CN117292197A - TBM contact interface identification method - Google Patents
TBM contact interface identification method Download PDFInfo
- Publication number
- CN117292197A CN117292197A CN202311301965.XA CN202311301965A CN117292197A CN 117292197 A CN117292197 A CN 117292197A CN 202311301965 A CN202311301965 A CN 202311301965A CN 117292197 A CN117292197 A CN 117292197A
- Authority
- CN
- China
- Prior art keywords
- interface
- image
- cutterhead
- rock
- slag
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 239000011435 rock Substances 0.000 claims abstract description 125
- 239000003245 coal Substances 0.000 claims abstract description 81
- 239000002893 slag Substances 0.000 claims abstract description 81
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 36
- 238000004458 analytical method Methods 0.000 claims abstract description 12
- 238000011010 flushing procedure Methods 0.000 claims abstract description 9
- 239000007788 liquid Substances 0.000 claims abstract description 6
- 238000013145 classification model Methods 0.000 claims description 20
- 238000012549 training Methods 0.000 claims description 17
- 238000010521 absorption reaction Methods 0.000 claims description 14
- 238000005406 washing Methods 0.000 claims description 14
- 238000011156 evaluation Methods 0.000 claims description 8
- 238000013526 transfer learning Methods 0.000 claims 1
- 230000005641 tunneling Effects 0.000 abstract description 21
- 230000008569 process Effects 0.000 abstract description 17
- 238000000926 separation method Methods 0.000 abstract description 3
- 239000002245 particle Substances 0.000 description 16
- 239000002689 soil Substances 0.000 description 13
- 238000009826 distribution Methods 0.000 description 12
- 238000001914 filtration Methods 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000005520 cutting process Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000006399 behavior Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 4
- 238000007599 discharging Methods 0.000 description 4
- 239000000428 dust Substances 0.000 description 4
- 238000003064 k means clustering Methods 0.000 description 4
- 239000000700 radioactive tracer Substances 0.000 description 4
- 238000003860 storage Methods 0.000 description 4
- 230000015572 biosynthetic process Effects 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000009977 dual effect Effects 0.000 description 3
- 238000005755 formation reaction Methods 0.000 description 3
- 238000013508 migration Methods 0.000 description 3
- 230000005012 migration Effects 0.000 description 3
- 238000009825 accumulation Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000009933 burial Methods 0.000 description 2
- 238000007635 classification algorithm Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000005303 weighing Methods 0.000 description 2
- 238000004873 anchoring Methods 0.000 description 1
- 238000005422 blasting Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005553 drilling Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
- G06V10/765—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects using rules for classification or partitioning the feature space
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/143—Sensing or illuminating at different wavelengths
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Excavating Of Shafts Or Tunnels (AREA)
Abstract
The invention relates to the field of tunneling, and discloses a TBM contact interface identification method, which comprises the following steps: obtaining cutterhead rock slag from a conveyor belt, obtaining an image of the cutterhead rock slag, determining a first characteristic of an interface based on the image of the cutterhead rock slag, performing mud-water separation on the cutterhead rock slag, determining a second characteristic of the interface based on the rock slag image after mud-water separation, determining a coal slime characteristic of the interface based on an analysis result of mud liquid after water flushing, and determining characteristics of a cutterhead contact surface based on the first characteristic of the interface, the second characteristic of the interface and the coal slime characteristic of the interface. According to the method and the device, the information of the interface can be better obtained, physical analysis is further carried out on the medium according to the information of the interface, and the property of the medium is accurately obtained and used for guiding the shield process.
Description
Technical Field
The invention relates to the field of tunneling, in particular to a TBM contact interface identification method.
Background
At present, coal mine roadway tunneling modes mainly comprise a comprehensive tunneling method, a drilling and blasting method and a continuous miner method (only suitable for coal roadway tunneling). The tunneling methods are easy to solve the problems of unbalanced tunneling, anchoring and transportation in actual construction. Aiming at the situation, some coal mines adopt shield tunneling systems, so that the tunneling efficiency of the coal mine rock roadway is greatly improved. The equipment mainly used in the shield tunneling system is a shield machine and is mainly used for tunneling tunnels. The modern shield machine has higher technological content, integrates a plurality of technologies such as light, machine, electricity, liquid, sensors and the like, can realize the functions of cutting and conveying rock and soil bodies, supporting forming roadways and the like, and has higher overall reliability and safety. However, the shield machine needs professional personnel to control, the shield machine of the coal mine also needs personnel to have quite professional knowledge, the workload of the personnel can be effectively reduced through intelligent assistance, but because the tunnel formation of the coal mine is inconsistent with the common shield process, a special solution is needed to be provided. The coal mine tunnel is mainly related to geological environment, the contact surface of the shield machine is difficult to directly acquire through a sensor, and the accurate acquisition of the contact surface is very important for guiding the shield process.
Disclosure of Invention
The invention aims to overcome one or more of the prior art problems and provide a method for identifying a TBM contact interface.
