Tunneling roadway ventilation-dust control cooperative adjustment method based on multi-mode intelligent control

Abstract

The invention provides a tunneling roadway ventilation-dust control cooperative adjustment method based on multi-modal intelligent control, and the method comprises the steps: obtaining multi-modal data, and obtaining multi-dimensional time sequence data; on the basis of the multi-dimensional time sequence data, predicting the multi-modal data through a prediction model to obtain a prediction result; performing pressure balance calculation based on the future change trend and fluctuation amplitude of each modal data in the multi-modal data and the multi-modal data, and generating a constraint condition; taking the constraint condition as a boundary constraint condition, solving the multi-target optimization model based on the future change trend and fluctuation amplitude of each modal data in the multi-modal data and real-time equipment operation parameters, and obtaining an optimal ventilation dust control adjustment strategy; and based on the expected control quantity and the multi-modal data, the frequency of a fan frequency converter, the opening degree of an air distribution valve and the angle of an air curtain are subjected to self-adaptive adjustment through a fuzzy PID algorithm, so that the ventilation and dust control effect of the tunneling roadway is improved.

Description

Technical Field

[0001] This application relates to the field of tunnel ventilation and dust control technology, and in particular to a method for coordinated adjustment of ventilation and dust control in tunneling tunnels based on multimodal intelligent control. Background Technology

[0002] Coal mine roadway excavation is a complex underground engineering project, facing numerous challenges such as poor ventilation and dust pollution. Ventilation and dust control technologies play a crucial role in coal mine roadway excavation, affecting not only the comfort of the working environment but also the safety of miners. Therefore, the application of ventilation and dust control technologies during coal mine roadway excavation is of great significance for ensuring miners' safety and improving work efficiency. However, the current ventilation and dust control effects in excavation roadways are inadequate and need improvement. Summary of the Invention

[0003] This application provides a method for coordinated adjustment of ventilation and dust control in tunneling roadways based on multimodal intelligent control, in order to improve the ventilation and dust control effect of tunneling roadways. The technical solution of this disclosure is as follows: In a first aspect, embodiments of this application propose a method for coordinated adjustment of ventilation and dust control in tunneling roadways based on multimodal intelligent control, including: The system acquires multimodal data collected in real time from various sensors, preprocesses the multimodal data, and obtains multidimensional time-series data. The multimodal data includes wind speed data, dust concentration data, gas concentration data, and ventilation pressure data. Based on the multidimensional time series data, a prediction model is used to predict the multimodal data to obtain prediction results, which include the future trends and fluctuation amplitudes of each modality in the multimodal data. Based on the future changing trends and fluctuation amplitudes of each modality in the multimodal data and the multimodal data, pressure balance calculations are performed to generate constraint conditions; Using the aforementioned constraints as boundary constraints, and based on the future trends and fluctuations of each modality in the multimodal data and the real-time equipment operating parameters, a multi-objective optimization model is solved to obtain the optimal ventilation and dust control adjustment strategy. The optimal ventilation and dust control adjustment strategy includes the desired control quantities of the fan frequency converter, the air distribution valve, and the air curtain angle adjustment mechanism. Based on the desired control quantity and the multimodal data, the frequency of the fan inverter, the opening degree of the air distribution valve, and the angle of the air curtain are adaptively adjusted using a fuzzy PID algorithm.

[0004] Secondly, embodiments of this application propose a multimodal intelligent control-based ventilation-dust control coordinated adjustment system for tunneling roadways, comprising: The multimodal sensing module is used to acquire multimodal data collected in real time by various sensors, preprocess the multimodal data to obtain multidimensional time-series data; the multimodal data includes wind speed data, dust concentration data, gas concentration data and ventilation pressure data; The trend prediction module is used to predict the multimodal data based on the multidimensional time series data through a prediction model, and obtain prediction results. The prediction results include the future change trend and fluctuation range of each modality in the multimodal data. The pressure balance control module is used to perform pressure balance calculations and generate constraint conditions based on the future changing trends and fluctuation amplitudes of each modality data in the multimodal data and the multimodal data. The multi-objective optimization module is used to take the constraints as boundary constraints, and solve the multi-objective optimization model based on the future change trend and fluctuation range of each modality data in the multi-modal data and the real-time equipment operating parameters to obtain the optimal ventilation and dust control adjustment strategy. The optimal ventilation and dust control adjustment strategy includes the expected control quantities of the fan frequency converter, the air distribution valve, and the air curtain angle adjustment mechanism. The fuzzy control module is used to adaptively adjust the frequency of the fan inverter, the opening degree of the air distribution valve, and the angle of the air curtain based on the desired control quantity and the multimodal data using a fuzzy PID algorithm.

