Intelligent cooperative control system for coal mining process based on digital twinning

Abstract

The invention discloses an intelligent cooperative control system for a coal mining process based on digital twinning, and particularly relates to the technical field of coal mining, which comprises a multi-source heterogeneous acquisition module, an anti-impact vibration sensor and a multi-protocol edge gateway implanted at key sites of a coal mine area, a mining system total factor perception system is constructed through an embedded sensor network and industrial Internet of Things protocol fusion technology, coal mining data is collected in real time, data robustness is enhanced through fault-tolerant Kalman filtering, and a spatial original data set is constructed; according to the invention, a total-factor perception system is constructed through the multi-source heterogeneous acquisition module to acquire data, and a multi-dimensional influence coefficient is output through fusion analysis of the data processing module, so that the problem of data independence is solved, deep association mining of equipment operation, environment and geological data is realized, and the data utilization rate is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of coal mining, and more specifically, to an intelligent collaborative control system for coal mining processes based on digital twins. Background Art

[0002] With the continuous advancement of science and technology, the coal mining industry is rapidly developing towards intelligentization. As a vital energy industry, coal mining efficiency and safety directly impact the nation's energy supply and the safety of miners. With the acceleration of intelligent transformation and the continuous innovation of coal mining technology, the demand for more advanced mining control systems is becoming increasingly urgent.

[0003] Traditional coal mining control systems usually include monitoring modules, control modules and communication modules; the monitoring module collects equipment operating parameters and underground environmental data in real time through sensors deployed at key locations of coal mining machines and hydraulic supports, and continuously monitors underground environmental data; the control module performs basic operations on coal mining machines and hydraulic support equipment based on preset programs; the communication module is responsible for data transmission with the network as the core, ensuring the transmission of information between equipment and the control center, realizing data transmission between underground equipment and the ground control center, ensuring the upload of equipment operation and environmental monitoring data, and the issuance of control instructions.

[0004] However, it still has some shortcomings in actual use, such as data independence. The modules of traditional coal mining control systems are relatively independent, and it is impossible to conduct in-depth integration and analysis of equipment operation data, environmental data, etc., resulting in low data utilization and difficulty in mining potential correlations between data; poor automation capabilities. The coordination of equipment in traditional coal mining control systems relies on manual scheduling. When underground equipment encounters sudden situations or mining conditions change, manual analysis and decision-making takes a long time and is prone to misjudgment, resulting in low equipment coordination efficiency; poor adaptability. Traditional coal mining control systems face complex geological conditions. Due to the lack of adaptive capabilities, the system cannot adjust mining parameters and equipment control strategies in a timely manner, resulting in waste of resources and equipment loss.

[0005] Therefore, there is an urgent need to provide an intelligent collaborative control system for coal mining processes based on digital twins to solve the problems of data independence, poor automation capabilities, and poor adaptability of existing coal mining control systems. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an intelligent collaborative control system for coal mining processes based on digital twins, which solves the problems raised in the above-mentioned background technology through the following scheme.

[0007] To achieve the above objectives, the present invention provides the following technical solution: an intelligent collaborative control system for coal mining processes based on digital twins, comprising:

[0008] A multi-source heterogeneous acquisition module implants shock-resistant vibration sensors and multi-protocol edge gateways at key locations in coal mining areas. By integrating embedded sensor networks with industrial IoT protocols, a full-factor perception system for the mining system is constructed. This system collects coal mining data in real time, enhances data robustness through fault-tolerant Kalman filtering, and constructs a spatial raw data set.

[0009] The data processing module uses a nonlinear coupling algorithm to output the impact coefficients of geological exploration layer data, environmental perception layer data, and equipment operation layer data in real time based on the coal mining data collected by the multi-source heterogeneous acquisition module, providing a quantitative decision-making basis for intelligent collaborative control.

[0010] The digital twin modeling module, based on underground geological exploration data, real-time mining face exposure data, and historical geological databases, integrates Kriging interpolation and hidden Markov models to construct a three-dimensional dynamic geological twin. It drives the twin model topology reconstruction in real time, provides a foundation for geological evolution prediction, and collects digital twin data.

