CN111520139B - Coal mining machine roller adjusting method based on coal rock recognition - Google Patents
Coal mining machine roller adjusting method based on coal rock recognition Download PDFInfo
- Publication number
- CN111520139B CN111520139B CN202010383241.4A CN202010383241A CN111520139B CN 111520139 B CN111520139 B CN 111520139B CN 202010383241 A CN202010383241 A CN 202010383241A CN 111520139 B CN111520139 B CN 111520139B
- Authority
- CN
- China
- Prior art keywords
- coal
- rock
- image
- wall
- characteristic vector
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000003245 coal Substances 0.000 title claims abstract description 334
- 239000011435 rock Substances 0.000 title claims abstract description 158
- 238000005065 mining Methods 0.000 title claims abstract description 62
- 238000000034 method Methods 0.000 title claims abstract description 25
- 239000013598 vector Substances 0.000 claims abstract description 98
- 230000003068 static effect Effects 0.000 claims abstract description 44
- 230000035515 penetration Effects 0.000 claims description 42
- 241001061225 Arcos Species 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 abstract description 2
- 239000000523 sample Substances 0.000 description 29
- 239000000428 dust Substances 0.000 description 7
- 238000002347 injection Methods 0.000 description 7
- 239000007924 injection Substances 0.000 description 7
- 238000007621 cluster analysis Methods 0.000 description 6
- 238000000605 extraction Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 4
- 238000007781 pre-processing Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000009434 installation Methods 0.000 description 3
- 238000011002 quantification Methods 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 238000009412 basement excavation Methods 0.000 description 2
- 241001497337 Euscorpius gamma Species 0.000 description 1
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 238000005299 abrasion Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 239000000243 solution Substances 0.000 description 1
- 239000007921 spray Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21C—MINING OR QUARRYING
- E21C35/00—Details of, or accessories for, machines for slitting or completely freeing the mineral from the seam, not provided for in groups E21C25/00 - E21C33/00, E21C37/00 or E21C39/00
- E21C35/24—Remote control specially adapted for machines for slitting or completely freeing the mineral
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21C—MINING OR QUARRYING
- E21C39/00—Devices for testing in situ the hardness or other properties of minerals, e.g. for giving information as to the selection of suitable mining tools
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mining & Mineral Resources (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geology (AREA)
- Artificial Intelligence (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geochemistry & Mineralogy (AREA)
- Mechanical Engineering (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The invention discloses a coal cutter roller adjusting method based on coal rock recognition, which comprises the following steps of S1: sequentially arranging n image acquisition sensors on the coal rock recognition device and numbering so as to gridde the coal wall to record corresponding space coordinates; s2: setting a fixed angle of an image acquisition sensor to shoot a coal wall to obtain an effective image; s3: extracting a characteristic vector of the effective image, and identifying coal and rock according to the extracted characteristic vector; s4: and adjusting the height of the roller of the coal mining machine according to the identified coal and rock results. The invention completes the advanced identification of coal rocks on the coal face through an image processing technology or a static sounding sensing array, adjusts the height of the coal mining machine according to the identification result, excavates the coal layer as much as possible, avoids hard rocks, prevents the cutter face of the coal mining machine from being damaged, prolongs the service life of coal mining equipment such as the coal mining machine and the like, and has important guiding significance for intelligent mining of the coal face.
Description
Technical Field
The invention relates to the technical field of coal mine intellectualization, in particular to a coal mining machine roller adjusting method based on coal rock recognition.
Background
Coal mining machines are one of the main devices of the comprehensive mechanized coal mining working face, and all kinds of coal mining machines work according to a set fixed mining route at present. When the mining head of the coal mining machine cuts rocks, on one hand, the abrasion of the mining head of the coal mining machine is aggravated, the power consumption is increased, and even gas is detonated at high temperature due to overlarge friction force of the mining head; on the other hand, the content of gangue in the mined coal is increased. At present, the problems are solved by timely stopping an on-site operator to adjust the position of the digging head, and the labor intensity and the danger of workers are high.
Disclosure of Invention
Aiming at the problem that the mining head cannot be adjusted in time in the prior art, the invention provides a coal mining machine roller adjusting method based on coal rock identification, which is characterized in that coal rock type identification is carried out by acquiring coal wall images in real time and extracting characteristic vectors; meanwhile, the penetration resistance of the coal wall is collected in real time through the static sounding sensor, coal and rock are identified according to the penetration resistance, advanced coal and rock identification of the coal face is achieved, and therefore the height of the roller of the coal mining machine is adjusted according to an identification result.
In order to achieve the purpose, the invention provides the following technical scheme:
a coal mining machine roller adjusting method based on coal rock recognition specifically comprises the following steps:
s1: sequentially arranging n image acquisition sensors on the coal rock recognition device and numbering so as to gridde the coal wall to record corresponding space coordinates;
s2: setting a fixed angle of an image acquisition sensor to shoot a coal wall to obtain an effective image;
s3: extracting a characteristic vector of the effective image, and identifying coal and rock according to the extracted characteristic vector;
s4: and adjusting the height of the roller of the coal mining machine according to the identified coal and rock results.
