithological combination of the rock formation in the water-resisting section, and other influencing factors, the adjacency coefficient proposed in 1964 still adopts the adjacency value proposed in 1964. Obviously, this is inappropriate.

The current calculation method of water inrush coefficient has the following problems:

1. Factors affecting the risk of water inrush are not fully considered

The calculation formula of water inrush coefficient only considers the water pressure and the thickness of the water-resisting layer, but does not take into account the damage degree of the coal seam floor, the lithology combination of the rock formation in the water-resisting section, the groundwater conduction height of the floor under pressure, and the water-rich index of the aquifer.

2. No machine learning method was used to establish the functional dependence between the risk of water inrush and its influencing factors

The calculation formula of water inrush coefficient in the early stage has few factors to consider, and the calculation formula is simple. There are many factors to be considered in the later calculation formula, and the functional dependence between the water inrush coefficient and the influencing factors cannot be fitted in the case of a small number of samples.

3. The water inrush coefficient only gives the risk of water inrush, but has nothing to do with the amount of water inrush

It can be seen from its calculation formula T=p/M that the current water inrush coefficient only considers the water pressure of the coal seam floor aquifer, the thickness of the coal seam floor water-resisting layer and other factors, and has nothing to do with the actual water inrush. In the actual mining process, if the water-richness of the floor aquifer is poor, even if the water inrush is small, it will not bring great harm to the coal mining face.

Contents of the invention

Aiming at the above-mentioned problems existing in the prior art, the present invention proposes a method for individually determining the risk level of water inrush from the floor of the coal mining face.

The technical solution adopted in the present invention is:

A method for individually determining the risk level of water inrush from the floor of a coal mining face is carried out according to the following steps:

Step 1: Collect the comprehensive data of the coal mining face in the coal mine production process, analyze the correlation between the maximum water inrush volume and various influencing factors, and select the corresponding type of influencing factors as the independent variable affecting the risk level of the floor water inrush;

Step 2: According to the comprehensive drainage capacity and maximum water inrush data of specific mines, individualize the risk level of floor water inrush;

Step 3: Denote the vector combination of independent variables as x and the risk level of water inrush from the floor as y, and establish the implicit functional dependence between the independent variable and the risk level of water inrush from the floor;

Step 4: Substituting the vector combination x of the newly collected independent variables into the implicit functional dependency established in step 3 to calculate the floor water inrush risk level y, and then determine the floor water inrush risk level according to step 2.

The above method also includes the following steps:

Step 5: Use the implicit functional dependence relationship established in Step 3 to subdivide and interpolate the respective variables. Within the scope of the water inrush hazard level of each floor, perform reverse calculation to obtain the maximum and minimum values of the respective variables, and establish the relationship between the respective variables and A comprehensive evaluation index system for the risk level of floor water inrush;

Step 6: According to the comprehensive evaluation index system of the respective variables and the risk level of floor water inrush established in step 5, the on-site staff can determine the risk level of floor water inrush by querying the comprehensive evaluation index system for the newly collected independent variable data.

Preferably, in step 1: the comprehensive data of the coal mining face include the maximum water inrush volume, water inrush coefficient, damage degree of coal seam floor pressure, rock formation lithology combination of water-resisting section, groundwater conduction height of floor under pressure and aquifer richness. Water index, etc.; analyze the correlation between the maximum water inrush volume and various influencing factors such as water inrush coefficient, coal seam floor pressure damage degree, rock formation lithology combination of water-resisting section, groundwater conduction height of floor under pressure, and water-rich index of aquifer , and according to the size of the correlation, the corresponding types of influencing factors such as the damage degree of the coal seam floor pressure, the height of groundwater under pressure on the floor, etc. are selected as independent variables that affect the risk level of floor water inrush.

Preferably, in step 2: the floor water inrush hazard level is divided into three levels: safe, relatively dangerous, and dangerous; the maximum water inrush volume is less than or equal to two-thirds of the comprehensive drainage capacity of the specific mine, which is defined as safety level 1; the maximum water inrush volume is greater than the specific mine. Two-thirds of the mine's comprehensive drainage capacity and less than or equal to the specific mine's comprehensive drainage capacity are defined as relatively dangerous level 2; the maximum water inrush is greater than the specific mine's comprehensive drainage capacity is defined as dangerous level 3.

Preferably, in Step 3: According to the specific mine, n observation samples (x ₁ , y ₁ ), (x ₂ , y ₂ )...(x _n , y _n ) are known in several functions {f(x,ω) } to find an optimal function f(x,ω ₀ ), and estimate the unknown dependencies, so that the formula R(ω)=∫L(y,f(x,ω))dF(x,y) shows Expect minimal risk.

