A mine intelligent ventilation control system based on digital twin and data-driven

Abstract

The present invention discloses a mine intelligent ventilation control system based on digital twin and data drive, which belongs to the technical field of coal mine intelligent ventilation, and comprises an intelligent perception module, a data processing module, a digital twin module, an intelligent decision-making module and an emergency control module. The intelligent perception module is used for data perception and dynamic monitoring; the data processing module is used for centralized processing of the perceived data of the mine ventilation system; the digital twin module is used for constructing a three-dimensional simulation model of the ventilation system; the intelligent decision-making module is used for making expert decisions on the ventilation control scheme; the emergency control module is used for monitoring mine disaster information and formulating emergency control measures. The present invention performs intelligent perception of data such as wind speed, air volume, gas and smoke through sensing technology, constructs a digital twin model and data drive of the mine ventilation system, and effectively improves the reliability of intelligent control of the mine ventilation system using digital twin technology according to the optimized intelligent control scheme.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent ventilation for coal mines, and in particular to an intelligent ventilation control system for mines based on digital twins and data drive. Background Art

[0002] An intelligent coal mine is based on existing coal development technology and equipment, and integrates new-generation information technology represented by the Internet of Things, cloud computing, artificial intelligence, big data, and automatic control. It forms a coal mine intelligent system that integrates self-perception, self-learning, dynamic prediction, intelligent decision-making and regulation, and realizes full-process and all-round safety protection for coal mine production.

[0003] The mine ventilation system is the basis for ensuring safe production in coal mines. Failures in the ventilation system often lead to safety hazards such as wind turbulence, wind short circuits, and insufficient air volume at air-using locations. In severe cases, it even leads to mine disasters. Especially when disasters such as external fires occur, the ventilation system is damaged. Rapidly restoring normal ventilation in the mine is the key to curbing the expansion of accidents and carrying out emergency rescue. As the basic link in mine safety production, the mine ventilation system has put forward higher requirements for the accuracy, timeliness, comprehensiveness, and reliability of mine ventilation in the context of intelligence. Intelligence is the mainstream development direction of mine ventilation in the future.

[0004] Digital twin technology makes full use of data such as physical models, sensor updates, and operation history, integrates multi-disciplinary, multi-physical quantity, multi-scale, and multi-probability interactive simulation processes, and realizes high-fidelity virtual-real mapping between digital virtual space and actual physical equipment. Its technical characteristics and application fields coincide with the needs of mine intelligent ventilation systems. Therefore, the present invention relies on digital twin technology to innovate mine intelligent ventilation control systems, solve the problem of linkage control from aspects such as sensor information perception, wind network intelligent decision-making, and fire emergency control, and realizes functions such as front-end perception of ventilation parameters, dynamic monitoring of ventilation networks, expert decision-making of ventilation control, and disaster emergency control. It ensures the economic and reliable operation of the ventilation system under normal conditions, realizes dynamic analysis of disasters and timely elimination of wind and smoke flows during fire disasters, and improves the level of intelligent development of mine ventilation. Summary of the invention

[0005] The purpose of the present invention is to provide a mine intelligent ventilation control system based on digital twins and data-driven, with big data as the center and digital twins as the method, for real-time monitoring, intelligent decision-making, and simulation control, to solve the problems of large volume of sensor data sources, imperfect control system, and difficult wind and smoke flow control in mine ventilation systems, thereby improving the stability of mine intelligent ventilation system operation and the reliability of control decisions, and ensuring ventilation safety and safe production in coal mines.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A mine intelligent ventilation control system based on digital twin and data-driven, including:

[0008] Intelligent sensing module, which uses sensing technology to form a network of multiple types of sensors for front-end real-time perception and processing of underground ventilation parameters, and transmits the integrated data collected by sensors through the Internet of Things technology;

[0009] The data processing module is used to centrally process the massive sensor data sensed by the mine ventilation system, build a perception database, realize on-demand air distribution under normal conditions, and realize timely warning under abnormal conditions, so as to achieve safe regulation of intelligent ventilation;

[0010] The digital twin module is used to realize three-dimensional simulation modeling in the mine ventilation system, build a twin model library, and establish a data interaction mechanism between physical entities and virtual models, realizing a new model data fusion mode that relies on twin data to drive the operation of physical entities;

[0011] Intelligent decision-making module, used to solve the mine ventilation network in real time, predict and analyze the changes in ventilation system monitoring information, conduct training and learning on multi-source parameter changes based on deep learning and other algorithms, and further formulate expert decision-making and control plans;

[0012] The emergency control module is used to monitor and synchronously locate mine fire disaster information and disaster source points in real time. When a fire occurs underground, it can formulate and implement emergency control measures for the ventilation system.

