Sensor-based coal mine dust removal equipment automatic control system

Abstract

The present application relates to dust control technical field, specifically for coal mine dust removal equipment automation control system based on sensor, system includes: source working condition sensing module, dust source intensity feedforward solution module, transport time lag compensation module and collaborative spray execution module, in the present application, through direct collection motor current data of cutting mechanism and establish its nonlinear mapping relationship with dust generation, before dust actually diffuses to sensor, the advance prediction of dust production intensity is completed, the detection lag and response delay defects existing in traditional feedback control are effectively overcome, by introducing wind speed as dynamic variable to participate in the calculation of the accurate transmission time of dust group from cutting point to spray point, the precise space-time matching of dust falling medium putting time and dust group arrival time is realized, while significantly improving the efficiency of respirable dust capture, the water resource waste and floor water accumulation hidden danger produced by invalid spraying are greatly reduced.

Description

Technical Field

[0001] This invention relates to the field of dust control technology, and in particular to an automated control system for coal mine dust removal equipment based on sensors. Background Technology

[0002] Dust control technology involves air quality management and occupational health protection systems for underground mining environments. This field aims to capture, settle, and isolate high-concentration dust generated during production through mechatronic equipment. Its core encompasses fully automated technology from environmental parameter monitoring and control strategy calculation to actuator operation, serving as a key support facility for ensuring safe and efficient coal mine production and preventing pneumoconiosis. Traditional automated control systems for coal mine dust removal equipment utilize a passive architecture that monitors environmental turbidity using fixed-point suspended optical sensors and controls the start and stop of water pumps based on a preset single threshold. Traditional technology relies on concentration probes placed in the return airflow to acquire data. When a dust cloud diffuses to the probe location and triggers the threshold, the control cabinet outputs a signal to open the spray solenoid valve, intercepting the dust through a constant-pressure water curtain. This method typically uses relay logic or a basic PLC controller to achieve simple on / off control. The operation of the nozzle array mainly depends on whether the sensor reading exceeds the critical point, lacking the ability to sense the state of the dust source or dynamically adjust spray parameters.

[0003] In existing technologies, the sensor installation location is far from the dust source, resulting in a significant time lag in the detection results. When the system detects high concentration data, the dust cloud has often already crossed the optimal interception area and spread downwind. The fixed threshold judgment logic cannot adapt to the changes in dust cloud migration speed caused by drastic fluctuations in underground wind speed, resulting in a serious misalignment between the spray action and the arrival time of the dust cloud. The constant pressure spray mode cannot adjust the water mist particle size and flow rate for different dust generation intensities, resulting in a large waste of water resources when dust generation is low or insufficient dust removal efficiency when dust generation is high. In addition, frequent hard start impacts will accelerate the wear and aging of solenoid valve groups and pump station mechanical components. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and propose a sensor-based automated control system for coal mine dust removal equipment.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a sensor-based automated control system for coal mine dust removal equipment includes:

[0006] The source working condition sensing module is used to collect three-phase stator current data of the coal mining machine cutting section, obtain the roadway airflow velocity vector and dust concentration value of the fully mechanized mining face, analyze the time domain amplitude variation law of the three-phase stator current data, and construct the cutting load fluctuation characteristics.

[0007] The dust source intensity feedforward calculation module is used to analyze the nonlinear coupling relationship between the truncated load fluctuation characteristics and the dust concentration value, construct a dust generation intensity prediction model, extract the transient change rate of the three-phase stator current data and input it into the dust generation intensity prediction model for calculation, and calculate the dust generation flux index.

[0008] The transport delay compensation module is used to calculate the dust cloud migration delay time based on the airflow velocity vector in the roadway and the physical spatial distance parameters from the cut-off point to the spray point, determine the spray intensity level based on the dust generation flux index, and construct the spray control command in combination with the dust cloud migration delay time.

[0009] The coordinated spray execution module is used to respond to the spray control command, adjust the water supply pressure parameters of the variable frequency pump station according to the dust generation flux index, and adjust the opening duty cycle value of the solenoid valve group according to the dust generation flux index.

[0010] As a further aspect of the present invention, the source condition sensing module specifically includes:

[0011] The multi-source heterogeneous data synchronous acquisition submodule is used to acquire the three-phase stator current data of the coal mining machine cutting section in real time through a high-frequency Hall current sensor, monitor the roadway airflow velocity vector of the fully mechanized mining face using an ultrasonic anemometer, and obtain the dust concentration value in combination with a laser scattering dust concentration sensor, and perform timestamp alignment processing on the multi-source data.

[0012] The current signal time-domain purification submodule is used to filter the three-phase stator current data using a wavelet threshold denoising algorithm, remove high-frequency electromagnetic interference noise, retain the fundamental component and characteristic harmonic component that characterize the change of cutting resistance, and generate a stator current time sequence.

[0013] The load fluctuation feature extraction submodule is used to perform sliding window segmentation on the stator current time series, calculate the effective value of the current, peak factor and waveform distortion rate in each time window, and combine the above statistics to form the truncated load fluctuation feature.

[0014] As a further aspect of the present invention, the dust source intensity feedforward calculation module specifically includes:

[0015] The working condition dust generation mechanism correlation analysis submodule is used to calculate the mutual information entropy between the cutting load fluctuation characteristics and the dust concentration values ​​at historical moments, identify the dust generation lag time constant under different cutting hardness, and determine the sensitivity coefficient of load change to dust generation intensity.