[0005] Thirdly, embodiments of this application provide an electronic device, including: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method described in the first aspect.

[0006] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method described in the first aspect.

[0007] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the method described in the first aspect.

[0008] This application provides a method and system for coordinated ventilation and dust control in tunneling roadways based on multimodal intelligent control. Utilizing multimodal data collected from various sensors, the system employs a collaborative prediction model, pressure balance control, multi-objective optimization, and fuzzy PID control to adaptively adjust the frequency of the fan inverter, the opening of the air distribution valve, and the angle of the air curtain through optimal ventilation and dust control adjustment strategies. This achieves dynamic and intelligent environmental control of dust velocity in tunneling roadways, improving ventilation and dust control effectiveness. When facing different environmental conditions, the system can quickly react and optimize control strategies, not only improving the efficiency and accuracy of dust control at the working face but also effectively reducing energy consumption. This ensures the safe, efficient, and environmentally friendly operation of coal mines, providing an intelligent environmental management solution for tunneling faces.

[0009] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0010] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 A flowchart illustrating a method for coordinated adjustment of ventilation and dust control in tunneling roadways based on multimodal intelligent control, provided in an embodiment of this application; Figure 2 A block diagram of a multimodal intelligent control-based ventilation-dust control coordinated adjustment system for tunneling roadways, provided as an embodiment of this application; Figure 3 This is a block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0011] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0012] The following description, with reference to the accompanying drawings, describes a method, apparatus, and equipment for coordinated adjustment of ventilation and dust control in tunneling roadways based on multimodal intelligent control, according to embodiments of this application.

[0013] Figure 1 This is a flowchart illustrating a method for coordinated adjustment of ventilation and dust control in tunneling roadways based on multimodal intelligent control, provided in an embodiment of this application.

[0014] It should be noted that the execution subject of the tunnel ventilation-dust control coordinated adjustment method based on multimodal intelligent control in this application embodiment is the tunnel ventilation-dust control coordinated adjustment system based on multimodal intelligent control in this application embodiment. The tunnel ventilation-dust control coordinated adjustment system based on multimodal intelligent control can be configured in an electronic device so that the electronic device can perform the tunnel ventilation-dust control coordinated adjustment function based on multimodal intelligent control.

[0015] like Figure 1 As shown, the method for coordinated adjustment of ventilation and dust control in tunneling roadways based on multimodal intelligent control includes the following steps: Step S101: Acquire multimodal data collected in real time by various sensors, preprocess the multimodal data to obtain multidimensional time-series data; the multimodal data includes wind speed data, dust concentration data, gas concentration data and ventilation pressure data.

[0016] In some embodiments, the various sensors include a wind speed sensor, an air volume sensor, a pressure sensor, a dust concentration sensor, and a gas concentration sensor. The wind speed sensor and air volume sensor are arranged in the intake and return air roadways, the pressure sensor is arranged in the main ventilation roadway and local branches, the dust concentration sensor is arranged in the working area and the return air side, and the gas concentration sensor is arranged in the roof, the return air inlet, and in front of the tunneling machine.

[0017] As one implementation method, a real-time data set of the tunneling face is acquired through a sensor array. Specifically, at the perception layer, sensors such as wind speed, air volume, pressure, gas concentration, and dust concentration are deployed at key nodes of the tunneling face and ventilation system to continuously collect environmental parameters with a sampling period of 2 to 5 seconds. The collected data is then time-synchronized, noise-reducing filtered, and anomaly-removing by the on-site data acquisition terminal before being packaged and transmitted to the control center. The control center first standardizes and fuses the data to form a multi-dimensional matrix for model input, providing a unified high-precision input for the upper-level algorithm.

[0018] Step S102: Based on multidimensional time series data, the multimodal data is predicted using a prediction model to obtain prediction results. The prediction results include the future trend and fluctuation range of each modality in the multimodal data.

[0019] In some embodiments, the method described above for predicting multimodal data based on multidimensional time-series data and obtaining prediction results through a prediction model includes: converting multidimensional time-series data into a multidimensional time-series matrix, inputting the multidimensional time-series matrix into a CNN-LSTM hybrid prediction model, and outputting prediction results. The prediction results include future predicted trajectory data of wind speed, dust concentration, gas concentration, and ventilation pressure, confidence intervals and rate of change information of each future predicted trajectory, and the coupling relationship between wind speed, dust concentration, gas concentration, and ventilation pressure.

