Workshop equipment intelligent operation regulation and control method and system based on load state

Abstract

The present application relates to the technical field of industrial control, and discloses a kind of workshop equipment intelligent operation regulation and control method and system based on load state, coal bunker surface point cloud, coal grinding motor stator current and active power time series data are collected, according to point cloud transient elevation difference determination coal bunker coal state, combined with triangulation and Simpson integral accurate calculation coal bunker effective coal volume.Volume reduction rate and current gradient are solved by time domain difference, combined with motor power to build basic consumption ratio, introduce resistance viscosity mapping function to complete loss gradient correction, then through linear extrapolation operator to solve coal material safe available time.When available time is lower than preset threshold, automatically issue replenishment instruction, linkage start coal belt frequency conversion operation, realize equipment intelligent regulation and control.The present application solves the problem that traditional material storage detection is not accurate and is prone to material breakage downtime, significantly improves the stability and intelligent control precision of workshop equipment operation.

Description

Technical Field

[0001] This invention relates to the field of industrial control technology, specifically to a method and system for intelligent operation control of workshop equipment based on load status. Background Technology

[0002] In the material distribution and circulation process of thermal power generation or large-scale industrial production workshops, the continuous and stable supply of materials from storage silos is the absolute core to ensure the overall production load and the safety of the system units.

[0003] Traditional workshop operation control mechanisms often rely on single-dimensional material height monitoring or setting simple drive motor current dead zone thresholds to determine the bottom-level material discharge situation and whether to activate the top-level replenishment equipment. However, under conditions of long-term storage of high-humidity, high-viscosity materials and frequent high-frequency load changes from the power grid, materials are prone to asymmetrical accumulation, localized wall hardening, or internal collapse within the silo. These complex hydrodynamic distortions result in a significant nonlinear deviation between the traditional single-point spatial height and the actual effective internal storage capacity.

[0004] Meanwhile, most existing control logics treat the three-dimensional spatial morphology at the top and the electrical dynamic characteristics of the mechanical discharge at the bottom in isolation, completely failing to extract in real time the sudden changes in physical resistance torque caused by humidity and compression of the material at the bottom. This makes it difficult for the control system to capture the actual consumption rate of the material when there are sudden changes in the discharge rate or large-area asymmetric collapse of the material, and thus it is unable to accurately predict the absolute time point when the actual usable material is completely exhausted. This can easily lead to delayed operation or fatal material shortage shutdown accidents in cascaded equipment in the workshop.

[0005] To address the aforementioned issues, there is an urgent need in this field for a data-driven control scheme that can deeply integrate spatial topographic and structural features with the electromagnetic physical state of underlying equipment. Summary of the Invention

[0006] In order to overcome the shortcomings of the existing technology, the purpose of this invention is to provide a method and system for intelligent operation control of workshop equipment based on load status, so as to solve the technical problem that the existing silo control cannot accurately measure the storage volume and is prone to material shortage and shutdown.

[0007] This invention is achieved through the following technical solution: In a first aspect, the present invention provides a method for intelligent operation control of workshop equipment based on load status, comprising: Obtain the point cloud set of the solid surface of the target coal bunker, the stator current time sequence of the coal mill drive motor, and the active power time sequence of the coal mill drive motor; The average transient elevation difference is calculated based on the point cloud set of the entity surface. Based on the relationship between the average transient elevation difference and the first preset threshold, it is determined whether the target coal bunker is in a coal-covered state. The point cloud set of the entity surface is subjected to Delaunay triangulation with preset constraint edges to obtain the entity connected domain point cloud mesh, and then the effective coal volume of the target coal bunker is obtained by Simpson double integral volume integral accumulation calculation. The effective coal volume is calculated using a first-order finite difference time-domain method to obtain the time reduction rate of the effective coal volume; the stator current time series of the coal mill drive motor is calculated using a first-order finite difference method to obtain the transient current gradient; and the basic consumption ratio is determined based on the ratio of the time reduction rate to the active power time series of the coal mill drive motor. The transient current gradient is input into a preset resistance viscosity mapping function to obtain a viscosity correction coefficient. The viscosity correction coefficient is then used to perform a product correction on the base consumption ratio to obtain the stock depletion time-series gradient. The effective coal volume and the time gradient of stock depletion are input into a first-order linear extrapolation operator for division extrapolation to obtain the safe available time characterizing the absolute depletion of coal volume. If the safe available time is less than or equal to the preset constraint time, a start feeding command is output to drive the cascade frequency converter of the coal conveyor belt in the target coal bunker to start, thus completing the intelligent operation control of the workshop equipment.

[0008] Preferably, the specific process of performing Delaunay triangulation of the point cloud set of the entity surface with preset constraint edges includes: Obtain the preset fixed elevation constant at the physical boundary between the upper cylindrical and lower conical parts of the target coal bunker, which characterizes the coal pile. Input the point cloud set of the entity surface into a preset three-dimensional coordinate filter, traverse and extract all boundary sampling points whose spatial Z-axis coordinates are equal to a preset fixed elevation constant, and output the boundary sampling point set. The set of boundary sampling points is input into the least squares circle fitting operator to fit the radius and center spatial coordinates, output the structural fracture edge of the spatial closed loop and lock it, and the structural fracture edge is recorded as the preset constraint edge.

[0009] Preferably, the specific process for obtaining the transient current gradient by performing first-order finite difference calculation on the stator current time series of the coal mill drive motor includes: The stator current timing sequence of the coal mill drive motor is input into a first-order discrete difference time-domain converter. The first-order numerical difference processing is performed on two adjacent current timing sampling points according to a preset fixed sampling period, and the transient current gradient is output.