[0016] The dust generation intensity nonlinear model construction submodule is used to train and generate the dust generation intensity prediction model by using the support vector machine regression algorithm, with the cut-off load fluctuation characteristics as input variables and the dust concentration value after lag time correction as output labels.

[0017] The real-time online inference submodule for flux index is used to calculate the first derivative value of the three-phase stator current data in real time to obtain the transient rate of change, input it into the dust generation intensity prediction model for feedforward calculation, and output the dust generation flux index.

[0018] As a further aspect of the present invention, the transport time delay compensation module specifically includes:

[0019] The wind flow field vector space analysis submodule is used to decompose the wind flow velocity vector in the roadway, extract the effective transport velocity component along the roadway axis, and combine it with the eddy current disturbance coefficient of the roadway cross section to correct the wind flow's ability to carry fine dust particles.

[0020] The dust cloud dynamic migration time delay calculation module is used to obtain the physical spatial distance parameter from the dust generation point of the cutting head to the spray point of the support in real time, calculate the theoretical time required for the dust-laden airflow to reach the spray area based on the corrected effective transport velocity, and generate the dust cloud migration delay time.

[0021] The spray timing coordination instruction generation submodule is used to map the dust generation flux index to a preset spray intensity level table, determine the target spray mode, and use the dust cloud migration delay time as a delay trigger parameter for action execution to synthesize the spray control instruction.

[0022] As a further aspect of the present invention, the coordinated spray execution module specifically includes:

[0023] The water supply pressure adaptive adjustment submodule is used to receive the spray control command, calculate the target water supply pressure value of the variable frequency pump station based on the dust generation flux index, and adjust the output frequency of the frequency converter through the PID closed-loop control algorithm to stabilize the pipeline pressure at the water supply pressure parameter.

[0024] The solenoid valve assembly duty cycle modulation submodule is used to determine the switching frequency and conduction duration of the solenoid valve assembly based on the dust generation flux index when the dust cloud migration delay time arrives, generate a corresponding pulse width modulation signal, and adjust the opening duty cycle value.

[0025] As a further aspect of the present invention, the specific function of the load fluctuation feature extraction submodule is as follows:

[0026] A fixed-length time sliding window is set. Within each sliding window, the root mean square value of the three-phase stator current data is calculated to characterize the average load level, the peak value of the signal amplitude is calculated to characterize the load impact intensity, and the fourth-order center moment kurtosis coefficient is calculated to characterize the steepness of the load distribution.

[0027] The root mean square value, the peak value, and the kurtosis coefficient are normalized according to a preset weight ratio to construct a multidimensional feature vector composed of these three dimensions, and this multidimensional feature vector is marked as the cut-off load fluctuation feature.

[0028] As a further aspect of the present invention, the specific calculation process of the real-time online inference submodule for the flux index includes:

[0029] The instantaneous value sequence of the three-phase stator current data is obtained, and the transient rate of change of current over time is calculated using a differential algorithm. At the same time, the root mean square value of current in the cut-off load fluctuation characteristics is obtained.

[0030] The dust generation flux index is calculated using the following formula:

[0031] ;

[0032] in, This represents the dust generation flux index. The transient rate of change representing the three-phase stator current data, This represents the current root mean square value of the current. This represents the reference current value of the coal mining machine when it is unloaded. Represents the transient impact dust generation coefficient. Represents the steady-state cutting dust generation coefficient. This represents a non-linear exponential correction factor.

[0033] As a further aspect of the present invention, the specific execution process of the dust cloud dynamic migration time delay solution submodule is as follows:

[0034] The effective airflow transport velocity along the roadway axis and the physical spatial distance parameters from the center of the cutting drum to the center of the hydraulic support nozzle array are obtained. Combined with the diffusion and drift characteristics of dust particles in the airflow, the migration delay time of the dust cloud is calculated.

[0035] The specific formula upon which the calculation is based is:

[0036] ;

[0037] in, This represents the migration delay time of the dust cloud. Represents the physical space distance parameter. This represents the axial component of the airflow velocity vector in the tunnel. This represents the drift correction factor that takes into account the Brownian motion of dust particles and turbulent diffusion. The inherent mechanical delay time between pipeline pressure build-up and solenoid valve response in the spray system.

[0038] As a further aspect of the present invention, the specific function of the solenoid valve group duty cycle modulation submodule is as follows:

[0039] Based on the value range of the dust generation flux index, the preset flux-duty cycle mapping table is queried to determine the baseline duty cycle value, and the baseline duty cycle is inversely compensated and corrected according to the real-time feedback value of the water supply pressure parameter to ensure the constantness of the spray flow rate.

[0040] A microprocessor generates a PWM pulse width modulation signal with a fixed frequency. The corrected duty cycle value is loaded into the PWM signal to drive the drive circuit of the solenoid valve group, thereby controlling the opening and closing time ratio of the valve core and outputting the opening duty cycle value.

[0041] As a further aspect of the present invention, the training process of the dust generation intensity nonlinear model construction submodule specifically includes:

[0042] A training dataset is constructed, which includes the truncated load fluctuation features at historical moments as input samples and the dust concentration values ​​calibrated by a high-precision dust meter at the corresponding moments as target samples. Outlier removal and standardization are then performed on the dataset.