In order to achieve the above object, the present invention provides a method for identifying a TBM contact interface, including:
acquiring cutterhead rock slag from a conveyor belt, and acquiring an image of the cutterhead rock slag;
determining a first feature of the interface based on the image of the cutterhead slag;
washing the cutterhead rock slag with water to separate mud from water;
determining a second characteristic of the interface based on the sludge-water separated rock image;
determining the coal slime characteristics of the interface based on the analysis result of the mud liquid after water flushing;
the characteristics of the cutterhead contact surface are determined based on the first characteristics of the interface, the second characteristics of the interface, and the slime characteristics of the interface.
According to one aspect of the invention, the image of cutterhead rock is acquired by means of dual wavelengths, including visible and infrared wavelengths, and after acquisition of the two wavelength images, the images are fused to obtain the image of cutterhead rock.
According to one aspect of the invention, infrared light is used as a light source, reflected light of an infrared light wave band is collected, an image of rock residue after washing is obtained, the boundary of the rock residue image is determined according to the reflection degree of visible light in the image, the absorption of rock residue components to infrared light is determined according to the brightness of each region in the image, the composition of each region is determined according to the absorption of infrared light, and the second characteristic of the interface is determined.
According to one aspect of the invention, an image of cutterhead rock slag obtained on a conveyor belt is taken as input, a rock slag image after water flushing and coal slime characteristics of an interface are taken as output, a migration learning training interface identification model is used, historical data of manual intervention operation is taken as a training set, and the output of the historical data or the interface identification model is taken as input of an interface operation network to construct a KNN neighbor classification model.
According to one aspect of the invention, taking an image of cutterhead slag as input, acquiring a coal slime characteristic of an interface and a second characteristic of the interface by using a pre-trained model; and taking the coal slime characteristics of the interface and the second characteristics of the interface as input of a KNN neighbor classification model to obtain classification of working conditions, when the Euclidean distance requirement is met, using manual operation of the classification as an operation recommended value, receiving evaluation of automatic operation of the interface by a user, and when the same classification of the working conditions is obtained, replacing the operation recommended value by the manual operation.
According to one aspect of the invention, the characteristics of the cutter head contact surface are used as input, and the historical data of the manual intervention operation are used as output of a training set to construct a KNN neighbor classification model.
According to one aspect of the invention, the characteristics of the contact surface of the cutterhead are used as the input of a KNN neighbor classification model, classification of working conditions is obtained, when the Euclidean distance requirement is met, the classification manual operation is used as an operation recommended value, and otherwise, the manual operation is used.
Based on the above, the invention has the beneficial effects that: according to the method and the device, the information of the interface can be better obtained, physical analysis is further carried out on the medium according to the information of the interface, and the property of the medium is accurately obtained and used for guiding the shield process.
Drawings
FIG. 1 is a flow chart of a method of identifying a TBM contact interface in accordance with the present invention.
Detailed Description
The present disclosure will now be discussed with reference to exemplary embodiments, it being understood that the embodiments discussed are merely for the purpose of enabling those of ordinary skill in the art to better understand and thus practice the present disclosure and do not imply any limitation to the scope of the present disclosure.
As used herein, the term "comprising" and variants thereof are to be interpreted as meaning "including but not limited to" open-ended terms. The terms "based on" and "based at least in part on" are to be construed as "at least one embodiment.
Fig. 1 is a flowchart of a method for identifying a TBM contact interface according to an embodiment of the present invention, and as shown in fig. 1, the method for identifying a TBM contact interface includes:
in order to achieve the above object, the present invention provides a method for identifying a TBM contact interface, including:
acquiring cutterhead rock slag from a conveyor belt, and acquiring an image of the cutterhead rock slag;
determining a first feature of the interface based on the image of the cutterhead slag;
washing the cutterhead rock slag with water to separate mud from water;
determining a second characteristic of the interface based on the sludge-water separated rock image;
determining the coal slime characteristics of the interface based on the analysis result of the mud liquid after water flushing;
the characteristics of the cutterhead contact surface are determined based on the first characteristics of the interface, the second characteristics of the interface, and the slime characteristics of the interface.