[0010] Preferably, the specific process of obtaining the viscosity correction coefficient by inputting the transient current gradient into a preset resistance viscosity mapping function includes: The preset drag viscosity mapping function is a first-order polynomial calibration equation for drag viscosity with transient current gradient as independent variable and viscosity correction coefficient as dependent variable. The transient current gradient of the current time series is substituted as an independent variable into the first-order polynomial equation. After performing algebraic multiplication with the preset viscosity sensitivity proportionality coefficient, and then superimposed with the preset reference zero-point bias constant, the calculated dependent variable value is the viscosity correction coefficient.

[0011] Preferably, the specific processing procedure for inputting the effective coal volume and the time gradient of stock depletion into a first-order linear extrapolation operator for division extrapolation to obtain the safe available time characterizing the absolute depletion of coal volume includes: Based on the principle of first-order linear extrapolation, a first-order linear evolution equation is constructed with the future prediction time as the independent variable and the coal volume at the future time as the dependent variable. Set the current effective coal volume as the initial intercept constant of the equation, and set the negative value of the time gradient of stock depletion as the slope parameter of the first derivative of the equation. Setting the boundary condition of the dependent variable of the equation to zero characterizes the ultimate state of complete depletion of the target coal bunker material. By solving the equation through algebraic rearrangement, the effective coal volume is divided by the absolute value of the stock depletion time gradient, and the resulting value is the safe available time.

[0012] Preferably, before outputting the start material replenishment command, a status check is also included, the specific process of which is as follows: Extract the power consumption token state vector and the operating status feature bits of the coal conveyor belt in the distributed workshop network of the target coal bunker; When the power token state vector is determined to be in a locked waiting state and the running state feature bit is in a standby pause state, the output start feeding command is executed to drive the cascade frequency converter start of the coal feeding belt of the target coal bunker.

[0013] Preferably, after outputting the start material replenishment command, the constraint edges are also dynamically updated, and the specific process is as follows: The point cloud set of the entity surface is used to calculate the partial derivative of the tangent plane using a local surface least squares plane fitting algorithm based on k-nearest neighbors, and the spatial height gradient vector corresponding to each sampling point is output. When the magnitude of the spatial height gradient vector is greater than the preset cliff shape determination threshold, the current working condition is determined to be an asymmetric local collapse working condition, and the following steps are performed: Sampling points larger than the cliff morphology determination threshold are selected to form a set of feature edge points of the slip surface; The set of feature edge points of the slip surface is reconstructed by concatenation using a nearest neighbor edge tracking algorithm based on Euclidean distance, and the dynamic constraint edge of the slip fault in the form of a polyline segment is output. The point cloud mesh of the entity connected domain in the current time series is traversed by a line segment intersection detection algorithm based on vector outer product, and the set of intersecting topological edges that geometrically intersect with each polyline segment of the dynamic constraint edge of the slip fault is retrieved and extracted. The set of intersecting topological edges is removed from the point cloud mesh of the entity connected domain by the mesh topology adjacency deletion algorithm, exposing the polygon hole domain; The dynamic constraint edge of the slip fault is used as a rigid, impenetrable boundary and fixedly embedded in the polygonal cavity domain, dividing the polygonal cavity domain into a left closed polygonal domain and a right closed polygonal domain. For the closed polygonal domains on the left and right sides respectively, the local topological mesh reconstruction is carried out by the mesh retriangulation algorithm based on the constrained empty circle criterion, thus completing the in-situ topological rewriting of the point cloud mesh of the entity connected domain. By using a one-dimensional dynamic array appending cascade algorithm, the polyline data matrix of the dynamic constraint edge of the slip fault is merged and written into the tail of the original structural fracture edge storage matrix, and the updated preset constraint edge is output for the Deloitte triangulation processing in the next time series.

[0014] Furthermore, after outputting the updated preset constraint edges, the process also includes volume data smoothing, as detailed below: The effective coal volume output continuously in time is input into a preset sliding short window. The time-domain variance of the effective coal volume is calculated using a sliding window statistical variance algorithm. If the time-domain variance is greater than the preset grid jitter tolerance, the current operating condition is determined to be an unbalanced high-frequency numerical jitter condition, and the following steps are executed: The digital low-pass Butterworth filtering algorithm using the bilinear transform method is used to filter and truncate all effective coal volumes within the sliding short window, eliminating high-frequency discrete numerical components caused by local switching of the grid topology, and outputting a smooth effective average coal volume. The effective coal volume of the current time series is replaced by the smoothed effective average coal volume.

[0015] Furthermore, after replacing the effective coal volume in the current time series, it also includes safety time compensation control, the specific process of which is as follows: The rate of change of the active power time series of the coal mill drive motor is extracted by the first-order backward finite difference algorithm, and the load change rate feature value is output. If the characteristic value of the load change rate is greater than the preset lower threshold for rapid material discharge, the current operating condition is determined to be a high-lag, undercompensated operating condition, and the following steps are executed: Multiply the characteristic value of the load variation rate by the preset phase lag constant corresponding to the digital low-pass Butterworth filter algorithm of the bilinear transform method to obtain the safe time deviation compensation amount; The corrected safe available time is obtained by subtracting the safe time deviation compensation from the safe available time output by the first-order linear extrapolation operator; If the corrected safe available time is less than or equal to the preset constraint time, a start-up feeding command carrying the target coal bunker identification code and the preset frequency converter start control word is generated through the preset industrial control network instruction code, and sent to the coal conveyor frequency converter controller to drive the coal conveyor cascade frequency converter start. If the revised safe available time is greater than the preset constraint time, the coal conveyor belt will remain in standby and suspended state.