[0043] The radial basis function is used to map the low-dimensional input samples to the high-dimensional feature space. The support vectors and the optimal hyperplane are determined by solving a quadratic programming problem. A nonlinear regression relationship between the cut-off load fluctuation characteristics and the dust concentration values ​​is established. The kernel function parameters and penalty factors are optimized by cross-validation to generate the dust generation intensity prediction model.

[0044] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0045] In this invention, by directly acquiring three-phase stator current data as a priori criterion, the system establishes an advanced mapping mechanism between load cutting and dust generation, generating a theoretical dust flux index before dust spreads, thus completely eliminating the time lag drawback of traditional feedback control. By introducing the airflow velocity vector in the roadway to participate in the dynamic calculation of dust cloud migration delay time, the system achieves precise synchronization between control commands and physical dust clouds in the spatiotemporal dimension, ensuring that the dust-suppressing medium always acts on the core area of ​​the high-concentration dust source. By adjusting the water supply pressure parameters and duty cycle value in real time according to the theoretical dust flux index, the system effectively avoids water waste and excessive equipment wear under low-load conditions while ensuring dust removal efficiency during high-intensity operations. Attached Figure Description

[0046] Figure 1 This is a block diagram illustrating the principle of the sensor-based automated control system for coal mine dust removal equipment according to the present invention.

[0047] Figure 2 This is a flowchart illustrating the internal operating logic of the original working condition sensing module of this invention.

[0048] Figure 3 This is a flowchart illustrating the internal operating logic of the dust source intensity feedforward calculation module of the present invention.

[0049] Figure 4 This is a flowchart illustrating the internal operating logic of the transport time delay compensation module of the present invention.

[0050] Figure 5 This is a flowchart illustrating the internal operating logic of the collaborative spray execution module of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this invention clearer, the software-based technical solution is described in detail below with reference to system architecture diagrams and embodiments. It should be understood that the specific embodiments described herein are only for explaining the technical solutions of this invention and do not constitute a limitation on the scope of protection.

[0052] In the description of this invention, the system architecture relationships or data processing flows indicated by terms such as "layer," "module," "interface," "data flow," "client," and "server" are all defined based on the architecture diagram or flowchart corresponding to the embodiments. This way of describing is only used to clearly illustrate the logical relationships between the elements in the technical solution, and not to limit the physical deployment form. The term "multiple" includes two or more technical units, including but not limited to multiple data nodes, processing threads, service instances, or functional components and other scalable elements. The specific number is determined according to the actual business scenario and needs to be specifically specified.

[0053] Please see Figure 1 and Figure 2This invention provides a technical solution: a sensor-based automated control system for coal mine dust removal equipment includes:

[0054] The source working condition sensing module is used to collect three-phase stator current data of the coal mining machine cutting section, obtain the roadway airflow velocity vector and dust concentration value of the fully mechanized mining face, analyze the time domain amplitude variation law of the three-phase stator current data, and construct the cutting load fluctuation characteristics.

[0055] Specifically, it includes:

[0056] The multi-source heterogeneous data synchronous acquisition submodule is used to acquire the three-phase stator current data of the coal cutting section in real time through a high-frequency Hall current sensor, monitor the airflow velocity vector of the roadway in the fully mechanized mining face using an ultrasonic anemometer, and obtain the dust concentration value by combining a laser scattering dust concentration sensor, and perform timestamp alignment processing on the multi-source data.

[0057] The current signal time-domain purification submodule is used to filter the three-phase stator current data using a wavelet threshold denoising algorithm, remove high-frequency electromagnetic interference noise, retain the fundamental component and characteristic harmonic component that characterize the change of cutting resistance, and generate a stator current time sequence.

[0058] The load fluctuation feature extraction submodule is used to perform sliding window segmentation on the stator current time series, calculate the effective value of the current, peak factor and waveform distortion rate in each time window, and combine the above statistics to form the truncated load fluctuation features.

[0059] The specific functions of the load fluctuation feature extraction submodule are as follows:

[0060] Set a fixed-length time sliding window, calculate the root mean square value of the three-phase stator current data in each sliding window to characterize the average load level, calculate the peak value of the signal amplitude to characterize the load impact intensity, and calculate the fourth-order center moment kurtosis coefficient to characterize the steepness of the load distribution.

[0061] The root mean square value, peak value, and kurtosis coefficient are normalized according to a preset weight ratio to construct a multidimensional feature vector composed of these three dimensions, and this multidimensional feature vector is marked as the truncated load fluctuation feature.

[0062] The multi-source heterogeneous data synchronous acquisition submodule deploys a HAS50-S high-frequency Hall current sensor at the three-phase output of the coal mining machine's frequency converter, with a sampling frequency set to 10kHz. This sensor acquires the analog three-phase stator current signal of the cutting section motor in real time and quantizes it into a digital signal via a 16-bit A / D converter. Simultaneously, a WindMasterPro three-dimensional ultrasonic anemometer is deployed every 5 meters below the top beam of the hydraulic support in the longwall mining face, outputting axial, lateral, and vertical wind speed vector data of the roadway airflow at a frequency of 50Hz. Furthermore, an LSD-01 laser scattering dust concentration sensor array acquires the current environmental dust concentration value every 200ms. The multi-source heterogeneous data synchronous acquisition submodule integrates an FPGA-based hardware timestamp alignment unit. Using the IEEE 1588PTP precision clock synchronization protocol, with the current acquisition time as the reference master clock, it performs linear interpolation alignment of the wind speed and dust concentration data, eliminating time phase deviations caused by differences in sampling rates between different sensors.