According to one embodiment of the invention, slag in a slag discharge groove in the shield machine cutterhead can drop downwards during the rotation of the shield machine cutterhead, and an upper slag slide machine and a lower slag slide machine can be arranged in the slag slide machine for collecting or discharging slag. Further, a conveyor belt is arranged in the mechanism and used for receiving the dregs discharged by the dreg discharging groove in the tray. The sensor can be arranged on the cutterhead and used for judging the position of the cutterhead, the position of the cutterhead can be any one point, when the cutterhead is at the 0 point moment, the cutterhead is used as the 0 point of the space, and the coal-rock interface information in the tunneling stroke can be obtained by analyzing the rock slag when the cutterhead returns to the 0 point moment. In the rotating process of the cutter head of the shield machine, slag and positions on a conveyor belt behind the cutter head of the shield machine are associated, but certain accumulation exists in a slag chute box, and the difference of positions obtained by images of the slag chute box and the conveyor belt is limited, the images obtained by the conveyor belt do not strictly belong to images of the current coal-rock interface, and certain incoherence exists, but the interface operated by the current position is associated with slag images on the conveyor belt, and the advancing speed of the TBM is in the cm/min level in the tunneling process, so that the TBM still has reference value for data analysis. In particular, if in a trip, interface information is obtained by collecting a portion of the coal rock in the trip for analysis. In order to implement the above-described procedure, the setting of the relevant apparatus will be explained first. Firstly, a cutter head of the shield machine comprises a position sensor, wherein the position sensor can be used for acquiring and judging the state of the cutter head, in particular the position of a 0 point relative to the cutter head;
and secondly, an image acquisition sensor, such as a camera or a video acquisition device, is arranged on a conveyor belt connected with the cutterhead slag chute and used for acquiring high-resolution images. The acquired images contain time stamps, the relation between the images and the actual tunneling process can be acquired in a tracing mode, for example, a tracer is added into a slag chute, the time of slag in an area reaching a conveying point or an image sensor is determined in a mode of tracking the maximum proportion of the tracer, and at the moment, the historical images and the historical working areas are corresponding, so that the coal-rock interface distribution of the actual working area is obtained. After the above configuration, the offset of the angle of the cutterhead relative to the 0 point is obtained first, further, an image of the conveyor belt is obtained, and the image and the position offset of the conveyor belt correspondingly form the input of the coal rock identification network. The image information on the conveyor belt is used for acquiring the distribution of the current interface, and the TBM pulverizes the coal and rock in the tunneling process, but the tunnel ledge still keeps the information of part of the interfaces, so that the coal and rock distribution information of the corresponding interfaces can be acquired by combining the image information on the conveyor belt. The cutter head position and the machine head position are covered and cannot be directly acquired, so that the available area and the image of the conveyor belt are valuable only after corresponding. When the coal rock is identified, firstly, the image is grayed; if the filtering processing is carried out during image acquisition, the actual image and the gray value image approximation are obtained, and the image is processed, namely, the image pixels are replaced by values of v= (r+g+b)/3 according to the positions; based on a coal-rock classification model, classifying the pixels in the image after the cut-off linear stretching treatment to obtain a coal-rock classification result image, wherein the coal-rock classification result image comprises coal pixels and rock pixels; and respectively calculating the proportion of the coal area and the rock area, and determining the composition of the coal-rock interface by combining the weighing result of the belt. Further, the belt image is subjected to boundary tracking processing to obtain the size of the coal block with the size larger than the threshold value. And (3) classifying all pixels contained in the image after linear stretching one by adopting a K-means clustering algorithm (K-means) classification algorithm. The classification results are shown in white for rock formations and black for coal. And selecting sample points of coal and sample points of rock soil in the slag image as training samples. In the K-means clustering algorithm, the three targets except coal, rock and soil of a tunneling surface are considered, so that the number k=3 of clustering centers, and the initial clustering centers are randomly selected. In this way, the coal and rock distribution information in the conveyor belt or the chute can be determined, and the accurate information can not be obtained only by relying on computer vision due to the non-uniformity of material distribution, so that in order to better obtain the information of the interface, the applicant also obtains further information of the interface through physical analysis of the medium. Firstly, washing cutterhead slag, wherein the cutterhead slag can remove small-particle coal particles, rock particles and soil; on this basis, the slime composition of small particles can be further obtained through the slime water, namely the soil and coal content is obtained based on density, after detection, the slime water can be mixed onto the rock slag to reduce dust, and after the water-washed rock slag is analyzed to obtain the second characteristic of the interface, the rock slag in the chute can be mixed and discharged. Here, the first feature is used to obtain the original image information after the interface cutting, in particular the apparent area distribution of coal, rock and earth; the second feature is used for obtaining coal block information and rock block information after interface cutting and determining the proportion of the coal block information and the rock block information; the slime characteristics are used to determine the proportion of slime particles in the rock slag.
According to one embodiment of the invention, the images of the cutterhead rock are acquired by means of dual wavelengths, including visible wavelengths and infrared wavelengths, and after the images of the two wavelengths are acquired, the images are fused to obtain the images of the cutterhead rock.
According to the embodiment of the invention, because dust and color differences exist in the underground coal and rock identification image, the acquisition position of the cutterhead rock slag image can also comprise a light source, and the light source preferably comprises two different types of light sources, wherein the first light source is a visible light source and the second light source is an infrared light source; or where coal and rock have a better discrimination (including edge discrimination and class discrimination) at one wavelength, a single wavelength light source may be used. The visible light source is used for acquiring a first image, the infrared light source is used for acquiring a second image, and in order to avoid interference, a filter can be further provided for filtering light rays with other wavelengths when the image acquisition is carried out, namely, the first light source is provided, reflected light rays are acquired, incoherent light rays with wavelengths are filtered and removed, and image information with target wavelengths is acquired; based on this, information of the interfaces of the rock slag of two different wavelengths can be obtained. Similarly, a first infrared light source may be provided, and then image information of the target infrared light source wavelength may be obtained by obtaining reflected infrared light, filtering to remove incoherent wavelengths. Then, the obtained images can be fused to obtain the ratio of the coal area to the rock area in the images, and when one area corresponds to the rock area and the coal area at the same time, weight calculation can be performed according to the brightness value (calculated according to RGB value) or according to the overlapping area and the weight calculation is performed according to 1: the weight of 1 is calculated.
