A method and system for controlling dust in a tunneling face

Abstract

The application provides a tunneling working face dust collection method and system. The method comprises the following steps: collecting real-time measured environmental parameters of a tunneling working face by using a sensor, and preprocessing the measured environmental parameters to obtain preprocessed measured environmental parameters; obtaining target environmental parameters of the tunneling working face, and determining an environmental parameter difference absolute value according to the target environmental parameters and the preprocessed measured environmental parameters; determining an actuator adjustment instruction value according to the environmental parameter difference absolute value and by using a PID control algorithm; and controlling a tunneling working face dust collection system based on the actuator adjustment instruction value. The actuator adjustment instruction value comprises an angle change amount of a guide vane in a negative pressure air duct body, an opening size change amount of a air distributor in the negative pressure air duct body, and a fan rotating speed. The technical scheme improves the safety and environmental protection of tunneling operation.

Description

Technical Field

[0001] This application relates to the field of safety protection and dust control technology in mining tunneling engineering, and in particular to a method and system for dust collection and control at a tunneling face. Background Technology

[0002] During mining tunneling operations, a large amount of dust is generated at the tunneling face. At the same time, factors such as ventilation conditions and gas leaks can easily lead to safety hazards such as excessive dust, gas accumulation, and abnormal temperature, which seriously threaten the health of workers and the safety of equipment. Traditional ventilation and dust control in tunneling faces often rely on fixed parameter control methods, such as manually adjusting fan speed and fixing guide vane angles. This approach suffers from several problems: Lagging response: Reliance on manual monitoring and adjustment makes it impossible to capture real-time changes in environmental parameters. When dust concentration, wind speed, or gas concentration becomes abnormal, adjustments are delayed, potentially leading to increased safety risks. Low control precision: Fixed parameter control cannot differentiate based on dust distribution and temperature variations in different areas of the working face. Some areas are prone to excessively high wind speeds (above 0.6 m / s, resulting in energy waste) or excessively low wind speeds (below 0.3 m / s, leading to dust accumulation). Insufficient safety: Lack of a rapid response mechanism to gas concentration fluctuations and the absence of automatic pressure relief protection for abnormal system pressure can easily lead to gas leaks, backflow, or equipment pressure damage. Weak adaptability: The inability to predict environmental parameter trends based on historical data limits responses to current anomalies, hindering proactive control. Therefore, there is an urgent need to propose a dust control system that can monitor in real time, predict intelligently, and regulate precisely, in order to solve the defects of traditional control methods and improve the safety protection and dust control efficiency of the tunneling face. Summary of the Invention

[0003] This application provides a method and system for dust control in tunneling faces, which at least solves the technical problems of slow response, low control accuracy, insufficient safety and weak adaptability in existing ventilation and dust control systems for tunneling faces.

[0004] The first aspect of this application provides a method for dust control at a tunneling face, the method comprising:

[0005] The measured environmental parameters of the tunneling face are collected in real time using sensors, and the measured environmental parameters are preprocessed to obtain the preprocessed measured environmental parameters.

[0006] Obtain the target environmental parameters of the tunneling face, and determine the absolute value of the difference between the environmental parameters based on the target environmental parameters and the preprocessed measured environmental parameters.

[0007] The adjustment command value of the actuator is determined based on the absolute value of the difference in the environmental parameters and using a PID control algorithm.

[0008] The dust control system at the tunneling face is controlled based on the adjustment command value of the actuator.

[0009] The actuator adjustment command values ​​include: the change in the angle of the guide vanes inside the negative pressure duct body, the change in the opening size of the air distributor inside the negative pressure duct body, and the fan speed.

[0010] Preferably, the measured environmental parameters include: dust concentration data, airflow temperature data, wind speed data, harmful gas concentration data, and pressure change data inside the duct.

[0011] The target environmental parameters include: target wind speed, target temperature, target dust concentration, target gas concentration, and target pressure.

[0012] Furthermore, the preprocessing of the measured environmental parameters to obtain preprocessed measured environmental parameters includes:

[0013] The wind speed data is smoothed using the moving average method, and missing values ​​are filled using mean fill and forward fill methods. Then the wind speed data is standardized to obtain preprocessed wind speed data.

[0014] The airflow temperature data is processed using the Kalman filter algorithm, and missing values ​​are filled using forward imputation or interpolation. Then, the airflow temperature data is standardized to obtain preprocessed airflow temperature data.

[0015] The IQR method is used to identify outliers in the dust concentration data and remove them. Then, the moving average method is used to smooth the dust concentration data after removing outliers. The smoothed dust concentration data is then processed using mean filling or interpolation to obtain filled dust concentration data. The filled dust concentration data is then standardized to obtain preprocessed dust concentration data.

[0016] The Z-Score method is used to identify outliers in the concentration data of the harmful gas and remove them. Then, linear interpolation or polynomial interpolation is used to process the concentration data of the harmful gas after removing outliers to obtain the concentration data of the harmful gas after filling. The concentration data of the harmful gas after filling is then standardized to obtain the concentration data of the harmful gas after preprocessing.

[0017] The pressure change data inside the ventilation duct is standardized to obtain pre-processed pressure change data inside the ventilation duct.

[0018] Furthermore, the step of determining the actuator adjustment command value based on the absolute value of the environmental parameter difference and using a PID control algorithm includes:

[0019] Determine whether the absolute value of the difference between each environmental parameter is greater than its threshold value for a given parameter. If so, use the PID calculation formula. Determine the actuator adjustment instruction value corresponding to the t-th environmental parameter; otherwise, no adjustment is made. Adjust the instruction value for the actuator corresponding to the t-th environmental parameter. This is the proportionality coefficient. This represents the difference between the target value and the measured value corresponding to the t-th environmental parameter. The integral coefficient is... is the differential coefficient.

[0020] Furthermore, the method also includes:

[0021] The current tunneling machine power, propulsion speed, and fan speed are obtained, and the current tunneling machine power, propulsion speed, and fan speed are normalized to obtain the pre-processed current tunneling machine power, propulsion speed, and fan speed.

[0022] The pre-processed current-time tunneling machine power, propulsion speed, fan speed, pre-processed wind speed data, pre-processed airflow temperature data, pre-processed dust concentration data, and pre-processed harmful gas concentration data are input into the pre-established dust concentration prediction model to obtain the dust concentration prediction values ​​at the first time, the second time, and the third time.

