A mining and transporting device with a dust suppression mechanism for a discharge port and a method thereof

Abstract

The ore mining and transporting device with a dust suppression mechanism for an outlet and a method thereof belong to the technical fields of mineral exploitation, material transportation and dust pollution control, and comprise the following steps: material flow characteristic soft measurement: the current signal of a power monitor, the vibration signal of an acoustic probe and the image signal of an industrial camera are fused by a controller to estimate the instantaneous flow and apparent moisture content of the material in real time; optimal atomization strategy predictive control: the controller combines the wind speed disturbance data of an anemometer to perform rolling optimization through a dynamic mechanism model based on the estimated instantaneous flow and apparent moisture content, and outputs a control instruction to an atomization spray gun; feedback constraint and closed loop: the controller takes the dust concentration measured by a turbidity sensor as a minimization target and takes the exhaust humidity measured by a humidity sensor as a constraint condition, so that the accuracy and robustness of the estimation result are improved, and a foundation is laid for subsequent accurate control.

Description

Technical Field

[0001] This invention relates to the fields of mineral mining, material transportation and dust pollution control, specifically to an ore mining transportation device and method with a dust suppression mechanism at the discharge port. Background Technology

[0002] With the rapid development and construction of ore mining projects, serious dust pollution problems arise in typical material transportation scenarios, especially at the discharge point. Traditional ore transportation methods mainly rely on fixed spraying strategies for dust suppression; however, in actual working conditions, material flow characteristics such as instantaneous flow rate and apparent moisture content change frequently and drastically. Under such volatile conditions, traditional fixed spraying strategies are difficult to adapt. Traditional dust suppression methods rely on a single spraying action to cope with complex material flow changes, and their shortcomings include:

[0003] Unstable suppression effect: When the material flow rate suddenly increases or the dryness rises, the fixed spray volume cannot provide sufficient suppression force, leading to a sharp increase in dust concentration and causing serious dust pollution; Risk of excessively wet materials: When the material flow rate decreases or the apparent moisture content is high, the fixed spray volume may waste water resources and cause the material to become excessively wet, affecting subsequent conveying and processing; In addition, traditional sensors used to provide feedback on dust suppression effect, such as turbidity sensors and humidity sensors, are hysteretic; they can only monitor the state after dust emission and increased exhaust humidity have occurred, and cannot provide early warning when the material flow state changes drastically, making the control system only able to take a hysteretic response and difficult to achieve proactive suppression.

[0004] Therefore, how to proactively and proactively adjust dust suppression strategies to effectively cope with the instantaneous changes in material flow while ensuring that the apparent moisture content of materials does not exceed a preset threshold is a key problem that urgently needs to be solved in the current ore transportation field.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide an ore mining and transportation device and method with a dust suppression mechanism at the discharge port, so as to solve the problems mentioned in the background art; the specific technical solution of this invention is as follows:

[0007] A method for transporting ore with a dust suppression mechanism at the discharge port includes the following steps:

[0008] Material flow characteristic soft measurement: By integrating the current signal from the power monitor, the vibration signal from the acoustic probe, and the image signal from the industrial camera, the instantaneous flow rate and apparent moisture content of the material are estimated in real time;

[0009] Predictive control of optimal atomization strategy: The controller performs rolling optimization based on the estimated instantaneous flow rate and apparent moisture content, combined with the wind speed disturbance data from the anemometer, through a dynamic mechanism model, and outputs control commands to the atomizing spray gun.

[0010] Feedback constraints and closed loop: When optimizing, the controller takes the dust concentration measured by the turbidity sensor as the minimum target and the exhaust humidity measured by the humidity sensor as the constraint condition that it does not exceed a preset threshold.

[0011] Preferably, the soft measurement step for material flow characteristics specifically includes:

[0012] Data acquisition: The power monitor acquires the motor load current as a reference mass flow rate; the acoustic probe picks up the structural vibration sound pattern of the material impact; and the industrial camera acquires the instantaneous cross-sectional profile of the material by exposing it with a high-brightness LED strobe light source.

[0013] Fusion estimation: The deep neural network model in the controller receives the current signal, the vibration sound pattern and the cross-sectional profile, and outputs the estimated values ​​of the instantaneous flow rate and the apparent moisture content through multimodal fusion calculation.

[0014] Preferably, the deep neural network model is established through the following steps:

[0015] During the debugging phase, historical operating condition datasets are collected, and the current signal of the power monitor, the vibration signal of the acoustic probe, and the image signal of the industrial camera are recorded simultaneously as input features of the model.