According to one embodiment of the invention, infrared light is used as a light source, reflected light rays of an infrared light wave band are collected, an image of rock residue after washing is obtained, the boundary of the rock residue image is determined according to the reflection degree of visible light in the image, the absorption of rock residue components to the infrared light is determined according to the brightness of each region in the image, the composition of each region is determined according to the absorption of the infrared light, and the second characteristic of the interface is determined.
According to one embodiment of the invention, particles with small particle sizes in coal rock after water washing are basically washed into coal slime water, the surfaces of coal and rock are exposed to form a detectable surface, and the surface of the coal and rock show absorption behavior of infrared light characteristics, so that the coal rock can show obvious absorption or reflection when an infrared incident light source is applied, namely the intensity of reflected light on the surface of an object with absorption behavior is obviously weakened, and the proportion of the coal rock in the image after water washing can be determined by monitoring the image.
According to one embodiment of the invention, an image of cutterhead rock slag obtained on a conveyor belt is taken as input, a rock slag image after water flushing and coal slime characteristics of an interface are taken as output, a migration learning training interface identification model is used, historical data of manual intervention operation is taken as a training set, and the output of the historical data or the interface identification model is taken as input of an interface operation network to construct a KNN neighbor classification model.
According to one embodiment of the present invention, by setting k=3, the distinction of the models is better optimized. For the present invention, k=2 can also be set, in which case a difficulty in distinguishing between slime and coal pieces may occur, but it may still provide a better distinction for large and small particle coal-rock interface distinction.
According to one embodiment of the invention, an image of cutterhead slag is taken as input, and a pre-trained model is used for acquiring coal slime characteristics of an interface and second characteristics of the interface; and taking the coal slime characteristics of the interface and the second characteristics of the interface as input of a KNN neighbor classification model to obtain classification of working conditions, when the Euclidean distance requirement is met, using manual operation of the classification as an operation recommended value, receiving evaluation of automatic operation of the interface by a user, and when the same classification of the working conditions is obtained, replacing the operation recommended value by the manual operation.
According to one embodiment of the invention, the method is based on the measured actual data and the output potential input as the input value of the working condition classification, and can increase the applicable working condition under the condition of low sample, thereby moderately reducing the burden of personnel. Other parameters which can be selected for KNN classification include shield machine cutterhead rotating speed, shield machine A group oil cylinder propelling force, shield machine B group oil cylinder propelling force, shield machine C group oil cylinder propelling force, shield machine D group oil cylinder propelling force, tunnel burial depth, underground water height, tunnel axis curvature, soil body compression modulus, soil body equivalent bearing capacity, shield machine cutterhead torque and shield machine propelling speed, manual evaluation on automatic driving operation can be provided, the evaluation can be used for labels, characteristics of collected cutterhead contact surfaces are used as input, a model can be constructed, training is carried out on the model, or incremental training is carried out on the basis of labels and actual working conditions so as to optimize a shield machine auxiliary driving model.
According to one embodiment of the invention, the characteristics of the contact surface of the cutterhead are used as input, the historical data of manual intervention operation are used as output of a training set, and a KNN neighbor classification model is constructed.
According to one embodiment of the invention, when the predicted operation and the characteristics at the cutter interface are considered unsuitable by an operator, the driving shield of the shield machine can be performed in a manual intervention mode, the characteristics of the large-disc interface corresponding to the process can be used as the input of a subsequent classification model, and corresponding data can be used as the output to train or increment train the network.
According to one embodiment of the invention, the characteristics of the contact surface of the cutterhead are used as the input of a KNN neighbor classification model, classification of working conditions is obtained, when the Euclidean distance requirement is met, the classification manual operation is used as an operation recommended value, and otherwise, the manual operation is used.
According to one embodiment of the invention, the division of the closest working conditions can be obtained by using Euclidean distance as a threshold value, and a corresponding driving model is adopted based on the closest working conditions; when the euclidean distance is greater than all the thresholds, then it is considered that there is no operation mode meeting the demand, and a manual mode should be employed.
Furthermore, to achieve the above object, the present invention provides a TBM contact interface identification system, including:
the cutterhead slag image acquisition module: acquiring cutterhead rock slag from a conveyor belt, and acquiring an image of the cutterhead rock slag;
a first feature acquisition module: determining a first feature of the interface based on the image of the cutterhead slag;
and a mud-water separation module: washing the cutterhead rock slag with water to separate mud from water;
and a second feature acquisition module: determining a second characteristic of the interface based on the sludge-water separated rock image;
the coal slime characteristic acquisition module: determining the coal slime characteristics of the interface based on the analysis result of the mud liquid after water flushing;
the characteristic determining module: the characteristics of the cutterhead contact surface are determined based on the first characteristics of the interface, the second characteristics of the interface, and the slime characteristics of the interface.
