A stocker path planning and segregation suppression control system

Abstract

The present application relates to the field of industrial automation digital programmable process control and unmanned yard intelligent control system based on programmable logic controller, and discloses a stockyard path planning and segregation suppression control system, which comprises a planning parameter configuration unit for setting stockpile target size and start-stop point initial value; a three-dimensional space alignment calibration unit for aligning the walking point cloud matrix of the trolley and the instantaneous beam signal of the cantilever unloading and resampling; a regional anti-noise state recognition unit for dividing the material falling influence area and the non-material falling monitoring area, configuring the point cloud data confidence weight, and outputting the surface three-dimensional height deviation; and a material flow speed trolley walking frequency matching unit for adjusting the trolley walking drive frequency control instruction according to the material instantaneous flow deviation.

Description

Technical Field

[0001] This invention relates to a stacker path planning and segregation suppression control system, belonging to the technical field of industrial automation digital programmable process control and unmanned stacker intelligent control system based on programmable logic controller. Background Technology

[0002] Currently, in bulk material stacking and transportation systems, the technical approach of using a trolley to spread materials along a track to construct regular bulk material strips is widely adopted in automated storage and transportation stations. This method is used to improve the geometric structure and density uniformity of the mixed materials. This approach relies on the long trolley and conveyor belt to lay the materials along the track, laying the foundation for material mixing. However, in material stacking operations, as the equipment repeatedly changes direction at the bidirectional endpoints and the material's natural sliding shape changes, the drop points and accumulation boundaries at the beginning and end of the stockpile become highly uncertain. The mechanical impact and material scattering generated by the material drop on-site result in a working environment filled with high concentrations of dust. The laser sensors used to detect the profile generate instantaneous noise and blind spots. Due to the overlapping and accumulation of materials in the end reversal area, the particle gradation distribution is altered, leading to deviations and beginning-end collapse defects, which cannot be directly adjusted based on the experience of the control personnel.

[0003] Optimizing hardware structures such as the boom or roller shape can improve material distribution stability, but purely hardware improvements do not solve the adaptive control problem. Software control methods also have shortcomings. For example, Chinese invention patent CN201780483U discloses a stacker control system that remotely issues action commands via a wireless remote control device. This approach implicitly relies on manual observation and manual button presses. In modern unmanned stacker yards with high dust levels and continuous stacking operations, this discrete adjustment based on human experience cannot provide continuous closed-loop compensation for instantaneous material flow fluctuations and pile edge collapse, nor can it suppress particle segregation. The core presuppositions and actual conditions are fundamentally flawed. Complex operating conditions and mismatches make it difficult for control accuracy to meet the requirements of fully automatic path planning. To solve the deviation of the profile at the end of the material pile, a linear improvement approach of simply increasing the number of fixed level gauges is adopted. This not only makes the hardware and software deployment cost high, but also makes it easy for high concentrations of dust to cause sensor optical path failure, making it impossible to remove false feature information in the measurement blind zone. If a fixed time filtering algorithm is used to smooth the sensing signal, the inherent delay in the system response will cause boundary overshoot in the trolley travel reversing control loop, causing abnormal impacts on the transmission components. In addition, the instantaneous fluctuation of the material flow in the upstream belt changes the cross-sectional area of ​​the fabric layer, making the static trajectory planning method unable to adapt to the stacked storage and transportation requirements under variable operating conditions.

[0004] Therefore, the technical problem to be solved by this invention is how to eliminate the feature blind spots caused by high-concentration dust environments by executing the spatiotemporal alignment of heterogeneous sensing signals through the programmable underlying logic instructions of the central digital controller, using programmable closed-loop control algorithms to calculate the deformation state of the material pile edge contour online, and adaptively correcting the speed adjustment step size and reversal boundary point of the trolley's bidirectional movement. This will construct a high-precision digital process control logic for an unmanned storage yard intelligent control system. Furthermore, by iteratively rewriting the digital closed-loop values ​​of multiple sets of target registers, the fully automatic digital process control system for unmanned storage yards will achieve centralized process control and collaborative data interaction for multiple heterogeneous sensing flows and multiple actuators. This will comprehensively solve the problem of insufficient multi-channel digital program processing efficiency and execution timing logic conflicts of industrial digital controllers in harsh industrial environments. Summary of the Invention

[0005] To address the problems in the background art, the technical solution of the present invention is as follows: A stacker path planning and segregation suppression control system, comprising:

[0006] The planning parameter configuration unit is used to set the target size of the stockpile and the initial register values ​​of the start and end points;

[0007] The three-dimensional spatiotemporal alignment calibration unit is used to align the point cloud matrix of the trolley travel with the instantaneous beam signal of the cantilever unloading in the in-situ space in the rectangular coordinate system, and resample according to a fixed sampling period to output the aligned multidimensional sensing dataset.

[0008] The regional noise reduction state recognition unit is used to divide the control space into the material dropping influence area and the non-material dropping monitoring area according to the real-time feedback coordinates of the trolley. When the real-time feedback coordinates of the trolley are located in the material dropping influence area, the confidence weight of the point cloud data in the multi-dimensional perception dataset is reduced, and the surface three-dimensional height deviation is output.

[0009] The material flow rate trolley travel frequency conversion ratio unit is used to dynamically change the trolley travel drive frequency control command based on the instantaneous flow rate deviation of the material measured by the continuous flow rate of the bulk material.

[0010] The boundary layered recursive weighted control unit is used to calculate the starting point correction and ending point correction based on the three-dimensional height deviation of the surface using a layered recursive weighted average rule, and dynamically modify the target register value of the next layer of reversal starting and ending points.

[0011] The limit over-limit protection unit is used to forcibly lock the current layer correction amplitude at the safety boundary threshold when the starting correction amount or the ending correction amount exceeds the safety boundary threshold.

[0012] Preferably, the target dimensions of the material pile set by the planning parameter configuration unit include the total length, the number of material layers, the design thickness of a single layer, and the repose angle of the loose material; the initial register values ​​of the start and end points set by the planning parameter configuration unit include the reference start coordinates and the reference end coordinates; the boundary layer recursive weighted control unit deduces the edge collapse slope of each layer of material pile according to the design thickness of a single layer and the repose angle of the loose material, and uses the start correction amount and the end correction amount to weighted scalar accumulation of the reference start coordinates and the reference end coordinates, and updates the target register values ​​of the next layer's reversal start and end points online.

[0013] Preferably, the three-dimensional spatiotemporal alignment calibration unit includes a spatial rigid transformation module and a time step resampling module; the spatial rigid transformation module is used to align the vehicle body pulse odometer reading, cantilever pitch angle sensor value, and trolley travel point cloud matrix in situ through a spatial rectangular transformation matrix; the time step resampling module uses the sampling period as the reference time axis to perform sliding window mean filtering on the high-frequency distributed cantilever unloading instantaneous beam signal, and outputs spatiotemporally locked material surface contour estimation data.

