An automatic feeding system

Abstract

The present invention provides an automatic feeding system, and the technical solution is as follows: it includes a feed bin, a pneumatic conveying pipeline, an air source device and a discharge bin, and also includes a distributed sensor group for monitoring the flux, pressure, temperature and vibration spectrum of preset points of the pneumatic conveying pipeline, including flow sensors, pressure sensors, temperature sensors and vibration acceleration sensors arranged at several preset points of the conveying pipeline, a digital twin model construction module, which constructs a three-dimensional modeling of the pneumatic conveying pipeline and maps the state parameters of each position of the pipeline distributed sensor group in real time; a blockage analysis module, which marks the blockage position to the three-dimensional modeling in real time based on the data collected by the distributed sensor group, extracts the multi-parameter change curve in the T period before the blockage and constructs a blockage feature library, and generates a blockage risk level through the difference in the weight change rate between the feed bin and the discharge bin; a graded dredging execution module, which is equipped with a differentiated dredging strategy and starts the differentiated dredging strategy according to the blockage risk level.

Description

Technical Field

[0001] The present invention relates to the field of powder material transportation, and in particular to an automatic feeding system. Background Art

[0002] Powder extrusion feeding systems are primarily used for conveying powders. Currently, the most common methods used for powder conveying on the market include screw feeders and pneumatic conveying systems. Pneumatic conveying systems utilize positive or negative air pressure to transport powders through pipelines. These systems are used for both long-distance dilute-phase conveying and short- to medium-distance dense-phase conveying. Dense-phase conveying uses low-speed airflow to transport highly concentrated powders. The solid-to-gas ratio is typically greater than 15, or even over 100. Powders are not fully suspended within the conveying pipeline, but rather form plugs, dunes, or fluidized beds. For extrusion feeding, medium-distance extrusion is more suitable, allowing for more gentle delivery of powders to the vicinity of the extruder compared to dilute-phase conveying. A pneumatic conveying system consists of an air source, a delivery tank, a conveying pipeline, a buffer hopper, and a control system. Because dense-phase conveying uses low-speed airflow, the risk of pipeline blockage is a common challenge. Pipeline blockages can be caused by improper pipeline design, incorrect operating parameters, changes in material properties, pipeline deformation, or residual material within the pipeline.

[0003] The existing technology usually adopts the method of setting pressure detection at intervals and elbows on the conveying pipeline, and determining the blockage point based on the principle that the pressure upstream of the blockage point rises sharply and the pressure downstream drops sharply to normal pressure. However, this method has obvious hysteresis and can only be detected after the blockage occurs. Maintenance personnel usually dredge the blockage through four methods: reverse airflow blowing, segmented disassembly and dredging, vibrator-driven loosening of compacted materials in conjunction with airflow blowing, and special dredging devices. However, these methods all require manual intervention, and a lot of time will be wasted from discovering the blockage point to completing the dredging, affecting production efficiency. In addition, the existing technology lacks a real-time monitoring and early warning mechanism for the blockage process, and it is impossible to predict the risk of blockage in advance, nor can it take differentiated dredging measures according to the degree of blockage. Summary of the Invention

[0004] The purpose of the present invention is to provide a pneumatic conveying pipeline blockage warning and graded dredging system based on digital twins, which has the advantages of real-time monitoring of pipeline status, early warning of blockage risks and automatic execution of graded dredging strategies, effectively reducing manual intervention and improving conveying efficiency.

[0005] In order to solve the above technical problems, the present invention is solved by the following technical solutions: an automatic feeding system, including a feeding bin, a pneumatic conveying pipeline, an air source device and a discharge bin, and also includes:

[0006] Distributed sensor group, used to monitor the flux, pressure, temperature and vibration spectrum of the preset points of the pneumatic conveying pipeline, including flow sensors, pressure sensors, temperature sensors and vibration acceleration sensors set at several preset points of the conveying pipeline,

[0007] The digital twin model construction module constructs a three-dimensional model of the pneumatic conveying pipeline and maps the status parameters of each position of the pipeline distribution sensor group in real time;

[0008] The congestion analysis module uses data collected by a distributed sensor group to mark the congestion location in real time and build a 3D model. It extracts the multi-parameter change curves in the T period before the congestion and builds a congestion feature library. The congestion risk level is generated based on the difference in the weight change rate between the feed bin and the discharge bin.

[0009] The graded dredging execution module is equipped with differentiated dredging strategies, which are activated according to the congestion risk level.

[0010] By adopting the above technical solution, the problems of passive response to blockage and low positioning efficiency of traditional systems are solved. Through real-time monitoring of multiple pipeline parameters by a distributed sensor group, combined with the three-dimensional mapping of the physical state by the digital twin model, accurate marking of the blockage location and automatic generation of risk levels are achieved, which triggers a graded dredging strategy and significantly reduces downtime and troubleshooting time.

[0011] The present invention is further configured to include: a congestion warning module, which is connected to the congestion analysis module and is used to calculate the dynamic similarity between the data of the real-time distributed sensor group and the historical congestion curve in the congestion feature library. When the similarity exceeds a preset matching threshold, a primary warning is generated. If the primary warning lasts for a preset time, it is upgraded to an advanced warning to trigger a graded dredging execution module.

[0012] By adopting the above technical solution, the false alarm defects caused by instantaneous fluctuations are addressed. Through dynamic similarity calculation and preset duration verification, the primary warning is upgraded to an advanced warning, ensuring that the warning signal is only triggered when the abnormality persists, thereby reducing the error operation rate.

[0013] The present invention is further configured as follows: the conditions for upgrading the blockage warning module to a high-level warning also include: the real-time air pressure fluctuation amplitude exceeds the standard deviation of the historical curve; the material flux continues to decrease at a rate greater than a preset risk threshold.

[0014] By adopting the above technical solution, the problem of single indicator missing high-risk blockage is solved. New complex conditions such as air pressure fluctuation amplitude exceeding the standard deviation and continuous flux decline rate exceeding the threshold are added to accelerate the response to rapidly worsening blockage and avoid the spread of blockage.

[0015] The present invention is further configured as follows: the feeding system further comprises:

[0016] A blockage simulation device includes a simulated blocker disposed inside a pipeline and a moving assembly for driving the simulated blocker to move, wherein the simulated blocker is used to create blockage conditions of different blockage assessment levels within the pipeline, and the moving assembly is used to drive the simulated blocker to move along the inner wall of the pipeline;

[0017] The blockage simulation module is used to control the blockage simulation device, drive the blockage simulation device to several positions in the pipeline, and drive the blockage simulation device to form several degrees of blockage assessment levels. The multi-parameter change curve generated by the blockage simulation module during simulated blockage is recorded in the blockage feature library through the blockage analysis module.

[0018] By adopting the above technical solution, the blockage feature library is dynamically optimized: different levels of blockage are generated at any position in the pipeline through a movable simulated plugger, and multi-parameter change curves are collected in real time and entered into the feature library, thereby improving the generalization ability and accuracy of the blockage prediction model.

[0019] The present invention is further configured as follows: the simulated plug includes a plugging elastic sheet, inner magnetic rings arranged at both ends of the plugging elastic sheet, and a movable outer magnetic ring magnetically matched with the inner magnetic ring; the inner magnetic ring and the movable outer magnetic ring are respectively arranged inside and outside the pneumatic conveying pipeline, and two inner magnetic rings and two movable outer magnetic rings are each provided; the movable outer magnetic ring is equipped with a movable component that drives it to move along the pneumatic conveying pipeline.

