A smart drainage control system for reverse slope in long tunnels

Abstract

This invention discloses an intelligent drainage control system for long tunnels with reverse slopes, relating to the field of tunnel construction. The system includes: a water volume sensing device, a construction information sensing device, a control host, and drainage equipment. All three devices are connected to the control host. The water volume sensing device collects images of the tunnel face, the drainage volume from mechanical construction, the water level in the sump well, and the water level in the supply well. The construction information sensing device collects information on personnel, machinery, materials, methods, and environment. The control host, based on the water output from the tunnel face, the drainage volume from mechanical construction, the water level in the sump well, the water level in the supply well, and the information on personnel, machinery, materials, methods, and environment, uses digital twin technology to determine the parameters of the drainage equipment and controls the power and the number of devices that can be started or stopped based on these parameters. This invention promotes timely and accurate drainage.

Description

Technical Field

[0001] This invention relates to the field of tunnel construction, and in particular to an intelligent drainage control system for long tunnels with reverse slope. Background Technology

[0002] Since the beginning of the 21st century, with the continuous improvement of China's economic development level, the pace of transportation infrastructure construction has also been accelerating. Tunnels play an important role in shortening travel distances, improving transportation capacity, and reducing safety accidents. However, during tunnel construction, the uncertainty of the surrounding rock hydrogeology causes water seepage at the tunnel face, and wastewater generated by construction machinery operations. If the accumulated water at the tunnel face is not drained in time, it will lead to excessive water accumulation at the tunnel face and other work areas, seriously affecting construction efficiency and safety. Compared with other slope tunnel types, reverse slope tunnels have a stronger water collection capacity at the tunnel face, making the need for accurate prediction of construction drainage volume and autonomous and safe drainage decision-making even more urgent.

[0003] Drainage during tunnel construction is a crucial safety guarantee for tunnel operations. Accurately predicting and automatically coordinating drainage volume has always been a major challenge in tunnel construction. During tunnel excavation, due to the uncertainty of hydrogeological conditions, varying degrees of water seepage occur at the tunnel face and surrounding rock. Large amounts of water accumulation in front of the working face reduce the bearing capacity of the surrounding rock and affect workers' operations, posing serious safety hazards. In recent years, with the continuous improvement of construction mechanization, the use of machinery in tunnel construction has also increased. While mechanized construction brings high efficiency, it also consumes water and generates large amounts of wastewater. Therefore, solving the drainage problem in long highway tunnel construction is extremely important. Furthermore, due to the uncertainty of hydrogeology and the complex overlapping and coupling of various construction cycles, conventional tunnel construction drainage systems cannot identify the water output at the tunnel face or monitor wastewater from mechanical construction, thus failing to efficiently control drainage. This can easily lead to delays in drainage, and in the event of an emergency, water accumulation inside the tunnel cannot be drained in time, easily causing flooding of the work area, reducing work efficiency, and resulting in loss of personnel and resources.

[0004] Especially for drainage of long-distance, large-section, reverse-slope tunnels, the overlapping effects of multiple construction work areas within the tunnel make the drainage of water from the surrounding rock at the tunnel face and the drainage of water from construction machinery more complex. In order to improve construction efficiency and ensure personnel safety, it is particularly important to efficiently select drainage equipment parameters, drainage layout schemes, and control the drainage process.

[0005] In conventional tunnel drainage systems, the power and head of the pumps are usually fixed, and the pump control is lagging, resulting in a short effective working time. This easily leads to construction safety risks due to untimely drainage and mechanical damage and resource waste caused by the drainage system running idle. Therefore, traditional tunnel drainage systems can no longer meet the current needs of tunnel construction and the development concept of green, environmentally friendly, and economical practices.

[0006] A search revealed Chinese Patent No. CN113503186A, which discloses a reverse slope drainage system for long tunnel inclined shaft entry construction. This system includes several sump pits distributed along the tunnel's inclination direction. Each sump pit contains a submersible pump, and the output end of each pump is connected to a connecting pipe. In adjacent sump pits, the submersible pump in the lower sump pit is connected to the upper sump pit via a connecting pipe. A water collection pool is located at the upper inclined end of the tunnel, below the tunnel entrance. The submersible pump in the highest sump pit is connected to the water collection pool via a connecting pipe. This invention effectively reduces the pumping distance of the submersible pumps.