[0011] The digital twin data processing module analyzes and processes the digital twin data to obtain the digital twin data influence coefficient and form the digital twin influence factor;

[0012] The comprehensive analysis module calculates the comprehensive safety effectiveness index based on the dynamic coupling of the four-dimensional influence coefficients of geological exploration layer data, environmental perception layer data, equipment operation layer data, and digital twin data, providing a quantitative decision-making basis for collaborative control;

[0013] The coal mining control module implements a four-level dynamic control strategy based on the comprehensive safety efficiency index. It triggers the corresponding control mode by real-time analysis of the comprehensive safety efficiency index value and issues control instructions in real time through the industrial ring network.

[0014] The human-computer interaction module transmits the obtained comprehensive safety effectiveness index, control mode strategy, and digital twin data to the user information terminal, providing reference data for making adjustment measures.

[0015] Preferably, the coal mining data includes geological exploration layer data, environmental perception layer data and equipment operation layer data.

[0016] Preferably, the geological exploration layer data includes the coal seam thickness variation, denoted as |ΔH|; lithologic impedance, denoted as Z; drilling rate, denoted as POR; microseismic event energy, denoted as σ; and environmental perception layer data includes gas concentration gradient, denoted as Dust settling rate, denoted as vd; temperature anomaly value, denoted as T1; roof separation displacement, denoted as δs; equipment operation layer data including chain tension deviation, denoted as |ΔF|; initial support force compliance rate, denoted as η; cutting energy entropy, denoted as Ec.

[0017] Preferably, the geological exploration layer data influence coefficient uses lithologic impedance to characterize the deformation resistance of the roof rock layer, and the drilling rate reflects the cuttability of the coal seam. The product of the two constitutes the basic geological stability indicator. Coal seam thickness variation is used as a risk factor, and stability is weakened using an exponential decay function: the greater the thickness variation, the more significant the reduction in stability. Microseismic energy reflects the risk of rock mass failure, and the coefficient is corrected in an exponential decay form.

[0018] Preferably, the environmental perception layer data impact coefficient uses the gas concentration gradient and the roof delamination displacement coupled through the Euclidean norm to form a composite disaster intensity, reflecting the risk of gas-rock interaction. The dust deposition rate is inversely regulated by the ventilation efficiency: when the ventilation volume increases, the dust impact is weakened, and the temperature anomaly is activated through the Sigmoid function to realize risk activation, and the coefficient is significantly amplified when the threshold is exceeded.

[0019] Preferably, the equipment operation layer data influence coefficient directly characterizes the reliability of the support system through the initial support force compliance rate and serves as the coefficient body. The cut vibration energy entropy reflects the health of the mechanism through an exponential decay function: the higher the entropy value, the more significant the health decay. The chain tension deviation is constrained by the hyperbolic tangent function, and the coefficient is sharply reduced when the deviation approaches the tolerance value to achieve hard protection of the transportation system overload. The support system serves as the core regulating factor, and vibration and tension serve as attenuation terms.

[0020] Preferably, the digital twin data includes cutting resistance deviation, recorded as ΔFc, support pressure deviation, recorded as ΔP, gas concentration deviation, recorded as ΔC, coal seam thickness deviation, recorded as ΔHm, model delay, recorded as τ, and confidence, recorded as β.

[0021] Preferably, the digital twin data influence coefficient reflects the prediction accuracy through the normalized robustness factor of the virtual-actual cutting resistance deviation, support pressure deviation, gas concentration deviation, and coal seam thickness deviation. The model delay is penalized by the hysteresis with piecewise exponential decay, and the historical fitting goodness and real-time deviation stability are weighted to form a dynamic confidence.

[0022] Preferably, the comprehensive safety effectiveness index is formed by the weighted average of the geological exploration layer data influence coefficient, the environmental perception layer data influence coefficient, and the equipment operation layer data influence coefficient to form the basic safety effectiveness, and then multiplied by the digital twin data influence coefficient to achieve global correction.