[0063] The aforementioned IEEE 1588PTP precision clock synchronization protocol refers to a precision clock synchronization protocol standard for network measurement and control systems. It achieves sub-microsecond master-slave clock synchronization accuracy in Ethernet networks through a hardware timestamp mechanism, ensuring strict consistency of distributed sensor data in the time dimension.

[0064] After receiving the raw three-phase stator current data, the current signal time-domain purification submodule calls the built-in digital signal processing core to execute a wavelet threshold denoising algorithm. The submodule uses Daubechies4 (db4) as the wavelet basis function and sets the decomposition level to 5 levels, decomposing the current signal into detail coefficients and approximation coefficients at different frequency scales. To address the high-frequency electromagnetic interference generated by the downhole frequency converter, the submodule uses the Minimax principle to calculate a universal threshold and applies a soft thresholding function to shrink the high-frequency detail coefficients; that is, when the absolute value of a wavelet coefficient is less than the threshold, it is set to zero, and when it is greater than the threshold, the threshold value is subtracted. Subsequently, the processed coefficients are used for inverse wavelet transform reconstruction, outputting a stator current time sequence with a signal-to-noise ratio better than 25dB.

[0065] The load fluctuation feature extraction submodule sets a sliding window of 200ms (containing 2000 sampling points) and slides it across the stator current time series with a step size of 50ms. Within each independent time window, the load fluctuation feature extraction submodule calculates three sets of statistics in parallel: the root mean square (RMS) value of the current is calculated through discretized integration to characterize the average load level; the difference between the maximum and minimum current values ​​within the search window is used to obtain the peak value to characterize the load impact intensity; and the fourth-order central moment kurtosis coefficient is calculated to characterize the steepness of the waveform distribution. Subsequently, the load fluctuation feature extraction submodule introduces a preset weight vector to perform weighted normalization on the above features, setting the weight of the RMS value to 0.4, the weight of the peak value to 0.3, and the weight of the kurtosis coefficient to 0.3. For example, when the calculated RMS within a certain window is 120A, the peak value is 45A, and the kurtosis coefficient is 3.8, the load fluctuation feature extraction submodule divides these three values ​​by their corresponding historical maximum values ​​of 200A, 100A, and 10 respectively to normalize them, and then multiplies them by the corresponding weights to finally construct a three-dimensional feature vector, and marks this multi-dimensional feature vector as the truncated load fluctuation feature.

[0066] The aforementioned fourth-order central moment kurtosis coefficient refers to a statistical measure that reflects the thickness of the tail and the sharpness of the peak of the probability density distribution curve. In the cutting load analysis, when the cutting drum encounters hard inclusions such as gangue, the current waveform will show an instantaneous peak, causing the kurtosis coefficient to deviate significantly from the standard value of 3 of the normal distribution, thus keenly indicating the sudden change in the working condition.

[0067] Please see Figure 1 and Figure 3 The dust source intensity feedforward calculation module is used to analyze the nonlinear coupling relationship between the characteristics of the cut load fluctuation and the dust concentration value, construct a dust generation intensity prediction model, extract the transient change rate of the three-phase stator current data and input it into the dust generation intensity prediction model for calculation, and calculate the dust generation flux index.

[0068] Specifically, it includes:

[0069] The working condition dust generation mechanism correlation analysis submodule is used to calculate the mutual information entropy between the characteristics of the cutting load fluctuation and the dust concentration values ​​at historical moments, identify the dust generation lag time constant under different cutting hardness, and determine the sensitivity coefficient of load change to dust generation intensity.

[0070] The nonlinear model construction submodule for dust generation intensity is used to train and generate a dust generation intensity prediction model using a support vector machine regression algorithm, with truncated load fluctuation characteristics as input variables and dust concentration values ​​corrected for lag time as output labels.

[0071] The real-time online inference submodule for flux index is used to calculate the first-order derivative value of the three-phase stator current data in real time to obtain the transient rate of change, input it into the dust generation intensity prediction model for feedforward calculation, and output the dust generation flux index.

[0072] The training process of the nonlinear model construction submodule for dust generation intensity specifically includes:

[0073] A training dataset is constructed, which includes the truncated load fluctuation features at historical moments as input samples and the dust concentration values ​​calibrated by a high-precision dust meter at the corresponding moments as target samples. Outlier removal and standardization are performed on the dataset.

[0074] A radial basis function kernel is used to map low-dimensional input samples to a high-dimensional feature space. The support vectors and the optimal hyperplane are determined by solving a quadratic programming problem. A nonlinear regression relationship between the truncated load fluctuation characteristics and the dust concentration value is established. The kernel function parameters and penalty factors are optimized by cross-validation to generate a dust generation intensity prediction model.