[0016] A high-precision weighing hopper is used for periodic offline calibration to obtain a baseline mass flow rate, and the material samples are dried to obtain a baseline apparent moisture content. The baseline flow rate and apparent moisture content data are used as supervision labels for the model.

[0017] Adjust the weights within the model to establish a nonlinear mapping relationship from the input features of the three sensors to the supervision label.

[0018] Preferably, the optimal atomization strategy prediction and control step specifically includes:

[0019] Prediction: Based on the rate of change of the instantaneous flow rate and the apparent moisture content, the controller establishes a short-term prediction model to predict the material flow status within a future control cycle;

[0020] Optimization solution: Based on the material flow state, the wind speed disturbance data, and the dynamic mechanism model describing the relationship between the spray gun action and the dust concentration and the exhaust humidity, the controller solves the objective function through a rolling optimization algorithm to determine the start and stop sequence of the atomizing spray gun.

[0021] Preferably, it also includes sensor status self-diagnosis and degradation steps:

[0022] Verification: The controller continuously runs consistency verification logic to cross-compare the data streams of the power monitor, the acoustic probe, and the industrial camera;

[0023] Diagnosis: When persistent data discrepancies are detected, it is determined that a specific sensor has malfunctioned;

[0024] Degradation: The controller immediately triggers the dynamic control strategy degradation process, and activates a pre-trained, degraded soft measurement model that does not depend on the fault sensor, based on the material characteristic parameters before the fault and failure.

[0025] A conveying device for ore mining with a dust suppression mechanism at the discharge port, the device comprising:

[0026] The end of a conveyor, used for conveying materials;

[0027] A dust suppression cover is fixedly attached to the discharge port at the end of the conveyor.

[0028] A power monitor is used to monitor the load current of the motor driving the conveyor;

[0029] The sensing array includes an industrial camera and an acoustic probe, wherein the industrial camera is fixed to the frame of the dust suppression hood and the acoustic probe is fixed to the end of the conveyor;

[0030] Atomizing spray guns are installed in an array on the inner wall of the dust suppression hood;

[0031] The status monitor includes a turbidity sensor and a humidity sensor installed inside the exhaust duct of the dust suppression hood;

[0032] An anemometer is installed inside the exhaust duct;

[0033] The controller is connected to the power monitor, the sensing array, the atomizing spray gun, the status monitor, and the anemometer.

[0034] Preferably, a discharge chute is installed at the end of the conveyor, and the acoustic probe is a piezoelectric contact microphone, which is tightly fixed to the outer wall of the bottom plate of the discharge chute through a rigid connecting seat.

[0035] Preferably, the industrial camera is a high-speed industrial camera, and a synchronously triggered high-brightness LED strobe light source is fixed next to the lens of the industrial camera via the same mounting bracket.

[0036] Preferably, the power monitor is a high-precision wideband current transformer, which is non-contactly clamped to the main power supply cable of the motor driving the conveyor.

[0037] Preferably, the atomizing spray gun is controlled by an independent electronically controlled high-frequency pulse width modulation valve.

[0038] This invention provides an improved ore mining and transportation device and method with a dust suppression mechanism at the discharge port, which has the following improvements and advantages compared with the prior art:

[0039] 1. This invention introduces a soft measurement step for material flow characteristics. By using a deep neural network model to multimodally fuse three heterogeneous data streams—motor current signal from a power monitor, vibration sound pattern from an acoustic probe, and cross-sectional profile image signal from an industrial camera—real-time and accurate estimates of instantaneous material flow rate and apparent moisture content are obtained. Compared to measurement methods that rely on a single sensor, this fusion method combines the stability of the current signal with the instantaneous nature of the acoustic / visual signal, improving the accuracy and robustness of the estimation results and laying the foundation for subsequent precise control.

[0040] 2. Optimal atomization strategy predictive control is adopted. Through a short-time prediction model and a rolling optimization algorithm based on a dynamic mechanism model, the start-up and shutdown sequence of the atomizing spray gun is determined based on the predicted future material flow state and wind speed disturbance data. This eliminates the lag response of traditional methods and can calculate and deploy a matching enhanced fog field in advance before high flow or dry materials actually reach the discharge port, significantly improving the initiative and effectiveness of dust suppression. In the optimization solution, the dust concentration measured by the turbidity sensor is used as the minimization target, while the exhaust humidity measured by the humidity sensor does not exceed the preset threshold as a constraint condition. This ensures that while achieving the best dust suppression effect, the material is not too wet, achieving a precise dynamic balance between dust suppression effect and apparent moisture content.

