An adaptive speed-regulating energy-saving belt conveyor control system

Abstract

This invention relates to the field of automated control technology for belt conveyors, and more particularly to an energy-saving belt conveyor control system with adaptive speed regulation. The system includes: a material feature extraction module, which collects real-time contour morphology information of the material above the belt conveyor and performs cross-sectional area integration calculation to extract and output real-time load parameters; and a spatiotemporal distribution deduction module, which synchronously matches the real-time load parameters with the real-time operating displacement of the conveyor belt to generate and output a material spatial distribution matrix. In this invention, the spatiotemporal distribution of the material is deeply integrated with the viscoelastic physical characteristics of the conveyor belt, accurately constructing a dynamic resistance map of micro-segments covering the entire line and an overall energy consumption objective function. This allows for the calculation of a smooth reference speed trajectory that minimizes energy consumption within different load ranges, achieving high-precision adaptive energy-saving speed regulation during belt conveyor operation.

Description

Technical Field

[0001] This invention relates to the field of automated control technology for belt conveyors, and in particular to an energy-saving belt conveyor control system with adaptive speed regulation. Background Technology

[0002] Belt conveyors, as indispensable continuous bulk material transport equipment in many heavy industrial sectors such as coal, metallurgy, ports, and chemicals, are widely used due to their advantages of long conveying distance, large carrying capacity, and strong operational continuity. However, with the increasing demand for long-distance and ultra-large-capacity conveying in modern industry, the enormous energy consumption of belt conveyor systems has become increasingly prominent, becoming a key bottleneck restricting related enterprises from achieving green, low-carbon, cost-reduction, and efficiency improvement. In order to effectively reduce ineffective energy consumption, the industry has generally begun to explore and apply the "on-demand speed regulation" operation control mode, that is, dynamically adjusting the operating frequency and speed of the drive motor according to the actual amount of material carried on the conveyor belt, in order to minimize mechanical friction and electrical energy loss under no-load or light-load conditions while ensuring timely material delivery. This control process highly depends on the accurate modeling and state perception of the overall machine's operating resistance and system energy consumption characteristics.

[0003] However, most existing belt conveyor speed control technologies rely solely on macroscopic average conveying volume or local single-point weighing feedback for simple proportional frequency regulation. They fail to deeply integrate the spatiotemporal distribution of bulk materials on the conveyor belt with the unique viscoelastic physical characteristics of flexible rubber belts. Consequently, they cannot accurately construct a dynamic resistance spectrum and a global energy consumption objective function covering the entire micro-segment of the conveyor. This results in the calculated speed control curve often exhibiting a step-like discrete abrupt change when switching between different load ranges. This not only easily causes severe instantaneous tensile impacts on the conveyor belt and mechanical transmission components but also makes it difficult to truly achieve adaptive optimal energy-saving speed regulation that aligns with the underlying physical structure of the system. Summary of the Invention

[0004] To overcome the above shortcomings, this invention provides an adaptive speed regulation energy-saving belt conveyor control system, which aims to improve the problem that most existing belt conveyor speed regulation control technologies rely only on macroscopic average conveying volume or local single-point weighing feedback for simple proportional frequency regulation.

[0005] This invention provides the following technical solution: an adaptive speed-regulating energy-saving belt conveyor control system, comprising: The material feature extraction module collects the contour and shape information of the material above the belt conveyor in real time, performs cross-sectional area integration calculation, and extracts and outputs real-time load parameters. The spatiotemporal distribution simulation module synchronously matches the real-time load parameters with the real-time operating displacement of the conveyor belt of the belt conveyor to generate and output a material spatial distribution matrix. The resistance assessment module, based on the material spatial distribution matrix and the pre-constructed discrete micro-element segment model of the belt conveyor, combines the conveyor belt indentation rolling resistance parameters and idler friction parameters to perform micro-element force analysis, and generates and outputs dynamic resistance maps. The decision module constructs an energy consumption model for the belt conveyor based on the dynamic resistance map, calculates the energy consumption objective function, and generates and outputs the reference speed trajectory. The tension control module performs longitudinal tension analysis and frequency band trajectory correction based on the reference speed trajectory and the preset viscoelastic dynamic equation of the conveyor belt, and generates and outputs the target speed command. The variable frequency execution module, based on preset drive slippage threshold and variable frequency overload threshold, limits and filters the acceleration and deceleration slope of the target speed command, and generates and outputs the variable frequency control frequency signal.

[0006] Preferably, in the material feature extraction module, the extraction and output of real-time load parameters specifically includes the following steps: The raw point cloud data of the material above the belt conveyor is obtained, filtered, and the material contour curve is extracted. The material profile curve is integrated relative to the reference surface of the empty conveyor belt to obtain the instantaneous material cross-sectional area; The real-time load parameters are calculated by multiplying the instantaneous material cross-sectional area by a preset material bulk density coefficient.

[0007] Preferably, in the spatiotemporal distribution extrapolation module, generating and outputting the material spatial distribution matrix specifically includes the following steps: Obtain the discrete time series of the real-time operating displacement of the conveyor belt; Using the discrete time series as a spatial index, the real-time load parameters are discretized and mapped to construct a one-dimensional spatial load vector; The displacement of the one-dimensional spatial load vector is updated over time based on the circular cycle length of the conveyor belt, and the material spatial distribution matrix is ​​generated by combining the results.

[0008] Preferably, in the resistance assessment module, generating and outputting the dynamic resistance map specifically includes the following steps: Align the pre-constructed discrete micro-segment model of the belt conveyor with the material space distribution matrix according to the spatial coordinates to obtain the local load mass corresponding to each micro-segment; Substitute the local load mass into the preset conveyor belt indentation rolling resistance equation and idler friction force equation to calculate the local running resistance value of each micro segment. The local operating resistance values ​​are spatially sequenced along the entire length of the conveyor belt to generate a dynamic resistance map.