[0016] Secondly, the present invention also provides an intelligent operation control system for workshop equipment based on load status, comprising: The data acquisition module acquires the point cloud set of the physical surface of the target coal bunker, the stator current time sequence of the coal mill drive motor, and the active power time sequence of the coal mill drive motor. The coal bunker determination module calculates the average transient elevation difference based on the point cloud set of the entity surface, and determines whether the target coal bunker is in a coal bunker state based on the relationship between the average transient elevation difference and a first preset threshold. The volume calculation module performs Delaunay triangulation processing on the point cloud set of the entity surface with preset constraint edges to obtain the point cloud mesh of the entity connected domain, and then calculates the effective coal volume of the target coal bunker by accumulating the volume integrals of the Simpson double integral. The feature calculation module performs a first-order finite difference calculation in the time domain on the effective coal volume to obtain the time loss rate of the effective coal volume; performs a first-order finite difference calculation on the stator current time series of the coal mill drive motor to obtain the transient current gradient; and determines the basic consumption ratio based on the ratio of the time loss rate to the active power time series of the coal mill drive motor. The gradient correction module inputs the transient current gradient into a preset resistance viscosity mapping function to obtain a viscosity correction coefficient. The viscosity correction coefficient is then used to perform a product correction on the base consumption ratio to obtain the stock depletion time-series gradient. The time calculation module inputs the effective coal volume and the time gradient of stock depletion into a first-order linear extrapolation operator for division extrapolation calculation to obtain the safe available time characterizing the absolute depletion of coal volume. If the safe available time is less than or equal to the preset constraint time, the equipment control module outputs a start feeding command to drive the cascade frequency converter of the coal conveyor belt in the target coal bunker to start, thus completing the intelligent operation control of the workshop equipment.

[0017] Compared with the prior art, the present invention has the following beneficial technical effects: This invention provides a method for intelligent operation control of workshop equipment based on load status. It collects point cloud data of the target coal bunker surface and time-series operating data of the current and power of the coal mill drive motor. It abandons the traditional single-point ranging detection mode and uses Delaunay triangulation with preset constraint edges combined with Simpson double integral volume integral accumulation calculation method. It can accurately adapt to the actual working conditions of wet and sticky coal, irregular stacking, asymmetric distribution, and complex morphological distortion inside the bunker. It effectively overcomes the problems of traditional detection methods being unable to characterize the complex morphology inside the coal bunker and the distortion of volume calculation. It realizes high-precision three-dimensional reconstruction and accurate measurement of the effective coal volume of the coal bunker, providing a precise data foundation for equipment control.

[0018] Furthermore, this invention integrates a time-domain difference algorithm to calculate the time loss rate of effective coal volume and the transient current gradient of motor operation, respectively. Based on a preset resistance-viscosity mapping function, a viscosity correction coefficient is generated to correct and optimize the basic consumption ratio. This achieves cross-domain fusion of the macroscopic discharge loss characteristics of coal bunker materials and the microscopic electromagnetic adhesion resistance characteristics of the motor. It completely changes the shortcomings of traditional equipment control that relies solely on load data monitoring and has a single parameter dimension, significantly improves the calculation accuracy of the stock depletion time-series gradient, and effectively solves the technical problem of large deviations between the predicted material discharge results and the actual working conditions.

[0019] Furthermore, by using a first-order linear extrapolation operator to solve the effective coal volume and the time gradient of stock depletion, the safe available time for complete depletion of coal bunker materials is accurately predicted. Based on this prediction result, the cascade frequency conversion feeding control command of the coal conveyor belt is triggered, breaking the traditional control mode of passive alarm and post-event remediation. This completes the technical upgrade to active feedforward prediction control, greatly improving the continuity of material supply of the workshop's coal milling equipment under high-frequency variable load and complex material conditions, effectively avoiding material shortage shutdowns, and significantly enhancing the operational stability and operational condition adaptability of the entire workshop equipment control system. Attached Figure Description

[0020] Figure 1 This is a flowchart of the intelligent operation control method for workshop equipment based on load status in an embodiment of the present invention; Figure 2 This is a logic diagram of the intelligent operation control method for workshop equipment based on load status in an embodiment of the present invention; Figure 3 This is a schematic diagram of the intelligent operation and control system for workshop equipment based on load status in an embodiment of the present invention; In the diagram: 1. Data acquisition module; 2. Coal briquette identification module; 3. Volume calculation module; 4. Feature calculation module; 5. Gradient correction module; 6. Time calculation module; 7. Equipment control module. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0022] The purpose of this invention is to provide a method and system for intelligent operation control of workshop equipment based on load status. Under specific material storage and supply conditions where complex wet and sticky coal causes asymmetric distortion of materials and drastic fluctuations in the discharge load of the bottom motor, the prediction of the absolute depletion time of materials is seriously delayed and distorted due to the disconnect between three-dimensional spatial volume monitoring and the electromagnetic physical characteristics of the bottom layer, which in turn leads to the technical problem of material shortage and shutdown of cascade feeding equipment.

