WO2020199290A1

WO2020199290A1 - Tbm-mounted advance geological forecast system and method based on recognition of lithology and unfavorable geological precursor characteristics

Info

Publication number: WO2020199290A1
Application number: PCT/CN2019/084654
Authority: WO
Inventors: 李术才; 许振浩; 谢辉辉; 余腾飞; 石恒; 王文扬; 潘东东; 林鹏
Original assignee: 山东大学
Priority date: 2019-04-04
Filing date: 2019-04-26
Publication date: 2020-10-08
Also published as: AU2019439936A1; AU2019439936B2; CN110043267A; CN110043267B

Abstract

The present disclosure provides a TBM-mounted advance geological forecast system and method based on recognition of lithology and unfavorable geological precursor characteristics. Said system comprises a base (2), the base (2) is provided on a TBM main conveyor belt (1), a telescopic mechanical gripper (4), a conveyor belt system (5), a crushing device (10), a control system (12) and a comprehensive data analysis platform (13) are sequentially provided on the base, and a drying mechanism (6), an image recognition device and an X-ray fluorescence analysis device (9) are sequentially provided above the conveyor belt system (5) to apply surface drying, lithology recognition and X-ray fluorescence analysis to a rock to be detected respectively. Information fusion and correction are performed by using a neural network according to an extracted mark image capable of reflecting unfavorable geological information, an unfavorable geological sensitive element and unfavorable geological mineral enrichment information in the rock, and then analysis is performed to obtain the change of lithology and unfavorable geological occurrence characteristics in front of a tunnel face, achieving an advance tunnel geological forecast based on smart recognition of lithology and unfavorable geological precursor characteristics.

Description

TBM-mounted advanced geological prediction system and method based on identification of lithology and poor geological precursor features

Technical field

The present disclosure relates to a TBM mounted advanced geological prediction system and method based on identification of lithology and poor geological precursor features.

Background technique

The statements in this section merely provide background information related to the present disclosure, and do not necessarily constitute prior art.

When the length-to-diameter ratio of a deep and long tunnel reaches or exceeds 600～1000, the TBM construction method should be used first in accordance with international common practice. The internationally recognized TBM construction method has fast excavation speed, low construction disturbance, high quality of hole formation, and comprehensive economic and social benefits. Higher advantage. However, the TBM construction method has poor adaptability to geological conditions. In the TBM construction, disasters such as water inrush, landslide, and large deformation are often encountered, resulting in major accidents such as jamming, damage, scrapping, and even casualties of the roadheader. In order to reduce the risk of the above-mentioned accidents during TBM construction, the most effective solution is to use advanced geological forecasting technology to detect the bad geological conditions in front of the master, and formulate reasonable disposal measures and construction plans in advance according to the geological conditions ahead to avoid and Prevent the occurrence of major disasters.

During the tunnel construction process, before the occurrence of fault fracture zones, karst caves, underground rivers, and silt zones, their own precursor signs generally appear. The appearance of these signs often indicates that the aforementioned bad geological bodies are approaching. However, most of the first-line construction workers in the tunnel lack good geological knowledge and cannot effectively identify the bad geological precursors in time, which leads to the occurrence of some disasters. At the same time, TBM construction requires high stratum lithology. In addition to paying attention to the possible unfavorable geology ahead, attention should also be paid to changes in stratum lithology along the excavation route in order to improve TBM construction parameters in time to ensure safe and rapid construction.

At present, there are many advanced geological prediction methods implemented in the construction process of TBM tunnels, mainly geophysical detection methods, such as seismic wave method, resistance method, infrared water detection, etc., and some TBM carrying equipment has also appeared. However, according to the inventor’s understanding, advanced geological forecasting is carried out from the perspective of geological analysis. In terms of identifying the bad geological precursors of disasters such as water inrush and mud inrush, the existing forecasting system and method cannot understand the lithology of surrounding rocks on the TBM tunnel in real time. , Changes in the types and contents of minerals and elements, and predict changes in geological conditions in front of the excavation.

Summary of the invention

In order to solve the above problems, the present disclosure proposes a TBM-mounted advanced geological prediction system and method based on the identification of lithology and poor geological precursor features. The present disclosure can understand the lithology, minerals and element types and types of surrounding rocks on the TBM tunnel in real time. The change of content, and predict the change of geological conditions in front of the excavation and possible bad geology.

According to some embodiments, the present disclosure adopts the following technical solutions:

A TBM-mounted advanced geological prediction system based on the identification of lithology and poor geological precursor features, including a base set on the TBM main conveyor belt, on which a retractable mechanical gripper, a conveyor belt system, and a crushing Equipment, control system and data comprehensive analysis platform, including:

The telescopic mechanical gripper is used to place the rock mass to be tested on the TBM main conveyor belt and place it on the conveyor belt system. A drying mechanism, an image recognition device, and an X-ray fluorescence analysis device are sequentially arranged above the conveyor belt system, respectively Apply surface drying, lithology identification and X-ray fluorescence analysis to the rock block to be tested;

The crushing device is arranged at the end of the conveyor belt system to crush the rock block to be tested, and the powder particles at the outlet of the crushing device can directly fall into the sample chamber of the X-ray diffraction analysis device set below;

The control system controls the actions of the retractable mechanical gripper, conveyor belt system, crushing device, image recognition device, X-ray fluorescence analysis device, and X-ray diffraction analysis device;

The data comprehensive analysis platform is configured to receive the collection results of the image recognition device, the X-ray fluorescence analysis device, and the X-ray diffraction analysis device, based on the sign images that can reflect the bad geological information, the bad geological sensitive elements, and the bad geological information in the extracted rocks. The enrichment information of geological minerals uses neural network for information fusion and correction, and analysis is performed to obtain the lithological changes and the occurrence characteristics of bad geology in front of the head, and realize the advanced geological prediction of the tunnel based on intelligent identification of lithology and bad geological precursors.

