Automatic edge material adjusting device for belt roaster

Abstract

The application discloses a kind of automatic adjusting device of edge material of belt roaster, it is related to the technical field of edge material distribution of belt roaster.The present application aims to solve the technical problems that uneven distribution of edge material, manual adjustment and real-time detection of distribution effect are needed in the prior art due to the deviation of trolley.The present application comprises: two range finders for measuring the distance from the trolley body sealing plate in real time;Control system for receiving measurement values and calculating the offset of the upper rail plate, and generating adjustment instructions according to the comparison results;Two fixed distribution hoppers are provided with guide shafts;Two movable distribution hoppers are slidingly arranged on the guide shafts;Two servo cylinders drive the corresponding movable distribution hoppers to move;Two radar level meters are arranged on the movable distribution hoppers for real-time detection of the actual height of the edge material and sending to the control system to trigger an alarm.The present application realizes online automatic adjustment of edge material position and real-time monitoring of distribution effect.

Description

Technical Field

[0001] This invention relates to the field of edge-laying material distribution technology for belt roasters. More specifically, this invention relates to an automatic edge-laying material adjustment device for belt roasters. Background Technology

[0002] Belt roasters are key equipment in the pellet production process, used to roast green pellets at high temperatures, solidifying them into finished pellets. During operation, upper guardrails are installed on both sides of the trolley to constrain the material layer boundaries. To prevent the upper guardrails from being directly exposed to high-temperature flue gas and causing overheating and burn-out, a layer of edge material is usually laid on the inside of the upper guardrails to form a heat insulation protective layer. The position and height of the edge material directly affect the protective effect of the upper guardrails.

[0003] Currently, the material feeding device for edge trimming typically employs a fixed structure, installed on the roasting machine frame. The relative position between the material feeding hopper's discharge port and the upper rail plate of the trolley is adjusted and aligned on-site during equipment installation through methods such as cutting and welding. However, in actual production, due to factors such as track wear, thermal expansion, and uneven material loading, the trolley may experience lateral deviation during operation, causing the position of the upper rail plate relative to the fixed material feeding hopper to change. When the deviation exceeds a certain range, misalignment occurs between the material feeding hopper's discharge port and the upper rail plate, preventing the edge trimming material from accurately falling inside the upper rail plate. This results in material segregation or even loss, leaving the upper rail plate unprotected by the edge trimming material and directly exposed to high-temperature heat radiation, making it susceptible to burning and deformation, thus affecting the equipment's service life and the roasting machine's operating rate.

[0004] To address the aforementioned issues, the current approach is as follows: when uneven distribution of the edge material or burnt upper plate is detected, the operator must stop the roasting machine, manually inspect for misalignment, and readjust the position of the material hopper through cutting, welding, or other methods. This method is reactive, the adjustment process is time-consuming, and requires production interruption, affecting the continuous operation of the roasting machine. Furthermore, since the misalignment gradually changes over time, a single adjustment is unlikely to be effective in the long term, requiring repeated shutdowns for correction, increasing maintenance workload and equipment downtime frequency.

[0005] On the other hand, during the edge fabric laying process, abnormalities such as blockages in the feed pipe and material caking may occur, resulting in the actual fabric laying height being lower than the process requirements, and the upper panel also faces the risk of exposure. Currently, the fabric laying effect mainly relies on manual inspection and observation, lacking real-time monitoring methods, making it difficult to detect fabric laying abnormalities in time. By the time problems are discovered, the upper panel has often already suffered some degree of burn damage.

[0006] Therefore, existing edge-laying material distribution devices suffer from problems such as untimely adjustments, reliance on manual intervention, and lack of real-time feedback in dealing with trolley deviation and abnormal material distribution, which affect the operational stability of the equipment and the service life of the upper panel. Summary of the Invention

[0007] One object of the present invention is to solve at least the above-mentioned problems and to provide at least the advantages that will be described later.

[0008] To achieve these objectives and other advantages according to the present invention, an automatic adjustment device for edge material of a belt roaster is provided, comprising: Two rangefinders are set at an interval relative to each other to measure the distance between themselves and the sealing plate of the trolley in real time; The control system is connected to two rangefinders to receive the measurement values ​​from the two rangefinders, calculates the offset and direction of the upper rail based on the measurement values, compares the absolute value of the upper rail offset with a preset action threshold, and generates an adjustment command based on the comparison result. The adjustment command includes the movement distance and movement direction. Two fixed fabric hoppers are set at intervals and opposite each other. The upper opening of each fixed fabric hopper is connected to the upstream discharge pipe, and its lower opening is sealed and connected to a flexible sealing connection pipe. The fixed fabric hopper is provided with a horizontal guide shaft perpendicular to the moving direction of the trolley. A movable fabric hopper is slidably mounted on the guide shaft. The upper opening of the movable fabric hopper is sealed and connected to the sealing connection pipe, and its lower opening is used to insert the edge material into the inner side of the upper panel of the trolley. Two servo electric cylinders are respectively mounted on two fixed fabric hoppers. Their push rod ends are parallel to the guide shaft and connected to the corresponding movable fabric hoppers. The servo electric cylinders are connected to the control system signal to receive adjustment commands and drive the movable fabric hoppers to move. Two radar level gauges are installed on two movable material hoppers and are connected to the control system. The probes of the radar level gauges are pointed downwards at the area where the edge material is falling from the movable material hoppers to detect the actual height of the edge material in real time and send the data to the control system. The control system compares the actual height with the preset nominal material level. When the actual height deviates from the nominal material level by more than the preset alarm threshold, an alarm is triggered.

[0009] Preferably, the rangefinder is a laser rangefinder, and the sealing connecting pipe has a fish-scale sealing structure.

[0010] Preferably, the servo electric cylinder has a built-in encoder, which is used to feed back the actual displacement signal of the servo electric cylinder to the control system.

[0011] Preferably, the action threshold is 2mm.

[0012] Preferably, the control system is used to compare the actual height of the edge material detected by the radar level gauge with the nominal level height. When the actual height is lower than the nominal level height and the difference exceeds the lower limit threshold, an edge material shortage alarm is triggered.

[0013] Preferably, there are two guide shafts arranged in parallel, with both ends fixed to the fixed fabric hopper, and the movable fabric hopper is mounted on the two guide shafts via linear bearings.

[0014] Preferably, the control system includes: The deviation calculation unit is connected to the signals of two laser rangefinders and is used to calculate the deviation of the upper guardrail in real time based on the measurement values ​​of the two laser rangefinders. The basic position command generation unit is connected to the deviation calculation unit and is used to generate a basic position command based on the offset of the upper panel. The basic position command corresponds to the position that aligns the movable hopper with the upper panel of the trolley. The material height deviation calculation unit is connected to the signals of two radar level gauges and is used to calculate the material height deviation in real time based on the difference between the actual edge material height detected by the radar level gauges and the preset nominal material level height. The final position command generation unit is signal-connected to the basic position command generation unit and is used to output the basic position command as the final position command. The servo drive unit is connected to the final position command generation unit and the servo cylinder signal respectively. It is used to drive the servo cylinder according to the final position command, control the movable cloth bucket to move to the target position, and form a position closed-loop control. The material height deviation calculation unit is connected to the alarm module in the control system and is used to trigger an alarm when the material height deviation exceeds a preset threshold.