[0023] The present invention will now be described in further detail with reference to the accompanying drawings: Example 1 See Figure 1 and Figure 2 In one embodiment of the present invention, a method for intelligent operation control of workshop equipment based on load status is provided, comprising: Step 1: Obtain the point cloud set of the solid surface of the target coal bunker, the stator current time sequence of the coal mill drive motor, and the active power time sequence of the coal mill drive motor; Step 2: Calculate the average transient elevation difference based on the point cloud set of the entity surface, and determine whether the target coal bunker is in a coal-covered state based on the relationship between the average transient elevation difference and the first preset threshold. Step 3: Perform Delaunay triangulation on the point cloud set of the entity surface with preset constraint edges to obtain the point cloud mesh of the entity connected domain. Then, calculate the effective coal volume of the target coal bunker by accumulating the volume integrals of the Simpson double integral. Step 4: Perform first-order finite difference calculation in the time domain on the effective coal volume to obtain the time reduction rate of the effective coal volume; perform first-order finite difference calculation on the stator current time series of the coal mill drive motor to obtain the transient current gradient; determine the basic consumption ratio based on the ratio of the time reduction rate to the active power time series of the coal mill drive motor. Step 5: Input the transient current gradient into the preset resistance viscosity mapping function to obtain the viscosity correction coefficient, and use the viscosity correction coefficient to perform product correction on the basic consumption ratio to obtain the stock depletion time gradient. Step 6: Input the effective coal volume and the time gradient of stock depletion into the first-order linear extrapolation operator for division extrapolation solution to obtain the safe available time characterizing the absolute depletion of coal volume; Step 7: If the safe available time is less than or equal to the preset constraint time, then output a start feeding command to drive the cascade frequency converter of the coal conveyor belt of the target coal bunker to start, and complete the intelligent operation control of the workshop equipment.

[0024] In the initial stage of execution, the system acquires a point cloud set of the target coal bunker's physical surface, a stator current time series of the coal mill drive motor, and a power time series of the coal mill drive motor. The physical surface point cloud set contains a large number of discrete three-dimensional spatial coordinate points, each in meters, used to reconstruct the geometric morphology of the material surface within the bunker with high fidelity. The stator current time series, in amperes, reflects the transient electromagnetic resistance of the motor rotor to the shear-viscosity friction of the bottom material. The power time series, in kilowatts, characterizes the real-time energy consumption intensity of the main pulverizer. The high-frequency electrical parameters of the stator current and power are collected strictly following the Nyquist sampling theorem, with a sampling period set to 20 milliseconds, and are written to memory in real time via an industrial fieldbus protocol. After acquiring the raw data, the average transient elevation difference between the geometric center point cloud elevation and the elevations of each edge point cloud is calculated. If the transient elevation difference is greater than a first preset threshold, the state is determined to be coal briquetted. The first preset threshold is the spatial distribution discrimination boundary obtained by nonlinear regression analysis of historical coal deformation data during the rainy season. The unit is meters, and its value range is set from 0.5 meters to 1.5 meters. The typical value in this embodiment is 0.8 meters. When the elevation difference exceeds the threshold, it indicates that a serious central tubular cavitation has occurred inside the material.

[0025] After confirming the occurrence of physical distortion, the point cloud set of the solid surface is processed by Delaunay triangulation according to the preset constraint edges to obtain a point cloud mesh of the solid connected domain. The effective coal volume is then calculated by accumulating the volume integrals using Simpson's double integral. Delaunay triangulation is a well-known spatial meshing technique in computational geometry. Its core operation logic is to maximize the minimum interior angle of all triangles to avoid elongated distortion. It takes discrete point cloud coordinates as input and outputs an optimal topological mesh connected graph that satisfies the characteristics of an empty circle. Simpson's double integral is a standard integration method in numerical analysis. It approximates the area of ​​a infinitesimal element using a quadratic polynomial and extends it to a two-dimensional spatial mesh. The envelope space of the previously output point cloud mesh is then accumulated for calculation. Through this dual mathematical operator processing, the effective coal volume is accurately output in cubic meters, completely eliminating hollow and wall-mounted interference areas.

[0026] To further extract the dynamic loss trend, the effective coal volume is calculated using a first-order finite-difference time-domain method, outputting the time-dependent loss rate of the effective coal volume. The stator current time series of the coal mill drive motor is also calculated using a first-order finite-difference method, outputting the transient current gradient. The ratio of the time-dependent loss rate to the active power time series of the coal mill drive motor is denoted as the basic consumption ratio. The first-order finite-difference time-domain calculation extracts the instantaneous rate of change by dividing the difference between the current value and the previous value by a fixed sampling period. The time-dependent loss rate is measured in cubic meters per second, the transient current gradient in amperes per second, and the basic consumption ratio in cubic meters per kilowatt-second, representing the rate of basic material consumption per unit of energy.

[0027] Subsequently, an electrophysical correction mechanism is introduced. The transient current gradient is input into a preset resistance viscosity mapping function, and the viscosity correction coefficient is output. The basic consumption ratio and the viscosity correction coefficient are multiplied for correction, and the stock depletion time series gradient is output. The effective coal volume and the stock depletion time series gradient are input into a first-order linear extrapolation operator for division extrapolation calculation, and the safe available time representing the absolute depletion of coal volume is output.

[0028] The physical dimension of the viscosity correction factor is configured as kilowatts (kW). After algebraic product correction with the basic consumption ratio in cubic meters per kilowatt-seconds, the dimension of the stock depletion time series gradient is strictly reduced and aligned to cubic meters per second, representing the global equivalent depletion rate that integrates the dual loss characteristics of material space loss and electro-adhesive resistance.

[0029] The first-order linear extrapolation operator uses the effective coal volume in cubic meters as the numerator and the stock depletion time gradient in cubic meters per second as the denominator. According to the algebraic operation rules, the output safe available time dimension is strictly aligned to seconds.

[0030] Finally, in the control execution phase, if the safe available time is less than or equal to the preset constraint time, a start-up material replenishment command is output to drive the cascaded frequency converter start of the coal conveyor belt in the target coal bunker. The preset constraint time is a rigid red line derived from the mechanical and physical lag time of the coal conveyor belt from cold start to rated speed plus the material flow time throughout the entire chain, measured in seconds, and typically ranging from 30 to 90 seconds, with a typical configuration value of 60 seconds.