[0023] Preferably, the coal mining control module has the following specific steps:

[0024] When CSEI > 0.8, it is regarded as the optimization mode. The digital twin optimization parameters are used to regulate the speed of the coal mining machine, the pressure of the support group, and the ventilation volume to the same parameters as the digital twin. Then, the prediction parameter library of the twin is called. When 0.6 < CSEI ≤ 0.8, it is regarded as the stable mode, and the preset safety parameters are executed. When 0.4 < CSEI ≤ 0.6, it is regarded as the production restriction mode, and the production restriction protocol is started. When CSEI ≤ 0.4, it is regarded as the emergency mode, and a five-step emergency response is triggered. At the same time, a safety emergency response interlock is deployed. When Ke > 2.0 or Kg < 0.3, it is also regarded as the emergency mode, directly overriding the CSEI decision.

[0025] The specific preset safety parameters are as follows: the coal mining machine is 3 m / min, the support is 30 MPa, and the ventilation is 50 m

[0030] , Figure 1 , ,

[0029] , ,

[0028] , ,

[0031] , , , / s. The specific production restriction protocol is as follows: the equipment power is 80%, and the ventilation volume is 125%. The specific five-step emergency response is as follows: power off, the support pressure is increased to 120%, the ventilation is strengthened to 150%, sound and light alarm, and personnel evacuation.

[0026] The technical effects and advantages of the present invention are as follows:

[0027] 1. The present invention constructs a full-element perception system through a multi-source heterogeneous acquisition module to collect data, and outputs a multi-dimensional influence coefficient through fusion analysis by a data processing module, solves the problem of independent data, realizes the deep correlation mining of equipment operation, environment, and geological data, and improves the data utilization rate.

[0028] 2. The present invention executes a four-level dynamic control strategy based on the comprehensive safety efficiency index, automatically triggers the control mode in combination with the prediction parameter library of the digital twin, replaces manual scheduling, solves the problem of poor automation ability, realizes the real-time intelligent regulation of mining parameters and equipment coordination, shortens the decision-making time and improves the accuracy.

[0029] 3. The present invention uses the digital twin modeling module to construct a three-dimensional dynamic geological digital twin, fuses Kriging interpolation and Hidden Markov model to drive the topological reconstruction of the model, and combines the four-dimensional influence coefficient to dynamically couple and calculate the comprehensive safety efficiency index, solves the problem of poor adaptability, realizes the adaptive adjustment to complex geological conditions and mining environments, optimizes resource utilization and reduces equipment loss. Brief Description of the Drawings

[0030] Figure 1 It is a schematic diagram of the overall structure of the present invention. Detailed Embodiment

[0031] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0032] As attached Figure 1 The digital twin-based intelligent collaborative control system for coal mining processes shown in the figure includes:

[0033] The multi-source heterogeneous acquisition module implants anti-shock vibration sensors and multi-protocol edge gateways at key locations in the coal mining area. Through the fusion technology of embedded sensor networks and industrial Internet of Things protocols, a full-factor perception system for the mining system is built to collect coal mining data in real time. The data robustness is enhanced through fault-tolerant Kalman filtering to construct a spatial original data set.

[0034] In this embodiment, it should be specifically explained that the coal mining data includes geological exploration layer data, environmental perception layer data, and equipment operation layer data.

[0035] In this embodiment, it should be specifically noted that the geological exploration layer data includes the coal seam thickness variation, denoted as |ΔH|; lithologic impedance, denoted as Z; drilling rate, denoted as POR; microseismic event energy, denoted as σ; environmental perception layer data includes gas concentration gradient, denoted as Dust settling rate, denoted as vd; temperature anomaly value, denoted as T1; roof separation displacement, denoted as δs; equipment operation layer data including chain tension deviation, denoted as |ΔF|; initial support force compliance rate, denoted as η; cutting energy entropy, denoted as Ec.

[0036] In this embodiment, it should be specifically noted that a 24-channel microseismic monitoring array is arranged on the tunneling working face to measure the coal seam thickness variation, including a P-wave / S-wave composite sensor, and the thickness variation is inverted by the first wave arrival time difference and the longitudinal wave velocity, specifically:

[0037] |ΔH|=|0.732Δt·vp-H1|,

[0038] Where Δt represents the arrival time difference of P wave / S wave, vp represents the longitudinal wave velocity, and H1 represents the mining thickness of the target coal seam.