The present disclosure is mounted on the TBM, and can analyze the changes of lithology and adverse geological precursor characteristics during the TBM tunneling work in real time. Based on the collected rock image information, the convolutional neural network is used to extract and characterize the color, texture, and shape of the rock through the convolutional layer and the maximum pooling layer; at the same time, combined with spectral analysis technology, it can identify minerals and minerals in the rock. The change of element content establishes a quantitative characterization relationship between sensitive elements and characteristic minerals and geological precursors of adverse geologically affected areas.

At the same time, the present disclosure uses the base to be set on the TBM, and uses a separate conveyor belt to perform analysis work, does not interfere with the TBM's own work, and realizes the integration and integration of tunneling, analysis, and forecasting.

As a further limitation, a protective shed is provided on the base. The protective shed can effectively prevent other interference factors.

The protective shed is welded and located above the base, supported by four angle steels as the roof. The roof is arched and made of stainless steel plate, which can prevent the falling rock and water seepage of the tunnel vault from damaging the instrument parts on the base. The two sides of the protective shed are covered by toughened glass panels, and are fixed on the supporting angle steel with fastening bolts.

As a further limitation, the pedestal is composed of angle steel, the outline of the pedestal is rectangular, and it is welded above the TBM conveyor belt near the main control room to provide support for other parts of the system.

As a further limitation, the telescopic mechanical gripper includes an electro-hydraulic column, a support rod, an electro-hydraulic arm, a mechanical arm, an electro-hydraulic rod, a rocker, a connecting rod, and a grab. The electro-hydraulic column is vertically welded to the base The upper part is used to install the electro-hydraulic arm and support rod, which can be extended up and down to control the lifting of the entire mechanical gripper device. One end of the support rod is vertically fixedly connected to the electro-hydraulic column, and one end is connected to the mechanical arm with a pin shaft. One end of the electro-hydraulic arm is connected to the electro-hydraulic column, and one end is connected to the mechanical arm with a pin. There are two grabs and they are symmetrically mounted on the front end of the mechanical arm with pins. At the same time, bucket teeth are installed at the front of the grab , Corresponding to the grab, there are two electro-hydraulic telescopic rods, which are connected with a rocker and a connecting rod to control the opening and closing of the grab.

As a further limitation, a laser profile measuring instrument is fixedly installed at the front end of the robotic arm and the grab bucket to measure the size of the rock mass on the TBM conveyor belt. When the measured rock mass meets the set range When, the signal is transmitted to the control system to control the mechanical gripper to grab the rock block.

As a further limitation, a high-pressure nozzle is installed in the grab, and the high-pressure nozzle is connected to an external water pump through a water pipe. When the grab grabs the rock block, the high-pressure nozzle will automatically spray water to wash the grabbed rock block. To ensure that the surface of the grabbed rock block is free from other slag powder pollution.

As a further limitation, the conveyor belt system is installed in the protective shed on the upper part of the base and includes a conveyor belt, a roller, a servo motor and a displacement sensor. There are at least two rollers, the conveyor belt is wound on the roller, and the servo motor The roller is driven to drive the movement of the conveyor belt. The displacement sensor is used to measure the transmission displacement of the rock block to be measured on the conveyor belt. The displacement sensor and servo motor are both communicated with the control system.

As a further limitation, the drying mechanism is a dryer, which air-drys the picked rock blocks to be tested. When the conveyor belt system transfers the rock blocks to be tested directly below the dryer, the control system starts to control the air dryer Work is carried out, and the hot air is released to the surface of the rock block to be tested. After a certain period of time, the dryer stops working and the surface of the rock block to be tested is dried.

As a further limitation, the image recognition device includes an image capture device, a lighting device, and an image recognition system. The image capture device is a digital camera configured to take high-definition images of the rock mass to be tested. The lighting device includes several industrial lighting devices. The light source is placed next to the image acquisition device. The image recognition system uses deep learning technology and uses the convolutional neural network intelligent recognition model to train the rock images in the existing rock specimen library, and compare it with the established rock specimen library image for information fusion and The analysis can finally give a preliminary lithology recognition result. When the image acquisition device captures an image of the rock block to be tested, it will transmit the acquired image to the data comprehensive analysis platform.