Preferably, the S2 includes:
s2-1: setting a fixed angle theta of an image acquisition sensor to obtain an effective image by shooting, wherein theta is more than or equal to arcos (ng/h) and less than 90 degrees, h represents the height of the coal wall, n represents the number of the image acquisition sensors, and g represents the vertical distance from the image acquisition sensors to the coal wall;
s2-2: calculating the advance distance OD between the effective image and the image acquisition sensor: OD ═ g × tan θ -h/2 n.
Preferably, the S3 includes:
s3-1: respectively extracting characteristic vectors of coal images and rock images in a sample library to obtain a coal characteristic vector set c ═ { c ═ c1,c2,…,ck},ckA set of feature vectors and rock feature vectors β ═ β representing the coal images in the kth sample bank1,β2,…,βm},βmRepresenting the characteristic vectors of the rock images in the mth sample library, and respectively calculating Euclidean distances between coal characteristic vector sets and rock characteristic vector sets to obtain a distance set T ═ { T {11,T12,…,Tkm}，TkmRepresenting the Euclidean distance between the characteristic vector of the coal image in the kth sample bank and the characteristic vector of the rock image in the mth sample bank;
s3-2: acquiring a coal wall image in real time, and extracting a characteristic vector of the coal wall image to obtain a coal wall characteristic vector set
And gamma nt represents a characteristic vector of the coal wall image acquired by the nth image acquisition sensor at the t moment, and the coal and rock of the coal wall are identified by a cluster analysis method.
Preferably, in S3-2, the cluster analysis method includes:
calculating any coal wall feature vector gammantAnd the distances of all coal feature vectors in the set c and taking the minimum value a; then any coal wall characteristic vector gamma is calculatedntAnd the distances of all the rock feature vectors in the set beta are collected, and the minimum value b is taken; if a < b, then γntThe corresponding coal wall image is coal, if b is less than a, gamma isntThe corresponding coal wall image is rock.
Preferably, the present invention further comprises the steps of:
s5: sequentially mounting m static sounding sensors on the coal rock recognition device, numbering from top to bottom, and meshing the coal wall to record corresponding space coordinates;
s6: the static sounding sensor applies acting force to the coal wall and collects penetration resistance electric signals, and then the collected penetration resistance electric signals are compared with a preset penetration resistance range, so that coal and rock identification is carried out;
s7: and controlling and adjusting the height of the roller of the coal mining machine to avoid the rock according to the coal and rock identification result.
In summary, due to the adoption of the technical scheme, compared with the prior art, the invention at least has the following beneficial effects:
the invention completes the advanced identification of coal rocks on the coal face through an image processing technology or a static sounding sensing array, adjusts the height of the coal mining machine according to the identification result, excavates the coal layer as much as possible, avoids hard rocks, prevents the cutter face of the coal mining machine from being damaged, prolongs the service life of coal mining equipment such as the coal mining machine and the like, and has important guiding significance for intelligent mining of the coal face.
Description of the drawings:
fig. 1 is a schematic diagram of a coal petrography identification system according to an exemplary embodiment of the present invention.
Fig. 2 is a schematic diagram of a coal petrography recognition apparatus according to an exemplary embodiment 1 of the present invention.
Fig. 3 is a schematic diagram of a coal petrography recognition apparatus according to an exemplary embodiment 2 of the present invention.
Fig. 4 is a schematic view of a dust-proof module installation according to an exemplary embodiment of the present invention.
Fig. 5 is a schematic structural diagram of a dust-proof module according to an exemplary embodiment of the present invention.
Fig. 6 is a flowchart illustrating a method for adjusting a shearer drum based on coal rock identification according to an exemplary embodiment of the present invention.
Fig. 7 is a schematic diagram of an efficient image acquisition according to an exemplary embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to examples and embodiments. It should be understood that the scope of the above-described subject matter is not limited to the following examples, and any techniques implemented based on the disclosure of the present invention are within the scope of the present invention.
In the description of the present invention, it is to be understood that the terms "longitudinal", "lateral", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like, indicate orientations or positional relationships based on those shown in the drawings, and are used merely for convenience of description and for simplicity of description, and do not indicate or imply that the referenced devices or elements must have a particular orientation, be constructed in a particular orientation, and be operated, and thus, are not to be construed as limiting the present invention.
As shown in fig. 1, the invention provides a coal rock identification system based on image and static sounding array sensing, which comprises two identification subsystems, wherein a first identification subsystem comprises an image acquisition module, an image processing module, an image identification module and a controller; the second identification subsystem comprises a static sounding array module, a signal identification module and a controller.
The first identification subsystem:
the output end of the image acquisition module is connected with the input end of the image processing module, the output end of the image processing module is connected with the input end of the image recognition module, the output end of the image recognition module is connected with the first input end of the controller, and the output end of the controller is connected with the coal mining machine roller so as to control the adjustment of the height of the coal mining machine roller;
the image acquisition module comprises a plurality of image acquisition sensors and is used for gridding the coal wall, establishing a space coordinate system, acquiring the coal rock image in the grid and recording the corresponding space coordinate, and the resolution of the coal rock image is generally over 800 ten thousand pixels in order to ensure the definition and facilitate the image processing;
and the image processing module is used for extracting the coal and rock image characteristic vectors through the existing image processing algorithm and performing cluster analysis on the coal and rock image characteristic vectors in the sample library.
In this embodiment, the image processing module includes two parts, an offline processing module and an online processing module.