[0007] A search revealed Chinese Patent No. CN110410143A, which discloses a drainage system for reverse-slope tunnel construction, belonging to the field of construction drainage systems. The system includes a mobile submersible pump, a mobile water tank pumping station, a sump device, an external sedimentation and drainage device, and a drainage pipe. The mobile submersible pump is connected to the mobile water tank pumping station via the drainage pipe. The mobile water tank pumping station is connected to the sump device via the drainage pipe. The sump device is connected to the external sedimentation and drainage device via the drainage pipe. The inlet of the mobile submersible pump is located at the bottom of the tunnel face during reverse-slope tunnel construction. This invention is inexpensive, easy to use, reusable, ensures stable tunnel face conditions, and reduces construction risks.

[0008] However, the above system did not consider the identification of drainage of construction machinery and seepage of surrounding rock in the tunnel face area under the mechanized construction environment of tunnel, and failed to accurately predict and assess the water output of each area in the tunnel face; moreover, it only drained the bottom of the tunnel face and did not consider the coordinated control of drainage in multiple areas and different construction stages. Therefore, there was a certain lag and inefficiency in the drainage process, which posed a construction safety hazard.

[0009] The drill-and-blast method for tunnel construction is flexible, low-cost, and adaptable to various geological conditions, making it widely used in tunnel engineering construction in China. Drainage system design, including the layout of drainage equipment and the selection of pump power and head, is a crucial aspect of drill-and-blast construction. Drill-and-blast construction involves drilling, charging explosives, blasting, hazard removal, muck removal, initial spraying, steel arch erection, secondary spraying, waterproofing membrane, secondary lining reinforcement binding, secondary lining pouring, and secondary lining curing. Existing drainage systems do not consider the multi-factor impact of personnel, machinery, materials, methods, and environment during tunnel construction. Tunnel construction is divided into multiple areas, including the face excavation area, invert arch working area, waterproofing membrane working area, reinforcement binding working area, secondary lining pouring working area, and secondary lining curing area. The existing tunnel construction drainage system does not monitor the drainage volume of each construction area, especially the most critical excavation area at the tunnel face. It does not consider the personnel, machinery, materials, methods, and environment at each stage of the operation, does not coordinate and control the drainage of multiple construction areas, and does not monitor the drainage efficiency when multiple areas work together. It cannot take into account the drainage needs of the entire tunnel construction process, which can easily lead to serious water accumulation in some construction areas, waste of costs and resources, and increased construction safety risks. Summary of the Invention

[0010] The purpose of this invention is to provide an intelligent drainage control system for long tunnels with reverse slopes, which can promote the timeliness and accuracy of the drainage process, reduce the waste of resources and control measures during the drainage process, and improve drainage safety during construction.

[0011] To achieve the above objectives, the present invention provides the following solution:

[0012] A smart drainage control system for long tunnels with reverse slope includes: a water volume sensing device, a construction information sensing device, a control host, and a drainage device; the water volume sensing device, the construction information sensing device, and the drainage device are all connected to the control host.

[0013] The water volume sensing device is used to collect images of the working face, the amount of water discharged during mechanical construction, the water level in the collection well, and the water level in the supply well.

[0014] The construction information sensing equipment is used to collect information on personnel, machinery, materials, methods, and environment.

[0015] The control host is used to determine the parameters of the construction drainage equipment based on the water output from the working face, the drainage volume of the mechanical construction, the water level of the collection well, the water level of the supply well, and the information of personnel, machinery, materials, methods, and environment, using digital twin technology, and to control the power and number of start-ups and shutdowns of the drainage equipment according to the construction drainage equipment parameters.

[0016] Optionally, the water volume sensing device includes a flow meter, a level gauge, and a multi-view camera; the flow meter and level gauge are arranged in each water collection well and water supply well; the multi-view camera is arranged in the tunnel face.

[0017] Optionally, the construction information sensing device includes multiple cameras; the multiple cameras are installed in each construction area.

[0018] Optionally, the control host includes a knowledge acquisition unit, a water volume sensing unit, a construction area sensing unit, an information fusion unit, a parameter optimization unit, a data twin system simulation and verification unit, and an intelligent drainage system self-adjustment unit.

[0019] The knowledge acquisition unit is used to store drainage equipment information, mechanical equipment information, warning water level information, and tunnel construction information; the drainage equipment information includes pump power, suction head, and sump capacity; the mechanical equipment information includes equipment type, water usage information, and drainage information; the warning water level information includes the highest water level that the sump can hold and the maximum allowable discharge from the tunnel face; the tunnel construction information includes tunnel geometry, excavation mileage information, the location of each construction area, and construction organization information;

[0020] The water volume sensing unit is connected to the water volume sensing device; the water volume sensing unit is used to store the working face image, mechanical construction drainage volume, water level of the collection well and water level of the supply well, and to perform three-dimensional reconstruction of the working face image to determine the water output of the working face, and to obtain the equipment construction drainage volume and equipment construction water volume of each mechanical equipment.