[0023] Determine whether the predicted dust concentration values ​​at the first time, the second time, and the third time are all less than or equal to a preset dust concentration threshold. If not, filter out the predicted dust concentration values ​​that are greater than the preset dust concentration threshold.

[0024] Based on the selected dust concentration prediction values ​​that are greater than the preset dust concentration threshold, the PID control algorithm is used to determine the actuator adjustment command value.

[0025] Furthermore, the process of establishing the dust concentration prediction model includes:

[0026] Acquire the pre-processed tunneling machine power, propulsion speed, fan speed, pre-processed wind speed data, pre-processed airflow temperature data, pre-processed dust concentration data, pre-processed harmful gas concentration data, and the measured dust concentration values ​​at the first, second, and third moments corresponding to each moment within the historical period.

[0027] The preprocessed data on tunneling machine power, propulsion speed, fan speed, preprocessed wind speed, preprocessed airflow temperature, preprocessed dust concentration, and preprocessed harmful gas concentration at each moment within the historical time period are used as inputs. The measured dust concentration values ​​at the first, second, and third moments corresponding to each moment within the historical time period are used as outputs. The initial neural network model is trained using mean square error as the loss function to obtain a trained dust concentration prediction model.

[0028] Furthermore, the method also includes:

[0029] The optimal wind speed, temperature, and dust concentration were determined using a multi-objective optimization algorithm.

[0030] The optimal wind speed, optimal temperature, and optimal dust concentration are respectively used as the optimized target wind speed, optimized target temperature, and optimized target dust concentration.

[0031] Based on the optimized target wind speed, optimized target temperature, and optimized target dust concentration, the PID control algorithm is used to determine the actuator adjustment command value.

[0032] Furthermore, the method also includes:

[0033] The optimal wind speed is determined based on the optimized target dust concentration. If the absolute value of the difference between the optimal wind speed and the optimized target wind speed is greater than a preset wind speed threshold, then the optimal wind speed is taken as the target wind speed.

[0034] Based on the pressure change data inside the duct, it is determined whether the pressure inside the duct is less than a preset first pressure threshold. If so, the fan speed is adjusted according to the preset first fan speed adjustment strategy.

[0035] Based on the pressure change data inside the duct, it is determined whether the pressure inside the duct is greater than the preset second pressure threshold. If so, the fan speed is adjusted according to the preset second fan speed adjustment strategy.

[0036] Furthermore, the first fan speed adjustment strategy includes: increasing the fan speed according to a preset speed increase value;

[0037] The second fan speed adjustment strategy includes: reducing the fan speed according to a preset speed reduction value.

[0038] A second aspect of this application provides a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the method described in the first aspect.

[0039] The technical solutions provided by the embodiments of this application bring at least the following beneficial effects:

[0040] This application proposes a method and system for dust control at a tunneling face. The method includes: real-time acquisition of measured environmental parameters at the tunneling face using sensors, and preprocessing the measured environmental parameters to obtain preprocessed measured environmental parameters; acquisition of target environmental parameters at the tunneling face, and determination of the absolute value of the difference between the target environmental parameters and the preprocessed measured environmental parameters; determination of adjustment command values ​​for the actuator based on the absolute value of the environmental parameter difference and using a PID control algorithm; and control of the dust control system at the tunneling face based on the actuator adjustment command values. The actuator adjustment command values ​​include: the change in the angle of the guide vanes within the negative pressure ventilation duct, the change in the opening size of the air distributor within the negative pressure ventilation duct, and the fan speed. The technical solution proposed in this application improves the safety and environmental friendliness of tunneling operations.

[0041] 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

[0042] 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:

[0043] Figure 1 This is a flowchart illustrating a dust control method for a tunneling face according to an embodiment of this application;

[0044] Figure 2 This is a structural diagram of a dust control system for a tunneling face according to an embodiment of this application. Detailed Implementation

[0045] 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.

[0046] This application proposes a method and system for dust control at a tunneling face. The method includes: real-time acquisition of measured environmental parameters at the tunneling face using sensors, and preprocessing the measured environmental parameters to obtain preprocessed measured environmental parameters; acquisition of target environmental parameters at the tunneling face, and determination of the absolute value of the difference between the target environmental parameters and the preprocessed measured environmental parameters; determination of actuator adjustment command values ​​based on the absolute value of the environmental parameter difference and using a PID control algorithm; and control of the dust control system at the tunneling face based on the actuator adjustment command values. The actuator adjustment command values ​​include: changes in the angle of the guide vanes within the negative pressure ventilation duct, changes in the opening size of the air distributor within the negative pressure ventilation duct, and the fan speed. The technical solution proposed in this application improves the safety and environmental friendliness of tunneling operations.

[0047] A method and system for dust control at a tunneling face, according to an embodiment of this application, is described below with reference to the accompanying drawings.

[0048] Example 1

[0049] Figure 1 This is a flowchart illustrating a dust control method for a tunneling face according to an embodiment of this application, as shown below. Figure 1 As shown, the method includes:

[0050] Step 1: Use sensors to collect real-time measured environmental parameters of the tunneling face, and preprocess the measured environmental parameters to obtain preprocessed measured environmental parameters;

[0051] It should be noted that the measured environmental parameters include: dust concentration data, airflow temperature data, wind speed data, harmful gas concentration data, and pressure change data inside the ventilation duct.

[0052] In this embodiment of the disclosure, the preprocessing of the measured environmental parameters to obtain preprocessed measured environmental parameters includes:

[0053] The wind speed data is smoothed using the moving average method, and missing values ​​are filled using mean fill and forward fill methods. Then the wind speed data is standardized to obtain preprocessed wind speed data.

[0054] The airflow temperature data is processed using the Kalman filter algorithm, and missing values ​​are filled using forward imputation or interpolation. Then, the airflow temperature data is standardized to obtain preprocessed airflow temperature data.

[0055] The IQR method is used to identify outliers in the dust concentration data and remove them. Then, the moving average method is used to smooth the dust concentration data after removing outliers. The smoothed dust concentration data is then processed using mean filling or interpolation to obtain filled dust concentration data. The filled dust concentration data is then standardized to obtain preprocessed dust concentration data.

[0056] The Z-Score method is used to identify outliers in the concentration data of the harmful gas and remove them. Then, linear interpolation or polynomial interpolation is used to process the concentration data of the harmful gas after removing outliers to obtain the concentration data of the harmful gas after filling. The concentration data of the harmful gas after filling is then standardized to obtain the concentration data of the harmful gas after preprocessing.