[0020] When the two inner magnetic rings approach each other, the blocking elastic sheet is driven to bend and deform. The deformation degree of the blocking elastic sheet is controlled according to the distance between the two inner magnetic rings. The different deformation degrees of the blocking elastic sheet are used to simulate different blockage degrees of the pneumatic conveying pipeline.

[0021] By adopting the above technical solution, non-contact precise blockage simulation is achieved, and the magnetic ring linkage design drives the deformation of the internal elastic sheet to avoid pipeline wear caused by mechanical contact; by adjusting the spacing between the internal magnetic rings, the bending degree of the elastic sheet is controlled, and different degrees of blockage are accurately simulated to ensure the authenticity of the feature library data.

[0022] The present invention is further configured such that the congestion warning module further includes a weight analysis unit for real-time monitoring of parameters and calculation of a comprehensive risk value, which is calculated as follows:

[0023] The dynamic similarity between the real-time sensor data and the historical congestion characteristics is multiplied by the first weight factor, plus the ratio of the number of times the air pressure change rate exceeds the set mutation threshold within the monitoring window to the length of the dynamic monitoring window, multiplied by the second weight factor, and the ratio of the temperature anomaly duration to the current transport task's running time, multiplied by the third weight factor;

[0024] Wherein, the first weighting factor is greater than the second weighting factor, the second weighting factor is greater than the third weighting factor, and the sum of the weights of the three is 1;

[0025] When the comprehensive risk value exceeds the warning threshold dynamically determined based on pipeline structural parameters and material characteristics, an advanced warning is triggered.

[0026] By adopting the above technical solution, the problem that the fixed risk threshold cannot adapt to different pipelines and materials is solved.

[0027] The present invention is further configured as follows: the dynamic determination rule of the warning threshold is:

[0028] Based on the preset benchmark value, periodic adjustments are made in combination with the ratio between the actual length of the pipeline and the maximum length of the system, and linear corrections are made based on the difference between the material fluidity index and the maximum and minimum fluidity values;

[0029] Among them, the influence of pipeline length is reflected by the periodic function, and the influence of material fluidity is reflected by the normalized difference.

[0030] By adopting the above technical solution, the problem that the fixed risk threshold cannot adapt to different pipelines and materials is solved.

[0031] The present invention is further configured such that the blockage risk levels include:

[0032] If the difference in weight change rate between the feed bin and the discharge bin is less than the preset value α, it is judged as a mild blockage level;

[0033] If the preset value α≤the difference in weight change rate between the feed bin and the discharge bin < the preset value β, it is determined to be a moderate blockage level;

[0034] If the difference in weight change rate between the feed bin and the discharge bin is ≥ the preset value β, it is determined to be a severe blockage level.

[0035] By adopting the above technical solution, the randomness of empirical judgment of the degree of blockage is eliminated, and a precise match between the dredging intensity and the degree of blockage is achieved.

[0036] The present invention is further configured such that when the congestion warning module issues a high-level warning, a corresponding differentiated unblocking strategy is triggered according to the congestion risk level. The differentiated unblocking strategy includes:

[0037] The mild unblocking strategy corresponds to the mild blockage level and is equipped with a vibrator that is triggered at the blockage location to vibrate the target section of the pneumatic conveying pipeline at a preset frequency.

[0038] The medium unblocking strategy corresponds to the medium blockage level. The pipeline is switched to inject reverse airflow into the pneumatic conveying pipeline. The air pressure of the reverse airflow is not less than the forward conveying pressure.

[0039] The severe dredging strategy corresponds to the severe blockage level and is equipped with a rotating dredging mechanism that extends into the blocked position, or drives the simulated blocker to move through a moving component to dredge the triggered blockage position.

[0040] By adopting the above technical solutions, a three-level progressive dredging method is designed to address the problem that a single dredging strategy is ineffective or damages the pipeline: mild vibration to loosen particles, moderate reverse airflow purge with air pressure ≥ positive pressure, and heavy telescopic rotation mechanism mechanical dredging, which protects the integrity of the pipeline while efficiently clearing blockages.

[0041] The present invention is further configured such that the feeding system also includes an adaptive cleaning module, which switches the pipeline to inject reverse airflow into the pneumatic conveying pipeline to perform reverse cleaning when the heavy dredging strategy is triggered or the cumulative conveying volume exceeds a preset quality threshold.

[0042] By adopting the above technical solution, secondary blockage caused by residues after dredging can be prevented. When heavy dredging is completed or the cumulative delivery volume exceeds the quality threshold, the gas path is automatically switched to perform reverse airflow cleaning to remove particles attached to the pipe wall, reduce the cleaning frequency and reduce energy consumption.

[0043] The present invention is further configured to include a transport efficiency evaluation module, which is connected to the congestion analysis module and generates optimization suggestions based on historical congestion locations and frequencies, including:

[0044] When the cumulative number of blockages at the same location exceeds the preset number of blockages, a suggestion is output to increase the curvature radius of the pipe bend to 1-10 times the pipe diameter;

[0045] When the wall vibration amplitude of a pneumatic conveying pipeline exceeds a preset safety threshold, it is marked as a damaged pipe section and a replacement recommendation is triggered;

[0046] It is recommended to add vibrators and temperature sensors at locations where blockages frequently occur, and the recommendations are displayed in the 3D modeling of the pneumatic conveying pipeline.

[0047] By adopting the above technical solutions, closed-loop optimization of pipeline design defects is carried out, and recommendations are given to increase the curvature radius to 1-10 times the pipe diameter for high-frequency blockage points. Pipe sections with excessive vibration (amplitude > 0.5mm) are marked for replacement, and pre-set sensor solutions are visualized in the digital twin model to reduce the probability of blockage from the root.

[0048] The present invention is further configured to include an operating parameter constraint module for interrupting the operation and requesting operation confirmation when the wind pressure or flow rate set before the pneumatic conveying pipe is activated exceeds a preset safety range.

[0049] By adopting the above technical solution, the risk of human error is blocked. When the wind pressure or flow exceeds the safe range (based on the pipeline pressure limit setting), the system is forced to interrupt and request confirmation, avoiding blockage or leakage accidents caused by incorrect parameter settings.

[0050] Due to the adoption of the above technical solution, the present invention has significant technical effects: the application provides a pneumatic conveying pipeline blockage warning and graded dredging system based on digital twins, which collects multi-dimensional pipeline data in real time through a distributed sensor group, dynamically maps the pipeline status in combination with a digital twin model, and performs risk classification and precise positioning based on a blockage feature library, while matching differentiated dredging strategies to achieve early warning and automated graded dredging, which has the advantage of significantly improving pipeline operation safety and maintenance efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a schematic diagram of the structure of part of the device in the embodiment;

[0052] Figure 2 This is a structural diagram of the automatic feeding system in the embodiment;

[0053] Figure 3 Schematic diagram of the structure of the simulated plug in the embodiment.

[0054] The names of the parts indicated by the numerical labels in the above drawings are as follows: 1. Feed bin; 2. Pneumatic conveying pipeline; 3. Discharge bin; 4. Distributed sensor group; 5. Simulated plug; 51. Blocking elastic sheet; 52. Inner magnetic ring; 53. Moving outer magnetic ring; 6. Moving component. DETAILED DESCRIPTION

[0055] The present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0056] Example:

[0057] Powder extrusion feeding systems often use dense-phase pneumatic conveying technology, which uses low-speed airflow to push high-concentration powder to form a plug or fluidized bed for conveying. Such systems are prone to pneumatic conveying pipeline blockage due to pipeline design defects, changes in material properties or improper operating parameters. Existing blockage detection methods usually rely on pressure sensors to monitor the upstream and downstream pressure differences to locate the blockage point, but this method has problems such as low positioning accuracy and response delay. When a blockage occurs, manual intervention is required to perform reverse airflow purge or disassembly and dredge. This leads to long production line downtime and significantly increased maintenance costs.