As a further limitation, the X-ray fluorescence analysis device includes a drive motor, a chassis, a connecting rod, an electro-hydraulic rod, a pin, a spherical hinge, and an X-ray fluorescence analyzer. The drive motor is connected to the chassis to control the rotation of the entire chassis. The chassis and the connecting rod, the chassis and the electro-hydraulic rod are connected by a pin shaft, the connecting rod and the electro-hydraulic rod are fixedly connected, and the X-ray fluorescence analyzer is connected with the other end of the connecting rod by a ball hinge to control The X-ray fluorescence analyzer detects the rock mass under test from different angles. The X-ray fluorescence analyzer emits X-rays and irradiates the rock mass to be tested, and receives the secondary characteristic X-rays generated. The types and contents of elements in the rock block to be tested are obtained, and the results are transmitted to the data comprehensive analysis platform.

As a further limitation, the crushing device includes a motor, a crushing chamber, a crushing shaft, a crushing knife, a feed port, a compartment, a discharge port, a screen, a valve, and a blower. The motor is installed on the upper part of the entire crushing device and is responsible for providing The power of high-speed rotation, the shaft of the servo motor is connected with a crushing shaft, the front end of the crushing shaft is equipped with a crushing knife, and the crushing shaft and the crushing knife are set in the crushing cabin, and the feeding port and conveyor belt system are provided on one side of the crushing cabin. The sample on the upper part can automatically fall to the feed inlet to enter the crushing chamber. A layer of screen is installed at the lower part of the crushing chamber. During the crushing process, the particles filtered by the screen will fall into the bucket set below the crushing chamber. There are two discharge ports at the lower part of the bucket, one is the powder discharge port and the other is the waste discharge port. The powder particles at the powder outlet can directly fall into the sample compartment of the X-ray diffraction analysis device below. The waste at the outlet directly falls into the main TBM conveyor belt below, and there is a valve at each of the two discharge ports, and the blower is installed in the upper part of the crushing cabin.

As a further limitation, the X-ray diffraction analysis device includes an X-ray diffraction analyzer and a sample compartment, the sample compartment has upper and lower openings, and a hair dryer is installed above the upper opening, and both the upper and lower openings are switched on and off. Control the opening and closing of the corresponding opening.

The forecast method based on the above system includes the following steps:

Start the system to obtain the rock image information, the type and content information of the elements in the rock, and the type and content information of the minerals in the rock collected during the TBM tunneling process;

Based on the collected rock image information, the convolutional neural network is used to extract and characterize the color, texture and shape of the rock through the convolutional layer and the maximum pooling layer;

The extracted and characterized rock image feature information is fused with the collected rock element type and content, rock mineral type and content, and input into the rock image recognition and classification model that has been trained in the intelligent rock recognition system. Analyze and compare, and finally give the rock name, bad geological mark and accuracy rate of the rock block to be tested;

Conduct data comparison and analysis on the types and contents of a large number of collected rock elements, apply data mining technology to extract the unfavorable geologically sensitive elements, and quantitatively study their changing laws;

Conduct data comparison and analysis on the types and contents of a large number of collected rocks and minerals, apply data mining techniques to extract minerals with poor geological characteristics, and quantitatively study their changing laws;

After continuous measurement of rock samples with a certain excavation distance, the extracted rock can reflect the sign images of bad geological information, bad geological sensitive elements and bad geological mineral enrichment information using neural network for information fusion and correction, and based on data mining technology Comparing and analyzing the established quantitative characterization relational database of adverse geological precursors characteristics, so as to give out the lithological changes and adverse geological occurrence characteristics in front of the tunnels, and realize the advanced geological prediction of tunnels based on the intelligent identification of lithology and adverse geological precursors. .

Compared with the prior art, the beneficial effects of the present disclosure are:

1) The present disclosure controls the sequence actions of the retractable mechanical gripper, conveyor belt system, crushing device, image recognition device, X-ray fluorescence analysis device and X-ray diffraction analysis device through the control system, which can realize full automation and continuity of work .

2) The present disclosure integrates image recognition technology and spectral analysis technology, which can accurately and quickly identify lithology, use spectral analysis technology to identify changes in minerals and element content in rocks, and establish sensitive elements and characteristic minerals and adverse geological influence areas Quantitative characterization relationship between geological precursor features.

3) The present disclosure is equipped with TBM and uses a separate conveyor belt to perform analysis work. It does not interfere with the TBM's own work, and can analyze the changes in lithology and poor geological precursor characteristics during the TBM tunneling work in real time.

5) The present disclosure can use neural network to integrate the sign images of bad geological information, sensitive elements of bad geology, and enrichment information of bad geological minerals, and then compare it with the quantitative characterization relation database of bad geological precursors established based on data mining technology Comparison and analysis can give out the lithological changes in front of the tunnel and the occurrence characteristics of bad geology, so as to finally realize the advanced geological prediction of the tunnel based on the intelligent identification of lithology and bad geological precursors.

Description of the drawings

The drawings of the specification forming a part of the application are used to provide a further understanding of the application, and the exemplary embodiments and descriptions of the application are used to explain the application, and do not constitute an improper limitation of the application.