[0021] The construction area sensing unit is connected to the water volume sensing device and the construction information sensing device; the construction area sensing unit is used to store the personnel, machinery, materials, methods, and environment information, and to determine the construction information based on the working face image and the construction area image using the YOLOv5 algorithm, and to record the entry and exit times of construction personnel and transport vehicles; the construction information includes the type of construction trolley, the type of transport vehicle, and the construction personnel.

[0022] The information fusion unit is connected to the knowledge acquisition unit, the water volume sensing unit, and the construction area sensing unit, respectively. The information fusion unit is used to preprocess the drainage equipment information, mechanical equipment information, warning water level information, tunnel construction information, tunnel face image, mechanical construction drainage volume, water collection well water level, water supply well water level, personnel, machinery, materials, methods, environment information, and construction information. Based on the preprocessed construction information and the entry and exit times of construction personnel and transport vehicles, the construction stage of the tunnel face is determined using a decision tree classification algorithm.

[0023] The parameter optimization unit is connected to the information fusion unit. The parameter optimization unit is used to predict the drainage volume of each construction stage and, based on the drainage volume, uses a particle swarm optimization algorithm to determine the construction drainage equipment parameters for each construction area. The construction drainage equipment parameters include: pump operating power, number of pumps to start and stop, head, and the arrangement spacing and number of collection wells.

[0024] The data twin system simulation verification unit is connected to the parameter optimization unit; the data twin system simulation verification unit is used to verify and iteratively optimize the equipment parameter scheme of the construction drainage system using digital twin technology;

[0025] The intelligent drainage system self-adjustment unit is connected to the data twin system simulation verification unit; the intelligent drainage system self-adjustment unit is used to control the power and number of start-ups and shutdowns of the drainage equipment according to the equipment parameter scheme of the construction drainage system.

[0026] Optionally, the information fusion unit includes a data processing subunit and a construction phase intelligent discrimination subunit;

[0027] The data processing subunit is used to preprocess information on drainage equipment, mechanical equipment, warning water level, tunnel construction, tunnel face images, drainage volume of mechanical construction, water level of collection well, water level of supply well, personnel, machinery, materials, methods, environment, and construction information; the preprocessing includes cleaning, completion, rejection, and noise reduction.

[0028] The intelligent construction stage discrimination subunit is used to determine the construction stage of the working face based on the processed construction information and the entry and exit times of construction personnel and transport vehicles, using a decision tree classification algorithm.

[0029] Optionally, the parameter optimization unit includes:

[0030] The drainage prediction subunit is used to predict the drainage volume of each construction stage based on the water output from the working face, the drainage volume of equipment construction, the total drainage volume of the sump, and the current construction stage, using the construction stage drainage prediction model. The construction stage drainage prediction model is obtained by training the GA-BP model using the water output from the working face, the drainage volume of equipment construction, the total drainage volume of the sump, the corresponding construction stage, and the training drainage volume of the corresponding construction stage.

[0031] The construction drainage equipment parameter optimization unit is used to determine the construction drainage equipment parameters for each construction area based on the drainage volume using a particle swarm optimization algorithm.

[0032] Optionally, the parameter optimization unit further includes:

[0033] The water demand prediction subunit is used to predict the water demand of each construction stage based on the water demand of equipment construction, the total water demand of the water supply well, and the corresponding construction stage, using the construction stage water demand prediction model. The construction stage water demand prediction model is obtained by training the GA-BP model using the water demand of equipment construction for training, the total water demand of the water supply well for training, the corresponding construction stage, and the training water demand for the corresponding construction stage.

[0034] The multi-construction area collaborative early warning subunit of the drainage system is used to assess the safety level of each construction area of ​​the tunnel based on the warning water level information and the parameters of the construction drainage equipment, according to the drainage volume, the water output at the tunnel face and the classification of the surrounding rock of the tunnel, and to determine the multi-construction area collaborative early warning water level, so as to provide early warning for the safety of the intelligent drainage control system for reverse slope of long tunnels.

[0035] Optionally, the control host further includes a data storage and retrieval unit; the data storage and retrieval unit is used to store the data output by each subunit in the information fusion unit and the parameter optimization unit.

[0036] Optionally, the drainage equipment includes a drain pipe, a collection well, a filter, a water supply well, a pump, and a pump controller.