[0075] The specific calculation process of the real-time online inference submodule for flux indicators includes:

[0076] The instantaneous value sequence of three-phase stator current data is obtained, and the transient rate of change of current over time is calculated using the differential algorithm. At the same time, the root mean square value of current in the cut-off load fluctuation characteristics is obtained.

[0077] Calculate the dust generation flux index using the following formula:

[0078] ;

[0079] in, This represents the dust generation flux index. The transient rate of change of the three-phase stator current data This represents the current root mean square value of the current. This represents the reference current value of the coal mining machine when it is unloaded. Represents the transient impact dust generation coefficient. Represents the steady-state cutting dust generation coefficient. This represents a non-linear exponential correction factor.

[0080] The working condition dust generation mechanism correlation analysis submodule first calls the mutual information entropy algorithm to quantify the statistical dependence between the cut-off load fluctuation characteristics and historical dust concentration values. This submodule then identifies the dust generation lag time constant under different cut-off hardnesses by calculating the difference between conditional entropy and edge entropy. For example, if the calculation results show that the mutual information value reaches its maximum value of 0.65 when the lag time is 1.5s, then the dust generation lag time constant is determined to be 1.5s. Simultaneously, the working condition dust generation mechanism correlation analysis submodule calculates the sensitivity coefficient of load characteristics to dust generation intensity.

[0081] The dust generation intensity nonlinear model construction submodule establishes a prediction model based on the Support Vector Machine Regression (SVR) algorithm. This submodule selects the Radial Basis Function (RBF) as the kernel function to map the low-dimensional truncated load fluctuation characteristics to a high-dimensional feature space. During model training, the submodule employs a grid search method combined with 5-fold cross-validation to optimize hyperparameters. When the penalty factor C is set to 100 and the kernel parameter Gamma is set to 0.1, the mean squared error (MSE) of the model on the validation set is minimized. The trained model establishes a nonlinear regression relationship between the truncated load fluctuation characteristics and dust concentration values ​​by solving a quadratic programming problem.

[0082] The real-time online inference submodule for flux indicators first extracts the latest instantaneous value sequence from the three-phase stator current data and calculates the transient rate of change of current over time using a first-order forward differential algorithm. Specifically, the real-time online inference submodule reads the current value at the current time t and the current value at the previous sampling time t-1, and divides the difference between the two by the sampling interval to obtain the transient rate of change. Simultaneously, the real-time online inference submodule obtains the current root mean square value of current from the load fluctuation feature extraction submodule. Subsequently, the real-time online inference submodule calculates the dust generation flux index according to the following formula:

[0083] ;

[0084] in, This represents the dust generation flux index; the higher the value, the greater the predicted dust generation. The transient rate of change of three-phase stator current data, in amperes per second; This represents the current root mean square value, in amperes. The reference current value represents the coal mining machine under no-load conditions. In this embodiment, the measured value is 45A. The transient impact dust generation coefficient is used to characterize the contribution weight of drastic current fluctuations to the generation of fine dust. In this embodiment, the value is 0.85 for medium-hard coal seams. The value represents the steady-state cutting dust generation coefficient, which is used to characterize the dust generation capacity under steady-state high-load cutting. In this embodiment, the value is 0.45. The value represents the nonlinear exponential correction factor, which is set to 1.2 in this embodiment based on the nonlinear growth relationship between dust generation and cutting energy.

[0085] Assume the real-time online flux index inference submodule monitors a transient rate of change of 60% for the three-phase stator current at time t, and simultaneously acquires a root mean square current value of 145A. The submodule first calculates the steady-state term, subtracting 45 from 145 to obtain 100, and then powering 100 to the power of 1.2, which is approximately 251.19. Next, a weighted calculation is performed: the first term is 0.85 multiplied by 60, which equals 51; the second term is 0.45 multiplied by 251.19, which is approximately 113.04. Finally, 51 and 113.04 are added together to obtain a dust flux index of 164.04. This result indicates that the current cutting operation will generate a dust flux significantly higher than the baseline level, requiring high-intensity spray suppression.

[0086] Please see Figure 1 and Figure 4 The transport delay compensation module is used to calculate the dust cloud migration delay time based on the airflow velocity vector in the roadway and the physical spatial distance parameters from the cut-off point to the spray point, determine the spray intensity level based on the dust generation flux index, and construct the spray control command in combination with the dust cloud migration delay time.

[0087] Specifically, it includes:

[0088] The wind flow field vector space analysis submodule is used to decompose the wind flow velocity vector in the roadway, extract the effective transport velocity component along the roadway axis, and combine it with the eddy current disturbance coefficient of the roadway cross section to correct the wind flow's ability to carry fine dust particles.

[0089] The dust cloud dynamic migration time delay calculation module is used to obtain the physical spatial distance parameters from the dust generation point of the cutting head to the spray point of the support in real time, calculate the theoretical time required for the dust-laden airflow to reach the spray area based on the corrected effective transport velocity, and generate the dust cloud migration delay time.

[0090] The spray timing coordination instruction generation submodule is used to map the dust generation flux index to a preset spray intensity level table, determine the target spray mode, and use the dust cloud migration delay time as the delay trigger parameter for action execution to synthesize spray control instructions.