The off-line processing module is used for processing the original coal and rock images in the sample library (the coal and the rock in the sample library are separated), and preprocessing the original coal and rock images, such as unifying the image size and removing noise; and then, performing feature extraction and quantification on the preprocessed coal image to obtain a coal feature vector set c { c }1,c2,…,ck},ckRepresenting the characteristic vector of the coal image in the kth sample library, and performing characteristic extraction and quantification on the preprocessed rock image to obtain a rock characteristic vector set beta ═ { beta ═ beta [ beta ])1,β2,…,βm},βmAnd representing the feature vector of the rock image in the mth sample library, wherein the extracted features comprise texture features, angular point features, Gaussian features and the like. The feature vectors of the coal and rock images of the sample library are classified once, and theoretically, the Hamming distance or the Euclidean distance of the feature vectors of different coal images and different rock images is very close, and the feature vector distance of different rock images is also very close.
In this embodiment, the hamming distance or euclidean distance T of the feature vector between the coal feature vector set and the rock feature vector set is calculated respectively (the calculation of the hamming distance or euclidean distance is an existing mathematical formula, and is not described here again), for example, c in the coal feature vector set c is calculated1And the feature vector ({ beta ] of each rock image in the rock feature vector set beta1,β2,…,βm}) separately distance, then c2And { beta ]1,β2,…,βmGet the distance until ckAnd beta1～βmRespectively calculating the distances to obtain a distance set T ═ T11,T12,…,Tkm}，TkmAnd representing the distance between the characteristic vector of the coal image in the kth sample bank and the characteristic vector of the rock image in the mth sample bank.
The on-line processing module is used for processing the coal wall images acquired during the working process of the coal mining machine in real time, changing the sizes of the on-line processing module and the off-line processing module to be consistent through image preprocessing, and then passing through the off-line processing module and the off-line processing moduleObtaining a coal wall feature vector set by extracting and quantizing the features of the method with the same physical module
Gamma nt represents the characteristic vector of the coal wall image acquired by the nth image acquisition touch sensor at the t time, and then the coal wall characteristic vector set gamma, the coal characteristic vector set c in the sample library and the rock characteristic vector set beta are subjected to cluster analysis, wherein the specific cluster analysis method comprises the following steps: calculating the distances (Hamming distance or Euclidean distance) of all the coal images in the set gamma and the set c and taking the minimum value a; then calculating the distances of all rock images in the set gamma and the set beta and taking the minimum value b; comparing the sizes of a and b, and classifying the image into a corresponding class according to which one is small. If a and b are both larger than T, the image is not classified as rock or coal, and the image is judged to be an invalid image; according to the actual situation, the T value is adjusted.
And the image identification module is used for classifying the coal and rock of the object in the coal and rock image according to clustering analysis.
In this embodiment, when the image recognition module determines that the coal rock image is a coal image, the image recognition module skips over the coal rock image; if the coal rock image is judged to be the rock image, storing the spatial coordinates corresponding to the coal rock image; if the image is judged to be not a coal image or a rock image, the front coal wall can be shielded by dust, and the identification is abandoned and skipped; if the coal rock image is judged to be both the coal image and the rock image, storing the corresponding spatial coordinates of the coal rock image; to ensure that the shearer's drum can avoid the rock during excavation, the controller adjusts the height of the drum (e.g., up or down) based on the stored spatial coordinates so that the drum avoids the rock.
In order to ensure the accuracy of system identification, in this embodiment, a second identification subsystem is further provided:
the electric signal output end of the static sounding array module is connected with the electric signal input end of the signal identification module, the output end of the signal identification module is connected with the second input end of the controller, and the controller controls the height of the coal mining machine roller to be adjusted according to the identification result;
the static sounding array module comprises a plurality of static sounding sensors and is used for gridding the coal wall and establishing a space coordinate system, acting force is applied to the coal wall through each static sounding sensor, and penetration resistance electric signals and corresponding space coordinates of the coal wall are collected;
and the signal identification module is used for carrying out coal rock identification by combining a qualitative relation between the injection resistance and the engineering geological characteristics of the coal rock (the strength is higher and the pressure applied to the probe is larger due to the resistance response change of the probe of the static sounding sensor because the hardness of the coal and the rock in the coal wall is different) and a statistical coal and rock correlation relation (namely, the injection resistance ranges corresponding to the coal and the rock in the coal wall respectively) according to the acquired injection resistance electric signals.
In this embodiment, the signal identification module may also be divided into a signal offline processing module and a signal online processing module. The signal off-line processing module is used for acquiring coal and rock injection resistance samples and analyzing the coal and rock injection resistance ranges (coal and rock correlation relations) corresponding to the coal and the rock; the signal online processing module is used for analyzing the penetration resistance electric signals acquired by the static cone penetration sensor probe in real time and judging whether the contact part of the probe is coal or rock according to the penetration resistance range of the coal and the rock obtained by the signal offline processing module.
In this embodiment, if the signal identification module determines that the corresponding position is a coal seam, the corresponding position is skipped; if the corresponding position is judged to be a rock stratum, storing the corresponding space coordinate; to ensure that the shearer's drum is able to avoid rock during excavation, the controller adjusts the height of the drum (e.g., up or down) based on the stored formation spatial coordinates so that rock is avoided, thereby reducing wear of the mining head in the drum.
In the embodiment, the dust-proof device further comprises a dust-proof module, a dust-proof curtain is formed between the image acquisition module or the static sounding array module and the roller of the coal mining machine, and the influence of dust generated when the roller cuts on the image acquired by the image acquisition module is reduced.