[0020] In one example, a CNN-LSTM hybrid prediction model is used as the prediction layer. Based on historical and real-time multi-source data such as wind speed, dust concentration, air volume, gas concentration, and pressure, it predicts the future trends and fluctuations of multiple variables such as wind speed, air volume, gas concentration, dust concentration, and pressure fluctuations. This CNN-LSTM hybrid prediction model extracts the spatial features of the input data through a convolutional neural network (CNN) and uses a long short-term memory network (LSTM) to model the time series data, predicting the possible future changes in multiple dimensions of the system, such as pressure fluctuations, wind speed changes, and gas concentration changes. The prediction results can be passed to the pressure balance control module 230 to provide a basis for early intervention, and also provide an assessment of the future system state for the multi-objective optimization algorithm.

[0021] As one implementation, the CNN-LSTM hybrid prediction model resides in the prediction layer. Its input is a fused data sequence of the most recent 5 minutes (10 sampling periods). The model outputs predicted trajectories of wind speed, air volume, pressure, gas concentration, and dust concentration for the next 3 minutes (6 prediction steps), along with confidence intervals and rate of change information. The prediction results are synchronously distributed to the pressure balance control module 230 and the multi-objective optimization algorithm via an intermediate cache module. For pressure balance control, the CNN-LSTM not only provides future pressure change trends but also the coupling relationships between variables, used to predict the impact of ventilation regulation on pressure stability in advance. For the multi-objective optimization algorithm, the prediction results serve as dynamic boundary inputs, providing a reference for the future system behavior in optimizing the objective function.

[0022] Step S103: Based on the future changing trends and fluctuation amplitudes of each modality in the multimodal data and the multimodal data, perform pressure balance calculations and generate constraint conditions.

[0023] In some embodiments, the method for generating constraints based on the future changing trends and fluctuation amplitudes of each modal data in multimodal data and the multimodal data includes: performing trend consistency analysis, rate of change judgment, and dynamic threshold correction on the future changing trends and fluctuation amplitudes of each modal data in multimodal data and the multimodal data to obtain the future pressure stability range, air volume adjustable range, gas safety threshold, and dust safety threshold; and generating pressure constraints, ventilation safety factor, target priority weight information, and amplitude limiting instructions based on the future pressure stability range, air volume adjustable range, gas safety threshold, and dust safety threshold.

[0024] As one implementation method, the pressure balance control module receives the predicted results of pressure, wind speed, air volume, and gas and dust concentrations output by the CNN-LSTM hybrid prediction model, and compares and analyzes them with real-time sensor data. This module assesses the risks of potential future pressure fluctuations, gas anomalies, and dust accumulation through trend consistency judgment and dynamic threshold calculation. When the risk level approaches or exceeds the safety threshold, the module automatically generates pressure constraints, ventilation safety factors, and target priority weight information, and passes these safety constraints as boundary parameters to the multi-objective optimization algorithm.

[0025] As an implementation method, pressure balance control is located in the evaluation layer, serving as a crucial intermediate link connecting the prediction and optimization layers. It continuously receives trends in pressure, wind speed, airflow, and gas concentration output from the CNN-LSTM model and dynamically compares and analyzes these trends with data collected from real-time sensors. The system performs a safety assessment in each cycle. When the difference between the prediction and actual measurements exceeds a threshold, it automatically shortens the assessment cycle and enters a high-priority mode, calculating future pressure stability ranges, airflow adjustment ranges, and safety margins for gas and dust. When the difference between the prediction and actual data exceeds a set threshold, or when the prediction results show pressure changes exceeding ±0.003 MPa, dust concentration exceeding 80 mg / m³, or a continuous upward trend in dust or gas concentration, it automatically increases the assessment frequency and activates a high-priority safety mode. This generates pressure constraints, ventilation safety factors, and target priority parameters, dynamically adjusting the upper and lower limits of pressure and airflow control. After the assessment, the module generates corresponding pressure constraints, ventilation safety factors, and target priority weights, and passes these data as safety boundary inputs to the multi-objective optimization algorithm to guide the optimization layer in performing control calculations within the safe and feasible domain.

[0026] In one example, the pressure balance control module 230 automatically performs a safety assessment in each control cycle without relying on manual triggering; when the difference between the prediction and the actual measurement exceeds the set threshold, it enters a high-priority mode, tightens the air volume and wind speed adjustment boundaries, and issues a temporary limiting command directly to the fuzzy PID controller when necessary.

[0027] Step S104: Using the constraints as boundary constraints, based on the future changing trends and fluctuation amplitudes of each modality in the multimodal data and the real-time equipment operating parameters, solve the multi-objective optimization model to obtain the optimal ventilation and dust control adjustment strategy. The optimal ventilation and dust control adjustment strategy includes the expected control quantities of the fan frequency converter, the air distribution valve, and the air curtain angle adjustment mechanism.