[0015] Preferably, the control system includes a deviation trend prediction unit, comprising: The data storage module is used to continuously record the historical measurement values ​​of the two laser rangefinders and form a time series data of deviation corresponding to the trolley running time; The first prediction model is an autoregressive moving average model based on time series analysis, which is used to predict the trolley deviation at a preset time point in the future based on the deviation time series data, and outputs the first prediction value and its corresponding first confidence level. The second prediction model is a neural network prediction model based on historical data. The neural network is pre-trained with historical deviation data and is used to predict the deviation of the trolley at a preset time point in the future based on the deviation time series data, and outputs the second prediction value and its corresponding second confidence level. The fusion decision module is connected to the first prediction model, the second prediction model, and the laser rangefinder signal, respectively. It is used to receive the first prediction value, the second prediction value, and the real-time deviation amount at the current moment. It also performs weighted fusion of the first prediction value and the second prediction value based on the first confidence level and the second confidence level to calculate the pre-adjustment displacement amount. The pre-adjustment displacement amount is less than the action threshold and the direction is consistent with the predicted deviation trend. In this system, when the absolute value of the real-time deviation does not exceed the action threshold, but the absolute value of the deviation will exceed the action threshold at a future preset time point based on the pre-adjusted displacement, the system outputs a pre-adjustment command to the servo cylinder in advance before the future preset time point arrives, and controls the movable cloth bucket to follow the deviation trend direction according to the pre-adjusted displacement.

[0016] Preferably, the control system has a self-learning calibration module, including: The calibration trigger unit is used to generate a calibration start signal during initial installation or after receiving a self-learning calibration command from an external input. The stroke scanning unit, connected to the calibration trigger unit and the servo cylinder signal, is used to control the servo cylinder to drive the movable fabric hopper to move along the guide shaft from the initial position in the first direction after receiving the calibration start signal, until the movable fabric hopper touches the first mechanical limit position, and record the first encoding value of the encoder built into the servo cylinder at this time; then, it controls the servo cylinder to drive the movable fabric hopper to move in the reverse direction until the movable fabric hopper touches the second mechanical limit position, and records the second encoding value of the encoder at this time; and calculates the actual total stroke of the movable fabric hopper and the encoder midpoint value corresponding to the stroke midpoint based on the first encoding value and the second encoding value. The reference calibration unit is connected to the laser rangefinder and the stroke scanning unit respectively, and is used to obtain the current distance value measured by the laser rangefinder as the laser reference value when the movable fabric hopper is located at the midpoint of the stroke. The zero-point correction unit is connected to the reference calibration unit and the stroke scanning unit. It is used to set the encoder midpoint value as the mechanical zero point of the movable cloth hopper and set the laser reference value as the measurement reference of the laser rangefinder. It calculates the deviation between the mechanical zero point and the preset electrical zero point and generates zero-point correction parameters. Non-volatile memory, connected to the zero-point correction unit signal, is used to store zero-point correction parameters; After the self-learning calibration mode ends, when the control system performs position control of the movable hopper, it calls the zero-point correction parameters in the non-volatile memory to compensate for the displacement commands sent to the servo cylinder.

[0017] Preferably, the control system includes an abnormal operating condition diagnosis and self-processing module, including: The data acquisition unit is connected to the encoder signal built into the radar level gauge and servo cylinder to collect the actual edge material height, the actual position of the servo cylinder, and the current deviation measured by the laser rangefinder in real time. The material shortage alarm triggering unit is connected to the data acquisition unit and is used to compare the actual edge material height with the preset nominal material level height. When the actual edge material height is lower than the nominal material level height and the difference exceeds the preset lower limit threshold, an edge material shortage alarm signal is generated. The status judgment unit is connected to the material shortage alarm trigger unit and the data acquisition unit. After receiving the edge material shortage alarm signal, it is used to obtain the current deviation amount and compare it with the preset action threshold. At the same time, it obtains the deviation between the actual position of the servo cylinder and the basic position command. If the absolute value of the current deviation amount is less than the action threshold and the position deviation is less than the preset position allowable error, it is determined that the material shortage is not caused by the trolley deviation and a suspected blockage signal is generated. The unblocking execution unit is connected to the status judgment unit. It is used to output a vibration unblocking command to the vibrator installed on the feed pipe after receiving a suspected blockage signal, and control the vibrator to start and stop according to the preset mode. The monitoring unit is connected to the data acquisition unit and the dredging execution unit respectively. It is used to continuously monitor the change in the actual edge material height after the dredging execution unit outputs the vibration dredging command, and start the timer. An alarm escalation unit, connected to the recovery monitoring unit, generates a severe alarm signal and sends it to the operator station when the timer reaches a preset recovery time threshold, but the actual edge material height is still lower than the nominal material level height and the difference exceeds the lower limit threshold. This invention provides at least the following advantages: First, this invention uses two rangefinders to detect the distance between the sealing plates on both sides of the trolley in real time. This allows the control system to accurately calculate the offset of the upper guardrail and generate adjustment commands, driving a servo cylinder to automatically move the movable fabric hopper along the guide shaft. This achieves dynamic alignment between the fabric hopper and the upper guardrail after deviation. Compared to existing technologies that require manual cutting, welding, and adjustment of the fabric hopper, this solution allows for real-time online response to trolley deviation without interrupting production, fundamentally solving the problem of upper guardrail burn-out due to untimely adjustments. Simultaneously, a radar level gauge on the movable fabric hopper monitors the height of the fabric in real time, changing the outdated manual inspection method and enabling online monitoring and alarm for abnormalities in the fabric application process. This significantly improves the automation level and operational reliability of the equipment.

[0018] Secondly, this invention triggers an alarm for missing edge material when the actual height of the edge material is lower than the nominal material level and the difference exceeds a lower threshold. This solution solves the problem in existing technologies where alarms based solely on excessive height cannot fully cover abnormal operating conditions, expanding the alarm logic from a single excessive height to the more critical fault type of insufficient height. In actual production, missing edge material means the upper panel is directly exposed to high-temperature heat radiation, making it highly susceptible to burning and deformation. By setting a lower threshold alarm, this invention can issue a warning at the first sign of missing edge material, preventing damage to the upper panel due to delayed detection of material shortages, demonstrating a profound understanding of production process safety.

[0019] Third, this invention upgrades the control logic from traditional feedback control to predictive control by setting up a deviation trend prediction unit. This unit employs a prediction algorithm that combines an autoregressive moving average model and a neural network model to comprehensively analyze the linear and nonlinear characteristics of the trolley deviation trend, improving the accuracy of deviation prediction under complex working conditions. When the real-time deviation is within the limit but is predicted to exceed it in the future, the control system outputs a small pre-adjustment command in advance, enabling the movable hopper to continuously micro-move in accordance with the deviation trend. This solution solves the "catching up" state and frequent start-stop shock problems caused by lagging adjustments based solely on real-time deviation in existing technologies, achieving dynamic following between the hopper and the upper guardrail, and significantly improving the stability of equipment operation.

[0020] Fourth, this invention, by incorporating a self-learning calibration module, endows the equipment with automatic calibration capabilities, solving the zero-point drift problem caused by mechanical vibration, temperature changes, and component wear during long-term operation. This module automatically locates mechanical limits and calculates the midpoint of the stroke to establish a precise mechanical zero-point reference through a stroke scanning unit. A reference calibration unit acquires a laser reference value at the midpoint of the stroke to eliminate the influence of sensor installation position changes. A zero-point correction unit generates correction parameters and stores them in non-volatile memory to ensure parameter retention after power failure. This solution fundamentally solves the maintenance problem of complex manual centering calibration required after replacing servo cylinders or laser rangefinders, ensuring the control accuracy and reliability of the automatic edge-laying material adjustment device throughout its entire lifecycle.