[0014] Preferably, the regional noise reduction status identification unit includes a geometric region division module and a time-varying weighted fusion module. The geometric region division module calculates the coordinates of the projection core point of the cantilever unloading point in a rectangular coordinate system based on the unloading trajectory. The spatial area with a radius of 3m around the coordinates of the projection core point is defined as the material drop influence zone, and the remaining open material surface area outside the material drop influence zone is defined as the non-material drop monitoring zone. The time-varying weighted fusion module is used to reduce the confidence weight of the point cloud data to 0.2 when the real-time feedback coordinates of the trolley enter the material drop influence zone, and simultaneously configures a numerical weight coefficient of 0.8 for the radar quantization data channel.

[0015] Preferably, the boundary layer recursive weighted control unit includes an initial boundary correction module and an end segregation control module; the initial boundary correction module is used to calculate the starting correction value required for the current reversal cycle based on the three-dimensional height deviation of the surface when the stockpile trolley is in the initial reversal stage; the end segregation control module is used to fine-tune the end correction value based on the collected end section particle segregation characteristics when the stockpile trolley is in the terminal reversal stage.

[0016] Preferably, the boundary layer recursive weighted control unit further includes a height deviation nonlinear normalization module and an interlayer recursive adjustment storage module; the height deviation nonlinear normalization module is used to input the surface three-dimensional height deviation into the nonlinear scaling formula to calculate the dimensionless height discrepancy factor; the interlayer recursive adjustment storage module is used to extract the dimensionless height discrepancy factor accumulated by the previous multiple layers within the reversal gap of the current layer of fabric, and update the target register value of the reversal start and end point of the next layer.

[0017] Preferably, the extreme boundary protection unit includes a soft limit amplitude interception module and a safety limit protection module; the soft limit amplitude interception module is used to monitor the starting point correction amount and the ending point correction amount online, and triggers the limit when the starting point correction amount or the ending point correction amount exceeds the 5% change threshold of the single design step size; the safety limit protection module is used to forcibly lock the correction amplitude to the safety boundary threshold when the trolley position is close to 2m from the edge of the track.

[0018] Preferably, the system further includes a load evolution trend monitoring module; the load evolution trend monitoring module includes a state time sequence storage unit, a differential numerical analysis unit, and an early warning signal generation unit; the state time sequence storage unit is used to continuously collect the root mean square value of the drive current output by the material flow rate trolley travel frequency conversion ratio unit with the increase of the number of fabric layers as the axis to construct a load time series feature library; the differential numerical analysis unit is used to perform sliding first-order differential calculation on the load time series feature library to extract the implicit mechanical resistance energy loss component caused by the expansion of the mechanical clearance of the wheel group transmission; the early warning signal generation unit is used to output a maintenance early warning status word when the implicit mechanical resistance energy loss component continuously exceeds the rated deviation threshold.

[0019] Preferably, the system also includes a high real-time control bus interconnection module and a monitoring terminal graphic output module; the high real-time control bus interconnection module is used to establish a full-duplex real-time data link through the industrial Ethernet network communication protocol, and synchronously interact with the output data of the planning parameter configuration unit, the control signal of the material flow rate trolley travel frequency conversion ratio unit, and the safety status word clock of the limit overrun protection unit, so that its network communication delay is less than 5ms; the monitoring terminal graphic output module is used to complete the dynamic matrix pixel mapping rendering output of the processed three-dimensional material surface contour and safety boundary amplitude data.

[0020] Compared with the prior art, the beneficial effects of the present invention are:

[0021] 1. In the path planning and segregation suppression control of the stacker, by collecting the material height deviation in the starting and ending areas of the stack, the material height deviation is used as the feedback excitation source of the control loop. The control unit calculates the starting and ending movement of the next layer of material distribution based on the height deviation. When the actual height of the current layer exceeds the reference height threshold, the bidirectional material distribution boundary of the next layer is spatially shifted outward or inward to compensate. This causes the trajectory of the material distribution plane to adaptively contract or expand as the material slippage shape and the stacking thickness change with the direction of material drop. This eliminates the problems of the stack's head and tail depression and particle segregation in the continuous reciprocating material distribution process, maintaining the consistency of the material distribution in the cross-section along the entire length of the material strip. This adjustment... The entire process relies entirely on the autonomous computation of digital controllers in a process control system with programmable digital logic. Through real-time issuance of material distribution program instructions and adaptive closed-loop parameter rewriting, highly integrated programmable digital process control is achieved. The process parameters of bulk material storage and transportation are seamlessly converted into digital instruction sequences for the controller, completing fully automated and fully digital path planning process control for unmanned storage yards. The multi-channel software execution sequence and interrupt handling mechanism of the general central digital controller are optimized, effectively preventing the calculation divergence of the control algorithm at extreme boundaries. The centralized control stability of industrial digital logic in harsh storage and transportation yard environments and the high real-time collaborative interaction performance of the network communication bus data link are improved.

[0022] 2. The system of this invention aligns laser point cloud data and radar material level signals through spatial three-dimensional rectangular coordinate transformation. It establishes an equivalent time window based on the sampling period of the laser scanner and calculates the arithmetic mean of the radar sampling values ​​of multiple points within the window, thereby achieving spatiotemporal unified alignment of heterogeneous data sources. Furthermore, it divides the material stacking work area into a material dropping zone and a non-material dropping zone by combining the real-time movement position of the trolley. Within the material dropping zone, radar data is given a high weight of 0.8 to capture the instantaneous dynamics of material dropping. In the non-material dropping zone, the system switches to overall contour monitoring based on laser data, eliminating random noise caused by cantilever material dropping impact and high-concentration dust dispersion on site, and outputting accurate surface tracking parameters for the control loop.

[0023] 3. The method of this invention constructs a proportional linkage model between the trolley's traveling speed and the instantaneous flow rate of the material on the belt conveyor. Based on the fabric width, the natural bulk density of the material, and the conveying efficiency coefficient, the initial benchmark value of the speed proportional coefficient is calculated. During the trolley's travel, the instantaneous material layer thickness deviation measured by the radar level gauge is introduced in real time. The traveling speed proportional coefficient is corrected online using an inverse proportional law. The deviation of the cross-sectional area caused by fluctuations in the material flow rate and changes in the compactness of the bulk material is converted into an adaptive adjustment of the feed speed of the trolley's variable frequency drive system. This eliminates the wave-like fluctuations in the thickness orientation, thereby spontaneously arranging a material layer with a constant bulk mass per unit length and a flat appearance. Attached Figure Description

[0024] Figure 1This is a flowchart illustrating the data interaction between the various units of the control system of this invention.

