Fully mechanized coal mining face equipment sensing data and geological exploration data fusion method

Abstract

The invention discloses a method for fusing equipment sensing data and geological exploration data of a fully mechanized coal mining face, and the method specifically comprises the steps: 1, carrying out the coordinate translation conversion of a currently established three-dimensional coal seam geological model, and building a geological data matrix with an origin coordinate as a starting point; 2, fully mechanized coal mining face equipment sensing data are obtained; step 3, establishing a coal machine real-time two-dimensional mining model; 4, calculating the current propulsion degree z of the coal mining machine; step 5, forming mined coal seam model data according to the current propulsion degree z of the coal mining machine and the model of the fully mechanized coal mining face equipment sensing data; and step 6, performing data fusion on the collected coal seam model data and the geologic model subjected to translation conversion through a wavelet transform Kalman filtering algorithm to form complete geologic model data. The method solves the problems that in the prior art, the cost of establishing a model by means of directional drilling and drilling geophysical prospecting is high, and the number of drillable holes is limited.

Description

Technical Field

[0001] The present invention belongs to the technical field of data processing methods for coal mining faces, and relates to a method for fusing the perception data of fully-mechanized coal mining face equipment and geological exploration data. Background Art

[0002] One of the key points in the construction of coal mine intelligentization is the intelligentization of coal mining faces. At present, the industry has proposed to generate a future planned cutting slice model of coal seams based on a three-dimensional geological model to achieve intelligent coal mining with fewer or no workers. However, the technology based on geological modeling mainly comprehensively utilizes technical means such as three-dimensional seismic exploration, mine geophysical prospecting, directional drilling, and borehole geophysical prospecting to obtain information such as the depth and thickness of coal and rock layers, geological structures, and hydrology, and generates a "geological similar shape" of the coal seam entity to establish an initial model of the working face; using the roof, floor, and coal thickness information revealed by the driving roadway as constraints, geologically dynamically interpret the three-dimensional seismic data, and at the same time, in the underground, combine data such as trough waves and crosshole detection to finely correct the working face to form a static geological model of the working face. Due to the high cost of exploration and the limitation of the number of drillable holes, the accuracy of the established geological model is not high, and its applicability for guiding mining needs to be improved. Summary of the Invention

[0003] The purpose of the present invention is to provide a method for fusing the perception data of fully-mechanized coal mining face equipment and geological exploration data, which solves the problems of high cost of establishing a model relying on directional drilling and borehole geophysical prospecting and limited number of drillable holes in the prior art.

[0004] The technical solution adopted by the present invention is a method for fusing the perception data of fully-mechanized coal mining face equipment and geological exploration data, which is specifically implemented according to the following steps:

[0005] Step 1: Perform coordinate translation transformation on the currently established three-dimensional geological model of the coal seam to establish a matrix of geological data with the origin coordinates as the starting point;

[0006] Step 2: Obtain the perception data of fully-mechanized coal mining face equipment and perform data processing;

[0007] Step 3: Establish a real-time two-dimensional mining model of the coal shearer according to the perception data of fully-mechanized coal mining face equipment processed in Step 2;

[0008] Step 4: Calculate the current advancement z of the shearer;

[0009] Step 5: Form the mined coal seam model data according to the current advancement z of the shearer and the model of the perception data of fully-mechanized coal mining face equipment;

[0010] Step 6: Perform data fusion on the mined coal seam model data and the geologically modeled data after translation transformation through the wavelet transform Kalman filter algorithm to form complete geological model data.

[0011] Preferably, step 1 is specifically implemented according to the following steps:

[0012] Step 1.1: Define the working face direction as X, the height of the shearer roof as YH, the height of the shearer floor as YL, and the advancing direction of the shearer as Z. Then the two coordinate matrices of the roof and floor of the geological body are respectively:

[0013]

[0014] Among them, gH is the coordinate matrix of the roof model, gL is the coordinate matrix of the floor model, and n represents the rock stratum number; X n represents the coordinate of the working face direction at the nth rock stratum, YH n is the height of the shearer roof at the nth rock stratum, YL n is the height of the shearer floor at the nth rock stratum, and Z n is the coordinate of the advancing direction of the shearer at the nth rock stratum;

[0015] Step 1.2: Define the origin coordinate as (0, 0, 0). Perform a translation transformation of the coordinate matrix gH of the geological body into gH′ with the row vector (X1, YH1, Z1), and perform a translation transformation of the coordinate matrix gL of the geological body into gL′ with the row vector (X1, YL1, Z1). Specifically:

[0016] gH′ = gH * (X1, YH1, Z1);

[0017] gL′ = gL * (X1, YL1, Z1);

[0018] Among them, gH′ and gL′ are respectively the matrices of geological data with the origin coordinate as the starting point after the coordinate matrices gH and gL are translated.