[0041] 3. A sensor status self-diagnosis and degradation step has been added. By continuously running consistency verification logic to cross-compare sensor data, and when a specific sensor failure is determined, a degraded soft measurement model that does not depend on the faulty sensor is activated. This ensures that the dust suppression system can continue to operate the adaptive dust suppression function even if a single sensing component fails in a harsh mining environment. This avoids the complete paralysis of traditional systems caused by sensor failures and the sacrifice of some control accuracy, but greatly improves the operational reliability and fault tolerance of the entire system. Attached Figure Description

[0042] The present invention will be further explained below with reference to the accompanying drawings and embodiments:

[0043] Figure 1 This is a schematic diagram of the overall external structure of the device;

[0044] Figure 2 This is a cross-sectional structural diagram of the dust suppression hood;

[0045] Figure 3 This is a schematic diagram of the exhaust duct and its connection structure;

[0046] Figure 4 This is a schematic diagram of the process flow of the method of the present invention.

[0047] In the diagram: 100, conveyor end; 110, discharge chute; 120, power monitor; 200, dust suppression hood; 210, exhaust duct; 220, anemometer; 230, atomizing spray gun; 300, sensing array; 310, industrial camera; 320, acoustic probe; 400, status monitor; 410, turbidity sensor; 420, humidity sensor; 500, controller. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0049] Example 1

[0050] Please see Figure 1-4 This invention provides a method for transporting ore from mining operations with a dust suppression mechanism at the discharge port, comprising the following steps:

[0051] Material flow characteristics soft measurement: By integrating the current signal from the power monitor 120, the vibration signal from the acoustic probe 320, and the image signal from the industrial camera 310, the instantaneous flow rate and apparent moisture content of the material are estimated in real time.

[0052] Predictive control of optimal atomization strategy: The controller 500 performs rolling optimization through a dynamic mechanism model based on the estimated instantaneous flow rate and apparent moisture content, combined with the wind speed disturbance data of the anemometer 220, and outputs control commands to the atomizing spray gun 230.

[0053] Feedback constraints and closed loop: When optimizing, the controller 500 takes the dust concentration measured by the turbidity sensor 410 as the minimum target and the exhaust humidity measured by the humidity sensor 420 as the constraint condition that it does not exceed the preset threshold.

[0054] The exhaust humidity not exceeding the preset threshold Hmax is a safety indicator used to define the maximum permissible exhaust humidity during dust suppression. Physically, when the exhaust humidity exceeds this value, it indicates that the atomization amount may be excessive, posing a risk of overly moist materials. This threshold needs to be determined offline during the commissioning phase through dynamic atomization testing of typical ore samples in the laboratory, combined with the requirements of subsequent processing techniques on the apparent moisture content of the materials. This threshold is a hard constraint condition for the rolling optimization algorithm in the optimal atomization strategy predictive control step, ensuring that while minimizing dust concentration, excessive material moisture is avoided.

[0055] This embodiment describes a transportation method for ore mining with a dust suppression mechanism at the discharge port. The soft measurement step, which assesses the material flow characteristics, serves as the starting point for the entire control logic. This step integrates sensor data from multiple sources to obtain a real-time estimate of the material's state, forming the basis for subsequent precise control. The optimal atomization strategy predictive control step utilizes this real-time estimate, combined with external environmental interference factors, to calculate a forward-looking spray action using a model. The feedback constraint and closed-loop steps set clear objectives and boundaries for the entire control process, namely, suppressing dust while preventing excessive material moisture. These three steps work together to enable the dust suppression system to proactively adapt to changes in operating conditions, overcoming the limitations of traditional fixed spray strategies and achieving a balance between dust suppression effectiveness and material moisture content.

[0056] The specific steps of soft measurement of material flow characteristics include:

[0057] Data acquisition: Power monitor 120 acquires motor load current to characterize reference mass flow rate; acoustic probe 320 picks up structural vibration acoustic patterns of material impact; industrial camera 310 acquires instantaneous cross-sectional profile of material through exposure using a high-brightness LED strobe light source.

[0058] Fusion estimation: The deep neural network model within the controller 500 receives current signals, vibration patterns, and cross-sectional profiles, and outputs estimated values ​​of instantaneous flow rate and apparent moisture content through multimodal fusion calculation.