[0058] The inventors observed that traditional blockage detection relies only on single-point pressure parameters and cannot fully reflect the material flow status in the pneumatic conveying pipeline. By analyzing the correlation changes in air pressure, flow and vibration signals before blockage occurs, they found that multi-parameter fusion monitoring can identify blockage risks earlier. Combined with digital twin technology, three-dimensional dynamic modeling of the pneumatic conveying pipeline status can track material flow anomalies in real time. In addition, a graded dredging mechanism is designed for different degrees of blockage to avoid energy waste or equipment damage caused by a single dredging method.

[0059] An automatic loading system includes an automatic loading system consisting of a feed bin 1, a pneumatic conveying pipeline 2, an air source device and a discharge bin 3. A distributed sensor group 4 is introduced to monitor the flux, pressure, temperature and vibration spectrum of the pneumatic conveying pipeline in real time. A digital twin model is constructed to map the status of the pneumatic conveying pipeline. The blockage location is marked and the risk level is generated through the blockage analysis module. Finally, a differentiated dredging strategy is initiated by the graded dredging execution module.

[0060] The distributed sensor group 4 refers to a combination of flow sensors, pressure sensors, temperature sensors and vibration acceleration sensors arranged at preset points along the pneumatic conveying pipeline. Specifically, it can be implemented by electromagnetic flowmeters, piezoelectric pressure transmitters, thermocouple sensors and MEMS accelerometers, and is used to collect dynamic parameters at different positions of the pneumatic conveying pipeline. The digital twin model construction module refers to the establishment of a pneumatic conveying pipeline geometric model through three-dimensional modeling software, and the integration of sensor data to achieve virtual-real mapping. Specifically, the finite element analysis algorithm can be used to simulate the material flow state. The blockage analysis module refers to the identification of abnormal patterns in multi-parameter change curves based on machine learning algorithms. Specifically, the time series analysis method can be used to extract blockage precursor features. The graded dredging execution module refers to a composite control system integrating vibrators, reverse airflow valves and mechanical dredging devices. Specifically, strategy switching can be achieved through PLC programming.

[0061] During the operation of the system, the distributed sensor group 4 continuously collects the flow, pressure, temperature and vibration signals of each section of the pneumatic conveying pipeline, and transmits the data to the digital twin model construction module to generate a three-dimensional visualization model. When the material flow rate of a certain section of the pneumatic conveying pipeline decreases, the blockage analysis module judges the blockage level based on the difference in weight change rate between the feed bin 1 and the discharge bin 3. For example, when the weight difference exceeds the preset threshold, a moderate blockage judgment is triggered. The graded dredging execution module selects the corresponding strategy based on the judgment result. For example, for mild blockage, a local vibrator is started to loosen the material, and for severe blockage, a rotary dredging mechanism is activated to physically remove the blockage. The entire process achieves closed-loop control through real-time data interaction and model update.

[0062] Traditional methods rely solely on single-point pressure detection to determine blockages, are unable to distinguish the degree of blockage, and have limited positioning accuracy. This solution uses multi-dimensional sensor fusion monitoring combined with a digital twin model to achieve panoramic visualization of the pneumatic conveying pipeline status, accurately identifying the location and development trend of the blockage. At the same time, the graded dredging strategy avoids the energy waste or excessive equipment wear that may be caused by traditional single dredging methods. For example, in the event of minor blockages, low-energy vibration is preferred over full-pressure reverse airflow flushing.

[0063] It achieves early warning and precise positioning of blockages in pneumatic conveying pipelines, shortens the response time for blockage processing, and multi-parameter collaborative analysis effectively reduces the misjudgment rate. The digital twin model provides operators with an intuitive status monitoring interface. The graded dredging mechanism optimizes energy consumption control while ensuring the dredging effect, extending the service life of key components. It is especially suitable for the continuous and stable operation requirements in high solid-gas ratio dense phase conveying scenarios.

[0064] The congestion warning module is connected to the congestion analysis module and is used to calculate the dynamic similarity between the data of the real-time distributed sensor group 4 and the historical congestion curve in the congestion feature library. When the similarity exceeds the preset matching threshold, a primary warning is generated. If the primary warning lasts for a preset time, it is upgraded to an advanced warning to trigger the graded dredging execution module.

[0065] The congestion warning module refers to a unit used to identify congestion risks and issue warnings. It can be implemented through an embedded processor and a data acquisition circuit, and is used to analyze the matching degree between sensor data and historical characteristics in real time. Dynamic similarity refers to the matching degree between real-time sensor data and historical congestion curves. It can be calculated using a dynamic time warping algorithm, and the similarity is judged by comparing the morphological differences in parameter change trends. Primary warning refers to a preliminary risk warning signal, which can be implemented through an audible and visual alarm or a pop-up window on the operating interface to remind operators to pay attention to potential congestion risks. Advanced warning refers to an emergency signal that requires immediate intervention. It can trigger the actuator action through a relay to automatically start the dredging equipment.

[0066] The congestion warning module continuously receives the flux, pressure, temperature and vibration spectrum data from the distributed sensor group 4, and dynamically matches the real-time collected multi-parameter change curve with the historical congestion curve stored in the congestion feature library. When the similarity calculation result between the real-time curve and the historical congestion curve exceeds the set threshold, the system generates a primary warning signal and starts timing. If the primary warning is not lifted within the set monitoring window and the duration reaches the preset value, such as one-fifth of a single conveying cycle, the system automatically upgrades the warning level to a high-level warning, and links the hierarchical dredging execution module to start the corresponding dredging strategy.

[0067] Existing technologies rely on locating the blockage point through sudden pressure changes after a blockage occurs, requiring manual intervention and resulting in extended downtime. This solution uses dynamic similarity analysis to identify blockage trends in advance. Combined with a hierarchical mechanism of primary and advanced warnings, it can trigger an automated response at the early stages of blockage formation, shortening the fault handling cycle.

[0068] This application can identify the evolution trend of blockage risks in real time during material transportation, implement risk classification management through a multi-level early warning mechanism, avoid false triggering or response delays caused by single threshold judgment, and improve the timeliness and accuracy of prevention and treatment of pneumatic conveying pipeline blockages.

[0069] The conditions for upgrading to an advanced warning in the blockage warning module also include the real-time air pressure fluctuation amplitude exceeding the standard deviation of the historical curve and the continuous decline rate of material flux exceeding the preset risk threshold.

[0070] The real-time air pressure fluctuation amplitude refers to the degree of discreteness of the air pressure value within the current monitoring period. Specifically, this can be achieved by statistically analyzing the historical operating data using a standard deviation calculation algorithm, and is used to quantify the degree of abnormal air pressure fluctuation. The continuous decline rate of material flux refers to the reduction in the amount of material passing through per unit time. Specifically, this can be achieved by calculating the first-order derivative of the real-time data collected by the flow sensor, and is used to reflect the accelerated state of obstructed material flow. The standard deviation of the historical curve refers to a benchmark model of pipeline operating parameters established under normal operating conditions. Specifically, this can be achieved by performing variance calculation on pressure data samples within a sliding time window, and is used to establish a basis for dynamic threshold judgment.