[0009] Preferably, the step of substituting the local load mass into the preset conveyor belt indentation rolling resistance equation and idler friction equation to calculate the local running resistance value of each micro-segment specifically includes the following steps: Based on the local load mass, the preset elastic modulus of the conveyor belt and the structural parameters of the idler rollers, the vertical indentation depth sequence of the conveyor belt in each idler roller contact area is calculated; The vertical indentation depth sequence and the relaxation time constant of the conveyor belt are input into a preset generalized Maxwell viscoelastic model to calculate the dynamic indentation friction coefficient. The local load mass, the dynamic indentation friction coefficient, and the roller friction parameters are multiplied and added together to generate the local running resistance values ​​for each micro-segment.

[0010] Preferably, in the decision-making module, generating and outputting the reference velocity trajectory specifically includes the following steps: Substitute the dynamic resistance map into the preset belt conveyor energy consumption model to construct an energy consumption objective function with the overall machine operating energy consumption as the dependent variable and the operating speed as the independent variable. Based on the preset upper and lower limits of the conveyor belt running speed constraints, the derivative of the energy consumption objective function is solved to obtain the discrete speed sequence corresponding to each preset load interval. The discrete velocity sequence is subjected to time-domain interpolation fitting to generate a baseline velocity trajectory.

[0011] Preferably, in the tension control module, generating and outputting the target speed command specifically includes the following steps: Using the reference velocity trajectory as the displacement excitation condition, and substituting it into the preset viscoelastic dynamic equation of the conveyor belt, the time-domain response waveform of the longitudinal tension of the conveyor belt is calculated. The time-domain response waveform is transformed in the frequency domain to extract the target frequency band whose amplitude exceeds the preset tension fluctuation threshold; A preset notch filter is used to filter and correct the target frequency band in the reference velocity trajectory to generate a target velocity command.

[0012] Preferably, the step of substituting the reference velocity trajectory as a displacement excitation condition into the preset viscoelastic dynamic equation of the conveyor belt to calculate the time-domain response waveform of the longitudinal tension of the conveyor belt specifically includes the following steps: Using the acceleration components of the reference velocity trajectory as the system boundary conditions, the differential equation of the conveyor belt distributed parameters constructed based on the Kelvin-Woyt viscoelastic theory is solved by finite difference discrete solution to obtain the longitudinal strain sequence of discrete nodes along the entire conveyor belt. The longitudinal strain sequence is multiplied by the dynamic stiffness coefficient and viscous damping coefficient of the conveyor belt to generate the time-domain response waveform of the longitudinal tension at each node location.

[0013] Preferably, in the frequency conversion execution module, generating and outputting the frequency conversion control signal specifically includes the following steps: The first derivative of the target speed command is calculated to obtain the command acceleration. When the command acceleration exceeds the upper limit of acceleration determined by the drive slippage threshold and the frequency converter overload threshold, the amplitude of the command acceleration is truncated. The target velocity command is reconstructed by integrating the truncated command acceleration to obtain the reconstructed velocity command. The reconstructed speed command is multiplied by a preset speed-frequency conversion coefficient to generate a frequency conversion control signal.

[0014] The present invention has the following beneficial effects: 1. In this invention, the spatiotemporal distribution of materials is deeply integrated with the viscoelastic physical characteristics of the conveyor belt, and a dynamic resistance map of micro-segments covering the entire line and the energy consumption objective function of the whole machine are accurately constructed. In this way, a smooth reference speed trajectory that minimizes energy consumption is solved in different load ranges. Under the premise of ensuring that the mechanical system is not affected by the sudden change of discrete commands, high-precision adaptive energy-saving speed regulation is realized in the operation of the belt conveyor.

[0015] 2. In this invention, the reference speed trajectory is substituted into the distributed parameter differential equation constructed based on viscoelastic theory. The specific excitation frequency band that induces longitudinal resonance of the conveyor belt is accurately captured by finite difference solution and sliding window frequency domain transformation. The dangerous frequency components in the speed command are filtered out by dynamically reconstructed notch filter, which effectively blocks the standing wave resonance and belt breakage risk caused by speed regulation and frequency conversion process, and improves the physical structural safety of flexible conveyor belt under transient acceleration and deceleration conditions.

[0016] 3. In this invention, dual electromechanical boundary constraints of anti-slipping of the drive drum and anti-overload of the frequency converter are introduced in the underlying frequency conversion execution stage. The running acceleration of the target speed command is strictly truncated and reconstructed by discrete integral. Combined with the rated electromagnetic slip mechanism of the asynchronous motor, electromechanical cross-domain mapping is performed to compensate for the actual speed loss during load operation. This ensures that the energy-saving and vibration-damping speed commands output by the upper-level algorithm can be stably and accurately reproduced at the underlying physical device. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the architecture of an energy-saving belt conveyor control system with adaptive speed regulation proposed in this invention. Detailed Implementation

[0018] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] This invention provides an adaptive speed regulation energy-saving belt conveyor control system, such as... Figure 1 As shown, it includes: The material feature extraction module collects the contour and shape information of the material above the belt conveyor in real time, performs cross-sectional area integration calculation, and extracts and outputs real-time load parameters. In this embodiment, the material feature extraction module extracts and outputs real-time load parameters, specifically including the following steps: The raw point cloud data of the material above the belt conveyor is obtained, filtered, and the material contour curve is extracted. The material profile curve is integrated relative to the reference plane of the empty conveyor belt to obtain the instantaneous material cross-sectional area; The real-time load parameters are calculated by multiplying the instantaneous material cross-sectional area by the preset material bulk density coefficient.