[0057] The pressure change data inside the ventilation duct is standardized to obtain pre-processed pressure change data inside the ventilation duct.

[0058] It should be noted that the processing of wind speed sensor data involves the following:

[0059] Wind speed sensor readings can be affected by environmental noise and sensor malfunctions, therefore data processing is necessary to ensure the stability and accuracy of the control system. This processing includes noise reduction, missing value imputation, outlier identification, and data standardization.

[0060] Processing steps:

[0061] 1. Noise Reduction (Smoothing): A moving average method is used to remove high-frequency noise from the sensor data. Specifically, the average value of the data within the moving window is taken to reduce the impact of individual measurements on the overall data.

[0062] Speed=[x1,x2,x3,x4,x5,x6,x7,x8,x9]

[0063] First smoothed data: x= ;

[0064] The second smoothed data: x= ;

[0065] This process continues until all smoothed data that meets the window requirements is calculated, ultimately yielding the smoothed wind speed data.

[0066] 2. Missing value imputation: The continuity of data is ensured by using the forward imputation method (replacing missing values ​​with the previous valid values).

[0067] 3. Outlier Detection: Outliers are detected using the quartile method. The upper and lower quartiles of the data are calculated, and data exceeding a predetermined threshold are removed. This prevents outliers from impacting overall analysis and control.

[0068] 4. Standardization Processing: Wind speed data is standardized to facilitate comparison and analysis with data from other sensors, such as temperature and pressure. The standardization formula is:

[0069] z=(x-μ) / σ

[0070] Where μ is the mean wind speed and σ is the standard deviation. Standardized data helps eliminate differences in dimensions.

[0071] Processing temperature sensor data:

[0072] Temperature sensors are generally not prone to severe noise, but they can still be affected by external interference, leading to abnormal or missing data.

[0073] Denoising (Kalman Filtering): Kalman filtering can effectively remove high-frequency noise from temperature data while retaining low-frequency variations.

[0074] Initialization: Set initial state mean and system noise covariance, etc.

[0075] Prediction Step: Based on the system's state transition equation, predict the current state estimate and error covariance. The state transition equation for the temperature system is as follows: Where A is the state transition matrix, The process noise follows a normal distribution with a mean of 0 and a covariance of Q; the observation equation is... Where C is the observation matrix, The observed noise follows a normal distribution with a mean of 0 and a covariance of R.

[0076] Update step: Calculate the Kalman gain based on the current observations. ;

[0077] Then update the state estimate: ;

[0078] Estimation and error covariance: ;

[0079] Iteration: Repeat the prediction and update steps to filter the entire temperature data sequence and obtain the filtered temperature data.

[0080] Missing value imputation: Filling missing data with neighboring data (such as forward imputation), suitable for continuously changing environmental data.

[0081] For temperature data with missing values, the missing value is filled with the value of the preceding non-missing data. This method is suitable for environmental data such as temperature that changes continuously and slowly.

[0082] Processing procedure: Temperature data with missing values ​​is as follows

[0083] The first missing value (located in the second position) is filled with the previous non-missing data t1.

[0084] The second missing value (located at the sixth position) is filled with the previous non-missing data t5, resulting in the temperature data after forward filling.

[0085] Interpolation: If data loss is severe, linear interpolation can be used to estimate the missing data points.

[0086] For missing data, a linear equation is constructed using the two sets of data before and after it to obtain the missing data.

[0087] Processing of dust sensor data:

[0088] Dust sensor data may fluctuate significantly, especially during mining operations when dust concentration may suddenly increase. The following method can be used to mitigate this:

[0089] Outlier detection (IQR method): The quartile method (IQR) is used to detect and remove outliers, ensuring that the data is not affected by extreme values.

[0090] Assume the dust concentration is [x1,x2,x3,x4,x5,x6,x7,x8,x9];

[0091] Sort the data from smallest to largest to obtain the sorted sequence x(1),x(2)……x(9),x(10);

[0092] Calculate the first quartile (Q1);

[0093] Determining the position of Q1: Quartiles are three values ​​that divide sorted data into four equal parts. Q1 is the first quartile (lower quartile), representing 25% of the values ​​in the data that are less than or equal to it. Since this set of data has n=10 values, according to the calculation rules, the position of Q1 is calculated using the formula "(n+1)÷4". Substituting n=10, we get: Q1 position = (10+1)÷4 = 2.75. This position indicates that Q1 is located between the second and third data points in the sorted sequence, and closer to the third data point.

[0094] Calculate the value of Q1: Considering the position of Q1 (2.75), its value needs to be obtained from the second sorted data d. (2) And the third data d (3) The calculation is as follows: Q1 = d(2) +0.75×(d (3) -d (2) The value of 0.75 is determined by the decimal part (0.75) of the 2.75 position in Q1. This means that 75% of the difference between the second and third data points needs to be taken and then added to the second data point to obtain the specific value of Q1.

[0095] IV. Calculate the third quartile (Q3):

[0096] Determine the position of Q3: Q3 is the third quartile (upper quartile), representing 75% of the data values ​​that are less than or equal to it. Its position is calculated using the formula "3 × (n + 1) ÷ 4". Substituting n = 10, we get: Q3's position = 3 × (10 + 1) ÷ 4 = 8.25. This position indicates that Q3 is located between the 8th and 9th data points in the sorted sequence, and closer to the 8th data point.

[0097] Calculate the value of Q3: Based on the position of Q3 (8.25), its value needs to be obtained from the 8th data point after sorting (d). (8) ) and the 9th data (d (9) ) Calculation. The specific formula is: Q3=d (8) +0.25×(d (9) -d (8) The 0.25 here refers to the decimal part (0.25) of 8.25 at position Q3. This means taking 25% of the difference between the 8th and 9th data points and adding it to the 8th data point to obtain the specific value of Q3.

[0098] V. Calculate the interquartile range (IQR)

[0099] The interquartile range (IQR) is the difference between Q3 and Q1. It reflects the dispersion of the middle 50% of the data. The calculation formula is: IQR = Q3 - Q1. Substituting the previously calculated values ​​of Q3 and Q1, the IQR of this set of dust concentration data can be obtained.

