Intelligent long-distance stock bin multi-point linkage dust removal and closing system and control method thereof

Abstract

The invention provides an intelligent long-distance stock bin multi-point linkage dust removal and closing system. The intelligent long-distance stock bin multi-point linkage dust removal and closing system comprises an air suction module, a closing module, a material level monitoring module and an automatic control module. The air suction module is provided with an air suction device on the bin opening side; the sealing module is provided with a cover plate with a switch module above or on the side surface of the stock bin opening; the material level monitoring module is provided with a radar material level meter to obtain material level data in real time. When the unloading trolley moves to a stock bin opening, the ejector rod abuts against the switch module to open the cover plate, and the automatic control module controls opening and closing of the air suction device according to material level data. The system controls opening and closing of the cover plate in real time, ensures tight sealing in the discharging process, prevents dust leakage, accurately manages the opening and closing time of the air suction device and the cover plate through intelligent control, reduces energy consumption and dust emission, and achieves the purposes of energy conservation and environmental protection.

Description

Technical Field

[0001] The present application relates to the field of warehouse dust removal, and in particular to an intelligent long-distance silo multi-point linkage dust removal and closure system and a control method thereof. Background Art

[0002] A large amount of dust is generated during the unloading process of the silo. If there is a lack of effective dust removal measures, it will not only cause environmental pollution, but also affect the normal operation of the equipment and the health of workers. Traditional dust removal methods often use a single dust removal device, which is difficult to cover the entire silo area, especially near the silo opening, where the dust problem is particularly prominent.

[0003] The silo opening needs to be opened during the unloading process to allow the unloading trolley to operate, but the traditional sealing method often has the problem of poor sealing, resulting in dust leakage. This not only affects the dust removal effect, but also causes a waste of resources.

[0004] Traditional silo dust removal and closed systems often rely on manual operation, which is not only inefficient but also prone to poor dust removal effects or waste of resources due to misoperation. Summary of the invention

[0005] The purpose of the present application is to overcome the above-mentioned prior art and provide an intelligent long-distance silo multi-point linkage dust removal and closure system and a control method thereof.

[0006] The present application provides an intelligent long-distance silo multi-point linkage dust removal and closure system, including: a suction module, a closure module, a material level monitoring module and an automation control module;

[0007] The suction module: a suction device is respectively arranged on the side of multiple silo openings on the preset path of the unloading trolley in the silo; the closing module: a cover plate is respectively installed above or on the side of the multiple silo openings, and a switch module is arranged on the cover plate; the material level monitoring module: a radar material level meter is arranged inside or on the top of the silo to obtain material level data in real time;

[0008] When the unloading trolley reaches the silo opening, the push rod provided on the unloading trolley presses against the switch module to open the cover plate;

[0009] The automatic control module receives the material level data and determines according to the material level data that if the unloading trolley reaches the silo port, the suction device is turned on and after the unloading is completed, the suction device is turned off.

[0010] Optionally, an optimization module is also included:

[0011] Deploy dust sensors at key points on the preset path where dust is likely to be generated to collect dust data; and establish a dust diffusion prediction model by combining historical dust data with the dust data;

[0012] Acquire dust prediction results of the key points according to the dust diffusion prediction model, and set dynamic priorities for the key points according to the dust prediction results;

[0013] The air volume of the air suction device corresponding to each of the key points is set according to the dynamic priority.

[0014] Optionally, combining historical dust data and the dust data to establish a dust diffusion prediction model includes:

[0015] Graph neural networks are used to jointly model the dust diffusion in the silo in space and time, and the interactions and dependencies between dust in different areas and time points are obtained.

[0016] Optionally, dust sensors are deployed at key points on the preset path where dust is easily generated, and also include: environmental sensors for monitoring wind speed and temperature.

[0017] Optionally, the dust diffusion prediction model is trained using a machine learning algorithm, including: LSTM, CNN or a combination thereof.

[0018] Optionally, combining historical dust data and currently monitored dust data, a dust diffusion prediction model is established, including:

[0019] The dust diffusion prediction model is constructed according to historical dust data, and the dust diffusion prediction model is adjusted according to newly acquired dust data.