[0037] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0038] The intelligent drainage control system for long tunnels with reverse slopes of this invention intelligently determines the personnel, machinery, materials, methods, and environment information of the current construction stage through water volume sensing and construction information sensing. It predicts the drainage volume based on the characteristics of the construction stage, and verifies the rationality and safety of the optimized parameters of the construction drainage equipment through digital twin system simulation. It adjusts the pumps through the pump controller to realize intelligent process regulation, promote the timeliness and accuracy of the drainage process, reduce the waste of resources and control measures during the drainage process, and improve the drainage safety during the construction process. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 The structural block diagram of the intelligent drainage control system for long tunnel reverse slope provided by the present invention;

[0041] Figure 2 A schematic diagram of the control host structure provided by the present invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] The purpose of this invention is to provide an intelligent drainage control system for long tunnels with reverse slopes, which can promote the timeliness and accuracy of the drainage process, reduce the waste of resources and control measures during the drainage process, and improve drainage safety during construction.

[0044] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0045] like Figure 1 As shown, the intelligent drainage control system for long tunnels with reverse slopes provided by this invention includes: a water volume sensing device, a construction information sensing device, a control host, and drainage equipment; the water volume sensing device, the construction information sensing device, and the drainage equipment are all connected to the control host. In practical applications, the intelligent drainage control system for long tunnels with reverse slopes includes a control host, drainage equipment, and water volume sensing devices and construction information sensing devices installed inside the tunnel. The drainage equipment includes a drainage pipe, a collection well, a filter, a supply well, a pump, and a pump controller. Collection wells between different construction areas are connected by drainage pipes, which are used for drainage. Drainage pipes within the construction area connect the collection well and the supply well, supplying water to the supply well. The collection well stores water from upstream. A filter is installed at the outlet of the collection well to remove impurities and other debris from the water, facilitating water supply to the supply well, improving utilization, and ensuring compliance with relevant emission standards when discharged outside the tunnel. The supply well stores construction water, receives usable water from the collection well, and provides water for each construction step and machinery. Pumps are installed in each collection well to drain water from the collection well to the next well. Pump controller: used to control the start / stop and operating time of the pumps.

[0046] The water volume sensing device is used to collect images of the tunnel face, the amount of water discharged during mechanical construction, and the water levels in the collection and supply wells. In practical applications, the water volume sensing device monitors information such as tunnel face images, mechanical construction drainage, and water levels in the collection and supply wells using monitoring instruments. These monitoring instruments include flow meters, level gauges, and multi-view cameras. The flow meters and level gauges are arranged in each collection and supply well; the multi-view cameras are arranged in the tunnel face.

[0047] The construction information sensing device is used to collect information on personnel, machinery, materials, methods, and environment. The device includes multiple cameras, which are positioned in various construction areas. In practical applications, the construction information sensing device acquires construction personnel, machinery, materials, methods, and environment information through the cameras in each construction area of ​​the tunnel.

[0048] The control host is used to determine the parameters of the construction drainage equipment based on the water output from the working face, the drainage volume of the mechanical construction, the water level of the collection well, the water level of the supply well, and the information of personnel, machinery, materials, methods, and environment, using digital twin technology, and to control the power and number of start-ups and shutdowns of the drainage equipment according to the construction drainage equipment parameters.

[0049] As an alternative implementation method, such as Figure 2 As shown, the control host includes a knowledge acquisition unit, a water volume sensing unit, a construction area sensing unit, an information fusion unit, a parameter optimization unit, a data storage and retrieval unit, a data twin system simulation and verification unit, and an intelligent drainage system self-adjustment unit.

[0050] The knowledge acquisition unit is used to store information on drainage equipment, mechanical equipment, warning water level, and tunnel construction.

[0051] In practical applications, the knowledge acquisition unit inputs the basic information required by the system (fixed attribute information during construction) and provides basic information for the subsequent information fusion unit and parameter optimization unit. The basic information includes drainage equipment information (pump power, suction head, and sump capacity), mechanical equipment information (equipment type, water usage and drainage information), warning water level information (the highest water level that the sump can hold, and the maximum permissible discharge from the tunnel face), and tunnel construction information (tunnel geometry, excavation mileage, location of each construction area, and construction organization information (organizational design of personnel, machinery, materials, methods, and environment)).

[0052] The water volume sensing unit is connected to the water volume sensing device; the water volume sensing unit is used to store the working face image, mechanical construction drainage volume, water level of the collection well and water level of the supply well, and to perform three-dimensional reconstruction of the working face image to determine the water output of the working face, and to obtain the equipment construction drainage volume and equipment construction water volume of each mechanical equipment.

[0053] In practical applications, the water volume sensing unit performs 3D reconstruction of the tunnel face image captured by the multi-view camera. It subtracts the equipment drainage volume from the bottom of the 3D reconstructed tunnel face image at two different times to obtain the water output volume corresponding to the time increment. It also acquires the equipment drainage volume and water consumption of each piece of machinery through mechanical equipment water volume monitoring. Furthermore, it calculates the total drainage volume upstream of the sump well based on water level changes and the total water consumption upstream of the supply well based on water level changes. Finally, it transmits the water output volume, equipment drainage volume, equipment water consumption, sump well water level monitoring data, and supply well water level monitoring data to the information fusion unit, providing subsequent multi-area tunnel construction water volume information data.