[0020] Specifically, the material feature extraction module uses a 3D laser scanner installed above the belt conveyor to collect raw point cloud data of the material above the conveyor belt in real time. Due to dust interference in the industrial environment, the system uses a Gaussian filtering algorithm to perform spatial filtering on the spatial coordinate values ​​in the raw point cloud data, removing free noise data in 3D space, thereby extracting a continuous material contour curve. This material contour curve is represented in the cross-sectional coordinate system as a function model of the absolute height of the material surface changing with the width of the conveyor belt.

[0021] The system pre-calibrates the spatial position of the conveyor belt surface under no-load conditions and records it as the reference plane for the no-load conveyor belt. Then, the extracted material profile curve is integrated relative to the no-load conveyor belt reference plane in one dimension to obtain the instantaneous material cross-sectional area at the current scanning section. The calculation formula is as follows: ; In the formula, S represents the instantaneous cross-sectional area of ​​the material; W represents the effective load-bearing width of the belt conveyor; x represents the lateral spatial coordinate along the width direction of the conveyor belt, with the centerline of the effective load-bearing width of the belt conveyor as the origin of the spatial coordinate system; This represents the absolute height of the material profile curve corresponding to the horizontal spatial coordinate x. This represents the absolute height of the reference plane of the unloaded conveyor belt corresponding to the lateral spatial coordinate x.

[0022] The system obtains a preset bulk density coefficient based on the physical properties of the conveyed material. The real-time load parameter is calculated by multiplying the instantaneous material cross-sectional area obtained in the above steps by this preset bulk density coefficient. The calculation formula is as follows: ; In the formula, This parameter represents the real-time load capacity of the material carried per unit length of the conveyor belt; S represents the instantaneous cross-sectional area of ​​the material; This indicates the preset bulk density coefficient of the material.

[0023] This enables non-contact continuous measurement of the weight of materials carried by the conveyor belt, providing basic data for system resistance assessment and speed control.

[0024] The spatiotemporal distribution simulation module synchronously matches the real-time load parameters with the real-time operating displacement of the conveyor belt of the belt conveyor to generate and output the material spatial distribution matrix. Furthermore, in the spatiotemporal distribution simulation module, generating and outputting the material spatial distribution matrix specifically includes the following steps: Obtain the discrete time series of the real-time running displacement of the conveyor belt; Using discrete time series as spatial indexes, the real-time load parameters are discretized and mapped to construct a one-dimensional spatial load vector; The displacement of the one-dimensional spatial load vector is updated over time based on the circular cycle length of the conveyor belt, and the material spatial distribution matrix is ​​generated by combining the two.

[0025] Specifically, the spatiotemporal distribution simulation module collects pulse signals in real time through a rotary encoder installed at the end of the drive roller shaft of the belt conveyor, converting them into the conveyor belt's running distance. The system sets a fixed spatial step size, and performs a data sampling when the accumulated running distance reaches this spatial step size, thereby obtaining a discrete-time series of the real-time running displacement of the conveyor belt corresponding to the fixed spatial step size. At each current sampling moment of this discrete-time series, the system extracts the real-time load parameters calculated and output by the previous module as the current input source.

[0026] The system divides the entire physical length of the belt conveyor into several equally spaced virtual spatial nodes according to the spatial discrete step length. The total number of nodes is the ratio of the conveyor belt's circular period length to the spatial discrete step length. The system uses the current moment of the discrete time series as the starting point of the spatial index, assigns the real-time load parameter acquired at the current moment to the first virtual spatial node, and simultaneously shifts the data of each virtual spatial node from the previous moment one node position backward. This discretizes and maps the real-time load parameter, constructing a one-dimensional spatial load vector for the current moment. The node extrapolation and update formula is: ; ; In the formula, Indicates the current time The load value corresponding to the i-th virtual space node in the generated one-dimensional spatial load vector; This represents the load value corresponding to the (i-1)th virtual space node in the one-dimensional spatial load vector formed at the previous time k-1. Indicates the index number of the current time step in the discrete time series; i represents the spatial index number of the virtual space node, and its value range is a positive integer greater than one and less than or equal to the total number of nodes N; N represents the total number of nodes, determined by the ratio of the conveyor belt's circular period length to its spatial distance. This represents the payload value corresponding to the first virtual space node at time k. This represents the real-time load parameter corresponding to the current time k output by the previous module.

[0027] Based on the circular cycle length of the conveyor belt, the system allocates a two-dimensional storage area in the controller's memory corresponding to the number of time-based simulation steps. The system first reassembles the generated load values ​​corresponding to each virtual spatial node at the current moment into a one-dimensional spatial load column vector with a length of N equal to the total number of nodes. The combination formula is: ; In the formula, This represents the one-dimensional spatial payload column vector generated at the current time k; T represents the transpose of a matrix or vector.

[0028] The system performs displacement rolling updates and stacking concatenation on the one-dimensional spatial load column vector generated at the current moment and the one-dimensional spatial load column vector accumulated within a preset number of historical steps in the time dimension, combining them to generate a material spatial distribution matrix. The matrix concatenation formula is as follows: ; In the formula, This represents the spatial distribution matrix of materials generated at the current moment; This represents the one-dimensional spatial load column vector corresponding to the current moment; This represents the one-dimensional spatial load column vector corresponding to the previous time step; This represents the one-dimensional spatial payload column vector corresponding to the (H-1)th time in history. H represents the preset number of matrix time simulation steps.

[0029] The above steps transform real-time load data at a single fixed location into two-dimensional spatiotemporal distribution data covering the entire length of the conveyor belt, providing a spatiotemporally aligned data source for subsequent accurate calculation of the local operating resistance of each micro-segment.