[0031] By introducing constrained Delaunay triangulation and Simpson integral, the measurement limitations of traditional single liquid level probes under linear assumptions are completely eliminated, effectively restoring the real material volume caused by non-Newtonian fluid characteristics.

[0032] Building upon this foundation, the scheme innovatively integrates the spatial volume loss state with the current and power sequences characterizing the electromagnetic properties of the underlying mechanical discharge, perfectly mapping the increased frictional resistance of the underlying layer due to wet and sticky coal using a viscosity correction coefficient. This complete data processing link completely solves the decision distortion problem caused by the inability of traditional control networks to perceive spatial topological distortions due to their adherence to static linear rules. It enables proactive feedforward issuance of anti-disruption commands under complex high-load consumption conditions, greatly improving the global operational robustness and physical safety of distributed storage and supply networks in extreme environments.

[0033] Furthermore, before outputting the start-up material replenishment command, a status verification step is also included: extracting the power token status vector of the target coal bunker in the distributed workshop network and the operating status feature bit of the coal conveyor belt; when it is determined that the power token status vector is in a locked waiting state and the operating status feature bit is in a standby pause state, the start-up material replenishment command is executed to drive the cascade frequency converter start of the coal conveyor belt of the target coal bunker.

[0034] The power token state vector is a core dynamic flag array that represents the allocation of permissions at the plant-wide macro-energy flow scheduling center. It is obtained in real time through the industrial Ethernet protocol.

[0035] The locked-wait state indicates that the current Boolean value of the vector is zero, which means that the control has actively deprived the device of its power supply rights; the standby pause state indicates that the belt conveyor drive frequency converter output frequency is zero Hertz and the machine is in a stopped and lingering state.

[0036] The status acquisition period for the aforementioned dynamic flag is set to 10 milliseconds to ensure absolute real-time status synchronization. From the perspective of the overall solution chain, defining this operating condition defines the extreme entry boundary of proactive intervention technology, enabling the highest level of interruption response rights to the underlying real physical exhaustion crisis at the most dangerous moment when the network is completely blocked by high-level scheduling instructions.

[0037] This step precisely identifies the sleep blind spots of underlying devices when network timing is limited. Through the above technical solution, this embodiment solves the specific technical problem of loss of underlying fuel supply authority caused by blind blocking of the macro-scheduling network under high load conditions by defining and strictly monitoring the concurrent logic of the power token state vector and the physical standby state. It achieves the unique technical effect of forcibly breaking through communication restrictions and ensuring that emergency refueling commands are absolutely delivered when the network is blocked in critical high-price segments.

[0038] Furthermore, in the scenario of reconstructing irregular coal bunker spaces, due to the geometric abrupt changes at the junction of the upper and lower cavities of the target coal bunker, conventional meshing algorithms tend to over-smooth the junction, resulting in severe volume redundancy errors.

[0039] To address this issue, this embodiment defines the specific processing procedure for the preset constraint edge as follows: Obtain a preset fixed elevation constant representing the physical boundary between the cylindrical upper part and conical lower part of the coal pile in the target coal bunker; input the point cloud set of the solid surface into a preset 3D coordinate filter, traverse and extract all boundary sampling points whose spatial Z-axis coordinates are equal to the preset fixed elevation constant, and output the boundary sampling point set; input the boundary sampling point set into a least-squares circular fitting operator to fit the radius and center spatial coordinates, output the structural fracture edge of the spatial closed loop and latch it, and record the structural fracture edge as the preset constraint edge. The preset fixed elevation constant is a rigid geometric position parameter obtained by analyzing the original architectural structural drawings of the target coal bunker, in meters, with a value range typically between 10 and 20 meters, representing the absolute physical boundary line where the bunker's shape undergoes a structural abrupt change. The preset 3D coordinate filter is a digital filtering program based on Z-axis spatial coordinate threshold truncation, with the rule that data points with an absolute deviation within ±5 centimeters are allowed to pass through the filter.

[0040] The least squares circle fitting operator is a classic algorithm in the field of computational geometry. Its input is a set of three-dimensional coordinates of scattered boundary sampling points extracted in the previous step. By solving the objective function that minimizes the sum of the squared errors of the distances from all sampling points to the preset circle center, the output is the optimal fitted circle center coordinates and radius trajectory.

[0041] In the complete technical chain, this step provides an irreplaceable foundational reinforcement for the core 3D reconstruction algorithm. Without pre-defined constraint edges, the subsequent integral volume will inevitably expand outwards and become distorted. For this specific scenario, this mechanism forcibly transforms fixed physical building parameters into insurmountable walls in the digital twin model.

[0042] Through the above technical solution, this embodiment solves the specific technical problem that the junction of the upper and lower cavities of the irregularly shaped compartment is easily over-smoothed by the grid algorithm, resulting in a false volume height, by introducing a combined mechanism of preset fixed elevation constant truncation and least squares circular space fitting. It achieves the unique technical effect of high-fidelity mapping of the rigid boundary of the physical building to the topological constraint boundary of the digital space.

[0043] Furthermore, for the electromagnetic feature extraction scenario of the bottom material discharge equipment, since the motor base load is often accompanied by low-frequency static fluctuations, directly using the absolute value of the current to extrapolate viscosity will lead to evaluation failure.

[0044] This embodiment defines the specific process for obtaining the transient current gradient as follows: the stator current time sequence of the coal mill drive motor is input into a first-order discrete difference converter in the time domain. Numerical first-order difference processing is performed on two adjacent current time sequence sampling points according to a preset fixed sampling period, and the transient current gradient is output. The preset fixed sampling period is set based on the data refresh rate of the field motor driver communication bus, and its value is constant at 20 milliseconds.