[0091] The specific execution process of the dust cloud dynamic migration time delay solution submodule is as follows:

[0092] The effective airflow transport velocity along the roadway axis and the physical spatial distance parameters from the center of the cutting drum to the center of the hydraulic support nozzle array are obtained. Combined with the diffusion and drift characteristics of dust particles in the airflow, the dust cloud migration delay time is calculated.

[0093] The specific formula used for the calculation is as follows:

[0094] ;

[0095] in, This represents the time delay in the migration of dust clumps. Represents the physical space distance parameter. This represents the axial component of the airflow velocity vector in the tunnel. This represents the drift correction factor that takes into account the Brownian motion of dust particles and turbulent diffusion. The inherent mechanical delay time between pipeline pressure build-up and solenoid valve response in the spray system.

[0096] The airflow field vector space analysis submodule reads the three-dimensional airflow velocity vector provided by the sensing module and extracts the effective transport velocity component along the tunnel axis, i.e., the main direction of dust migration, using the trigonometric function projection method. The airflow field vector space analysis submodule introduces an eddy current disturbance coefficient to correct the measured wind speed. In a typical high Reynolds number turbulent mine environment, the eddy current disturbance coefficient is taken as 0.85. For example, if the axial center wind speed measured by the sensor is 3.3 m / s, the average effective wind speed for dust transport calculated by the airflow field vector space analysis submodule is 3.3 multiplied by 0.85, which equals 2.805 m / s, approximated as 2.8 m / s.

[0097] The dust cloud dynamic migration time-delay calculation module reads in real time the physical spatial distance parameters between the coal mining machine's cutting section (dust source) and the target hydraulic support spray point (dust suppression zone). Assuming the cutting section is located at the 50th support in the current operating condition, and the spray needs to be activated beforehand at the 58th support on the leeward side, with a center-to-center distance of 1.5m, the physical spatial distance parameter is 8 multiplied by 1.5, equaling 12m. The dust cloud dynamic migration time-delay calculation module, combined with the previously calculated effective transport velocity, uses the principles of physical kinematics to calculate the time it takes for the dust cloud to reach the spray area.

[0098] The specific formula used for the calculation is as follows:

[0099] ;

[0100] in, This represents the delay time for dust cloud migration, measured in seconds. This represents the physical spatial distance parameter, which is 12m in this embodiment; The axial component of the airflow velocity vector in the tunnel is represented by 2.8 m / s in this embodiment; The drift correction factor is 0.9, which takes into account the velocity lag of dust particles under the effects of Brownian motion and turbulent diffusion. The inherent mechanical delay time representing the pipeline pressure build-up and solenoid valve response of the spray system is calibrated to 0.2s.

[0101] The dust cloud dynamic migration time delay calculation module substitutes the above parameters into the formula for calculation: the denominator is 2.8 multiplied by 0.9, which equals 2.52; the fractional part is 12 divided by 2.52, which is approximately 4.76; finally, adding 0.2, we get the dust cloud migration delay time as 4.96s. This result indicates that from the current dust generation time, the high-concentration dust cloud will reach the target spray area after 4.96s.

[0102] The spray timing coordination command generation submodule determines the target spray mode based on the input dust generation flux index and a preset spray intensity level table. As shown in Table 1, when the dust generation flux index is 164.04 in the aforementioned example, this value falls within the range of 120 to 200. The spray timing coordination command generation submodule determines that the current mode is "high-power spray," with a corresponding target water supply pressure of 6.0 MPa and a solenoid valve reference duty cycle of 80%. The spray timing coordination command generation submodule combines this mode parameter with a delay time of 4.96 seconds to synthesize a spray control command.

[0103] Table 1 Spray Intensity Level Mapping Table

[0104]

[0105] As shown in Table 1, Table 1 lists the spray patterns, target water supply pressures, and solenoid valve reference duty cycles corresponding to different dust generation flux ranges, which serve as the basis for system decision-making.

[0106] Please see Figure 1 and Figure 5 The coordinated spray execution module is used to respond to spray control commands, adjust the water supply pressure parameters of the variable frequency pump station according to the dust generation flux index, and adjust the opening duty cycle value of the solenoid valve group according to the dust generation flux index.

[0107] Specifically, it includes:

[0108] The adaptive water supply pressure adjustment submodule is used to receive spray control commands, calculate the target water supply pressure value of the variable frequency pump station based on the dust generation flux index, and adjust the output frequency of the frequency converter through the PID closed-loop control algorithm to stabilize the pipeline pressure at the water supply pressure parameter.

[0109] The solenoid valve assembly duty cycle modulation submodule is used to determine the switching frequency and conduction duration of the solenoid valve assembly based on the dust generation flux index when the dust cloud migration delay time arrives, generate the corresponding pulse width modulation signal, and adjust the opening duty cycle value.

[0110] The specific functions of the solenoid valve assembly duty cycle modulation submodule are as follows:

[0111] Based on the value range of the dust generation flux index, the preset flux-duty cycle mapping table is consulted to determine the baseline duty cycle value. The baseline duty cycle is then inversely compensated and corrected based on the real-time feedback value of the water supply pressure parameter to ensure the constantness of the spray flow rate.

[0112] A microprocessor generates a PWM pulse width modulation signal with a fixed frequency. The corrected duty cycle value is loaded into the PWM signal to drive the drive circuit of the solenoid valve assembly, thereby controlling the opening and closing time ratio of the valve core and outputting the opening duty cycle value.