[0020] The present application also provides an intelligent long-distance silo multi-point linkage dust removal and closed control method, including:

[0021] Suction devices are respectively arranged on the sides of multiple silo openings on the preset path of the unloading trolley in the silo; cover plates are respectively installed above or on the sides of the multiple silo openings, and switch modules are arranged on the cover plates; a radar level meter is arranged inside or on the top of the silo to obtain material level data in real time;

[0022] When the unloading trolley reaches the silo opening, the push rod provided on the unloading trolley presses against the switch module to open the cover plate;

[0023] Receive the material level data; determine according to the material level data: if the unloading trolley reaches the silo port, control to open the air suction device, and after unloading is completed, control to close the air suction device;

[0024] Deploy dust sensors at key points on the preset path where dust is likely to be generated to collect dust data; and establish a dust diffusion prediction model by combining historical dust data with the dust data;

[0025] Acquire dust prediction results of the key points according to the dust diffusion prediction model, and set dynamic priorities for the key points according to the dust prediction results;

[0026] The air volume of the air suction device corresponding to each of the key points is set according to the dynamic priority.

[0027] Optionally, combining historical dust data and currently monitored dust data, a dust diffusion prediction model is established, including:

[0028] Graph neural networks are used to jointly model the dust diffusion in the silo in space and time, and the interactions and dependencies between dust in different areas and time points are obtained.

[0029] Optionally, dust sensors are deployed along preset paths and key points in the silo where dust is easily generated, and also include: environmental sensors for monitoring wind speed and temperature.

[0030] Optionally, the dust diffusion prediction model is trained using a machine learning algorithm, including: LSTM, CNN or a combination thereof.

[0031] The beneficial effects of this application are:

[0032] The present application provides an intelligent long-distance silo multi-point linkage dust removal and closure system, including: a suction module, a closure module, a material level monitoring module and an automatic control module; the suction module: a suction device respectively arranged on the side of multiple silo openings on the preset path of the unloading trolley in the silo; the closure module: a cover plate respectively installed above or on the side of the multiple silo openings, and a switch module is arranged on the cover plate; the material level monitoring module: a radar level meter is arranged inside or on the top of the silo to obtain the material level data in real time; if the unloading trolley reaches the silo opening, the top rod arranged on the unloading trolley presses against the switch module to open the cover plate; the automatic control module: receives the material level data; judges according to the material level data: if the unloading trolley reaches the silo opening, the suction device is controlled to be turned on, and after the unloading is completed, the suction device is controlled to be turned off. The present application ensures a tight closure during the unloading process by controlling the opening and closing of the cover plate in real time according to the position of the unloading trolley, thereby preventing the leakage of dust. At the same time, intelligent control is achieved by introducing material level monitoring module and automatic control module. By accurately controlling the opening and closing time of the suction device and cover, unnecessary energy consumption and dust emission are avoided, thus achieving the goal of energy saving and environmental protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a schematic diagram of the intelligent long-distance silo multi-point linkage dust removal and closure system in this application;

[0034] Figure 2 , is a schematic diagram of a switch module in this application;

[0035] Figure 3 It is a schematic diagram of the ejector in this application;

[0036] Figure 4 This is a flow chart of the intelligent long-distance silo multi-point linkage dust removal and closed control method of this application. DETAILED DESCRIPTION

[0037] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0038] like Figure 1 As shown, the present application also provides an intelligent long-distance silo multi-point linkage dust removal and closure system, including:

[0039] The suction module 101 is composed of: a suction device is installed on each side of the silo opening, and these devices are connected to the central dust removal system through pipelines. When the unloading trolley arrives at the silo opening for unloading, the suction device is started, and the dust generated near the silo opening is sucked into the pipeline through the negative pressure generated, and then enters the dust removal system for purification.

[0040] The sealing module 102 is composed of: a cover plate which can be opened and closed automatically is installed above or on the side of the silo opening.