[0021] Other advantages, objectives and features of the present invention will become apparent in part from the following description, and in part from those skilled in the art through study and practice of the invention. Attached Figure Description

[0022] Figure 1 This is a side view of the present invention.

[0023] Figure 2 This is a schematic diagram of another aspect of the structure of the present invention.

[0024] Figure 3This is a partially enlarged structural schematic diagram of the present invention.

[0025] Figure 4 This is a flowchart illustrating the automatic adjustment of the edge material in the belt roasting machine of the present invention.

[0026] Figure 5 This is a control logic block diagram of the self-learning calibration module of the present invention.

[0027] The following are the reference numerals in the instruction manual: 1. Trolley; 2. Rangefinder; 3. Carriage sealing plate; 4. Upper sideboard; 5. Fixed hopper; 6. Discharge pipe; 7. Sealing connection pipe; 8. Guide shaft; 9. Movable hopper; 10. Radar level gauge; 11. Servo electric cylinder. Detailed Implementation

[0028] The present invention will now be described in further detail with reference to examples, so that those skilled in the art can implement it based on the description.

[0029] It should be noted that, unless otherwise specified, the experimental methods described in the following implementation plan are all conventional methods, and the reagents and materials described are all commercially available unless otherwise specified.

[0030] like Figures 1 to 5 As shown, this invention provides an automatic adjustment device for edge material in a belt roasting machine. During the operation of the roasting machine, it automatically adjusts the position of the edge material feeding hopper based on the deviation of the trolley 1, and monitors and alarms the material feeding height in real time. The device includes: two laser rangefinders 2, a control system, two fixed feeding hoppers 5, two movable feeding hoppers 9, two servo electric cylinders 11, and two radar level gauges 10.

[0031] Two laser rangefinders are fixedly mounted on the roasting machine frame, symmetrically arranged, and respectively facing the left and right side sealing plates 3 of the trolley. The sealing plates are fixedly connected to the sides of the trolley and are located on the same trolley body as the upper panel 4. Therefore, the lateral displacement of the sealing plates represents the lateral displacement of the upper panel. The left laser rangefinder is denoted as LC, and the right laser rangefinder is denoted as RC. Both are used to measure the distance from themselves to the corresponding side sealing plate in real time and transmit the measured values ​​to the control system in real time.

[0032] The control system is connected to two laser rangefinders to receive real-time measurements from LC and RC sensors. The control system includes a calculation unit, a comparison unit, and a command generation unit to calculate the upper panel offset and direction based on the measured values, and to generate adjustment commands according to preset logic. Simultaneously, the control system is also connected to a radar level gauge and a servo cylinder to receive feedback signals and output control commands.

[0033] Two fixed feeding hoppers are symmetrically arranged on the roasting machine frame, and their positions are fixed. The upper opening of each fixed feeding hopper is connected to the upstream discharge pipe 6, and the discharge pipe is equipped with a discharge valve to control the material flow. The lower opening of the fixed feeding hopper is sealed and connected to the upper opening of the movable feeding hopper through a flexible sealing connection pipe 7. The movable feeding hopper is slidably mounted on the guide shaft 8, and its lower opening is aligned with the inner side of the upper rail plate of the trolley, for spreading the edge material into the inner side of the upper rail plate to form a protective layer.

[0034] The guide shafts are horizontally positioned perpendicular to the direction of the trolley's movement, serving to guide the lateral movement of the movable fabric hopper and bear its load. In this embodiment, there are two guide shafts, arranged in parallel, with both ends fixed to the outer wall of the fixed fabric hopper via supports. Linear bearings are mounted on the sides of the movable fabric hopper, sleeved on the two guide shafts, allowing the movable fabric hopper to slide smoothly along the guide shafts, preventing jamming or shaking.

[0035] Two servo electric cylinders are respectively mounted on the sides of two fixed fabric hoppers, with their push rod ends parallel to the guide shaft and fixedly connected to the corresponding movable fabric hoppers via connectors. The servo electric cylinders are connected to the control system to receive adjustment commands and drive the movable fabric hoppers to move laterally along the guide shaft. Each servo electric cylinder has a built-in encoder, which feeds back the actual displacement signal of the servo electric cylinder to the control system in real time, forming a closed-loop position control to ensure that the movable fabric hoppers move precisely to the target position.

[0036] A fish-scale seal is installed at the gap between the fixed and movable fabric hoppers, serving as a flexible sealing connection. The fish-scale seal is composed of multiple layers of flexible metal sheets overlapping each other. One side is fixed to the outer wall of the fixed fabric hopper by a pressure plate and bolts, while the other side naturally conforms to the outer wall of the movable fabric hopper. When the movable fabric hopper moves, the fish-scale seal undergoes elastic deformation but always covers the gap between the two, achieving a dynamic seal and preventing material spillage under dusty conditions.

[0037] Two radar level gauges are fixedly installed on the outer walls of two movable material hoppers, with their probes pointing vertically downwards or slightly tilted downwards, aimed at the area where the edge material is falling from the movable material hoppers. The radar level gauges are connected to the control system to detect the actual height of the edge material pile in real time and send the detected value to the control system. The left radar level gauge is denoted as HL, and the right one as HR.

[0038] The control system executes the following logic according to the preset program: 1. Calculation of deviation and determination of direction The control system receives the measurement values ​​from the left laser rangefinder LC and the right laser rangefinder RC in real time, and calculates the offset C of the upper panel according to formula (1): (1) Where C>0 indicates that the trolley veers to the right (LC side), and C<0 indicates that the trolley veers to the left (RC side). This represents the absolute value of the deviation.

[0039] 2. Action threshold comparison and instruction generation The control system has a preset action threshold of 2 mm. The action threshold is compared, and an adjustment instruction is generated based on the comparison result. when When the deviation is considered to be within the allowable range, a no-action command is generated, and the servo cylinder remains in its original position. When C>0 and At that time, a rightward movement command is generated, driving the two movable cloth hoppers to move to the right by Cmm; When C<0 and At that time, a leftward movement command is generated, driving the two movable cloth hoppers to move to the left. mm.

[0040] 3. Fabric height detection and alarm logic The control system receives the real-time detection values ​​HL and HR from the radar level gauge and compares them with the preset nominal level height H: When HL - H>0 or HR - H>0, it means that the actual height of the edge material is higher than the nominal value, which may be due to the risk of overflow due to excessive material accumulation, and the control system will trigger an over-high alarm. When H - HL > preset lower threshold or H - HR > preset lower threshold, it indicates that the actual edging material height is lower than the nominal value, posing a risk of missing edging material. The upper panel may be directly exposed to the high-temperature zone, triggering an edging material shortage alarm in the control system. The lower threshold can be set according to process requirements, for example, 10% to 20% of H.

[0041] During operation of the belt roaster, the trolley moves continuously, and the edge material is distributed from the feed pipe through the fixed and movable feed hoppers into the inner side of the upper plate of the trolley. During this process, the device performs the following automatic adjustment and monitoring steps: 1. Real-time detection of deviation The left-side laser rangefinder LC and the right-side laser rangefinder RC continuously measure the distance to the vehicle body sealing plate and send the measured values ​​LC and RC to the control system in real time.