[0113] The adaptive water supply pressure regulation submodule receives the target water supply pressure value of 6.0 MPa from the spray control command as the setpoint and collects feedback values ​​from the pipeline pressure sensor in real time as process variables. The submodule utilizes a PID closed-loop control algorithm, setting the proportional coefficient Kp to 1.5, the integral coefficient Ki to 0.08, and the derivative coefficient Kd to 0.2. When the pipeline pressure needs to be increased, the submodule adjusts the output frequency of the frequency converter to stabilize the pipeline pressure at 6.0 MPa.

[0114] The solenoid valve group duty cycle modulation submodule is activated when the dust cloud migration delay time is reached. The submodule first determines a baseline duty cycle value of 80% based on the dust generation flux index of 164.04. To compensate for the impact of water supply pressure fluctuations on flow rate, the submodule introduces inverse compensation correction logic. Assuming the system rated pressure is 6.0 MPa, if the real-time feedback water supply pressure is only 5.5 MPa, the submodule calculates the correction coefficient: 6.0 divided by 5.5 yields approximately 1.0909, the square root of which is approximately 1.0444; multiplying the baseline duty cycle of 0.8 by 1.0444 yields 0.8355. Therefore, the final output duty cycle is 83.5%. The solenoid valve assembly duty cycle modulation submodule uses a microprocessor to generate a PWM pulse width modulation signal with a frequency of 10Hz. The calculated 83.5% duty cycle is loaded into this signal, driving the solenoid valve assembly to open for 83.5ms within a 100ms cycle, ensuring the constant spray flow rate.

[0115] The above embodiments illustrate preferred embodiments of the present invention. Any equivalent adjustments to the technical solution based on software engineering methods are within the scope of protection, including but not limited to: implementing algorithm logic using different programming languages, refactoring functional modules into services, adjusting data interaction protocols, and optimizing resource scheduling strategies. Any implementation scheme derived from reasonable modifications to the data processing flow, service call chain, or system architecture layer without departing from the core technology of the present invention should be considered within the protection scope defined by the technical solution of the present invention.

Claims

1. A sensor-based automation control system for coal mine dust removal equipment, characterized in that, The system includes: The source working condition sensing module is used to collect three-phase stator current data of the coal mining machine cutting section, obtain the roadway airflow velocity vector and dust concentration value of the fully mechanized mining face, analyze the time domain amplitude variation law of the three-phase stator current data, and construct the cutting load fluctuation characteristics. The dust source intensity feedforward calculation module is used to analyze the nonlinear coupling relationship between the truncated load fluctuation characteristics and the dust concentration value, construct a dust generation intensity prediction model, extract the transient change rate of the three-phase stator current data and input it into the dust generation intensity prediction model for calculation, and calculate the dust generation flux index. The transport delay compensation module is used to calculate the dust cloud migration delay time based on the airflow velocity vector in the roadway and the physical spatial distance parameters from the cut-off point to the spray point, determine the spray intensity level based on the dust generation flux index, and construct the spray control command in combination with the dust cloud migration delay time. The coordinated spray execution module is used to respond to the spray control command, adjust the water supply pressure parameters of the variable frequency pump station according to the dust generation flux index, and adjust the opening duty cycle value of the solenoid valve group according to the dust generation flux index.

2. The sensor-based automated control system for coal mine dust removal equipment according to claim 1, characterized in that, The source condition sensing module specifically includes: The multi-source heterogeneous data synchronous acquisition submodule is used to acquire the three-phase stator current data of the coal mining machine cutting section in real time through a high-frequency Hall current sensor, monitor the roadway airflow velocity vector of the fully mechanized mining face using an ultrasonic anemometer, and obtain the dust concentration value in combination with a laser scattering dust concentration sensor, and perform timestamp alignment processing on the multi-source data. The current signal time-domain purification submodule is used to filter the three-phase stator current data using a wavelet threshold denoising algorithm, remove high-frequency electromagnetic interference noise, retain the fundamental component and characteristic harmonic component that characterize the change of cutting resistance, and generate a stator current time sequence. The load fluctuation feature extraction submodule is used to perform sliding window segmentation on the stator current time series, calculate the effective value of the current, peak factor and waveform distortion rate in each time window, and combine the above statistics to form the truncated load fluctuation feature.

3. The sensor-based automated control system for coal mine dust removal equipment according to claim 1, characterized in that, The dust source intensity feedforward calculation module specifically includes: The working condition dust generation mechanism correlation analysis submodule is used to calculate the mutual information entropy between the cutting load fluctuation characteristics and the dust concentration values ​​at historical moments, identify the dust generation lag time constant under different cutting hardness, and determine the sensitivity coefficient of load change to dust generation intensity. The dust generation intensity nonlinear model construction submodule is used to train and generate the dust generation intensity prediction model by using the support vector machine regression algorithm, with the cut-off load fluctuation characteristics as input variables and the dust concentration value after lag time correction as output labels. The real-time online inference submodule for flux index is used to calculate the first derivative value of the three-phase stator current data in real time to obtain the transient rate of change, input it into the dust generation intensity prediction model for feedforward calculation, and output the dust generation flux index.