[0019] Preferably, step 2 is specifically:

[0020] The fully-mechanized mining face obtains the relative position coordinates of the shearer, the cutting height and undercutting coordinates of the shearer, and the body attitude information of the shearer, and then performs data processing.

[0021] Preferably, the data processing in step 2 includes performing data deduplication, interpolation method processing, outlier processing, and data standardization processing on the relative position coordinate data of the shearer, the cutting height and undercutting coordinate data of the shearer, and the body attitude information data of the shearer in sequence.

[0022] Preferably, step 3 is specifically:

[0023] Define the relative position coordinate matrices of the roof and floor relative to the shearer after data processing as zH and zL. With the working face direction reference X as the accuracy of 1 cm, record the current mining height and bottom value of the shearer at each position in the reference X direction:

[0024]

[0025] And convert it through the coal mining machine's fuselage posture information and fuselage length ml:

[0026] YH1′=YH1+ml*sinθ, YH2′=YH2+ml*sinθ,...,YH n ′=YH n +ml*sinθ,

[0027] YL1′=YL1+ml*sinθ, YL2′=YL2+ml·sinθ,...,YL n ′=YL n +ml*sinθ,

[0028] Get the real-time two-dimensional mining model of the coal machine:

[0029]

[0030] Among them, θ is the pitch angle in the fuselage attitude information, and zH′ and zL′ are the coordinate matrices after the translation transformation of zH and zL.

[0031] Preferably, step 4 is specifically:

[0032] Calculate the current number of cutting blades d of the coal mining machine, specifically:

[0033] Assume that the coal mining machine passes through k sample points A2, A3, ..., A k , the number of points in each sample is L, namely a1, a2, a3, ..., a L ,but:

[0034]

[0035] Define the cutting depth m of the coal mining machine, then the current advancement of the coal mining machine z = d*m.

[0036] Preferably, the coal seam model data mined in step 5 is:

[0037]

[0038] Among them, cH is the coordinate matrix of the mined coal seam roof model, and cL is the coordinate matrix of the mined coal seam floor model.

[0039] Preferably, step 6 is specifically:

[0040] The geological model matrix and equipment-aware model matrix The fused data is obtained through wavelet transform Kalman filter algorithm, which is the complete geological model data.

[0041] Preferably, when performing data fusion in step 6, use w i Represents the i-th data x i The weight in the fusion process, due to w i The value reflects the effect of other data on the i-th data x i Comprehensive trust process

[0042] Degrees, you can use w i x i Perform weighted summation to obtain the expression of data fusion:

[0043]

[0044] Among them, the weight coefficient satisfies

[0045] w i Should be integrated with a i In the trust system of each subsystem x1, x2, ..., x n All the information of , so we need to find a set of non-negative numbers a1, a2, ..., a n , so that w i =a1x1+a2x2+...a n x n , i=1,2,...,n

[0046] Rewritten as matrix form W = XA, where W = [w1,w2,…,w n ] T ,A=[a1,a2,…,a n ] T , in the matrix cH, there exists a maximum modulus eigenvalue λ>0, so that λA=cHA, find λ and the corresponding eigenvector A, satisfying a i ≥0, then W=λA;

[0047]

[0048] to w i Perform normalization and obtain

[0049] The final result of the fusion estimation of the data measured by all sensors is

[0050] The final result of all data fusion estimations is obtained as

[0051] Similarly

[0052] The beneficial effects of the present invention are as follows:

[0053] The present invention utilizes the perception data of the fully mechanized coal mining face equipment and the geological exploration data to obtain static data (geological model after translation transformation) and dynamic data (coal seam model data of the mined coal seam), and fuses them to obtain complete geological model data. There is no need for directional drilling and borehole geophysical exploration, which saves costs and is not limited by the number of boreholes, and complete geological model data can be obtained. Specific implementation manners

[0054] The present invention will be described in detail below in conjunction with specific implementation manners.