[0030] The resistance assessment module, based on the material spatial distribution matrix and the pre-constructed discrete micro-element model of the belt conveyor, combines the conveyor belt indentation rolling resistance parameters and idler friction parameters to perform micro-element force analysis, and generates and outputs dynamic resistance maps. Furthermore, the generation and output of the dynamic resistance map in the resistance assessment module specifically includes the following steps: Align the pre-built discrete micro-segment model of the belt conveyor with the material space distribution matrix according to the spatial coordinates to obtain the local load mass corresponding to each micro-segment. Substitute the local load mass into the preset conveyor belt indentation rolling resistance equation and idler friction equation to calculate the local running resistance value of each micro segment. The local operating resistance values ​​along the entire length of the conveyor belt are spatially spliced ​​to generate a dynamic resistance map.

[0031] Furthermore, the calculation of the local running resistance value of each micro-element segment by substituting the local load mass into the preset conveyor belt indentation rolling resistance equation and idler friction equation includes the following steps: Based on the local load mass, the preset elastic modulus of the conveyor belt and the structural parameters of the idler, the vertical indentation depth sequence of the conveyor belt in the contact area of ​​each idler is calculated; The vertical indentation depth sequence and the relaxation time constant of the conveyor belt are input into the preset generalized Maxwell viscoelastic model to calculate the dynamic indentation friction coefficient. The local load mass, dynamic indentation friction coefficient, and idler roller friction parameters are multiplied and added together to generate the local running resistance values ​​for each micro-segment.

[0032] Specifically, the resistance assessment module pre-constructs a discrete micro-segment model of the belt conveyor. The discrete partitioning method of this model is consistent with the virtual space nodes of the aforementioned spatiotemporal distribution deduction module. The system aligns the pre-constructed discrete micro-segment model of the belt conveyor with the material spatial distribution matrix output by the previous module according to spatial coordinates. Specifically, it extracts the one-dimensional spatial load column vector output by the previous module at the current moment and performs node-by-node matching. Combined with the unloaded mass parameters of the conveyor belt itself, it obtains the local load mass corresponding to each micro-segment. The mass alignment conversion formula is as follows: ; In the formula, This represents the local load mass corresponding to the i-th infinitesimal element at the current time; k represents the index number of the current time step in the discrete time series; i represents the spatial index number of the micro segment, and its value range is a positive integer greater than or equal to one and less than or equal to the total number of nodes N of the micro segment; N represents the total number of micro-segment nodes along the entire belt conveyor line; This represents the load value corresponding to the i-th virtual space node in the one-dimensional space load column vector at the current moment, as output by the previous module. This indicates the preset mass parameter per unit length of the unloaded conveyor belt; This indicates the spatial distance of the infinitesimal segment from the walk.

[0033] Based on the obtained local load mass, the system, combined with the preset elastic modulus of the conveyor belt and the structural parameters of the idlers, calculates the sequence of vertical indentation depths of the conveyor belt in the contact areas of each idler according to contact mechanics theory. The formula for calculating the vertical indentation depth of a single micro-segment is as follows: ; In the formula, This represents the vertical indentation depth of the i-th micro-element in the contact area of ​​the idler roller at time k. Indicates the current time The local load mass corresponding to the i-th micro-element segment; Represents the gravitational acceleration constant; E represents the preset elastic modulus of the conveyor belt; R represents the preset idler structure parameters, i.e., the idler radius.

[0034] The system then inputs the calculated vertical indentation depth sequence and the inherent relaxation time constant of the conveyor belt material into a preset generalized Maxwell viscoelastic model. The system introduces a preset idler contact characteristic time and, by extracting the hysteresis physical characteristics of the rubber material's deformation recovery in the model, calculates the dynamic indentation friction coefficient for each micro-segment. The calculation formula is as follows: ; In the formula, Indicates the current time The dynamic indentation friction coefficient corresponding to the i-th micro-element segment; This represents the dimensionless contact damping proportionality constant pre-defined in the generalized Maxwell viscoelastic model; Indicates the current time No. The vertical indentation depth of each micro-segment in the contact area of ​​the idler roller; R represents the preset idler radius; Indicates the preset contact characteristic time of the idler roller; T represents the preset relaxation time constant of the conveyor belt material; This represents an exponential function with the natural constant as its base.

[0035] After obtaining the dynamic indentation friction coefficient, the system performs a multiplication-addition combination operation on the local load mass, the dynamic indentation friction coefficient, and the preset idler roller friction parameters to calculate the local running resistance value of each micro-segment. The calculation formula is as follows: ; In the formula, Indicates the current time The local running resistance value corresponding to the i-th micro-element segment; Indicates the current time The local load mass corresponding to the i-th micro-element segment; Represents the gravitational acceleration constant; Indicates the current time The dynamic indentation friction coefficient corresponding to the i-th micro-element segment; This indicates the preset roller friction parameter, i.e., the roller rotational equivalent friction coefficient.

[0036] After obtaining the local operating resistance values ​​of all micro-segments at the current moment, the system performs spatial sequence splicing of the local operating resistance values ​​along the entire length of the conveyor belt to generate a dynamic resistance map covering the entire belt conveyor. The splicing matrix formula is as follows: ; In the formula, This represents the column vector of the dynamic resistance map generated at the current time k; to These represent the local running resistance values ​​corresponding to the first to Nth infinitesimal segments at the current time k, respectively. T denotes the transpose of a vector.

[0037] The above steps transform the spatiotemporal load data into a dynamic resistance set with physical deformation characteristics, providing a high-precision local resistance input for the subsequent construction of the energy consumption equation of the system.