4. The sensor-based automated control system for coal mine dust removal equipment according to claim 1, characterized in that, The transport time delay compensation module specifically includes: The wind flow field vector space analysis submodule is used to decompose the wind flow velocity vector in the roadway, extract the effective transport velocity component along the roadway axis, and combine it with the eddy current disturbance coefficient of the roadway cross section to correct the wind flow's ability to carry fine dust particles. The dust cloud dynamic migration time delay calculation module is used to obtain the physical spatial distance parameter from the dust generation point of the cutting head to the spray point of the support in real time, calculate the theoretical time required for the dust-laden airflow to reach the spray area based on the corrected effective transport velocity, and generate the dust cloud migration delay time. The spray timing coordination instruction generation submodule is used to map the dust generation flux index to a preset spray intensity level table, determine the target spray mode, and use the dust cloud migration delay time as a delay trigger parameter for action execution to synthesize the spray control instruction.

5. The sensor-based automated control system for coal mine dust removal equipment according to claim 1, characterized in that, The coordinated spray execution module specifically includes: The water supply pressure adaptive adjustment submodule is used to receive the spray control command, calculate the target water supply pressure value of the variable frequency pump station based on the dust generation flux index, and adjust the output frequency of the frequency converter through the PID closed-loop control algorithm to stabilize the pipeline pressure at the water supply pressure parameter. The solenoid valve assembly duty cycle modulation submodule is used to determine the switching frequency and conduction duration of the solenoid valve assembly based on the dust generation flux index when the dust cloud migration delay time arrives, generate a corresponding pulse width modulation signal, and adjust the opening duty cycle value.

6. The sensor-based automated control system for coal mine dust removal equipment according to claim 2, characterized in that, The specific functions of the load fluctuation feature extraction submodule are as follows: A fixed-length time sliding window is set. Within each sliding window, the root mean square value of the three-phase stator current data is calculated to characterize the average load level, the peak value of the signal amplitude is calculated to characterize the load impact intensity, and the fourth-order center moment kurtosis coefficient is calculated to characterize the steepness of the load distribution. The root mean square value, the peak value, and the kurtosis coefficient are normalized according to a preset weight ratio to construct a multidimensional feature vector composed of these three dimensions, and this multidimensional feature vector is marked as the cut-off load fluctuation feature.

7. The sensor-based automated control system for coal mine dust removal equipment according to claim 3, characterized in that, The specific calculation process of the real-time online inference submodule for the flux index includes: The instantaneous value sequence of the three-phase stator current data is obtained, and the transient rate of change of current over time is calculated using a differential algorithm. At the same time, the root mean square value of current in the cut-off load fluctuation characteristics is obtained. The dust generation flux index is calculated using the following formula: ； in, This represents the dust generation flux index. The transient rate of change representing the three-phase stator current data, This represents the current root mean square value of the current. This represents the reference current value of the coal mining machine when it is unloaded. Represents the transient impact dust generation coefficient. Represents the steady-state cutting dust generation coefficient. This represents a non-linear exponential correction factor.

8. The sensor-based automated control system for coal mine dust removal equipment according to claim 4, characterized in that, The specific execution process of the dust cloud dynamic migration time delay calculation submodule is as follows: The effective airflow transport velocity along the roadway axis and the physical spatial distance parameters from the center of the cutting drum to the center of the hydraulic support nozzle array are obtained. Combined with the diffusion and drift characteristics of dust particles in the airflow, the migration delay time of the dust cloud is calculated. The specific formula upon which the calculation is based is: ； in, This represents the migration delay time of the dust cloud. Represents the physical space distance parameter. This represents the axial component of the airflow velocity vector in the tunnel. This represents the drift correction factor that takes into account the Brownian motion of dust particles and turbulent diffusion. The inherent mechanical delay time between pipeline pressure build-up and solenoid valve response in the spray system.

9. The sensor-based automated control system for coal mine dust removal equipment according to claim 5, characterized in that, The specific function of the solenoid valve group duty cycle modulation submodule is as follows: Based on the value range of the dust generation flux index, the preset flux-duty cycle mapping table is queried to determine the baseline duty cycle value, and the baseline duty cycle is inversely compensated and corrected according to the real-time feedback value of the water supply pressure parameter to ensure the constantness of the spray flow rate. A microprocessor generates a PWM pulse width modulation signal with a fixed frequency. The corrected duty cycle value is loaded into the PWM signal to drive the drive circuit of the solenoid valve group, thereby controlling the opening and closing time ratio of the valve core and outputting the opening duty cycle value.

10. The sensor-based automated control system for coal mine dust removal equipment according to claim 3, characterized in that, The training process of the nonlinear model construction submodule for dust generation intensity specifically includes: A training dataset is constructed, which includes the truncated load fluctuation features at historical moments as input samples and the dust concentration values ​​calibrated by a high-precision dust meter at the corresponding moments as target samples. Outlier removal and standardization are then performed on the dataset. The radial basis function is used to map the low-dimensional input samples to the high-dimensional feature space. The support vectors and the optimal hyperplane are determined by solving a quadratic programming problem. A nonlinear regression relationship between the cut-off load fluctuation characteristics and the dust concentration values ​​is established. The kernel function parameters and penalty factors are optimized by cross-validation to generate the dust generation intensity prediction model.

Images