[0059] Furthermore, in another embodiment, the deep neural network model within the controller 500 can be replaced with an adaptive fusion algorithm based on Kalman filtering. This algorithm establishes a state-space model, using the signal from the power monitor 120 as the primary measurement input, and utilizes the instantaneous signals from the acoustic probe 320 and the industrial camera 310 as correction terms for model prediction, to achieve multi-sensor state estimation of instantaneous flow rate and apparent moisture content, thus obtaining accurate and robust estimation results.

[0060] The input sources for multimodal fusion computation are the current signal from the power monitor 120, the vibration sound pattern picked up by the acoustic probe 320, and the cross-sectional profile acquired by the industrial camera 310—three heterogeneous data streams. The logical steps include: Step 1: The controller 500 performs synchronous timestamp alignment and noise reduction preprocessing on the three signals. Step 2: The processed signals are input into a deep neural network model. Step 3: The network performs weighted and nonlinear combinations of the extracted features from the three signals through a feature-level or decision-level fusion layer. The final output of the process is a real-time estimate of the instantaneous flow rate and apparent moisture content of the material, which serves as the core input in the optimal atomization strategy predictive control step.

[0061] In this embodiment, the soft measurement steps for material flow characteristics are further refined; the acquisition process utilizes three sensing methods with different physical principles; the power monitor 120 provides a relatively stable low-frequency signal that reflects the overall mass of the material by measuring the load current of the conveyor motor, which is used to characterize the baseline mass flow rate; the acoustic probe 320 is used to capture the high-frequency structural vibration acoustic pattern generated by the collision and friction between the material and the equipment, and this acoustic pattern characteristic is closely related to the particle size and impact kinetic energy of the material; the industrial camera 310, in conjunction with a high-brightness LED strobe light source, can capture clear images without motion blur, which are used to calculate the instantaneous flow of the material. The data streams, including the cross-sectional profile and surface texture, are simultaneously input into a deep neural network model deployed within the controller 500. In the fusion estimation step, these three data streams, each containing information of different dimensions—current signal, vibration signature, and cross-sectional profile—are processed through multimodal fusion calculations to output real-time estimates of instantaneous flow rate and apparent moisture content. This fusion method utilizes the stability of the current signal to calibrate the mean of the results, while leveraging the transience of acoustic and visual signals to capture dramatic flow fluctuations. The resulting estimates outperform those relying on single-sensor measurements in both accuracy and robustness.

[0062] A deep neural network model is built through the following steps:

[0063] During the commissioning phase, historical operating condition datasets are collected, and the current signal of the power monitor 120, the vibration signal of the acoustic probe 320, and the image signal of the industrial camera 310 are recorded simultaneously as input features of the model.

[0064] A high-precision weighing hopper is used for periodic offline calibration to obtain the baseline mass flow rate, and the sampled material is dried to obtain the baseline apparent moisture content. The baseline flow rate and moisture content data are used as the supervision labels for the model.

[0065] Adjust the internal weights of the model to establish a nonlinear mapping relationship from the input features of the three sensors to the supervision label.

[0066] In this embodiment, the deep neural network model is established during the debugging phase before the equipment is officially put into operation. The purpose of this process is to teach the model how to deduce the accurate material state from the sensor input signals. Specifically, it first requires collecting historical operating condition datasets. Under various typical material transportation conditions, the system synchronously records raw data from the power monitor 120, acoustic probe 320, and industrial camera 310. This data constitutes the model's input features. Simultaneously, the actual state of the material under these operating conditions must be obtained as the answer or supervision label for model learning. This is achieved through offline calibration: a high-precision weighing hopper is used to actually measure the baseline mass flow rate of the material at the discharge port, and the sampled material is dried to obtain its baseline apparent moisture content. After obtaining sufficient input features and corresponding supervision labels, model training begins. During training, the model continuously adjusts its internal weights, essentially optimizing its complex internal computational logic. The goal is to continuously reduce the error between the results inferred from the input features of the three sensors and the actual supervision labels, namely the baseline flow rate and moisture content. This process continues until the model converges and forms a stable nonlinear mapping relationship. At this point, the model has the ability to accurately estimate the material state from the sensor signals.

[0067] The specific steps of the optimal atomization strategy predictive control include:

[0068] Prediction: The controller 500 establishes a short-time prediction model based on the rate of change of instantaneous flow rate and apparent moisture content to predict the material flow status within the next control cycle;

[0069] Optimization solution: Based on the material flow status, wind speed disturbance data, and a dynamic mechanism model describing the relationship between spray gun action and dust concentration and exhaust humidity, the controller 500 solves the objective function through a rolling optimization algorithm to determine the start-stop sequence of the atomizing spray gun 230.