[0042] 2. Calculate the deviation and generate adjustment instructions. The control system calculates the C value according to formula (1), and based on... The comparison result with 2 mm and the sign of C generate the corresponding adjustment command. For example, if LC = 98 mm and RC = 102 mm, then C = -2 mm, indicating that the trolley deviates 2 mm to the left, which satisfies... And since C < 0, the control system generates a command to move 2 mm to the left.

[0043] 3. Perform movement adjustments The control system sends adjustment commands to two servo cylinders. The push rods of the servo cylinders extend or retract according to the commands, pushing the movable fabric hopper smoothly along the guide shaft. During the movement, the fish-scale seal maintains a tight seal between the fixed and movable fabric hoppers, preventing dust leakage. Once in position, the encoders built into the servo cylinders feed back the actual displacement to the control system. After confirming the position is correct, the cylinders are locked, realigning the discharge port of the movable fabric hopper with the misaligned upper panel.

[0044] 4. Real-time monitoring of fabric height During the fabric placement process, radar level gauges fixed to the movable fabric hopper continuously measure the height HL and HR of the edge-laying material pile and send the data to the control system. The control system compares the measured height with the nominal material level height H. Once an abnormal height (too high or too low) is detected that exceeds the preset range, the corresponding alarm is immediately triggered, prompting the operator to check whether the discharge pipe is blocked, whether the hopper is empty, or whether the adjustment is in place.

[0045] Through the aforementioned closed-loop control of detection-execution-feedback, this device can automatically and promptly adjust the position of the edge material hopper when the trolley deviates from its course, ensuring that the edge material is always accurately placed inside the upper plate. At the same time, it monitors the material placement effect in real time to avoid damage to the upper plate due to missing material. This effectively solves the problems of needing to stop the machine for manual adjustment and lacking real-time feedback in the existing technology, thereby improving the operating rate of the roasting machine and the service life of the upper plate.

[0046] In another embodiment of the present invention, during the operation of the belt roaster, trolley misalignment and abnormal material distribution height are two independent but potentially simultaneous abnormal conditions. Coupled control of these two conditions may cause the servo cylinder to erroneously respond to material height deviations during misalignment adjustment, resulting in ineffective shaking of the material distribution hopper position and even exacerbating uneven material distribution. To solve the above technical problems, this embodiment provides a dual closed-loop decoupling control system, specifically including the following units: The deviation calculation unit is connected to the signals of two laser rangefinders and is used to calculate the deviation based on the signal from the left laser rangefinder L. C And the laser rangefinder R on the right C The real-time measured value is used to calculate the offset C of the upper panel according to formula (1).

[0047] The basic position command generation unit is signal-connected to the deviation calculation unit, and is used to generate basic position commands based on the upper panel offset C. P base This instruction corresponds to the target position where the movable fabric hopper is aligned with the upper sideboard of the trolley, i.e.: In the formula, P 0 represents the initial centering position of the movable fabric hopper (determined by the self-learning calibration module).

[0048] The material height deviation calculation unit is connected to the signals of two radar level gauges and is used to calculate the material level deviation based on the signals from the left radar level gauge. H L and the radar level gauge on the right H R The detected actual edge material height, compared to the preset nominal material height. H nom Compare the results and calculate the material height deviation in real time according to formula (3). : The material height deviation calculation unit is connected to the alarm module in the control system. When or Exceeding the preset alarm threshold (e.g., ±10%) H nom When the position is too high or missing, the corresponding alarm (such as an over-temperature alarm or a missing alarm) is immediately triggered. This alarm is independent of the position control loop and does not participate in the generation of displacement commands for the servo cylinder.

[0049] The final position command generation unit is signal-connected to the basic position command generation unit, and is used to generate the basic position command. P base Directly used as the final position instruction P final Output, i.e.: P final = P base This unit does not compensate for or correct material height deviations, ensuring that deviation adjustments are unaffected by material height fluctuations.

[0050] The servo drive unit is connected to both the final position command generation unit and the servo electric cylinder signal unit, and is used to receive the final position command. P final It drives the servo electric cylinder to move the movable fabric hopper to the target position. The servo electric cylinder has a built-in encoder that provides real-time feedback on the actual position. P act The servo drive unit adjusts according to the deviation. P final - P act PID control is applied to form a closed-loop position control, ensuring that the feed hopper accurately follows the deviation.

[0051] Through the above-described dual-closed-loop decoupling control architecture, the present invention achieves the following technical effects: Misalignment control and material height monitoring operate independently: the misalignment calculation unit and the basic position command generation unit form an outer position control loop to ensure that the material hopper is always aligned with the upper guardrail; the material height deviation calculation unit independently monitors the material distribution effect and is only used for alarm triggering, without intervening in position adjustment, thus avoiding unnecessary adjustments caused by material height fluctuations.

[0052] Eliminating control loop coupling interference: During the deviation adjustment process, even if the material height fluctuates instantaneously, the target position of the feeding hopper will not change, ensuring the accuracy of the adjustment action and the stability of the system.

[0053] Improve fault location efficiency: When an alarm is triggered, since there is no control coupling between material height deviation and deviation amount, operators can quickly determine the source of the abnormality (such as excessive deviation or material blockage), which facilitates precise intervention.

[0054] For example, during normal operation of the calcining machine, if the laser rangefinder on the left measures... L C = 102mm, right side R C = 98mm, then the deviation calculation unit obtains C = 2mm, basic position command generation unit output P base = P 0 + 2mm, the final position command generation unit directly forwards the command, and the servo drive unit drives the movable hopper to move 2mm to the right. Simultaneously, if the left-side radar level gauge detects... H L Below H nom If the deviation reaches 15%, the material deviation calculation unit immediately triggers an alarm for missing edge material, but this alarm will not change. P final The movable fabric hopper remains in its adjusted position, thus simultaneously achieving deviation compensation and material shortage warning, avoiding the drawback of deviation adjustment being interfered with by high material alarm.

[0055] In another embodiment of the present invention, the control system includes a deviation trend prediction unit to achieve predictive adjustment of the trolley deviation.

[0056] During long-term continuous operation of a belt roaster, trolley misalignment is often not a sudden change, but rather a gradual trend. For example, with the accumulation of track wear, changes in equipment thermal expansion, or fluctuations in material load, the trolley may slowly drift to one side over a period of time. Threshold-based feedback control only adjusts after the misalignment exceeds the limit, which is a reactive adjustment. This results in the movable feeding hopper always being in a "catching up" state, leading to brief periods of material mismatch during continuous misalignment. Furthermore, frequent start-stop adjustments can impact the mechanical structure and affect the equipment's lifespan. To address these issues, this implementation uses a predictive control mechanism to enable the system to anticipate changes and achieve dynamic following between the feeding hopper and the upper slab.

[0057] In addition to performing basic deviation calculations and generating adjustment commands, the control system also integrates a deviation trend prediction unit, which consists of the following modules: The data storage module is connected to two laser rangefinders to continuously record their historical measurements and calculate the deviation in real time based on these measurements. Specifically, the control system uses the formula... The offset C of the upper panel is calculated in real time, where LC is the measurement value of the left laser rangefinder and RC is the measurement value of the right laser rangefinder. The data storage module records the value of C at a fixed sampling period (e.g., every 100 ms), forming a time series data of deviation corresponding to the trolley running time, and stores it in a circular buffer, retaining the most recent N data points (e.g., N=1000) for use by the prediction model.