[0054] The three-dimensional reconstruction behind the working face includes:

[0055] 1. Use a multi-view camera to acquire two-dimensional images of the tunnel face from multiple points and perspectives. Ensure an overlap rate of more than 50% between each image.

[0056] 2. Feature extraction and matching: Align the images, use the SIFT (Scale-invariant feature transform) algorithm to extract feature points in the images, and perform matching to establish relationships between the images.

[0057] 3. Sparse point cloud reconstruction: Based on the incremental SFM (Structure from Motion) algorithm, a sparse 3D point cloud structure is generated, a unified coordinate system is established, and reliable matching point pairs are provided.

[0058] 4. Dense Point Cloud Reconstruction: The MVS (MultipleView Stereo) algorithm is used to generate a dense 3D point cloud structure.

[0059] 5. 3D Reconstruction: The point cloud is meshed using the Delaunay triangulation algorithm and texture mapping is performed to achieve 3D reconstruction.

[0060] The construction area sensing unit is connected to the water volume sensing device and the construction information sensing device; the construction area sensing unit is used to store the personnel, machinery, materials, methods, and environment information, and to determine the construction information based on the working face image and the construction area image using the YOLOv5 algorithm, and to record the entry and exit times of construction personnel and transport vehicles; the construction information includes the type of construction trolley, the type of transport vehicle, and the construction personnel.

[0061] In practical applications, the construction area perception unit statistically analyzes information on personnel, machinery, materials, methods, and environment during construction and builds a YOLOv5 deep learning model. It collects frame images from the working face monitoring camera and other construction area cameras, and uses the labelimg program to label and identify construction information (types of construction trolleys, transport vehicles, and construction personnel) in the images for tagging and classification. The model is then trained using the YOLOv5 algorithm to automatically recognize these tags, while simultaneously recording the entry and exit times of construction personnel and transport vehicles, providing data for subsequent construction phase identification.

[0062] The information fusion unit is connected to the knowledge acquisition unit, the water volume sensing unit, and the construction area sensing unit. The information fusion unit is used to preprocess drainage equipment information, mechanical equipment information, warning water level information, tunnel construction information, tunnel face image, mechanical construction drainage volume, water level of collection well, water level of supply well, personnel, machinery, materials, methods, environment information, and construction information. Based on the preprocessed construction information and the entry and exit times of construction personnel and transport vehicles, the unit uses a decision tree classification algorithm to determine the construction stage of the tunnel face.

[0063] In practical applications, the information fusion unit processes the acquired information and links the information on personnel, machinery, materials, methods, and environment to each construction stage (drilling-charging-blasting-hazard removal-slag removal-initial spraying-steel arch erection-re-spraying-waterproofing membrane-secondary lining reinforcement binding-secondary lining pouring-secondary lining curing) based on a decision tree classification algorithm using the information stored in the construction area perception unit. The information fusion unit includes a data preprocessing unit and a construction stage intelligent discrimination unit.

[0064] The data processing subunit is used to preprocess information on drainage equipment, mechanical equipment, warning water level, tunnel construction, tunnel face images, drainage volume of mechanical construction, water level of collection well, water level of supply well, personnel, machinery, materials, methods, environment, and construction information; the preprocessing includes cleaning, completion, rejection, and noise reduction.

[0065] In practical applications, the data processing subunit cleans, completes, and removes data collected from the upper part; and performs noise reduction and cleaning on image information.

[0066] The intelligent construction stage discrimination subunit is used to determine the construction stage of the working face based on the processed construction information and the entry and exit times of construction personnel and transport vehicles, using a decision tree classification algorithm.

[0067] In practical applications, the intelligent discrimination subunit for construction stages analyzes the characteristics of each construction stage and the construction organization plan, obtaining the characteristics of personnel, vehicles, and construction trolleys in each stage. Using a decision tree classification algorithm, and taking the types of construction trolleys, transport vehicles, and personnel marking information and entry / exit time information stored in the construction process perception unit as input, the information from the construction process perception unit is mapped to the construction stages, thereby establishing a "construction stage intelligent discrimination model." For example: Drilling stage (3 personnel, red clothing, red safety helmets, three-arm rock drilling trolley, no obvious material characteristics); Charging stage (12 charging personnel - yellow safety helmets - orange vests, 1 blasting engineer - red safety helmet - insulated clothing, 1 charging platform, 1 transport machine, 1 loader, No. 2 rock emulsion explosive); Blasting stage (no construction personnel, no construction machinery, no materials, obvious blasting detonator connection lines at the working face).