[0028] In some embodiments, the method described above, which uses constraints as boundary constraints and solves a multi-objective optimization model based on the future trends and fluctuations of each modality in the multi-modal data and real-time equipment operating parameters, includes: using pressure constraints and ventilation safety factors as boundary constraints, solving the multi-objective optimization model using a multi-objective particle swarm optimization algorithm based on the future trends and fluctuations of each modality in the multi-modal data and real-time equipment operating parameters to obtain the optimal ventilation and dust control adjustment strategy; determining whether the expected control quantity in the optimal ventilation and dust control adjustment strategy leads to pressure or dust exceeding limits; if so, adjusting the objective weights of the multi-objective optimization model according to the objective priority weight information and solving it again.

[0029] In some embodiments, the multi-objective optimization model is represented as follows:

[0030] in, This indicates the minimization of total energy consumption, reflecting the total power consumption of the wind turbine equipment; This indicates that the fan shaft power fluctuation is minimized, and the control system is stable while maintaining wind pressure balance. This indicates maximizing ventilation efficiency and improving the matching of fan operating efficiency and wind resistance. This means minimizing the safe deviation of gas concentration and keeping the gas concentration below the safe threshold. This indicates that the safety deviation of dust concentration is minimized, keeping the dust concentration below the health and explosion-proof standard limits; This indicates the goal of minimizing pressure fluctuation deviation and ensuring pressure safety.

[0031] The constraints include: The intake air volume and the exhaust air volume are conserved; The intake air volume and the exhaust air volume are conserved; respectively flow Boundary, wind turbine efficiency constraint; pressure Fluctuation constraints; For gas safety threshold, This refers to the dust concentration limit.

[0032] As an implementation approach, multi-objective optimization is used as the core decision-making layer, serving as a crucial node connecting the upper-level prediction and evaluation module with the lower-level execution module. It receives safety constraints, ventilation safety factors (i.e., dynamic boundary parameters), and target priority weights from the pressure balance control output, while simultaneously reading trend data predicted by the CNN-LSTM model and real-time equipment status. A multi-objective optimization function is established with the objectives of maximizing ventilation efficiency, minimizing energy consumption, minimizing pressure fluctuations, minimizing gas concentration deviation, and minimizing dust concentration deviation. An improved multi-objective particle swarm optimization algorithm (MOPSO) is used for iterative optimization within the feasible region. The algorithm calculates the optimal combination of fan frequency, air distribution valve opening, and air curtain angle within each 30-second control cycle. When the optimization result may trigger new pressure fluctuations or dust rebound risks, the algorithm automatically adjusts the multi-objective weights and iterates again based on the target priority weights provided by the pressure balance module, ensuring that the final solution achieves comprehensive optimization of energy efficiency, ventilation, and dust control performance within the safety boundaries. The optimization results are converted into executable parameters via the interface module and transmitted to the fuzzy PID control with a safety level label.

[0033] In one example, the multi-objective optimization solution employs a hybrid approach combining an improved multi-objective particle swarm optimization algorithm and sequential quadratic programming (SQP) to improve the convergence speed and computational accuracy.

[0034] This step involves multi-objective joint optimization of the predicted results of parameters such as wind pressure, wind speed, air volume, dust concentration, and gas concentration to output the optimal ventilation adjustment scheme, thereby achieving safe, energy-saving, and efficient coordinated control of the mine ventilation system. It can also provide a basis for early intervention for the pressure balance control module 230.

[0035] Step S105: Based on the desired control quantity and multimodal data, the frequency of the fan inverter, the opening degree of the air distribution valve, and the angle of the air curtain are adaptively adjusted using a fuzzy PID algorithm.

[0036] In some embodiments, the method described above for adaptively adjusting the frequency of the fan inverter, the opening degree of the air distribution valve, and the angle of the air curtain based on the desired control quantity and multimodal data using a fuzzy PID algorithm includes: determining whether there is a conflict between the desired control quantity and the limiting command; if so, adaptively adjusting the frequency of the fan inverter, the opening degree of the air distribution valve, and the angle of the air curtain based on the limiting command using a fuzzy PID algorithm; if not, adaptively adjusting the frequency of the fan inverter, the opening degree of the air distribution valve, and the angle of the air curtain based on the desired control quantity using a fuzzy PID algorithm.

[0037] In one example, based on the desired control quantity output by the multi-objective optimization of the fan inverter, the air distribution valve, the air volume of the curtain and the air curtain angle adjustment mechanism, the temporary limiting command and safety weight output by the pressure balance module, and the real-time sensor feedback data, the control measures of at least one component in the dust control system are obtained. The components of the dust control system include the fan inverter, the air distribution valve, the air volume of the curtain and the air curtain angle adjustment mechanism, etc., and may also include gas over-limit early warning.