[0038] The decision module calculates the energy consumption objective function based on the dynamic resistance map and the preset belt conveyor energy consumption model, and generates and outputs the reference speed trajectory. Furthermore, in the decision-making module, generating and outputting the baseline velocity trajectory specifically includes the following steps: Substitute the dynamic resistance graph into the preset belt conveyor energy consumption model to construct an energy consumption objective function with the overall machine operating energy consumption as the dependent variable and the operating speed as the independent variable. Based on the preset upper and lower limits of the conveyor belt running speed constraints, the derivative of the energy consumption objective function is solved to obtain the discrete speed sequence corresponding to each preset load interval. Time-domain interpolation fitting is performed on the discrete velocity sequence to generate a baseline velocity trajectory.

[0039] Specifically, the decision module first extracts the column vector of the dynamic resistance map output by the previous module at the current moment, and spatially accumulates the local operating resistance values ​​of all the micro-segments contained within it to obtain the overall dynamic operating resistance of the entire line at the current moment. The system substitutes this overall dynamic operating resistance into the preset energy consumption model of the belt conveyor. This energy consumption model comprehensively considers the nonlinear loss of air damping of the belt conveyor, the effective mechanical power to overcome physical resistance, and the low-speed domain penalty mechanism, and constructs an energy consumption objective function with the overall operating energy consumption as the dependent variable and the operating speed as the independent variable. The function formula is as follows: ; In the formula, This represents the scalar value of the total operating energy consumption of the machine, with the operating speed at the current time k as the independent variable. k represents the index number of the current time step in the discrete time series; This represents the current running speed of the problem at time k. This represents the preset air resistance and mechanical nonlinear friction power loss coefficient; This represents the preset transmission efficiency coupling constant; N represents the total number of micro-segment nodes along the entire belt conveyor line; i represents the spatial index number of the micro segment, and its value range is a positive integer greater than or equal to one and less than or equal to the total number of nodes N of the micro segment; This represents the local running resistance value corresponding to the i-th micro-segment in the dynamic resistance map column vector generated at the current time k in the output of the previous module. This represents the preset low-speed domain penalty weight coefficient to prevent excessively low running speeds.

[0040] To determine the optimal energy-saving speed, the system, based on preset upper and lower limits of conveyor belt operating speed constraints, calculates the first derivative of the energy consumption objective function with respect to operating speed, establishing an extreme value solution equation where the derivative equals zero. The calculation formula is as follows: ; In the formula, This represents the first derivative of the scalar energy consumption of the entire machine with respect to the operating speed.

[0041] The system solves the above extreme value equation to obtain the theoretical optimal speed and compares this theoretical optimal speed with the preset upper and lower limits of the conveyor belt running speed constraints. When the theoretical optimal speed exceeds the constraint range, the system forcibly truncates it to the corresponding upper or lower limit value, thereby obtaining the discrete optimal speed value mapped to the current moment. Simultaneously, the system extracts the local load mass corresponding to each micro-segment in the previous module and accumulates it across the entire line to obtain the current total load of the entire machine. The system divides the preset rated load of the conveyor into multiple preset load intervals with equal gradients and determines which preset load interval the current total load belongs to. Within one control cycle, the system continuously executes the above calculations, extracting and combining multiple discrete optimal speed values ​​belonging to the same preset load interval in chronological order to obtain the discrete speed sequence corresponding to each preset load interval.

[0042] After obtaining the discrete velocity sequence, since directly issuing discrete values ​​would cause abrupt changes in system control commands, the system employs a cubic spline interpolation algorithm to perform time-domain interpolation fitting on the discrete velocity sequence. The system constructs a continuous and smooth cubic polynomial between adjacent discrete time nodes, transforming the discrete velocity set into a continuous reference velocity trajectory in the time domain. The fitting formula is as follows:

[0043] In the formula, This represents the baseline velocity trajectory value corresponding to the time variable t; m represents the index number of the time node in the discrete velocity sequence, and its value range is a positive integer greater than or equal to one and less than the total length of the discrete velocity sequence; t represents a continuous time variable, and its value range is limited to the absolute time label of the m-th time point. And less than the absolute time label of the (m+1)th node Within the closed and open intervals; Represents the absolute time label of the m-th time node; Represents the absolute time label of the (m+1)th time node; , , , The system uses a cubic spline interpolation algorithm to obtain zeroth to third-order polynomial fitting coefficients based on numerical solutions of adjacent discrete velocities, where the zeroth-order coefficients are... The value is strictly equal to the discrete optimal velocity value corresponding to the m-th time node in the discrete velocity sequence.

[0044] This step transforms the dynamic resistance data of the entire machine into the optimal control speed curve that meets the theoretical minimum energy consumption, and eliminates the mechanical shock caused by discrete commands through time-domain smooth interpolation, providing a continuous and energy-saving reference speed control source for subsequent systems.

[0045] The tension control module performs longitudinal tension analysis and frequency band trajectory correction based on the reference speed trajectory and the preset viscoelastic dynamic equation of the conveyor belt, and generates and outputs the target speed command. Furthermore, in the tension control module, generating and outputting the target speed command specifically includes the following steps: Using the reference velocity trajectory as the displacement excitation condition, and substituting it into the preset viscoelastic dynamic equation of the conveyor belt, the time-domain response waveform of the longitudinal tension of the conveyor belt is calculated. Perform frequency domain transformation on the time-domain response waveform to extract the target frequency band whose amplitude exceeds the preset tension fluctuation threshold; A preset notch filter is used to filter and correct the target frequency band in the reference velocity trajectory to generate the target velocity command.