[0039] In this embodiment, it should be specifically noted that the lithologic impedance coefficient is calculated by multiplying the target shear wave velocity and rock density by the acoustic wave emitted by the MWD acoustic wave transmitter.

[0040] In this embodiment, it should be specifically noted that the penetration rate is calculated by collecting the drill pipe rotation speed and the bit pressure in real time through the Hall effect rotation speed sensor and the hydraulic pressure transmitter of the measurement while drilling system, and is specifically:

[0041]

[0042] Where k represents the drilling efficiency coefficient, which is directly obtained from the drilling rig calibration parameters, ω represents the drill pipe speed, and F represents the bit pressure.

[0043] In this embodiment, it should be specifically noted that the microseismic event energy is calculated by recording the vibration velocity using a three-component geophone, integrating the displacement amplitude, and calculating the microseismic event energy, specifically:

[0044] σ=2π 2 f 2 ρA 2 ,

[0045] Where f represents the main frequency of microseismicity, ρ represents the rock density, and A represents the vibration displacement amplitude.

[0046] In this embodiment, it should be specifically explained that the gas concentration gradient is obtained by collecting the east-west concentration difference and the north-south concentration difference through a laser methane sensor grid, and the gas concentration gradient is calculated based on the vector synthesis of the east-west concentration difference and the north-south concentration difference.

[0047] In this embodiment, it should be specifically noted that the dust settling rate is obtained by collecting the dust mass increment per unit time using a β-ray dust meter and calculating it in combination with the sampling area, specifically:

[0048]

[0049] Where ρd represents the dust density, A2 represents the sampling area, and Δm represents the dust mass increment per unit time.

[0050] In this embodiment, it should be specifically noted that the roof separation displacement is calculated by installing an inclination sensor on the roof to record the angle, and a laser rangefinder to record the displacement of the anchor end, based on the geometric decoupling of the anchor rod angle and the displacement of the anchor end, specifically:

[0051] δs=dsinθ+L(1-cosθ),

[0052] Where L is the anchor length, θ is the anchor angle, and d is the displacement of the anchor end.

[0053] In this embodiment, it should be specifically explained that the temperature anomaly value is collected by collecting the tunnel cross-section temperature matrix through a distributed optical fiber temperature measurement system, and the ambient reference temperature is measured at the air inlet, and the maximum value of the difference between the two is taken as the ambient reference temperature.

[0054] In this embodiment, it should be specifically explained that the real-time tension is collected by the strain gauge group installed on the scraper chain tensioner, and the difference is calculated by comparing it with the set value.

[0055] In this embodiment, it should be specifically explained that the initial support force reaching standard rate is directly calculated and obtained by array-type pressure monitoring and statistical probability model using a 16-point fiber Bragg grating pressure array arranged on the top beam of the hydraulic support.

[0056] In this embodiment, it should be specifically noted that the cutting energy entropy is obtained by collecting vibration acceleration through a three-axis vibration sensor on the cutting part of the coal mining machine, and then a 1024-point fast Fourier transform is performed to obtain a spectrum, and the power spectrum density is normalized and calculated, specifically:

[0057]

[0058] Where pk represents the frequency energy ratio.

[0059] The data processing module collects coal mining data through the multi-source heterogeneous acquisition module, and uses a nonlinear coupling algorithm to output the geological exploration layer data influence coefficient, the environmental perception layer data influence coefficient, and the equipment operation layer data influence coefficient in real time, providing a quantitative decision-making basis for intelligent collaborative control and forming a decision-making basis.