[0025] Figure 2 This is a structural diagram of the control system of the present invention.

[0026] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0027] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0028] A stacker path planning and segregation suppression control system, comprising:

[0029] The planning parameter configuration unit is used to set the target size of the stockpile and the initial register values ​​of the start and end points;

[0030] The three-dimensional spatiotemporal alignment calibration unit is used to align the point cloud matrix of the trolley travel with the instantaneous beam signal of the cantilever unloading in the in-situ space in the rectangular coordinate system, and resample according to a fixed sampling period to output the aligned multidimensional sensing dataset.

[0031] The regional noise reduction state recognition unit is used to divide the control space into the material dropping influence area and the non-material dropping monitoring area according to the real-time feedback coordinates of the trolley. When the real-time feedback coordinates of the trolley are located in the material dropping influence area, the confidence weight of the point cloud data in the multi-dimensional perception dataset is reduced, and the surface three-dimensional height deviation is output.

[0032] The material flow rate trolley travel frequency conversion ratio unit is used to dynamically change the trolley travel drive frequency control command based on the instantaneous flow rate deviation of the material measured by the continuous flow rate of the bulk material.

[0033] The boundary layered recursive weighted control unit is used to calculate the starting point correction and ending point correction based on the three-dimensional height deviation of the surface using a layered recursive weighted average rule, and dynamically modify the target register value of the next layer of reversal starting and ending points.

[0034] The limit over-limit protection unit is used to forcibly lock the current layer correction amplitude at the safety boundary threshold when the starting correction amount or the ending correction amount exceeds the safety boundary threshold.

[0035] Preferably, the target dimensions of the material pile set by the planning parameter configuration unit include the total length, the number of material layers, the design thickness of a single layer, and the repose angle of the loose material; the initial register values ​​of the start and end points set by the planning parameter configuration unit include the reference start coordinates and the reference end coordinates; the boundary layer recursive weighted control unit deduces the edge collapse slope of each layer of material pile according to the design thickness of a single layer and the repose angle of the loose material, and uses the start correction amount and the end correction amount to weighted scalar accumulation of the reference start coordinates and the reference end coordinates, and updates the target register values ​​of the next layer's reversal start and end points online.

[0036] Preferably, the three-dimensional spatiotemporal alignment calibration unit includes a spatial rigid transformation module and a time step resampling module; the spatial rigid transformation module is used to align the vehicle body pulse odometer reading, cantilever pitch angle sensor value, and trolley travel point cloud matrix in situ through a spatial rectangular transformation matrix; the time step resampling module uses the sampling period as the reference time axis to perform sliding window mean filtering on the high-frequency distributed cantilever unloading instantaneous beam signal, and outputs spatiotemporally locked material surface contour estimation data.

[0037] Preferably, the regional noise reduction status identification unit includes a geometric region division module and a time-varying weighted fusion module. The geometric region division module calculates the coordinates of the projection core point of the cantilever unloading point in a rectangular coordinate system based on the unloading trajectory. The spatial area with a radius of 3m around the coordinates of the projection core point is defined as the material drop influence zone, and the remaining open material surface area outside the material drop influence zone is defined as the non-material drop monitoring zone. The time-varying weighted fusion module is used to reduce the confidence weight of the point cloud data to 0.2 when the real-time feedback coordinates of the trolley enter the material drop influence zone, and simultaneously configures a numerical weight coefficient of 0.8 for the radar quantization data channel.

[0038] Preferably, the boundary layer recursive weighted control unit includes an initial boundary correction module and an end segregation control module; the initial boundary correction module is used to calculate the starting correction value required for the current reversal cycle based on the three-dimensional height deviation of the surface when the stockpile trolley is in the initial reversal stage; the end segregation control module is used to fine-tune the end correction value based on the collected end section particle segregation characteristics when the stockpile trolley is in the terminal reversal stage.

[0039] Preferably, the boundary layer recursive weighted control unit further includes a height deviation nonlinear normalization module and an interlayer recursive adjustment storage module; the height deviation nonlinear normalization module is used to input the surface three-dimensional height deviation into the nonlinear scaling formula to calculate the dimensionless height discrepancy factor; the interlayer recursive adjustment storage module is used to extract the dimensionless height discrepancy factor accumulated by the previous multiple layers within the reversal gap of the current layer of fabric, and update the target register value of the reversal start and end point of the next layer.

[0040] Preferably, the extreme boundary protection unit includes a soft limit amplitude interception module and a safety limit protection module; the soft limit amplitude interception module is used to monitor the starting point correction amount and the ending point correction amount online, and triggers the limit when the starting point correction amount or the ending point correction amount exceeds the 5% change threshold of the single design step size; the safety limit protection module is used to forcibly lock the correction amplitude to the safety boundary threshold when the trolley position is close to 2m from the edge of the track.

[0041] Preferably, the system further includes a load evolution trend monitoring module; the load evolution trend monitoring module includes a state time sequence storage unit, a differential numerical analysis unit, and an early warning signal generation unit; the state time sequence storage unit is used to continuously collect the root mean square value of the drive current output by the material flow rate trolley travel frequency conversion ratio unit with the increase of the number of fabric layers as the axis to construct a load time series feature library; the differential numerical analysis unit is used to perform sliding first-order differential calculation on the load time series feature library to extract the implicit mechanical resistance energy loss component caused by the expansion of the mechanical clearance of the wheel group transmission; the early warning signal generation unit is used to output a maintenance early warning status word when the implicit mechanical resistance energy loss component continuously exceeds the rated deviation threshold.

[0042] Preferably, the system also includes a high real-time control bus interconnection module and a monitoring terminal graphic output module; the high real-time control bus interconnection module is used to establish a full-duplex real-time data link through the industrial Ethernet network communication protocol, and synchronously interact with the output data of the planning parameter configuration unit, the control signal of the material flow rate trolley travel frequency conversion ratio unit, and the safety status word clock of the limit overrun protection unit, so that its network communication delay is less than 5ms; the monitoring terminal graphic output module is used to complete the dynamic matrix pixel mapping rendering output of the processed three-dimensional material surface contour and safety boundary amplitude data.

[0043] Example 1: In a stockpile operation scenario, a stacker is used to pile bulk materials into a specific shape while maintaining the cross-sectional uniformity of the pile along its length to prevent particle segregation. In the process parameter initialization phase of the stacker's path planning and segregation suppression control system, the control unit sets the target pile length to 300m, the number of stacking layers to 50, the thickness of a single layer to 0.3m, and the angle of repose to 8.53°, based on the mixing process plan. In the trolley drive system, the preset starting point coordinates are 10m, the preset ending point coordinates are 290m, and the starting point fluctuation range is... The endpoint floating range is 15m. It is 15m.