[0070] The dynamic mechanism model aims to establish the dynamic causal relationship between the actions of the atomizing spray gun 230, its inputs, and the final dust concentration and exhaust humidity, thereby supporting predictive control. Logically, the model includes a multi-input multi-output time-series prediction structure. It receives the future material flow state from the prediction sub-step, the wind speed disturbance from the anemometer 220, and the spray gun start-up and shutdown timing u(t+k) to be optimized as inputs. The model characterizes the physical laws governing the collision, condensation, and settling of water mist particles and dust particles, as well as the dynamic changes in gas humidity within the dust suppression hood 200, considering the coupling effect of material properties and environmental wind disturbances.

[0071] In this embodiment, the core of the optimal atomization strategy predictive control step lies in achieving anticipatory suppression rather than lag response; the prediction sub-step is a prerequisite for achieving this goal; the short-term prediction model within the controller 500 continuously analyzes the instantaneous flow rate and apparent moisture content provided by the soft measurement step, especially their rate of change; if a rapid upward trend in flow rate and dryness is detected, the model will infer that the material flow state will be high flow rate and dry in the future control cycle, for example, 0.5 seconds later; the subsequent optimization solution sub-step makes decisions based on this predicted material flow state, rather than the current state; the controller 500 calls the dynamic mechanism model, which integrates the predicted material state, wind speed disturbance data obtained from the anemometer 220, and spray gun action, i.e., the mathematical relationship between the flow rate of different spray gun combinations and the final dust concentration and exhaust humidity; this dynamic mechanism model is established during the equipment commissioning phase through system identification methods. Its establishment logic steps are as follows:

[0072] Data acquisition: During debugging, the controller 500 applies a series of preset, dynamically changing test control signals to the atomizing spray gun 230, such as rapid start and stop at different duty cycles or applying pseudo-random sequence signals to simulate different spraying actions.

[0073] Response recording: During this process, wind speed disturbance data from anemometer 220, dust concentration data from turbidity sensor 410, and exhaust humidity data from humidity sensor 420 are recorded simultaneously at high speed.

[0074] Model identification: The control signal of the spray gun used for testing and the measured wind speed disturbance are used as the input data of the model, and the measured dust concentration and exhaust humidity are used as the output data of the model. Data-driven identification algorithms are used, such as ARX model identification based on least squares method or training a recurrent neural network (RNN) to fit these input-output time series data.

[0075] Model solidification: A mathematical model, i.e. a dynamic mechanism model, is obtained that can accurately describe how the spray gun action and wind speed disturbance jointly and dynamically affect dust concentration and exhaust humidity. This model is solidified in the controller 500 and can be called in real time for the optimization solution step.

[0076] Optimizing the computational logic of the solution sub-steps essentially involves solving a constrained optimization problem. The objective function is to minimize the short-term future predicted by the dynamic mechanism model, such as the future, while satisfying the constraints. The cumulative dust concentration within each control cycle.

[0077] The computational logic can be expressed as:

[0078] Minimize objective:

[0079] ;

[0080] Constraints:

[0081] For all ;

[0082] For all ;

[0083] The meanings of the letters involved are not clearly stated in the original text:

[0084] The dynamic mechanism model predicts the future... Dust concentration for each control cycle.

[0085] The dynamic mechanism model predicts the future... The exhaust humidity is controlled for each cycle.

[0086] : The preset maximum allowable threshold for exhaust humidity, i.e., the constraint condition.

[0087] The controller 500 will output in the future... The spray gun control action for each control cycle, such as PWM duty cycle.

[0088] and The physical limitations of spray gun control actions, such as , .

[0089] The number of control cycles contained in the prediction time domain of rolling optimization.

[0090] Controller 500 in the current At any given moment, based on this mathematical framework, the future is calculated iteratively. Optimal control sequence of steps However, only the calculated first step of the control action is considered. Send to atomizing spray gun 230 for execution; in the next control cycle When the time comes, the system repeats the entire calculation process described above.

[0091] The controller 500 iteratively calculates using a rolling optimization algorithm to solve the objective function: under the premise of satisfying the exhaust humidity constraint, what start-stop sequence of the atomizing spray gun 230 should be used to minimize the predicted dust concentration; the calculated optimal start-stop sequence is immediately sent to the atomizing spray gun 230 for execution. In this way, when the high-flow-rate dried material actually arrives at the discharge port, the matched and enhanced mist field has already been pre-positioned, thereby effectively suppressing dust generation.