[0013] As a further solution of the present invention, the intelligent sensing module includes:

[0014] The sensor submodule includes a new type of multi-source parameter sensor with high sensitivity and good reliability. Through the optimized layout strategy of the sensor, the ventilation parameters and disaster information are comprehensively collected, mainly including deep intelligent perception of the working condition data of the mine ventilation system equipment, multi-source information collection sensing of the ventilation network and perception of disaster precursor information, so as to achieve full mine area coverage and instant front-end perception, and collect real-time data, historical data and related derived data of various sensors of the physical entity of the wind network, so as to overcome the problems of dense sensor data sources, large volume and difficult ventilation control traceability;

[0015] The communication submodule includes the bottom terminal base station, the intermediate monitoring substation, and the top Internet of Things component. By constructing the "terminal-tunnel-working face-mining area-ground" hierarchical Internet of Things, it completes the efficient networking of underground multi-source parameter monitoring sensors and connects to the coal mine Internet of Things big data platform to realize real-time data transmission and safe storage of multi-source heterogeneous information, mainly including monitoring data networking, data anti-interference high-fidelity stable transmission, and disaster precursor information analysis and collaborative control.

[0016] As a further solution of the present invention, the data processing module includes:

[0017] Big data center is used to store, mine, analyze and apply sensor data of ventilation parameters. It uses intelligent computing technology based on big data to conduct in-depth mining, analysis and fusion processing of multi-source massive dynamic perception data and information to achieve intelligent decision-making and control. It mainly includes multi-source massive dynamic information aggregation, data mining model construction and update, and demand-driven wind network control service knowledge system construction;

[0018] The early warning center is used to realize real-time, advanced and accurate early warning of mine ventilation anomalies and disasters. It mainly includes the construction of a ventilation anomaly knowledge base with reasoning ability and semantic consistency, intelligent identification of disaster risks and intelligent disaster early warning based on deep learning, etc. It can judge whether the monitored sensor data exceeds the set numerical range. If it exceeds the set numerical range, timely early warning will be given to the abnormal data source of the ventilation system.

[0019] As a further solution of the present invention, the digital twin module includes:

[0020] The simulation modeling submodule is used to establish a matching digital twin model for the physical entity of the intelligent ventilation system, use 3D modeling software for simulation modeling and rendering optimization, build a digital and visual simulation model, and iteratively update the virtual model through simulation iterative calculation;

[0021] The intelligent computing submodule uses deep learning and neural network algorithms to train and learn the operation status and control scheme of the ventilation system and determine the optimal control method;

[0022] The data interaction submodule realizes the overall coordination and intelligence of the underground wind network through the intelligent interaction mechanism between entities and models, thereby realizing the data fusion interaction between physical entities and digital twins.

[0023] As a further solution of the present invention, the intelligent decision-making module includes:

[0024] The wind network solver module stores, analyzes and displays ventilation parameters, imports the mine ventilation system wind network solver software to perform network solver, and provides a basis for formulating control decision-making plans. The wind network solver imports the ventilation system diagram and network diagram into the mine ventilation simulation software, inputs the ventilation resistance measurement results, including air volume and resistance coefficient, and performs mine ventilation system network solver to predict changes in tunnel air volume.

[0025] The wind network control sub-module develops an expert decision-making solution library and determines the mine air volume adjustment method through deep data mining of the big data center and collaborative interaction with the twin model, while providing services for users to issue decision-making instructions.

[0026] As a further solution of the present invention, the emergency control module includes:

[0027] The disaster monitoring submodule senses, processes and analyzes real-time sensor data to monitor fire precursor information and detect fire sources, and to identify disaster locations in real time.

[0028] The emergency decision-making submodule formulates emergency control decision-making plans by analyzing the evolution of wind and smoke flow during the fire period;

[0029] The emergency air adjustment sub-module remotely controls multiple groups of air adjustment facilities. When a fire occurs underground, it will direct the wind and smoke flow into unrelated tunnels, thereby minimizing the impact of the wind and smoke flow on the ventilation system and ensuring the safety of personnel lives.

[0030] As a further solution of the present invention, the digital twin and data-driven mathematical description are as follows:

[0031] R = X ∪ Y;

[0032] X={x 1 ,x 2 ,x 3 ,…,x t} and Y={y 1 ,y 2 ,y 3 ,…,y t};

[0033] S={A,B,C} and

[0034] A={V,Q,P,Q 2 Concentration, CH 4 concentration...}; B = {T, CO concentration, CO 2 concentration,smoke...};

[0035] C={I,U,P,speed,shaft temperature,frequency...};

[0036]

[0037]

[0038] In the above formula: R is the fusion verification mechanism, X is the model constraint relationship, Y is the reasonable data threshold; S is the mine ventilation system data set; A is the ventilation parameter data set, and the ventilation parameters include wind speed, air volume, pressure, oxygen concentration, gas concentration, etc.; B is the fire information data set, and the fire information includes temperature, carbon monoxide concentration, carbon dioxide concentration, smoke, etc.; C is the equipment operating condition data set, and the equipment operating condition includes current, voltage, power, speed, shaft temperature, frequency, etc.; F is the generation of the control plan; T is the control plan verification mechanism; Indicates a further relationship, It is a collaborative driving relationship between the fusion verification mechanism and the regulation scheme verification mechanism.