In the third step above, the implicit functional dependence between the established independent variable and the risk level of water inrush on the floor is realized with the help of machine learning theory, specifically:

(1) {f(x,ω)} is the prediction function set, ω is called the generalized parameter, and L(y,f(x,ω)) is the loss function;

(2) The loss function adopts L(y,f(x,ω))=(yf(x,ω)) ² ;

(3) The information that can be used in the training process is only sample data, so the expected risk

R(ω)＝∫L(y,f(x,ω))dF(x,y) cannot be calculated, use estimate it;

(4) In the calculation process, the risk level predicted by regression training is a continuous value, which is rounded to an integer.

Preferably, in step 4: the vector combination x of the independent variables is actually obtained during the construction exploration and coal mining; for the section without physical exploration, the interpolation prediction is performed through the three-dimensional mine digital model, and the prediction is performed first and then adjusted.

Preferably, in step five, specifically:

(1) Analyze the maximum and minimum values of each component of the vector combination x of the independent variable;

(2) Take one percent of the difference between the maximum value minus the minimum value as the unit, and subdivide and interpolate;

(3) Obtain the water inrush hazard level of the bottom plate at each interpolation point, and the hazard level at this time will not be rounded;

(4) When the water inrush risk level is in the interval [0, 1.5), [1.5, 2.5), [2.5, 3.5), the maximum and minimum values of the respective variables are established, and the synthesis of the respective variables and the floor water inrush risk level is established Evaluation System.

The beneficial technical effect of the present invention is:

Compared with the prior art, the present invention comprehensively considers the specific mine drainage capacity and various factors affecting the water inrush volume, establishes the implicit functional dependence relationship between the water inrush risk rating and the influencing factors, and adopts parameter interval interpolation to determine the index framework, which solves the problem of Technical problems such as small sample modeling of water inrush data and difficulties in the application of implicit functions make the evaluation of water inrush hazard levels more scientific and reasonable, which is in line with the actual mine production.

Description of drawings

Fig. 1 is a flow chart of a method for individually determining the risk level of water inrush from the floor of a coal mining face according to the present invention.

Detailed ways

The invention provides a method for individually determining the risk level of water inrush from the floor of a coal mining face. This method comprehensively considers the factors such as the water pressure of the coal mining face floor, the thickness of the water-resisting layer, the damage phenomenon of the coal seam floor pressure, the lithology combination of the rock strata in the water-resisting section, and the development of the original conduction of groundwater under pressure on the floor, and establishes the relationship between these influencing factors and the outburst. The implicit function dependence of the water inrush hazard level is analyzed through interpolation to establish a comprehensive evaluation index system for evaluating the water inrush hazard level of the floor. This method fully considers various factors that affect the threat of water inrush from the coal mining floor, and establishes the functional dependence relationship between each influencing factor and the water inrush coefficient, which solves the problem that the traditional method only considers the water pressure and the thickness of the water-resisting layer or cannot establish the relationship between the influencing factors and the water inrush. The difficult problem of coefficient function dependence provides a new method for the calculation of water inrush coefficient.

Below in conjunction with accompanying drawing and specific embodiment the present invention is described in further detail:

As shown in Figure 1, a method for individually determining the risk level of water inrush from the floor of a coal mining face follows the steps below:

Step 1: Collect the maximum water inrush volume of the coal mining face during the coal mine production process, the water inrush coefficient, the damage degree of the coal seam floor pressure, the lithology combination of the rock strata in the water-resisting section, the groundwater conduction height of the floor under pressure, and the water-rich index of the aquifer, etc. According to the data, the correlation between the maximum water inrush and other influencing factors such as the water inrush coefficient and the damage degree of coal seam floor pressure is analyzed, and the factors with higher correlation are selected as the independent variables affecting the maximum water inrush.

Step 2: According to the comprehensive drainage capacity and maximum water inrush data of specific mines, the floor water inrush risk level of each group of data is individually divided. Generally, it is divided into three levels: safe, relatively dangerous, and dangerous. The maximum water inrush is less than or equal to two-thirds of the comprehensive drainage capacity of the specific mine as safety level 1; If the water volume is greater than or equal to the comprehensive drainage capacity of the specific mine, it is classified as hazard level 3.

Step 3: Denote the vector combination of independent variables as x, and denote the water inrush hazard level of the floor as y. According to the specific mine known n observation samples (x ₁ ,y ₁ ),(x ₂ ,y ₂ )...(x _n ,y _n ) find an optimal function among several functions {f(x,ω)} f(x,ω ₀ ), estimate the unknown dependencies, and minimize the expected risk shown by the formula R(ω)=∫L(y,f(x,ω))dF(x,y).