[0055] Example 1

[0056] The method for fusing the perception data of the fully mechanized coal mining face equipment and the geological exploration data of the present invention is specifically implemented according to the following steps:

[0057] Step 1: Perform coordinate translation transformation on the currently established geological model of the three-dimensional coal seam to establish a matrix of geological data with the origin coordinates as the starting point;

[0058] Step 2: Obtain the perception data of the fully mechanized coal mining face equipment and perform data processing;

[0059] Step 3: Establish a real-time two-dimensional mining model of the coal shearer according to the perception data of the fully mechanized coal mining face equipment processed in Step 2;

[0060] Step 4: Calculate the current advancement z of the coal shearer;

[0061] Step 5: Form the mined coal seam model data according to the current advancement z of the coal shearer and the model of the perception data of the fully mechanized coal mining face equipment;

[0062] Step 6: Perform data fusion on the mined coal seam model data and the geological model after translation transformation through the wavelet transform Kalman filter algorithm to form complete geological model data.

[0063] Example 2

[0064] On the basis of Example 1, Step 1 is specifically implemented according to the following steps:

[0065] Step 1.1: Define the working face direction as X, the roof height of the coal shearer as YH, the floor height of the coal shearer as YL, and the advancement direction of the coal shearer as Z. Then the two coordinate matrices of the roof and floor of the geological body are respectively:

[0066]

[0067] Among them, gH is the coordinate matrix of the roof model, gL is the coordinate matrix of the floor model, and n represents the rock stratum number; X n represents the coordinate in the working face direction at the nth rock stratum, YH n is the height of the shearer roof at the nth rock stratum, YL n is the height of the shearer floor at the nth rock stratum, Z n is the coordinate in the shearer advancing direction at the nth rock stratum;

[0068] Step 1.2: Define the origin coordinates as (0, 0, 0), perform a translation transformation of the coordinate matrix gH of the geological body into a row vector (X1, YH1, Z1) to obtain gH′, and perform a translation transformation of the coordinate matrix gL of the geological body into a row vector (X1, YL1, Z1) to obtain gL′. Specifically:

[0069] gH′ = gH * (X1, YH1, Z1);

[0070] gL′ = gL * (X1, YL1, Z1);

[0071] Among them, gH′ and gL′ are respectively the matrices of geological data with the starting point at the origin coordinates after the translation of the coordinate matrix gH and the coordinate matrix gL.

[0072] Example 3

[0073] Based on Example 2, Step 2 is specifically as follows:

[0074] The fully mechanized coal mining face obtains the relative position coordinates of the shearer, the cutting height and undercutting coordinates of the shearer, and the body attitude information of the shearer, and then performs data processing.

[0075] The data processing in Step 2 includes performing data deduplication, interpolation method processing, outlier processing, and data standardization processing on the relative position coordinate data of the shearer, the cutting height and undercutting coordinate data of the shearer, and the body attitude information data of the shearer in sequence.

[0076] Example 4

[0077] Based on Example 3, Step 3 is specifically as follows:

[0078] Define the relative position coordinate matrices of the roof and floor after data processing with respect to the shearer as zH and zL, and record the current cutting height value and undercutting value of the shearer at each position in the reference X direction of the working face with an accuracy of 1 cm:

[0079]

[0080] And it is converted by the body attitude information and the body length ml of the shearer:

[0081] YH1′ = YH1 + ml * sinθ, YH2′ = YH2 + ml * sinθ,... YH n ′ = YH n + ml * sinθ,

[0082] YL1′ = YL1 + ml * sinθ, YL2′ = YL2 + ml * sinθ,... YL n ′ = YL n + ml * sinθ,

[0083] Obtain the real-time two-dimensional mining model of the coal shearer:

[0084]

[0085] Where θ is the pitch angle in the body attitude information, and zH′ and zL′ are the coordinate matrices after translational transformation of zH and zL.

[0086] Example 5

[0087] On the basis of Example 4, step 4 is specifically: Calculate the current number of cutting passes d of the shearer, specifically:

[0088] Assume that the cutting path of the shearer passes through k sample points A2, A3,..., A k , and the number of points of each sample is L, which are a1, a2, a3,..., a L , then:

[0089]

[0090] Define the cutting depth m of the shearer, then the current advancement z of the shearer = d * m.