Figure 1 is a schematic diagram of the present disclosure;

Figure 2 is a structural diagram of a retractable mechanical gripper;

Figure 3 is a three-dimensional structure diagram of the grab;

Figure 4 is a diagram of an X-ray fluorescence analysis device;

Figure 5 is a structural diagram of the grinding device;

Figure 6 is a schematic diagram of the structure of an X-ray diffraction analyzer;

Among them: 1. TBM main conveyor belt; 2. base; 3. protective shed; 4. retractable mechanical gripper; 5. conveyor belt system; 6. dryer; 7. image acquisition device; 8. lighting device; 9.X X-ray fluorescence analysis device; 10. Crushing device; 11. X-ray diffraction analysis device; 12. Control system; 13. Data comprehensive analysis platform; 14. TBM main control room; 15. Electro-hydraulic column; 16. Electro-hydraulic arm; 17 Robotic arm; 18. Support rod; 19. Electro-hydraulic rod; 20. Rocker; 21. Connecting rod; 22. Pin shaft; 23. Grab; 24. Bucket teeth 25. High pressure nozzle; 26. Laser profile measuring instrument 27. Servo motor; 28. Pin shaft; 29. Chassis 30. Electro-hydraulic rod; 31. Connecting rod; 32. Spherical hinge; 33. X-ray fluorescence analyzer; 34. Servo motor; 35. Crushing shaft; 36. Feed port; 37. Crushing chamber; 38. Crushing knife; 39. Screen; 40. Bucket; 41. Valve; 42. Powder discharge port; 43. Waste discharge port; 44. Hair dryer.

detailed description:

The disclosure will be further described below in conjunction with the drawings and embodiments.

It should be pointed out that the following detailed descriptions are all illustrative and are intended to provide further explanations for the application. Unless otherwise indicated, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this application belongs.

It should be noted that the terms used here are only for describing specific embodiments, and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should also be understood that when the terms "comprising" and/or "including" are used in this specification, they indicate There are features, steps, operations, devices, components, and/or combinations thereof.

In the present disclosure, terms such as "upper", "lower", "left", "right", "front", "rear", "vertical", "horizontal", "side", "bottom", etc. indicate The azimuth or positional relationship is based on the azimuth or positional relationship shown in the drawings, and is only a relationship term determined to facilitate the description of the structural relationship of each component or element in the present disclosure. It does not specifically refer to any component or element in the present disclosure, and cannot be understood as a reference to the present disclosure. Disclosure restrictions.

In the present disclosure, terms such as "fixed connection", "connected", "connected", etc. should be understood in a broad sense, indicating that it can be a fixed connection, an integral connection or a detachable connection; it can be directly connected, or through an intermediate connection. The medium is indirectly connected. For relevant scientific research or technical personnel in the field, the specific meanings of the above terms in the present disclosure can be determined according to specific circumstances, and cannot be understood as a limitation of the present disclosure.

A TBM-mounted advanced geological prediction system and method based on the intelligent recognition of lithology and poor geological precursor features, including a base, a protective shed, a retractable mechanical gripper, a conveyor belt system, and a drying Dryer, image acquisition device, X-ray fluorescence analysis device, crushing device, X-ray diffraction analysis device, control system and data comprehensive analysis platform. The base mainly provides support for other components of the system. The protective shed is welded and located above the base to prevent rockfall and water seepage on the tunnel vault from damaging the instrument parts on the base. The retractable mechanical gripper can realize the grabbing of the rock block to be tested and the washing of the rock block to be tested. The dryer is responsible for drying the washed rock block to be tested, and then the image acquisition device completes the collection of the image of the rock block to be tested, and transmits the image information to the rock intelligent identification system in the data comprehensive analysis platform. The X-ray fluorescence analysis device can realize the identification of the type and content of elements in the rock block, and the X-ray diffraction device can realize the diffraction analysis of the powder processed by the crushing device, and give the result of the mineral type and content in the rock block to be tested. Finally, these test results are transmitted to the data comprehensive analysis system for information fusion and correction, and the recognition results of lithological features are given.

After continuously measuring rock samples with a certain excavation distance, the data comprehensive analysis platform will use neural network to perform information fusion and correction on the extracted rocks that can reflect the sign images of bad geological information, bad geological sensitive elements, and bad geological mineral enrichment information. Then compare and analyze the relational database based on the quantitative characterization of bad geological precursor features established based on data mining technology, so as to give the lithological changes and the occurrence characteristics of bad geological occurrence in front of the face, and finally realize the characteristics based on lithology and bad geological precursors Advanced geological forecast of tunnels with intelligent identification. At the same time, with the continuous advancement of TBM excavation work, the information in the data comprehensive analysis platform is continuously improved, and the final lithology identification results and the tunnel advanced geological forecast results will be continuously revised and become more accurate.

Specifically, a TBM-mounted advanced geological prediction system based on the intelligent recognition of lithology and poor geological precursor features, as shown in Figure 1, specifically includes a base 2, a protective shed 3. Shrinkable mechanical gripper 4, conveyor belt system 5, dryer 6, image acquisition device 7, lighting device 8, X-ray fluorescence analysis device 9, crushing device 10, X-ray diffraction analysis device 11, control system 12 and comprehensive data analysis Platform 13.

The base 2 is mainly composed of angle steel, the base is rectangular in outline, and is welded above the TBM conveyor belt near the main control room to provide support for other parts of the system.

The protective shed 3 is welded and located above the base, and is supported by four angle steels as the roof. The roof is arched and made of stainless steel, which can prevent falling rocks and water seepage on the tunnel vault from damaging the instrument parts on the base. The front and back of the protective shed 3 are covered by toughened glass plates, and are fixed on the supporting angle steel with fastening bolts.