According to one embodiment of the invention, slag in a slag discharge groove in the shield machine cutterhead can drop downwards during the rotation of the shield machine cutterhead, and an upper slag slide machine and a lower slag slide machine can be arranged in the slag slide machine for collecting or discharging slag. Further, a conveyor belt is arranged in the mechanism and used for receiving the dregs discharged by the dreg discharging groove in the tray. The sensor can be arranged on the cutterhead and used for judging the position of the cutterhead, the position of the cutterhead can be any one point, when the cutterhead is at the 0 point moment, the cutterhead is used as the 0 point of the space, and the coal-rock interface information in the tunneling stroke can be obtained by analyzing the rock slag when the cutterhead returns to the 0 point moment. In the rotating process of the cutter head of the shield machine, slag and positions on a conveyor belt behind the cutter head of the shield machine are associated, but certain accumulation exists in a slag chute box, and the difference of positions obtained by images of the slag chute box and the conveyor belt is limited, the images obtained by the conveyor belt do not strictly belong to images of the current coal-rock interface, and certain incoherence exists, but the interface operated by the current position is associated with slag images on the conveyor belt, and the advancing speed of the TBM is in the cm/min level in the tunneling process, so that the TBM still has reference value for data analysis. In particular, if in a trip, interface information is obtained by collecting a portion of the coal rock in the trip for analysis. In order to implement the above-described procedure, the setting of the relevant apparatus will be explained first. Firstly, a cutter head of the shield machine comprises a position sensor, wherein the position sensor can be used for acquiring and judging the state of the cutter head, in particular the position of a 0 point relative to the cutter head;
and secondly, an image acquisition sensor, such as a camera or a video acquisition device, is arranged on a conveyor belt connected with the cutterhead slag chute and used for acquiring high-resolution images. The acquired images contain time stamps, the relation between the images and the actual tunneling process can be acquired in a tracing mode, for example, a tracer is added into a slag chute, the time of slag in an area reaching a conveying point or an image sensor is determined in a mode of tracking the maximum proportion of the tracer, and at the moment, the historical images and the historical working areas are corresponding, so that the coal-rock interface distribution of the actual working area is obtained. After the above configuration, the offset of the angle of the cutterhead relative to the 0 point is obtained first, further, an image of the conveyor belt is obtained, and the image and the position offset of the conveyor belt correspondingly form the input of the coal rock identification network. The image information on the conveyor belt is used for acquiring the distribution of the current interface, and the TBM pulverizes the coal and rock in the tunneling process, but the tunnel ledge still keeps the information of part of the interfaces, so that the coal and rock distribution information of the corresponding interfaces can be acquired by combining the image information on the conveyor belt. The cutter head position and the machine head position are covered and cannot be directly acquired, so that the available area and the image of the conveyor belt are valuable only after corresponding. When the coal rock is identified, firstly, the image is grayed; if the filtering processing is carried out during image acquisition, the actual image and the gray value image approximation are obtained, and the image is processed, namely, the image pixels are replaced by values of v= (r+g+b)/3 according to the positions; based on a coal-rock classification model, classifying the pixels in the image after the cut-off linear stretching treatment to obtain a coal-rock classification result image, wherein the coal-rock classification result image comprises coal pixels and rock pixels; and respectively calculating the proportion of the coal area and the rock area, and determining the composition of the coal-rock interface by combining the weighing result of the belt. Further, the belt image is subjected to boundary tracking processing to obtain the size of the coal block with the size larger than the threshold value. And (3) classifying all pixels contained in the image after linear stretching one by adopting a K-means clustering algorithm (K-means) classification algorithm. The classification results are shown in white for rock formations and black for coal. And selecting sample points of coal and sample points of rock soil in the slag image as training samples. In the K-means clustering algorithm, the three targets except coal, rock and soil of a tunneling surface are considered, so that the number k=3 of clustering centers, and the initial clustering centers are randomly selected. In this way, the coal and rock distribution information in the conveyor belt or the chute can be determined, and the accurate information can not be obtained only by relying on computer vision due to the non-uniformity of material distribution, so that in order to better obtain the information of the interface, the applicant also obtains further information of the interface through physical analysis of the medium. Firstly, washing cutterhead slag, wherein the cutterhead slag can remove small-particle coal particles, rock particles and soil; on this basis, the slime composition of small particles can be further obtained through the slime water, namely the soil and coal content is obtained based on density, after detection, the slime water can be mixed onto the rock slag to reduce dust, and after the water-washed rock slag is analyzed to obtain the second characteristic of the interface, the rock slag in the chute can be mixed and discharged. Here, the first feature is used to obtain the original image information after the interface cutting, in particular the apparent area distribution of coal, rock and earth; the second feature is used for obtaining coal block information and rock block information after interface cutting and determining the proportion of the coal block information and the rock block information; the slime characteristics are used to determine the proportion of slime particles in the rock slag.
According to one embodiment of the invention, the images of the cutterhead rock are acquired by means of dual wavelengths, including visible wavelengths and infrared wavelengths, and after the images of the two wavelengths are acquired, the images are fused to obtain the images of the cutterhead rock.