[0038] As an implementation method, the fuzzy PID controller resides at the execution layer and is the final implementation unit of the system's actions. The controller receives the optimal parameter combination output by the multi-objective optimization algorithm and the limiting command issued by the pressure balance module, and prioritizes execution according to safety priorities. When conflicts arise, the system prioritizes the safety constraint command from the pressure balance module to ensure operational safety. The fuzzy PID controller calculates the deviation and deviation change rate in real time, dynamically adjusting the PID parameters through fuzzy inference to achieve coordinated control of the fan frequency and air curtain angle. Execution commands are sent to field devices via the industrial bus, including the main fan inverter, the distributor valve actuator, and the air curtain angle actuator system valve group, adjusting the equipment's operating status in real time. During operation, if the detected gas concentration approaches 1.0%, when the gas concentration monitoring value reaches the alarm threshold (1.0%), the safety interlock logic module immediately triggers a gas over-limit alarm and sends a coordination signal to the gas extraction system, prompting the extraction system to adjust the extraction negative pressure according to the concentration change. Simultaneously, it issues airflow limitation and speed reduction execution commands to the fuzzy PID controller. At this point, the fuzzy PID controller switches to safety mode, prioritizing the execution of speed limiting and ventilation stabilization control strategies to ensure that the gas concentration at the working face gradually decreases and the ventilation system pressure balance is maintained.

[0039] As one implementation method, the optimal combination parameters obtained from multi-objective optimization and the pressure constraint command are input into the fuzzy PID controller. The controller adaptively adjusts the PID parameters based on the real-time deviation and the rate of change of deviation, and precisely adjusts the fan frequency, the opening of the air distribution valve and the angle of the air curtain. When the gas concentration is detected to exceed 1.0%, the gas over-limit alarm is automatically triggered and speed limiting and air reduction measures are implemented.

[0040] In some embodiments, after control execution is complete, the system feeds back the results to the prediction and evaluation layers for model self-learning and weight update optimization. After execution, the feedback layer collects new real-time data and automatically compares it with the CNN-LSTM prediction results to form error samples for online model updates. When the error exceeds a preset threshold, the system triggers a self-learning mechanism to adaptively correct the prediction model weights and multi-objective optimization algorithm weight factors. Through this self-feedback mechanism, the system can gradually correct prediction biases and optimization biases in each cycle, achieving continuous improvement in global performance.

[0041] As one implementation method, after each control cycle, the system automatically compares the actual operating data with the prediction results of the CNN-LSTM model and calculates the prediction error of each ventilation parameter. The system maintains the data samples of the most recent N control cycles in real time through a rolling window data update mechanism. When the prediction error exceeds a set threshold, an online fine-tuning mechanism is triggered to automatically adjust some parameters of the CNN-LSTM model and simultaneously correct the weights of each objective in the multi-objective optimization algorithm, thereby achieving an adaptive closed-loop update of the model's prediction accuracy and optimization decision weights.

[0042] In some embodiments, the method of this application further includes: determining whether the gas concentration data has reached a warning threshold; when the gas concentration reaches the warning threshold, the system enters a warning state, the safety interlock logic increases the monitoring frequency and sends an airflow increase limit and speed reduction preparation command to the fuzzy PID controller; when the gas concentration reaches the warning threshold (e.g., 1.0%), the gas safety interlock module immediately triggers a gas over-limit alarm and sends a wind-limiting and speed-reducing execution command to the ventilation control side, while simultaneously sending a coordination signal to the gas extraction system through an interface. Subsequently, the fuzzy PID controller switches to a safe mode and executes measures such as speed limiting and wind reduction, valve position limiting, and air curtain stabilization under the priority of safety constraints; the alarm information is simultaneously reported to the upper-level monitoring system for manual confirmation and handling. In the above linkage, the ventilation control system does not directly control the extraction equipment, but only sends a coordination signal to achieve safe coordination between ventilation and extraction, thereby causing the gas concentration to drop without disrupting the pressure balance.

[0043] In some embodiments, the method of this application further includes: determining whether gas abnormalities and dust abnormalities exist simultaneously based on gas concentration data and dust concentration data. When the gas concentration is detected to reach or exceed the warning threshold, regardless of whether dust abnormalities are accompanied by it, the system prioritizes gas safety, immediately triggers a gas over-limit alarm, restricts airflow adjustment, and sends a regional closure and personnel evacuation prompt to the upper-level monitoring system to ensure that ventilation and safety control are coordinated.

[0044] The control cycle of this embodiment is 30 seconds, the prediction time window is 3 minutes, the response time for abnormal states is no more than 2 minutes, and the control adjustment efficiency is high.