As shown in Figure 2, the retractable mechanical gripper 4 is mainly composed of an electro-hydraulic column 15, a support rod 18, an electro-hydraulic arm 16, a mechanical arm 17, an electro-hydraulic rod 19, a rocker 20, a connecting rod 21, and a grab 23 . The electro-hydraulic column 15 is vertically welded on the base 2 to install the electro-hydraulic arm 16 and the support rod 18, which can be extended up and down to control the lifting of the entire mechanical gripper device. One end of the support rod 18 is vertically and fixedly connected to the electro-hydraulic column 15 and one end is connected to the mechanical arm 17 by a pin. One end of the electro-hydraulic arm 16 is connected to the electro-hydraulic column 15 and one end is connected to the mechanical arm 17 by a pin. As shown in Figure 3, bucket teeth 24 are installed at the front end of the grab. In addition to ensuring that the rock block is easy to grab, when the rock block is washed after the grab is closed, the washed water and slag can flow out. A laser profile measuring instrument 25 is fixedly installed at the front end of the robotic arm connected to the grab 23 to measure the size of the rock mass on the TBM conveyor belt. When the measured rock mass meets the set range, the The signal is transmitted to the control system to control the mechanical grab to grab the rock block. A high-pressure nozzle 26 is also installed in the grab. The high-pressure nozzle is connected to an external water pump through a water pipe. When the grab grabs a rock block, the high-pressure nozzle 26 will automatically spray water to wash the grabbed rock block to ensure grabbing. There is no pollution from other slag powder on the surface of the rock block.

The conveyor belt system 5 is installed in the protective shed on the upper part of the base, and includes a conveyor belt, a roller, a servo motor and a displacement sensor. The dryer 6 is placed at a height of 5 cm above the conveyor belt of the system, and is mainly used for air-drying the grabbed rock blocks to be tested. The servo motor is responsible for driving the movement of the conveyor belt, and the displacement sensor is used to measure the transmission displacement of the rock block to be measured on the conveyor belt. The two are combined with the control system to realize real-time control of the rock block displacement to be measured.

The dryer 6 is placed at a height of 5 cm above the conveyor belt of the system, and is mainly used for air-drying the grabbed rock blocks to be tested. When the conveyor belt system transports the rock block to be tested directly below the dryer 6, the control system starts to control the air dryer to work, and release hot air to blow to the surface of the rock block to be tested. After a certain period of time, the dryer stops working. The surface of the rock block is dried. The conveyor belt system starts to work again and continues to transport the rock blocks to be tested forward.

The image recognition device is mainly composed of an image acquisition device 7, a lighting device 8, and an image recognition system. The image acquisition device 7 is a high-definition digital camera, which is mainly responsible for shooting high-definition images of the rock mass to be tested and transmitting the information to the data comprehensive analysis platform 13. The lighting device 8 is mainly composed of two industrial lighting light sources, which are arranged on both sides of the image acquisition device. In this embodiment, the image recognition system is a rock intelligent recognition software developed based on the Windows system. The software applies deep learning technology and uses a convolutional neural network intelligent recognition model to train the rock images in the existing rock specimen library, and The established rock specimen library image comparison, information fusion and analysis, can finally give a preliminary lithology recognition results. When the image acquisition device captures an image of the rock block to be tested, it will transmit the collected image to the rock intelligent identification system. At the same time, the conveyor belt continues to work to transfer the rock block to be tested to the X-ray fluorescence analysis device.

As shown in Figure 4, the X-ray fluorescence analysis device 9 is mainly composed of a servo motor 27, a chassis 29, a connecting rod 31, an electro-hydraulic rod 30, a pin 28, a spherical hinge 32, and an X-ray fluorescence analyzer 33. The servo motor 27 and The chassis 29 is connected to control the rotation of the entire chassis 29. The chassis 29 and the connecting rod 31, the chassis 29 and the electro-hydraulic rod 30 are connected by a pin, the connecting rod 31 and the electro-hydraulic rod 30 are fixedly connected, and the X-ray fluorescence analyzer and the other end of the connecting rod 31 are connected with a ball hinge 32-phase connection, which can control the X-ray fluorescence analyzer to detect the rock mass under test from different angles. The X-ray fluorescence analyzer emits X-rays and irradiates them on the rock block to be tested, and receives the secondary characteristic X-rays generated, gives the element types and content in the rock block to be tested, and transmits the results to the comprehensive data analysis Platform 13.

As shown in Figure 5, the crushing device 10 is composed of a crusher, which mainly includes a servo motor 34, a crushing cabin 37, a crushing shaft 35, a crushing knife 38, a feed port 36, a bucket 40, and discharge

ports

42 and 43, The screen 39, the valve 41 and the blower 44 are composed. The servo motor 34 is installed in the upper part of the whole device and is responsible for providing high-speed rotation power. A pulverizing shaft 35 is connected to the shaft of the servo motor 34, a pulverizing knife 38 is installed at the front end of the pulverizing shaft 35, and both the pulverizing shaft 35 and the pulverizing knife 38 are in the pulverizing cabin 37. A feed port 36 is provided on one side of the crushing cabin 37, and the sample on the conveyor belt system can automatically fall to the feed port 36 to enter the crushing cabin 37. A thicker 200-mesh screen 39 is installed at the lower part of the crushing cabin 37, and particles smaller than 200-mesh are automatically transferred into the lower bucket 40 during the crushing process. There are two

discharge ports

42, 43 at the lower part of the bucket 40, one is the powder discharge port and the other is the waste discharge port. The powder particles at the powder outlet can directly fall into the sample compartment of the X-ray diffraction analyzer below. In the middle, the waste at the waste outlet directly falls into the main TBM conveyor belt below, and there is a valve at each of the two outlets. The blower 44 is installed in the upper part of the crushing cabin 37, and is mainly responsible for cleaning the residual waste in the crushing cabin 37 and the screen 39. The rock block to be tested can be ground into powder to provide test materials for the X-ray diffraction device 11.