According to the embodiment of the invention, because dust and color differences exist in the underground coal and rock identification image, the acquisition position of the cutterhead rock slag image can also comprise a light source, and the light source preferably comprises two different types of light sources, wherein the first light source is a visible light source and the second light source is an infrared light source; or where coal and rock have a better discrimination (including edge discrimination and class discrimination) at one wavelength, a single wavelength light source may be used. The visible light source is used for acquiring a first image, the infrared light source is used for acquiring a second image, and in order to avoid interference, a filter can be further provided for filtering light rays with other wavelengths when the image acquisition is carried out, namely, the first light source is provided, reflected light rays are acquired, incoherent light rays with wavelengths are filtered and removed, and image information with target wavelengths is acquired; based on this, information of the interfaces of the rock slag of two different wavelengths can be obtained. Similarly, a first infrared light source may be provided, and then image information of the target infrared light source wavelength may be obtained by obtaining reflected infrared light, filtering to remove incoherent wavelengths. Then, the obtained images can be fused to obtain the ratio of the coal area to the rock area in the images, and when one area corresponds to the rock area and the coal area at the same time, weight calculation can be performed according to the brightness value (calculated according to RGB value) or according to the overlapping area and the weight calculation is performed according to 1: the weight of 1 is calculated.
According to one embodiment of the invention, infrared light is used as a light source, reflected light rays of an infrared light wave band are collected, an image of rock residue after washing is obtained, the boundary of the rock residue image is determined according to the reflection degree of visible light in the image, the absorption of rock residue components to the infrared light is determined according to the brightness of each region in the image, the composition of each region is determined according to the absorption of the infrared light, and the second characteristic of the interface is determined.
According to one embodiment of the invention, particles with small particle sizes in coal rock after water washing are basically washed into coal slime water, the surfaces of coal and rock are exposed to form a detectable surface, and the surface of the coal and rock show absorption behavior of infrared light characteristics, so that the coal rock can show obvious absorption or reflection when an infrared incident light source is applied, namely the intensity of reflected light on the surface of an object with absorption behavior is obviously weakened, and the proportion of the coal rock in the image after water washing can be determined by monitoring the image.
According to one embodiment of the invention, an image of cutterhead rock slag obtained on a conveyor belt is taken as input, a rock slag image after water flushing and coal slime characteristics of an interface are taken as output, a migration learning training interface identification model is used, historical data of manual intervention operation is taken as a training set, and the output of the historical data or the interface identification model is taken as input of an interface operation network to construct a KNN neighbor classification model.
According to one embodiment of the present invention, by setting k=3, the distinction of the models is better optimized. For the present invention, k=2 can also be set, in which case a difficulty in distinguishing between slime and coal pieces may occur, but it may still provide a better distinction for large and small particle coal-rock interface distinction.
According to one embodiment of the invention, an image of cutterhead slag is taken as input, and a pre-trained model is used for acquiring coal slime characteristics of an interface and second characteristics of the interface; and taking the coal slime characteristics of the interface and the second characteristics of the interface as input of a KNN neighbor classification model to obtain classification of working conditions, when the Euclidean distance requirement is met, using manual operation of the classification as an operation recommended value, receiving evaluation of automatic operation of the interface by a user, and when the same classification of the working conditions is obtained, replacing the operation recommended value by the manual operation.
According to one embodiment of the invention, the method is based on the measured actual data and the output potential input as the input value of the working condition classification, and can increase the applicable working condition under the condition of low sample, thereby moderately reducing the burden of personnel. Other parameters which can be selected for KNN classification include shield machine cutterhead rotating speed, shield machine A group oil cylinder propelling force, shield machine B group oil cylinder propelling force, shield machine C group oil cylinder propelling force, shield machine D group oil cylinder propelling force, tunnel burial depth, underground water height, tunnel axis curvature, soil body compression modulus, soil body equivalent bearing capacity, shield machine cutterhead torque and shield machine propelling speed, manual evaluation on automatic driving operation can be provided, the evaluation can be used for labels, characteristics of collected cutterhead contact surfaces are used as input, a model can be constructed, training is carried out on the model, or incremental training is carried out on the basis of labels and actual working conditions so as to optimize a shield machine auxiliary driving model.
According to one embodiment of the invention, the characteristics of the contact surface of the cutterhead are used as input, the historical data of manual intervention operation are used as output of a training set, and a KNN neighbor classification model is constructed.
According to one embodiment of the invention, when the predicted operation and the characteristics at the cutter interface are considered unsuitable by an operator, the driving shield of the shield machine can be performed in a manual intervention mode, the characteristics of the large-disc interface corresponding to the process can be used as the input of a subsequent classification model, and corresponding data can be used as the output to train or increment train the network.
According to one embodiment of the invention, the characteristics of the contact surface of the cutterhead are used as the input of a KNN neighbor classification model, classification of working conditions is obtained, when the Euclidean distance requirement is met, the classification manual operation is used as an operation recommended value, and otherwise, the manual operation is used.
According to one embodiment of the invention, the division of the closest working conditions can be obtained by using Euclidean distance as a threshold value, and a corresponding driving model is adopted based on the closest working conditions; when the euclidean distance is greater than all the thresholds, then it is considered that there is no operation mode meeting the demand, and a manual mode should be employed.
In order to achieve the above object, the present invention also provides an electronic device including: the TBM contact interface identification method comprises a processor, a memory and a computer program stored in the memory and capable of running on the processor, wherein the computer program is executed by the processor to realize the TBM contact interface identification method.