[0045] This application presents a method for coordinated ventilation and dust control in tunneling roadways based on multimodal intelligent control. Utilizing multimodal data collected from various sensors, the method employs a coordinated prediction model, pressure balance control, multi-objective optimization, and fuzzy PID control. This allows for adaptive adjustment of the frequency of the fan inverter, the opening of the air distribution valve, and the angle of the air curtain through optimal ventilation and dust control strategies. This enables dynamic and intelligent environmental control of dust velocity in tunneling roadways, improving ventilation and dust control effectiveness. The system can quickly respond and optimize control strategies under different environmental conditions, not only improving the efficiency and accuracy of dust control at the working face but also effectively reducing energy consumption. This ensures safe, efficient, and environmentally friendly operation of coal mines, providing an intelligent environmental management solution for tunneling faces.

[0046] To achieve the above embodiments, this application also proposes a multimodal intelligent control-based ventilation-dust control coordinated adjustment system for tunneling roadways. Figure 2 This is a schematic diagram of a multimodal intelligent control-based ventilation-dust control coordinated adjustment system for tunneling roadways, provided as an embodiment of this application. Figure 2 As shown, the multimodal intelligent control-based ventilation-dust control coordinated adjustment system for tunneling can include: The multimodal sensing module 210 is used to acquire multimodal data collected in real time by various sensors, preprocess the multimodal data, and obtain multidimensional time-series data; the multimodal data includes wind speed data, dust concentration data, gas concentration data, and ventilation pressure data; The trend prediction module 220 is used to predict multimodal data based on multidimensional time series data through a prediction model, and obtain prediction results. The prediction results include the future change trend and fluctuation range of each modality in the multimodal data. The pressure balance control module 230 is used to perform pressure balance calculations and generate constraint conditions based on the future changing trends and fluctuation amplitudes of each modality data in the multimodal data and the multimodal data. The multi-objective optimization module 240 is used to take the constraints as boundary constraints, and solve the multi-objective optimization model based on the future change trend and fluctuation range of each modality data in the multi-modal data and the real-time equipment operating parameters to obtain the optimal ventilation and dust control adjustment strategy. The optimal ventilation and dust control adjustment strategy includes the expected control quantities of the fan frequency converter, the air distribution valve, and the air curtain angle adjustment mechanism. The fuzzy control module 250 is used to adaptively adjust the frequency of the fan inverter, the opening degree of the air distribution valve, and the angle of the air curtain based on the desired control quantity and multimodal data using a fuzzy PID algorithm.

[0047] Furthermore, in one possible implementation of this application embodiment, the pressure balance control module 230 is specifically used for: The future trends and fluctuations of each modality in the multimodal data are analyzed for trend consistency, rate of change and dynamic threshold correction, to obtain the future pressure stability range, air volume adjustable range, gas safety threshold and dust safety threshold. Based on the future pressure stability range, air volume adjustable range, gas safety threshold and dust safety threshold, pressure constraints, ventilation safety factor, target priority weight information and limit instructions are generated.

[0048] Furthermore, in one possible implementation of this application embodiment, the multi-objective optimization module 240 is specifically used for: Using pressure constraints and ventilation safety factor as boundary constraints, and based on the future trends and fluctuations of each modality in the multimodal data and real-time equipment operating parameters, the multi-objective optimization model is solved by the multi-objective particle swarm optimization algorithm to obtain the optimal ventilation and dust control adjustment strategy. Determine whether the desired control quantity in the optimal ventilation and dust control adjustment strategy leads to excessive pressure or dust levels. If so, adjust the objective weights of the multi-objective optimization model according to the objective priority weight information and solve it again; The multi-objective optimization model is represented as follows:

[0049] in, This represents minimizing total energy consumption. This indicates that the fan shaft power fluctuation is minimized. This indicates that ventilation efficiency is maximized. This indicates that the safety deviation of gas concentration is minimized. This indicates that the safety deviation of dust concentration is minimized. This indicates that the pressure fluctuation deviation is minimized.

[0050] Furthermore, in one possible implementation of this application embodiment, the trend prediction module 220 is specifically used for: Multidimensional time series data is transformed into a multidimensional time series matrix. The multidimensional time series matrix is ​​then input into a CNN-LSTM hybrid prediction model, which outputs prediction results. The prediction results include future predicted trajectory data of wind speed, dust concentration, gas concentration and ventilation pressure, confidence intervals and rate of change information of each future predicted trajectory, and the coupling relationship between wind speed, dust concentration, gas concentration and ventilation pressure.

[0051] Furthermore, in one possible implementation of this application embodiment, the fuzzy control module 250 is specifically used for: Determine if there is a conflict between the desired control quantity and the limiting instruction; If present, based on the limiting command, the frequency of the fan inverter, the opening of the air distribution valve, and the angle of the air curtain are adaptively adjusted using a fuzzy PID algorithm. If not, based on the desired control quantity, the frequency of the fan inverter, the opening degree of the air distribution valve, and the angle of the air curtain are adaptively adjusted using a fuzzy PID algorithm.