[0068] The parameter optimization unit is connected to the information fusion unit. The parameter optimization unit is used to predict the drainage volume of each construction stage and, based on the drainage volume, uses a particle swarm optimization algorithm to determine the construction drainage equipment parameters for each construction area. The construction drainage equipment parameters include: pump operating power, number of pumps to start and stop, head, and the arrangement spacing and number of collection wells.

[0069] As an optional implementation, the parameter optimization unit includes:

[0070] The drainage prediction subunit is used to predict the drainage volume of each construction stage based on the water output from the working face, the drainage volume of equipment construction, the total drainage volume of the sump, and the current construction stage, using a construction stage drainage prediction model. The construction stage drainage prediction model is obtained by training the GA-BP model using the water output from the working face, the drainage volume of equipment construction, the total drainage volume of the sump, the corresponding construction stage, and the training drainage volume of the corresponding construction stage.

[0071] In practical applications, the "Construction Stage Drainage Prediction Model" is established using the information on water output from the working face, equipment construction drainage, total drainage recorded in the collection well, and the construction stage discrimination results of the "Construction Stage Intelligent Discrimination Model" as inputs and the drainage volume of each construction stage as output.

[0072] The water demand prediction subunit is used to predict the water demand of each construction stage based on the water consumption of equipment construction, the total water consumption of the water supply well, and the corresponding construction stage, using the construction stage water demand prediction model. The construction stage water demand prediction model is obtained by training the GA-BP model using the water consumption of equipment construction for training, the total water consumption of the water supply well for training, the corresponding construction stage, and the training water demand for the corresponding construction stage.

[0073] In practical applications, the "construction stage water demand prediction model" is established using the equipment construction water consumption, the total water consumption recorded by the water supply well, and the construction stage discrimination results of the "construction stage intelligent discrimination model" as inputs and the water demand of each construction stage as output.

[0074] The construction drainage equipment parameter optimization unit is used to determine the construction drainage equipment parameters for each construction area based on the drainage volume using a particle swarm optimization algorithm.

[0075] In practical applications, the varying drainage volumes at different construction stages result in different requirements for pump power, number of pumps to start / stop, and pump head in different construction areas. Therefore, based on the predicted "construction stage drainage volume prediction model," and considering the varying drainage needs of different areas, a particle swarm optimization algorithm is used to optimize and select the construction drainage equipment parameters that meet the needs of each construction area. The construction drainage equipment parameters include: pump power, number of pumps to start / stop, and pump head, as well as the spacing and number of sump pits.

[0076] The drainage system and water supply system coordinated control subunit is used to supply wastewater stored in the drainage system that meets the construction water conditions to the water supply system after passing through the drainage system filtration device, and to record the amount of water used, thereby reducing drainage volume, improving water use efficiency, and reducing resource waste.

[0077] The multi-construction area collaborative early warning subunit of the drainage system is used to assess the safety level of each construction area of ​​the tunnel based on the warning water level information (generally not exceeding 0.3m above the top of the sump) and the parameters of the construction drainage equipment, according to the drainage volume, the water output at the tunnel face (based on the "construction stage drainage volume prediction model" and the monitoring results of the water output at the tunnel face), and the classification of the surrounding rock of the tunnel, and to determine the multi-construction area collaborative early warning water level, so as to provide early warning for the safety of the intelligent drainage control system for reverse slope of long tunnels.

[0078] In practical applications, the safety level is determined based on the actual construction conditions, surrounding rock grade, surrounding rock water output, and drainage volume of construction equipment. Based on this safety level and warning water level information, the actual warning water level of the working face sump is lowered. For other construction areas, the warning water level is determined based on the drainage situation of construction equipment in that area and changes in upstream drainage volume, with the principle of reducing pumping costs. The specific water level should be adjusted according to the actual engineering construction conditions, geological and hydrological conditions, construction methods, and the construction equipment used. The following water level adjustment methods are for reference only:

[0079] 1. For Class I or II surrounding rock with no seepage at the working face, the warning water level of the working face sump should not be lowered; 2. For Class III surrounding rock or with a small amount of seepage at the working face, the warning water level of the working face sump should be lowered by 0.1m; 3. For Class IV surrounding rock or with significant seepage at the working face, the warning water level of the working face sump should be lowered by 0.2m; 4. For Class V and Class VI surrounding rock or with water inflow at the working face, the warning water level of the working face sump should be lowered by 0.3m.