[0071] When the pressure fluctuation amplitude in the pneumatic conveying pipeline is monitored to exceed the standard deviation range formed by the historical operating data, it indicates that abnormal pressure accumulation occurs in the local area. At the same time, if the material flux shows a continuous downward trend and the decline rate exceeds the preset risk threshold, it can be determined that the material flow obstruction is rapidly worsening. When the above two conditions are met, the early warning module will upgrade the primary warning to an advanced warning, triggering the graded dredging execution module to take active intervention measures.

[0072] Traditional blockage detection relies solely on the upstream and downstream pressure difference of the pressure sensor for single-point judgment, which is prone to misjudgment due to fluctuations in operating conditions. This technical solution introduces the standard deviation of historical air pressure fluctuations as a dynamic benchmark, combined with the dual criteria of material flux change rate, to achieve dynamic tracking and trend prediction of the blockage formation process.

[0073] This application can identify potential blockage risks earlier and trigger the unblocking mechanism before the material flow is completely stagnant, avoiding the problem of complete blockage of the pneumatic conveying pipeline caused by detection lag in traditional methods. At the same time, it reduces the probability of false triggering of a single sensor through a multi-parameter collaborative judgment mechanism, reducing the number of unnecessary shutdowns for maintenance.

[0074] The congestion warning module also includes a weight analysis unit, which is used to monitor parameters in real time and calculate the comprehensive risk value R, which is calculated as follows:

[0075] The dynamic similarity between the real-time sensor data and the historical congestion characteristics is multiplied by the first weight factor, plus the ratio of the number of times the air pressure change rate exceeds the set mutation threshold within the monitoring window to the length of the dynamic monitoring window, multiplied by the second weight factor, and the ratio of the temperature anomaly duration to the current transport task's running time, multiplied by the third weight factor;

[0076] That is, its expression is R=k1S+k2(Nmutation / tmonitoring)+k3*(tabnormal / ttotal), where S is the dynamic similarity between the real-time calculated data of the distributed sensor group and the historical congestion curve and its value range is 0 to 1, tmonitoring is the duration of the dynamic monitoring window and its value range is one-tenth to one-half of the total duration of a single material conveying task, Nmutation is the number of times the air pressure change rate exceeds the set mutation threshold within the monitoring window, tabnormal is the cumulative time that the temperature sensor reading continuously exceeds the material tolerance temperature threshold, ttotal is the running time of the current conveying task, the first weight factor k1 is greater than the second weight factor k2, the second weight factor k2 is greater than the third weight factor k3, and the sum of the three weights is 1. When the comprehensive risk value exceeds the warning threshold dynamically determined according to the structural parameters of the pneumatic conveying pipeline and the material characteristics, an advanced warning is triggered.

[0077] The dynamic monitoring window duration refers to the data collection time range that is dynamically adjusted according to the transportation task cycle. It can be achieved by using a time window sliding algorithm. By dynamically matching the changing trend of the transportation cycle, the monitoring sensitivity is improved. Dynamic similarity refers to the degree of matching between real-time sensor data and historical congestion curves in the time series. It can be calculated using a dynamic time warping algorithm. Accurate comparison is achieved by eliminating the time axis offset difference. The weight factor refers to the contribution ratio of each parameter to the comprehensive risk value. It can be assigned using the hierarchical analysis method. The decision weight of the core indicators is strengthened by distinguishing the priorities of different parameters. Dynamic adjustment of the R threshold refers to establishing a functional relationship between the pipeline length and the material fluidity index. It can be calculated using the linear interpolation method. The adaptability of the warning trigger conditions is improved by associating physical parameters with material properties.

[0078] During the operation of the transportation task, the weight analysis unit continuously collects the flow, pressure, temperature and vibration data of the distributed sensor group 4, and calculates the similarity S between the current data and the historical blockage curve through the dynamic time warping algorithm. At the same time, the number of air pressure mutations N mutations in the monitoring window is counted, and the temperature exceeding the limit time t anomaly is accumulated. The current transportation task duration t total is used as the denominator, and the N mutations and t anomalies are normalized respectively. The three parameters are weighted and summed according to the preset weight factors k1, k2, and k3 to generate a comprehensive risk value R. When the value exceeds the R threshold calculated based on the pneumatic conveying pipeline length L and the material fluidity index μ, the system automatically upgrades the warning level. For example, for a pneumatic conveying pipeline with a length of 10 meters and a fluidity index of 0.8, the R threshold can be set to 0.75; for a pipeline with a length of 20 meters and a fluidity index of 0.5, the R threshold can be adjusted to 0.68.

[0079] Traditional congestion warning methods only make threshold judgments based on a single pressure parameter and are unable to distinguish between occasional fluctuations and real congestion risks. This solution builds a comprehensive risk assessment model by integrating multi-dimensional sensor data. Multiple related parameters such as time series similarity analysis, air pressure mutation frequency statistics, and temperature anomaly duration are introduced into the calculation. Combined with the weight distribution mechanism, a composite judgment criterion is formed. For example, when the material fluidity decreases and the friction coefficient of the pneumatic conveying pipeline increases, the system reduces the probability of misjudgment by dynamically adjusting the R threshold, thereby improving the warning accuracy compared to the fixed threshold method.

[0080] This application realizes a multi-dimensional quantitative assessment of blockage risks, effectively identifies early signs of blockage through dynamic similarity matching, enhances the response speed to sudden blockages by combining the frequency statistics of air pressure mutations, and can detect the risk of chronic blockages caused by material adhesion in advance through cumulative monitoring of temperature anomaly time. The differentiated setting of weight factors strengthens the dominant role of core parameters and avoids interference from secondary parameters in judgment. The dynamic adjustment mechanism of the R threshold enables the warning trigger conditions to adapt to different pneumatic conveying pipeline lengths and material characteristics, reducing the false alarm rate while ensuring timely warning of high-risk working conditions, thereby shortening the dredging response time and reducing production interruption losses.

[0081] The dynamic determination rule for the warning threshold is as follows: based on a preset baseline value, periodic adjustment is performed in combination with the ratio between the actual length of the pneumatic conveying pipeline and the maximum length of the system, and linear correction is performed based on the difference between the material fluidity index and the maximum and minimum fluidity values. The influence of the pneumatic conveying pipeline length is reflected through a periodic function, and the influence of material fluidity is reflected through the normalized difference, as follows:

[0082] The threshold Rthreshold of the comprehensive risk value R in the blockage warning module is dynamically adjusted according to the physical length of the pneumatic conveying pipeline and the material fluidity index. The dynamic adjustment rule is Rthreshold = 0.5×(L / Lmax)+0.5×(μ−μmin) / (μmax−μmin), where L is the physical length of the pneumatic conveying pipeline, Lmax is the maximum pneumatic conveying pipeline length of the system, μ is the material fluidity index, μmax is the maximum material fluidity, and μmin is the minimum material fluidity.