[0092] It also includes sensor status self-diagnosis and degradation steps:

[0093] Verification: The controller 500 continuously runs the consistency verification logic to cross-compare the data streams of the power monitor 120, acoustic probe 320 and industrial camera 310;

[0094] Diagnosis: When persistent data discrepancies are detected, it is determined that a specific sensor has malfunctioned;

[0095] Degradation: The controller 500 immediately triggers the dynamic control strategy degradation process, which activates a pre-trained, degraded soft measurement model that does not rely on fault sensors, based on the material characteristic parameters before the fault and failure.

[0096] In this embodiment, to improve the system's operational reliability in harsh mining environments, a sensor status self-diagnosis and degradation step is added. The verification logic runs continuously in the background of the controller 500, performing real-time cross-comparison of data from the power monitor 120, acoustic probe 320, and industrial camera 310. Under normal operating conditions, the data from these three sources should have a certain correlation. The diagnostic step is triggered when a persistent data discrepancy is detected. For example, if the image from the industrial camera 310 shows a zero material curtain cross-section due to lens contamination, but the signals from the power monitor 120 and acoustic probe 320 both indicate a high flow rate of material, the system determines that the industrial camera 310 has malfunctioned after several seconds of persistent discrepancy. Once the diagnosis is confirmed, the system does not stop operating but executes a degradation process. The controller 500, based on the fault, which sensor failed, and the material characteristic parameters before the failure (e.g., determining whether the material was dry, hard ore or wet, sticky fine ore before the failure), activates the most suitable degraded soft measurement model from a pre-stored model library. This degradation model is pre-trained during the commissioning phase. Its characteristic is that it does not rely on data from faulty sensors; for example, it only uses power and acoustic data for estimation. Although this dynamic degradation method sacrifices some control accuracy, it ensures that the entire adaptive dust suppression function can continue to operate even if a single sensing component fails, thus avoiding complete system paralysis.

[0097] Example 2

[0098] Please see Figure 1-3 A conveying device for ore mining with a dust suppression mechanism at the discharge port, the device comprising:

[0099] The end of the conveyor 100 is used for conveying materials;

[0100] Dust suppression hood 200 is fixedly attached to the discharge port of the end of the conveyor 100;

[0101] Power monitor 120 is used to monitor the load current of the motor driving the conveyor;

[0102] The sensing array 300 includes an industrial camera 310 and an acoustic probe 320. The industrial camera 310 is fixed to the frame of the dust suppression cover 200, and the acoustic probe 320 is fixed to the end of the conveyor 100.

[0103] Atomizing spray gun 230 is installed in an array on the inner wall of dust suppression hood 200;

[0104] The status monitor 400 includes a turbidity sensor 410 and a humidity sensor 420 installed in the exhaust duct 210 of the dust suppression hood 200.

[0105] Anemometer 220 is installed inside exhaust duct 210;

[0106] The controller 500 is connected to the power monitor 120, the sensing array 300, the atomizing spray gun 230, the status monitor 400, and the anemometer 220.

[0107] This embodiment provides an ore mining and transportation device with a dust suppression mechanism at the discharge port. The structure of this device provides a physical carrier for implementing the above method. The end of the conveyor 100 is the source of the material. The dust suppression hood 200 covers the discharge port, forming a basic space to suppress dust diffusion. The atomizing spray gun 230 is installed on its inner wall to generate a mist field. The power monitor 120 is installed at the conveyor drive motor to obtain the motor load current. The sensing array 300 is the core input unit, and its industrial camera 310 and acoustic probe 320 are used to capture the morphology and vibration information of the material flow. The exhaust pipe 2... Unit 10 is used to maintain the air pressure balance inside the hood. It is equipped with an anemometer 220 to monitor environmental wind disturbances and a status monitor 400. The turbidity sensor 410 and humidity sensor 420 in the status monitor 400 measure the final dust concentration and humidity as feedback signals for system control, respectively. The controller 500 is the central hub of the entire device. It can be a programmable logic controller 500 such as Siemens S7-1500 or an industrial computer such as Beckhoff C6930. It connects to all the above-mentioned sensing and actuating components and is responsible for data collection, calculation and instruction issuance.

[0108] A discharge chute 110 is installed on the end of the conveyor 100. The acoustic probe 320 is a piezoelectric contact microphone, which is tightly fixed to the outer wall of the bottom plate of the discharge chute 110 through a rigid connecting seat.