[0044] During the initial stacking phase, the control unit sends control commands to the trolley drive mechanism, driving the trolley to move along the length of the target stockpile. Simultaneously, the cantilever conveyor belt outputs material at a preset flow rate of 1500 t / h. During the stacking operation, the laser scanner in the sensing system collects point cloud data of the stockpile surface in real time, and the radar level gauge collects the stockpile height information in real time. Within the impact zone of falling material, the system configures the confidence weight of the laser point cloud data to 0.2 and the confidence weight of the radar quantization data to 0.8. Based on weighted fusion logic, the instantaneous data of the current stacking area is calculated. The process of collecting and controlling the characteristics of particle segregation at the end section of the stockpile height deviation is as follows: The particle segregation characteristics at the end section are explicitly mapped and converted into the geometric distortion state of the three-dimensional contour line of the stockpile's end surface in a spatial rectangular coordinate system at the level of physical entities and control data flow. The specific objectively measurable input parameters are the laser scanning distance matrix and radar height vector collected by the three-dimensional spatiotemporal alignment calibration unit. The action topology executed by the control unit is: extracting discrete height spatial points within a 1.5m span of the lower edge of the transverse slope of the end section of the trolley track. The system continuously calculates the spatial geometric gradient of adjacent spatial points. It uses the microscopic protrusions, slope abrupt change rate, and surface roughness caused by the gravity-induced accumulation of large particles on the local slope as characteristic features. A dimensionless segregation characteristic evaluation scalar is then calculated using this closed-loop solution as the trigger state for the next-level control loop. After this mapping transformation, the specific physical flow process is as follows: When the stockpile trolley reaches the terminal reversing stage, the three-dimensional spatiotemporal alignment calibration unit extracts the continuous point cloud shape of the terminal section in the lateral width direction. The control unit calculates the shape in a specific local area at the lower edge of the slope. The slope change rate and profile roughness values ​​of the domain are related to the fact that large particles tend to slide and roll towards the bottom edge of the stockpile under gravity. When particle segregation occurs, the local packing slope at the bottom of the stockpile will be distorted and raised, which means that the local slope change rate exceeds the preset extreme threshold of 12%. At this time, the control unit uses the excess proportion of the local slope change rate as a feature quantity and inputs it into the segregation evaluation logic. When it is determined that the segregation feature quantity exceeds the standard in a positive direction, it indicates that coarse particles are enriched at the terminal edge. The system fine-tunes and reduces the current endpoint correction amount, so that the reversing boundary of the next layer shrinks appropriately inward by 0.1 meter, by increasing the coverage of the bottom edge with subsequent fine-grained layers to suppress particle segregation, thereby stabilizing the uniformity of particle distribution. The regional noise-resistant state recognition unit acquires a multi-dimensional sensing dataset, extracts a continuous set of spatial point coordinates along the transverse width direction of the trolley running track, and uses the three-point moving window method to calculate the second-order central difference value of adjacent spatial points to obtain the local slope change rate. In specific data processing, the control unit does not need to rely on abstract function operators, but directly performs data calculation through the following discrete steps: First, within the local distribution window containing 5 spatial laser detection points, the least squares method is used to perform first-order linear regression fitting on the spatial rectangular coordinates of these 5 detection points to obtain a baseline feature line; then, the actual measured height value of each detection point in the window is extracted in sequence, and the fitted height value of the corresponding horizontal coordinate position of the point on the baseline feature line is subtracted to obtain the height residual of each detection point; then, the 5 height residuals are squared respectively, and the squared results are summed, and the summation result is divided by With a degree of freedom parameter of 4, the discrete variance value of the profile height is obtained. Finally, the square root of this discrete variance value is calculated to perform a closed-loop calculation within the digital register space and output a precise profile fluctuation scalar. The control unit directly uses this profile fluctuation scalar as the data source and employs a standard deviation filter operator to calculate the root mean square deviation between the actual measured height of each point within the sliding window and the corresponding height on the least squares fitted line. The root mean square deviation is determined as the profile roughness index. When the profile roughness index exceeds a preset extreme threshold of 12%, coarse grain enrichment segregation is determined at the terminal edge. Subsequent calculations divide the absolute value of the measured actual height deviation by the single-layer design thickness of 0.3m, inputting the ratio into a hyperbolic tangent scaling function. When the ratio is between 0 and 0.1, a 1:1 linear response output is maintained. When the ratio exceeds 0.2, the output is progressively compressed to the range of 0.95 to 1.0 through this function, completing the nonlinear normalization processing of the surface three-dimensional height deviation and outputting a dimensionless height discrete factor.

[0045] After the first layer of material is piled up, the boundary layer recursive weighted control unit calculates the correction amount between the start and end points, based on the measured deviation of the initial height of the material pile. relative to reference height The difference between them is calculated by determining the height dispersion factor. Specifically, in the data logic loop of the boundary layered recursive weighted control unit, this model corresponds to a closed-loop iterative data flow control flow. Its bound objectively measurable characteristic input parameters include the normalized height deviation component, the instantaneous flow fluctuation component of the conveyor belt material, the trolley travel speed fluctuation component, and the historical comprehensive correction coefficient register value of the previous reversing cycle. The action topology executed by the control unit is as follows: divide the absolute value of the actual height deviation measured in the current layer by the single-layer design thickness of 0.3m to complete the dimensionless normalization process; at the same time, divide the absolute value of the difference between the current instantaneous flow of the belt and the rated flow of 1500t / h by the rated flow to obtain the flow fluctuation normalization value; then divide the absolute value of the difference between the actual travel speed of the trolley and the process reference speed by the reference speed to obtain the speed fluctuation normalization value; the control unit calls the sign function to determine the positive value of the actual height deviation. In the negative direction, when the height deviation is positive (i.e., the material pile is too high), the sign function outputs a positive sign. At this time, the normalized values ​​of flow fluctuation and velocity fluctuation are taken as positive increments, multiplied by their respective preset weighting coefficients of 0.25 and 0.15, and summed with the height deviation component multiplied by a weighting coefficient of 0.6. This summation is then multiplied by a control coefficient of 0.6 and accumulated with the comprehensive correction coefficient register value of the previous historical cycle multiplied by a control coefficient of 0.4, thus outputting the comprehensive correction coefficient for the current layer's reversal cycle in a closed loop. If the height deviation is negative (i.e., the material pile is concave), the sign function outputs a negative sign. The flow and velocity fluctuations are converted into negative decreases and subtracted. The control unit ultimately writes the calculated comprehensive correction coefficient as a deterministic register update state into the hardware storage area. This is implemented using this specific discrete conditional flow and employs a hierarchical recursive weighted average model to calculate the correction amount. The solution formula is: ,in, This is the recursive smoothing coefficient, with a value of 0.4; This is the height deviation weighting coefficient, with a value of 0.6; This is the traffic stability weighting coefficient, with a value of 0.25. The weighting coefficient for walking speed fluctuations is 0.15. This represents the normalized deviation of the stockpile height. This is the normalized value for traffic fluctuation; This represents the normalized value for walking speed fluctuations. As a sign function, to ensure the comparability of heterogeneous parameters across channels, the underlying calculation and processing paths for the above-mentioned normalized parameters and factors are as follows: The normalized stockpile height deviation is obtained by dividing the absolute value of the currently measured actual height deviation by the single-layer design thickness of 0.3 meters for dimensionless processing; the normalized flow fluctuation value is obtained by collecting the current instantaneous flow rate of the cantilever conveyor, calculating its absolute difference from the design rated flow rate of 1500 tons per hour, and then dividing the difference by the design rated flow rate; the normalized travel speed fluctuation value is obtained by calculating the difference between the current actual travel speed of the trolley and the process preset reference speed, and then dividing the absolute value of the difference by the reference speed; the height dispersion factor directly corresponds to the normalized stockpile height deviation. Through the above proportional mapping, all heterogeneous state parameters involved in the calculation are equivalently converted into dimensionless scalar data between 0 and 1, eliminating the absolute scale differences between different physical dimensions.