The functional dependence between each influencing factor established in step 3 and the risk level of floor water inrush is an implicit functional dependence, which can be realized with the help of machine learning theories such as support vector machine theory.

specifically:

(1) {f(x,ω)} is the prediction function set, ω is called the generalized parameter, and L(y,f(x,ω)) is the loss function.

(2) The loss function adopts L(y,f(x,ω))=(yf(x,ω)) ² .

(3) The information that can be used in the training process is only sample data, so the expected risk R(ω)=∫L(y,f(x,ω))dF(x,y) cannot be calculated, using Estimate it.

(4) In the calculation process, the risk level predicted by regression training is a continuous value, which is rounded to an integer.

Step 4: According to the implicit function dependency relationship established in Step 3, the data vector of the newly collected influencing factors, that is, the vector combination x of independent variables, can be substituted into the implicit function to calculate its floor water inrush risk level. The factors affecting the risk level of floor water inrush can be actually obtained in the process of construction exploration and coal mining. For the sections that have not been physically explored, interpolation prediction can be performed through the 3D mine digital model, and the prediction is made first and then adjusted.

Step 5: In order to facilitate the actual application on site, use the implicit function dependence relationship established in Step 3 to subdivide and interpolate the respective variables, and perform reverse calculation to obtain the maximum and minimum values of the respective variables within the range of the water inrush hazard level of each floor value, and establish a comprehensive evaluation index system for the respective variables and the risk level of floor water inrush.

specifically:

(1) Analyze the maximum and minimum values of each component of the factors affecting the risk level of floor water inrush (the vector combination x of independent variables).

(2) Subdividing and interpolating in units of one percent of the difference between the maximum value minus the minimum value.

(3) Obtain the water inrush hazard level of the bottom plate at each interpolation point, and the hazard level at this time will not be rounded.

(4) When the water inrush risk level is in the interval [0, 1.5), [1.5, 2.5), [2.5, 3.5), the maximum and minimum values of the respective variables are established, and the synthesis of the respective variables and the floor water inrush risk level is established Evaluation System.

Step 6: According to the comprehensive evaluation index system of the respective variables and the risk level of floor water inrush established in step 5, the on-site staff can determine the risk level of floor water inrush by querying the comprehensive evaluation index system for the newly collected data of influencing factors.

The present invention proposes that the key to the risk of water inrush in step one is the water inrush volume, not just the ratio of the water pressure on the bottom plate to the thickness of the water-resisting layer. And in the first step, factors such as the water inrush coefficient affecting the water inrush volume, the damage degree of the coal seam floor pressure, the lithology combination of the rock formation in the water-resisting section, the groundwater conduction height of the floor under pressure, and the water-rich index of the aquifer are considered comprehensively.

In the second step, according to the specific mine drainage capacity, the water inrush risk level of the coal mining face is divided individually, rather than all mine water inrush risk levels are uniformly divided.

Under the condition that the explicit functional dependence relationship between the influencing factors of water inrush and the maximum water inrush volume cannot be given, the implicit functional dependence relationship is used to describe the functional dependence relationship between the maximum water inrush volume and each influencing factor in step 3. The functional dependence relationship requires: For data samples (x ₁ ,y ₁ ),(x ₂ ,y ₂ ),(x ₃ ,y ₃ )...(x _n ,y _n ), find one of several functions {f(x,ω)} The optimal function f(x,ω ₀ ) estimates the unknown dependencies so that the expected risk shown in the following formula is the smallest.

R(ω)＝∫L(y,f(x,ω))dF(x,y)

In step 5, subdivide and interpolate each influencing factor, within the scope of each floor water inrush risk level, reverse calculation to obtain the maximum and minimum values of the respective variables, and establish a comprehensive evaluation index for each variable and the floor water inrush risk level system.

The present invention proposes a method for individually determining the risk level of water inrush on the floor of a coal mining face. Compared with the prior art, the present invention comprehensively considers the specific mine drainage capacity and various factors that affect the amount of water inrush, and establishes the water inrush risk level. The implicit functional dependency between risk rating and influencing factors, and the use of parameter interval interpolation to determine the index framework solve technical problems such as small sample modeling of water inrush data and difficulties in the application of implicit functions, making the evaluation of water inrush risk levels more scientific and reasonable. In line with the actual production of mines.

Parts not mentioned in the above methods can be realized by adopting or referring to existing technologies.

The above descriptions are not intended to limit the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by those skilled in the art within the scope of the present invention shall also belong to the scope of the present invention. protected range.