[0052] Furthermore, in one possible implementation of this application embodiment, the fuzzy control module 250 is also used for: Determine whether the gas concentration data is greater than the gas concentration threshold; If the gas concentration data exceeds the gas concentration threshold, a coordination signal, air volume limitation strategy, and gas over-limit alarm are sent to the gas extraction system.

[0053] Furthermore, in one possible implementation of this application embodiment, the fuzzy control module 250 is also used for: Based on gas concentration data and dust concentration data, determine whether gas abnormalities and dust abnormalities occur simultaneously. If both gas and dust anomalies are present, gas safety should be the highest priority. Implement measures such as limiting airflow and reducing speed, closing off hazardous areas, and maintaining adjacent air ducts within safe airflow ranges.

[0054] Furthermore, in one possible implementation of the embodiments of this application, the various sensors include a wind speed sensor, an air volume sensor, a pressure sensor, a dust concentration sensor, and a gas concentration sensor. The wind speed sensor and the air volume sensor are arranged in the intake and return air roadways, the pressure sensor is arranged in the main ventilation roadway and local branch roadway, the dust concentration sensor is arranged in the working area and the return air side, and the gas concentration sensor is arranged in the roof, the return air inlet, and in front of the tunneling machine.

[0055] It should be noted that the foregoing explanation of the embodiment of the ventilation-dust control coordinated adjustment method for tunneling roadways based on multimodal intelligent control also applies to the ventilation-dust control coordinated adjustment system for tunneling roadways based on multimodal intelligent control in this embodiment, and will not be repeated here.

[0056] To implement the above embodiments, this application also proposes an electronic device. Please see [link to relevant documentation]. Figure 3 , Figure 3 This is a schematic diagram of the structure of the electronic device provided in an embodiment of this application. For example... Figure 3 As shown, the electronic device 300 includes: a processor 301, and a memory 302 communicatively connected to the processor 301; the memory 302 stores computer execution instructions; the processor 301 executes the computer execution instructions stored in the memory to implement the method provided in the foregoing embodiments.

[0057] To implement the above embodiments, this application also proposes a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the methods provided in the foregoing embodiments.

[0058] To implement the above embodiments, this application also proposes a computer program product, including a computer program that, when executed by a processor, implements the methods provided in the foregoing embodiments.

[0059] In the foregoing descriptions of the embodiments, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0060] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0061] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for coordinated adjustment of ventilation and dust control in tunneling roadways based on multimodal intelligent control, characterized in that, Includes the following steps: The system acquires multimodal data collected in real time from various sensors, preprocesses the multimodal data, and obtains multidimensional time-series data. The multimodal data includes wind speed data, dust concentration data, gas concentration data, and ventilation pressure data. Based on the multidimensional time series data, a prediction model is used to predict the multimodal data to obtain prediction results, which include the future trends and fluctuation amplitudes of each modality in the multimodal data. Based on the future changing trends and fluctuation amplitudes of each modality in the multimodal data and the multimodal data, pressure balance calculations are performed to generate constraint conditions; Using the aforementioned constraints as boundary constraints, and based on the future trends and fluctuations of each modality in the multimodal data and the real-time equipment operating parameters, a multi-objective optimization model is solved to obtain the optimal ventilation and dust control adjustment strategy. The optimal ventilation and dust control adjustment strategy includes the desired control quantities of the fan frequency converter, the air distribution valve, and the air curtain angle adjustment mechanism. Based on the desired control quantity and the multimodal data, the frequency of the fan inverter, the opening degree of the air distribution valve, and the angle of the air curtain are adaptively adjusted using a fuzzy PID algorithm.

2. The method according to claim 1, characterized in that, The step involves performing a pressure balance calculation based on the future trends and fluctuation amplitudes of each modality in the multimodal data, and generating constraint conditions, including: The future change trends and fluctuation amplitudes of each modality in the multimodal data are analyzed for trend consistency, change rate is judged, and dynamic threshold is corrected to obtain the future pressure stability range, air volume adjustable range, gas safety threshold, and dust safety threshold. Based on the future pressure stability range, air volume adjustable range, gas safety threshold, and dust safety threshold, pressure constraints, ventilation safety factor, target priority weight information, and amplitude limiting instructions are generated.