[0058] The first prediction model is signal-connected to the data storage module to receive time-series data on deviation. This model is an autoregressive moving average (ARMA) model based on time-series analysis. The ARMA model utilizes the autocorrelation and moving average characteristics of historical data to model the linear variation of deviation, effectively capturing the periodic components and short-term inertia of the deviation trend. Based on current and historical deviation data, the first prediction model predicts the trolley deviation at a preset future time point (e.g., 5, 10, or 15 seconds) and outputs a first predicted value and its corresponding first confidence level. The first confidence level is calculated based on the model's goodness of fit, for example, determined by the prediction interval width or model residual variance. The confidence level ranges from 0 to 1, with higher values ​​indicating more reliable prediction results.

[0059] The second prediction model is signal-connected to the data storage module and is also used to receive time-series data on deviation. This model is a neural network prediction model based on historical data, employing a multi-layer feedforward neural network (such as a backpropagation neural network) or a recurrent neural network (such as an LSTM). This neural network is pre-trained with a large amount of historical deviation data and corresponding actual deviation results. The training data covers deviation patterns under different operating conditions, including monotonic deviations caused by track wear, periodic fluctuations caused by thermal expansion, and random disturbances caused by material misalignment. The trained neural network can extract nonlinear features from the time series and predict the deviation of the trolley at a preset future time point. The second prediction model outputs a second predicted value and its corresponding second confidence level. The second confidence level can be estimated based on the probability distribution of the neural network output or the validation set error, and its value ranges from 0 to 1.

[0060] The fusion decision module is connected to the first prediction model, the second prediction model, and the laser rangefinder signal, respectively, to receive the first predicted value, the second predicted value, and the real-time deviation. The fusion decision module performs a weighted fusion of the first and second predicted values ​​based on a first confidence level and a second confidence level to calculate the pre-adjustment displacement. The weighted fusion uses the following formula: in, P This represents the predicted deviation after fusion. P 1 and P 2 represents the first predicted value and the second predicted value, respectively. w 1 and w 2 represents the first confidence level and the second confidence level, respectively. The fusion decision module further determines the pre-adjustment displacement based on the predicted deviation trend. D The direction of this displacement is consistent with the predicted deviation trend, and its absolute value is less than the preset action threshold. In this embodiment, the action threshold is set to 2 mm, therefore the pre-adjusted displacement satisfies... Specifically, the pre-adjusted displacement can be extracted from the fused prediction values ​​according to the scaling factor k, i.e. , where 0 < k <1, and must ensure that under any circumstances Not exceeding 1.5 mm (with a safety margin). For example, it can be taken as follows: k = 0.5, and set a limiting condition: if Then take .

[0061] The control system continues to monitor the prediction results as long as the absolute value of the real-time deviation does not exceed the action threshold. When based on the fused prediction value... PWhen the system determines that the absolute value of the deviation will exceed the action threshold at a future preset time point (e.g., 5 seconds later), it outputs a pre-adjustment command to the servo cylinder before the preset time point arrives, controlling the movable cloth hopper to move according to the pre-adjusted displacement. D Make movements that follow the direction of the deviation trend.

[0062] The specific judgment condition is as follows: Let the current time be... t 0, real-time deviation is C 0, and The integrated decision-making module predicts the future. T The deviation amount after 5 seconds (T is a preset time, which can be adjusted according to the rate of change of the trolley deviation, for example, 5 s, 10 s or 15 s) is P .like If this occurs, the system determines that an over-range deviation is imminent and immediately initiates pre-adjustment. Pre-adjustment displacement. D direction and P The signs are the same, meaning it moves in the predicted direction of deviation.

[0063] After the pre-adjustment command is sent to the servo cylinder, the movable fabric hopper slowly moves along the guide shaft to a new position, fine-tuning the relative position of the fabric hopper and the upper plate. Because the pre-adjustment displacement is small and its direction aligns with the trend, this action does not cause drastic changes in the fabric position, but rather achieves smooth following. When the deviation actually exceeds the threshold, the system will still perform a formal adjustment according to the logic described above, but because it has already been pre-moved, the magnitude of the formal adjustment is reduced, the action is smoother, and the impact on the mechanical structure is minimized.

[0064] Through the aforementioned predictive adjustment mechanism, this invention solves the problem of lagging adjustment based solely on real-time deviation in existing technologies. First, it employs a fusion of ARMA and neural network models for prediction, taking into account both the linear and nonlinear characteristics of the deviation trend, thus improving prediction accuracy and avoiding misjudgments by a single model under complex working conditions. Second, it performs small-scale pre-adjustments before the deviation exceeds the limit, enabling the movable feeding hopper to continuously micro-move in accordance with the deviation trend, reducing the impact on the mechanical structure caused by frequent start-stop adjustments. Finally, this solution is particularly suitable for working conditions where the trolley deviation exhibits a gradual trend, significantly improving the following performance of the edge material feeding hopper, ensuring that the edge material is always accurately placed inside the upper plate, thereby effectively protecting the upper plate from high-temperature burning and improving the operating rate and service life of the roasting machine.

[0065] In another embodiment of the present invention, the control system includes a self-learning calibration module to achieve automatic calibration of the zero-point drift of the equipment.

[0066] During long-term continuous operation of a belt roaster, factors such as mechanical vibration, temperature changes, and component wear may cause slight loosening or deformation at the mechanical connection between the movable feeding hopper and the servo cylinder. This results in a deviation between the position signal fed back by the encoder built into the servo cylinder and the actual position of the movable feeding hopper, i.e., zero-point drift. Furthermore, when components such as the servo cylinder and laser rangefinder are replaced due to malfunction, complex manual alignment and calibration operations are usually required on-site, which is time-consuming and difficult to guarantee in terms of accuracy. To address these issues, this implementation uses a self-learning calibration module to enable the system to automatically calibrate, ensuring control accuracy during long-term operation.

[0067] In addition to performing basic deviation calculations and generating adjustment commands, the control system also integrates a self-learning calibration module, which consists of the following units: calibration trigger unit, stroke scanning unit, reference calibration unit, zero-point correction unit, and non-volatile memory.

[0068] The calibration trigger unit generates a calibration start signal during initial installation or upon receiving an externally input self-learning calibration command. Specifically, the calibration trigger unit is connected to the operator station signal, allowing operators to input self-learning calibration commands through the human-machine interface. Simultaneously, the calibration trigger unit can also automatically generate a calibration start signal when the equipment is first powered on or when a replacement signal for a critical component (such as a servo cylinder or laser rangefinder) is detected.

[0069] The stroke scanning unit is connected to the calibration trigger unit and the servo cylinder signal, and is used to execute the stroke scanning program after receiving the calibration start signal. The stroke scanning unit first controls the servo cylinder to drive the movable fabric hopper to move along the guide shaft from its current position in a first direction (e.g., to the left) until the movable fabric hopper touches the first mechanical limit position. This first mechanical limit position is defined by a mechanical stop or limit switch located on the left side of the fixed fabric hopper. When the movable fabric hopper touches the first mechanical limit position, the servo cylinder drive current increases or the limit switch is triggered, and the stroke scanning unit records the first encoded value of the servo cylinder's built-in encoder at this time. E 1. Subsequently, the stroke scanning unit controls the servo electric cylinder to drive the movable fabric hopper to move in the reverse direction, i.e., to the second direction (e.g., to the right), until the movable fabric hopper touches the second mechanical limit position, and records the second encoding value of the encoder at this time. E 2. The travel scanning unit scans the data according to the first encoded value. E 1 and second encoded value E 2. Calculate the actual total stroke of the movable fabric hopper. and the encoder midpoint value corresponding to the midpoint of the travel. E mid : The reference calibration unit is connected to both the laser rangefinder and the stroke scanning unit. It is used to acquire the current distance value measured by the laser rangefinder as the laser reference value when the movable fabric hopper is at the midpoint of its stroke. Specifically, after calculating the encoder midpoint value, the stroke scanning unit controls the servo cylinder to drive the movable fabric hopper to the position corresponding to the encoder midpoint value. E mid The location is the midpoint of the journey. At this point, the left and right laser rangefinders measure their distances to the corresponding side vehicle body sealing plates, and the reference calibration unit obtains the measurement value from the left laser rangefinder at that moment. L ref And the measurement value of the laser rangefinder on the right R ref This serves as the laser reference value for subsequent control.