In this embodiment, a coal rock recognition device includes n supports 10, and each support 10 is equipped with a static sounding array module 20 and an image acquisition module 30, and n supports 10 all are connected with the coal mining machine.
In the embodiment, the coal mining machine rollers are respectively arranged at the left side and the right side of the coal mining machine so as to cut the coal bed forwards and backwards; the left side and the right side of the coal mining machine are installed in a mirror image mode, so in the embodiment, for convenience of description, the installation structure of the left side system of the coal mining machine is described.
Example 1
As shown in fig. 2, a support 10 for rigid connection with a shearer loader; the support 10 comprises two side faces, one side face connected with the coal mining machine is a first side face 101, one side face contacted with a coal wall is a second side face 102, an included angle between the second side face 102 and the first side face 101 is alpha, alpha is more than or equal to 0 degrees and less than 180 degrees, and the included angle is generally set to be more than or equal to 60 degrees and less than 150 degrees.
In this embodiment, the n brackets 10 are fixed together by the connecting rods 50, that is, the connecting rods 50 are installed at the intersections of the first side surface 101 and the second side surface 102 in each bracket 10 to fix the n brackets 10, and the image capturing sensors 20 are installed on the connecting rods 50 at the intersections of the first side surface 101 and the second side surface 102, and capture corresponding images of the coal rock and transmit the images to the controller for coal rock recognition.
In this embodiment, the static cone penetration sensor 30 needs to contact with the coal wall, and identifies the coal rock by applying an acting force to the coal wall to generate penetration resistance, so that the static cone penetration sensor 30 is installed at the top end of the second side surface 102 (i.e., the end of the second side surface 102 contacting with the coal wall, and the other end of the second side surface 102 is connected with the first side surface 101); in the movement process, the static sounding sensor 30 penetrates into the coal wall for a certain distance (for example, 2cm) to generate penetration resistance, and then electric signals of the penetration resistance are collected and transmitted to the controller for coal rock identification. The number of static cone penetration sensors 30 and the number of image capturing sensors 20 may be the same or different, since the two sensors operate differently.
The image processing module, the image recognition module and the signal recognition module are arranged on the bracket, and the controller is arranged on the coal mining machine; or the image processing module, the image recognition module, the signal recognition module and the controller can be installed on the coal mining machine in a centralized manner, so that the weight of the bracket 10 can be reduced. The transmission between the modules can adopt cable or wireless transmission.
Example 2
As shown in fig. 3, a support 10 for rigid connection with a shearer loader; the bracket 10 comprises three side surfaces, wherein one side surface connected with the coal mining machine is a first side surface 101, one side surface close to a coal wall (not contacting the coal wall) is a second side surface 102, an included angle between the second side surface 102 and the first side surface 101 is alpha, and alpha is more than or equal to 0 degree and less than 180 degrees; one side surface in contact with the coal wall is a third side surface 103, an included angle between the third side surface 103 and the second side surface 102 is beta, alpha is more than beta and less than alpha plus 90 degrees, and the third side surface 103 deflects towards the direction far away from the coal mining machine, so that the static sounding sensor can be conveniently installed on the third side surface 103, and the coal and rock in the coal wall can be identified in advance. The third side surface 103 has the advantages that the static sounding sensor 30 can be conveniently installed, the static sounding sensor 30 can detect and identify coal and rock in a coal wall in advance when the coal mining machine moves, and the coal mining machine can timely make judgment to adjust the height of a roller of the coal mining machine.
In this embodiment, the n brackets 10 are fixed together by the connecting rods 50, that is, the connecting rods 50 are installed at the intersections of the first side surface 101 and the second side surface 102 in each bracket 10 to fix the n brackets 10, and the image capturing sensors 20 are installed on the connecting rods 50 at the intersections of the first side surface 101 and the second side surface 102, and capture corresponding images of the coal rock and transmit the images to the controller for coal rock recognition.
In this embodiment, the static cone penetration sensor 30 needs to contact with the coal wall, and identifies the coal rock by applying an acting force to the coal wall to generate penetration resistance, so that the static cone penetration sensor 30 is installed at the top end of the second side surface 102 (i.e., the end of the second side surface 102 contacting with the coal wall, and the other end of the second side surface 102 is connected with the first side surface 101); in the movement process, the static sounding sensor 30 penetrates into the coal wall for a certain distance (for example, 2cm) to generate penetration resistance, and then electric signals of the penetration resistance are collected and transmitted to the controller for coal rock identification. The number of static cone penetration sensors 30 and the number of image capturing sensors 20 may be the same or different, since the two sensors operate differently.
The image processing module, the image recognition module and the signal recognition module are arranged on the bracket, and the controller is arranged on the coal mining machine; or the image processing module, the image recognition module, the signal recognition module and the controller can be installed on the coal mining machine in a centralized manner, so that the weight of the bracket 10 can be reduced. The transmission between the modules can adopt cable or wireless transmission.
As shown in fig. 4, in order to reduce the influence of dust generated when the shearer drum cuts the coal wall on the image acquisition sensor 20 to acquire images or the static sounding sensor 30 to apply force to the coal wall, a dust prevention module 40 may be installed between the shearer drum and the image acquisition sensor 20 to acquire images or the static sounding sensor 30. To facilitate the installation of the dust-proof module 40, a fixed holder 104 may be installed on the first side 101, and then the dust-proof module 40 may be installed on the fixed holder 104.