3. The method according to claim 2, characterized in that, The step of using the constraints as boundary constraints, and solving the multi-objective optimization model based on the future changing trends and fluctuation amplitudes of each modality in the multimodal data and the real-time equipment operating parameters, includes: Using the pressure constraint and ventilation safety factor as boundary constraints, and based on the future changing trends and fluctuation amplitudes of each modality in the multimodal data and the real-time equipment operating parameters, the multi-objective optimization model is solved by the multi-objective particle swarm optimization algorithm to obtain the optimal ventilation and dust control adjustment strategy. Determine whether the desired control quantity in the optimal ventilation and dust control adjustment strategy leads to excessive pressure or dust levels. If so, adjust the target weights of the multi-objective optimization model according to the target priority weight information and solve it again; The multi-objective optimization model is represented as follows: in, This represents minimizing total energy consumption. This indicates that the fan shaft power fluctuation is minimized. This indicates that ventilation efficiency is maximized. This indicates that the safety deviation of gas concentration is minimized. This indicates that the safety deviation of dust concentration is minimized. This indicates that the pressure fluctuation deviation is minimized.

4. The method according to claim 1, characterized in that, The process of using a prediction model to predict multimodal data based on the multidimensional time-series data to obtain prediction results includes: The multidimensional time series data is transformed into a multidimensional time series matrix, and the multidimensional time series matrix is ​​input into a CNN-LSTM hybrid prediction model to output prediction results. The prediction results include future predicted trajectory data of wind speed, dust concentration, gas concentration and ventilation pressure, confidence interval and rate of change information of each future predicted trajectory, and the coupling relationship between wind speed, dust concentration, gas concentration and ventilation pressure.

5. The method according to claim 2, characterized in that, The adaptive adjustment of the frequency of the fan inverter, the opening degree of the air distribution valve, and the angle of the air curtain based on the desired control quantity and the multimodal data using a fuzzy PID algorithm includes: Determine whether there is a conflict between the desired control value and the limiting command; If present, based on the limiting command, the frequency of the fan inverter, the opening of the air distribution valve, and the angle of the air curtain are adaptively adjusted using a fuzzy PID algorithm. If not, based on the desired control quantity, the frequency of the fan inverter, the opening degree of the air distribution valve, and the angle of the air curtain are adaptively adjusted using a fuzzy PID algorithm.

6. The method according to claim 5, characterized in that, The method further includes: Determine whether the gas concentration data is greater than the gas concentration threshold; If the gas concentration data exceeds the gas concentration threshold, a coordination signal and a gas over-limit alarm are sent to the gas extraction system. The ventilation control system does not directly control the extraction equipment, but only sends a coordination signal to achieve safe coordination between ventilation and extraction.

7. The method according to claim 1, characterized in that, The method further includes: Based on the gas concentration data and the dust concentration data, determine whether gas abnormality and dust abnormality occur simultaneously. If both gas and dust anomalies occur simultaneously, gas safety should be the highest priority. Implement measures such as limiting airflow and reducing speed, closing off hazardous areas, and maintaining adjacent air ducts within safe airflow ranges.

8. The method according to claim 1, characterized in that, The various sensors include a wind speed sensor, an air volume sensor, a pressure sensor, a dust concentration sensor, and a gas concentration sensor. The wind speed sensor and air volume sensor are arranged in the intake and return air roadways, the pressure sensor is arranged in the main ventilation roadway and local branches, the dust concentration sensor is arranged in the working area and the return air side, and the gas concentration sensor is arranged in the roof, the return air inlet, and in front of the tunneling machine.

9. A multimodal intelligent control-based ventilation-dust control coordinated adjustment system for tunneling roadways, characterized in that, include: The multimodal sensing module is used to acquire multimodal data collected in real time by various sensors, preprocess the multimodal data to obtain multidimensional time-series data; the multimodal data includes wind speed data, dust concentration data, gas concentration data and ventilation pressure data; The trend prediction module is used to predict the multimodal data based on the multidimensional time series data through a prediction model, and obtain prediction results. The prediction results include the future change trend and fluctuation range of each modality in the multimodal data. The pressure balance control module is used to perform pressure balance calculations and generate constraint conditions based on the future changing trends and fluctuation amplitudes of each modality data in the multimodal data and the multimodal data. The multi-objective optimization module is used to take the constraints as boundary constraints, and solve the multi-objective optimization model based on the future change trend and fluctuation range of each modality data in the multi-modal data and the real-time equipment operating parameters to obtain the optimal ventilation and dust control adjustment strategy. The optimal ventilation and dust control adjustment strategy includes the expected control quantities of the fan frequency converter, the air distribution valve, and the air curtain angle adjustment mechanism. The fuzzy control module is used to adaptively adjust the frequency of the fan inverter, the opening degree of the air distribution valve, and the angle of the air curtain based on the desired control quantity and the multimodal data using a fuzzy PID algorithm.

10. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1-8.

Images