[0070] The zero-point correction unit is connected to the reference calibration unit and the stroke scanning unit to transmit the encoder midpoint value. E mid Set as the mechanical zero point of the movable fabric hopper and set the laser reference value. L ref , R ref This is set as the measurement reference for the laser rangefinder. The zero-point correction unit further calculates the deviation between the mechanical zero point and the preset electrical zero point. The electrical zero point is the coded value corresponding to the theoretical zero-point position set during initial equipment installation. E zero The zero-point correction unit calculates the zero-point deviation Δ. E : It also generates zero-point correction parameters, which include deviation values. And the direction of correction.

[0071] The non-volatile memory is connected to the zero-point correction unit and is used to store zero-point correction parameters. The non-volatile memory uses EEPROM or Flash memory to ensure that the parameters are not lost after the system is powered off. After the self-learning calibration mode ends, when the control system performs position control of the movable hopper, it calls the zero-point correction parameters in the non-volatile memory to compensate for the displacement commands sent to the servo cylinder.

[0072] After receiving the calibration start signal, the self-learning calibration module performs automatic calibration according to the following steps: 1. Stroke Scan: The stroke scanning unit controls the servo electric cylinder to drive the movable fabric hopper to move to the left until it touches the left mechanical limit, and records the first coded value. E 1; Then control the movable fabric hopper to move to the right until it touches the right mechanical limit switch, and record the second code value. E2; Calculate the total stroke S and encoder midpoint value. E mid .

[0073] 2. Reference Calibration: The stroke scanning unit drives the movable fabric hopper to the encoder midpoint value. E mid At the corresponding position, the reference calibration unit reads the measurement value from the laser rangefinder on the left. L ref And the measurement value of the laser rangefinder on the right R ref It is stored as a laser reference value.

[0074] 3. Zero point correction: The zero point correction unit reads the preset electrical zero point code value. E zero Calculate the zero-point deviation Generate zero-point correction parameters.

[0075] 4. Parameter storage: Store zero-point correction parameters (including deviation values). Laser reference value L ref , R ref Write to non-volatile memory.

[0076] 5. Exit Calibration: After calibration is completed, the control system exits the self-learning calibration mode and enters the normal operation mode.

[0077] After the self-learning calibration mode ends, the control system enters the normal operation mode. When the control system calculates the offset C of the upper panel based on the measurement value of the laser rangefinder and generates an adjustment command, it first reads the zero-point correction parameter from the non-volatile memory to compensate for the displacement command.

[0078] Specifically, let the theoretical adjustment displacement calculated by the control system according to the formula C = (LC – RC) / 2 be... D cmd (Signed values: positive indicates leftward, negative indicates rightward). The deviation value Δ provided by the zero-point correction unit. E This represents the offset of the mechanical zero point relative to the electrical zero point. It then represents the actual displacement command sent to the servo cylinder. D comp for: The minus sign indicates that the deviation of the mechanical zero point is compensated in reverse to the instruction. For example, if the mechanical zero point is offset to the left by Δ relative to the electrical zero point... E = +0.5 mm (i.e., the encoder midpoint value is greater than the electrical zero point), then in actual control, the command needs to be corrected to the right by 0.5 mm so that the actual position of the movable cloth hopper is consistent with the theoretical target.

[0079] At the same time, the control system also uses the laser reference value when calculating the deviation. L ref and R ref For reference, the difference between the real-time measurements LC and RC from the laser rangefinder and the laser reference value reflects the trolley's offset relative to the calibration time. That is, the actual value used to calculate the deviation is: Through the above compensation, it is ensured that even after a slight displacement or replacement of mechanical parts, the control system can still accurately control the position of the movable cloth hopper without manual intervention.

[0080] For example, after six months of continuous operation, the right-side servo electric cylinder of the belt roaster needs to be replaced due to mechanical wear. After replacement, the operator inputs self-learning calibration commands through the human-machine interface.

[0081] The calibration trigger unit generates a calibration start signal, and the stroke scanning unit begins to execute the stroke scan. The movable fabric hopper moves to the left from its current position, and the encoder reading is [value missing] when it touches the left limit switch. E 1 = 5000 pulses; then move to the right, and the encoder reading is when it touches the right limit. E 2 = 15000 pulses. Calculate the total stroke. S = 10000 pulses, encoder midpoint value: E mid =10000 pulses.

[0082] The stroke scanning unit drives the movable fabric hopper to the position corresponding to the encoder's midpoint value of 10000 pulses. At this time, the reading of the laser rangefinder on the left is... L ref = 100.0 mm, the laser rangefinder reading on the right is R ref = 100.0 mm (assuming the trolley is in the center position). The reference calibration unit records this set of values ​​as the laser reference value.

[0083] The zero-point correction unit reads the preset electrical zero-point code value. E zero = 9950 pulses, calculate zero-point deviation: Δ E =+50 pulses.

[0084] Based on the conversion of 10,000 pulses corresponding to a 2 mm lead screw pitch in a servo electric cylinder, 50 pulses correspond to 0.01 mm, meaning the mechanical zero point is offset to the left by 0.01 mm relative to the electrical zero point. The zero-point correction unit generates zero-point correction parameters, including the deviation value + 50 pulses.

[0085] The zero-point correction parameters are written to non-volatile memory. Calibration is complete, and the system exits self-learning mode.

[0086] The system then transitioned to normal operation. The laser rangefinder measured LC = 102.0 mm and RC = 98.0 mm. The control system first subtracted the laser reference value: LC' =+2.0mm, RC' = -2.0mm, C = 2.0mm. This indicates that the trolley has deviated 2.0 mm to the right. The control system generates a command to move 2.0 mm to the left. D cmd = +2.0 mm. Δ read from non-volatile memory. E = +0.01mm, calculated compensation command: D comp = 1.99mm. The actual command sent to the servo cylinder is to move 1.99 mm to the left. Due to a 0.01 mm deviation in the mechanical zero point, compensation ensures that the movable cloth hopper moves precisely to the target position, guaranteeing control accuracy.

[0087] Through the aforementioned self-learning calibration module, this invention solves the problems of mechanical zero-point drift caused by long-term operation and the need for complex manual calibration after component replacement. First, the stroke scanning unit automatically locates the mechanical limit and calculates the midpoint of the stroke, establishing a precise mechanical zero-point reference. Second, the reference calibration unit acquires the laser reference value at the midpoint of the stroke, eliminating the influence of changes in sensor installation position. Third, the zero-point correction unit generates correction parameters and stores them in non-volatile memory, ensuring that parameters are not lost after system power failure. Finally, during normal operation, zero-point compensation automatically corrects zero-point drift without manual intervention. This solution significantly reduces equipment maintenance difficulty and downtime, ensures the control accuracy and reliability of the automatic edge-laying material adjustment device throughout its entire lifecycle, effectively protects the upper plate from high-temperature burns, and improves the operating rate of the roasting machine.