[0039] As a further solution of the present invention, the step of collaboratively driving the model data includes:

[0040] First, the constraint relationship X between the virtual-real mapping of the digital twin model and the physical entity is formed through three-dimensional feature analysis, including the models of the main ventilator, local fan, wind tube, etc., which should have the same size constraint x as the physical entity. 1 , structural constraints x 2 , shape constraint x 3 , operating condition constraints x 4 , in order to ensure the rationality of the twin model; at the same time, according to the mine ventilation system data set S, determine the reasonable thresholds Y of each model and data in the system, including ventilation parameters and fire information reasonable thresholds, that is, the allowable wind speed y 1 、Air volume 2 , pressure 3 , oxygen concentration y 4 , gas concentration y 5 , carbon monoxide concentration y 6 , and according to the actual range of equipment operating conditions, establish reasonable operating thresholds for each equipment operating condition, including current y 7 , speed y 8 , frequency y 9 ;

[0041] Secondly, based on the mapping and collaborative relationship between each model and the data, a fusion verification mechanism R including the model constraint relationship X and the reasonable data threshold Y is constructed. On this basis, the dynamic operation data of the mine intelligent ventilation system data set S is deeply analyzed, focusing on the real-time changes of ventilation parameters, fire information and equipment operating condition data in the ventilation system. At the same time, the dynamic changes of other data are analyzed, and a real-time mapping relationship between the model and the data is established. Further, a control plan F is generated, and a control plan verification mechanism T is established through expert decision-making and deep learning.

[0042] Finally, according to the control scheme verification mechanism T, the mine ventilation system entity and the digital twin system are compared and judged, and the dynamic operation data and the rationality of the control measures are analyzed to provide a data basis for the accurate mapping of the twin model and the simulation of the control scheme; the fusion verification mechanism R is used to judge the rationality of the constraint relationship X and the data threshold Y in the twin model, so that the operation of the twin model is consistent with the physical entity; then the equipment and wind network operation data of the ventilation system are feature matched with the reasonable data threshold Y to ensure the correctness of the real-time operation data, and finally the data-driven model and the virtual-real mapping of the real-time data are realized through the data transmission channel, completing the model data fusion drive of the fusion verification mechanism R and the control scheme verification mechanism T.

[0043] Compared with the prior art, the beneficial effects of the present invention are as follows: first, the present invention obtains the sensory data of all elements of the ventilation system through sensing technology, including physical entities such as mine tunnels, fans, air ducts, air doors and windows, and coal mining machines; secondly, based on the mine ventilation system plan, stereogram, network diagram, mining engineering plan and other information, a three-dimensional simulation model of the mine wind network solution is established; according to the basic information of the physical entities and the sensor monitoring data, a digital twin model of the whole mine ventilation system is obtained; then, the ventilation system operation data is continuously updated through the optimized arrangement of sensors, and the data is uploaded to the ground big data center in time through the mine Internet of Things, so as to realize the dynamic update of the digital twin data and obtain the updated digital twin model; through the health behavior analysis of ventilation parameter monitoring and control schemes, the control behavior is trained and learned using deep learning and other algorithms, and the digital twin model is formulated. Expert decision-making program library determines the adaptive optimal control strategy; at the same time, during the operation of the entire control system, it dynamically monitors disaster precursor information and disaster source points, and can formulate and implement emergency control plans in the event of a fire, control the remote linkage of air conditioning facilities, reduce the impact of fire wind and smoke on the ventilation system, optimize the escape path of personnel, and guide personnel to escape through safe tunnels in time; ultimately, a full-scene simulation of virtual-real mapping, simulation, collaborative interaction, and other physical entities and three-dimensional models of the mine ventilation system is formed, and the model is displayed through a visual interface. The entire process from parameter monitoring to data analysis and decision-making control is comprehensively improved, thereby achieving intelligent control of the mine ventilation system driven by model data fusion, using digital twin models as technical means and multi-source sensor information as data support. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a schematic diagram of the organizational structure of a mine intelligent ventilation control system based on digital twin and data drive in the present invention.

[0045] Figure 2 This is a schematic diagram of constructing a digital twin model of a mine intelligent ventilation control system based on digital twin and data-driven in the present invention.

[0046] Figure 3 This is a flow chart of a decision-making and control scheme for a mine intelligent ventilation control system based on digital twin and data-driven in the present invention. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0048] The digital twin-based mine intelligent ventilation control system of the present invention organically combines multiple fields of knowledge such as sensing technology, Internet of Things technology, simulation modeling technology, digital twin technology, neural network algorithm, etc. The intelligent control system relies on sensors to instantly sense tunnel ventilation parameters (wind speed, air volume, gas, oxygen concentration, etc.) and fire precursor information (smoke, temperature, carbon monoxide, etc.), and completes dynamic data transmission through Internet of Things network technology (industrial Ethernet, WIFI protocol, 5G, etc.); uses simulation modeling software (3DMax, CAD, UG, Unity3D, etc.) to perform three-dimensional visualization modeling of physical entity equipment and tunnel information in the system, establishes a twin data information fusion model based on the virtual-real mapping relationship, and constructs a digital twin control system for mine ventilation system; according to algorithms such as deep learning, the decision-making simulation of the wind flow adjustment scheme is carried out to determine the optimal strategy for wind network control; during the operation of the entire system, the fire status and the evolution law of wind and smoke flow can be judged in real time, and prevention and control plans and wind and smoke flow control plans can be formulated to guide underground personnel to escape safely. Realizing intelligent control of mine ventilation systems can effectively prevent and control disasters and accidents caused by various abnormal ventilation parameters and fire risks.