[0045] As a conventional digital signal processing module in industrial control, the first-order discrete-time differential converter in the time domain has the core logic of reading the current value of the current cycle, subtracting the current value of the previous cycle, dividing by a fixed sampling period, and outputting the instantaneous rate of change of the current physical quantity on the time axis.

[0046] From an overall technical perspective, this step is a crucial mathematical transformation link connecting macroscopic volumetric losses with underlying microscopic electrical load characteristics, providing a unique, clean electrical benchmark free from static noise for the subsequent accurate calculation of the viscosity correction coefficient. From an independent technical perspective, this operation effectively shields the evaluation of frictional resistance from the contamination caused by normal no-load fluctuations in the equipment.

[0047] Through the above technical solution, this embodiment performs high-frequency numerical differential processing on the current time sequence by using a first-order discrete differential converter in the time domain. This solves the specific technical problem that the traditional method of using the absolute value of the current is easily affected by the background noise of the motor foundation and cannot accurately assess the degree of obstruction of the actual material discharge. It achieves the unique technical effect of stripping away static load interference and extracting the transient viscous electrical variables of the material.

[0048] Furthermore, this embodiment is a deeper progression based on the previously output structural fracture edge. In the scenario of gravity flow of wet, sticky coal, the material often undergoes random asymmetric avalanche collapse, generating a near-vertical dynamic cliff.

[0049] Fixed building structure constraints cannot track randomly generated material slip faults, causing the mesh algorithm to smooth the cliff into a gentle slope again, resulting in local volume calculation distortion. Therefore, this embodiment, after outputting the material replenishment command, also includes: calculating the partial derivative of the tangent plane using a k-nearest neighbor-based local surface least squares plane fitting algorithm for the entity surface point cloud set, and outputting the spatial height gradient vector corresponding to each sampling point; when the magnitude of the spatial height gradient vector is greater than a preset cliff morphology judgment threshold, the current working condition is determined to be an asymmetric local collapse condition, and the following steps are performed: filtering and outputting sampling points greater than the cliff morphology judgment threshold as a slip surface feature edge point set; reconstructing the slip surface feature edge point set through a nearest neighbor edge tracking algorithm based on Euclidean distance, outputting the slip fault dynamic constraint edge in the form of a polyline segment; traversing the entity connected domain point cloud mesh of the current time series using a line segment intersection detection algorithm based on vector outer product, retrieving and extracting the set of intersecting topological edges that geometrically intersect with each polyline segment in the slip fault dynamic constraint edge; and then performing mesh topology... The adjacency deletion algorithm removes all topological edges from the set of intersecting topological edges in the point cloud mesh of the entity connected domain, breaking the original triangular topological relationships in the intersection region and exposing open polygonal voids. The dynamic constraint edges of the slip fault are fixedly embedded as rigid, impenetrable boundaries into the polygonal voids to physically separate the polygonal voids into left-side closed polygonal domains and right-side closed polygonal domains. The left-side and right-side closed polygonal domains are respectively reconstructed locally using a mesh retriangulation algorithm based on the constraint empty circle criterion to generate a new triangular mesh filled with polygonal voids, completing the in-situ topological rewriting of the entity connected domain point cloud mesh. The polyline data matrix of the dynamic constraint edges of the slip fault is merged into the tail of the storage matrix of the existing structural fracture edges using a one-dimensional dynamic array appending cascade algorithm, and the updated preset constraint edges are output for the Deloitte triangulation processing in the next time series.

[0050] The local surface least squares plane fitting algorithm based on k-nearest neighbor takes scattered local 3D coordinates as input and outputs the tangent derivative of each node by approximating the infinitesimal plane; the nearest neighbor edge tracking algorithm based on Euclidean distance uses the shortest distance between points to output closed polygonal polylines; the line segment intersection detection algorithm based on vector cross product uses the sign of the cross product to determine the geometric penetration relationship; the mesh retriangulation algorithm based on the constrained empty circle criterion is specifically designed to handle polygonal holes with embedded rigid boundaries and regenerates triangular mesh clusters that do not intersect on both sides.

[0051] The preset cliff topography judgment threshold is a dimensionless spatial gradient limit parameter calibrated based on the non-Newtonian fluid repose angle limit test, set to 1.732. This value corresponds to the tangent limit value of a spatial inclination angle of 60 degrees. During the dynamic rewriting process, the one-dimensional dynamic array appending cascade algorithm performs array splicing operations in the matrix memory, ensuring that each topology rewriting strictly converges within one control cycle (50 milliseconds). The termination condition is set when all polygonal voids are completely filled with triangular meshes and there are no suspended or free edge points. After the calculation is completed, the cache is released to avoid memory overflow.

[0052] As a progressive subdivision scheme, this embodiment progresses from solving static "dead constraints" to solving dynamic "live constraints." This embodiment is specifically designed to deal with sudden and severe working conditions such as asymmetric avalanche collapse, completely blocking soft topological stretching distortion under extreme asymmetric collapse conditions.

[0053] Through the above technical solution, this embodiment solves the specific refinement problem of the volume falsely high caused by the excessive smoothing of the steep slip surface formed by the traditional static mesh during the asymmetric collapse of wet and sticky materials by introducing a joint topology rewriting architecture of gradient vector solution and adaptive incremental insertion of dynamic constraint edges. It achieves the unique technical effect of millisecond-level precise wrapping and adaptive recombination of complex abrupt physical faults by digital mesh.

[0054] Furthermore, the frequent insertion and deletion of dynamic constraint edges inevitably leads to frequent disconnection and reconnection of local mesh connections, which in turn causes the volume data calculated by integration to exhibit high-frequency numerical jumps. This pseudo-oscillation seriously interferes with the accuracy of subsequent time-domain differential calculations.