[0088] In another embodiment of the present invention, during the operation of the belt roaster, the alarm for insufficient edge material may be caused by various reasons, including misalignment of the feeding hopper and the upper guardrail due to trolley deviation, material interruption due to blockage of the discharge pipe, and empty hopper. Blindly adjusting the feeding hopper position based solely on the material height alarm may exacerbate blockage or mislead operation. To solve the above technical problems, this embodiment provides an abnormal operating condition diagnosis and self-processing module integrated into the control system, specifically including the following units: The data acquisition unit is connected to the radar level gauge, the servo cylinder's built-in encoder, and the laser rangefinder to collect the following parameters in real time: Actual height measured by the radar level gauge on the left H L Measured height of radar level gauge on the right side H R The unit is mm; Actual position of the left servo electric cylinder P act，L Actual position of the right-side servo electric cylinder P act，R (Feedback from the encoder), unit: mm; Current deviation C The value is calculated from the laser rangefinder measurement according to formula (1), and the unit is mm.

[0089] The data acquisition unit updates the above data at a fixed sampling period (e.g., 100 ms) and caches it for subsequent units to call.

[0090] The material shortage alarm trigger unit is connected to the data acquisition unit to compare the actual edge material height with the preset nominal material level height. H nom A comparison is performed. An alarm signal for missing edge material is generated when the following conditions are met: ,or , where Δ H low The preset lower threshold value can be set according to process requirements, for example, taking... H low The value is 10% to 20%, in mm. This alarm signal is sent to the status assessment unit and operator station for initial notification.

[0091] The status judgment unit is connected to the material shortage alarm trigger unit and the data acquisition unit, and is used to execute the following diagnostic logic after receiving the edge material shortage alarm signal: 1. Obtain the current deviation amount C and with the preset action threshold T 动作 (For example T 动作 Compare with (e.g., 2mm); 2. Obtain the actual position of the servo electric cylinder P act With basic position instructions P base The deviation Δ between P = P act - P base and the preset position allowable error Δ P (e.g., Δ) P Compare with (e.g., 0.5mm); 3. If both conditions are met: and If the current missing edge material is not caused by the trolley running off track or the hopper not being aligned, it may be due to non-tracking factors such as blockage of the discharge pipe or material caking. At this time, a suspected blockage signal is generated.

[0092] If any of the above conditions are not met (i.e., the deviation exceeds the standard or the position deviation of the cloth hopper is too large), it indicates that the material shortage may be related to the deviation. The system will not start automatic unblocking, but will only maintain the original deviation adjustment logic and prompt the operator to check the deviation.

[0093] The unblocking execution unit is signal-connected to the status judgment unit and is used to output a vibration unblocking command to the vibrator installed on the feed pipe after receiving a suspected blockage signal. The vibrator can be electromagnetic or pneumatic, and its start and stop mode is controlled by a preset program, for example: vibration frequency: 25 Hz; vibration time: continuous for 5 s, interval for 2 s, repeated 3 times; total unblocking time: approximately 21 s.

[0094] After the unblocking execution unit outputs the command, it simultaneously notifies the recovery monitoring unit to start monitoring.

[0095] The monitoring unit is connected to both the data acquisition unit and the dredging execution unit, and is used to perform the following operations after the dredging execution unit outputs a vibration dredging command: Continuously monitor changes in the actual height of the edge material; Start the timer to record the time elapsed since the unblocking began; If the height of the edge material returns to normal during the timing process... If the blockage is cleared, the suspected obstruction signal is removed, the timer is stopped, and the event log is recorded. If the height of the edge material is still lower than the alarm threshold after the timer reaches the preset recovery time threshold (e.g., 30 seconds), the automatic unblocking is deemed ineffective, and the alarm is escalated.

[0096] The alarm escalation unit is signal-connected to the recovery monitoring unit and is used to generate a critical alarm signal when automatic unblocking fails. This signal contains the following information: Alarm type: Automatic unblocking failed due to missing edge material; Fault location: Left or right side; Current material high value and nominal value; Recommended steps: Check if the feed pipe is severely blocked, if the hopper is empty, or if the vibrator is malfunctioning.

[0097] A critical alarm signal is sent to the operator station via industrial Ethernet and displayed on the HMI in a flashing red manner, while simultaneously triggering an audible and visual alarm to prompt the operator to intervene in a timely manner.

[0098] Through the above-mentioned abnormal operating condition diagnosis and self-processing module, the present invention achieves the following technical effects: Automatically distinguish the cause of material shortage to avoid false alarms and malfunctions caused by deviation; Automatically activate vibration to clear blockages, reducing the frequency of manual inspections; If unblocking fails, the alarm will be upgraded to ensure that the fault is not ignored and to protect the equipment safety.

[0099] For example, during a certain operation, the left-side radar level gauge measures 45mm, the nominal material height is 50mm, and the lower limit threshold is 5mm, thus triggering a material shortage alarm. At this time, the laser rangefinder measures... L C = 100mm, R C = 100mm, therefore C = 0, which is less than the action threshold of 2mm; the actual position of the servo electric cylinder is 10.2mm, which is the base position command. P _base，L = 10.0mm, deviation 0.2mm, less than the allowable error of 0.5mm. The status judgment unit determines that the condition is met and generates a suspected blockage signal. The unblocking execution unit starts the vibrator to work in the preset mode. After the recovery monitoring unit times for 30 seconds, if the material height still does not recover, the alarm escalation unit issues a serious alarm, prompting the operator to check the left feed pipe. The operator promptly handles the blockage to avoid prolonged exposure and damage to the upper plate. This module realizes a closed-loop management of the entire process from alarm triggering, cause diagnosis, automatic handling to anomaly escalation, significantly improving the intelligence level and operational reliability of the roasting machine edge material laying system.

[0100] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. Other modifications can be easily made by those skilled in the art. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and examples shown and described herein.

Claims

1. An automatic adjustment device for edge material laying in a belt roasting machine, characterized in that, include: Two rangefinders are set at an interval relative to each other to measure the distance between themselves and the sealing plate of the trolley in real time; The control system is connected to two rangefinders to receive the measurement values ​​from the two rangefinders, calculates the offset and direction of the upper rail based on the measurement values, compares the absolute value of the upper rail offset with a preset action threshold, and generates an adjustment command based on the comparison result. The adjustment command includes the movement distance and movement direction. Two fixed fabric hoppers are set at intervals and opposite each other. The upper opening of each fixed fabric hopper is connected to the upstream discharge pipe, and its lower opening is sealed and connected to a flexible sealing connection pipe. The fixed fabric hopper is provided with a horizontal guide shaft perpendicular to the moving direction of the trolley. A movable fabric hopper is slidably mounted on the guide shaft. The upper opening of the movable fabric hopper is sealed and connected to the sealing connection pipe, and its lower opening is used to insert the edge material into the inner side of the upper panel of the trolley. Two servo electric cylinders are respectively mounted on two fixed fabric hoppers. Their push rod ends are parallel to the guide shaft and connected to the corresponding movable fabric hoppers. The servo electric cylinders are connected to the control system signal to receive adjustment commands and drive the movable fabric hoppers to move. Two radar level gauges are installed on two movable material hoppers and are connected to the control system. The probes of the radar level gauges are pointed downwards at the area where the edge material is falling from the movable material hoppers to detect the actual height of the edge material in real time and send the data to the control system. The control system compares the actual height with the preset nominal material level. When the actual height deviates from the nominal material level by more than the preset alarm threshold, an alarm is triggered.