The X-ray diffraction analysis device 11 is mainly composed of an X-ray diffraction analyzer, as shown in FIG. 6, and its main structure includes an X-ray tube, an X-ray detector, a goniometer, a sample chamber, a data acquisition system, and a data processing analysis system. The sample compartment has upper and lower openings, and a small blower is installed above the upper opening. The upper opening is opened when loading samples, and the upper and lower openings are closed when working. After the test is completed, the lower opening is opened, and the blower starts working and blows the rock dust in the sample chamber. Sweep onto the TBM main conveyor belt and transport away.

The control system 12 mainly receives control information from the retractable mechanical gripper 4, the conveyor belt system 5, the dryer 6, the image acquisition device 7, the lighting device 8, the X-ray fluorescence analysis device 9 and the crushing device 10, and controls the above devices Work and stop.

The data comprehensive analysis platform 13 includes a lithology intelligent recognition system based on image and spectroscopy technology, a bad geological precursor feature database, and an advanced geological prediction system. The platform 13 is installed in the main control room 14 of the TBM and is connected with the above devices by data connection lines.

Based on the intelligent recognition of lithology and poor geological precursor features, the TBM-mounted advanced geological prediction system based on the recognition of lithology and poor geological precursor features. The specific implementation process includes the following steps:

(1) After the TBM starts to dig and the cut slag is conveyed under the system device by the TBM main conveyor belt 1, press the working button of the entire system in the TBM main control room 14, and the system starts to work.

(2) First, the retractable mechanical gripper 4 starts to work, the front end of the mechanical arm 17, and the laser profile measuring instrument installed at the junction of the grab 23 to measure the size of the rock mass on the TBM conveyor belt, and then select a piece that meets the system requirements Set the size range of rock blocks to grab. After grabbing the rock block, the high-pressure nozzle 25 will automatically spray water to wash the rock block in the grab bucket. After the flushing is completed, the electro-hydraulic arm 16 and electro-hydraulic rod 19 begin to extend and move and place the rock block to be tested on the conveyor belt On system 5.

(3) When the rock block is moved directly below the dryer 6, the conveyor belt system 5 stops working, and the control system 12 starts to control the air dryer 6 to work, blowing out hot air. After 2 minutes, the surface of the rock block to be tested is After being dried, the dryer stops working.

(4) When the conveyor belt system 5 starts to work again and conveys the rock mass to be measured to the bottom of the image acquisition device 7. The conveyor belt system 5 stops working again, the lighting device 8 starts to light up, the image acquisition device continuously takes pictures of the rock block to be tested, and transmits the collected pictures to the lithology intelligent recognition system of the data comprehensive analysis platform 13.

(5) The conveyor belt system 5 continues to work, transports the rock mass to be tested under the X-ray fluorescence analysis device 9 and then stops working again. At this time, the X-ray fluorescence analysis device 9 starts to work, firstly controlled by the servo motor 27 and the electro-hydraulic rod 30 The irradiation direction of X-ray fluorescence analysis is to irradiate different parts of the rock block under test multiple times, and finally give the element type and content in the rock block to be tested, and transmit the result to the data comprehensive analysis platform 13.

(6) After the X-ray fluorescence analysis of the rock mass to be tested is completed, the conveyor belt system 5 continues to work, and the rock mass to be tested will be fed into the feed port of the crushing device 10, and then the crusher will start working, firstly the powder discharge port 42 The valve of the waste discharge port 43 is closed. During the crushing process, the powder particles falling from the screen 39 are automatically transferred into the sample compartment of the X-ray diffraction analyzer. When the powder in the sample compartment reaches a certain amount, the pulverizer stops working, the valve of the powder outlet 42 is closed, the valve of the waste outlet 43 is opened, the screen will be drawn out, and the blower 44 will start purging at the same time. The remaining waste is purged from the waste discharge port to the main TBM conveyor 1 below.

(7) After the rock dust falling into the sample chamber reaches a certain amount, the opening of the sample chamber is closed, and the X-ray diffraction analysis device 11 starts to work, and then the mineral types and contents in the rock block to be tested are given, and the result is transmitted To the data comprehensive analysis platform 13. After the test is completed, the lower opening of the sample compartment is opened, and the blower starts to work and blows the rock dust in the sample compartment to the TBM main conveyor belt 1 for transportation.

(8) The data comprehensive analysis platform 13 provides the lithological characteristics according to the image intelligent recognition system, the type and content of each element in the rock sample given by the X-ray fluorescence analysis device, and the mineral types and the mineral types given by the X-ray diffraction analysis device 11. The content result is compared with the established rock image, element and mineral standard database, and finally the accurate lithology identification result of the rock block to be tested is given.