[0060] In this embodiment, it is specifically necessary to explain that the influence coefficient of the geological exploration layer data characterizes the deformation resistance of the roof rock layer through lithologic impedance, and the drilling rate reflects the cuttability of the coal seam. The product of the two constitutes the basic geological stability index. The variation in coal seam thickness is used as a risk factor, and the stability is weakened through an exponential decay function: the greater the thickness variation, the more significant the reduction in stability. Microseismic energy reflects the risk of rock mass rupture, and the coefficient is corrected in the form of exponential decay to achieve a triple dynamic balance of rock layer bearing capacity, cuttability, and rupture risk; the maximum drilling rate is used as a normalized benchmark to eliminate dimensions, ultimately achieving a dimensionless quantitative assessment of geological structure stability, specifically:

[0061]

[0062] in Represents the tensor product operation, specifically:

[0063]

[0064] Where |ΔH| represents the variation of coal seam thickness, Z represents lithologic impedance, POR represents drilling rate, σ represents microseismic event energy, and POR represents the microseismic event energy. max represents the historical maximum drilling rate, H1 represents the target coal seam mining thickness, and σ0 represents the critical energy of the rock mass.

[0065] In this embodiment, it should be specifically noted that the impact coefficient of the environmental perception layer data uses the gas concentration gradient and the roof separation displacement to couple through the Euclidean norm to form a composite disaster intensity, reflecting the risk of gas-rock interaction. The dust deposition rate is inversely regulated by the ventilation efficiency: when the ventilation volume increases, the dust impact is weakened. The temperature anomaly is activated by the Sigmoid function to significantly amplify the coefficient when it exceeds the threshold, thus constructing a dynamic assessment system for the gas-solid-thermal disaster chain. Specifically,

[0066]

[0067] in represents the gas concentration gradient, vd represents the dust settling rate, T1 represents the temperature anomaly value, δs represents the roof delamination displacement, C0 represents the upper limit of the standard safety concentration, δ0 represents the delamination safety limit, v0 represents the settlement reference value, and T0 represents the critical temperature for thermal damage activation.

[0068] In this embodiment, it should be specifically noted that the equipment operation layer data influence coefficient directly characterizes the reliability of the support system through the initial support force compliance rate and serves as the coefficient body. The cut vibration energy entropy reflects the health of the mechanism through an exponential decay function: the higher the entropy value, the more significant the health decay; the chain tension deviation is constrained by a hyperbolic tangent function, and the coefficient is sharply reduced when the deviation approaches the tolerance value to achieve hard overload protection of the transportation system. The support system serves as the core adjustment factor, and vibration and tension serve as attenuation terms to construct an overall health assessment model for the equipment system, specifically:

[0069]

[0070] Where |ΔF| represents the chain tension deviation, η represents the initial support force compliance rate, Ec represents the cutting energy entropy, F0 represents the tension safety tolerance, and E0 represents the critical energy entropy of mechanical health.

[0071] The digital twin modeling module, based on underground geological exploration data, real-time exposure data of mining faces and historical geological databases, integrates Kriging interpolation and hidden Markov models to construct a three-dimensional dynamic geological twin, drives the twin model topology reconstruction in real time, provides a basis for geological evolution prediction, and collects digital twin data.

[0072] In this embodiment, it should be specifically explained that the digital twin data includes the cutting resistance deviation, denoted as ΔFc, the support pressure deviation, denoted as ΔP, the gas concentration deviation, denoted as ΔC, the coal seam thickness deviation, denoted as ΔHm, the model delay, denoted as τ, and the confidence, denoted as β.

[0073] In this embodiment, it is specifically necessary to explain that the cutting resistance deviation is calculated by a discrete element simulation engine, and the spatial trajectory of the tooth tip of the coal mining machine drum and the hardness distribution of the coal seam are input. The cutting resistance value is output through parallel calculation by the GPU, and then subtracted from the current value collected in real time by the three-phase current sensor of the cutting motor, combined with the torque coefficient and the actual cutting resistance obtained by the drum radius.

[0074] In this embodiment, it should be specifically explained that the support pressure deviation is calculated through multi-body dynamics simulation, the hydraulic support group posture matrix and the top plate load distribution are input, the hydraulic system fluid-structure coupling equation is solved, the pressure cloud map mean is output, and the virtual support pressure obtained is subtracted from the actual support pressure collected by the embedded fiber grating sensor.

[0075] In this embodiment, it should be specifically noted that the gas concentration deviation is solved by a computational fluid dynamics simulation engine, a three-dimensional grid model of the tunnel is constructed in the digital twin, the real-time ventilation vector and the gas outburst source intensity are input, the Navier-Stokes-diffusion equation is solved to obtain the virtual gas concentration, and then a laser methane sensor grid is used for real-time collection, and the spatial gradient weighted fusion algorithm is used to calculate the actual gas concentration, and the two are subtracted to obtain the result.