[0083] The physical length of the pneumatic conveying pipeline refers to the actual extension distance of the conveying pipeline from the feed bin 1 to the discharge bin 3, which can be measured by a laser rangefinder or a tape measure. The introduction of the physical length of the pipeline enables the threshold adjustment to adapt to the risk changes under different conveying distances. The maximum pipeline length of the system refers to the maximum length of the pipeline allowed to be installed in the system design, which can be determined according to the equipment technical manual or engineering specifications. The setting of the maximum pipeline length of the system is used to normalize the physical length of the pipeline. The material fluidity index refers to the parameter that characterizes the flow properties of powder or granular materials during pneumatic conveying, which can be measured by an angle of repose tester or a shear tester. The introduction of the material fluidity index enables the threshold adjustment to reflect the impact of the material's own characteristics on the blockage risk. The maximum and minimum values ​​of material fluidity refer to the material fluidity limit values ​​allowed by the system for conveying, which can be determined based on historical conveying data or a material property database. The setting of the maximum and minimum values ​​of material fluidity is used to normalize the material fluidity index.

[0084] During the blockage warning process, the threshold of the comprehensive risk value R is not fixed, but is dynamically calculated based on the actual length of the current conveying pipeline and the fluidity characteristics of the conveyed material. When the pipeline length increases, the pressure attenuation and material retention risk increase due to the extension of the conveying path, and the threshold will increase accordingly to avoid false alarms; when the material fluidity decreases, the material is prone to form deposits or agglomerations in the pipeline, so the threshold will decrease accordingly to improve the warning sensitivity. By incorporating the two key parameters of pipeline length and material fluidity into the threshold calculation model, the risk judgment benchmark can automatically adapt to different working conditions.

[0085] Traditional blockage warning systems typically use fixed thresholds for risk assessment, which cannot effectively address the differences in operating conditions caused by different pipeline lengths and material properties. For example, when transporting low-fluidity materials in long pipelines, fixed thresholds may cause the warning to be triggered prematurely, resulting in unnecessary unblocking operations. When transporting highly fluid materials in short pipelines, fixed thresholds may cause warning delays and miss the optimal unblocking opportunity. This solution uses a dynamic adjustment mechanism to enable the risk assessment benchmark to match the current transportation conditions in real time, significantly improving the adaptability and accuracy of the warning system.

[0086] Intelligent adjustment of the blockage warning threshold is achieved, which effectively solves the problem of the traditional fixed threshold method that it is difficult to balance the warning sensitivity and false alarm rate under different pipeline lengths and material characteristics. This solution can automatically optimize the warning trigger conditions according to the actual working conditions, avoiding frequent false alarms under low-risk conditions and ensuring timely issuance of warning signals under high-risk conditions, thereby improving the operational reliability and maintenance efficiency of the entire pneumatic conveying system.

[0087] The method for dividing the blockage risk level includes: when the difference in the weight change rate between the feed silo 1 and the discharge silo 3 is less than the preset value α, it is judged as a mild blockage level; when the difference is in the range of α to β, it is judged as a moderate blockage level; when the difference exceeds β, it is judged as a severe blockage level.

[0088] The difference in weight change rate between feed silo 1 and discharge silo 3 refers to the difference in material weight change per unit time in the two silos. This can be achieved by collecting data in real time through a high-precision weighing sensor and calculating the instantaneous weight change rate. This difference directly reflects the degree of obstruction to material flow in the pipeline. The preset values ​​α and β refer to the classification thresholds set according to the pipeline transportation capacity, material characteristics and historical operation data. For example, α can be set to 20% of the theoretical flux, and β can be set to 50% of the theoretical flux. Accurate classification under different working conditions can be achieved by dynamically adjusting the thresholds.

[0089] During the material conveying process, the weighing sensor continuously monitors the material reduction rate of the feed silo 1 and the material increase rate of the discharge silo 3. When the difference between the two silo rates is continuously lower than α, it indicates that there is only a slight delay in the material flow, and the system then determines it as a mild blockage. When the difference rises to between α and β, it indicates that the degree of obstruction of material flow has intensified, and the system upgrades to a moderate blockage level. When the difference exceeds β, it is judged to be a severe blockage state where the material flow is close to stagnation. This grading logic is executed in real time by the embedded controller and forms a closed-loop control with the dredging actuator.

[0090] Traditional blockage detection relies solely on pressure changes to determine if a blockage has occurred. This approach cannot quantify the extent of the blockage and carries the risk of misjudgment. This solution establishes a multi-level blockage determination system through dynamic monitoring of the difference in weight change rates between the two bins, enabling accurate identification of blockage conditions of varying severity.

[0091] It achieves refined identification of blockage status and provides a reliable basis for the selection of subsequent unblocking strategies. Compared with the traditional single processing method, the graded judgment mechanism can avoid excessive energy consumption in low-risk blockages, while ensuring that strong unblocking measures are activated in time in high-risk blockages, effectively balancing production safety and operational economy.

[0092] When the congestion warning module issues a high-level warning, the corresponding differentiated unblocking strategy is triggered according to the congestion risk level. The differentiated unblocking strategy includes a mild unblocking strategy, corresponding to the mild congestion level, equipped with a vibrator, triggering the vibrator at the blocked position to vibrate the target pipe section of the pneumatic conveying pipeline 2 at a preset frequency; a moderate unblocking strategy, corresponding to the moderate congestion level, switching the pipeline to inject reverse airflow into the pneumatic conveying pipeline 2, and the air pressure of the reverse airflow is not less than the forward conveying pressure; a severe unblocking strategy, corresponding to the severe congestion level, equipped with a rotating unblocking mechanism extending into the blocked position, or driving the simulated blocker 5 to move through a moving component to unblock the triggered blockage position.

[0093] The blockage risk level refers to the blockage severity classification based on the difference in weight change rate between the feed bin 1 and the discharge bin 3. The blockage level is judged by quantifying the degree of imbalance in the material flow. Specifically, it can be implemented by a preset threshold interval matching algorithm to accurately match different dredging intensities. A vibrator refers to a mechanical vibration device installed on the outer wall of the pipeline. Specifically, it can be implemented by an electromagnetic or pneumatic vibrator. The blocked material is loosened by generating high-frequency vibration. Reverse airflow refers to an airflow opposite to the material conveying direction. Specifically, it can be achieved by switching the direction of the air source pipeline and adjusting the air pressure. It is used to destroy the material accumulation structure in the blocked area. A rotary dredging mechanism refers to a mechanical device with a rotary cutting function. Specifically, it can be implemented by a pneumatic actuator with a built-in spiral cutter head. Stubborn blockages are removed by physical cutting. The simulated blocker 5 can also be driven by a moving component to perform reciprocating dredging actions, which is a further utilization of the blockage simulation device.

[0094] When the system detects a high-level warning signal, it first determines the blockage level based on the difference in weight change rate between the feed bin 1 and the discharge bin 3. For mild blockage, the vibrator corresponding to the blockage position is activated to generate high-frequency vibration to cause periodic deformation of the pipe wall, prompting the material to resume flow; in case of moderate blockage, the direction of the air source is switched and a reverse airflow with a pressure not lower than the forward conveying pressure is generated, and the impact force of the airflow is used to destroy the static balance of the blocked material; in case of severe blockage, the rotary dredging mechanism is started to extend into the blocked section, and the rotating cutter head is used to cut and compact the material layer to form a dredging channel.

[0095] The existing technology only relies on a single dredging method and requires manual judgment of the degree of blockage. This solution establishes a multi-level dredging strategy library to automatically match the optimal dredging intensity according to real-time data, avoiding energy waste caused by excessive dredging or secondary blockage caused by insufficient dredging. The traditional method requires manual operation of the dredging device after shutdown, while this solution achieves rapid dredging through hierarchical actuators without stopping the machine.