(9) After continuous measurement of rock samples with a certain excavation distance, the extracted rock can reflect the sign images of bad geological information, bad geological sensitive elements and bad geological mineral enrichment information using neural network for information fusion and correction, and then Comparing and analyzing the relational database based on the quantitative characterization of the bad geological precursor features established based on data mining technology, to give out the lithological changes and the occurrence of bad geological features in the front of the face, and finally realize the intelligence based on the lithology and the bad geological precursor features Advance geological forecast of identified tunnels.

The above descriptions are only preferred embodiments of the application, and are not used to limit the application. For those skilled in the art, the application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this application shall be included in the protection scope of this application.

Although the specific embodiments of the present disclosure are described above in conjunction with the accompanying drawings, they do not limit the scope of protection of the present disclosure. Those skilled in the art should understand that on the basis of the technical solutions of the present disclosure, those skilled in the art do not need to make creative efforts. Various modifications or variations that can be made are still within the protection scope of the present disclosure.

Claims

A TBM-mounted advanced geological prediction system based on the identification of lithology and poor geological precursor features, which is characterized by comprising a base, which is arranged on the TBM main conveyor belt, on which a retractable mechanical gripper, Conveyor belt system, crushing device, control system and data comprehensive analysis platform, including:

The telescopic mechanical gripper is used to place the rock mass to be tested on the TBM main conveyor belt and place it on the conveyor belt system. A drying mechanism, an image recognition device, and an X-ray fluorescence analysis device are sequentially arranged above the conveyor belt system, respectively Apply surface drying, lithology identification and X-ray fluorescence analysis to the rock block to be tested;

The crushing device is arranged at the end of the conveyor belt system to crush the rock block to be tested, and the powder particles at the outlet of the crushing device can directly fall into the sample chamber of the X-ray diffraction analysis device set below;

The control system controls the actions of the retractable mechanical gripper, conveyor belt system, crushing device, image recognition device, X-ray fluorescence analysis device, and X-ray diffraction analysis device;

The data comprehensive analysis platform is configured to receive the collection results of the image recognition device, the X-ray fluorescence analysis device, and the X-ray diffraction analysis device, based on the sign images that can reflect the bad geological information, the bad geological sensitive elements, and the bad geological information in the extracted rocks. The enrichment information of geological minerals uses neural network for information fusion and correction, and analysis is performed to obtain the lithological changes and the occurrence characteristics of bad geology in front of the head, and realize the advanced geological prediction of the tunnel based on intelligent identification of lithology and bad geological precursors.
A TBM-mounted advanced geological prediction system based on identification of lithology and poor geological precursor features according to claim 1, characterized in that: a protective shed is arranged on the base, and the protective shed is welded and located on the base. The roof is supported by angle steel, the roof shape is arched, and the material is stainless steel plate. The two sides of the protective shed are covered by tempered glass plates, and they are fixed on the supporting angle steel with fastening bolts.
A TBM-mounted advanced geological prediction system based on the identification of lithology and poor geological precursor features as claimed in claim 1, characterized in that: the base is composed of angle steel, the base contour is rectangular, and it is welded to the main control room Above the nearby TBM conveyor belt, it mainly provides support for other parts of the system.
The TBM-mounted advanced geological prediction system based on the identification of lithology and poor geological precursor features according to claim 1, characterized in that: the retractable mechanical gripper includes an electro-hydraulic column, a support rod, an electro-hydraulic Arms, mechanical arms, electro-hydraulic rods, rockers, connecting rods, and grabs. The electro-hydraulic column is welded vertically on the base to install the electro-hydraulic arm and support rod. It can be extended up and down to control the entire mechanical gripper device. One end of the support rod is vertically fixedly connected to the electro-hydraulic column, one end is connected to the pin of the mechanical arm, one end of the electro-hydraulic arm is connected to the electro-hydraulic column, and one end is connected to the pin of the mechanical arm. There are two grab buckets, which are symmetrically installed on the front end of the robotic arm with a pin axis. At the same time, bucket teeth are installed at the front end of the grab bucket. Corresponding to the grab bucket, there are two electro-hydraulic telescopic rods, which are connected by a rocker and a connecting rod. , Used to control the opening and closing of the grab;

Or, a laser profile measuring instrument is fixedly installed at the front end of the robotic arm where the grab bucket is connected to measure the size of the rock mass on the TBM conveyor belt. When the measured rock mass meets the set range, the The signal is transmitted to the control system to control the mechanical grab to grab the rock block;