[0046] Based on the geometric collapse slope of a single-layer stockpile, the single-layer correction amount The calculation formula is: ,in, The pre-set stacking thickness for each layer, The control unit is based on the material's angle of repose. The target register values ​​corresponding to the next layer's dynamic start and end points are updated. The boundary layered recursive weighted control unit modifies the target register values ​​of the next layer's commutation start and end points. An independent location is deployed in the central controller program to update the interrupt vector. Within the current layer's commutation gap, the pre-comparison logic is called to read the dimensionless height discrete factor accumulated from the previous multiple layers. The start and end point corrections are calculated according to the layered recursive weighted average rule. The start and end point corrections are compared with the preset 5% change threshold limit verification of the single design step size in the current layer's memory. If the 5% change threshold is not exceeded, a data write instruction is triggered when the next clock rising edge signal arrives during the commutation gap, latching the updated target register values ​​of the next layer's commutation start and end points into the programmable memory. The logic controller hardware storage area uses a feedback calibration loop to periodically correct boundary measurement deviations caused by the evolution of the material flow channel. This avoids divergence or overshoot oscillation between the trolley speed control command and the boundary control loop due to high-concentration dust point cloud noise. In the closed-loop iterative calculation process, the physical linkage mechanism in which the flow stability term and the walking speed fluctuation term are multiplied by the height deviation sign function is as follows: The height deviation sign function is responsible for providing the direction pointer for spatial compensation. When the measured height deviation is positive, i.e. the material pile is locally too high, the sign function outputs a positive sign. At this time, the fluctuation components of flow and speed will be superimposed into the comprehensive correction coefficient in the form of positive increments, driving the starting and ending point coordinates of the next layer to perform spatial translation expansion to the outside, thereby flattening the excessively high area by expanding the material distribution range.

[0047] When the height deviation is negative, indicating a concave defect in the material pile, the sign function outputs a negative sign, transforming flow and velocity fluctuations into negative reductions. This causes the reversal start and end point coordinates to contract inwards, concentrating more material at the defect location during subsequent material placement. Ultimately, this directional modulation logic achieves closed-loop control between process dynamics disturbances and boundary space position compensation. If the calculated correction value exceeds 5% of the pile length, the limit exceedance protection unit locks the correction amplitude within the maximum allowable range to ensure operational stability. During material laying, the trolley's travel speed... With instantaneous material flow The linkage, and its control commands satisfy the following relationship: ,in, For walking speed, As a speed proportional coefficient, the system uses the instantaneous material layer thickness fed back by the radar level gauge in real time. Comparison coefficient Perform closed-loop correction, let: Adjust the scaling factor online The system compensates for height fluctuations on the surface of the material pile and maintains the consistency of the material layer thickness. When continuously piling material back and forth to the 50th layer, the system maintains the cross-sectional consistency of the material strip length direction by performing spatiotemporal unified alignment of the heterogeneous sensor data of laser and radar. Finally, the height deviation of the beginning and end of the material pile is stabilized within ±0.2 m. Through the digital logic closed-loop command flow, the online correction and segregation suppression of path planning are realized.

[0048] Example 2: In a certain ore blending yard, a stacker is used to mix and pile up various mineral powders to reduce the fluctuation of ore composition. To verify the control performance of the path planning and segregation suppression control scheme of the present invention, the test platform selected the same model stacker and the matching belt conveyor control system. High-precision laser scanners and radar level gauges were respectively arranged at the beginning and end of the stockpile to obtain the surface contour and height data of the stockpile in real time. The test environment simulated the actual production conditions and actively input the system with a material flow fluctuation interference signal with an amplitude of ±5% to test the system's path correction capability under non-steady-state flow conditions.

[0049] To establish the engineering rationality of the control parameters, this experiment presupposes a stockpile length of 300m, 50 stockpile layers, and a single layer design thickness of 0.3m. Parameter settings consider both system stability and data processing efficiency: the smoothing coefficient of the layered recursive weighted average model... The value is set to 0.4, a choice based on a trade-off between the need to filter out data acquisition noise and the real-time performance of control response; when When the value is below 0.3, the control system is too sensitive to short-term flow fluctuations, causing frequent vibrations in the trolley; when When the value is higher than 0.5, the system lags in correcting the cumulative deviation of the stockpile height. Based on this logic, the test group determined... The value is set to 0.4, which serves as the optimal working window parameter for the current operational scenario. During the experiment, the control unit calculates the initial height deviation of the material pile based on the sensor data. relative to reference height For each layer of material, the system updates the comprehensive correction coefficient for the current reversing cycle based on the following layer-by-layer recursive weighted formula. : ,in, This is to normalize the stockpile height deviation; This is the normalized value for traffic fluctuation; This represents the normalized value for walking speed fluctuations. The height deviation weight is set to 0.6. As a weight for traffic stability, the value is 0.25; The weight for walking speed fluctuation is set to 0.15.