As shown in fig. 5, the dust-proof module 40 includes an inlet passage 401, a housing 402, a duct 403, and a spout 404. In this embodiment, the housing 402 is in an inverted cone shape, the housing 402 includes a side surface 4021 and a bottom surface 4022, and the bottom surface 4022 is in an arc shape; the nozzles 404 are distributed on the bottom surface 4022 evenly, and each nozzle 404 is spaced at a certain distance; the plurality of pipes 403 are arranged in one-to-one correspondence with the nozzles 404 and are arranged inside the housing 402 to connect the nozzles 404 with the inlet channel 401. In this embodiment, each nozzle 404 may spray a fan-shaped water curtain or air curtain, and the plurality of nozzles 404 form a dustproof water curtain or air curtain, so that when the coal mining machine drum cuts coal walls to generate dust, the water curtain or air curtain sprayed by the dustproof module 40 can effectively prevent the dust from drifting to the image acquisition sensor 20 or the static cone penetration sensor 30, thereby improving the accuracy of image acquisition.
In this embodiment, based on the above coal petrography recognition system and apparatus, a coal petrography recognition method is further provided, as shown in fig. 6, the method specifically includes the following steps:
s1: and (3) installing n image acquisition sensors on the coal rock recognition device and numbering from top to bottom so as to gridding the coal wall.
In the embodiment, the height of the coal wall is h, and when the n image acquisition sensors grid the coal wall, the length of each grid is h/n, and the width of each grid is h/n. Assuming that the maximum speed of the coal mining machine is V, the acquisition frequency f is Vn/h, and when the visual angle of the image sensor is greater than or equal to 90 degrees, the acquired coal rock image can completely cover the coal wall, so that the identification error caused by incomplete coverage of the coal wall is avoided, and the accuracy is improved.
S2: and setting a fixed angle of the image acquisition sensor to shoot the coal wall to obtain an effective image.
In this embodiment, the controller needs a certain response time for processing data, that is, there is a time difference between the image acquisition and the control and adjustment of the height of the shearer drum, so that in order to eliminate the time difference to ensure that the height of the shearer drum can be adjusted in time, the image acquisition needs to be performed in advance. In this embodiment, the image capturing sensor is mounted on the bracket and forms an angle θ with the second side surface 102, and θ is greater than 0 °, so as to shoot the coal wall obliquely, thereby obtaining an effective image.
S2-1: a fixed angle theta of the image capturing sensor is set.
In this embodiment, as shown in fig. 7, if it is desired that the image acquired by the image acquisition sensor module within the 90 ° viewing angle completely covers the coal wall, in order to facilitate the coal mining machine to acquire the coal rock image and identify the coal and rock during the movement process, the coal rock image needs to be acquired in advance, and therefore a fixed angle θ of the image acquisition sensor needs to be set so as to form a certain forward inclination angle θ with the coal wall. After the fixed angle theta of the image acquisition sensor is set, the image acquisition sensor is taken as a vertex to make an isosceles right triangle AOF, namely the vertical intersection point of the A point where the image acquisition sensor is located and the coal wall is taken as an O point, and the F point is the midpoint of an image area DE, namely the virtual intersection point of the vertex of the image acquisition sensor and the coal wall. When the image acquisition sensor and the coal wall form a certain forward inclination angle theta, the shot coal wall coal rock image area is DE, namely the effective image; and the effective image corresponds to a square area of the coal wall, and the OD is the advance distance between the effective image and the image acquisition sensor.
In this embodiment, the hypotenuse AF should be greater than or equal to the width h/n of the grid corresponding to the image capturing sensor (i.e. AF is greater than or equal to h/n), and in the isosceles right triangle AOF, cos θ is OA/AF less than or equal to g/(h/n) less than or equal to 1, so that the solution: when g is less than or equal to h/n, theta is equal to arcos (ng/h), namely the minimum value of theta; when g is more than or equal to h/n, theta is more than or equal to 0 degree; so the range of theta is 0 DEG theta < 90 DEG or arcos (ng/h) theta < 90 DEG, h represents the height of the coal wall, n represents the number of image capturing sensors, and g represents the vertical distance of the image capturing sensors from the coal wall.
S2-2: and calculating the advance distance OD between the effective image and the image acquisition sensor.
In this embodiment, as shown in fig. 7, the effective image is a grid coal rock image shot by the image capture sensor, and corresponds to a square area of the coal wall, where the length DE is the length h/n of the grid and the width is also h/n. The advance distance is also an interval, i.e., OD to OE, where OD ═ OF-DF ═ g ═ tan θ -h/2n, and OE ═ OF + EF ═ g ═ tan θ + h/2n, and if OD < 0, the lower limit is 0.
In this embodiment, the controller needs a certain response time for processing data, that is, there is a time difference between the image acquisition and the control adjustment of the height of the shearer drum, so that the advance distance is to ensure that the time difference is eliminated so that the height of the shearer drum can be adjusted in time, and the real-time performance is improved.
S3: and extracting the characteristic vector of the effective image, and identifying the coal and rock according to the characteristic vector.