[0091] Preferably, the mined coal seam model data in step 5 is:

[0092]

[0093] Where cH is the coordinate matrix of the roof model of the mined coal seam, and cL is the coordinate matrix of the floor model of the mined coal seam.

[0094] Example 6

[0095] On the basis of Example 5, step 6 is specifically: Perform fusion on the geological model matrix and the mined model matrix sensed by the equipment through the wavelet transform Kalman filtering algorithm to obtain the fused data, that is, the complete geological model data.

[0096] When performing data fusion in step 6, use w i to represent the weight of the i-th data x i in the fusion process. Since the value of w i reflects the comprehensive trust degree of other data in the i-th data x i , we can use w i to perform weighted summation on x i to obtain the expression for data fusion:

[0097]

[0098] where the weight coefficients satisfy

[0099] w i should comprehensively consider all information of each subsystem x1, x2,..., x i in a trust degree system regarding x n . Therefore, a set of non-negative numbers a1, a2,..., a n needs to be obtained such that w i = a1x1 + a2x2 + … a n x n , i = 1, 2, …, n

[0100] Rewrite it in matrix form W = XA. Here, W = [w1, w2, …, w n T , A = [a1, a2, …, a n T . In the matrix cH, there exists a maximum modulus eigenvalue λ > 0 such that λA = cHA. Calculate λ and the corresponding eigenvector A, satisfying a i ≥0, then W = λA;

[0101]

[0102] Normalize w i to obtain

[0103] The final result of the data fusion estimation for all sensor measurements is

[0104] The final result of the data fusion estimation for all data is

[0105] Similarly

[0106] ​​During the entire life cycle of working face production, this invention uses a data acquisition service platform to obtain and process all dynamic data such as the position coordinates, attitude data of the shearer in space-time, the undulation changes of the coal seam floor obtained by augmented reality technology, and the actual measured profile of the cutting eye. It studies the geological information extraction algorithm and the dynamic construction technology of the coal seam model, and through multi-source data fusion technology, dynamically optimizes the three-dimensional geological model to form the real-time sharing and dynamic feedback of multi-source heterogeneous information such as geological models, unmined models, and mined models.

[0107] After preprocessing the data such as the attitude, spatial position information, and roof and floor conditions of the coal shearer and cleaning abnormal data, an established mined model based on equipment perception is established, and the unified coordinate system of the three-dimensional geological model established by the geological system is completed. Through feature extraction, a fusion attribute based on feature vectors is formed, and it is processed by a function connection type network fusion algorithm to finally realize the fusion establishment of the working face geological model.

Claims

1. A method for fusing the perception data of the fully-mechanized mining face equipment and the geological exploration data, characterized in that The specific implementation steps are as follows: Step 1: Perform coordinate translation transformation on the currently established three-dimensional coal seam geological model to establish a matrix of geological data with the origin coordinate as the starting point; Step 2: Acquire the sensing data of the fully mechanized mining face equipment and process the data; Step 3: Establish a real-time two-dimensional mining model of the coal mining machine based on the fully mechanized mining face equipment perception data processed in step 2; Step 4, calculate the current advancement z of the coal mining machine; Step 5: Generate mined coal seam model data based on the current advancement z of the shearer and the model of the fully mechanized mining face equipment perception data; Step 6: The mined coal seam model data and the translated geological model are fused by wavelet transform Kalman filter algorithm to form complete geological model data.

2. The fusion method of the perception data of the fully-mechanized mining face equipment and the geological exploration data according to claim 1, wherein The step 1 is specifically implemented as follows: In step 1.1, define the working face direction as X, the shearer roof height as YH, the shearer floor height as YL, and the shearer advancement direction as Z. The coordinate matrices of the top and bottom plates of the geological body are: Among them, gH is the coordinate matrix of the roof model, gL is the coordinate matrix of the floor model, and n represents the rock stratum number; X n represents the coordinate in the working face direction at the nth rock stratum, YH n is the height of the shearer roof at the nth rock stratum, YL n is the height of the shearer floor at the nth rock stratum, Z n is the coordinate in the shearer advancement direction at the nth rock stratum; Step 1.2: Define the origin coordinates as (0, 0, 0), perform a translation transformation gH′ on the coordinate matrix gH of the geological body by the row vector (X1, YH1, Z1), and perform a translation transformation gL′ on the coordinate matrix gL of the geological body by the row vector (X1, YL1, Z1). Specifically, gH′=gH*(X1,YH1,Z1); gL′=gL*(X1,YL1,Z1); Among them, gH′ and gL′ are coordinate matrices gH and gL, which are transformed into matrices of geological data with the origin coordinates after translation.