[0041] like Figure 2 As shown, the cover plate is provided with a switch module. The switch module comprises: a force-bearing sheet 201 and a fixing sheet 202, wherein the force-bearing sheet 201 and one side of the fixing sheet 202 are connected by a hinge 203, a control rod is provided on one side of the force-bearing sheet, and the control rod comprises a first rod 204 and a second rod 206 fixed perpendicularly to each other, a control rod slot 205 is provided on the fixing sheet 202, and the control rod passes through the control rod slot 205 and extends into the other side of the fixing sheet 202. The inward angle of the end of the second rod 206 is set to an arc shape.

[0042] like Figure 3As shown, corresponding to the switch module, a push rod is provided on the unloading trolley. The push rod includes a base rod 207 and an action rod 210. The action rod 210 extends into a blind hole provided on the base rod 207. A spring 208 is provided between one end of the action rod 210 extending into the blind hole and the bottom of the blind hole. A snap ring 211 is provided on the action rod 210, and a U-shaped fixing piece 209 is used on the outside of the snap ring to significantly fix the action rod 210 in a position where it does not leave the blind hole. The U-shaped fixing piece 209 is threadedly connected to one side of the blind hole of the base rod 207.

[0043] The cover plate is slidably fixed at the silo opening, and when the front side of the unloading trolley in the advancing direction reaches the cover plate, the push rod abuts against the action piece 201. When the unloading trolley continues to move, the push rod slides the cover plate open through the switch module.

[0044] The unloading trolley continues to move forward, the second rod 206 enters the preset groove on the sliding groove of the cover plate, the action piece 201 tilts, the push rod compresses the spring 208 and passes over the action piece 201.

[0045] A spring is provided between the cover plate and the silo opening, and the spring 208 is compressed when the cover plate is opened. When the push rod passes over the action piece 201, the spring pushes the cover plate to close, and at this time, the arc provided on the second rod causes the second rod to slide out of the preset groove on the slide slot.

[0046] The material level monitoring module 103 is composed of: a radar level meter installed inside or on the top of the silo, which is connected to the automation control module by wireless or wired means. The radar level meter measures the material height in the silo in real time and sends the data to the automation control module. These data are used to determine the position of the unloading trolley and the unloading progress.

[0047] The automation control module 104 is composed of a central processing unit (CPU), an input / output interface (I / O interface), a memory and a communication module. The automation control module receives the material level data from the material level monitoring module and determines the position of the unloading trolley according to a preset algorithm. When it is determined that the unloading trolley is about to reach the silo port, the control module sends a control signal to the suction module through the I / O interface to start the suction device. After unloading is completed, the control module sends a control signal again to turn off the suction device.

[0048] The unloading trolley moves in the silo along the preset path. The radar level meter measures the material height in the silo in real time and sends the data to the automatic control module. The automatic control module determines the position of the unloading trolley based on the material level data. When the unloading trolley is about to reach the silo opening, the automatic control module sends a control signal to the suction module, including: the suction device is started, allowing the unloading trolley to carry out unloading operations. After unloading is completed, the automatic control module sends a control signal again to turn off the suction device.

[0049] Furthermore, the system further includes: an optimization module 105, including:

[0050] Arrangement unit, deploy dust sensors along the preset path and key points where dust is likely to be generated in the silo to collect dust data;

[0051] The model unit combines the historical dust data and the currently monitored dust data to establish a dust diffusion prediction model, and the dust diffusion prediction model is expressed as follows:

[0052]

[0053] Where C is the dust concentration, t is the time, u is the wind speed vector, D is the dust diffusion coefficient, S is the source term, which indicates the rate of dust generation. is the gradient operator, is the divergence operator;

[0054] The prediction unit sets dynamic priorities for each dust removal point based on the results of the dust diffusion prediction model;

[0055] The deployment unit sets the air volume and cleaning cycle of each dust sensor deployment position according to the dynamic priority.

[0056] The execution unit controls the cover of the silo to open and starts the dust removal equipment when the unloading vehicle starts working; after the unloading is completed, the cover is controlled to close, and the dust removal equipment performs the cleaning task according to the current air volume and cleaning cycle.