[0050] The experiment monitored that when the material flow rate fluctuated, the material layer thickness reported by the radar level gauge changed. The change occurs; at this moment, the speed of the large vehicle changes. via formula Real-time control, among which, For instantaneous material flow rate, a proportional coefficient is used to achieve dynamic correction. The update follows The closed-loop logic and data show that when the actual material flow rate deviates from the design value by 8%, the experimental group controlled the material layer thickness fluctuation within 2% by adjusting the trolley travel frequency in real time. The comparative test showed that the control group without path correction was affected by the material flow rate fluctuation, and the two ends of the material pile showed obvious collapse and sharp corners, with a peak height deviation of 1.1m. Under the same flow rate fluctuation interference, the experimental group stabilized the height deviation at the beginning and end within 0.15m by correcting the start and end points online. The test data showed a clear performance inflection point. When the thickness of a single layer of material exceeds 0.5m, even with the above path correction strategy, the particle slippage phenomenon on the slope of the material pile still intensifies with the increase of thickness, and the segregation index increases. This confirms that the layer thickness range of 0.3m to 0.5m defined by the present invention is the optimal process range under physical flow constraints. The material distribution behavior outside this range is limited by the natural stacking slope of the material. The above test process is autonomously realized through the closed-loop command flow integrated into the digital controller, which effectively maintains the stability of the cross-sectional composition of the mixed material strip.

[0051] Example 3: This example combines Figures 1 to 2 This paper describes a stacker path planning and segregation suppression control system, such as... Figure 1As shown, point cloud and beam signals are input to the 3D spatiotemporal alignment calibration unit to output a multidimensional sensing dataset. The trolley feedback coordinates and the multidimensional sensing dataset output by the 3D spatiotemporal alignment calibration unit are both input to the area noise reduction status recognition unit to output the surface height deviation. The planned parameter configuration unit sets the size and initial value, which are respectively input to the boundary layer recursive weighted control unit to calculate the start and end point correction amount and the material flow rate trolley travel frequency conversion ratio unit to dynamically change the command. The surface height deviation output by the area noise reduction status recognition unit is input to the boundary layer recursive weighted control unit to calculate the start and end point correction amount. The boundary layer recursive weighted control unit calculates the start and end point correction amount and inputs it to the limit over-limit protection unit to lock the correction amplitude. The correction amplitude locked by the limit over-limit protection unit is input to the safety boundary threshold. The instantaneous material flow deviation is input to the material flow rate trolley travel frequency conversion ratio unit to dynamically change the command. The output of the material flow rate trolley travel frequency conversion ratio unit is the travel drive frequency command.

[0052] like Figure 2 As shown, the planned parameter configuration unit outputs to the three-dimensional spatiotemporal alignment calibration unit, the three-dimensional spatiotemporal alignment calibration unit outputs to the regional noise immunity status identification unit, the regional noise immunity status identification unit outputs to the material flow rate trolley travel frequency conversion ratio unit, the material flow rate trolley travel frequency conversion ratio unit outputs to the boundary layer recursive weighted control unit, the boundary layer recursive weighted control unit outputs to the limit cross-boundary protection unit, the limit cross-boundary protection unit outputs to the operating load evolution trend monitoring module, the operating load evolution trend monitoring module outputs to the high real-time control bus interconnection module, and the high real-time control bus interconnection module outputs to the monitoring terminal graphic output module.

[0053] Example 4: In the stockpiling operation of a steel plant's blending yard, the system needs to cope with the instantaneous signal distortion caused by multipath reflection from the sensors and the dynamic load fluctuations caused by the collapse of edge materials. A material flow fluctuation compensation model is constructed to maintain the morphological stability of the material strip cross-section. During the initialization phase of the sensing unit, the laser scanner and radar level gauge complete coordinate system calibration, establishing a three-dimensional rectangular coordinate system with the trolley's running track as the Y-axis and the cantilever drop point as the origin O. In actual calibration, to eliminate the influence of the dynamic changes in the spatial position of the drop point caused by the cantilever movement on the fixed spatial coordinate system, the origin is not a moving origin, but rather a fixed initial origin when the trolley is at the initial zero position of the track and the cantilever pitch angle is 0 degrees. The projection point of the cantilever material drop point on the ground serves as the absolute geometric origin of the three-dimensional Cartesian coordinate system. During system operation, when the trolley travels or the cantilever pitches, the control unit reads the readings from the vehicle's pulse odometer and the values ​​from the cantilever pitch angle sensor in real time. It then calculates the real-time three-dimensional spatial offset of the current cantilever material drop point relative to this absolute geometric origin. This offset is used to compensate for the spatial coordinate backward translation of the instantaneously acquired point cloud matrix and beam signal. This seamlessly converts and aligns the dynamic material drop trajectory and sensing data to a unified fixed Cartesian coordinate system, maintaining absolute self-consistency between the global spatial point cloud matrix and the geometric topological relationship of the material surface height matrix. The sensing system then collects data from the material surface... The three-dimensional point cloud data of each spatial point is processed, and spatiotemporal alignment is performed on each frame of data. The three-dimensional spatiotemporal alignment calibration unit uses the time-stamp information of each scanning point to map the radar quantization data onto the spatial grid of the laser point cloud through linear interpolation, generating a three-dimensional height matrix of the material pile surface. The system executes dynamic load balancing logic when the material flow rate becomes... When the material flow rate fluctuates, the variable frequency drive unit of the trolley travels adjusts the frequency according to a preset compensation function, thereby adjusting the trolley travel drive frequency. The calculation path is as follows: ,in, This is the dynamic gain compensation coefficient. for The instantaneous flow rate of materials detected at all times. Determine the target thickness for the material layer. This is the effective fabric width of the cantilever.

[0054] parameter The calibration process employs an offline segmented fitting method, and experiments are conducted to record different flow fluctuation rates. Extracting the fluctuation value of the cross-sectional height of the material pile. The response curve in the range of 0.05 to 0.20, hour, Set to a fixed value of 0.5; when hour, It increases linearly to 0.75; when hour, The value is locked at 0.80 to prevent overheating of the trolley's travel motor due to excessively high system adjustment frequency. The boundary layered recursive weighted control unit monitors the stockpile boundary in real time. If the fluctuation rate of the material level data exceeds 5% for three consecutive sampling periods, the path correction algorithm is triggered, and the control unit calculates the position offset value of the cantilever coordinate register. The system updates the position and interrupt vector. This adjustment command is transmitted to the actuator via the industrial control bus within a 120ms response time. Under extreme conditions with a flow fluctuation of 20%, the above correction logic limits the material strip cross-sectional height deviation to within 0.12m, confirming the technical effectiveness of the dynamic compensation model in suppressing edge segregation and height fluctuation.