[0076] In this embodiment, it should be specifically explained that the coal seam thickness deviation is obtained by transmitting electromagnetic wave pulses through a mining geological radar, receiving reflected signals to generate a tunnel section point cloud, combining the SLAM algorithm to construct a three-dimensional geological model, and subtracting the extracted virtual coal seam thickness from the actual coal seam thickness collected by the microseismic monitoring array.

[0077] In this embodiment, it should be specifically noted that the model delay is recorded through the system clock stamp. The time is marked when the digital twin receives the physical sensor data packet and when the simulation calculation is completed and the instruction is output. The time difference between the two is the model delay.

[0078] In this embodiment, it should be specifically explained that the confidence level is calculated by storing historical physical sensor data and the twin simulation values ​​at the corresponding moments, calculating the ratio of the sum of squares of the deviations between the physical value and the virtual value to the variance of the physical value, and then averaging the ratio to obtain the confidence level.

[0079] The digital twin data processing module analyzes and processes the digital twin data to obtain the digital twin data influence coefficient and form the digital twin influence factor.

[0080] In this embodiment, it should be specifically noted that the digital twin data influence coefficient reflects the prediction accuracy through the normalized robustness factor of the virtual-actual cutting resistance deviation, support pressure deviation, gas concentration deviation, and coal seam thickness deviation. The model delay is penalized by piecewise exponential decay hysteresis. The historical fit goodness and real-time deviation stability are weighted to form a dynamic confidence level, realizing the triple coupling of prediction accuracy, delay, and reliability. It is used to quantify the correction weight of the digital twin on the control decision, specifically:

[0081]

[0082] in

[0083]

[0084] in

[0085] Δi={ΔFc, ΔP, ΔC, ΔHm},

[0086] X0={Fc0,P0,C0,Hm0},

[0087] Where τ represents the model delay, β represents the confidence level, λ(τ) represents the delay attenuation coefficient, |Δi| represents the set of absolute deviation values ​​of the four sets of physical quantities, X0 represents the set of maximum allowable threshold values ​​of the four sets of physical quantities, ΔFc represents the cutting resistance deviation, ΔP represents the support pressure deviation, ΔC represents the gas concentration deviation, ΔHm represents the coal seam thickness deviation, Fc0 represents the maximum allowable cutting resistance deviation, P0 represents the maximum allowable support pressure deviation, C0 represents the maximum allowable gas concentration deviation, and Hm0 represents the maximum allowable coal seam thickness deviation.

[0088] The comprehensive analysis module calculates the comprehensive safety effectiveness index based on the dynamic coupling of the four-dimensional influence coefficients of geological exploration layer data, environmental perception layer data, equipment operation layer data and digital twin data, providing a quantitative decision-making basis for collaborative control.

[0089] In this embodiment, it should be specifically noted that the comprehensive safety effectiveness index is formed by weighted average of the influence coefficient of geological exploration layer data, the influence coefficient of environmental perception layer data, and the influence coefficient of equipment operation layer data to form a basic safety effectiveness, which is then multiplied by the influence coefficient of digital twin data to achieve global correction, and is used to quantitatively evaluate the safety and efficiency coordination status of the coal mining system, specifically:

[0090]

[0091] Among them, Kg represents the impact coefficient of geological exploration layer data, Ke represents the impact coefficient of environmental perception layer data, Kq represents the impact coefficient of equipment operation layer data, and Kd represents the impact coefficient of digital twin data.

[0092] The coal mining control module executes a four - level dynamic control strategy based on the comprehensive safety efficiency index, triggers the corresponding control mode by real - time analyzing the value of the comprehensive safety efficiency index, and issues control instructions in real - time through the industrial ring network.