[0049] like Figure 1As shown, a mine intelligent ventilation control system based on digital twin and data-driven is characterized in that it includes an intelligent perception module, a data processing module, a digital twin module, an intelligent decision-making module, and an emergency control module. The intelligent perception module uses sensing technology to form a network of multiple types of sensors for real-time front-end perception and processing of underground ventilation parameters, and transmits the fused data collected by the sensors through the Internet of Things technology; the data processing module is used to centrally process the massive sensor data perceived by the mine ventilation system, build a perception database, realize on-demand air distribution under normal conditions, and realize timely warning under abnormal conditions, so as to achieve safe regulation of intelligent ventilation; the digital twin module is used to realize three-dimensional simulation modeling within the mine ventilation system, build a twin model library, and establish a data interaction mechanism between physical entities and virtual models to realize model data fusion that relies on twin data to drive the operation of physical entities; the intelligent decision-making module is used to solve the mine air network in real time, predict and analyze changes in ventilation system monitoring information, carry out training and learning on multi-source parameter changes based on algorithms such as deep learning, and further formulate expert decision-making and regulation plans; the emergency regulation module is used to monitor and synchronously locate mine fire disaster information and disaster source points in real time. When a fire occurs underground, it can formulate and implement emergency regulation measures for the ventilation system.

[0050] The intelligent sensing module includes a sensor submodule and a communication submodule.

[0051] The sensor submodule includes a new type of multi-source parameter sensor with high sensitivity and good reliability, etc. Through the optimized layout strategy of the sensor, the ventilation parameters and disaster information are fully collected, mainly including deep intelligent perception of the working condition data of the mine ventilation system equipment, multi-source information collection sensing of the ventilation network, and perception of disaster precursor information. It realizes full mine area coverage and instant front-end perception, and collects real-time data, historical data and related derived data of various sensors of the physical entity of the wind network, overcoming the problems of dense sensor data sources, large volume, and difficult ventilation control traceability;

[0052] The communication submodule includes a bottom terminal base station, an intermediate monitoring substation, and a top-level Internet of Things component. By constructing the "terminal-tunnel-working face-mining area-ground" hierarchical Internet of Things, the efficient networking of underground multi-source parameter monitoring sensors is completed, and the Internet of Things big data platform of the coal mine is connected to realize real-time data transmission and safe storage of multi-source heterogeneous information, mainly including monitoring data networking, data anti-interference stable transmission, disaster precursor information analysis and collaborative control, etc.;

[0053] The mine Internet of Things communication network technology includes Wi-Fi, 5G network, ZigBee, LoRa, NB-IoT, GPRS, RFID, RS232, RS485, TCP / IP, HTTP, etc.

[0054] The data processing module includes a big data center and an early warning center.

[0055] The big data center is used to store, mine, analyze and apply the sensor data of ventilation parameters. By using intelligent computing technology based on big data, it conducts in-depth mining, analysis and fusion processing of multi-source massive dynamic perception data and information to achieve intelligent decision-making and control. It mainly includes multi-source massive dynamic information aggregation, data mining model construction and update, and demand-driven wind network control service knowledge system construction;

[0056] The early warning center is used to realize real-time, advanced and accurate early warning of mine ventilation anomalies and disasters, mainly including the construction of a ventilation anomaly knowledge base with reasoning ability and semantic consistency, intelligent identification of disaster risks based on deep learning, and intelligent early warning of disasters, etc. It is used to determine whether the monitored sensor data exceeds the set numerical range. If it exceeds the set numerical range, a timely early warning will be issued for the abnormal data source of the ventilation system.

[0057] The digital twin module includes a simulation modeling submodule, an intelligent computing submodule, and a data interaction submodule.

[0058] The simulation modeling submodule is used to establish a matching digital twin model for the physical entity of the intelligent ventilation system, use three-dimensional modeling software for simulation modeling and rendering optimization, build a digital and visual simulation model, and iteratively update the virtual model through simulation iterative calculation;

[0059] The intelligent computing submodule trains and studies the ventilation system operation status and control scheme through deep learning, neural network algorithms, etc., and determines the optimal control method

[0060] The data interaction submodule realizes the overall coordination and intelligence of the underground wind network through the intelligent interaction mechanism between the entity and the model, thereby realizing the data fusion interaction between the physical entity and the digital twin.

[0061] The intelligent decision-making module includes a wind network solver module and a wind network control submodule.

[0062] The wind network solver module stores, analyzes and displays ventilation parameters, imports the mine ventilation system wind network solver software to perform network solver, and provides a basis for formulating control decision-making plans. The wind network solver is a mine ventilation simulation software that imports mine ventilation system diagrams, network diagrams, etc., inputs ventilation resistance measurement results, including air volume, resistance coefficient, etc., and can perform mine ventilation system network solver to predict changes in tunnel air volume;

[0063] The wind network control submodule develops an expert decision-making solution library and determines the mine air volume adjustment method through deep data mining of the big data center and collaborative interaction with the twin model, while providing services for users to issue decision-making instructions.