Or, a high-pressure nozzle is installed in the grab, and the high-pressure nozzle is connected to an external water pump through a water pipe. After the grab grabs the rock block, the high-pressure nozzle will automatically spray water to wash the grabbed rock block to ensure that it is grabbed. There is no pollution from other slag powder on the surface of the rock block.
A TBM-mounted advanced geological prediction system based on the identification of lithology and poor geological precursor features according to claim 1, wherein the conveyor belt system is installed in a protective shed on the upper part of the base and includes a conveyor belt, a roller, Servo motors and displacement sensors, the rollers have at least two, the conveyor belt is wound on the rollers, the servo motor drives the rollers to drive the movement of the conveyor belt, and the displacement sensors are used to measure the conveying displacement of the rock mass on the conveyor belt, Both the displacement sensor and the servo motor communicate with the control system.
A TBM-mounted advanced geological prediction system based on the identification of lithology and poor geological precursor features according to claim 1, wherein the image recognition device includes an image acquisition device, a lighting device, and an image recognition system. The image acquisition device is a digital camera, which is configured to take high-definition images of the rock mass to be tested. The lighting device includes several industrial lighting sources and is placed next to the image acquisition device. The image recognition system uses deep learning technology and uses convolutional neural network for intelligent recognition. The model trains the rock images in the existing rock specimen library, compares them with the established rock specimen library images, performs information fusion and analysis, and finally gives a preliminary lithology recognition result. When the image acquisition device captures the image of the rock block to be tested At the time, the collected images will be transmitted to the data comprehensive analysis platform.
A TBM-mounted advanced geological prediction system based on the identification of lithology and poor geological precursor features according to claim 1, wherein the X-ray fluorescence analysis device includes a drive motor, a chassis, a connecting rod, and an electro-hydraulic rod , Pin, spherical hinge and X-ray fluorescence analyzer, the drive motor is connected with the chassis to control the rotation of the entire chassis, the chassis and the connecting rod, the chassis and the electro-hydraulic rod are connected by a pin shaft, and the connecting rod is connected with the electric The hydraulic rods are fixedly connected, and the X-ray fluorescence analyzer is connected with the other end of the connecting rod by a spherical hinge, thereby controlling the X-ray fluorescence analyzer to detect the rock mass under test from different angles. The X-ray fluorescence analysis The instrument will emit X-rays and irradiate them to the rock block to be tested, and receive the secondary characteristic X-rays generated, give the element types and content in the rock block to be tested, and transmit the results to the data comprehensive analysis platform.
The TBM-mounted advanced geological prediction system based on the identification of lithology and poor geological precursor features as claimed in claim 1, characterized in that: the drying mechanism is a dryer, and the captured rock block to be tested Air-drying. When the conveyor belt system transfers the rock block to be tested directly below the dryer, the control system starts to control the air dryer to work, and release hot air to blow to the surface of the rock block to be tested. After a certain period of time, the dryer stops working , The surface of the rock block to be tested is dried.
A TBM-mounted advanced geological prediction system based on the identification of lithology and poor geological precursor features according to claim 1, wherein the crushing device includes a motor, a crushing cabin, a crushing shaft, a crushing knife, and a feed port , Hopper, discharge port, screen, valve and blower, the motor is installed on the upper part of the whole crushing device, responsible for providing high-speed rotation power, the servo motor shaft is connected with a crushing shaft, and the front end of the crushing shaft is equipped with a crushing knife, and The crushing shaft and crushing knife are both installed in the crushing cabin, and a feed port is provided on one side of the crushing cabin. The sample on the conveyor belt system can automatically fall to the feed port to enter the crushing cabin. The lower part of the crushing cabin is installed There is a layer of screen. During the crushing process, the particles filtered by the screen will fall into the bucket set below the crushing cabin. There are two outlets at the bottom of the bucket, one for the powder outlet and the other for the waste outlet. The powder particles at the outlet and the powder outlet can directly fall into the sample chamber of the X-ray diffraction analysis device below, and the waste at the waste outlet directly falls on the main TBM conveyor belt below, and there is one at each of the two outlets Valve, the blower is installed in the upper part of the crushing cabin;

Or, the X-ray diffraction analysis device includes an X-ray diffraction analyzer and a sample compartment, the sample compartment has upper and lower openings, and a hair dryer is installed above the upper opening, and the upper and lower openings have switches to control the corresponding openings The opening and closing.
The forecast method based on the system according to any one of claims 1-9, characterized in that it comprises the following steps:

Start the system to obtain the rock image information, the type and content information of the elements in the rock, and the type and content information of the minerals in the rock collected during the TBM tunneling process;

Based on the collected rock image information, the convolutional neural network is used to extract and characterize the color, texture and shape of the rock through the convolutional layer and the maximum pooling layer;

The extracted and characterized rock image feature information is fused with the collected rock element type and content, rock mineral type and content, and input into the rock image recognition and classification model that has been trained in the intelligent rock recognition system. Analyze and compare, and finally give the rock name, bad geological mark and accuracy rate of the rock block to be tested;

Conduct data comparison and analysis on the types and contents of a large number of collected rock elements, apply data mining technology to extract the unfavorable geologically sensitive elements, and quantitatively study their changing laws;

Conduct data comparison and analysis on the types and contents of a large number of collected rocks and minerals, apply data mining techniques to extract minerals with poor geological characteristics, and quantitatively study their changing laws;

After continuous measurement of rock samples with a certain excavation distance, the extracted rock can reflect the sign images of bad geological information, bad geological sensitive elements and bad geological mineral enrichment information using neural network for information fusion and correction, and based on data mining technology Comparing and analyzing the established quantitative characterization relational database of adverse geological precursors characteristics, so as to give out the lithological changes and adverse geological occurrence characteristics in front of the tunnels, and realize the advanced geological prediction of tunnels based on the intelligent identification of lithology and adverse geological precursors. .