[0055] This embodiment, after defining the preset constraint edges after outputting the updated values, further includes: inputting the effective coal volume of the continuous time-series output into a preset sliding short window, calculating the time-domain variation variance of the output effective coal volume using a sliding window statistical variance algorithm; if it is determined that the time-domain variation variance is greater than a preset grid jitter tolerance, then the current operating condition is determined to be an unbalanced high-frequency numerical jitter operating condition, and the following steps are performed: filtering and truncating all the effective coal volumes within the preset sliding short window using a bilinear transform digital low-pass Butterworth filtering algorithm to remove high-frequency discrete numerical components caused by local grid topology switching, and outputting a smoothed effective average coal volume; replacing the current time-series effective coal volume with the smoothed effective average coal volume.

[0056] The preset sliding window is a dynamic time window used to constrain the length of real-time data stream statistics. The length is set to 5 seconds, and the queue is updated using a first-in, first-out (FIFO) mechanism.

[0057] The preset grid jitter tolerance is a dimensionless coefficient calculated in advance under steady-state conditions, representing the upper limit of the natural volume fluctuation variance. A typical configuration uses 1.5 times the base fluctuation. The bilinear transform digital low-pass Butterworth filter algorithm follows standard digital signal processing specifications. Its difference equation is constructed using a discretized continuous transfer function, exhibiting the characteristic of maximum flat amplitude-frequency response within the passband.

[0058] The convergence logic of this filtering step is to reduce the residual to the steady-state allowable range within a maximum of 3 sliding window periods after the volume signal undergoes a step change, and continuously output a smooth and effective average coal volume during this period.

[0059] In terms of progressive linkage logic, this embodiment is specifically designed to smooth out the secondary noise introduced by dynamic mesh reconstruction. Through the above technical solution, this embodiment solves the specific technical problem of discrete pseudo-fluctuations in the integral volume caused by the drastic switching of local mesh topology due to the high-frequency insertion of dynamic constraint edges by using the combined application of sliding variance determination and bilinear transform digital low-pass Butterworth filtering algorithm. It achieves the unique technical effect of effectively filtering out high-frequency computational noise while maintaining the true trend of macroscopic entity loss of the volume.

[0060] Furthermore, this embodiment addresses the accompanying defects arising from low-pass filtering through a closed-loop progressive approach. Filtering in signal processing inevitably introduces group delay in the time domain.

[0061] When the underlying equipment is in a rapid material discharge condition, a filtering delay of tens of seconds will result in an excessively long output safe availability time and a severely delayed warning. Therefore, this embodiment, after replacing the effective coal volume of the current time series with the smoothed effective average coal volume, further includes: extracting the rate of change of the active power time series of the coal mill drive motor using a first-order backward finite difference algorithm, and outputting a load change rate characteristic value; if the load change rate characteristic value is determined to be greater than a preset rapid material discharge lower limit threshold, then the current operating condition is determined to be a high-lag, undercompensated operating condition, and the following steps are performed: multiplying the load change rate characteristic value with a preset phase lag constant corresponding to the bilinear transform digital low-pass Butterworth filter algorithm using a multiplicative algebra algorithm, and outputting a safe availability time. The full-time deviation compensation amount is calculated by subtracting the safe time deviation compensation amount from the safe available time output by the first-order linear extrapolation operator using a subtraction algebra algorithm, and outputting the corrected safe available time. If the corrected safe available time is less than or equal to the preset constraint time, a start-up feeding command carrying the target coal bunker identification code and the preset frequency converter start control word is generated through the preset industrial control network instruction code, and the start-up feeding command is sent to the frequency converter controller of the coal conveyor belt to drive its cascaded frequency converter start. If the corrected safe available time is greater than the preset constraint time, the current standby pause state of the coal conveyor belt is maintained.

[0062] The preset rapid discharge lower limit threshold is a power decrease slope boundary calibrated based on the maximum load ramping capability required by the unit's automatic power generation control command, with the dimension in kilowatts per second. The first-order backward finite difference algorithm extracts the instantaneous power change rate by reading the current power, subtracting the power of the previous cycle, and dividing by the time difference. The preset phase lag constant corresponding to the bilinear transform method digital low-pass Butterworth filter algorithm is a group delay correction equivalent constant derived by directly analyzing the phase frequency characteristics of the filter transfer function. To ensure dimensional homogeneity, the unit of this phase lag constant is strictly configured as seconds squared per kilowatt, thus ensuring that the output safe time deviation compensation amount, after being multiplied by the load change rate characteristic value in kilowatts per second, strictly has the time dimension of seconds. This subtraction operation is completed instantly within the current calculation cycle, and the compensation amount for each cycle is refreshed independently, without generating infinite cumulative drift across cycles, ensuring the physical safety boundary of time correction.

[0063] In the progressive correlation, this step is the decisive compensation mechanism for eliminating the side effects of the denoising scheme. It introduces the load variation slope through the feedforward channel to accurately make up for the deficit in the time dimension, so as to have zero-latency response sensitivity even in extremely fast material discharge scenarios.

[0064] Through the above technical solution, this embodiment solves the specific technical problem that low-pass filtering noise reduction methods inevitably introduce time domain response delay under rapid load change conditions, leading to blindly optimistic prediction of available time by introducing a feedforward compensation mechanism of power change rate extraction and phase lag constant product. It achieves the unique technical effect of completely eliminating filter phase lag and ensuring absolute advance warning of material shortage under extremely fast material discharge crisis conditions.