[0064] The emergency control module includes a disaster monitoring submodule, an emergency decision-making submodule, and a wind control facility submodule.

[0065] The disaster monitoring submodule senses, processes and analyzes real-time sensor data to monitor fire precursor information and detect fire sources, and to determine the disaster location in real time.

[0066] The emergency decision-making submodule formulates an emergency control decision-making plan by analyzing the evolution law of wind and smoke flow during the fire period;

[0067] The emergency air-adjusting submodule remotely controls multiple groups of air-adjusting facilities. When a fire occurs underground, the air and smoke flow is introduced into irrelevant tunnels to minimize the impact of the air and smoke flow on the ventilation system and ensure the safety of personnel.

[0068] The air regulating facilities include an intelligent damper and window monitoring system, a mine-use flameproof and intrinsically safe control box, and remotely controllable dampers and windows; the mine-use flameproof and intrinsically safe control box includes a Siemens PLC, a safety barrier, an emergency power supply, etc.

[0069] The mathematical description of digital twins and data-driven is as follows:

[0070] R = X ∪ Y;

[0071] X={x 1 ,x 2 ,x 3 ,…,x t} and Y={y 1 ,y 2 ,y 3 ,…,y t};

[0072] S = {A, B, C}, and

[0073] A={V,Q,P,Q 2 Concentration, CH 4 concentration...}; B = {T, CO concentration, CO 2 concentration,smoke...};

[0074] C={I,U,P,speed,shaft temperature,frequency...};

[0075]

[0076]

[0077] Where: R is the fusion verification mechanism, X is the model constraint relationship, Y is the reasonable data threshold; S is the mine ventilation system dataset; A is the ventilation parameter dataset, and the ventilation parameters include wind speed, air volume, pressure, oxygen concentration, gas concentration, etc.; B is the fire information dataset, and the fire information includes temperature, carbon monoxide concentration, carbon dioxide concentration, smoke, etc.; C is the equipment operating condition dataset, and the equipment operating condition includes current, voltage, power, speed, shaft temperature, frequency, etc.; F is the generated control plan; T is the control plan verification mechanism; Indicates a further relationship, It is a collaborative driving relationship between the fusion verification mechanism and the regulation scheme verification mechanism.

[0078] The steps of collaboratively driving the model data include:

[0079] First, the constraint relationship X between the virtual-real mapping of the digital twin model and the physical entity is formed through three-dimensional feature analysis, including but not limited to the main ventilator, local fan, wind tube and other models, which should have the same size constraint x as the physical entity. 1 , structural constraints x 2 , shape constraint x 3 , operating condition constraints x 4 etc., in order to ensure the rationality of the twin model; at the same time, according to the mine ventilation system data set S, determine the reasonable thresholds Y of each model and data in the system, including but not limited to ventilation parameters and fire information reasonable thresholds, that is, the allowable wind speed y 1 、Air volume 2 , pressure 3 , oxygen concentration y 4 , gas concentration y 5 , carbon monoxide concentration y 6 etc., and establish reasonable operating thresholds for each equipment operating condition based on the actual range of equipment operating conditions, including but not limited to current y 7 , speed y 8 , frequency y 9 wait.

[0080] Secondly, based on the mapping and synergistic relationship between each model and the data, a fusion verification mechanism R is constructed, including the model constraint relationship X and the reasonable data threshold Y. On this basis, the dynamic operation data of the mine intelligent ventilation system data set S is deeply analyzed, focusing on the real-time changes of ventilation parameters, fire information and equipment operating condition data in the ventilation system, while analyzing the dynamic changes of other data, establishing a real-time mapping relationship between model and data, and further generating a control plan F. Through expert decision-making and deep learning, a control plan verification mechanism T is established.

[0081] Finally, according to the control scheme verification mechanism T, the mine ventilation system entity and the digital twin system are compared and judged, and the dynamic operation data and the rationality of the control measures are analyzed to provide a data basis for the accurate mapping of the twin model and the simulation of the control scheme; the fusion verification mechanism R is used to judge the rationality of the constraint relationship X and the data threshold Y in the twin model, so that the operation of the twin model is consistent with the physical entity; then the equipment and wind network operation data of the ventilation system are feature matched with the reasonable data threshold Y to ensure the correctness of the real-time operation data, and finally the data-driven model and the virtual-real mapping of the real-time data are realized through the data transmission channel, completing the model data fusion drive of the fusion verification mechanism R and the control scheme verification mechanism T.