[0100] VI. Determine the boundaries for outlier detection:

[0101] To distinguish between normal and abnormal data, IQR needs to be used to determine the judgment boundaries, including the lower bound and the upper bound:

[0102] Calculate the lower boundary: The lower boundary is the critical value for determining whether the data is an "outlier". The calculation formula is: lower boundary = Q1 - 1.5 × IQR, that is, subtract 1.5 times IQR from Q1 to get the lower boundary value.

[0103] Calculate the upper boundary: The upper boundary is the critical value for judging whether the data is an "excessively high outlier". The calculation formula is: upper boundary = Q3 + 1.5 × IQR, that is, add 1.5 times IQR to Q3 to get the upper boundary value.

[0104] 7. Identify outliers and determine their locations:

[0105] The original 10 dust concentration data (d1 to d) 10 Each element is compared with the lower and upper boundaries:

[0106] If a data point is less than the lower boundary or greater than the upper boundary, then that data point is considered an outlier.

[0107] If a data point is greater than or equal to the lower boundary and less than or equal to the upper boundary, then that data point is considered normal.

[0108] Record all outliers in the original data sequence [d1, d2, ..., d... 10 The position in ] (i.e., the outlier is from d1 to d) 10 The set of these locations (which number of data points in the dataset) is called outlier locations.

[0109] Smoothing (moving average): Similar to wind speed processing, smoothing helps to remove sudden spikes in data.

[0110] Missing value imputation: Missing data is handled using the same mean imputation or interpolation method. The smoothing principle and calculation process are consistent with the moving average smoothing of wind speed sensor data, with a window size of 3, to smooth the dust concentration data.

[0111] Processing gas sensor data:

[0112] Gas sensors (such as CO2 and CO sensors) are used to monitor the concentration of harmful gases in mines. Gas concentration fluctuations can be significant, therefore noise reduction and outlier detection are necessary. Methods include:

[0113] Outlier detection (Z-Score method): Z-Score can identify data points that exceed a set threshold and mark them as outliers.

[0114] (I) Outlier Detection (Z-Score Method) Principle: The Z-Score method calculates the Z-Score value of each data point by dividing the deviation of the data point from the mean by the standard deviation. Data points with an absolute Z-Score greater than 3 are generally considered outliers because in a normal distribution, approximately 99.7% of the data will fall within the range of the mean plus or minus 3 standard deviations.

[0115] Calculate the average gas concentration: x = ;

[0116] Calculate the sample standard deviation μ

[0117] Calculate the Z-Score for each data point: ;

[0118] Determine if the absolute value of each z-value is greater than 3. If it is, the data point is considered an outlier, and the location of the outlier is determined.

[0119] It should be noted that data standardization and normalization are:

[0120] Because different sensors output different units (e.g., wind speed in m / s, temperature in °C), the data needs to be standardized or normalized to ensure they have the same dimensions so that they can be processed together in the same model.

[0121] Map the original data to the interval [0,1] x= .

[0122] Step 2: Obtain the target environmental parameters of the tunneling face, and determine the absolute value of the difference between the environmental parameters based on the target environmental parameters and the pre-processed measured environmental parameters;

[0123] It should be noted that the target environmental parameters include: target wind speed, target temperature, target dust concentration, target gas concentration, and target pressure.

[0124] Step 3: Determine the actuator adjustment command value based on the absolute value of the difference in the environmental parameters and using the PID control algorithm; wherein, the actuator adjustment command value includes: the change in the angle of the guide vanes in the negative pressure duct body, the change in the opening size of the air distributor in the negative pressure duct body, and the fan speed.

[0125] In this embodiment of the disclosure, step 3 specifically includes:

[0126] Determine whether the absolute value of the difference between each environmental parameter is greater than its threshold value for a given parameter. If so, use the PID calculation formula. Determine the actuator adjustment instruction value corresponding to the t-th environmental parameter; otherwise, no adjustment is made. Adjust the instruction value for the actuator corresponding to the t-th environmental parameter. This is the proportionality coefficient. This represents the difference between the target value and the measured value corresponding to the t-th environmental parameter. The integral coefficient is... is the differential coefficient.

[0127] It should be noted that the PID control algorithm is the core of the system's "real-time monitoring and dynamic adjustment." Using "wind speed, dust concentration below safety standards, and temperature as specified in coal mine safety regulations" as target values, the algorithm calculates the control quantity through the coordinated calculation of proportional (P), integral (I), and derivative (D) components, driving the actuators such as guide vanes and air distributors. The specific process is as follows:

[0128] The parameters are defined as follows:

[0129] Target value S: Preset safety environment parameters, such as target wind speed. Dust concentration target Measured value Y: Real-time sensor data after preprocessing, such as current wind speed. Current dust concentration Deviation value The difference between the target value and the measured value, i.e. Control quantity The algorithm outputs adjustment commands for the actuators (such as changes in guide vane angle and changes in distributor opening size). The proportional coefficient directly outputs the control quantity based on the magnitude of the deviation, quickly offsetting the deviation. The integral coefficient is used to eliminate static deviations (such as a slight leak in the air distributor after long-term ventilation, causing the air velocity to remain below the target value). By accumulating the deviation through integration, when e(t) is not zero for a long period, the integral term gradually increases, driving the control variable to continuously adjust until e(t) = 0. The differential coefficient predicts the trend of deviation changes and suppresses overshoot. When the deviation increases rapidly (e.g., a sudden increase in dust concentration after an blast), the differential term calculates the rate of change of deviation and outputs the reverse control quantity in advance. To avoid excessive adjustments to the implementing agency.

[0130] Specifically:

[0131] 1. Real-time calculation of deviation ,like (δ is the allowable error), then the control quantity is not adjusted.

[0132] 2. If Then, the control quantity u(t) is calculated using the PID formula and converted into a specific action command for the actuator.

[0133] 3. After the action is executed, the measured value Y(t) is fed back in real time by the sensor. The deviation is recalculated, and the next round of the PID loop begins, until... .

[0134] Step 4: Control the dust control system at the tunneling face based on the adjustment command value of the actuator;

[0135] In this embodiment of the disclosure, the method further includes:

[0136] The current tunneling machine power, propulsion speed, and fan speed are obtained, and the current tunneling machine power, propulsion speed, and fan speed are normalized to obtain the pre-processed current tunneling machine power, propulsion speed, and fan speed.

[0137] The pre-processed current-time tunneling machine power, propulsion speed, fan speed, pre-processed wind speed data, pre-processed airflow temperature data, pre-processed dust concentration data, and pre-processed harmful gas concentration data are input into the pre-established dust concentration prediction model to obtain the dust concentration prediction values ​​at the first time, the second time, and the third time.