[0057] Furthermore, combining historical dust data with the currently monitored dust data, a dust diffusion prediction model is established, including:

[0058] Graph neural networks are used to jointly model the dust diffusion in the silo in space and time, and the interactions and dependencies between dust in different areas and time points are obtained.

[0059] Furthermore, dust sensors are deployed along preset paths and key points in the silo where dust is easily generated, and also include: environmental sensors for monitoring wind speed and temperature.

[0060] Furthermore, the dust diffusion prediction model is trained using a machine learning algorithm, including but not limited to LSTM, CNN or a combination thereof.

[0061] Furthermore, combining historical dust data with the currently monitored dust data, a dust diffusion prediction model is established, including:

[0062] The dust diffusion prediction model is constructed according to historical dust data, and the dust diffusion prediction model is adjusted according to newly acquired dust data.

[0063] Please refer to Figure 2 The method for intelligent long-distance silo multi-point linkage dust removal and closure control comprises the following steps:

[0064] S201. Air suction devices are respectively arranged at the sides of multiple silo openings on a preset path of a discharge trolley in the silo.

[0065] A set of suction devices is installed on each silo opening side on the preset path of the unloading trolley in the silo. These devices should be close to the silo opening to ensure that the dust generated during the unloading process can be effectively sucked.

[0066] The air suction device includes a fan, a dust suction pipe and a dust filter bag. The fan provides suction, and the dust suction pipe guides the dust into the dust filter bag for filtration to ensure that the exhaust gas meets environmental protection standards.

[0067] S202. Cover plates are installed above or on the sides of the plurality of silo openings.

[0068] An automatically openable and closable cover is installed above or on the side of each silo opening. The size and shape of the cover should ensure that it can completely cover the silo opening to prevent dust from leaking out.

[0069] The cover is controlled by an electric or pneumatic device and can receive signals from the automation control module to achieve automatic opening and closing.

[0070] S203, a radar level meter is arranged inside or on the top of the silo to obtain material level data in real time.

[0071] Inside or on the top of the silo, choose a suitable location to install the radar level meter according to the shape and size of the silo. The radar level meter should be able to accurately measure the height of the material in the silo. The radar level meter obtains the material level data in real time and transmits the data to the automation control module via wired or wireless means. These data are used to determine the position of the unloading trolley and the unloading progress.

[0072] S204, receiving the material level data; determining the position of the unloading trolley according to the material level data: if the unloading trolley reaches the silo opening, controlling the air suction device and the cover plate to be turned on; after unloading is completed, controlling the air suction device and the cover plate to be turned off.

[0073] The automation control module receives the material level data from the radar level meter via wired or wireless means.

[0074] The automation control module determines whether the unloading trolley is about to reach the silo port based on the received material level data, combined with the preset algorithm and the preset path of the unloading trolley.

[0075] When it is determined that the unloading trolley is about to reach the silo opening, the automatic control module sends a control signal to the suction device and the cover plate to start the suction device and open the cover plate to prepare for the unloading operation.

[0076] After unloading is completed, the automatic control module sends a control signal again to close the suction device and the cover to prevent dust leakage and energy waste.

[0077] Furthermore, before the silo starts unloading, ensure that all equipment (including the suction device, cover plate and radar level meter) has been properly installed and debugged. Then start the automation control module to start receiving the material level data and control it.

[0078] S205: deploy dust sensors at key points on the preset path where dust is likely to be generated, and collect dust data.

[0079] Conduct a detailed analysis of the overall structure of the silo, material type, flow path and historical dust generation to identify key areas where dust is prone to generation and the main paths of material flow.

[0080] High-precision dust sensors are installed at key points where dust is easily generated, such as material inlets, outlets, turning points, stockpiling areas, and high drop areas. The sensors must have real-time data collection and remote transmission functions. The deployed dust sensors collect dust concentration data at each monitoring point in real time and transmit the data to the central control system.

[0081] The collected dust data is cleaned to remove outliers and noise to ensure data quality.