[0082] The specific working method of the digital twin and data-driven intelligent mine ventilation control system is as follows:

[0083] Step 1, first, determine the physical entity of the mine ventilation intelligent control system. The physical entity refers to the real ventilation scene underground in the coal mine, specifically the physical models of the physical tunnels, ventilation equipment, electromechanical equipment, air conditioning facilities and multi-source parameter monitoring sensors through which the wind flows during the mine ventilation process; the physical tunnel information mainly refers to the shape and cross-section of the main tunnels through which the wind flows; the ventilation equipment includes the main ventilator, local fan, wind tube and other equipment that generates and guides the wind flow; the electromechanical equipment includes coal mining machines, tunneling machines, etc., and the air conditioning facilities include wind bridges, dampers, wind windows and other facilities for regulating wind flow; the multi-source parameters include ventilation parameters and fire information parameters, and the monitoring sensors include wind speed sensors, pressure sensors, temperature sensors, etc., as well as various concentration sensors of harmful gases such as gas, carbon monoxide, carbon dioxide, sulfur dioxide, smoke, hydrocarbons, etc. Select high-precision and highly adaptable sensors, optimize the sensor layout strategy, complete the layout and collection in mine tunnels and ventilation equipment facilities, and use the sensor sub-module to transmit the front-end perception data of ventilation equipment facility information and multi-source parameter monitoring information in real time to the mine Internet of Things through the unified network protocol of the communication sub-module (Wi-Fi, NB-IoT, 5G network, etc.), and further connect to the mine big data platform for centralized processing.

[0084] Step 2, further, such as Figure 2As shown, an intelligent digital twin control system for mine ventilation is constructed. The real mapping of all elements of the physical entity in the computer is established in real time and accurately, providing a virtual modeling environment for the digital twin system, an operating environment for the data flow, and an intelligent decision-making operating platform. The digital twin system includes but is not limited to the working face ventilation twin, the excavation ventilation twin, the pressure-equalizing ventilation twin, etc. The three-dimensional solution model and simulation model of the mine ventilation system are designed through three-dimensional simulation modeling software (3DMax, CAD software, UG, Unity3D, etc.), and the interface program is written using programming software (C# language, etc.) to establish data interaction, and the model data is connected in real time with various sensors of the physical entity through the OPC UA protocol.

[0085] The specific steps of solving the three-dimensional wind network model of the mine ventilation system are as follows:

[0086] The mine ventilation system is modeled by means of mine ventilation simulation software.

[0087] Step 2.1, based on the mining plan and ventilation system diagram of the mine, the data are simplified and imported into the software to generate a basic three-dimensional model;

[0088] Step 2.2, according to the spatial relationship of the ventilation tunnels, the model is specifically adjusted to make the tunnel position relationship in the model consistent with the actual situation, and the specific parameters of the tunnels in different sections are input into the model;

[0089] Step 2.3, assign values ​​to the friction resistance coefficient, cross section, air volume and other parameters of the tunnel according to the ventilation resistance measurement results, and input the location and parameters of the ventilator into the model. According to the actual situation of the mine, add ventilation adjustment equipment inside the tunnel to make the ventilation model consistent with the actual situation;

[0090] Step 2.4, by measuring the ventilation resistance of the mine and comparing it with the model calculation results, verify whether the model establishment is reasonable and accurate.

[0091] The specific steps of the digital twin simulation modeling of the mine ventilation system are as follows:

[0092] According to the virtual-real mapping relationship between the physical entity and the virtual body of the mine ventilation system, the ventilation control system is simulated and modeled.

[0093] Step 2.5, determine the three-dimensional characteristics of the structure, geometry, material and state of the physical entities in the mine ventilation system, and use CAD geometric modeling software to perform three-dimensional geometric modeling and simulation design on them, including equipment and facilities such as tunnels, main fans, local fans, wind pipes, dampers, wind windows, and monitoring devices;

[0094] Step 2.6, using 3D model software to render and optimize the model, and establish a digital model of the virtual entity of the twin control system;

[0095] Step 2.7, import the digital model of the ventilation control system into Unity3D software for virtual-reality fusion and visualization modeling, and build a virtual simulation model of the visualized physical entity.

[0096] Step 2.8, use C# language to write an interface program to realize information interaction and perform parameterized control on the virtual simulation model;

[0097] Step 2.9, design the interactive interface through the controls in the UGUI class of the Unity3D platform, develop the virtual scene and UI interface of the control system, and be able to observe and control the operating status of the twin control system from multiple angles.

[0098] Step 3: The digital twin system perceives the status data of physical entities in real time based on a unified network protocol, builds a mine ventilation information fusion data model, and establishes a virtual-real mapping relationship with the simulation model formed by the physical entity by calling the corresponding model in the data-driven virtual entity. By continuously iterating and optimizing the sensor perception process, a real-time perception and collaborative interaction mechanism between physical and virtual entities can be realized.

[0099] Step 4: Based on the three-dimensional solution model of the mine air network and the digital twin information fusion model, advance simulation and intelligent analysis of mine air volume changes are carried out. According to multi-source parameter monitoring information, the fan air supply, working face air volume, tunnel air volume, air volume and pressure of air regulating facilities, and gas outburst volume are predicted in advance to determine the best air volume matching and control plan. Based on deep learning, neural network algorithms, etc., intelligent decision-making and simulation control are realized, and expert decision-making and control plans are formulated. Simulation can be carried out on virtual entities first, and then applied to the physical entities of the ventilation system to implement control measures for the physical entities of the ventilation system.

[0100] The ventilation control scheme generation process is as follows: Figure 3 shown.