In order to achieve the above object, the present invention further provides a computer readable storage medium, on which a computer program is stored, which when executed by a processor implements a method for identifying a TBM contact interface as described above.
Based on the information, the method and the device have the beneficial effects that the interface information can be better obtained, the physical analysis is further carried out on the medium according to the interface information, and the property of the medium is accurately obtained and used for guiding the shield process.
Those of ordinary skill in the art will appreciate that the modules and algorithm steps described in connection with the embodiments disclosed herein can be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
It will be clearly understood by those skilled in the art that, for convenience and brevity of description, specific working procedures of the apparatus and device described above may refer to corresponding procedures in the foregoing method embodiments, which are not described herein again.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be additional divisions when actually implemented, for example, multiple modules or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or modules, which may be in electrical, mechanical, or other forms.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical modules, i.e., may be located in one place, or may be distributed over a plurality of network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the embodiment of the invention.
In addition, each functional module in the embodiment of the present invention may be integrated in one processing module, or each module may exist alone physically, or two or more modules may be integrated in one module.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method for energy saving signal transmission/reception of the various embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk, etc.
The foregoing description is only of the preferred embodiments of the present application and is presented as a description of the principles of the technology being utilized. It will be appreciated by persons skilled in the art that the scope of the invention referred to in this application is not limited to the specific combinations of features described above, but it is intended to cover other embodiments in which any combination of features described above or equivalents thereof is possible without departing from the spirit of the invention. Such as the above-described features and technical features having similar functions (but not limited to) disclosed in the present application are replaced with each other.
It should be understood that, the sequence numbers of the steps in the summary and the embodiments of the present invention do not necessarily mean the order of execution, and the execution order of the processes should be determined by the functions and the internal logic, and should not be construed as limiting the implementation process of the embodiments of the present invention.
Claims (7)
1. A method for identifying a TBM contact interface, comprising:
acquiring cutterhead rock slag from a conveyor belt, and acquiring an image of the cutterhead rock slag;
determining a first feature of the interface based on the image of the cutterhead slag;
washing the cutterhead rock slag with water to separate mud from water;
determining a second characteristic of the interface based on the sludge-water separated rock image;
determining the coal slime characteristics of the interface based on the analysis result of the mud liquid after water flushing;
the characteristics of the cutterhead contact surface are determined based on the first characteristics of the interface, the second characteristics of the interface, and the slime characteristics of the interface.
2. The method for identifying a TBM contact interface as defined in claim 1, wherein the image of the cutterhead slag is acquired by two wavelengths, the two wavelengths including a visible wavelength and an infrared wavelength, and the images are fused after the images of the two wavelengths are acquired.
3. The method for identifying a TBM contact interface according to claim 2, wherein infrared light is used as a light source, reflected light of an infrared light wave band is collected, an image of the rock residue after washing is obtained, a boundary of the rock residue image is determined according to a reflection degree of visible light in the image, absorption of the rock residue component to infrared light is determined according to brightness of each region in the image, composition of each region is determined according to absorption of infrared light, and a second characteristic of the interface is determined.
4. The method for identifying a TBM contact interface according to claim 3, wherein an image of cutterhead slag acquired on a conveyor belt is taken as input, a slag image after water flushing and a coal slime characteristic of the interface are taken as output, a transfer learning interface identification model is used, historical data of manual intervention operation is taken as a training set, and the output of the historical data or the interface identification model is taken as input of an interface operation network, so that a KNN neighbor classification model is constructed.
5. The method for identifying a TBM contact interface of claim 4 wherein the image of the cutterhead slag is used as input and a pre-trained model is used to obtain a slime characteristic of the interface and a second characteristic of the interface; and taking the coal slime characteristics of the interface and the second characteristics of the interface as input of a KNN neighbor classification model to obtain classification of working conditions, when the Euclidean distance requirement is met, using manual operation of the classification as an operation recommended value, receiving evaluation of automatic operation of the interface by a user, and when the same classification of the working conditions is obtained, replacing the operation recommended value by the manual operation.
6. The method for identifying a TBM contact interface according to claim 5, wherein characteristics of a cutter contact surface are used as input, and historical data of manual intervention operation is used as output of a training set to construct a KNN neighbor classification model.