[0096] It effectively solves the problem of low unblocking efficiency caused by inaccurate judgment of blockage level during dense phase conveying. Through the multi-stage synergistic effect of vibration, reverse airflow flushing and mechanical cutting, it significantly shortens the processing time of blockages of different severity, while reducing the frequency of manual intervention and the risk of equipment damage.

[0097] The automatic feeding system also includes a blockage simulation device and a blockage simulation module. The blockage simulation device includes a simulated blocker 5 arranged on the inside of the pipeline and a moving component for driving the simulated blocker 5 to move. The simulated blocker 5 is used to form blockage conditions with different blockage assessment levels in the pipeline, and the moving component is used to drive the simulated blocker 5 to move along the inner wall of the pipeline; the blockage simulation module is used to control the blockage simulation device, drive the blockage simulation device to be at several positions in the pipeline, and drive the blockage simulation device to form several degrees of blockage assessment levels. The multi-parameter change curve generated by the blockage simulation module during simulated blockage is recorded in the blockage feature library through the blockage analysis module.

[0098] Among them, the congestion simulation device refers to a device that actively creates a controllable congestion scene, which can be specifically achieved by using a mechanical component with a deformable structure in conjunction with a magnetic drive device.

[0099] The simulated blocker in this embodiment includes a blocking elastic sheet 51, an inner magnetic ring 52 arranged at both ends of the blocking elastic sheet 51, and a movable outer magnetic ring 53 magnetically coupled with the inner magnetic ring 52. The inner magnetic ring 52 and the movable outer magnetic ring 53 are respectively arranged inside and outside the pneumatic conveying pipe 2, and there are two inner magnetic rings 52 and two movable outer magnetic rings 53. The movable outer magnetic ring 53 is equipped with a moving component for driving it to move along the pneumatic conveying pipe 2. The moving component in this embodiment is specifically a driving motor with a driving wheel and is fixedly connected to the movable outer magnetic ring 53. A plurality of moving components are arranged in an annular array on the outer ring of the movable outer magnetic ring 53. When the two inner magnetic rings 52 approach each other, the blocking elastic sheet 51 is driven to bend and deform. The deformation degree of the blocking elastic sheet 51 is controlled according to the distance between the two inner magnetic rings 52. Different deformation degrees of the blocking elastic sheet 51 are used to simulate different blockage degrees of the pneumatic conveying pipe 2.

[0100] The blocking elastic sheet 51 is a sheet-like structure with elastic deformation capabilities. Specifically, it can be made of polyurethane or silicone materials. It produces blocking effects of varying cross-sectional areas through bending deformation, and is used to form a controllable local blockage inside the pipeline. The inner magnetic ring 52 is an annular magnetic component fixed to the end of the elastic sheet. Specifically, it can be a neodymium iron boron permanent magnet. It forms a magnetic attraction with the external movable outer magnetic ring 53, and is used to convert the mechanical movement of the external movable component into deformation control of the elastic sheet. The movable outer magnetic ring 53 is a magnetic drive component disposed on the outer wall of the pipeline. Specifically, it can be an electromagnet or a permanent magnet. It changes its relative position with the inner magnetic ring 52 by moving along the axial direction of the pipeline, and is used to precisely adjust the spacing between the two inner magnetic rings 52. The moving assembly is an actuator that drives the movement of the outer magnetic ring. Specifically, it can be a linear motor or a ball screw device. It is used to achieve the positioning and continuous displacement of the outer magnetic ring on the outer wall of the pipeline.

[0101] The inner magnetic ring 52 and the blocking elastic sheet 51 form an integrated structure and are installed on the inner wall of the pipe. The movable outer magnetic ring 53 is driven axially along the outer wall of the pipe by a movable assembly. When the two outer magnetic rings are driven closer by the movable assembly, their magnetic fields cause the corresponding two inner magnetic rings 52 to attract each other, causing the blocking elastic sheet 51 to bend. As the spacing between the inner magnetic rings 52 decreases, the degree of bending of the elastic sheet increases, forming blockage areas of varying cross-sectional areas within the pipe. By controlling the displacement of the movable assembly, the spacing between the inner magnetic rings 52 can be precisely adjusted, thereby achieving a continuous simulation state from slight blockage to complete blockage. This structure allows adjustable blockage to be generated at any location in the pipe, without intruding into the pipe interior.

[0102] Traditional blockage simulation devices usually use fixed throttle valves or plug-in baffles, which cannot achieve variable adjustment of the pipeline's axial position, and the adjustment of the blockage degree requires manual intervention and component replacement. This solution realizes non-contact blockage simulation through the synergy of magnetic coupling and external mobile components. It can dynamically change the blockage location and severity without interrupting the transportation process, significantly improving the efficiency of collecting blockage feature data and the accuracy of model training.

[0103] Through the above technical solution, this application achieves controllable and repeatable simulation of blockage in pneumatic conveying pipelines. It can accurately generate blockage conditions at different locations and levels, providing multi-dimensional data support for the construction of a blockage feature library. The magnetic drive method avoids the damage to pipeline integrity caused by traditional mechanical structures. The precise positioning capability of the moving component ensures the spatial resolution of the blockage simulation, and the continuous deformation characteristics of the elastic sheet enable stepless adjustment of the blockage level.

[0104] The function of the mobile component is to accurately position the simulated plug 5 to any position in the pipeline, covering the characteristics of different pipe sections. The blockage simulation module controls the combination of blockage position and blockage level through a preset program to generate multi-dimensional data samples to expand the coverage of the blockage feature library.

[0105] During the initial system operation or maintenance phase, the blockage simulation module drives the mobile assembly to move the simulated blocker 5 to the target position. By adjusting the spacing between the inner magnetic rings 52, the blockage elastic sheet 51 undergoes a preset deformation, forming a blockage of varying cross-sectional areas. At this point, the distributed sensor group 4 collects flux, pressure, temperature, and vibration spectrum data under this simulated blockage state. After processing by the blockage analysis module, a corresponding multi-parameter variation curve is generated and stored in the blockage feature library. For example, when the simulated blocker 5 moves to a bend, it can simulate the increased local resistance caused by material accumulation and record the pressure fluctuation characteristics in this scenario. When the deformation of the blockage elastic sheet 51 reaches a severe blockage, it can collect abnormal peak data of the high-frequency components in the vibration spectrum.

[0106] Existing technologies rely on passive data collection after actual blockages occur, failing to cover a wide range of blockage scenarios. This results in a single feature library sample and limited warning accuracy. This solution proactively simulates blockage conditions at different locations and levels, systematically capturing blockage response data from all areas of the pipeline. This particularly captures the differentiated characteristics of blockage-prone areas like elbows and reducers, providing more comprehensive sample coverage for the blockage feature library and improving the reliability of dynamic similarity calculations.

[0107] This application can proactively generate diverse congestion training data during system debugging or idle phases, resolving the misjudgment or missed detection issues inherent in existing technologies due to insufficient samples. For example, by simulating the congestion patterns corresponding to different material properties, the system's adaptability to materials with varying fluidity can be enhanced. By recording the dynamic parameter changes during the movement of the simulated clog 5, a mapping relationship between the congestion location and sensor data can be established, shortening the time required to locate an actual congestion.

[0108] An automatic feeding system includes a feed bin 1, a pneumatic conveying pipeline 2, an air source device and a discharge bin 3, and also includes an adaptive cleaning module. When the heavy unblocking strategy is triggered or the cumulative conveying volume exceeds a preset quality threshold, the pipeline is switched to inject reverse airflow into the pneumatic conveying pipeline 2 to perform reverse cleaning.