[0101] Step 4.1: Based on the ventilation parameters, fire information, and equipment operating data sensed and monitored in real time, analyze the changes in air volume, wind speed, and harmful gas concentration, and receive preliminary control requirements for the ventilation system;

[0102] Step 4.2, through big data computing and processing, select the digital twin virtual model environment, including but not limited to the working face twin, tunneling face twin, etc.;

[0103] Step 4.3, further, design a preliminary control plan, verify its rationality and optimize it through expert decision-making, and then form a preliminary plan;

[0104] Step 4.4, further, by simulating the implementation of twin data and decision-making solutions on the digital twin virtual body, the entity data and virtual body data are collected synchronously to form virtual-real interaction;

[0105] Step 4.5: Further, the solution is verified through intelligent computing analysis to decide whether to continue adjustment. If the result is yes, return to step 4 and continue simulation on the virtual control system. If not, then adjust the physical entity and test run the ventilation system.

[0106] Step 4.6, finally, analyze the results of the control plan. If it does not meet the requirements, return to step 3. If it meets the requirements, the process ends.

[0107] Step 5: During the operation of the mine ventilation real system and digital twin system, the big data center stores, analyzes, and deeply mines the sensor data source, monitors the abnormal ventilation status in real time, including abnormal working conditions of ventilation equipment and facilities, abnormal changes in air volume and pressure, abnormal concentration of harmful gases, and abnormal fire precursor information, and issues early warnings in a timely manner. When an abnormality or fire is detected, the ventilation system intelligently calculates the control plan to eliminate the risk, and the twin data simulation of the ventilation network solution and the effect after adjustment is performed at a high speed to verify and analyze the adjustment plan, monitor the ventilation parameter changes and the effect of eliminating fire hazards in the intelligent adjustment process of the ventilation system in real time, avoid accidents or minimize the scope of accidents as much as possible, reduce the loss of life and property, and reduce the impact on the mine ventilation system, so as to ensure safe production in the mine.

[0108] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0109] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0110] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A mine intelligent ventilation control system based on digital twin and data driven, characterized in that: include: Intelligent sensing module, which uses sensing technology to form a network of multiple types of sensors for front-end real-time perception and processing of underground ventilation parameters, and transmits the integrated data collected by sensors through the Internet of Things technology; The data processing module is used to centrally process the massive sensor data sensed by the mine ventilation system, build a perception database, realize on-demand air distribution under normal conditions, and realize timely warning under abnormal conditions, so as to achieve safe regulation of intelligent ventilation; The digital twin module is used to realize three-dimensional simulation modeling in the mine ventilation system, build a twin model library, and establish a data interaction mechanism between physical entities and virtual models, realizing a new model data fusion mode that relies on twin data to drive the operation of physical entities; Intelligent decision-making module, used to solve the mine ventilation network in real time, predict and analyze the changes in ventilation system monitoring information, conduct training and learning on multi-source parameter changes based on deep learning algorithms, and further formulate expert decision-making and control plans; The emergency control module is used to monitor and locate the mine fire disaster information and disaster source in real time. When a fire occurs underground, it can formulate and implement emergency control measures for the ventilation system. The mathematical description of digital twins and data-driven is as follows: R = X ∪ Y; X = {x1, x2, x3, …, x t} and Y = {y1, y2, y3, …, y t}; S={A,B,C} and A={V,Q,P,Q2 concentration,CH4 concentration...};B={T,CO concentration,CO2 concentration,smoke...}; C={I,U,P,speed,shaft temperature,frequency...}; In the above formula: R is the fusion verification mechanism, X is the model constraint relationship, Y is the reasonable data threshold; S is the mine ventilation system data set; A is the ventilation parameter data set, and the ventilation parameters include wind speed, air volume, pressure, oxygen concentration, and gas concentration; B is the fire information data set, and the fire information includes temperature, carbon monoxide concentration, carbon dioxide concentration, and smoke; C is the equipment operating condition data set, and the equipment operating condition includes current, voltage, power, speed, shaft temperature, and frequency; F is the generated control plan; T is the control plan verification mechanism; Indicates a further relationship, To provide a collaborative driving relationship between the fusion verification mechanism and the control scheme verification mechanism; The steps of collaboratively driving the model data include: First, through three-dimensional feature analysis, a constraint relationship X of virtual-real mapping between the digital twin model and the physical entity is formed, including the main fan, local fan, and wind tube model, which should have the same size constraint x1, structure constraint x2, shape constraint x3, and operating condition constraint x4 as the physical entity to ensure the rationality of the twin model; at the same time, according to the mine ventilation system data set S, the reasonable thresholds Y of each model and data in the system are determined, including ventilation parameters and fire information reasonable thresholds, that is, allowable wind speed y1, air volume y2, pressure y3, oxygen concentration y4, gas concentration y5, carbon monoxide concentration y6, and according to the actual range of equipment operating conditions, the reasonable operating thresholds of each equipment operating condition are established, including current y7, speed y8, and frequency y9; Secondly, based on the mapping and collaborative relationship between each model and the data, a fusion verification mechanism R including the model constraint relationship X and the reasonable data threshold Y is constructed. On this basis, the dynamic operation data of the mine intelligent ventilation system data set S is deeply analyzed, focusing on the real-time changes of ventilation parameters, fire information and equipment operating condition data in the ventilation system. At the same time, the dynamic changes of other data are analyzed, and a real-time mapping relationship between the model and the data is established. Further, a control plan F is generated, and a control plan verification mechanism T is established through expert decision-making and deep learning. Finally, according to the control scheme verification mechanism T, the mine ventilation system entity and the digital twin system are compared and judged, and the dynamic operation data and the rationality of the control measures are analyzed to provide a data basis for the accurate mapping of the twin model and the simulation of the control scheme; the fusion verification mechanism R is used to judge the rationality of the constraint relationship X and the data threshold Y in the twin model, so that the operation of the twin model is consistent with the physical entity; then the equipment and wind network operation data of the ventilation system are feature matched with the reasonable data threshold Y to ensure the correctness of the real-time operation data, and finally the data-driven model and the virtual-real mapping of the real-time data are realized through the data transmission channel, completing the model data fusion drive of the fusion verification mechanism R and the control scheme verification mechanism T.