[0138] Determine whether the predicted dust concentration values ​​at the first time, the second time, and the third time are all less than or equal to a preset dust concentration threshold. If not, filter out the predicted dust concentration values ​​that are greater than the preset dust concentration threshold.

[0139] Based on the selected dust concentration prediction values ​​that are greater than the preset dust concentration threshold, the PID control algorithm is used to determine the actuator adjustment command value.

[0140] Furthermore, the process of establishing the dust concentration prediction model includes:

[0141] Acquire the pre-processed tunneling machine power, propulsion speed, fan speed, pre-processed wind speed data, pre-processed airflow temperature data, pre-processed dust concentration data, pre-processed harmful gas concentration data, and the measured dust concentration values ​​at the first, second, and third moments corresponding to each moment within the historical period.

[0142] The preprocessed data on tunneling machine power, propulsion speed, fan speed, preprocessed wind speed, preprocessed airflow temperature, preprocessed dust concentration, and preprocessed harmful gas concentration at each moment within the historical time period are used as inputs. The measured dust concentration values ​​at the first, second, and third moments corresponding to each moment within the historical time period are used as outputs. The initial neural network model is trained using mean square error as the loss function to obtain a trained dust concentration prediction model.

[0143] It should be noted that the dust concentration prediction specifically includes:

[0144] 1. Data Preparation and Preprocessing

[0145] Build a historical database: Collect historical data under different operating conditions over the past year, including input features and output labels.

[0146] Input features (independent variables): tunneling machine power, propulsion speed, fan speed, current wind speed, current temperature, current dust concentration, and gas concentration (a total of 7 types of parameters, with 1 data point collected every 10 seconds to form a time series).

[0147] Output labels (dependent variable): Dust concentration values ​​for the next 5, 10, and 15 minutes (corresponding to 3 prediction dimensions to meet different control lead time requirements).

[0148] Data normalization: Mapping all feature data to the [0,1] interval to avoid the impact of unit differences on model training. The formula is as follows: ;

[0149] Time series partitioning: The normalized data is partitioned into input sequences according to "time step = 30" (i.e., 300 seconds of data), and the corresponding output label is the dust concentration in the next 5 minutes, forming training samples.

[0150] 2. LSTM Model Structure and Training

[0151] Model layer count: "Input layer + 2 LSTM hidden layers + fully connected output layer" structure

[0152] Input layer: Receives sequence data with 30 time steps × 7 features, with input dimensions of (None, 30, 7).

[0153] The first LSTM hidden layer has 64 neurons, uses the tanh activation function, preserves sequence information, and has an output dimension of (None, 30, 64).

[0154] The second LSTM hidden layer has 32 neurons and uses the tanh activation function to further extract deep features. The output dimension is (None, 32).

[0155] Fully connected output layer: 3 neurons (corresponding to the predicted dust concentration values ​​for the next 5, 10, and 15 minutes), with a linear activation function and an output dimension of (None, 3).

[0156] Training parameters: The Adam optimizer is used, with a learning rate of 0.001; the loss function is mean squared error (MSE), which measures the deviation between the predicted and actual values; the number of training epochs is set to 100, and the batch size is set to 32.

[0157] Model validation and optimization: The training and test sets are divided in an 8:2 ratio. The model performance is validated every 10 training rounds. If the MSE of the test set does not decrease for 5 consecutive rounds, an early stopping strategy is adopted to prevent overfitting, and the number of neurons or the learning rate is adjusted to optimize the model.

[0158] 3. Prediction and Regulation Triggers

[0159] Real-time prediction: Input the preprocessed data from the 30 time steps prior to the current moment into the trained LSTM model to obtain predicted dust concentration values ​​for the next 5, 10, and 15 minutes.

[0160] Control judgment: If any predicted value exceeds the safety standard, an "advance control instruction" is triggered, and the predicted value is used as the "pre-deviation" of the PID control algorithm to adjust the control quantity in advance.

[0161] In this embodiment of the disclosure, the method further includes:

[0162] The optimal wind speed, temperature, and dust concentration were determined using a multi-objective optimization algorithm.

[0163] The optimal wind speed, optimal temperature, and optimal dust concentration are respectively used as the optimized target wind speed, optimized target temperature, and optimized target dust concentration.

[0164] Based on the optimized target wind speed, optimized target temperature, and optimized target dust concentration, the PID control algorithm is used to determine the actuator adjustment command value.

[0165] It should be noted that the multi-objective optimization algorithm is used to solve the multi-objective conflict problem of "wind speed, temperature, and dust concentration" (e.g., increasing airflow can reduce dust concentration, but may lead to excessively low temperature or excessively high energy consumption). The algorithm determines the optimal control parameters by finding the optimal parameters, and the specific process is as follows:

[0166] Objectives and constraints: Minimize dust concentration and system energy consumption, ensure wind speed is within the specified range, temperature is within the safe range, and gas concentration is ≤1% LEL and system pressure is ≤0.15MPa.

[0167] Simplify the execution steps:

[0168] Randomly generate initial control parameter solutions (such as blade angle, distributor opening, fan speed);

[0169] For each solution, calculate the objective function value and sort them according to "dominance relationship";

[0170] The optimal solution (prioritizing dust control while considering energy consumption, wind speed, and temperature) is selected as the basis for adjusting the target value of the PID algorithm.

[0171] In this embodiment of the disclosure, the method further includes:

[0172] The optimal wind speed is determined based on the optimized target dust concentration. If the absolute value of the difference between the optimal wind speed and the optimized target wind speed is greater than a preset wind speed threshold, then the optimal wind speed is taken as the target wind speed.

[0173] Based on the pressure change data inside the duct, it is determined whether the pressure inside the duct is less than a preset first pressure threshold. If so, the fan speed is adjusted according to the preset first fan speed adjustment strategy.

[0174] Based on the pressure change data inside the duct, it is determined whether the pressure inside the duct is greater than the preset second pressure threshold. If so, the fan speed is adjusted according to the preset second fan speed adjustment strategy.

[0175] The first fan speed adjustment strategy includes: increasing the fan speed according to a preset speed increase value;

[0176] The second fan speed adjustment strategy includes: reducing the fan speed according to a preset speed reduction value.

[0177] It should be noted that for wind speed optimization: a wind speed-air volume-blade angle correlation model is established, and the required air volume is deduced from the dust target value of multi-objective optimization to calculate the optimal wind speed; if the deviation from the PID target wind speed is large, the PID target value is corrected.