[0080] The data twin system simulation verification unit is connected to the parameter optimization unit; the data twin system simulation verification unit is used to verify and iteratively optimize the equipment parameter scheme of the construction drainage system using digital twin technology.

[0081] In practical applications, the digital twin system simulation verification unit uses a tunnel construction drainage digital twin system to put the optimized and selected construction drainage equipment parameters into a virtual digital twin model, simulates the drainage process characteristics of each construction area, verifies the construction drainage equipment parameters and layout scheme, and iteratively optimizes the drainage system by repeatedly adjusting the equipment layout within the digital twin system based on the principle of reducing the operating cost of the pumps.

[0082] The intelligent drainage system self-adjustment unit is connected to the data twin system simulation verification unit; the intelligent drainage system self-adjustment unit is used to control the power and number of start-ups and shutdowns of the drainage equipment according to the equipment parameter scheme of the construction drainage system.

[0083] In practical applications, the self-regulating unit of the intelligent drainage system transmits construction drainage equipment parameters, layout schemes, water output predictions, and sump water level monitoring data to the input interface of the LSTM deep learning model, and transmits construction drainage equipment parameters (pump working power, number of starts and stops, head, etc.) to the output interface of the LSTM deep learning model. This establishes a link between drainage system parameters and sump water level monitoring, realizing an "intelligent drainage system self-regulating model." This model guides the pump controller to perform frequency conversion, timing, counting, and safety early warning pre-drainage (when the predicted water output is large, the drainage equipment discharges a portion of the water in advance, thus increasing the overall water storage and drainage capacity of the drainage equipment), thereby achieving the purpose of automatically adjusting the pump working power and the number of starts and stops.

[0084] The data storage and retrieval unit is used to store the data output by each subunit in the information fusion unit and the parameter optimization unit.

[0085] The current drainage control system is slow to respond and inaccurate. It cannot adjust the drainage according to the actual construction situation and cannot make full use of existing and past data to make reasonable and effective adjustments to the drainage control.

[0086] Compared with existing technologies, the intelligent drainage control system for long tunnels with reverse slopes of the present invention establishes a data storage unit for the operation of the tunnel drainage system to facilitate data mining, integration and application. It senses information on personnel, machinery, materials, methods and environment during construction, and identifies the construction stage based on this. It establishes an intelligent prediction of drainage water consumption that takes into account the construction stage, and coordinates and optimizes drainage control and monitoring and early warning for multiple construction areas. It puts drainage decision information into the tunnel drainage digital twin system for simulation verification, improves the pertinence and rationality of construction drainage, improves drainage efficiency and reduces safety risks in various construction areas of the tunnel.

[0087] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0088] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A smart drainage control system for reverse slope in long tunnels, characterized in that, include: The system includes a water volume sensing device, a construction information sensing device, a control host, and a drainage device; the water volume sensing device, the construction information sensing device, and the drainage device are all connected to the control host. The water volume sensing device is used to collect images of the working face, the amount of water discharged during mechanical construction, the water level in the collection well, and the water level in the supply well. The construction information sensing equipment is used to collect information on personnel, machinery, materials, methods, and environment. The control host is used to determine the parameters of the construction drainage equipment based on the working face image, the mechanical construction drainage volume, the water level of the collection well, the water level of the supply well, and the information of personnel, machinery, materials, methods, and environment, using digital twin technology, and to control the power and number of start-ups and shutdowns of the drainage equipment according to the construction drainage equipment parameters; The control host includes a knowledge acquisition unit, a water volume sensing unit, a construction area sensing unit, an information fusion unit, a parameter optimization unit, a data twin system simulation and verification unit, and an intelligent drainage system self-adjustment unit. The knowledge acquisition unit is used to store drainage equipment information, mechanical equipment information, warning water level information, and tunnel construction information; the drainage equipment information includes pump power, suction head, and sump capacity; the mechanical equipment information includes equipment type, water usage information, and drainage information; the warning water level information includes the highest water level that the sump can hold and the maximum allowable discharge from the tunnel face; the tunnel construction information includes tunnel geometry, excavation mileage information, the location of each construction area, and construction organization information; The water volume sensing unit is connected to the water volume sensing device; the water volume sensing unit is used to store the working face image, mechanical construction drainage volume, water level of the collection well and water level of the supply well, and to perform three-dimensional reconstruction of the working face image to determine the water output of the working face, and to obtain the equipment construction drainage volume and equipment construction water volume of each mechanical equipment. The construction area sensing unit is connected to the water volume sensing device and the construction information sensing device; the construction area sensing unit is used to store the personnel, machinery, materials, methods, and environment information, and to determine the construction information based on the working face image and the construction area image using the YOLOv5 algorithm, and to record the entry and exit times of construction personnel and transport vehicles; the construction information includes the type of construction trolley, the type of transport vehicle, and the construction personnel. The information fusion unit is connected to the knowledge acquisition unit, the water volume sensing unit, and the construction area sensing unit, respectively. The information fusion unit is used to preprocess the drainage equipment information, mechanical equipment information, warning water level information, tunnel construction information, tunnel face image, mechanical construction drainage volume, water collection well water level, water supply well water level, personnel, machinery, materials, methods, environment information, and construction information. Based on the preprocessed construction information and the entry and exit times of construction personnel and transport vehicles, the construction stage of the tunnel face is determined using a decision tree classification algorithm. The parameter optimization unit is connected to the information fusion unit. The parameter optimization unit is used to predict the drainage volume of each construction stage and, based on the drainage volume, uses a particle swarm optimization algorithm to determine the construction drainage equipment parameters for each construction area. The construction drainage equipment parameters include: pump operating power, number of pumps to start and stop, head, and the arrangement spacing and number of collection wells. The data twin system simulation verification unit is connected to the parameter optimization unit; the data twin system simulation verification unit is used to verify and iteratively optimize the parameters of the construction drainage equipment using digital twin technology; The intelligent drainage system self-adjustment unit is connected to the data twin system simulation verification unit; the intelligent drainage system self-adjustment unit is used to control the power and number of start-ups and shutdowns of the drainage equipment according to the parameters of the construction drainage equipment.