In this embodiment, at the same time, the n image capturing sensors capture the coal wall, and n effective images can be obtained, so that preprocessing can be performed before feature vector extraction is performed on the n effective images, including naming the effective images according to the number and the spatial position of the capturing device and unifying the sizes of the effective images, and thus, corresponding feature vectors can be obtained conveniently, and coal and rock recognition can be performed according to the feature vectors.
S3-1: respectively extracting characteristic vectors of the coal images and the rock images in the sample library to obtain a coal characteristic vector set c and a rock characteristic vector set beta, and calculating the distance between the coal characteristic vector set and the rock characteristic vector set to obtain a distance set T.
In this embodiment, the coal images in the sample library are subjected to feature extraction (including texture features, angular point features, gaussian features, and the like) and quantized to obtain a coal feature vector set c ═ { c ═ c1,c2,…,ck},ckRepresenting the characteristic vector of the coal image in the kth sample library, and performing characteristic extraction and quantification on the preprocessed rock image to obtain a rock characteristic vector set beta ═ { beta ═ beta [ beta ])1,β2,…,βm},βmAnd representing the feature vector of the rock image in the mth sample library. The feature vectors of the coal and rock images of the sample library are classified once, and theoretically, the Hamming distance or the Euclidean distance of the feature vectors of different coal images and different rock images is very close, and the feature vector distance of different rock images is also very close.
In this embodiment, the hamming distance or euclidean distance T of the feature vector between the coal feature vector set and the rock feature vector set is calculated respectively (the calculation of the hamming distance or euclidean distance is an existing mathematical formula, and is not described here again), for example, c in the coal feature vector set c is calculated1And the feature vector ({ beta ] of each rock image in the rock feature vector set beta1,β2,…,βm}) separately distance, then c2And { beta ]1,β2,…,βmGet the distance until ckAnd beta1～βmRespectively calculating the distances to obtain a distance set T ═ T11,T12,…,Tkm}，TkmRepresenting the distance between the characteristic vector of the coal image in the kth sample bank and the characteristic vector of the rock image in the mth sample bank; selecting the maximum value T from the distance setmaxAs a threshold value.
S3-2: and acquiring a coal wall image in real time, extracting a characteristic vector of the coal wall image to obtain a coal wall characteristic vector set gamma, and identifying the coal and rock of the coal wall by a cluster analysis method.
In this embodiment, the coal wall image is collected in real time, and through image preprocessing, feature extraction and quantization are performed to obtain a coal wall feature vector set
Gamma nt represents the characteristic vector of the coal wall image acquired by the nth image acquisition sensor for the t time, and then the coal wall characteristic vector set gamma and the coal characteristic vector set c and the rock in the sample library are combinedAnd performing clustering analysis on the characteristic vector set beta.
The specific clustering analysis method comprises the following steps:
in this embodiment, the n image capturing sensors simultaneously capture coal wall images at the same time t, so for real-time image analysis, the clustering object is the coal wall feature vector { r } corresponding to the coal wall images captured by the n image capturing sensors at the same time t1t,r2t,...,rnt}。
Calculation of { r1t,r2t,...,rntAny one of the coal wall feature vectors gammantAnd the distances (Hamming distance or Euclidean distance) of all coal feature vectors in the set c and taking the minimum value a; recalculate { r1t,r2t,...,rntAny one of the coal wall feature vectors gammantAnd the distances of all the rock feature vectors in the set beta are collected, and the minimum value b is taken; comparing the sizes of a and b, and classifying the image into a corresponding class according to which one is small. Comparing the sizes of a and b, if a < b, then gammantThe corresponding coal wall image is coal; if b < a, then γntThe corresponding coal wall image is rock. If Tmax< a < b, or TmaxB < a, then gammantThe corresponding image is classified as neither rock nor coal, i.e. gamma is determinedntThe corresponding image is an invalid image; according to the actual situation, TmaxThe value will be adjusted.
S4: if the coal wall image is judged to be the coal image, keeping silence; if the coal wall image is judged to be the rock image, sending the spatial position data corresponding to the coal wall image to a controller, and adjusting the height of a roller of the coal mining machine by the controller according to the spatial position to avoid the spatial position; if the coal wall image is judged to be neither the coal image nor the rock image, the front coal wall is possibly shielded by dust, the coal wall image is discarded, the silence is kept, the height of the roller is not adjusted, and the coal wall is cut while the original direction is kept; and if the coal wall image is judged to be the coal image and the rock image, sending data of the corresponding position of the coal wall image to the controller, and adjusting the height of the roller of the coal mining machine by the controller according to the space position to avoid the space position.
In order to improve the accuracy of the coal mining machine for identifying the coal and the rock in the coal wall, the method further comprises the following steps:
s5: and (3) installing m static sounding sensors on the coal rock recognition device, numbering from top to bottom, and meshing the coal wall, wherein the length and the width of each grid are h/m.
In this embodiment, when the m static sounding sensors gridd the coal wall, the static sounding sensors respectively correspond to each grid one by one, so that the positions of the grids can be conveniently known.
S6: the static sounding sensor applies acting force to the coal wall and collects penetration resistance electric signals, and then the collected penetration resistance electric signals are transmitted to the signal identification module to be compared with the penetration resistance range, so that coal and rock identification is carried out.