[0093] In this embodiment, it should be specifically noted that the specific steps of the coal mining control module are as follows:

[0094] When CSEI > 0.8, it is regarded as the optimization mode. The digital twin optimization parameters are used to regulate the speed of the coal mining machine, the pressure of the support group, and the ventilation volume to the same parameters as the digital twin body, and then the twin body prediction parameter library is called; when 0.6 < CSEI ≤ 0.8, it is regarded as the stable mode, and the preset safety parameters are executed; when 0.4 < CSEI ≤ 0.6, it is regarded as the production - limiting mode, and the production - limiting protocol is started; when CSEI ≤ 0.4, it is regarded as the emergency mode, triggering a five - step emergency response, and at the same time deploying a safety emergency response interlock. When Ke > 2.0 or Kg < 0.3, it is also regarded as the emergency mode, directly overriding the CSEI decision.

[0095] In this embodiment, it should be specifically noted that the preset safety parameters are specifically: the coal mining machine at 3 m / min, the support at 30 MPa, and the ventilation at 50 m 3 / s; the production - limiting protocol is specifically: 80% of the equipment power and 125% of the ventilation volume; the five - step emergency response is specifically: power off, the support pressure is increased to 120%, the ventilation is strengthened to 150%, sound and light alarm, and personnel evacuation.

[0096] The human - machine interaction module transmits the obtained comprehensive safety efficiency index, control mode strategy, and digital twin body data to the user information terminal, providing reference data for making adjustment measures.

[0097] Secondly: In the attached drawings of the disclosed embodiments of the present invention, only the structures related to the disclosed embodiments are involved. For other structures, the usual designs can be referred to. Without conflict, the same embodiment and different embodiments of the present invention can be combined with each other;

[0098] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. The intelligent collaborative control system for coal mining process based on digital twin is characterized by: include: A multi-source heterogeneous acquisition module implants shock-resistant vibration sensors and multi-protocol edge gateways at key locations in coal mining areas. By integrating embedded sensor networks with industrial IoT protocols, a full-factor perception system for the mining system is constructed. This system collects coal mining data in real time, enhances data robustness through fault-tolerant Kalman filtering, and constructs a spatial raw data set. The data processing module uses a nonlinear coupling algorithm to output the impact coefficients of geological exploration layer data, environmental perception layer data, and equipment operation layer data in real time based on the coal mining data collected by the multi-source heterogeneous acquisition module, providing a quantitative decision-making basis for intelligent collaborative control. The digital twin modeling module, based on underground geological exploration data, real-time mining face exposure data, and historical geological databases, integrates Kriging interpolation and hidden Markov models to construct a three-dimensional dynamic geological twin. It drives the twin model topology reconstruction in real time, provides a foundation for geological evolution prediction, and collects digital twin data. The digital twin data processing module analyzes and processes the digital twin data to obtain the digital twin data influence coefficient and form the digital twin influence factor; The comprehensive analysis module calculates the comprehensive safety effectiveness index based on the dynamic coupling of the four-dimensional influence coefficients of geological exploration layer data, environmental perception layer data, equipment operation layer data, and digital twin data, providing a quantitative decision-making basis for collaborative control; The coal mining control module implements a four-level dynamic control strategy based on the comprehensive safety efficiency index. It triggers the corresponding control mode by real-time analysis of the comprehensive safety efficiency index value and issues control instructions in real time through the industrial ring network. The human-computer interaction module transmits the obtained comprehensive safety effectiveness index, control mode strategy, and digital twin data to the user information terminal, providing reference data for making adjustment measures.

2. The digital twin-based intelligent collaborative control system for coal mining processes according to claim 1 is characterized by: The coal mining data includes geological exploration layer data, environmental perception layer data and equipment operation layer data.

3. The digital twin-based intelligent collaborative control system for coal mining processes according to claim 1 is characterized by: The geological exploration layer data includes the coal seam thickness variation, denoted as |ΔH|; lithologic impedance, denoted as Z; drilling rate, denoted as POR; microseismic event energy, denoted as σ; environmental perception layer data includes gas concentration gradient, denoted as Dust settling rate, denoted as vd; temperature anomaly value, denoted as T1; roof separation displacement, denoted as δs; equipment operation layer data including chain tension deviation, denoted as |ΔF|; initial support force compliance rate, denoted as η; cutting energy entropy, denoted as Ec.