[0178] Pressure optimization: Establish a pressure-speed-airflow correlation model to monitor the pressure difference in real time; if the pressure is too low, increase the fan speed, and if it is too high, decrease the speed; feed the speed adjustment amount back to the PID controller to ensure that the airflow still meets the standard after the pressure is adjusted.

[0179] In this embodiment of the disclosure, the method further includes:

[0180] Determine whether the preprocessed wind speed data is greater than or equal to a preset minimum wind speed and less than or equal to a preset maximum wind speed. If the preprocessed wind speed data is less than the preset minimum wind speed, then the preset minimum wind speed is used as the initial preprocessed wind speed data. If the preprocessed wind speed data is greater than the preset maximum wind speed, then the preset maximum wind speed is used as the initial preprocessed wind speed data.

[0181] If the pressure change data inside the pre-processed air duct is greater than or equal to the preset minimum pressure and less than or equal to the preset maximum pressure, and if the pressure change data inside the pre-processed air duct is less than the preset minimum pressure, then the initial pre-processed wind speed data is increased by 0.1, and the value of the initial pre-processed wind speed data after increasing by 0.1 is less than or equal to the preset maximum wind speed, and the initial pre-processed wind speed data after increasing by 0.1 is used as the pre-processed wind speed data.

[0182] If the pressure change data inside the pre-processed duct is greater than the preset minimum pressure value, then the initial pre-processed wind speed data is reduced by 0.1, and the value of the initial pre-processed wind speed data after the reduction of 0.1 is greater than or equal to the preset minimum wind speed value, and the initial pre-processed wind speed data after the reduction of 0.1 is used as the pre-processed wind speed data.

[0183] It should be noted that regarding the adjustment of the guide vane angle and wind speed control: Using a PID control algorithm, the system dynamically adjusts the guide vane angle based on real-time monitored wind speed data. When the wind speed is too low (e.g., below 0.3 m / s), the vanes will adjust to a larger angle (e.g., +45°) to increase airflow; when the wind speed is too high (above 0.6 m / s), the vanes will tighten to a smaller angle (e.g., -15°) to reduce airflow, ensuring the wind speed remains between 0.3 and 0.6 m / s. Furthermore, the vane angle adjustment can guide airflow to high-dust areas, improving dust collection efficiency. Distributor opening size adjustment: Airflow distribution optimization: The system automatically adjusts the distributor opening size based on real-time changes in dust concentration and temperature. For example, when dust concentration is high (e.g., PM10 > 200 mg / m³), the system will increase the airflow by enlarging the distributor opening to accelerate dust dilution and emission; conversely, when the temperature is too high (exceeding the set upper temperature limit), the system will appropriately reduce the opening to limit excessive airflow, avoid energy waste, and ensure the temperature remains within a safe range. Wind speed feedback control: Wind speed adjustment: Wind speed sensors monitor wind speed changes after the distributor in real time and feed the data back to the central control system. If the monitored wind speed exceeds the preset range (below 0.3 m / s or above 0.6 m / s), the system will immediately stabilize the wind speed by adjusting the guide vane angle or the distributor opening size. The wind speed control strategy continuously optimizes airflow through the feedback system, ensuring stable operation of the system under changing working conditions. Pressure differential sensor adjustment: Pressure balance adjustment: Pressure differential sensors are installed before and after the distributor to monitor airflow pressure changes. Through linkage with the fan, the system adjusts the fan speed in real time. When the pressure difference is small (e.g., below a preset threshold), the system increases the fan speed and air volume to improve airflow velocity and effective ventilation. When the pressure is too high (e.g., exceeding a preset upper limit), the system reduces the fan speed to avoid energy waste caused by excessive airflow and ensure efficient and stable system operation. Backflow Prevention Device: Gas Flow Direction Control: When the gas concentration approaches a safe threshold (e.g., 1% LEL), the system automatically activates the backflow prevention device. The flap valve closes quickly to prevent backflow into the work area or equipment area, ensuring unidirectional airflow and preventing leakage of harmful gases. The backflow prevention device has a very short response time, providing rapid safety assurance during gas concentration fluctuations. Explosion Relief Structure (Pressure Protection): Automatic Pressure Relief Function: When the internal system pressure exceeds a set threshold (e.g., 0.15 MPa), the explosion relief structure automatically activates, releasing excess pressure to a safe area via a rupture disc, preventing equipment damage due to excessive pressure. This device responds quickly to pressure fluctuations, ensuring system safety. Multi-channel diversion design: The system adopts a multi-channel diversion design, which can flexibly adjust the airflow direction to different areas according to the needs.For example, when the dust concentration is high on the right side of the working face, the guide vanes on the right will expand to +30°, guiding more airflow to the right, while the vanes on the left will automatically retract to -10°, reducing the airflow on the left, thus effectively guiding the dust to the dust collection port. Each guide vane is precisely controlled by a servo motor and can be independently adjusted according to the instructions of the central control system to achieve precise airflow distribution. Wind speed sensors, temperature sensors, dust sensors, and other equipment monitor environmental changes in various areas of the working face in real time and calculate the optimal airflow distribution strategy through intelligent algorithms.

[0184] The method proposed in this application achieves efficient ventilation and dust control at the tunneling face by combining real-time monitoring with multi-sensor technology, dynamic control with intelligent algorithms, and precise response from actuators. In terms of algorithms, the system integrates PID control, a machine learning prediction model (LSTM), a multi-objective optimization algorithm, and a wind speed and pressure optimization algorithm. The PID control algorithm dynamically adjusts wind speed, airflow, and airflow direction based on real-time data such as wind speed, temperature, and dust concentration, ensuring that the wind speed at the working face is within the range of 0.3~0.6 m / s and the dust concentration is below the safety standard. The machine learning prediction model (LSTM) predicts future dust concentration trends using historical data, thereby adjusting airflow and wind speed in advance to prevent dust concentration from exceeding the standard. The multi-objective optimization algorithm considers multiple factors such as wind speed, temperature, and dust concentration, optimizing airflow and wind speed through real-time feedback data to ensure optimal ventilation and dust control effects. The system adjusts the operation of each component based on real-time data to ensure the accuracy of wind speed and dust control.

[0185] In summary, the dust control method proposed in this embodiment is a dust control system that enables real-time monitoring, intelligent prediction, and precise regulation, thereby addressing the shortcomings of traditional control methods and improving the safety protection and dust control efficiency of the tunneling face.