2. According to claim 1, a digital twin and data-driven intelligent mine ventilation control system is characterized in that: The intelligent sensing module comprises: The sensor submodule includes a new type of multi-source parameter sensor with high sensitivity and good reliability. Through the optimized layout strategy of the sensor, the ventilation parameters and disaster information are comprehensively collected, mainly including deep intelligent perception of the working condition data of the mine ventilation system equipment, multi-source information collection sensing of the ventilation network and perception of disaster precursor information, so as to achieve full mine area coverage and instant front-end perception, and collect real-time data, historical data and related derived data of various sensors of the physical entity of the wind network, so as to overcome the problems of dense sensor data sources, large volume and difficult ventilation control traceability; The communication submodule includes the bottom terminal base station, the intermediate monitoring substation, and the top Internet of Things component. By constructing the "terminal-tunnel-working face-mining area-ground" hierarchical Internet of Things, it completes the efficient networking of underground multi-source parameter monitoring sensors and connects to the coal mine Internet of Things big data platform to realize real-time data transmission and safe storage of multi-source heterogeneous information, mainly including monitoring data networking, data anti-interference high-fidelity stable transmission, and disaster precursor information analysis and collaborative control.

3. According to claim 1, a digital twin and data-driven intelligent mine ventilation control system is characterized in that: The data processing module comprises: Big data center is used to store, mine, analyze and apply sensor data of ventilation parameters. It uses intelligent computing technology based on big data to conduct in-depth mining, analysis and fusion processing of multi-source massive dynamic perception data and information to achieve intelligent decision-making and control. It mainly includes multi-source massive dynamic information aggregation, data mining model construction and update, and demand-driven wind network control service knowledge system construction; The early warning center is used to realize real-time, advanced and accurate early warning of mine ventilation anomalies and disasters. It mainly includes the construction of a ventilation anomaly knowledge base with reasoning ability and semantic consistency, intelligent identification of disaster risks and intelligent disaster early warning based on deep learning, and analysis of whether the monitored sensor data exceeds the set numerical range. If it exceeds the set numerical range, timely early warning of the abnormal data source of the ventilation system will be issued.

4. According to claim 1, a digital twin and data-driven intelligent mine ventilation control system is characterized in that: The digital twin module includes: The simulation modeling submodule is used to establish a matching digital twin model for the physical entity of the intelligent ventilation system, use 3D modeling software for simulation modeling and rendering optimization, build a digital and visual simulation model, and iteratively update the virtual model through simulation iterative calculation; The intelligent computing submodule uses deep learning and neural network algorithms to train and learn the operation status and control scheme of the ventilation system and determine the optimal control method; The data interaction submodule realizes the overall coordination and intelligence of the underground wind network through the intelligent interaction mechanism between entities and models, thereby realizing the data fusion interaction between physical entities and digital twins.

5. According to claim 3, a digital twin and data-driven intelligent mine ventilation control system is characterized in that: The intelligent decision-making module includes: The wind network solver module stores, analyzes and displays ventilation parameters, imports the mine ventilation system wind network solver software to perform network solver, and provides a basis for formulating control decision-making plans. The wind network solver imports the ventilation system diagram and network diagram into the mine ventilation simulation software, inputs the ventilation resistance measurement results, including air volume and resistance coefficient, and performs mine ventilation system network solver to predict changes in tunnel air volume. The wind network control sub-module develops an expert decision-making solution library and determines the mine air volume adjustment method through deep data mining of the big data center and collaborative interaction with the twin model, while providing services for users to issue decision-making instructions.

6. The digital twin and data-driven intelligent mine ventilation control system according to claim 3 is characterized in that: The emergency control module includes: The disaster monitoring submodule senses, processes and analyzes real-time sensor data to monitor fire precursor information and detect fire sources, and identify disaster locations in real time. The emergency decision-making submodule formulates emergency control decision-making plans by analyzing the evolution of wind and smoke flow during the fire period; The emergency air adjustment sub-module remotely controls multiple groups of air adjustment facilities. When a fire occurs underground, it will direct the wind and smoke flow into unrelated tunnels, thereby minimizing the impact of the wind and smoke flow on the ventilation system and ensuring the safety of personnel lives.

Images