[0055] Example 5: In a coking coal blending yard of a coking plant, the stacker faces the risk of segregation due to uneven particle size distribution of the material. Furthermore, environmental dust interference causes intermittent drift in the sensing layer data. The system eliminates implicit bias in the sensing signal by introducing offline calibration and pre-calibration procedures. Before the stacker is deployed to the work site, the control unit executes the offline calibration procedure. In a material-free state, the system scans a calibration object of known geometric dimensions using a laser scanner, collecting distance data from 1000 sampling points. The mean and standard deviation are calculated, and the systematic deviation of the laser signal is determined by comparing it with the true geometric value of the calibration object. During system operation, radar level gauge data is acquired in real time. Based on the calibration deviation, the calibrated stockpile height is generated. In addition, the system sets the initial zero-point drift correction factor for the flow rate sensor. This factor is the instantaneous current value obtained by the flow rate sensor when no material is passing through after the stacker has stopped and the conveyor belt has finished unloading. With preset zero-point current The calculation shows that, The system outputs current to the sensor during operation. Corrections were made to obtain flow velocity data. ,in, This is the flow conversion coefficient.

[0056] Before the material stacking operation begins, a pre-calibration procedure is performed to address hardware thermal drift caused by ambient temperature. The system records the drive unit under no-load conditions at different ambient temperatures. Current reference sequence below When the ambient temperature When a change occurs, the system uses linear interpolation to determine the corresponding current reference value, loads it into the buffer, and the control unit then uses the current ambient temperature as the reference value. Real-time locked current reference This leads to the correction of the load judgment logic for the walking drive motor. If the real-time monitoring value of the motor current is... With load reference The difference exceeds The system determines that the driving fluctuation is caused by uneven material load, and then adjusts the gain parameter to adjust the trolley travel speed to the preset safe travel range. This calibration process is completed in the self-test stage before the stacker starts. By incorporating the ambient temperature parameter into the load judgment logic, the system maintains stable perception and path planning of the material accumulation state under different climatic conditions, thereby ensuring the consistency of the stacking trajectory.

[0057] Example 6: In a coking plant's blending yard stockpiling scenario, to suppress segregation caused by particle size differences during material unloading and to mitigate interference from environmental dust on sensing accuracy, the system executes standardized engineering procedures to construct a stockpiling path model. During the process parameter initialization phase, the target stockpile length is set to 300.0m, the number of stockpile layers is set to 50, and the design thickness for each layer is [not specified]. The angle of repose of the material is set to 0.30m. Set to 8.53°, establish the direction of the stacker's travel track as... The origin is the material drop point of the shaft and cantilever. The material yard uses a three-dimensional rectangular coordinate system. The system executes the following sensing calibration procedure: zero-point calibration of the laser scanner's spatial coordinate system is performed; under no-material conditions, the trolley is moved to the center of the track, and the cantilever pitch angle is adjusted. The angle is 0°. A full-angle scan is performed on the preset standard calibration block to obtain... Distance of each sampling point from the original data Calculate the average distance measurement Establish a zero-point reference for the flow velocity sensor. After the stacker stops and the belt conveyor finishes unloading, read the instantaneous current value corresponding to the flow velocity sensor. With preset zero-point current Calculate the zero-point drift correction factor The calculation formula is: ,in, This is the zero-point drift correction factor for the flow velocity sensor. This represents the instantaneous current value under conditions where no material is passing through. To establish a preset zero-point calibration current value, a load reference matrix for the trolley drive unit is constructed, recording the load under different ambient temperatures in the trolley drive unit's no-load operation state. Current reference value sequence The control unit reads the current ambient temperature. Linear interpolation is then performed to determine the corresponding current reference value. It is then loaded into the memory cache to correct the load determination threshold of the walking drive motor.

[0058] The path planning and segregation suppression control logic is as follows: During the movement of the trolley, the motor current value is monitored in real time. With load reference When the absolute value of the difference between the two satisfies When the material load fluctuation is determined to exceed the set range, the control unit locks the current reference for that operating condition. and the speed of the large vehicle Adjust to the preset speed range of 0.10 m / s to 0.50 m / s. If an instantaneous deviation in material flow rate is detected... The material flow rate of the trolley travel frequency conversion ratio unit is based on the dynamic gain compensation coefficient. Adjust drive frequency ,frequency The calculation path is as follows: ,in, For the driving frequency of the large vehicle, This is the dynamic gain compensation coefficient. for The instantaneous flow rate of materials detected at all times. Determine the target thickness for the material layer. The effective fabric width of the cantilever. Calibration is performed using a piecewise fitting method when hour, Set to 0.50; when hour, It increases linearly to 0.75; when hour, The material flow rate trolley travel frequency conversion unit is locked at 0.80. When a sudden change in material flow rate exceeding 15% occurs, it detects the instantaneous material flow rate change rate of the upstream conveyor belt and determines the dynamic gain compensation coefficient based on the segmented fitting path. When the instantaneous material flow rate deviation is less than 8%, the dynamic gain compensation coefficient is set to a fixed value of 0.50. When the deviation is between 8% and 15%, the dynamic gain compensation coefficient is controlled to increase linearly from 0.50 to 0.75 as the flow rate deviation increases. When the deviation is greater than 15%, the dynamic gain compensation coefficient is locked at 0.80. The calculated trolley travel drive frequency is sent to the travel drive frequency conversion system via the industrial Ethernet network communication protocol with a network communication delay of less than 5ms. When the incoming material flow rate increases, the trolley travel speed is controlled to increase proportionally and synchronously. When the incoming material flow rate decreases, the trolley travel speed is controlled proportionally and synchronously. Synchronous reduction, utilizing the variable frequency drive system to adjust the feed speed, continuously adaptively compensates for the cross-sectional area deviation caused by fluctuations in the incoming material flow rate. During the continuous reciprocating material distribution process, the operating status of the bulk material equipment is stabilized within the safe travel range. The selection of the above core thresholds and weight coefficients has clear physical and engineering boundary basis: defining the 3-meter radius around the core point of the projection as the material falling influence zone is based on the limit diffusion radius of the bulk material falling trajectory of the cantilever belt conveyor at the highest conveying flow rate, which can accurately cover the core interference area where the material dust is most concentrated; configuring the confidence weight of the point cloud data to 0.2 and the radar channel to 0.8 is because under harsh working conditions where the dust concentration exceeds 80 grams per cubic meter, the laser beam will scatter and undergo multipath attenuation, while the radar beam has high penetration. This weight ratio can control the comprehensive material level measurement error within 3%.

[0059] The critical values ​​for flow fluctuation rate are set at 0.08 and 0.15 based on the mechanical hysteresis characteristics of the conveying system. Fluctuations below 8% can be naturally smoothed out by the volumetric elasticity of the belt itself, without the need to adjust the trolley. Sudden changes exceeding 15% trigger maximum frequency conversion compensation to prevent the pile collapse. The boundary layered recursive weighted control unit monitors the stockpile boundary in real time. If the fluctuation rate of the material level data exceeds 5% for three consecutive sampling periods, the path correction algorithm is triggered to update the cantilever position interruption vector. This correction command is transmitted to the actuator within 120ms via the industrial control bus. In this stockpile, the cross-sectional height deviation of the material strip is controlled within 0.12m, and the height deviation of the beginning and end of the stockpile is kept within ±0.20m.