[0082] The cleaned data is stored in chronological order to form a historical dust database.

[0083] S206: Establish a dust diffusion prediction model by combining historical dust data and the dust data.

[0084] Combining historical data (such as dust concentration, wind speed, temperature, etc.) and current monitoring data, a dust diffusion prediction model is established using machine learning algorithms (such as LSTM, CNN, etc.) to predict the distribution and change trend of dust in the future.

[0085] According to the actual layout of the silo and the location of the monitoring points, a graph structure is constructed. The nodes in the graph represent different monitoring points or areas, and the edges represent the physical connections or spatial relationships between the nodes.

[0086] Each node is associated with a certain feature vector, including environmental parameters such as historical dust concentration, wind speed, and temperature.

[0087] Design a graph neural network model suitable for dust diffusion prediction. The model should be able to process spatial and temporal data simultaneously and capture the dynamic process of dust diffusion.

[0088] At each time step, the model receives the node features at the current time point as input and updates the node representation through methods such as graph convolution or graph attention mechanism, while taking into account the interactions and dependencies between nodes.

[0089] Over time, the model is able to gradually accumulate and propagate information between nodes to generate a comprehensive prediction of dust dispersion.

[0090] The graph neural network model is trained using historical dust data and environmental parameter data. During the training process, the model optimizes its parameters through the back-propagation algorithm to minimize the prediction error.

[0091] After training, the model can predict dust diffusion in the future based on real-time monitored dust data and environmental parameter data.

[0092] The dust diffusion prediction model expression is as follows:

[0093]

[0094] Where C is the dust concentration, t is the time, u is the wind speed vector, D is the dust diffusion coefficient, S is the source term, which indicates the rate of dust generation. is the gradient operator, is the divergence operator.

[0095] To simplify the solution process, this application optionally uses an LSTM learning model to "learn" the solution to this equation.

[0096] The LSTM model can capture long-term dependencies in time series data through its internal state (cell state) and mechanisms such as forget gate, input gate, and output gate, thereby predicting future dust concentration distribution.

[0097] Although the LSTM model itself is not directly given in the form of a mathematical formula (it is implemented through a neural network architecture), its prediction process can be simplified into a function mapping:

[0098]

[0099] in, is the predicted dust concentration at the next moment, f LSTM is the function implemented by the LSTM model, θ is the parameter of the LSTM model (learned through training data), and other symbols represent input features, which can be determined according to actual conditions (such as temperature, humidity, etc.).

[0100] S207: Obtain dust prediction results of the key points according to the dust diffusion prediction model, and set dynamic priorities for the key points according to the dust prediction results.

[0101] Conduct in-depth analysis on the data output by the model to identify the changing trends of dust concentration, including rising speed, peak time, duration, etc.

[0102] Based on the evaluation results of the dust concentration change rate and potential impact range, dynamic priority is set for each dust removal point. This step considers the following key factors:

[0103] 1. Areas with fast concentration growth, especially those about to reach or exceed safety thresholds, should be given higher priority.

[0104] 2. Evaluate areas that may be affected by dust diffusion, including densely populated areas, environmentally sensitive areas (such as water sources, nature reserves), etc. Dust removal points in these areas should also be given priority.

[0105] 3. Combined with the concentration change rate and potential impact range, comprehensively evaluate the urgency of each dust removal point to determine its priority order.

[0106] According to the set dynamic priorities, formulate and implement corresponding dust removal plans. Areas with high priority will be given priority for dust removal to reduce the negative impact of dust on personnel health, environmental safety and production efficiency.

[0107] S208, setting the air volume of the air suction device corresponding to each of the key points according to the dynamic priority.

[0108] According to the dynamic priority, the fan air volume at each dust sensor deployment location is adjusted to ensure that the high-priority area has sufficient air volume to effectively control dust diffusion. At the same time, according to the dust concentration change trend and dust removal effect, the cleaning cycle of each dust removal point is dynamically adjusted to reduce energy consumption while ensuring dust removal efficiency.