[0186] Example 2

[0187] To achieve the above embodiments, this disclosure also provides a structural diagram of a dust control system for a tunneling face, as shown below. Figure 2 As shown, the system includes:

[0188] The acquisition module 100 is used to acquire measured environmental parameters of the tunneling face in real time using sensors, and to preprocess the measured environmental parameters to obtain preprocessed measured environmental parameters.

[0189] The measured environmental parameters include: dust concentration data, airflow temperature data, wind speed data, harmful gas concentration data, and pressure change data inside the ventilation duct.

[0190] The first determining module 200 is used to acquire the target environmental parameters of the tunneling face and determine the absolute value of the environmental parameter difference based on the target environmental parameters and the pre-processed measured environmental parameters.

[0191] The target environmental parameters include: target wind speed, target temperature, target dust concentration, target gas concentration, and target pressure.

[0192] The second determining module 300 is used to determine the actuator adjustment command value based on the absolute value of the difference in the environmental parameters and using a PID control algorithm.

[0193] Control module 400 is used to control the dust control system at the tunneling face based on the adjustment command value of the actuator;

[0194] The actuator adjustment command values ​​include: the change in the angle of the guide vanes inside the negative pressure duct body, the change in the opening size of the air distributor inside the negative pressure duct body, and the fan speed.

[0195] In this embodiment of the disclosure, the acquisition module 100 is further configured to:

[0196] The wind speed data is smoothed using the moving average method, and missing values ​​are filled using mean fill and forward fill methods. Then the wind speed data is standardized to obtain preprocessed wind speed data.

[0197] The airflow temperature data is processed using the Kalman filter algorithm, and missing values ​​are filled using forward imputation or interpolation. Then, the airflow temperature data is standardized to obtain preprocessed airflow temperature data.

[0198] The IQR method is used to identify outliers in the dust concentration data and remove them. Then, the moving average method is used to smooth the dust concentration data after removing outliers. The smoothed dust concentration data is then processed using mean filling or interpolation to obtain filled dust concentration data. The filled dust concentration data is then standardized to obtain preprocessed dust concentration data.

[0199] The Z-Score method is used to identify outliers in the concentration data of the harmful gas and remove them. Then, linear interpolation or polynomial interpolation is used to process the concentration data of the harmful gas after removing outliers to obtain the concentration data of the harmful gas after filling. The concentration data of the harmful gas after filling is then standardized to obtain the concentration data of the harmful gas after preprocessing.

[0200] The pressure change data inside the ventilation duct is standardized to obtain pre-processed pressure change data inside the ventilation duct.

[0201] In this embodiment of the disclosure, the second determining module 300 is further configured to:

[0202] Determine whether the absolute value of the difference between each environmental parameter is greater than its threshold value for a given parameter. If so, use the PID calculation formula. Determine the actuator adjustment instruction value corresponding to the t-th environmental parameter; otherwise, no adjustment is made. Adjust the instruction value for the actuator corresponding to the t-th environmental parameter. This is the proportionality coefficient. This represents the difference between the target value and the measured value corresponding to the t-th environmental parameter. The integral coefficient is... is the differential coefficient.

[0203] In this embodiment of the disclosure, the second determining module 300 is further configured to:

[0204] The current tunneling machine power, propulsion speed, and fan speed are obtained, and the current tunneling machine power, propulsion speed, and fan speed are normalized to obtain the pre-processed current tunneling machine power, propulsion speed, and fan speed.

[0205] The pre-processed current-time tunneling machine power, propulsion speed, fan speed, pre-processed wind speed data, pre-processed airflow temperature data, pre-processed dust concentration data, and pre-processed harmful gas concentration data are input into the pre-established dust concentration prediction model to obtain the dust concentration prediction values ​​at the first time, the second time, and the third time.

[0206] Determine whether the predicted dust concentration values ​​at the first time, the second time, and the third time are all less than or equal to a preset dust concentration threshold. If not, filter out the predicted dust concentration values ​​that are greater than the preset dust concentration threshold.

[0207] Based on the selected dust concentration prediction values ​​that are greater than the preset dust concentration threshold, the PID control algorithm is used to determine the actuator adjustment command value.

[0208] The process of establishing the dust concentration prediction model includes:

[0209] Acquire the pre-processed tunneling machine power, propulsion speed, fan speed, pre-processed wind speed data, pre-processed airflow temperature data, pre-processed dust concentration data, pre-processed harmful gas concentration data, and the measured dust concentration values ​​at the first, second, and third moments corresponding to each moment within the historical period.

[0210] The preprocessed data on tunneling machine power, propulsion speed, fan speed, preprocessed wind speed, preprocessed airflow temperature, preprocessed dust concentration, and preprocessed harmful gas concentration at each moment within the historical time period are used as inputs. The measured dust concentration values ​​at the first, second, and third moments corresponding to each moment within the historical time period are used as outputs. The initial neural network model is trained using mean square error as the loss function to obtain a trained dust concentration prediction model.

[0211] In this embodiment of the disclosure, the second determining module 300 is further configured to:

[0212] The optimal wind speed, temperature, and dust concentration were determined using a multi-objective optimization algorithm.

[0213] The optimal wind speed, optimal temperature, and optimal dust concentration are respectively used as the optimized target wind speed, optimized target temperature, and optimized target dust concentration.

[0214] Based on the optimized target wind speed, optimized target temperature, and optimized target dust concentration, the PID control algorithm is used to determine the actuator adjustment command value.

[0215] In this embodiment of the disclosure, the second determining module 300 is further configured to:

[0216] The optimal wind speed is determined based on the optimized target dust concentration. If the absolute value of the difference between the optimal wind speed and the optimized target wind speed is greater than a preset wind speed threshold, then the optimal wind speed is taken as the target wind speed.

[0217] Based on the pressure change data inside the duct, it is determined whether the pressure inside the duct is less than a preset first pressure threshold. If so, the fan speed is adjusted according to the preset first fan speed adjustment strategy.

[0218] Based on the pressure change data inside the duct, it is determined whether the pressure inside the duct is greater than the preset second pressure threshold. If so, the fan speed is adjusted according to the preset second fan speed adjustment strategy.

[0219] The first fan speed adjustment strategy includes: increasing the fan speed according to a preset speed increase value;

[0220] The second fan speed adjustment strategy includes: reducing the fan speed according to a preset speed reduction value.