2. The intelligent drainage control system for long tunnels with reverse slope as described in claim 1, characterized in that, The water volume sensing device includes a flow meter, a level gauge, and a multi-view camera; the flow meter and level gauge are arranged in each water collection well and water supply well; the multi-view camera is arranged in the tunnel face.

3. The intelligent drainage control system for long tunnels with reverse slope as described in claim 1, characterized in that, The construction information sensing device includes multiple cameras; the multiple cameras are set in each construction area.

4. The intelligent drainage control system for long tunnels with reverse slope as described in claim 1, characterized in that, The information fusion unit includes a data processing subunit and a construction phase intelligent discrimination subunit; The data processing subunit is used to preprocess information on drainage equipment, mechanical equipment, warning water level, tunnel construction, tunnel face images, drainage volume of mechanical construction, water level of collection well, water level of supply well, personnel, machinery, materials, methods, environment, and construction information; the preprocessing includes cleaning, completion, rejection, and noise reduction. The intelligent construction stage discrimination subunit is used to determine the construction stage of the working face based on the processed construction information and the entry and exit times of construction personnel and transport vehicles, using a decision tree classification algorithm.

5. The intelligent drainage control system for long tunnels with reverse slope as described in claim 1, characterized in that, The parameter optimization unit includes: The drainage prediction subunit is used to predict the drainage volume of each construction stage based on the water output from the working face, the drainage volume of equipment construction, the total drainage volume of the sump, and the current construction stage, using the construction stage drainage prediction model. The construction stage drainage prediction model is obtained by training the GA-BP model using the water output from the working face, the drainage volume of equipment construction, the total drainage volume of the sump, the corresponding construction stage, and the training drainage volume of the corresponding construction stage. The construction drainage equipment parameter optimization unit is used to determine the construction drainage equipment parameters for each construction area based on the drainage volume using a particle swarm optimization algorithm.

6. The intelligent drainage control system for long tunnels with reverse slope as described in claim 1, characterized in that, The parameter optimization unit further includes: The water demand prediction subunit is used to predict the water demand of each construction stage based on the water demand of equipment construction, the total water demand of the water supply well, and the corresponding construction stage, using the construction stage water demand prediction model. The construction stage water demand prediction model is obtained by training the GA-BP model using the water demand of equipment construction for training, the total water demand of the water supply well for training, the corresponding construction stage, and the training water demand for the corresponding construction stage. The multi-construction area collaborative early warning subunit of the drainage system is used to assess the safety level of each construction area of ​​the tunnel based on the warning water level information and the parameters of the construction drainage equipment, according to the drainage volume, the water output at the tunnel face and the classification of the surrounding rock of the tunnel, and to determine the multi-construction area collaborative early warning water level, so as to provide early warning for the safety of the intelligent drainage control system for reverse slope of long tunnels.

7. The intelligent drainage control system for long tunnels with reverse slope as described in claim 1, characterized in that, The control host also includes a data storage and retrieval unit; the data storage and retrieval unit is used to store the data output by each subunit in the information fusion unit and the parameter optimization unit.

8. The intelligent drainage control system for long tunnels with reverse slope as described in claim 1, characterized in that, The drainage equipment includes a drain pipe, a collection well, a filter, a water supply well, a pump, and a pump controller.

Images