[0109] In this embodiment, the installation method of the acoustic probe 320 has a clear functional purpose. The probe is a piezoelectric contact microphone; it is not suspended in the air, but is tightly fixed to the outer wall of the bottom plate of the discharge chute 110 via a rigid connecting seat, such as a metal L-shaped bracket. The discharge chute 110 itself is used to collect material flow. The purpose of this installation method is to utilize the metal bottom plate of the discharge chute 110 as an acoustic waveguide. When the material rubs and impacts the bottom plate of the chute, the resulting structural vibration is directly transmitted to the rigidly connected acoustic probe 320. This design can effectively pick up high signal-to-noise ratio structural vibration acoustic signals, while mechanically filtering out interference noise propagating from the air, such as the high-frequency wind noise and water spray noise when the atomizing spray gun 230 is working, ensuring that the vibration signal acquired by the controller 500 mainly originates from the impact of the material itself.

[0110] The vibration sound pattern of this structure directly reflects the particle size distribution and impact kinetic energy of the material, and is one of the key input features for the controller 500 to estimate instantaneous flow rate and apparent moisture content;

[0111] Industrial camera 310 is a high-speed industrial camera. A high-brightness LED strobe light source that is synchronously triggered is fixed next to the lens of industrial camera 310 through the same mounting bracket.

[0112] In this embodiment, the industrial camera 310 is configured to solve the problem of clear imaging of high-speed moving objects. A high-speed industrial camera, such as the Basler acA1920-155um model, is selected, featuring a high frame rate. More importantly, a high-brightness LED strobe light source, synchronously triggered, is fixed next to the lens of the industrial camera 310 via a mounting bracket, ensuring the light source and lens angle are aligned. This high-brightness LED strobe light source is controlled by the controller 500, ensuring its emission pulses are strictly synchronized with the camera's shutter opening. Its function is to provide high-intensity illumination for an extremely short instant during camera exposure, such as tens of microseconds. This brief exposure time is sufficient to freeze the high-speed falling ore material, effectively eliminating motion blur. This makes every frame acquired by the controller 500 from the camera clearly visible, providing a high-quality data foundation for subsequent image segmentation, material curtain cross-sectional contour extraction, and surface texture analysis.

[0113] The power monitor 120 is a high-precision, wide-frequency current transformer that is non-contactly clamped to the main power supply cable of the motor driving the conveyor.

[0114] In this embodiment, the selection and installation method of the power monitor 120 balances accuracy and convenience. The monitor uses a high-precision, wide-frequency current transformer, such as the IT400-S series from the Swiss company LEM. This type of transformer can accurately respond to current fluctuations caused by changes in motor load. Its installation method is a non-contact clamp structure. This means that during installation, only the induction ring of the transformer needs to be opened and clamped onto the main power supply cable of the motor driving the conveyor, without cutting the cable or connecting it in series in the main circuit. This design ensures both safety and convenience in installation and maintenance. Furthermore, the non-contact measurement does not cause any electrical interference to the existing power system, enabling it to stably measure the load current flowing through the motor. This current value is used to calculate the reference mass flow rate of the material carried by the conveyor.

[0115] The atomizing spray gun 230 is controlled by an independent electronically controlled high-frequency pulse width modulation valve.

[0116] In this embodiment, the atomizing spray gun 230 has precise adjustment capabilities. Each atomizing spray gun 230 constituting the array has an independent electrically controlled high-frequency pulse width modulation valve connected to its water inlet pipe; this valve, for example a small high-speed solenoid valve, receives a pulse width modulation control signal from the controller 500. The controller 500 does not adjust the valve opening by analog voltage, but rather by sending high-frequency digital pulse signals and adjusting the duty cycle of the pulses, i.e., the proportion of energized time within one cycle, to control the average opening time of the valve. For example, a 50% duty cycle means that the valve is open for a cumulative period of 0.5 seconds per second; due to the high valve switching frequency, the water flow response exhibits a smooth flow rate change. This independent control method allows the controller 500 to quickly and precisely adjust the flow rate of any spray gun in the array from 0 to 100%, thereby dynamically reconstructing the morphology, concentration, and coverage area of ​​the mist field to match the complex start-stop sequence calculated by the optimal atomization strategy predictive control steps.