[0109] This application adopts a simplified version of Genetic Algorithm (GA) to optimize the operating parameters of the dust removal equipment.

[0110] Genetic algorithms include three basic operations: selection, crossover, and mutation. This application provides a simplified fitness function example to evaluate the quality of dust removal strategies:

[0111] Fitness function:

[0112]

[0113] Where x is the parameter vector of the dust removal strategy (such as air volume, cleaning cycle, etc.), E(x) is the dust removal efficiency (the higher the better, so the inverse is taken as part of the optimization target), C avg (x) is the average dust concentration (the lower the better), Cost(x) is the operating cost (the lower the better), and α and β are weight coefficients used to balance the importance of different optimization objectives.

[0114] The goal of the genetic algorithm is to find the parameter vector x that maximizes the fitness function F(x). * .

[0115] When the unloading vehicle starts working, the dust removal equipment is started; after the unloading is completed, the dust removal equipment performs the cleaning task according to the current air volume and cleaning cycle.

[0116] When the unloading vehicle receives the instruction to start working, its built-in control system will be activated and it will start to move along the preset track or path to the designated unloading location.

[0117] While the unloading truck is moving, the system will transmit the status of the unloading truck (such as position, speed, etc.) to the central control system in real time through sensors or signal transmission devices.

[0118] When the cover is opened, the central control system will also send a start-up command to the dust removal equipment.

[0119] After receiving the command, the dust removal equipment will immediately start the key components such as the fan and the blowing device to start the dust removal operation in the unloading area. When the unloading vehicle completes the unloading task, its control system will send a signal of unloading completion to the central control system.

[0120] After the cover is closed, the dust removal equipment will perform subsequent cleaning operations based on the current air volume and cleaning cycle. At this time, the dust removal equipment may continue to operate for a period of time to ensure that the residual dust in the unloading area is completely removed.

[0121] Subsequently, the dust removal equipment will gradually adjust to standby or low-power mode according to the instructions of the central control system, waiting for the next unloading task.

[0122] like Figure 4 As shown, the present application also provides an intelligent long-distance silo multi-point linkage dust removal and closed control method, including:

[0123] S301, air suction devices are respectively arranged at the sides of multiple silo openings on the preset path of the unloading trolley in the silo;

[0124] S302, cover plates are respectively installed above or on the sides of the plurality of silo openings, wherein the cover plates are provided with switch modules;

[0125] S303, arranging a radar level meter inside or on the top of the silo to obtain material level data in real time;

[0126] S304: When the unloading trolley reaches the silo opening, a push rod provided on the unloading trolley presses against the switch module to open the cover plate;

[0127] S305, receiving the material level data; judging according to the material level data: when the unloading trolley reaches the silo port, controlling to open the air suction device, and after unloading is completed, controlling to close the air suction device;

[0128] S306 deploys dust sensors at key points on the preset path where dust is likely to be generated, to collect dust data;

[0129] S307 combines historical dust data and the dust data to establish a dust diffusion prediction model;

[0130] S308, obtaining dust prediction results of the key points according to the dust diffusion prediction model, and setting dynamic priorities for the key points according to the dust prediction results;

[0131] S309: setting the air volume of the air suction device corresponding to each of the key points according to the dynamic priority.

[0132] Furthermore, combining historical dust data with the currently monitored dust data, a dust diffusion prediction model is established, including:

[0133] Graph neural networks are used to jointly model the dust diffusion in the silo in space and time, and the interactions and dependencies between dust in different areas and time points are obtained.

[0134] Furthermore, dust sensors are deployed along preset paths and key points in the silo where dust is easily generated, and also include: environmental sensors for monitoring wind speed and temperature.

[0135] Furthermore, the dust diffusion prediction model is trained using a machine learning algorithm, including: LSTM, CNN or a combination thereof.