4. The digital twin-based intelligent collaborative control system for coal mining processes according to claim 1 is characterized by: The geological exploration layer data influence coefficient characterizes the deformation resistance of the roof rock layer through lithologic impedance, and the drilling rate reflects the cuttability of the coal seam. The product of the two constitutes the basic geological stability index. The thickness variation of the coal seam is used as a risk factor to weaken the stability through an exponential decay function: the greater the thickness variation, the more significant the reduction in stability. The microseismic energy reflects the risk of rock mass failure and is corrected in the form of an exponential decay coefficient.

5. The digital twin-based intelligent collaborative control system for coal mining processes according to claim 1 is characterized in that: The impact coefficient of the environmental perception layer data uses the gas concentration gradient and the roof delamination displacement coupled through the Euclidean norm to form a composite disaster intensity, reflecting the risk of gas-rock interaction. The dust deposition rate is inversely regulated by the ventilation efficiency: when the ventilation volume increases, the dust impact is weakened. The temperature anomaly realizes risk activation through the Sigmoid function, and the coefficient is significantly amplified when it exceeds the threshold.

6. The digital twin-based intelligent collaborative control system for coal mining processes according to claim 1 is characterized by: The data influence coefficient of the equipment operation layer directly characterizes the reliability of the support system through the initial support force compliance rate and serves as the main body of the coefficient. The cutting vibration energy entropy reflects the health of the mechanism through an exponential decay function: the higher the entropy value, the more significant the health decay. The chain tension deviation is constrained by a hyperbolic tangent function, and the coefficient drops sharply when the deviation approaches the tolerance limit to achieve hard protection against overload of the transportation system. The support system is the core adjustment factor, and vibration and tension are the decay terms.

7. The digital twin-based intelligent collaborative control system for coal mining processes according to claim 1 is characterized by: The digital twin data includes the cutting resistance deviation, denoted as ΔFc, the support pressure deviation, denoted as ΔP, the gas concentration deviation, denoted as ΔC, the coal seam thickness deviation, denoted as ΔHm, the model time delay, denoted as τ, and the confidence level, denoted as β.

8. The digital twin-based intelligent collaborative control system for coal mining processes according to claim 1 is characterized by: The data influence coefficient of the digital twin reflects the prediction accuracy through the normalized robust factors of the virtual-actual cutting resistance deviation, support pressure deviation, gas concentration deviation, and coal seam thickness deviation. The model time delay punishes the lag with a piecewise exponential decay, and the historical goodness of fit and the stability of the real-time deviation are weighted to form the dynamic confidence level.

9. The digital twin-based intelligent collaborative control system for coal mining processes according to claim 1 is characterized by: The comprehensive safety efficiency index is composed of the weighted average of the data influence coefficients of the geological exploration layer, the environmental perception layer, and the equipment operation layer to form the basic safety efficiency, and then multiplied by the data influence coefficient of the digital twin to achieve global correction.

10. The digital twin-based intelligent collaborative control system for coal mining processes according to claim 1, characterized in that: The specific steps of the coal mining control module are as follows: When CSEI > 0.8, it is regarded as the optimization mode. The digital twin optimization parameters are used to regulate the speed of the coal shearer, the pressure of the support group, and the ventilation volume to the same parameters as the digital twin, and then the twin prediction parameter library is called. When 0.6 < CSEI ≤ 0.8, it is regarded as the stable mode, and the preset safety parameters are executed. When 0.4 < CSEI ≤ 0.6, it is regarded as the production restriction mode, and the production restriction protocol is started. When CSEI ≤ 0.4, it is regarded as the emergency mode, and a five-step emergency response is triggered. At the same time, a safety emergency response interlock is deployed. When Ke > 2.0 or Kg < 0.3, it is also regarded as the emergency mode, directly overriding the CSEI decision. The preset safety parameters are: coal mining machine 3m / min, support 30MPa, ventilation 50m 3 / s; the specific production limit agreement is: equipment power 80%, ventilation volume 125%; the five-step emergency response is: power off, support pressurization to 120%, ventilation enhancement to 150%, sound and light alarm, and personnel evacuation.

Images