[0046] Furthermore, by using the reference velocity trajectory as the displacement excitation condition and substituting it into the preset viscoelastic dynamics equation of the conveyor belt, the time-domain response waveform of the longitudinal tension of the conveyor belt is calculated, specifically including the following steps: Using the acceleration components of the reference velocity trajectory as the system boundary conditions, the differential equation of the conveyor belt distributed parameters constructed based on the Kelvin-Woyt viscoelastic theory is solved by finite difference discrete solution to obtain the longitudinal strain sequence of discrete nodes along the entire conveyor belt. The longitudinal strain sequence is multiplied by the dynamic stiffness coefficient and viscous damping coefficient of the conveyor belt, and then combined to generate the time-domain response waveform of the longitudinal tension at each node location. Specifically, the tension control module first extracts the continuous reference velocity trajectory generated by the previous module, performs a first-order derivative operation on it with respect to the time variable, and obtains the acceleration component of the reference velocity trajectory. The system establishes a differential equation for the distributed parameters of the conveyor belt based on Kelvin-Woythe viscoelastic theory, and inputs the acceleration component as the system boundary condition at the drive roller into this differential equation. The system uses a finite difference algorithm to perform spatial and temporal meshing of the entire belt conveyor, iteratively solves the differential equation, and obtains the longitudinal strain sequence of discrete nodes along the entire conveyor belt.

[0047] After obtaining the longitudinal strain sequence, the system multiplies it by the dynamic stiffness coefficient of the conveyor belt material and simultaneously calculates the time-varying rate of change of the longitudinal strain sequence, multiplies it by the viscous damping coefficient, and adds the two together to generate the time-domain response waveform of the longitudinal tension at each node location. The formula for calculating the viscoelastic tension is as follows: ; In the formula, This indicates the longitudinal tension value at the position of the s-th discrete node at the current discrete time step n; n represents the index number of the discrete time step in the finite difference solution process; s represents the spatial index number of the discrete node in the finite difference solution process, and its value range is a positive integer greater than or equal to one and less than or equal to the total number of nodes in the entire grid. This indicates the preset dynamic stiffness coefficient of the conveyor belt; This represents the longitudinal strain value at the s-th discrete node at the current discrete time step n; This indicates the preset viscous damping coefficient of the conveyor belt; This represents the longitudinal strain value at the s-th discrete node in the previous discrete time step n-1; This represents the discrete time step in the finite difference solution process.

[0048] To accurately capture the excitation frequency that induces resonance in the conveyor belt, the system selects a preset characteristic node at the junction of the drive rollers of the belt conveyor as a constant observation point. A fixed-length historical time window, ending at the current discrete time step, is extracted. The time-domain response waveform of the longitudinal tension output by this preset characteristic node is then subjected to a sliding-window discrete Fourier transform with amplitude normalization. The calculation formula is as follows: ; In the formula, Represents discrete frequency components The corresponding actual physical tension amplitude; Represents the discrete frequency independent variable after frequency domain transformation; The length of the time-domain waveform sampling window L is defined, and its range of values ​​limits the total number of discrete time steps involved in the transformation. The local historical time backward index number within the j-moving window has a value range of integers that are greater than or equal to zero and less than the sampling window length; This indicates the preset feature nodes when tracing back j steps in the historical direction from the current discrete time step. The corresponding longitudinal tension value; The spatial index number of the preset feature node at the junction of the drive rollers of the belt conveyor; e represents the exponent with the natural constant as the base; i represents the imaginary unit in mathematics; Pi is a constant. This represents the discrete time step in the finite difference solution process.

[0049] The system compares the actual physical tension amplitude corresponding to all discrete frequency components obtained by the solution with the preset tension fluctuation threshold, extracts the target frequency band whose tension amplitude is strictly greater than the tension fluctuation threshold, and extracts the center frequency of the target frequency band as the dynamic stopband center frequency that needs to be suppressed.

[0050] Based on the obtained dynamic stopband center frequency and the discrete sampling period of the control system, the system reconstructs the tap coefficients of the preset notch filter online. The system performs equally spaced time-discrete sampling on the reference speed trajectory generated by the previous module. Using the reconstructed notch filter, it performs time-domain convolution on the discrete-time sequence of the reference speed trajectory to filter out specific frequency components that can excite destructive longitudinal tension in the conveyor belt, generating the final output target speed command. The convolution calculation formula is as follows: ; In the formula, This represents the target speed command value corresponding to the z-th sampling point in the current control cycle; z indicates the time index number of the sampling point within the current control cycle; M is the pre-defined order of the notch filter; p represents the filter tap index number in the convolution summation operation, and its value range is an integer greater than or equal to zero and less than or equal to M. This indicates the dynamic stopband center frequency that needs to be suppressed, extracted from the target frequency band. Indicated by the center frequency of the dynamic stopband The dimensionless filter coefficients corresponding to the p-th head in the notch filter that is dynamically reconstructed with independent variables; This means that the reference velocity trajectory generated by the previous module is discretized according to the discrete sampling period of the control system, and the reference velocity value corresponding to the p-th sampling point is calculated backward in the historical direction.

[0051] This step transforms the idealized energy-saving reference speed into a physical control command that is fully adapted to the viscoelastic dynamics of the flexible conveyor belt, effectively eliminating the dangerous excitation frequency that could cause standing wave resonance in the system, completely blocking the risk of belt breakage during speed regulation, and ensuring the physical safety of the entire machine.

[0052] The variable frequency execution module, based on preset drive slippage threshold and variable frequency overload threshold, limits and filters the acceleration and deceleration slope of the target speed command, and generates and outputs the variable frequency control frequency signal.