In this embodiment, the signal identification module may be divided into a signal offline processing module and a signal online processing module. The signal off-line processing module is used for acquiring coal and rock injection resistance samples and analyzing the coal and rock injection resistance ranges (coal and rock correlation relations) corresponding to the coal and the rock; the signal online processing module is used for analyzing the penetration resistance electric signals acquired by the static cone penetration sensor probe in real time, namely comparing the penetration resistance electric signals with the penetration resistance ranges of coal and rock obtained by the signal offline processing module, so as to judge whether the contact part of the probe is coal or rock.
In this embodiment, the static sounding sensor is driven to apply an acting force to the coal wall as the coal mining machine advances, penetration resistance electric signals of the static sounding sensor corresponding to the position of the coal wall are collected, the collected electric signals are preprocessed, a static sounding curve graph is respectively established according to the number of the static sounding sensor, the electric signal strength and the position corresponding to the collected electric signals, and then coal rock identification is performed through a qualitative relation between penetration resistance and engineering geological characteristics of the coal rock (because the hardness of coal and rock in the coal wall is different, the resistance response change of a probe of the static sounding sensor is high, the higher the strength is, the higher the pressure is, the higher the strength is of the corresponding penetration resistance electric signal strength is) and statistical coal and rock correlation relations (namely, the penetration resistance ranges corresponding to the coal and rock in the coal wall respectively).
For example, the strength of the coal in the coal wall is low, and can be best taken from the static cone penetration graphSmall value and maximum value, and the obtained coal penetration resistance range is f1～f2In the unit of N; similarly, the penetration resistance range of the rock is f3～f4In the unit of N; and because the strength of coal is much less than that of rock, i.e. f2＜f3. When the static sounding sensor acquires the coal wall in real time to obtain the penetration resistance f, when f is less than f2Then the coal wall corresponding to the grid is coal; when f > f3And in time, the coal wall corresponding to the grid is rock.
In the embodiment, the advance distance is the horizontal distance from the roller to the static sounding sensor, and the position of the advance distance in front of the roller is coal or rock, wherein the coal is mined and the rock is avoided.
S7: if the penetration resistance electric signal acquired by the sensor is judged to correspond to the coal bed, keeping silence, namely, not adjusting the height of the roller of the coal mining machine; and if the electric signal of the penetration resistance acquired by the sensor is judged to correspond to the rock stratum, transmitting the data (space position coordinates) of the position corresponding to the acquired signal to the controller, and adjusting the height of the roller of the coal mining machine by the controller according to the space position so as to avoid the space position.
It will be understood by those of ordinary skill in the art that the foregoing embodiments are specific examples for carrying out the invention, and that various changes in form and details may be made therein without departing from the spirit and scope of the invention in practice.
Claims (4)
1. A coal mining machine roller adjusting method based on coal rock recognition is characterized by comprising the following steps:
s1: sequentially arranging n image acquisition sensors on the coal rock recognition device and numbering so as to gridde the coal wall to record corresponding space coordinates;
s2: setting a fixed angle of an image acquisition sensor to shoot a coal wall to obtain an effective image;
s2-1: setting a fixed angle theta of an image acquisition sensor to obtain an effective image by shooting, wherein theta is more than or equal to arcos (ng/h) and less than 90 degrees, h represents the height of the coal wall, n represents the number of the image acquisition sensors, and g represents the vertical distance from the image acquisition sensors to the coal wall;
s2-2: calculating the advance distance OD between the effective image and the image acquisition sensor: OD ═ g × tan θ -h/2 n;
s3: extracting a characteristic vector of the effective image, and identifying coal and rock according to the extracted characteristic vector;
s4: and adjusting the height of the roller of the coal mining machine according to the identified coal and rock results.
2. The shearer drum adjustment method based on coal petrography recognition as claimed in claim 1, wherein the S3 includes:
s3-1: respectively extracting characteristic vectors of coal images and rock images in a sample library to obtain a coal characteristic vector set c ═ { c ═ c1,c2,…,ck},ckA set of feature vectors and rock feature vectors β ═ β representing the coal images in the kth sample bank1,β2,…,βm},βmRepresenting the characteristic vectors of the rock images in the mth sample library, and respectively calculating Euclidean distances between coal characteristic vector sets and rock characteristic vector sets to obtain a distance set T ═ { T {11,T12,…,Tkm}，TkmRepresenting the Euclidean distance between the characteristic vector of the coal image in the kth sample bank and the characteristic vector of the rock image in the mth sample bank;
s3-2: acquiring a coal wall image in real time, and extracting a characteristic vector of the coal wall image to obtain a coal wall characteristic vector set
γntAnd (4) representing the characteristic vector of the coal wall image acquired by the nth image acquisition sensor at the t moment, and identifying the coal and rock of the coal wall by a clustering analysis method.
3. The shearer drum adjusting method based on coal petrography recognition as claimed in claim 2, wherein in S3-2, the clustering method is:
calculating any coal wall feature vector gammantAnd all coals in the aggregate cTaking the distance of the characteristic vector and taking the minimum value a; then any coal wall characteristic vector gamma is calculatedntAnd the distances of all the rock feature vectors in the set beta are collected, and the minimum value b is taken; if a < b, then γntThe corresponding coal wall image is coal, if b is less than a, gamma isntThe corresponding coal wall image is rock.