3. The fusion method of the perception data of the fully-mechanized mining face equipment and the geological exploration data according to claim 2, characterized in that The step 2 is specifically as follows: The fully mechanized mining face obtains the relative position coordinates of the coal mining machine, the mining height and bottom coordinates of the coal mining machine, and the body posture information of the coal mining machine, and then performs data processing.

4. The fusion method of the perception data of the fully-mechanized mining face equipment and the geological exploration data according to claim 3, wherein, The data processing in step 2 includes deduplication, interpolation, outlier processing, and data standardization of the relative position coordinate data of the coal mining machine, the high and low mining coordinate data of the coal mining machine, and the body posture information data of the coal mining machine.

5. The fusion method of the perception data of the fully mechanized coal mining face equipment and the geological exploration data according to claim 4, characterized in that The step 3 is specifically as follows: Define the relative position coordinate matrices of the roof and floor relative to the shearer after data processing as zH and zL. With the working face direction reference X as the accuracy of 1 cm, record the current mining height and bottom value of the shearer at each position in the reference X direction: And convert it through the coal mining machine's fuselage posture information and fuselage length ml: YH1′ = YH1 + ml * sinθ, YH2′ = YH2 + ml * sinθ,..., YH n ′ = YH n + ml * sinθ, YL1′ = YL1 + ml * sinθ, YL2′ = YL2 + ml * sinθ,..., YL n ′ = YL n + ml * sinθ, Get the real-time two-dimensional mining model of the coal machine: Among them, θ is the pitch angle in the fuselage attitude information, and zH′ and zL′ are the coordinate matrices after the translation transformation of zH and zL.

6. The method for fusing the perception data of the fully-mechanized mining face equipment and the geological exploration data according to claim 5, wherein The step 4 is specifically as follows: Calculate the current number of cutting blades d of the coal mining machine, specifically: Suppose the cutting path of a shearer passes through k sample points A2, A3, …, A k , and the number of points in each sample is L, which are a1, a2, a3, …, a L , then: Define the cutting depth m of the coal mining machine, then the current advancement of the coal mining machine z = d*m.

7. The method for fusing the perception data of the fully-mechanized mining face equipment and the geological exploration data according to claim 6, wherein The coal seam model data mined in step 5 is: Among them, cH is the coordinate matrix of the mined coal seam roof model, and cL is the coordinate matrix of the mined coal seam floor model.

8. The fusion method of the perception data of the fully-mechanized mining face equipment and the geological exploration data according to claim 7, characterized in that, The step 6 is specifically as follows: The geological model matrix and the mined model matrix with equipment perception are fused through the wavelet transform Kalman filtering algorithm to obtain the fused data, which is the complete geological model data.

9. The method for fusing the perception data of the fully-mechanized mining face equipment and the geological exploration data according to claim 8, wherein When performing data fusion in step 6, use w i to represent the weight of the i-th data x i in the fusion process. Since the value of w i reflects the comprehensive trust level of other data in the i-th data x i , we can use w i to perform a weighted sum on x i to obtain the expression for data fusion: wherein, the weight coefficients satisfy w i In a trust system regarding x i , all information of each subsystem x1, x2,..., x n should be integrated. Therefore, a set of non - negative numbers a1, a2,..., a n needs to be obtained such that w i = a1x1 + a2x2 + … a n x n , where i = 1, 2, …, n Rewrite it in matrix form \(W = XA\), where \(W=[w_1, w_2,\cdots, w\) n T , \(A = [a_1, a_2,\cdots, a\) n T , in the matrix \(cH\), there exists a maximum modulus eigenvalue \(\lambda>0\) such that \(\lambda A = cHA\). Find \(\lambda\) and the corresponding eigenvector \(A\) satisfying \(a\) i \(\geq0\), then \(W = \lambda A\);​​ Normalize w i to obtain The final result of the fusion estimation of the data measured by all sensors is obtained as The final result of the fusion estimation of all data is obtained as Similarly

Images