[0053] Furthermore, the generation and output of the frequency conversion control signal in the frequency conversion execution module specifically includes the following steps: The first derivative of the target speed command is calculated to obtain the command acceleration. When the command acceleration exceeds the upper limit of acceleration determined by the drive slippage threshold and the frequency converter overload threshold, the amplitude of the command acceleration is truncated. The target velocity command is reconstructed by integrating the truncated command acceleration to obtain the reconstructed velocity command. The reconstructed speed command is multiplied by a preset speed-frequency conversion coefficient to generate a frequency control signal.

[0054] Specifically, the frequency converter execution module first extracts the discrete-time sequence of the target speed command generated by the previous module, and then performs a first-order differential derivative calculation on the target speed command in conjunction with the discrete sampling period of the control system. To avoid the integral drift error introduced by direct differential calculation, the system calculates the difference between the target speed command received in the current control cycle and the reconstructed speed command output in the previous control cycle, and divides it by the discrete sampling period to obtain the command acceleration required at the current moment. The calculation formula is as follows: ; In the formula, This represents the command acceleration value corresponding to the z-th sampling point in the current control cycle; z represents the time index number of the sampling point within the current control cycle, and its value range is a positive integer greater than or equal to one; This represents the target speed command value corresponding to the z-th sampling point of the current control cycle output by the previous module; This represents the reconfiguration speed command value corresponding to point z-1 of the previous control cycle. When the system is in the first control cycle of initial startup, i.e., z-1, and there is no historical data, this value is forced to be equal to the actual physical running line speed currently fed back by the belt conveyor. This represents the discrete sampling period of the control system, and its physical dimension is defined as seconds.

[0055] The system pre-extracts the drive slippage acceleration threshold required for physical anti-slip at the drive drum of the belt conveyor, and the maximum output torque of the frequency converter required for safe operation of the electrical frequency converter. The system extracts the maximum output torque of the frequency converter and multiplies it by the gear ratio of the reducer, then divides it by the product of the drive drum radius and the current equivalent translational mass of the entire belt conveyor. Based on Newton's second law, it converts this into an electrical-dimensional frequency converter overload acceleration threshold. The system selects the smaller value between the drive slippage acceleration threshold and the frequency converter overload acceleration threshold using an extreme value comparison function, reducing the electromechanical dual physical boundaries to generate the upper limit of the overall absolute acceleration. The constraint generation formula is as follows: ; In the formula, This represents the upper limit of the operating acceleration of the belt conveyor, which is jointly determined by the drive slippage threshold and the frequency converter overload threshold. In mathematics, a function for finding the minimum and extreme values; This indicates the preset drive slippage acceleration threshold to prevent relative sliding between the conveyor belt and the drive roller; This indicates the preset maximum output torque of the frequency converter; I indicates the preset gear ratio constant of the reducer; R indicates the preset physical radius of the drive roller; This represents the equivalent translational mass of the entire belt conveyor at the current moment, calculated based on real-time load.

[0056] The system compares the calculated command acceleration with the aforementioned upper limit of acceleration. When the absolute value of the command acceleration exceeds this upper limit, its amplitude is truncated. The truncation calculation formula is as follows: ; In the formula, This represents the command acceleration value at point z of the current control cycle after amplitude truncation.

[0057] After obtaining the truncated command acceleration, the system performs discrete integral reconstruction of the target velocity command in the time domain based on this truncated command acceleration. The system multiplies the truncated command acceleration by the discrete sampling period to obtain the velocity increment of the current control cycle, and accumulates it with the reconstructed velocity command retained in the previous control cycle to obtain the reconstructed velocity command that is actually allowed to be issued in the current control cycle. The integral reconstruction formula is as follows: ; In the formula, This represents the reconstructed speed command value obtained at the z-th sampling point of the current control cycle.

[0058] After completing the speed domain command reconstruction, the system constructs a high-precision electromechanical cross-domain mapping model based on the number of pole pairs of the belt conveyor drive motor, the transmission ratio of the reducer, the physical radius of the drive drum, and the rated electromagnetic slip rate of the asynchronous motor. The system substitutes the reconstructed speed command acquired in the current control cycle into this model for slip compensation and kinematic transformation, converting the linear velocity value in the mechanical kinematic dimension into the drive frequency in the electrical dimension, generating a frequency conversion control signal directly sent to the underlying hardware actuator. The conversion formula is as follows: ; In the formula, This represents the value of the frequency control signal generated and output to the inverter at the z-th sampling point in the current control cycle. p represents the preset number of pole pairs of the drive motor; Pi is a constant. This indicates the preset rated electromagnetic slip rate of the drive motor.

[0059] This step transforms the theoretical target speed command generated by the aforementioned module into a safe drive frequency signal that strictly conforms to the boundaries of motor overload prevention and drum anti-slipping. It also corrects the speed loss during operation through electromagnetic slip compensation, ensuring that the smooth speed regulation command that suppresses tension fluctuations can be accurately reproduced at the underlying physical frequency converter.

[0060] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An adaptive speed-regulating energy-saving belt conveyor control system, characterized in that, include: The material feature extraction module collects the contour and shape information of the material above the belt conveyor in real time, performs cross-sectional area integration calculation, and extracts and outputs real-time load parameters. The spatiotemporal distribution simulation module synchronously matches the real-time load parameters with the real-time operating displacement of the conveyor belt of the belt conveyor to generate and output a material spatial distribution matrix. The resistance assessment module, based on the material spatial distribution matrix and the pre-constructed discrete micro-element segment model of the belt conveyor, combines the conveyor belt indentation rolling resistance parameters and idler friction parameters to perform micro-element force analysis, and generates and outputs dynamic resistance maps. The decision module constructs an energy consumption model for the belt conveyor based on the dynamic resistance map, calculates the energy consumption objective function, and generates and outputs the reference speed trajectory. The tension control module performs longitudinal tension analysis and frequency band trajectory correction based on the reference speed trajectory and the preset viscoelastic dynamic equation of the conveyor belt, and generates and outputs the target speed command. The variable frequency execution module, based on the preset anti-slip acceleration threshold of the drive drum and the variable frequency overload torque threshold, limits and filters the acceleration and deceleration slope of the target speed command, and generates and outputs the variable frequency control frequency signal.

2. The adaptive speed regulation energy-saving belt conveyor control system according to claim 1, characterized in that, In the material feature extraction module, the extraction and output of real-time load parameters specifically includes the following steps: The raw point cloud data of the material above the belt conveyor is obtained, filtered, and the material contour curve is extracted. The material profile curve is integrated relative to the reference surface of the empty conveyor belt to obtain the instantaneous material cross-sectional area; The real-time load parameters are calculated by multiplying the instantaneous material cross-sectional area by a preset material bulk density coefficient.

3. The adaptive speed regulation energy-saving belt conveyor control system according to claim 1, characterized in that, In the spatiotemporal distribution deduction module, generating and outputting the material spatial distribution matrix specifically includes the following steps: Obtain the discrete time series of the real-time operating displacement of the conveyor belt; Using the discrete time series as a spatial index, the real-time load parameters are discretized and mapped to construct a one-dimensional spatial load vector; The displacement of the one-dimensional spatial load vector is updated over time based on the circular cycle length of the conveyor belt, and the material spatial distribution matrix is ​​generated by combining the results.

4. The adaptive speed regulation energy-saving belt conveyor control system according to claim 1, characterized in that, In the resistance assessment module, generating and outputting the dynamic resistance map specifically includes the following steps: Align the pre-constructed discrete micro-segment model of the belt conveyor with the material space distribution matrix according to the spatial coordinates to obtain the local load mass corresponding to each micro-segment; Substitute the local load mass into the preset conveyor belt indentation rolling resistance equation and idler friction force equation to calculate the local running resistance value of each micro segment. The local operating resistance values ​​are spatially sequenced along the entire length of the conveyor belt to generate a dynamic resistance map.

5. The adaptive speed regulation energy-saving belt conveyor control system according to claim 4, characterized in that, The step of substituting the local load mass into the preset conveyor belt indentation rolling resistance equation and idler friction equation to calculate the local running resistance value of each micro-segment specifically includes the following steps: Based on the local load mass, the preset elastic modulus of the conveyor belt and the structural parameters of the idler rollers, the vertical indentation depth sequence of the conveyor belt in each idler roller contact area is calculated; The vertical indentation depth sequence and the relaxation time constant of the conveyor belt are input into a preset generalized Maxwell viscoelastic model to calculate the dynamic indentation friction coefficient. The local load mass, the dynamic indentation friction coefficient, and the roller friction parameters are multiplied and added together to generate the local running resistance values ​​for each micro-segment.

6. The adaptive speed regulation energy-saving belt conveyor control system according to claim 1, characterized in that, In the decision-making module, generating and outputting the reference velocity trajectory specifically includes the following steps: Substitute the dynamic resistance map into the preset belt conveyor energy consumption model to construct an energy consumption objective function with the overall machine operating energy consumption as the dependent variable and the operating speed as the independent variable. Based on the preset upper and lower limits of the conveyor belt running speed constraints, the derivative of the energy consumption objective function is solved to obtain the discrete speed sequence corresponding to each preset load interval. The discrete velocity sequence is subjected to time-domain interpolation fitting to generate a baseline velocity trajectory.

7. The adaptive speed regulation energy-saving belt conveyor control system according to claim 1, characterized in that, In the tension control module, generating and outputting the target speed command specifically includes the following steps: Using the reference velocity trajectory as the displacement excitation condition, and substituting it into the preset viscoelastic dynamic equation of the conveyor belt, the time-domain response waveform of the longitudinal tension of the conveyor belt is calculated. The time-domain response waveform is transformed in the frequency domain to extract the target frequency band whose amplitude exceeds the preset tension fluctuation threshold; A preset notch filter is used to filter and correct the target frequency band in the reference velocity trajectory to generate a target velocity command.

8. The adaptive speed regulation energy-saving belt conveyor control system according to claim 7, characterized in that, The step of using the reference velocity trajectory as the displacement excitation condition and substituting it into the preset viscoelastic dynamics equation of the conveyor belt to calculate the time-domain response waveform of the longitudinal tension of the conveyor belt specifically includes the following steps: Using the acceleration components of the reference velocity trajectory as the system boundary conditions, the differential equation of the conveyor belt distributed parameters constructed based on the Kelvin-Woyt viscoelastic theory is solved by finite difference discrete solution to obtain the longitudinal strain sequence of discrete nodes along the entire conveyor belt. The longitudinal strain sequence is multiplied by the dynamic stiffness coefficient and viscous damping coefficient of the conveyor belt to generate the time-domain response waveform of the longitudinal tension at each node location.

9. The adaptive speed regulation energy-saving belt conveyor control system according to claim 1, characterized in that, In the frequency conversion execution module, generating and outputting the frequency conversion control signal specifically includes the following steps: The first derivative of the target speed command is calculated to obtain the command acceleration. When the command acceleration exceeds the upper limit of acceleration determined by the anti-slip acceleration threshold of the drive drum and the anti-overload torque threshold of the frequency converter, the amplitude of the command acceleration is truncated. The target velocity command is reconstructed by integrating the truncated command acceleration to obtain the reconstructed velocity command. The reconstructed speed command is multiplied by a preset speed-frequency conversion coefficient to generate a frequency conversion control signal.

Images