[0109] The adaptive cleaning module refers to a functional unit that automatically triggers the cleaning action according to the system operation status. It can be implemented by linking the solenoid valve group with the airflow direction controller. It is used to switch the pipeline connection mode under specific conditions and change the airflow direction. Reverse airflow refers to the airflow in the opposite direction of material conveying. It can be achieved by adjusting the output port of the air source device or adding an independent back-blowing air pump. It is used to flush the deposited materials in the pipeline to the sending tank or discharge port. Reverse cleaning refers to the operation of flushing the inner wall of the pipeline through reverse airflow. Specifically, pulsed airflow or continuous high-pressure airflow mode can be used to remove residual materials and potential blockages on the pipe wall.

[0110] When the system executes a heavy unclogging strategy, for example, there is still a risk of material residue after the rotary unclogging mechanism intervenes, or when the cumulative conveying volume reaches a preset quality threshold, for example, the total conveying volume reaches 80%-95% of the set value, the adaptive cleaning module will automatically switch the connection path of the pneumatic conveying pipeline 2, close the forward airflow valve and open the reverse airflow valve to allow the high-pressure airflow to flow in the reverse direction of the pipeline. During this process, the airflow pressure can be dynamically adjusted according to the length of the pipeline. For example, a pressure sensor is used to provide real-time feedback and control the output power of the air source device to ensure that the airflow speed during reverse cleaning is sufficient to peel off material clumps attached to the pipe wall. After cleaning is completed, the system automatically resumes the forward conveying state to avoid extended downtime caused by manual intervention.

[0111] In existing technologies, cleaning operations rely on manual judgment of the blockage location and manual switching of pipelines, and usually only local dredging is performed after the blockage occurs. However, this solution triggers full pipeline cleaning by linking preset quality thresholds with dredging strategies. It can actively remove potential deposits before the material is completely blocked. At the same time, it combines the cumulative conveying volume to predict internal pipeline wear or residual risks, thereby achieving preventive maintenance.

[0112] It can simultaneously clear dredging residues after severe blockage occurs to avoid the risk of secondary blockage. At the same time, it automatically triggers periodic cleaning according to the total amount of material transported to prevent the pipe wall from shrinking in diameter or decreasing in fluidity due to long-term material accumulation, thereby reducing maintenance frequency and improving continuous operation stability.

[0113] It also includes a transportation efficiency evaluation module, which is connected to the blockage analysis module and generates optimization suggestions based on historical blockage locations and frequencies, including: when the cumulative number of blockages at the same location is greater than the preset number of blockages, outputting a suggestion to increase the curvature radius of the pipe bending section to 1-10 times the pipe diameter; when the pipe wall amplitude of the pneumatic conveying pipeline 2 exceeds the preset safety threshold, it is marked as a damaged pipe section and a replacement suggestion is triggered; it is recommended to add vibrators and temperature sensors at frequently blocked locations, and the suggestions are displayed in the three-dimensional modeling of the pneumatic conveying pipeline 2.

[0114] The transportation efficiency assessment module refers to a logical unit that analyzes system performance based on historical operating data. Specifically, it can use data mining algorithms to perform pattern recognition on the time series of blockage events and generate optimization suggestions by correlating the relationship between pipeline structural parameters and blockage frequency. This module provides data support for pipeline structure improvements by continuously learning the system operating status.

[0115] The cumulative number of blockages at the same location refers to the statistical number of blockage events that occur at the same physical coordinate within a set time period. This can be achieved by obtaining position mark data through the distributed sensor group 4 and establishing a spatial coordinate system database. This parameter is used to determine whether there are structural design defects in a specific pipe section.

[0116] The suggestion to increase the curvature radius of the pipe bend to 1-10 times the pipe diameter refers to a structural optimization plan for the elbow. Specifically, computer-aided engineering software can be used to simulate and verify the fluid mechanics model of the elbow. This suggestion can reduce the flow resistance of the material at the elbow.

[0117] The transportation efficiency assessment module continuously records the location information and environmental parameters of each blockage event to establish a time series database of pipeline operation status. When the blockage frequency of a specific pipe section exceeds a set threshold, the module automatically calls the pipeline 3D modeling data, analyzes the correlation between structural parameters such as the bending radius and inclination angle at that location and the material flow characteristics, and generates structural optimization suggestions. For pipe sections experiencing abnormal vibration, the module combines the spectral characteristics of the vibration acceleration sensor to identify pipeline damage status and trigger maintenance instructions. All optimization suggestions are visualized through the 3D modeling interface, allowing operators to quickly locate problem areas.

[0118] In some specific embodiments, when three consecutive blockages are detected at a certain elbow position, the system automatically marks the elbow as a red warning state in the three-dimensional model, and at the same time pops up a modification suggestion dialog box indicating "It is recommended to expand the curvature radius from the current 50mm to the range of 150-500mm". In another embodiment, when the peak value of the pipe wall vibration acceleration exceeds 10g, the system automatically generates a maintenance work order and pushes it to the equipment management department.

[0119] Existing dense-phase conveying systems typically only perform passive clearing after a blockage occurs and lack the ability to proactively analyze pipeline structural defects. This solution, by establishing a correlation model between historical operating data and pipeline design, can proactively identify structural defects in easily blocked sections of pipes, transforming post-processing into pre-emptive prevention and effectively reducing maintenance costs.

[0120] It can automatically generate pipeline optimization plans based on actual operating data, avoid the limitations of manual experience judgment, and reduce repeated blockage problems caused by unreasonable pipeline structure. At the same time, through the visualization suggestion display of three-dimensional modeling, it significantly improves the maintenance personnel's efficiency in identifying system defects and ensures the long-term stable operation of the pneumatic conveying pipeline 2.

[0121] The operation parameter constraint module is used to interrupt the operation and request operation confirmation when the wind pressure or flow rate set before the pneumatic conveying pipe is activated exceeds the preset safety range.

[0122] The operating parameter constraint module refers to a logical control unit used to limit the startup parameters of the pneumatic conveying system. It can be implemented by linking a programmable controller with a sensor. Its function is to avoid the risk of pipeline blockage due to human operational errors or equipment debugging errors by predicting abnormal parameter inputs. The preset safety range refers to the wind pressure or flow threshold set according to different material characteristics and pipeline specifications. It can be specifically generated by calculating the correlation function between the material fluidity index and the pipeline diameter. Its function is to provide a dynamic parameter boundary for the system to prevent material accumulation due to excessive pressure or flow. Interrupting operation means immediately stopping the startup instruction of the pneumatic conveying system when it is detected that the parameters exceed the preset range. It can be achieved by cutting off the power supply circuit of the air source device or closing the air intake valve. Its function is to block potential risks from the source. Requesting operation confirmation means generating a manual review request signal. Specifically, it can pop up a parameter abnormality warning and lock the operation authority through the human-machine interface. Its function is to reduce the probability of misoperation through a secondary confirmation mechanism.

[0123] Before the pneumatic conveying system is started, the operating parameter constraint module compares the wind pressure value and flow value set by the operator with the preset safety range. For example, the preset safety range can be dynamically generated according to the pipeline length, material solid-gas ratio and fluidity index. When the input wind pressure exceeds the maximum tolerance pressure corresponding to the current material, the module automatically triggers the interruption mechanism, suspends the start of the air source device, and displays a parameter over-limit warning on the control terminal. At this time, the operator needs to re-enter the parameters or adjust the material property data. The lock can only be released after the system is re-verified. For the flow parameter, if the set value is lower than the minimum flow required to maintain dense phase conveying, the module also performs an interruption operation and generates a prompt message, requiring the material filling rate to be checked or the sending tank pressure parameters to be adjusted.