[0117] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An ore mining and conveying device with a dust suppression mechanism at the discharge port, characterized in that, The device includes: The end of the conveyor (100) is used for conveying materials; A dust suppression hood (200) is fixedly attached to the discharge port of the end of the conveyor (100); A power monitor (120) is used to monitor the load current of the motor driving the conveyor; The sensing array (300) includes an industrial camera (310) and an acoustic probe (320), the industrial camera (310) being fixed to the frame of the dust suppression hood (200) and the acoustic probe (320) being fixed to the end of the conveyor (100). Atomizing spray guns (230) are installed in an array on the inner wall of the dust suppression hood (200); The status monitor (400) includes a turbidity sensor (410) and a humidity sensor (420) installed in the exhaust duct (210) on the dust suppression cover (200). An anemometer (220) is installed inside the exhaust duct (210); The controller (500) is connected to the power monitor (120), the sensing array (300), the atomizing spray gun (230), the status monitor (400), and the anemometer (220), respectively. The ore transportation method using a transportation device includes the following steps: Material flow characteristics soft measurement: The instantaneous flow rate and apparent moisture content of the material are estimated in real time by integrating the current signal of the power monitor (120), the vibration signal of the acoustic probe (320) and the image signal of the industrial camera (310) through the controller (500); Predictive control of optimal atomization strategy: The controller (500) performs rolling optimization through a dynamic mechanism model based on the estimated instantaneous flow rate and apparent moisture content, combined with the wind speed disturbance data of the anemometer (220), and outputs control commands to the atomizing spray gun (230). Feedback constraints and closed loop: When optimizing, the controller (500) takes the dust concentration measured by the turbidity sensor (410) as the minimum target and the exhaust humidity measured by the humidity sensor (420) as the constraint condition that it does not exceed the preset threshold. The soft measurement steps for material flow characteristics specifically include: Data acquisition: The power monitor (120) acquires the motor load current as a reference mass flow rate, the acoustic probe (320) picks up the structural vibration acoustic pattern of the material impact, and the industrial camera (310) acquires the instantaneous cross-sectional profile of the material by exposing it with a high-brightness LED strobe light source. Fusion estimation: The deep neural network model in the controller (500) receives the current signal, the vibration sound pattern and the cross-sectional profile, and outputs the estimated values ​​of the instantaneous flow rate and the apparent water content through multimodal fusion calculation; The optimal atomization strategy prediction and control steps specifically include: Prediction: The controller (500) establishes a short-time prediction model based on the rate of change of the instantaneous flow rate and the apparent moisture content to predict the material flow status within a future control cycle; Optimization solution: The controller (500) determines the start-stop sequence of the atomizing spray gun (230) by solving the objective function through a rolling optimization algorithm based on the material flow state, the wind speed disturbance data, and the dynamic mechanism model describing the relationship between the spray gun action and the dust concentration and the exhaust humidity. A discharge chute (110) is installed on the end of the conveyor (100). The acoustic probe (320) is a piezoelectric contact microphone, which is tightly fixed to the outer wall of the bottom plate of the discharge chute (110) by a rigid connecting seat. The industrial camera (310) is a high-speed industrial camera (310), and a synchronously triggered high-brightness LED strobe light source is fixed next to the lens of the industrial camera (310) through the same mounting bracket; The atomizing spray gun (230) is controlled by an independent electronically controlled high-frequency pulse width modulation valve.

2. The ore mining and conveying device with a dust suppression mechanism at the discharge port according to claim 1, characterized in that, The deep neural network model is established through the following steps: During the commissioning phase, historical operating condition datasets are collected, and the current signal of the power monitor (120), the vibration signal of the acoustic probe (320), and the image signal of the industrial camera (310) are recorded synchronously as input features for the deep neural network model. A high-precision weighing hopper is used for periodic offline calibration to obtain a baseline mass flow rate, and the material sample is dried to obtain a baseline apparent moisture content. The baseline mass flow rate and apparent moisture content data are used as supervision labels for the deep neural network model. Adjust the weights within the deep neural network model to establish a nonlinear mapping relationship from the input features of the power monitor (120), acoustic probe (320), and industrial camera (310) to the supervision label.

3. The ore mining and conveying device with a discharge port dust suppression mechanism according to claim 1, characterized in that, It also includes sensor status self-diagnosis and degradation steps: Verification: The controller (500) continuously runs consistency verification logic to cross-compare the data streams of the power monitor (120), the acoustic probe (320), and the industrial camera (310); Diagnosis: When persistent data discrepancies are detected, it is determined that a specific sensor has malfunctioned; Degradation: The controller (500) immediately triggers the dynamic control strategy degradation process, and activates a pre-trained, degraded soft measurement model that does not depend on the specific sensor that caused the failure, based on the material characteristic parameters before the failure.

4. The ore mining and conveying device with a dust suppression mechanism at the discharge port according to claim 1, characterized in that, The power monitor (120) is a high-precision wideband current transformer, which is non-contactly clamped to the main power supply cable of the motor driving the conveyor.

Images