Claims (7)

1. A method for determining the water inrush risk level of the floor of the coal mining face individuated, characterized in that it is carried out in the following steps: Step 1: Collect the comprehensive data of the coal mining face in the coal mine production process, analyze the correlation between the maximum water inrush volume and various influencing factors, and select the corresponding type of influencing factors as the independent variable affecting the risk level of the floor water inrush; Step 2: According to the comprehensive drainage capacity and maximum water inrush data of specific mines, individualize the risk level of floor water inrush; Step 3: Denote the vector combination of independent variables as x and the risk level of water inrush from the floor as y, and establish the implicit functional dependence between the independent variable and the risk level of water inrush from the floor; Step 4: Substituting the vector combination x of the newly collected independent variables into the implicit functional dependency established in step 3, calculating the floor water inrush risk level y, and then determining the floor water inrush risk level according to step 2; Step 5: Use the implicit functional dependence relationship established in Step 3 to subdivide and interpolate the respective variables. Within the scope of the water inrush hazard level of each floor, perform reverse calculation to obtain the maximum and minimum values of the respective variables, and establish the relationship between the respective variables and A comprehensive evaluation index system for the risk level of floor water inrush; Step 6: According to the comprehensive evaluation index system of the respective variables and the risk level of floor water inrush established in step 5, the on-site staff can determine the risk level of floor water inrush by querying the comprehensive evaluation index system for the newly collected independent variable data. 2. A method for determining the risk level of water inrush from the floor of the coal mining face according to claim 1, characterized in that, in step 1: the comprehensive data of the coal mining face includes the maximum water inrush volume and the water inrush coefficient , damage degree of coal seam floor pressure, lithological combination of rock strata in the water-resisting section, height of groundwater under pressure on the floor, and water-rich index of the aquifer; analyze the relationship between the maximum water inrush and the water inrush coefficient, the degree of rock pressure damage of the coal seam floor, and the water-resisting section The correlation between the lithological combination of rock formations, the groundwater conduction height of the floor under pressure, and the water-richness index of the aquifer, and the corresponding types of influencing factors are selected as the independent variables affecting the risk level of floor water inrush according to the magnitude of the correlation. 3. A method for individually determining the risk level of water inrush from the floor of the coal mining face according to claim 1, characterized in that in step 2: the risk level of water inrush from the floor is divided into three levels: safe, relatively dangerous, and dangerous ; The maximum water inrush is less than or equal to two-thirds of the comprehensive drainage capacity of the specific mine as safety level 1; the maximum water inrush is greater than two-thirds of the comprehensive drainage capacity of the specific mine and less than or equal to the comprehensive drainage capacity of the specific mine is defined as the more dangerous level 2 ; The maximum water inrush is greater than the comprehensive drainage capacity of the specific mine, which is defined as hazard level 3. 4. A method for determining the risk level of water inrush from the floor of the coal mining face according to claim 1, characterized in that, in step 3: n observation samples (x ₁ , y ₁ ) are known according to specific mines ,(x ₂ ,y ₂ )……(x _n ,y _n ) Find an optimal function f(x,ω ₀ ) among several functions {f(x,ω)}, and estimate the unknown dependencies, so that The expected risk shown by the formula R(ω)=∫L(y,f(x,ω))dF(x,y) is the smallest. 5. A method for determining the risk level of water inrush from the floor of the coal mining face according to claim 4, characterized in that: the implicit functional dependence between the established independent variable and the risk level of water inrush from the floor, with the help of machine learning Theoretical implementation, specifically: (1) {f(x,ω)} is the prediction function set, ω is called the generalized parameter, and L(y,f(x,ω)) is the loss function; (2) The loss function adopts L(y,f(x,ω))=(yf(x,ω)) ² ; (3) The information that can be used in the training process is only sample data, so the expected risk R(ω)=∫L(y,f(x,ω))dF(x,y) cannot be calculated, using estimate it; (4) In the calculation process, the risk level predicted by regression training is a continuous value, which is rounded to an integer. 6. A kind of method for determining the risk level of water inrush from the bottom plate of the coal mining face according to claim 1, characterized in that, in step 4: the vector combination x of the independent variable is actually used in construction exploration and coal mining work process Acquisition; For the sections that have not been physically explored, interpolation prediction is performed through the 3D mine digital model, and the prediction is made first and then adjusted. 7. A kind of method for determining the risk level of water inrush from the floor of the coal mining face according to claim 1, characterized in that in step five, specifically: (1) Analyze the maximum and minimum values of each component of the vector combination x of the independent variable; (2) Take one percent of the difference between the maximum value minus the minimum value as the unit, and subdivide and interpolate; (3) Obtain the water inrush hazard level of the bottom plate at each interpolation point, and the hazard level at this time will not be rounded; (4) When the water inrush risk level is in the interval [0, 1.5), [1.5, 2.5), [2.5, 3.5), the maximum and minimum values of the respective variables are established, and the synthesis of the respective variables and the floor water inrush risk level is established Evaluation System.