[0124] Traditional dense-phase conveying systems rely on manual experience to set operating parameters and lack a real-time verification mechanism, which can easily lead to blockages due to improper parameter settings. However, this solution uses a preset safety range and automatic interruption mechanism to complete parameter compliance verification before the system starts, avoiding material compaction due to excessive pressure or local stagnation caused by insufficient flow, thereby reducing the probability of blockage from the source.

[0125] It effectively solves the problem of pipeline blockage caused by incorrect operating parameter settings. Through the pre-inspection mechanism and mandatory review process, it significantly reduces system abnormal shutdowns caused by human operating errors, ensuring that the pneumatic conveying system operates stably within the safety parameter range.

Claims

1. An automatic feeding system, comprising a feeding bin, a pneumatic conveying pipeline, an air source device and a discharging bin, characterized in that: Also includes: A distributed sensor group for monitoring the flux, pressure, temperature and vibration spectrum of predetermined points on the pneumatic conveying pipeline, including flow sensors, pressure sensors, temperature sensors and vibration acceleration sensors installed at several predetermined points on the pneumatic conveying pipeline; The digital twin model construction module constructs a three-dimensional model of the pneumatic conveying pipeline and maps the data collected by several preset points of the distributed sensor group on the pneumatic conveying pipeline to the three-dimensional model in real time; The congestion analysis module analyzes data collected by a distributed sensor group in real time and marks the congestion location for 3D modeling. It extracts the multi-parameter change curves in the T period before the congestion and constructs a congestion feature library. The congestion assessment level is generated based on the difference in the weight change rate between the feed and discharge silos. The hierarchical dredging execution module is equipped with differentiated dredging strategies, which are activated according to the congestion assessment level; The congestion warning module is connected to the congestion analysis module and is used to calculate in real time the dynamic similarity between the data of the distributed sensor group and the historical congestion curve in the congestion feature library. When the dynamic similarity exceeds the preset matching threshold, a primary warning is generated. If the primary warning lasts for a preset time, it is upgraded to an advanced warning to trigger the graded dredging execution module.

2. The automatic feeding system according to claim 1, characterized in that: The conditions for upgrading to a high-level warning in the congestion warning module also include: The real-time air pressure fluctuation exceeds the standard deviation of the historical curve; The material flux continues to decrease at a rate greater than the preset risk threshold.

3. The automatic feeding system according to claim 2, characterized in that: The feeding system also includes: A blockage simulation device includes a simulated blocker disposed inside a pipeline and a moving assembly for driving the simulated blocker to move, wherein the simulated blocker is used to create blockage conditions with different blockage assessment levels within the pneumatic conveying pipeline, and the moving assembly is used to drive the simulated blocker to move along the inner wall of the pneumatic conveying pipeline; The blockage simulation module is used to control the blockage simulation device, drive the blockage simulation device to several positions in the pneumatic conveying pipeline, and drive the blockage simulation device to form several degrees of blockage assessment levels. The multi-parameter change curve generated by the blockage simulation module during simulated blockage is recorded in the blockage feature library through the blockage analysis module.

4. An automatic feeding system according to claim 3, characterized in that: The simulated plug includes a plugging elastic sheet, inner magnetic rings arranged at both ends of the plugging elastic sheet, and a movable outer magnetic ring magnetically matched with the inner magnetic ring. The inner magnetic ring and the movable outer magnetic ring are respectively arranged inside and outside the pneumatic conveying pipe, and there are two inner magnetic rings and two movable outer magnetic rings. The movable outer magnetic ring is equipped with a movable component that drives it to move along the pneumatic conveying pipe. When the two inner magnetic rings approach each other, the blocking elastic sheet is driven to bend and deform. The deformation degree of the blocking elastic sheet is controlled according to the distance between the two inner magnetic rings. The different deformation degrees of the blocking elastic sheet are used to simulate different blockage degrees of the pneumatic conveying pipeline.

5. The automatic feeding system according to claim 2, characterized in that: The congestion warning module also includes a weight analysis unit for real-time monitoring of parameters to calculate a comprehensive risk value, which is calculated as follows: The dynamic similarity between the real-time distributed sensor group data and the historical congestion curve is multiplied by the first weight factor, plus the ratio of the number of times the air pressure change rate exceeds the set mutation threshold within the monitoring window to the length of the dynamic monitoring window, multiplied by the second weight factor, and the ratio of the temperature anomaly duration to the running time of the current transport task, multiplied by the third weight factor; Wherein, the first weighting factor is greater than the second weighting factor, the second weighting factor is greater than the third weighting factor, and the sum of the weights of the three is 1; When the comprehensive risk value exceeds the warning threshold dynamically determined based on pipeline structural parameters and material characteristics, an advanced warning is triggered.

6. The automatic feeding system according to claim 5, characterized in that: The dynamic determination rule of the warning threshold is: Based on the preset benchmark value, periodic adjustment is performed in combination with the proportional relationship between the actual length of the pneumatic conveying pipeline and the maximum length of the system, and linear correction is performed according to the difference between the material fluidity index and the maximum and minimum fluidity values.

7. The automatic feeding system according to claim 3, characterized in that: The congestion assessment levels include: The light blockage level is determined by the following method: the difference in weight change rate between the feed bin and the discharge bin is less than the preset value α; Moderate blockage level, which is determined by: preset value α≤the difference in weight change rate between the feed bin and the discharge bin<preset value β; The severe blockage level is determined as follows: the difference in weight change rate between the feed bin and the discharge bin ≥ the preset value β.

8. The automatic feeding system according to claim 7, characterized in that: When the congestion warning module issues a high-level warning, the corresponding differentiated unblocking strategy is triggered according to the congestion assessment level. The differentiated unblocking strategy includes: The mild unblocking strategy corresponds to the mild blockage level and is equipped with a vibrator that is triggered at the blockage location to vibrate the target section of the pneumatic conveying pipeline at a preset frequency. The medium unblocking strategy corresponds to the medium blockage level. The pipeline is switched to inject reverse airflow into the pneumatic conveying pipeline. The air pressure of the reverse airflow is not less than the forward conveying pressure. The severe dredging strategy corresponds to the severe blockage level and is equipped with a rotating dredging mechanism that extends into the blocked position, or drives the simulated blocker to move through a moving component to dredge the triggered blockage position.

9. The automatic feeding system according to claim 8, characterized in that: The feeding system also includes an adaptive cleaning module. When the heavy unclogging strategy is triggered or the cumulative conveying volume exceeds the preset quality threshold, the pipeline is switched to inject reverse airflow into the pneumatic conveying pipeline to perform reverse cleaning.

10. The automatic feeding system according to claim 1, characterized in that: The system also includes a transport efficiency evaluation module, which is connected to the congestion analysis module and generates optimization suggestions based on historical congestion locations and frequencies, including: When the cumulative number of blockages at the same location exceeds the preset number of blockages, a suggestion is output to increase the curvature radius of the pipe bend to 1-10 times the pipe diameter; When the wall vibration amplitude of a pneumatic conveying pipeline exceeds a preset safety threshold, it is marked as a damaged pipe section and a replacement recommendation is triggered; It is recommended to add preset points and vibrators at locations where congestion occurs frequently; The optimization proposals are displayed in a 3D model of the pneumatic conveying pipeline.

Images