4. The shearer drum adjustment method based on coal petrography recognition as claimed in claim 1, further comprising the steps of:
s5: sequentially mounting m static sounding sensors on the coal rock recognition device, numbering from top to bottom, and meshing the coal wall to record corresponding space coordinates;
s6: the static sounding sensor applies acting force to the coal wall and collects penetration resistance electric signals, and then the collected penetration resistance electric signals are compared with a preset penetration resistance range, so that coal and rock identification is carried out;
s7: and controlling and adjusting the height of the roller of the coal mining machine to avoid the rock according to the coal and rock identification result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010383241.4A CN111520139B (en) | 2020-05-08 | 2020-05-08 | Coal mining machine roller adjusting method based on coal rock recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010383241.4A CN111520139B (en) | 2020-05-08 | 2020-05-08 | Coal mining machine roller adjusting method based on coal rock recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111520139A CN111520139A (en) | 2020-08-11 |
| CN111520139B true CN111520139B (en) | 2021-07-23 |
Family
ID=71906735
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010383241.4A Active CN111520139B (en) | 2020-05-08 | 2020-05-08 | Coal mining machine roller adjusting method based on coal rock recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111520139B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103927514A (en) * | 2014-04-09 | 2014-07-16 | 中国矿业大学(北京) | Coal and rock recognition method based on random local image features |
| CN105005081A (en) * | 2015-08-11 | 2015-10-28 | 西安科技大学 | Dynamic tomography system and method for fully mechanized face near field coal rock under exploiting excitation of coal machine |
| CN107120116A (en) * | 2017-07-03 | 2017-09-01 | 中国矿业大学 | A kind of coal mining machine roller automatic height regulating device and method based on image recognition |
| CN208858354U (en) * | 2018-08-23 | 2019-05-14 | 江苏中机矿山设备有限公司 | Coalcutter cutting control system |
| CN110175517A (en) * | 2019-04-19 | 2019-08-27 | 中国煤炭科工集团太原研究院有限公司 | A kind of Coal-rock identification method based on transform domain numerical characteristic |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9650893B2 (en) * | 2011-04-01 | 2017-05-16 | Joy Mm Delaware, Inc. | Imaging-based interface sensor and control device for mining machines |
-
2020
- 2020-05-08 CN CN202010383241.4A patent/CN111520139B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103927514A (en) * | 2014-04-09 | 2014-07-16 | 中国矿业大学(北京) | Coal and rock recognition method based on random local image features |
| CN105005081A (en) * | 2015-08-11 | 2015-10-28 | 西安科技大学 | Dynamic tomography system and method for fully mechanized face near field coal rock under exploiting excitation of coal machine |
| CN107120116A (en) * | 2017-07-03 | 2017-09-01 | 中国矿业大学 | A kind of coal mining machine roller automatic height regulating device and method based on image recognition |
| CN208858354U (en) * | 2018-08-23 | 2019-05-14 | 江苏中机矿山设备有限公司 | Coalcutter cutting control system |
| CN110175517A (en) * | 2019-04-19 | 2019-08-27 | 中国煤炭科工集团太原研究院有限公司 | A kind of Coal-rock identification method based on transform domain numerical characteristic |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111520139A (en) | 2020-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111411961A (en) | Coal petrography identification system and device | |
| WO2019007439A1 (en) | Automatic height adjusting apparatus and method for shearer based on advanced detection of shearer seismic source | |
| CN109630110B (en) | Fully-mechanized coal mining face coal seam thickness self-adaptive cutting control method and electronic equipment | |
| CN109113741B (en) | Automatic coal caving control system based on video monitoring image recognition | |
| CN106194181B (en) | Intelligent working face coal-rock interface recognition method based on geological data | |
| CN110214120B (en) | Belt inspection system and method | |
| CN212614710U (en) | Coal petrography identification system and device | |
| CN110080766A (en) | Fully-mechanized mining working coal petrography identification device and method | |
| CN108510539A (en) | A kind of coal production monitoring method based on image analysis | |
| CN108564628A (en) | A kind of cutterhead vision positioning orientation system towards development machine automation | |
| CN110159199B (en) | Coal mine water exploration drilling positioning method based on image recognition | |
| CN108119140A (en) | A kind of coalcutter driving attitude adjusts system | |
| CN113137230B (en) | Coal-rock interface recognition system | |
| CA2112030A1 (en) | Process and a device for the continuous nondestructive control of rails on a railway line by ultrasonics | |
| CN111520139B (en) | Coal mining machine roller adjusting method based on coal rock recognition | |
| CN203744928U (en) | Device for measuring external geometrical sizes and surface quality of steel rail | |
| CN110691889A (en) | Adaptive pitch control in longwall mining systems | |
| WO2010077970A2 (en) | System and method for reducing signature variation of seismic sources | |
| CN112211657B (en) | Method for intelligently judging closing of coal discharge port of top coal caving hydraulic support | |
| CN113945333A (en) | Method for detecting full water pipeline by comprehensively utilizing multiple devices | |
| CN108106569B (en) | Fully mechanized coal mining face straightness detection method and system based on vision | |
| CN113256551B (en) | Roadway roof rigid belt drilling identification and positioning system and method based on machine vision | |
| CN112764113B (en) | Method for detecting coal seam top and bottom plate interface by using drilling radar in underground coal mine | |
| US20200211203A1 (en) | Systems and methods for monitoring offshore structures | |
| US11204108B2 (en) | Apparatus and method for sonar imaging and managing of undersea cable laying |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |