Automatic feeding system

Abstract

According to the technical scheme, the automatic feeding system comprises a feeding bin, a pneumatic conveying pipeline, an air source device and a discharging bin, and further comprises a distributed sensor set used for monitoring the flux, pressure, temperature and vibration frequency spectrum of a preset point of the pneumatic conveying pipeline; comprising a flow sensor, a pressure sensor, a temperature sensor and a vibration acceleration sensor which are arranged at a plurality of preset points of a conveying pipeline, and a digital twin model building module which is used for building three-dimensional modeling of the pneumatic conveying pipeline and mapping state parameters of all positions of a pipeline distribution sensor group in real time; the blockage analysis module is used for marking a blockage position to three-dimensional modeling in real time based on data acquired by the distributed sensor group, extracting a multi-parameter change curve in a time period T before blockage, constructing a blockage feature library, and generating a blockage risk level through a weight change rate difference value of a feeding bin and a discharging bin; and the graded dredging execution module is configured with a differentiated dredging strategy, and the differentiated dredging strategy is started according to the blockage risk grade.

Description

Technical Field

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

[0002] The powder extrusion feeding system is mainly used for the transmission of powder. At present, there are generally two common types for powder transmission on the market, namely a screw feeder or a pneumatic conveying system. The pneumatic conveying system mainly uses the positive or negative pressure of the air flow to convey powder in a pipeline, including dilute-phase conveying for long distances and dense-phase conveying for medium and short distances. Among them, dense-phase conveying uses a low-speed air flow to convey high-concentration powder, and the solid-gas ratio is usually greater than 15, and even reaches more than 100. The powder is not completely suspended in the conveying pipeline, but forms plugs, sand dunes or fluidized beds and advances forward. In extrusion feeding, it is more appropriate to use medium-distance extrusion feeding, which can convey the powder to the vicinity of the extruder more gently compared with dilute-phase. The pneumatic conveying pipe system includes an air source, a sending tank, a conveying pipeline, a buffer hopper and a control system. Since dense-phase conveying uses a low-speed air flow, the most difficult problem it encounters during transmission is the risk of pipeline blockage. The reasons for pipeline blockage may be due to improper pipeline design of the conveying pipeline, incorrect operating parameters, changes in material properties, pipeline deformation or residues in the pipeline.

[0003] In the prior art, pressure detection is usually set at intervals on the conveying pipeline and at elbows. According to the principle that the pressure upstream of the blockage point rises sharply and the pressure downstream drops suddenly to atmospheric pressure, the blockage point is determined. However, this method has obvious hysteresis and can only be detected after the blockage occurs. Maintenance personnel usually use four types of methods for dredging, namely reverse air flow purging, segmental disassembly and dredging, vibrating the compaction material to loosen it and cooperating with air flow purging, and a special dredging device. But these methods all require manual intervention, and a large amount of time is wasted from discovering the blockage point to completing the dredging, which affects production efficiency. In addition, the prior art lacks a real-time monitoring and early warning mechanism for the blockage process, and cannot predict the blockage risk in advance, nor can it take differential dredging measures according to the blockage degree. Summary of the Invention

[0004] The purpose of the present invention is to provide a pneumatic conveying pipeline blockage early warning and hierarchical dredging system based on digital twin, which has the advantages of real-time monitoring of the pipeline state, early warning of the blockage risk and automatic execution of hierarchical dredging strategies, effectively reducing manual intervention and improving the 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 includes a feed bin, a pneumatic conveying pipeline, an air source device and a discharge bin, and further includes: A distributed sensor group for monitoring the flux, pressure, temperature and vibration spectrum at preset points of a pneumatic conveying pipeline, including a flow sensor, a pressure sensor, a temperature sensor and a vibration acceleration sensor arranged at several preset points of the conveying pipeline. A digital twin model construction module that constructs a three-dimensional model of the pneumatic conveying pipeline and real-time maps the state parameters of each position of the pipeline distribution sensor group. A blockage analysis module that based on the data collected by the distributed sensor group, marks the blockage position in real time to the three-dimensional model, extracts the multi-parameter change curves within a period T before blockage and constructs a blockage feature library, and generates a blockage risk level through the difference in the weight change rates of the feeding bin and the discharging bin. A hierarchical dredging execution module configured with differentiated dredging strategies, and starts the differentiated dredging strategies according to the blockage risk level.

[0006] By adopting the above technical solutions, the problems of passive response to blockages and low positioning efficiency in traditional systems are solved. The multi-parameters of the pipeline are monitored in real time by the distributed sensor group, and the physical state is three-dimensionally mapped in combination with the digital twin model, realizing accurate marking of the blockage position and automatic generation of the risk level, and triggering the hierarchical dredging strategy, significantly reducing the downtime for troubleshooting.

[0007] The present invention is further configured as: further including a blockage warning module, connected to the blockage analysis module, for calculating the dynamic similarity between the data of the real-time distributed sensor group and the historical blockage curves in the blockage feature library, generating a primary warning when the similarity exceeds a preset matching threshold, and if the duration of the primary warning reaches a preset time, promoting it to a high-level warning to trigger the hierarchical dredging execution module.

[0008] By adopting the above technical solutions, aiming at the defect of false alarms caused by instantaneous fluctuations, through dynamic similarity calculation and verification of the preset duration, the primary warning is upgraded to a high-level warning to ensure that the warning signal triggers dredging only when there is continuous abnormality, reducing the misoperation rate.

[0009] The present invention is further configured as: the conditions for upgrading to a high-level warning in the blockage warning module further include: the amplitude of the real-time air pressure fluctuation exceeds the standard deviation of the historical curve; the continuous decline rate of the material flux is greater than the preset risk threshold.

[0010] By adopting the above technical solutions, the problem of missed reporting of high-risk blockages by a single index is solved, and composite conditions such as the amplitude of the air pressure fluctuation exceeding the standard deviation and the continuous decline rate of the flux exceeding the threshold are added to accelerate the response to rapidly deteriorating blockages and avoid blockage spread.

[0011] The present invention is further configured as: the feeding system further includes: The blockage simulation device includes a simulation blockage device arranged inside the pipeline and a moving component for driving the simulation blockage device to move. The simulation blockage device is used to form blockage situations with different blockage evaluation levels inside the pipeline, and the moving component is used to drive the simulation blockage device 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 in several positions of the pipeline, and drive the blockage simulation device to form several degrees of blockage evaluation levels. The multi-parameter change curve during the simulated blockage generated by the blockage simulation module is recorded in the blockage feature library through the blockage analysis module.

[0012] By adopting the above technical solutions, the blockage feature library is dynamically optimized: different levels of blockages are generated at any position in the pipeline by the movable simulation blockage device, the multi-parameter change curves are collected in real time and entered into the feature library, improving the generalization ability and accuracy of the blockage prediction model.

[0013] The present invention is further configured as: the blockage simulator includes a blockage elastic sheet, inner magnetic rings arranged at both ends of the blockage elastic sheet, and moving outer magnetic rings magnetically coupled with the inner magnetic rings. The inner magnetic rings and the moving outer magnetic rings are respectively arranged inside and outside the pneumatic conveying pipeline, and there are two inner magnetic rings and two moving outer magnetic rings. Moving components for driving them to move along the pneumatic conveying pipeline are respectively arranged on the moving outer magnetic rings. When the two inner magnetic rings are close to each other, the blockage elastic sheet is driven to bend and deform, and the deformation degree of the blockage elastic sheet is controlled according to the distance between the two inner magnetic rings. Different deformation degrees of the blockage elastic sheet are used to simulate different blockage degrees of the pneumatic conveying pipeline.

[0014] By adopting the above technical solutions, non-contact and precise blockage simulation is achieved. The magnetic ring linkage design drives the deformation of the internal elastic sheet, avoiding pipeline wear caused by mechanical contact; by adjusting the distance between the inner magnetic rings, the bending degree of the elastic sheet is controlled to accurately simulate different degrees of blockages, ensuring the authenticity of the data in the feature library.

[0015] The present invention is further configured as: the blockage warning module further includes a weight analysis unit for real-time monitoring of parameter calculation of the comprehensive risk value, and its calculation method is as follows: Multiply the dynamic similarity between the real-time sensor data and the historical blockage characteristics by the first weight factor, add the ratio of the number of times the air pressure change rate exceeds the set mutation threshold within the monitoring window to the duration of the dynamic monitoring window multiplied by the second weight factor, and then add the ratio of the duration of the temperature anomaly to the running time of the current conveying task multiplied by the third weight factor; Wherein, the first weight factor is greater than the second weight factor, the second weight factor is greater than the third weight factor, and the sum of the three weights is 1; When the comprehensive risk value exceeds the warning threshold dynamically determined according to the pipeline structure parameters and material characteristics, a high-level warning is triggered.

[0016] By adopting the above technical solution, the problem that the fixed risk threshold cannot adapt to different pipelines and materials is solved.

[0017] The present invention is further configured that: the dynamic determination rule of the warning threshold is: Based on a preset reference value, it is periodically adjusted in combination with the proportional relationship between the actual length of the pipeline and the maximum length of the system, and linearly corrected according to the difference between the material fluidity index and the maximum and minimum fluidity values; Among them, the influence of the pipeline length is reflected by a periodic function, and the influence of the material fluidity is reflected by a normalized difference.

[0018] By adopting the above technical solution, the problem that the fixed risk threshold cannot adapt to different pipelines and materials is solved.

[0019] The present invention is further configured that: the blockage risk levels include: If the difference in the weight change rate between the feed bin and the discharge bin < preset value α, it is determined as a mild blockage level; If preset value α ≤ the difference in the weight change rate between the feed bin and the discharge bin < preset value β, it is determined as a moderate blockage level; If the difference in the weight change rate between the feed bin and the discharge bin ≥ preset value β, it is determined as a severe blockage level.

[0020] By adopting the above technical solution, the randomness of the empirical judgment of the blockage degree is eliminated, and the accurate matching between the dredging intensity and the blockage degree is realized.

[0021] The present invention is further configured that: when the blockage warning module issues a high-level warning, corresponding differentiated dredging strategies are triggered according to the blockage risk levels, and the differentiated dredging strategies include: A mild dredging strategy, corresponding to the mild blockage level, is equipped with a vibrator, and the vibrator at the blockage position is triggered to oscillate the target pipe section of the pneumatic conveying pipeline at a preset frequency; A moderate dredging strategy, corresponding to the moderate blockage level, switches the pipeline to inject reverse air flow into the pneumatic conveying pipeline, and the air pressure of the reverse air flow is not less than the forward conveying pressure; A severe dredging strategy, corresponding to the severe blockage level, is equipped with a rotary dredging mechanism to extend into the blockage position, or a simulation blockage device is driven by a moving component to move to dredge the triggered blockage position.

[0022] By adopting the above technical solution, aiming at the problem that a single dredging strategy is ineffective or damages the pipeline, a three-level progressive dredging is designed: mild vibration to loosen particles, moderate reverse air flow blowing with air pressure ≥ forward pressure, and severe telescopic rotary mechanism mechanical dredging, which protects the integrity of the pipeline while efficiently clearing blockages.

[0023] The present invention is further configured that: the feeding system further includes an adaptive cleaning module, which, when triggering a heavy dredging strategy or when the cumulative conveying volume exceeds a preset mass threshold, switches the pipeline to inject a reverse air flow into the pneumatic conveying pipeline to perform reverse cleaning.

[0024] By adopting the above technical solution, it prevents secondary blockage caused by residues after dredging. When the heavy dredging is completed or the cumulative conveying volume exceeds the mass threshold, it automatically switches the air path to perform reverse air flow cleaning, removes the particles attached to the pipe wall, reduces the cleaning frequency and lowers the energy consumption.

[0025] The present invention is further configured that: it further includes a conveying efficiency evaluation module, which is connected to the blockage analysis module and generates optimization suggestions based on the historical blockage positions and frequencies, including: When the cumulative blockage times at the same position are greater than the preset blockage times, it outputs a suggestion to increase the curvature radius of the pipeline bend section to 1 - 10 times the pipe diameter; When the wall amplitude of the pneumatic conveying pipeline exceeds the preset safety threshold, it marks it as a damaged pipe section and triggers a replacement suggestion; Suggestions to add vibrators and temperature sensors at frequently blocked positions, and display the suggestions in the 3D modeling of the pneumatic conveying pipeline.

[0026] By adopting the above technical solution, it closes the loop to optimize the pipeline design defects, outputs a suggestion to increase the curvature radius to 1 - 10 times the pipe diameter for high-frequency blockage points, marks the pipe sections with excessive vibration (amplitude > 0.5mm) for replacement, and visualizes the pre-set sensor scheme in the digital twin model, reducing the blockage probability from the root cause.

[0027] The present invention is further configured that: it further includes an operating parameter constraint module, which is used to interrupt the operation and request operation confirmation when the wind pressure or flow rate set before the pneumatic conveying pipe is enabled exceeds the preset safety range.

[0028] By adopting the above technical solution, it blocks the risk of human misoperation, forcibly interrupts and requests confirmation when the wind pressure or flow rate exceeds the safety range (set based on the pipeline pressure resistance limit), and avoids blockage or leakage accidents caused by incorrect parameter settings.

[0029] Due to adopting the above technical solutions, the present invention has remarkable technical effects: a pneumatic conveying pipeline blockage early warning and hierarchical dredging system based on digital twin provided by the present application collects multi-dimensional data of the pipeline in real time through a distributed sensor group, dynamically maps the pipeline state in combination with the digital twin model, and performs risk grading and precise positioning based on the blockage feature library. At the same time, it matches different dredging strategies to achieve early warning and automatic hierarchical dredging, having the advantages of significantly improving the pipeline operation safety and maintenance efficiency. Description of the Drawings

[0030] Figure 1It is a schematic structural diagram of a part of the device in the embodiment; Figure 2 It is a schematic structural diagram of the automatic feeding system in the embodiment; Figure 3 It is a schematic structural diagram of the simulated blockage device in the embodiment.

[0031] The names of the parts referred to by each digital label in the above drawings are as follows: 1, feed bin; 2, pneumatic conveying pipeline; 3, discharge bin; 4, distributed sensor group; 5, simulated blockage device; 51, blockage elastic sheet; 52, inner magnetic ring; 53, moving outer magnetic ring; 6, moving component. Specific embodiments

[0032] The present invention will be further described in detail below with reference to the drawings and embodiments.

[0033] Embodiment: The powder extrusion feeding system often adopts the dense-phase pneumatic conveying technology, and the high-concentration powder is pushed by the low-speed air flow to form a plug or fluidized bed for conveying. Such a system is prone to pipeline blockage due to pipeline design defects, material property changes or improper operation parameters. The 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 blockage occurs, manual intervention is required for reverse air flow purging or disassembly and dredging, resulting in long production line downtime and significantly increased maintenance costs.

[0034] The inventor observed that traditional blockage detection only relies on single-point pressure parameters and cannot comprehensively reflect the material flow state in the pipeline. By analyzing the correlation changes of air pressure, flow rate and vibration signals before blockage occurs, it is found that multi-parameter fusion monitoring can identify blockage risks earlier. Combining digital twin technology to perform three-dimensional dynamic modeling of the pipeline state can real-time track abnormal material flow. In addition, a hierarchical dredging mechanism is designed for different blockage degrees, which can avoid energy waste or equipment damage caused by a single dredging method.

[0035] An automatic feeding system includes an automatic feeding system of a feed bin 1, a pneumatic conveying pipeline 2, a gas source device and a discharge bin 3, and introduces a distributed sensor group 4 to monitor the pipeline flux, pressure, temperature and vibration spectrum in real time, constructs a digital twin model to map the pipeline state, marks the blockage position through a blockage analysis module and generates a risk level, and finally starts a differential dredging strategy by a hierarchical dredging execution module.

[0036] 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 conveying pipeline. Specifically, it can be implemented using electromagnetic flowmeters, piezoelectric pressure transmitters, thermocouple sensors, and MEMS accelerometers, and is used to collect dynamic parameters at different positions of the pipeline. The digital twin model construction module refers to establishing a pipeline geometric model through 3D modeling software and integrating sensor data to achieve virtual-real mapping. Specifically, it can use finite element analysis algorithms to simulate the material flow state. The blockage analysis module refers to identifying abnormal patterns in the multi-parameter change curves based on machine learning algorithms. Specifically, it can use time series analysis methods to extract blockage precursor features. The hierarchical dredging execution module refers to a composite control system integrating vibrators, reverse air flow valves, and mechanical dredging devices, and can specifically achieve strategy switching through PLC programming.

[0037] 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 pipeline, and transmits the data to the digital twin model construction module to generate a 3D visualization model. When the material flow rate in a certain section of the pipeline decreases, the blockage analysis module combines the weight change rate difference between the feed bin 1 and the discharge bin 3 to determine the blockage level. For example, when the weight difference exceeds the preset threshold, a medium blockage determination is triggered. The hierarchical dredging execution module selects corresponding strategies according to the determination results. For example, it starts a local vibrator to loosen the material for a minor blockage, and enables a rotary dredging mechanism to physically remove the blockage for a severe blockage. The entire process realizes closed-loop control through real-time data interaction and model update.

[0038] Traditional methods only rely on single-point pressure detection for blockage judgment, cannot distinguish the blockage degree, and have limited positioning accuracy. This solution realizes panoramic visualization of the pipeline state through multi-dimensional sensor fusion monitoring and combines with a digital twin model, can accurately identify the blockage position and development trend. At the same time, the hierarchical dredging strategy avoids the energy waste or excessive wear of equipment that may be caused by traditional single dredging methods. For example, in the case of a minor blockage, a low-energy vibration method is preferred rather than full-pressure reverse air flow scouring.

[0039] It realizes early warning and precise positioning of pipeline blockages, shortens the blockage handling response time, the multi-parameter collaborative analysis effectively reduces the misjudgment rate, the digital twin model provides an intuitive state monitoring interface for operators, and the hierarchical dredging mechanism optimizes energy consumption control while ensuring the dredging effect, extends the service life of key components, and is especially suitable for the continuous and stable operation requirements in the high solid-gas ratio dense-phase conveying scenario.

[0040] The blockage warning module is connected to the blockage analysis module and is used to calculate the dynamic similarity between the data of the real-time distributed sensor group 4 and the historical blockage curves in the blockage feature library. When the similarity exceeds the preset matching threshold, a primary warning is generated. If the duration of the primary warning reaches the preset time, it is upgraded to a high-level warning to trigger the hierarchical dredging execution module.

[0041] 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, which can be triggered by a relay to automatically start the dredging equipment.

[0042] 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 graded dredging execution module to start the corresponding dredging strategy.

[0043] The existing technology relies on locating the blockage point through pressure mutation after the blockage occurs, which requires manual intervention and leads to prolonged downtime. This solution identifies the blockage trend in advance through dynamic similarity analysis, and combines the grading mechanism of primary and advanced warnings to trigger an automated response in the early stage of blockage, shortening the fault handling cycle.

[0044] This application can identify the evolution trend of blockage risk 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 pipeline blockage prevention and treatment.

[0045] 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 decrease rate of material flux exceeding the preset risk threshold.

[0046] The real-time air pressure fluctuation amplitude refers to the degree of dispersion of air pressure values within the current monitoring period. Specifically, it can be achieved by statistically analyzing historical operation data using the standard deviation calculation algorithm, and is used to quantify the degree of abnormal air pressure fluctuations. The continuous decline rate of the material flux refers to the reduction amplitude of the material throughput per unit time. Specifically, it can be achieved by calculating the first derivative of the real-time data collected by the flow sensor, and is used to reflect the accelerating state of the blocked material flow. The standard deviation of the historical curve refers to the pipeline operation parameter benchmark model established under normal working conditions. Specifically, it can be achieved by calculating the variance of the pressure data samples within the sliding time window, and is used to establish the basis for dynamic threshold judgment.

[0047] When it is detected that the air pressure fluctuation amplitude in the pipeline exceeds the standard deviation range formed by the historical operation data, it indicates that there is an abnormal pressure aggregation phenomenon 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 blocked material flow is deteriorating rapidly. When the above two conditions are met simultaneously, the early warning module will upgrade the primary warning to a high-level warning, triggering the hierarchical dredging execution module to take active intervention measures.

[0048] Traditional blockage detection only relies on the differential pressure between the upstream and downstream of the pressure sensor for single-point judgment, and is prone to misjudgment due to working condition fluctuations. This technical solution realizes the dynamic tracking and trend prediction of the blockage formation process by introducing the historical air pressure fluctuation standard deviation as a dynamic benchmark and combining the dual criteria of the material flux change rate.

[0049] This application can identify potential blockage risks earlier, trigger the dredging mechanism before the material flow is completely stagnant, avoid the problem of complete pipeline blockage caused by detection lag in traditional methods, and at the same time reduce the probability of false triggering of a single sensor through the multi-parameter collaborative judgment mechanism, reducing the number of unnecessary shutdown and maintenance times.

[0050] The blockage early warning module also includes a weight analysis unit, which is used to monitor and calculate the comprehensive risk value R of the parameters in real time. The calculation method is as follows: Multiply the dynamic similarity between the real-time sensor data and the historical blockage characteristics by the first weight factor, add the ratio of the number of times the air pressure change rate in the monitoring window exceeds the set mutation threshold to the duration of the dynamic monitoring window multiplied by the second weight factor, and then add the ratio of the abnormal temperature duration to the running time of the current conveying task multiplied by the third weight factor; That is, its expression is R = k1S + k2(N mutation / t monitoring) + k3*(t anomaly / t total), where S is the dynamic similarity between the real-time sensor data and the historical blockage curve and its value range is from 0 to 1, t monitoring is the duration of the dynamic monitoring window and its value range is from one-tenth to one-half of the total duration of a single material conveying task, N mutation is the number of times the air pressure change rate in the monitoring window exceeds the set mutation threshold, t anomaly is the cumulative time when the temperature sensor reading continuously exceeds the material tolerance temperature threshold, t total is the time that the current conveying task has been running. 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 weights of the three is 1. When the comprehensive risk value exceeds the early warning threshold dynamically determined according to the pipeline structure parameters and material characteristics, an advanced early warning is triggered.

[0051] The duration of the dynamic monitoring window refers to the data acquisition time range dynamically adjusted according to the conveying task cycle. Specifically, it can be implemented by using the time window sliding algorithm. By dynamically matching the change trend of the conveying cycle, the monitoring sensitivity is improved. The dynamic similarity refers to the matching degree between the real-time sensor data and the historical blockage curve in the time series. Specifically, it can be calculated by using the dynamic time warping algorithm. By eliminating the time axis offset difference, accurate comparison is achieved. The weight factor refers to the contribution ratio of each parameter to the comprehensive risk value. Specifically, the analytic hierarchy process can be used for assignment. By distinguishing the priorities of different parameters, the decision-making weight of the core indicators is strengthened. The dynamic adjustment of the R threshold refers to establishing a functional relationship based on the pipeline length and the material fluidity index. Specifically, the linear interpolation method can be used for calculation. By correlating the physical parameters with the material characteristics, the adaptability of the early warning trigger condition is improved.

[0052] During the operation of the conveying task, the weight analysis unit continuously collects the flow rate, pressure, temperature and vibration data of the distributed sensor group 4, calculates the similarity S between the current data and the historical blockage curve through the dynamic time warping algorithm, simultaneously counts the number of air pressure mutations N mutation in the monitoring window, and accumulates the temperature overrun time t anomaly. Using the current conveying task duration t total as the denominator, N mutation and t anomaly are respectively normalized, and the three parameters are weighted and summed according to the preset weight factors k1, k2, k3 to generate the comprehensive risk value R. When this value exceeds the R threshold calculated based on the pipeline length L and the material fluidity index μ, the system automatically upgrades the early warning level. For example, for a 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.

[0053] Traditional blockage warning methods only make threshold judgments based on a single pressure parameter and cannot distinguish between accidental fluctuations and real blockage risks. This solution constructs a comprehensive risk assessment model by integrating multi-dimensional sensor data, introduces multiple correlation parameters such as time series similarity analysis, air pressure mutation frequency statistics, and temperature anomaly duration in the calculation, and forms a composite criterion in combination with a weight allocation mechanism. For example, when the material fluidity decreases and the pipeline friction coefficient increases, the system dynamically adjusts the R threshold to reduce the misjudgment probability, improving the warning accuracy compared with the fixed threshold method.

[0054] This application realizes the multi-dimensional quantitative assessment of blockage risks, effectively identifies early blockage signs through dynamic similarity matching, enhances the response speed to sudden blockages by combining air pressure mutation frequency statistics, can detect chronic blockage risks caused by material adhesion in advance through cumulative monitoring of temperature anomaly time, and the differential setting of weight factors strengthens the leading role of core parameters and avoids interference from secondary parameters in judgment. The dynamic adjustment mechanism of the R threshold enables the warning trigger condition to adapt to different pipeline lengths and material characteristics, ensuring timely warning of high-risk working conditions while reducing the false alarm rate, thereby shortening the dredging response time and reducing production interruption losses.

[0055] The dynamic determination rule of the warning threshold is: Based on a preset reference value, it is periodically adjusted in combination with the proportional relationship between the actual pipeline length and the maximum system length, and linearly corrected according to the difference between the material fluidity index and the maximum and minimum fluidity values; among them, the influence of the pipeline length is reflected by a periodic function, and the influence of the material fluidity is reflected by a normalized difference, specifically as follows: The threshold R threshold of the comprehensive risk value R in the blockage warning module is dynamically adjusted according to the pipeline physical length and the material fluidity index. The dynamic adjustment rule is R threshold = 0.5×(L / Lmax) + 0.5×(μ−μmin) / (μmax−μmin), where L is the pipeline physical length, Lmax is the maximum pipeline length of the system, μ is the material fluidity index, μmax is the maximum material fluidity value, and μmin is the minimum material fluidity value.

[0056] The physical length of the pipeline refers to the actual extended distance of the conveying pipeline from the feed bin 1 to the discharge bin 3. Specifically, it can be measured using 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 that the system design allows to be installed. Specifically, it 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 performance of powder or granular materials during pneumatic conveying. Specifically, it can be measured using a repose angle tester or a shear tester. The introduction of the material fluidity index enables the threshold adjustment to reflect the influence of the material's own characteristics on the blockage risk. The maximum and minimum values of material fluidity refer to the limit values of the material fluidity that the system allows to convey. Specifically, it can be determined according to historical conveying data or the material property database. The setting of the maximum and minimum values of material fluidity is used to normalize the material fluidity index.

[0057] During the blockage warning process, the threshold of the comprehensive risk value R is not fixed, but is dynamically calculated according to the actual length of the current conveying pipeline and the fluidity characteristics of the conveyed material. When the pipeline length increases, due to the extended conveying path, the air pressure attenuation and the risk of material retention increase, and the threshold will increase accordingly to avoid false alarms; when the material fluidity decreases, since the material is likely to form deposits or lumps in the pipeline, 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.

[0058] Traditional blockage warning systems usually use fixed thresholds to judge risks and cannot effectively cope with the working condition differences brought about by different pipeline lengths and material characteristics. For example, when conveying low-fluidity materials through a long pipeline, the fixed threshold may cause the warning to be triggered prematurely, resulting in unnecessary dredging operations; while when conveying high-fluidity materials through a short pipeline, the fixed threshold may cause the warning to be delayed, missing the best dredging opportunity. Through the dynamic adjustment mechanism of this solution, the risk judgment benchmark can match the current conveying conditions in real time, significantly improving the adaptability and accuracy of the warning system.

[0059] It realizes the intelligent adjustment of the blockage warning threshold and effectively solves the problem that it is difficult to balance the warning sensitivity and false alarm rate of the traditional fixed threshold method in scenarios with 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 in low-risk working conditions and ensuring timely warning signals in high-risk working conditions, thereby improving the operation reliability and maintenance efficiency of the entire pneumatic conveying system.

[0060] Method for classifying blockage risk levels, including determining a mild blockage level when the difference in the weight change rates of the feed bin 1 and the discharge bin 3 is less than a preset value α, determining a moderate blockage level when the difference is in the range of α to β, and determining a severe blockage level when the difference exceeds β.

[0061] The difference in the weight change rates of the feed bin 1 and the discharge bin 3 refers to the difference in the weight change of the material in the two bins per unit time. Specifically, it can be achieved by collecting data in real time through high-precision weighing sensors and calculating the instantaneous weight change rate. This difference directly reflects the degree of obstruction of the material flow in the pipeline. The preset values α and β refer to the classification thresholds set according to the pipeline conveying capacity, material characteristics, and historical operation data. For example, α can be set to 20% of the theoretical throughput, and β can be set to 50% of the theoretical throughput. Precise classification under different working conditions is achieved by dynamically adjusting the thresholds.

[0062] During the material transportation process, the weighing sensors continuously monitor the material reduction rate of the feed bin 1 and the material increase rate of the discharge bin 3. When the difference in the rates of the two bins continuously remains below α, it indicates that there is only a slight delay in the material flow. At this time, the system determines it as a mild blockage; when the difference rises between α and β, it indicates that the degree of obstruction of the material flow has increased, and the system upgrades to a moderate blockage level; when the difference breaks through β, it is judged as a severe blockage state where the material flow is close to stagnation. This classification logic is executed in real time by an embedded controller and forms a closed-loop control with the dredging actuator.

[0063] Traditional blockage detection only relies on pressure changes to judge the occurrence of blockage, unable to quantify the degree of blockage and there is a risk of misjudgment. This solution establishes a multi-level blockage determination system through the dynamic monitoring of the difference in the weight change rates of the two bins, and can accurately identify blockage conditions of different severities.

[0064] It realizes the refined identification of the blockage state, provides a reliable basis for the selection of subsequent dredging strategies. Compared with the traditional single processing method, the classification determination mechanism can avoid excessive energy consumption in case of low-risk blockages, and at the same time ensure the timely activation of strong dredging means in case of high-risk blockages, effectively balancing production safety and operation economy.

[0065] When the blockage warning module issues a high-level warning, corresponding differentiated dredging strategies are triggered according to the blockage risk level. The differentiated dredging strategies include a mild dredging strategy, corresponding to the mild blockage level, equipped with a vibrator, triggering the vibrator at the blockage position to oscillate the target section of the pneumatic conveying pipeline 2 at a preset frequency; a moderate dredging strategy, corresponding to the moderate blockage level, switching the pipeline to inject reverse air flow into the pneumatic conveying pipeline 2, and the air pressure of the reverse air flow is not less than the forward conveying pressure; a severe dredging strategy, corresponding to the severe blockage level, equipped with a rotary dredging mechanism to extend into the blockage position, or driving the simulated blockage device 5 to move through a moving component to dredge the triggered blockage position.

[0066] The blockage risk level refers to the classification of the severity of blockage based on the difference in the weight change rates of the feed bin 1 and the discharge bin 3. The blockage level is judged by quantifying the degree of imbalance in material flow. Specifically, it can be achieved by using a preset threshold interval matching algorithm for accurately matching different dredging intensities. The vibrator refers to a mechanical vibration device installed on the outer wall of the pipeline, which can be specifically realized by an electromagnetic or pneumatic vibrator. By generating high-frequency vibrations, the blocked material is loosened. The reverse air flow refers to the air flow opposite to the material conveying direction, which can be specifically achieved by switching the direction of the gas source pipeline and adjusting the air pressure, and is used to destroy the material accumulation structure in the blocked area. The rotary dredging mechanism refers to a mechanical device with a rotary cutting function, which can be specifically realized by a pneumatic actuator with an internal spiral cutter head. By means of physical cutting, stubborn blockages are removed. It can also drive the simulation blocker 5 to perform reciprocating dredging actions through the moving component, for further utilization of the blockage simulation device.

[0067] When the system detects a high-level warning signal, first determine the blockage level according to the difference in the weight change rates of the feed bin 1 and the discharge bin 3. For mild blockages, activate the vibrator corresponding to the blocked position to generate high-frequency vibrations to cause periodic deformation of the pipe wall, prompting the material to resume flow; when it is a moderate blockage, switch the gas source direction and generate a reverse air flow not lower than the forward conveying pressure, using the impact force of the air flow to destroy the static balance of the blocked material; for severe blockages, start the rotary dredging mechanism to extend into the blocked section, and cut the compacted material layer through the rotary cutter head to form a dredging channel.

[0068] In the prior art, only a single dredging method is relied on and manual judgment of the blockage degree is required. This solution automatically matches the optimal dredging intensity according to real-time data by establishing a multi-level dredging strategy library, avoiding energy waste caused by over-dredging or secondary blockages caused by insufficient dredging. The traditional method requires manual operation of the dredging device after shutdown, while this solution realizes rapid dredging through a hierarchical execution mechanism in a non-stop state.

[0069] It effectively solves the problem of low dredging efficiency caused by inaccurate judgment of the blockage level during the dense-phase conveying process. Through the multi-level cooperative action of vibration, reverse air flow scouring and mechanical cutting, it significantly shortens the treatment time for blockages of different severities, and at the same time reduces the frequency of manual intervention and the risk of equipment damage.

[0070] The automatic feeding system further includes a blockage simulation device and a blockage simulation module. The blockage simulation device includes a simulation blocker 5 disposed inside the pipeline and a moving component for driving the simulation blocker 5 to move. The simulation blocker 5 is used to form blockage situations with different blockage evaluation levels inside the pipeline, and the moving component is used to drive the simulation 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 in several positions of the pipeline, and drive the blockage simulation device to form several degrees of blockage evaluation levels. The multi-parameter change curve during the simulated blockage generated by the blockage simulation module is recorded in the blockage feature library through the blockage analysis module.

[0071] Among them, the blockage simulation device refers to a device that actively creates a controllable blockage scenario, and specifically can be realized by the cooperation of a mechanical component with a deformation structure and a magnetic attraction driving device.

[0072] The blockage simulator in this embodiment includes a blockage elastic sheet 51, inner magnetic rings 52 disposed at both ends of the blockage elastic sheet 51, and a moving outer magnetic ring 53 magnetically attracted and cooperated with the inner magnetic rings 52. The inner magnetic rings 52 and the moving outer magnetic ring 53 are respectively disposed inside and outside the pneumatic conveying pipeline 2, and there are two inner magnetic rings 52 and two moving outer magnetic rings 53. A moving component for driving it to move along the pneumatic conveying pipeline 2 is cooperatively provided on the moving outer magnetic ring 53. The moving component in this embodiment is specifically a driving motor with a driving wheel and is fixedly connected to the moving outer magnetic ring 53. A plurality of moving components are annularly arranged on the outer ring of the moving outer magnetic ring 53. When the two inner magnetic rings 52 are close to each other, the blockage elastic sheet 51 is driven to bend and deform, and the deformation degree of the blockage elastic sheet 51 is controlled according to the distance between the two inner magnetic rings 52. Different deformation degrees of the blockage elastic sheet 51 are used to simulate different blockage degrees of the pneumatic conveying pipeline 2.

[0073] The blockage elastic sheet 51 refers to a sheet-like structure with elastic deformation ability, and specifically can be made of polyurethane or silicone material. It generates different cross-sectional area blocking effects through bending deformation, and is used to form a controllable local blockage inside the pipeline. The inner magnetic ring 52 refers to an annular magnetic component fixed to the end of the elastic sheet, and specifically can be made of neodymium iron boron permanent magnet, which forms a magnetic attraction cooperation with the external moving outer magnetic ring 53, and is used to convert the mechanical movement of the external moving component into the deformation control of the elastic sheet. The moving outer magnetic ring 53 refers to a magnetic driving component disposed on the outer wall of the pipeline, and specifically can be an electromagnet or a permanent magnet. By moving along the axial direction of the pipeline, the relative position with the inner magnetic ring 52 is changed, and is used to accurately adjust the distance between the two inner magnetic rings 52. The moving component refers to an actuator for driving the outer magnetic ring to move, and specifically can be a linear motor or a ball screw device, and is used to realize the positioning and continuous displacement of the outer magnetic ring on the outer wall of the pipeline.

[0074] The inner magnetic ring 52 and the blocking elastic sheet 51 form an integrated structure and are installed on the inner wall of the pipeline. The movable outer magnetic ring 53 slides axially on the outer wall of the pipeline driven by a moving component. When the two outer magnetic rings are driven by the moving component to approach each other, their magnetic fields cause the corresponding two inner magnetic rings 52 to attract each other, resulting in the bending deformation of the blocking elastic sheet 51. As the distance between the inner magnetic rings 52 decreases, the bending degree of the elastic sheet increases, thereby forming blocking areas with different cross-sectional areas in the pipeline. By controlling the displacement of the moving component, the distance between the inner magnetic rings 52 can be accurately adjusted, and thus a continuous simulation state from slight blockage to complete blockage can be obtained. This structure allows adjustable blockage to be generated at any position in the pipeline and can be operated without invading the interior of the pipeline.

[0075] Traditional blockage simulation devices usually adopt fixed throttle valves or inserted baffles, which cannot achieve variable adjustment of the axial position of the pipeline, and manual intervention is required to replace components for adjusting the blockage degree. Through the synergistic effect of magnetic coupling and the external moving component, this solution realizes non-contact blockage simulation, can dynamically change the blockage position and severity without interrupting the conveying process, and significantly improves the acquisition efficiency of blockage characteristic data and the model training accuracy.

[0076] Through the above technical solutions, this application realizes the controllability and repeatability of blockage simulation in the pneumatic conveying pipeline 2, can accurately generate blockage conditions at different positions and different levels, and provides multi-dimensional data support for the construction of the blockage characteristic library. The magnetic drive method avoids the damage to the pipeline integrity caused by traditional mechanical structures. The precise positioning ability of the moving component ensures the spatial resolution of blockage simulation, and the continuous deformation characteristic of the elastic sheet realizes stepless adjustment of the blockage degree.

[0077] The function of the moving component is to accurately position the simulation blockage device 5 to any position in the pipeline to cover the characteristics of different pipe sections. The blockage simulation module controls the combination of the blockage position and the blockage level through a preset program to generate multi-dimensional data samples for expanding the coverage range of the blockage characteristic library.

[0078] In the initial stage of system operation or during the maintenance phase, the blockage simulation module drives the moving component to move the simulation blockage device 5 to the target position, and adjusts the distance between the inner magnetic rings 52 to cause the blocking elastic sheet 51 to generate a preset deformation, forming blockages with different cross-sectional areas. At this time, the distributed sensor group 4 collects the flux, pressure, temperature, and vibration spectrum data in this simulated blockage state, generates corresponding multi-parameter change curves after being processed by the blockage analysis module, and stores them in the blockage characteristic library. For example, when the simulation blockage device 5 moves to the elbow section, the phenomenon of increased local resistance caused by material accumulation can be simulated, and the air pressure fluctuation characteristics in this scenario can be recorded; when the deformation degree of the blocking elastic sheet 51 reaches severe blockage, the abnormal peak data of the high-frequency component in the vibration spectrum can be collected.

[0079] The prior art relies on passive data collection after actual blockages occur, unable to cover various blockage scenarios, resulting in a single sample in the feature library and limited warning accuracy. This solution can systematically obtain the blockage response data of each area of the pipeline by actively simulating blockage conditions at different positions and levels, especially the differential characteristics of easily blocked areas such as elbows and diameter-changing sections, enabling the blockage feature library to have a more comprehensive sample coverage and improving the reliability of dynamic similarity calculation.

[0080] This application can actively generate diverse blockage training data during system debugging or idle phases, solving the problems of misjudgment or missed judgment caused by insufficient samples in the prior art. For example, by simulating the blockage forms corresponding to different material characteristics, the adaptability of the system to materials with different fluidity can be enhanced; by recording the dynamic parameter changes during the movement of the simulated blockage device 5, the mapping relationship between the blockage position and sensor data can be established, shortening the positioning time when an actual blockage occurs.

[0081] An automatic feeding system includes a feed bin 1, a pneumatic conveying pipeline 2, a gas source device, and a discharge bin 3, and further includes an adaptive cleaning module. When a severe dredging strategy is triggered or the cumulative conveying amount exceeds a preset mass threshold, the pipeline is switched to inject a reverse airflow into the pneumatic conveying pipeline 2 to perform reverse cleaning.

[0082] The adaptive cleaning module refers to a functional unit that automatically triggers a cleaning action according to the system operation state. Specifically, it can be realized by the linkage of a solenoid valve group and an airflow direction controller, used to switch the pipeline connection mode under specific conditions and change the airflow direction. The reverse airflow refers to the airflow opposite to the material conveying direction, which can be specifically realized by adjusting the output port of the gas source device or adding an independent back-blowing air pump, used to reverse-scour the deposited materials in the pipeline to the sending tank or the discharge port. Reverse cleaning refers to the operation of scouring the inner wall of the pipeline by the reverse airflow, which can specifically adopt a pulsed airflow or a continuous high-pressure airflow mode, used to remove the residual materials and potential blockages on the pipe wall.

[0083] When the system executes the severe dredging strategy, for example, there is still a risk of material residue after the rotation dredging mechanism intervenes, or when the cumulative conveying amount reaches the preset mass threshold, for example, the total conveying amount 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, so that the high-pressure airflow flows in the reverse direction along the pipeline. During this process, the airflow pressure can be dynamically adjusted according to the pipeline length. For example, a pressure sensor is used to provide real-time feedback and control the output power of the gas source device to ensure that the airflow speed during reverse cleaning is sufficient to peel off the material clumps attached to the pipe wall. After the cleaning is completed, the system automatically resumes the forward conveying state, avoiding the extension of downtime caused by manual intervention.

[0084] In the prior art, the cleaning operation relies on manual judgment of the blockage position and manual switching of pipelines, and usually only local dredging is carried out after the blockage occurs. However, in this solution, the full-pipeline cleaning is triggered by the linkage of the preset quality threshold and the dredging strategy, which can actively remove potential accumulations before the material is completely blocked. At the same time, the internal wear or residue risk of the pipeline is predicted by combining the cumulative conveying volume, so as to achieve preventive maintenance.

[0085] It can synchronously remove the dredging residues after a severe blockage occurs, avoiding the risk of secondary blockage. At the same time, it automatically triggers periodic cleaning according to the total amount of material conveyed, preventing the reduction of the pipe diameter or the decrease of fluidity due to long-term material accumulation on the pipe wall, thereby reducing the maintenance frequency and improving the stability of continuous operation.

[0086] It also includes a conveying efficiency evaluation module. The conveying efficiency evaluation module is connected to the blockage analysis module and generates optimization suggestions based on the historical blockage positions and frequencies, including: when the cumulative blockage times at the same position are greater than the preset blockage times, output a suggestion to increase the curvature radius of the pipeline bending section to 1-10 times the pipe diameter; when the wall amplitude of the pneumatic conveying pipeline 2 exceeds the preset safety threshold, mark it as a damaged pipe section and trigger a replacement suggestion; suggestions to add vibrators and temperature sensors at frequently blocked positions, and display the suggestions in the 3D modeling of the pneumatic conveying pipeline 2.

[0087] The conveying efficiency evaluation module refers to a logic unit that analyzes the system performance based on historical operation data. Specifically, data mining algorithms can be used to perform pattern recognition on the time series of blockage events, and optimization suggestions are generated by associating the relationship between pipeline structure parameters and blockage frequencies. This module provides data support for pipeline structure improvement by continuously learning the system operation status.

[0088] The cumulative blockage times at the same position refer to the statistical quantity of the number of blockage events occurring at the same physical coordinate within a set time period. Specifically, it can be achieved by obtaining position marking data through a distributed sensor group 4 and establishing a spatial coordinate system database. This parameter is used to judge whether there are structural design defects in a specific pipe section.

[0089] The suggestion to increase the curvature radius of the pipeline bending section to 1-10 times the pipe diameter refers to a structural optimization scheme for the elbow part. Specifically, computer-aided engineering software can be used to simulate and verify the fluid mechanics model at the elbow. This suggestion can reduce the flow resistance of the material at the elbow.

[0090] The conveying efficiency evaluation module establishes a time - series database of the pipeline operation status by continuously recording the location information and environmental parameters of each blockage event. When the blockage frequency of a specific pipe section exceeds the set threshold, the module automatically calls the three - dimensional modeling data of the pipeline, analyzes the correlation between structural parameters such as the bending radius and inclination angle at this location and the material flow characteristics, generates structure optimization suggestions. For pipe sections with abnormal vibration, it combines the frequency spectrum characteristics of the vibration acceleration sensor to identify the pipeline damage status and triggers a maintenance instruction. All optimization suggestions are visually displayed through the three - dimensional modeling interface, facilitating the operator to quickly locate the problem area.

[0091] In some specific embodiments, when it is detected that blockages occur three times consecutively 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 showing "It is recommended to expand the curvature radius from the current 50 mm to the range of 150 - 500 mm". In another embodiment, when the peak value of the pipe wall vibration acceleration exceeds 10 g, the system automatically generates a maintenance work order and pushes it to the equipment management department.

[0092] Existing dense - phase conveying systems usually only perform passive dredging after blockages occur and lack the ability to actively analyze pipeline structure defects. This solution can identify structural defects in easily blocked pipe sections in advance by establishing an association model between historical operation data and pipeline design, turning post - event processing into pre - event prevention and effectively reducing maintenance costs.

[0093] It can automatically generate a pipeline optimization plan based on actual operation data, avoid the limitations of manual experience judgment, reduce repeated blockage problems caused by unreasonable pipeline structures, and at the same time, through the visual suggestion display of three - dimensional modeling, significantly improve the defect recognition efficiency of maintenance personnel and ensure the long - term stable operation of the pneumatic conveying pipeline 2.

[0094] 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 enabled exceeds the preset safety range.

[0095] The operation parameter constraint module refers to a logic control unit used to limit the starting parameters of the pneumatic conveying system. Specifically, it can be implemented by the linkage of a programmable controller and sensors. Its function is to avoid the risk of pipeline blockage caused by human operation 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. Specifically, it can be calculated and generated by the correlation function of the material fluidity index and the pipeline diameter. Its function is to provide a dynamic parameter boundary for the system to prevent material accumulation caused by over-limit pressure or flow and interrupt the operation. When it is detected that the parameter exceeds the preset range, the operation is to immediately stop the starting instruction of the pneumatic conveying system. Specifically, it can be achieved by cutting off the power supply circuit of the air source device or closing the intake valve. Its function is to block potential risks at the source. Requesting operation confirmation means generating an artificial review request signal. Specifically, it can pop up a parameter abnormality warning on the human-machine interface and lock the operation permission. Its function is to reduce the probability of misoperation through a secondary confirmation mechanism.

[0096] Before the pneumatic conveying system starts, the operation parameter constraint module compares the wind pressure value and flow value set by the operator with the preset safety range respectively. 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 tolerable pressure corresponding to the current material, the module automatically triggers the interruption mechanism, pauses the start of the air source device, and at the same time displays a parameter over-limit warning at the control terminal. At this time, the operator needs to re-enter the parameters or adjust the material property data, and the lock can be released only after passing the system verification again. 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, asking to check the material filling rate or adjust the pressure parameter of the sending tank.

[0097] Traditional dense-phase conveying systems rely on manual experience to set operation parameters and lack a real-time verification mechanism, which is prone to blockage due to improper parameter settings. However, this solution completes the parameter compliance verification before the system starts through the preset safety range and the automatic interruption mechanism, avoiding material compaction caused by excessive pressure or local stagnation caused by insufficient flow, and reducing the probability of blockage from the source.

[0098] It effectively solves the problem of pipeline blockage caused by incorrect setting of operation parameters. Through the pre-inspection mechanism and the forced review process, it significantly reduces the abnormal shutdown of the system caused by human operation errors and ensures the stable operation of the pneumatic conveying system within the safe parameter range.

Claims

1. An automatic feeding system, comprising a feed bin, a pneumatic conveying pipeline, a gas source device and a discharge bin, characterized in that, It further includes: A distributed sensor group for monitoring the flux, pressure, temperature, and vibration spectrum at preset monitoring points of the pneumatic conveying pipeline, including a flow sensor, a pressure sensor, a temperature sensor, and a vibration acceleration sensor arranged at several preset monitoring points of the conveying pipeline; A digital twin model construction module that constructs a three-dimensional model of the pneumatic conveying pipeline and maps the data collected at several preset monitoring points of the pipeline distributed sensor group to the three-dimensional model in real time; A blockage analysis module that analyzes and marks the blockage position to the three-dimensional model in real time based on the data collected by the distributed sensor group, extracts the multi-parameter change curves within a period T before the blockage and constructs a blockage feature library, and generates a blockage evaluation level through the difference in the weight change rates of the feeding bin and the discharging bin; A hierarchical dredging execution module configured with different dredging strategies and activates different dredging strategies according to the blockage evaluation level; A blockage warning module connected to the blockage analysis module, which is used to calculate the dynamic similarity between the data of the distributed sensor group and the historical blockage curves in the blockage feature library in real time. When the dynamic similarity exceeds the preset matching threshold, a primary warning is generated. If the duration of the primary warning reaches the preset time, it is upgraded to a high-level warning to trigger the hierarchical dredging execution module.

2. An automatic feeding system according to claim 1, wherein The conditions for upgrading to a high-level warning in the blockage warning module further include: The real-time air pressure fluctuation amplitude exceeds the standard deviation of the historical curve; The continuous decline rate of the material flux is greater than the preset risk threshold.

3. An automatic feeding system according to claim 2, characterized in that, The feeding system further includes: A blockage simulation device, including a simulation blockage device arranged inside the pipeline and a moving component for driving the simulation blockage device to move. The simulation blockage device is used to form blockage situations with different blockage evaluation levels in the pipeline, and the moving component is used to drive the simulation blockage device to move along the inner wall of the pipeline; A blockage simulation module for controlling the blockage simulation device, driving the blockage simulation device to be in several positions of the pipeline, and driving the blockage simulation device to form several degrees of blockage evaluation levels. The multi-parameter change curves during the simulated blockage generated by the blockage simulation module are recorded in the blockage feature library through the blockage analysis module.

4. An automatic feeding system according to claim 3, characterized in that, The blockage simulator includes a blockage elastic sheet, inner magnetic rings arranged at both ends of the blockage elastic sheet, and moving outer magnetic rings magnetically coupled with the inner magnetic rings. The inner magnetic rings and the moving outer magnetic rings are respectively arranged inside and outside the pneumatic conveying pipeline, and there are two inner magnetic rings and two moving outer magnetic rings. Moving components for driving them to move along the pneumatic conveying pipeline are arranged on the moving outer magnetic rings; When the two inner magnetic rings approach each other, it drives the blockage elastic sheet to bend and deform, and controls the deformation degree of the blockage elastic sheet according to the distance between the two inner magnetic rings. Different deformation degrees of the blockage elastic sheet are used to simulate different blockage degrees of the pneumatic conveying pipeline.

5. An automatic feeding system according to claim 2, wherein The blockage warning module further includes a weight analysis unit for monitoring the parameters in real time to calculate the comprehensive risk value, and its calculation method is: Multiply the dynamic similarity between real-time sensor data and historical blockage characteristics by a first weight factor, add the ratio of the number of times the air pressure change rate in the monitoring window exceeds a set mutation threshold to the duration of the dynamic monitoring window multiplied by a second weight factor, and then add the ratio of the abnormal temperature duration to the running time of the current conveying task multiplied by a third weight factor; Among them, the first weight factor is greater than the second weight factor, the second weight factor is greater than the third weight factor, and the sum of the three weights is 1; When the comprehensive risk value exceeds the early warning threshold dynamically determined according to the pipeline structure parameters and material characteristics, an advanced early warning is triggered.

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

7. An automatic feeding system according to claim 3, characterized in that, The blockage assessment levels include: Mild blockage level, the determination method is: the difference in the weight change rates of the feeding bin and the discharging bin < preset value α; Moderate blockage level, the determination method is: preset value α ≤ the difference in the weight change rates of the feeding bin and the discharging bin < preset value β; Severe blockage level, the determination method is: the difference in the weight change rates of the feeding bin and the discharging bin ≥ preset value β.

8. An automatic feeding system according to claim 7, characterized in that, When the blockage early warning module issues an advanced early warning, trigger corresponding differential dredging strategies according to the blockage assessment level. The differential dredging strategies include: Mild dredging strategy, corresponding to the mild blockage level, equipped with a vibrator, trigger the vibrator at the blockage position to oscillate the target pipe section of the pneumatic conveying pipeline at a preset frequency; Moderate dredging strategy, corresponding to the moderate blockage level, switch the pipeline to inject reverse air flow into the pneumatic conveying pipeline, and the air pressure of the reverse air flow is not less than the forward conveying pressure; Severe dredging strategy, corresponding to the severe blockage level, equipped with a rotating dredging mechanism to extend into the blockage position, or drive a simulated blockage device to move through a moving component to dredge the triggered blockage position.

9. An automatic feeding system according to claim 8, wherein, The feeding system further includes an adaptive cleaning module. When the severe dredging strategy is triggered or the cumulative conveying volume exceeds a preset mass threshold, switch the pipeline to inject reverse air flow into the pneumatic conveying pipeline to perform reverse cleaning.

10. An automatic feeding system according to claim 1, wherein, It also includes a conveying efficiency evaluation module. The conveying efficiency evaluation module is connected to the blockage analysis module and generates optimization suggestions based on the historical blockage positions and frequencies, including: When the cumulative blockage times at the same position are greater than the preset blockage times, output a suggestion to increase the curvature radius of the pipeline bending section to 1-10 times the pipe diameter; When the wall amplitude of the pneumatic conveying pipeline exceeds the preset safety threshold, mark it as a damaged pipe section and trigger a replacement suggestion; Suggestion to add monitoring preset points and vibrators at frequently blocked positions; The optimization suggestions are displayed in the 3D modeling of the pneumatic conveying pipeline.

Images