[0221] In summary, the dust control system proposed in this embodiment is a dust control method for tunneling faces that enables real-time monitoring, intelligent prediction, and precise regulation, thereby addressing the shortcomings of traditional control methods and improving the safety protection and dust control efficiency of tunneling faces.

[0222] Example 3

[0223] To implement the above embodiments, this disclosure also proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in Embodiment 1.

[0224] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is 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.

[0225] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0226] 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 dust control at a tunneling face, characterized in that, The method includes: The measured environmental parameters of the tunneling face are collected in real time using sensors, and the measured environmental parameters are preprocessed to obtain the preprocessed measured environmental parameters. Obtain the target environmental parameters of the tunneling face, and determine the absolute value of the difference between the environmental parameters based on the target environmental parameters and the preprocessed measured environmental parameters. The adjustment command value of the actuator is determined based on the absolute value of the difference in the environmental parameters and using a PID control algorithm. The dust control system at the tunneling face is controlled based on the adjustment command value of the actuator. The actuator adjustment command values ​​include: the change in the angle of the guide vanes inside the negative pressure duct body, the change in the opening size of the air distributor inside the negative pressure duct body, and the fan speed; The method further includes: The current tunneling machine power, propulsion speed, and fan speed are obtained, and the current tunneling machine power, propulsion speed, and fan speed are normalized to obtain the pre-processed current tunneling machine power, propulsion speed, and fan speed. The pre-processed current-time tunneling machine power, propulsion speed, fan speed, pre-processed wind speed data, pre-processed airflow temperature data, pre-processed dust concentration data, and pre-processed harmful gas concentration data are input into the pre-established dust concentration prediction model to obtain the dust concentration prediction values ​​at the first time, the second time, and the third time. Determine whether the predicted dust concentration values ​​at the first time, the second time, and the third time are all less than or equal to a preset dust concentration threshold. If not, filter out the predicted dust concentration values ​​that are greater than the preset dust concentration threshold. Based on the selected dust concentration prediction values ​​that are greater than the preset dust concentration threshold, the PID control algorithm is used to determine the actuator adjustment command value; The method further includes: The optimal wind speed, temperature, and dust concentration were determined using a multi-objective optimization algorithm. The optimal wind speed, optimal temperature, and optimal dust concentration are respectively used as the optimized target wind speed, optimized target temperature, and optimized target dust concentration. Based on the optimized target wind speed, optimized target temperature, and optimized target dust concentration, the PID control algorithm is used to determine the actuator adjustment command value.

2. The method as described in claim 1, characterized in that, The measured environmental parameters include: dust concentration data, airflow temperature data, wind speed data, harmful gas concentration data, and pressure change data inside the ventilation duct. The target environmental parameters include: target wind speed, target temperature, target dust concentration, target gas concentration, and target pressure.

3. The method as described in claim 2, characterized in that, The preprocessing of the measured environmental parameters to obtain preprocessed measured environmental parameters includes: The wind speed data is smoothed using the moving average method, and missing values ​​are filled using mean fill and forward fill methods. Then the wind speed data is standardized to obtain preprocessed wind speed data. The airflow temperature data is processed using the Kalman filter algorithm, and missing values ​​are filled using forward imputation or interpolation. Then, the airflow temperature data is standardized to obtain preprocessed airflow temperature data. The IQR method is used to identify outliers in the dust concentration data and remove them. Then, the moving average method is used to smooth the dust concentration data after removing outliers. The smoothed dust concentration data is then processed using mean filling or interpolation to obtain filled dust concentration data. The filled dust concentration data is then standardized to obtain preprocessed dust concentration data. The Z-Score method is used to identify outliers in the concentration data of the harmful gas and remove them. Then, linear interpolation or polynomial interpolation is used to process the concentration data of the harmful gas after removing outliers to obtain the concentration data of the harmful gas after filling. The concentration data of the harmful gas after filling is then standardized to obtain the concentration data of the harmful gas after preprocessing. The pressure change data inside the ventilation duct is standardized to obtain pre-processed pressure change data inside the ventilation duct.

4. The method as described in claim 3, characterized in that, The step of determining the actuator adjustment command value based on the absolute value of the difference in environmental parameters and using a PID control algorithm includes: Determine whether the absolute value of the difference between each environmental parameter is greater than its threshold value for a given parameter. If so, use the PID calculation formula. Determine the actuator adjustment instruction value corresponding to the t-th environmental parameter; otherwise, no adjustment is made. Adjust the instruction value for the actuator corresponding to the t-th environmental parameter. This is the proportionality coefficient. This represents the difference between the target value and the measured value corresponding to the t-th environmental parameter. The integral coefficient is... is the differential coefficient.

5. The method as described in claim 4, characterized in that, The process of establishing the dust concentration prediction model includes: Acquire the pre-processed tunneling machine power, propulsion speed, fan speed, pre-processed wind speed data, pre-processed airflow temperature data, pre-processed dust concentration data, pre-processed harmful gas concentration data, and the measured dust concentration values ​​at the first, second, and third moments corresponding to each moment within the historical period. The preprocessed data on tunneling machine power, propulsion speed, fan speed, preprocessed wind speed, preprocessed airflow temperature, preprocessed dust concentration, and preprocessed harmful gas concentration at each moment within the historical time period are used as inputs. The measured dust concentration values ​​at the first, second, and third moments corresponding to each moment within the historical time period are used as outputs. The initial neural network model is trained using mean square error as the loss function to obtain a trained dust concentration prediction model.

6. The method as described in claim 5, characterized in that, The method further includes: The optimal wind speed is determined based on the optimized target dust concentration. If the absolute value of the difference between the optimal wind speed and the optimized target wind speed is greater than a preset wind speed threshold, then the optimal wind speed is taken as the target wind speed. Based on the pressure change data inside the duct, it is determined whether the pressure inside the duct is less than a preset first pressure threshold. If so, the fan speed is adjusted according to the preset first fan speed adjustment strategy. Based on the pressure change data inside the duct, it is determined whether the pressure inside the duct is greater than the preset second pressure threshold. If so, the fan speed is adjusted according to the preset second fan speed adjustment strategy.

7. The method as described in claim 6, characterized in that, The first fan speed adjustment strategy includes: increasing the fan speed according to a preset speed increase value; The second fan speed adjustment strategy includes: reducing the fan speed according to a preset speed reduction value.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-7.

Images