Claims

1. An intelligent long-distance silo multi-point linkage dust removal and closure system, characterized in that: include: Air suction module, sealing module, material level monitoring module and automation control module; The suction module: a suction device is respectively arranged on the side of multiple silo openings on the preset path of the unloading trolley in the silo; the closing module: a cover plate is respectively installed above or on the side of the multiple silo openings, and a switch module is arranged on the cover plate; the material level monitoring module: a radar material level meter is arranged inside or on the top of the silo to obtain material level data in real time; When the unloading trolley reaches the silo opening, the push rod provided on the unloading trolley presses against the switch module to open the cover plate; The automation control module receives the material level data; According to the material level data, it is judged that: when the unloading trolley reaches the silo port, the suction device is controlled to be turned on; after the unloading is completed, the suction device is controlled to be turned off.

2. According to claim 1, the intelligent long-distance silo multi-point linkage dust removal and closure system is characterized in that: Also includes, optimization module: Deploy dust sensors at key points on the preset path where dust is likely to be generated to collect dust data; Combining historical dust data with the dust data, a dust diffusion prediction model is established; Acquire dust prediction results of the key points according to the dust diffusion prediction model, and set dynamic priorities for the key points according to the dust prediction results; The air volume of the air suction device corresponding to each of the key points is set according to the dynamic priority.

3. According to claim 2, the intelligent long-distance silo multi-point linkage dust removal and closure system is characterized in that: Combining historical dust data with the dust data, a dust diffusion prediction model is established, including: Graph neural networks are used to jointly model the dust diffusion in the silo in space and time, and the interactions and dependencies between dust in different areas and time points are obtained.

4. According to claim 2, the intelligent long-distance silo multi-point linkage dust removal and closure system is characterized in that: Dust sensors are deployed at key points on the preset path where dust is likely to be generated, and also include: environmental sensors for monitoring wind speed and temperature.

5. According to claim 2, the intelligent long-distance silo multi-point linkage dust removal and closure system is characterized in that: The dust diffusion prediction model is trained using a machine learning algorithm, including: LSTM, CNN or a combination thereof.

6. The intelligent long-distance silo multi-point linkage dust removal and closure system according to claim 2 is characterized in that: Combining historical dust data with the currently monitored dust data, a dust diffusion prediction model is established, including: The dust diffusion prediction model is constructed according to historical dust data, and the dust diffusion prediction model is adjusted according to newly acquired dust data.

7. An intelligent long-distance silo multi-point linkage dust removal and closure control method, characterized in that: include: Suction devices are respectively arranged on the sides of multiple silo openings on the preset path of the unloading trolley in the silo; cover plates are respectively installed above or on the sides of the multiple silo openings, and switch modules are arranged on the cover plates; a radar level meter is arranged inside or on the top of the silo to obtain material level data in real time; When the unloading trolley reaches the silo opening, the push rod provided on the unloading trolley presses against the switch module to open the cover plate; receiving the material level data; According to the material level data, it is judged that: when the unloading trolley reaches the silo port, the suction device is controlled to be turned on; after the unloading is completed, the suction device is controlled to be turned off; Deploy dust sensors at key points on the preset path where dust is likely to be generated to collect dust data; Combining historical dust data with the dust data, a dust diffusion prediction model is established; Acquire dust prediction results of the key points according to the dust diffusion prediction model, and set dynamic priorities for the key points according to the dust prediction results; The air volume of the air suction device corresponding to each of the key points is set according to the dynamic priority.

8. The intelligent long-distance silo multi-point linkage dust removal and closure control method according to claim 7 is characterized in that: Combining historical dust data with the currently monitored dust data, a dust diffusion prediction model is established, including: Graph neural networks are used to jointly model the dust diffusion in the silo in space and time, and the interactions and dependencies between dust in different areas and time points are obtained.

9. The intelligent long-distance silo multi-point linkage dust removal and closure control method according to claim 7 is characterized in that: Dust sensors are deployed along preset paths and key points in the silo where dust is likely to be generated, and also include: environmental sensors for monitoring wind speed and temperature.

10. The intelligent long-distance silo multi-point linkage dust removal and closure control method according to claim 7 is characterized in that: The dust diffusion prediction model is trained using a machine learning algorithm, including: LSTM, CNN or a combination thereof.

Images