[0060] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0061] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A stacker path planning and segregation suppression control system, characterized in that, include: The planning parameter configuration unit is used to set the target size of the stockpile and the initial register values ​​of the start and end points; The three-dimensional spatiotemporal alignment calibration unit is used to align the point cloud matrix of the trolley travel with the instantaneous beam signal of the cantilever unloading in the in-situ space in the rectangular coordinate system, and resample according to a fixed sampling period to output the aligned multidimensional sensing dataset. The regional noise reduction state recognition unit is used to divide the control space into the material dropping influence area and the non-material dropping monitoring area according to the real-time feedback coordinates of the trolley. When the real-time feedback coordinates of the trolley are located in the material dropping influence area, the confidence weight of the point cloud data in the multi-dimensional perception dataset is reduced, and the surface three-dimensional height deviation is output. The material flow rate trolley travel frequency conversion ratio unit is used to dynamically change the trolley travel drive frequency control command based on the instantaneous flow rate deviation of the material measured by the continuous flow rate of the bulk material. The boundary layered recursive weighted control unit is used to calculate the starting point correction and ending point correction based on the three-dimensional height deviation of the surface using a layered recursive weighted average rule, and dynamically modify the target register value of the next layer of reversal starting and ending points. The limit over-limit protection unit is used to forcibly lock the current layer correction amplitude at the safety boundary threshold when the starting correction amount or the ending correction amount exceeds the safety boundary threshold.

2. The stacker path planning and segregation suppression control system according to claim 1, characterized in that, The target dimensions of the stockpile set by the planning parameter configuration unit include the total length, the number of fabric layers, the design thickness of a single layer, and the angle of repose of the loose material; the initial register values ​​of the start and end points set by the planning parameter configuration unit include the reference start coordinates and the reference end coordinates. The boundary layer recursive weighted control unit deduces the edge collapse slope of each layer of material pile based on the single-layer design thickness and the repose angle of the bulk material, and uses the starting point correction amount and the ending point correction amount to accumulate the weighted scalar of the reference starting point coordinate and the reference ending point coordinate, and updates the target register value of the next layer reversal starting and ending points online.

3. The stacker path planning and segregation suppression control system according to claim 1, characterized in that, The three-dimensional spatiotemporal alignment calibration unit includes a spatial rigid transformation module and a time step resampling module. The spatial rigid transformation module is used to align the vehicle pulse odometer readings, cantilever pitch angle sensor values, and the trolley travel point cloud matrix in situ using a spatial rectangular transformation matrix. The time step resampling module uses the sampling period as the reference time axis to perform sliding window mean filtering on the high-frequency distributed instantaneous beam signal of the cantilever unloading, and outputs spatiotemporally locked material surface contour estimation data.

4. The stacker path planning and segregation suppression control system according to claim 1, characterized in that, The regional noise reduction status identification unit includes a geometric region division module and a time-varying weighted fusion module. The geometric region division module calculates the coordinates of the projection core point of the cantilever unloading point in a rectangular coordinate system based on the unloading trajectory. It defines the spatial area with a radius of 3m around the coordinates of the projection core point as the material drop influence zone and defines the remaining open material surface area outside the material drop influence zone as the non-material drop monitoring zone. The time-varying weighted fusion module is used to reduce the confidence weight of the point cloud data to 0.2 when the real-time feedback coordinates of the trolley enter the material drop influence zone, and simultaneously configures a numerical weight coefficient of 0.8 for the radar quantization data channel.

5. The stacker path planning and segregation suppression control system according to claim 1, characterized in that, The boundary layer recursive weighted control unit includes an initial boundary correction module and an end segregation control module. The initial boundary correction module is used to calculate the starting correction value required for the current reversal cycle based on the three-dimensional height deviation of the surface when the stockpile trolley is in the initial reversal stage. The end segregation control module is used to fine-tune the end correction value based on the collected particle segregation characteristics of the terminal section when the stockpile trolley is in the terminal reversal stage.

6. The stacker path planning and segregation suppression control system according to claim 1, characterized in that, The boundary layer recursive weighted control unit also includes a height deviation nonlinear normalization module and an interlayer recursive adjustment storage module; the height deviation nonlinear normalization module is used to input the three-dimensional height deviation of the surface into the nonlinear scaling formula to calculate the dimensionless height discrepancy factor; the interlayer recursive adjustment storage module is used to extract the dimensionless height discrepancy factor accumulated by the previous multiple layers within the reversal gap of the current layer of fabric and update the target register value of the reversal start and end point of the next layer.

7. The stacker path planning and segregation suppression control system according to claim 1, characterized in that, The extreme boundary protection unit includes a soft limit amplitude interception module and a safety limit protection module. The soft limit amplitude interception module is used to monitor the starting point correction amount and the ending point correction amount online, and triggers the limit when the starting point correction amount or the ending point correction amount exceeds the 5% change threshold of the single design step size. The safety limit protection module is used to forcibly lock the correction amplitude to the safety boundary threshold when the trolley position is close to 2m from the edge of the track.

8. The stacker path planning and segregation suppression control system according to claim 1, characterized in that, The system also includes a load evolution trend monitoring module; the load evolution trend monitoring module includes a status time sequence storage unit, a differential value analysis unit, and an early warning signal generation unit; The state time sequence storage unit is used to continuously collect the root mean square value of the drive current output by the material flow rate trolley travel frequency conversion ratio unit with the increase of the number of fabric layers as the axis to build a load time series feature library. The differential numerical analysis unit is used to perform sliding first-order differential calculation on the load time series feature library and extract the implicit mechanical resistance energy loss component caused by the expansion of the mechanical clearance of the wheel group transmission; the early warning signal generation unit is used to output a maintenance early warning status word when the implicit mechanical resistance energy loss component continuously exceeds the rated deviation threshold.

9. A stacker path planning and segregation suppression control system according to claim 1, characterized in that, The system also includes a high real-time control bus interconnection module and a monitoring terminal graphic output module. The high real-time control bus interconnection module is used to establish a full-duplex real-time data link through the industrial Ethernet network communication protocol, and synchronously interacts the output data of the planning parameter configuration unit, the control signal of the material flow rate trolley travel frequency conversion ratio unit, and the safety status word clock of the limit overrun protection unit, so that its network communication latency is less than 5ms. The monitoring terminal graphic output module is used to complete the dynamic matrix pixel mapping rendering output of the processed three-dimensional material surface contour and safety boundary amplitude data.

Images