7. The method for identifying a TBM contact interface according to claim 6, wherein characteristics of a contact surface of a cutterhead are used as input of a KNN neighbor classification model to obtain classification of working conditions, and when euclidean distance requirements are met, manual operation of the classification is used as an operation recommended value, otherwise manual operation is used.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311301965.XA CN117292197A (en) | 2023-10-10 | 2023-10-10 | TBM contact interface identification method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311301965.XA CN117292197A (en) | 2023-10-10 | 2023-10-10 | TBM contact interface identification method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117292197A true CN117292197A (en) | 2023-12-26 |
Family
ID=89244232
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311301965.XA Pending CN117292197A (en) | 2023-10-10 | 2023-10-10 | TBM contact interface identification method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117292197A (en) |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH1136785A (en) * | 1997-07-17 | 1999-02-09 | Kajima Corp | Photographic processing system for tbm cutting image |
CN111894615A (en) * | 2020-08-10 | 2020-11-06 | 中国电建集团铁路建设有限公司 | Air pressure balance shield machine and construction method |
CN112818952A (en) * | 2021-03-11 | 2021-05-18 | 中国科学院武汉岩土力学研究所 | Coal rock boundary recognition method and device and electronic equipment |
CN112990227A (en) * | 2021-02-08 | 2021-06-18 | 中国铁建重工集团股份有限公司 | Face geology detection method |
US20210248428A1 (en) * | 2020-02-11 | 2021-08-12 | Schlumberger Technology Corporation | Automated pixel-wise labeling of rock cuttings based on convolutional neural network-based edge detection |
US20220341831A1 (en) * | 2020-04-20 | 2022-10-27 | Shandong University | Horizontal jet-mechanical combined rock breaking test device and method |
CN115342873A (en) * | 2022-07-05 | 2022-11-15 | 中铁隧道局集团有限公司 | Large-diameter slurry shield slag metering method and slag state monitoring method |
CN115422822A (en) * | 2022-07-22 | 2022-12-02 | 北京交通大学 | Tunnel rock mass parameter prediction method and device |
-
2023
- 2023-10-10 CN CN202311301965.XA patent/CN117292197A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH1136785A (en) * | 1997-07-17 | 1999-02-09 | Kajima Corp | Photographic processing system for tbm cutting image |
US20210248428A1 (en) * | 2020-02-11 | 2021-08-12 | Schlumberger Technology Corporation | Automated pixel-wise labeling of rock cuttings based on convolutional neural network-based edge detection |
US20220341831A1 (en) * | 2020-04-20 | 2022-10-27 | Shandong University | Horizontal jet-mechanical combined rock breaking test device and method |
CN111894615A (en) * | 2020-08-10 | 2020-11-06 | 中国电建集团铁路建设有限公司 | Air pressure balance shield machine and construction method |
CN112990227A (en) * | 2021-02-08 | 2021-06-18 | 中国铁建重工集团股份有限公司 | Face geology detection method |
CN112818952A (en) * | 2021-03-11 | 2021-05-18 | 中国科学院武汉岩土力学研究所 | Coal rock boundary recognition method and device and electronic equipment |
CN115342873A (en) * | 2022-07-05 | 2022-11-15 | 中铁隧道局集团有限公司 | Large-diameter slurry shield slag metering method and slag state monitoring method |
CN115422822A (en) * | 2022-07-22 | 2022-12-02 | 北京交通大学 | Tunnel rock mass parameter prediction method and device |
Non-Patent Citations (2)
Title |
---|
HUANG XING, ET.AL: "A real-time prediction method for tunnel boring machine cutter-head torque using bidirectional long short-term memory networks optimized by multi-algorithm", 《JOURNAL OF ROCK MECHANICS AND GEOTECHNICAL ENGINEERING》, vol. 14, no. 3, 30 June 2022 (2022-06-30), pages 798 - 812 * |
付柯: "TBM掘进参数相关性分析及掘进速度预测", 《中国优秀硕士学位论文全文数据库 (工程科技Ⅱ辑)》, no. 3, 15 March 2019 (2019-03-15), pages 034 - 194 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102509087B (en) | Coal-rock identification method based on image gray level co-occurrence matrixes | |
CN111047555B (en) | Ore image granularity detection algorithm based on image processing technology | |
CN102496004B (en) | Coal-rock interface identifying method and system based on image | |
CN112990227B (en) | Face geology detection method | |
Schleifer et al. | FRAGSCAN: A tool to measure fragmentation of blasted rock | |
WO2003069561A2 (en) | Rock fragmentation analysis system | |
CA3050922C (en) | Detection and quantification of proppant for optimized fracture treatment design in in-fill and new wells | |
CN112127896B (en) | Automatic acquisition and analysis system and method for TBM excavation rock slag information | |
CN102930253A (en) | Coal and rock identification method based on image discrete multi-wavelet transform | |
CN113685188B (en) | TBM tunneling optimization method based on physical characteristics of rock slag | |
CN107862675A (en) | A kind of real-time vision detection method for electric bucket tooth missing | |
CN110519566B (en) | Belt motion state monitoring method based on video processing | |
CN202383714U (en) | Coal petrography interface identification system based on image | |
CN117292197A (en) | TBM contact interface identification method | |
CN114120218A (en) | River course floater monitoring method based on edge calculation | |
CN103942576A (en) | Method for identifying coal and rock through airspace multiscale random characteristics | |
CN110130921B (en) | TBM and belt feeder slag information acquisition device thereof | |
Wang | Rock particle image segmentation and systems | |
CN113569943B (en) | Slag piece large-block early warning method, system and device based on deep neural network | |
CN105350963B (en) | A kind of Coal-rock identification method learnt based on relativity measurement | |
CN116660263A (en) | Grouting effect evaluation method based on digital image processing | |
JP3485300B2 (en) | TBM excavation shear state measurement method and apparatus by image processing | |
CN112990169B (en) | Coal-rock interface identification method and coal cutting track determination method and device | |
CN117037135A (en) | TBM tunneling optimization method based on relevance information | |
Lemy et al. | Image analysis of drill core |
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 |