2. The automatic adjustment device for edge material laying in a belt roaster as described in claim 1, characterized in that, The rangefinder is a laser rangefinder, and the sealed connecting pipe has a fish-scale sealing structure.

3. The automatic adjustment device for edge material laying in a belt roaster as described in claim 1, characterized in that, The servo electric cylinder has a built-in encoder, which is used to feed back the actual displacement signal of the servo electric cylinder to the control system.

4. The automatic adjustment device for edge material laying in a belt roaster as described in claim 1, characterized in that, The action threshold is 2mm.

5. The automatic adjustment device for edge material laying in a belt roaster as described in claim 1, characterized in that, The control system compares the actual height of the edge material detected by the radar level gauge with the nominal level height. When the actual height is lower than the nominal level height and the difference exceeds the lower limit threshold, an edge material shortage alarm is triggered.

6. The automatic adjustment device for edge material laying in a belt roaster as described in claim 1, characterized in that, There are two guide shafts, arranged in parallel, with both ends fixed to the fixed fabric hopper. The movable fabric hopper is mounted on the two guide shafts via linear bearings.

7. The automatic adjustment device for edge material laying in a belt roaster as described in claim 1, characterized in that, The control system includes: The deviation calculation unit is connected to the signals of two laser rangefinders and is used to calculate the deviation of the upper guardrail in real time based on the measurement values ​​of the two laser rangefinders. The basic position command generation unit is connected to the deviation calculation unit and is used to generate a basic position command based on the offset of the upper panel. The basic position command corresponds to the position that aligns the movable hopper with the upper panel of the trolley. The material height deviation calculation unit is connected to the signals of two radar level gauges and is used to calculate the material height deviation in real time based on the difference between the actual edge material height detected by the radar level gauges and the preset nominal material level height. The final position command generation unit is signal-connected to the basic position command generation unit and is used to output the basic position command as the final position command. The servo drive unit is connected to the final position command generation unit and the servo cylinder signal respectively. It is used to drive the servo cylinder according to the final position command, control the movable cloth bucket to move to the target position, and form a position closed-loop control. The material height deviation calculation unit is connected to the alarm module in the control system and is used to trigger an alarm when the material height deviation exceeds a preset threshold.

8. The automatic adjustment device for edge material laying in a belt roaster as described in claim 1, characterized in that, The control system includes a deviation trend prediction unit, which includes: The data storage module is used to continuously record the historical measurement values ​​of the two laser rangefinders and form a time series data of deviation corresponding to the trolley running time; The first prediction model is an autoregressive moving average model based on time series analysis, which is used to predict the trolley deviation at a future preset time point based on the deviation time series data, and outputs the first prediction value and its corresponding first confidence level. The second prediction model is a neural network prediction model based on historical data. The neural network is pre-trained with historical deviation data and is used to predict the deviation of the trolley at a preset time point in the future based on the deviation time series data, and outputs the second prediction value and its corresponding second confidence level. The fusion decision module is connected to the first prediction model, the second prediction model, and the laser rangefinder signal, respectively. It is used to receive the first prediction value, the second prediction value, and the real-time deviation amount at the current moment. It also performs weighted fusion of the first prediction value and the second prediction value based on the first confidence level and the second confidence level to calculate the pre-adjustment displacement amount. The pre-adjustment displacement amount is less than the action threshold and the direction is consistent with the predicted deviation trend. In this system, when the absolute value of the real-time deviation does not exceed the action threshold, but the absolute value of the deviation will exceed the action threshold at a future preset time point based on the pre-adjusted displacement, the system outputs a pre-adjustment command to the servo cylinder in advance before the future preset time point arrives, and controls the movable cloth bucket to follow the deviation trend direction according to the pre-adjusted displacement.

9. The automatic adjustment device for edge material laying in a belt roaster as described in claim 1, characterized in that, The control system has a self-learning calibration module, including: The calibration trigger unit is used to generate a calibration start signal during initial installation or after receiving a self-learning calibration command from an external input. The stroke scanning unit, connected to the calibration trigger unit and the servo cylinder signal, is used to control the servo cylinder to drive the movable fabric hopper to move along the guide shaft from the initial position in the first direction after receiving the calibration start signal, until the movable fabric hopper touches the first mechanical limit position, and record the first encoding value of the encoder built into the servo cylinder at this time; then, it controls the servo cylinder to drive the movable fabric hopper to move in the reverse direction until the movable fabric hopper touches the second mechanical limit position, and records the second encoding value of the encoder at this time; and calculates the actual total stroke of the movable fabric hopper and the encoder midpoint value corresponding to the stroke midpoint based on the first encoding value and the second encoding value. The reference calibration unit is connected to the laser rangefinder and the stroke scanning unit respectively, and is used to obtain the current distance value measured by the laser rangefinder as the laser reference value when the movable fabric hopper is located at the midpoint of the stroke. The zero-point correction unit is connected to the reference calibration unit and the stroke scanning unit. It is used to set the encoder midpoint value as the mechanical zero point of the movable cloth hopper and set the laser reference value as the measurement reference of the laser rangefinder. It calculates the deviation between the mechanical zero point and the preset electrical zero point and generates zero-point correction parameters. Non-volatile memory, connected to the zero-point correction unit signal, is used to store zero-point correction parameters; After the self-learning calibration mode ends, when the control system performs position control of the movable hopper, it calls the zero-point correction parameters in the non-volatile memory to compensate for the displacement commands sent to the servo cylinder.

10. The automatic adjustment device for edge material laying in a belt roaster as described in claim 1, characterized in that, The control system includes an abnormal operating condition diagnosis and self-processing module, including: The data acquisition unit is connected to the encoder signal built into the radar level gauge and servo cylinder to collect the actual edge material height, the actual position of the servo cylinder, and the current deviation measured by the laser rangefinder in real time. The material shortage alarm triggering unit is connected to the data acquisition unit and is used to compare the actual edge material height with the preset nominal material level height. When the actual edge material height is lower than the nominal material level height and the difference exceeds the preset lower limit threshold, an edge material shortage alarm signal is generated. The status judgment unit is connected to the material shortage alarm trigger unit and the data acquisition unit. After receiving the edge material shortage alarm signal, it is used to obtain the current deviation amount and compare it with the preset action threshold. At the same time, it obtains the deviation between the actual position of the servo cylinder and the basic position command. If the absolute value of the current deviation amount is less than the action threshold and the position deviation is less than the preset position allowable error, it is determined that the material shortage is not caused by the trolley deviation and a suspected blockage signal is generated. The unblocking execution unit is connected to the status judgment unit. It is used to output a vibration unblocking command to the vibrator installed on the feed pipe after receiving a suspected blockage signal, and control the vibrator to start and stop according to the preset mode. The monitoring unit is connected to the data acquisition unit and the dredging execution unit respectively. It is used to continuously monitor the change in the actual edge material height after the dredging execution unit outputs the vibration dredging command, and start the timer. The alarm escalation unit is connected to the recovery monitoring unit. When the timer reaches the preset recovery time threshold and the actual edge material height is still lower than the nominal material level height and the difference exceeds the lower limit threshold, a serious alarm signal is generated and sent to the operator station.

Images