[0065] Furthermore, for the characteristic mapping scenario where electrical physical quantities are transformed into microscopic viscous resistance, directly establishing a high-order nonlinear dynamic model would consume enormous underlying computational resources and is prone to divergence. This embodiment defines the specific processing procedure for the output viscosity correction coefficient as follows: the preset resistance viscosity mapping function is specifically a preset first-order polynomial equation for resistance viscosity, a calibration function with transient current gradient as the independent variable and viscosity correction coefficient as the dependent variable; the transient current gradient of the current time series is substituted as an independent variable into the first-order polynomial equation for resistance viscosity, and after performing algebraic multiplication with a preset viscosity sensitivity proportionality coefficient, a preset reference zero-point bias constant is superimposed to calculate the corresponding dependent variable value, and this dependent variable value is used as the viscosity correction coefficient.

[0066] The preset first-order polynomial equation for drag viscosity is a set of standard linear algebraic mapping baselines. The preset viscosity sensitivity proportionality coefficient represents the slope gain of this polynomial, and its value is obtained by performing least-squares regression calibration on massive historical motor output torque and current evolution data. Since the dimension of transient current gradient is amperes per second (A / s), to ensure that the dimension of the final dependent variable (viscosity correction coefficient) is kilowatts (kW), the physical unit of this viscosity sensitivity proportionality coefficient is strictly assigned to kilowatt-seconds per ampere (kW·s / A). The preset reference zero-point offset constant serves as the polynomial intercept, and its dimension is also strictly configured to kilowatts (kW), representing the inherent equivalent frictional power reference under standard dry coal no-load conditions.

[0067] The preset reference zero-point bias constant, serving as the polynomial intercept, is dimensionless and characterizes the inherent friction reference under standard dry coal unloaded conditions. Within the complete technical chain, this embodiment decomposes the abstract and complex physical resistance black box into the purest addition-multiplication algebraic operations, providing a mapping hub with extremely low computational overhead for cross-domain fusion channels. For its specific application dimension, this enables the edge gateway to linearly extract the resistance correction ratio immediately upon capturing the initial transient of current jumps.

[0068] Through the above technical solution, this embodiment solves the specific refinement technical problem of huge computational overhead and extremely poor real-time performance in dynamically quantifying shear force mutations by simply relying on fluid mechanism models by constructing a first-order polynomial equation of drag viscosity with a clear sensitivity proportional coefficient and zero-point bias constant. It achieves the unique technical effect of linearly converting the viscous drag of non-Newtonian fluids, which is difficult to measure directly, into a digital control rate factor with low delay and high reliability.

[0069] Furthermore, in scenarios where multidimensional data converges and extrapolates to the final countdown scale, complex curve fitting or high-order derivative prediction is easily affected by input noise in industrial settings, leading to oscillations and divergences in the prediction time. This embodiment defines the specific processing procedure for inputting the effective coal volume and the stock depletion time series gradient into a first-order linear extrapolation operator for division extrapolation: Based on the principle of first-order linear extrapolation, a first-order linear evolution equation is constructed with the future prediction time as the independent variable and the coal volume at the future time as the dependent variable; in the first-order linear evolution equation, the effective coal volume at the current time is set as the initial intercept constant, and the negative value of the stock depletion time series gradient is set as the first derivative slope parameter; the boundary condition of the dependent variable in the first-order linear evolution equation is set to zero to characterize the limit state of complete physical depletion of materials inside the target coal bunker; the equation is solved using the algebraic equation transposition rule, and the effective coal volume is divided by the absolute value of the stock depletion time series gradient, and the numerical variable of the future prediction time obtained from the solution is output as the safe available time.

[0070] The first-order linear extrapolation principle is a classic prediction model in the field of numerical approximation based on the first-order expansion of Taylor series. Its core assumption is that the rate of decay change within an extremely short monitoring window is approximately constant. In the dimensional homogeneity review, the initial intercept constant inherits the cubic meter dimension of the effective coal volume, and the first derivative slope parameter inherits the cubic meter per second dimension of the stock depletion time series gradient. By setting the future volume boundary condition to zero, and using algebraic rearrangement to divide cubic meters by cubic meters per second, the dimensions of the output numerical variable are strictly restored to seconds, representing time.

[0071] In the complete technical chain, this solution serves as the central normalization hub of the entire algorithm architecture, directly reducing the dimensionality of the composite large gradient vector, which has undergone spatial topology correction, electrical correction, and filtering, into a scalar time scale. For the specific scenario, the simplified mathematical dimensionality reduction and term relocation ensures the absolute response speed of the CPU interrupt cycle in industrial real-time control.

[0072] Through the above technical solution, this embodiment solves the specific technical problem of countdown prediction oscillation deadlock caused by computing power overload and noise amplification in complex high-order differential derivation in industrial control embedded systems by constructing a first-order linear evolution equation and hard-zeroing the boundary of the target dependent variable and performing algebraic transposition and division. It achieves the unique technical effect of converting the transient of the composite gradient vector into a precise countdown safety scale with extremely simple deterministic algebraic operation overhead.

[0073] This application embodiment solves the technical problem in the prior art that, under specific material storage and supply conditions caused by asymmetric material distortion due to complex wet and sticky coal quality and drastic fluctuations in the discharge load of the bottom motor, the prediction of the absolute material depletion time is severely delayed and distorted due to the disconnect between three-dimensional spatial volume monitoring and the electromagnetic physical characteristics of the bottom layer, which in turn leads to the shutdown of cascaded feeding equipment due to material shortage.

[0074] This embodiment aims to solve the serious problem of passive lag or erroneous material interruption in feeding equipment caused by material deformation, moisture adhesion, and drastic load fluctuations in complex industrial scenarios. This embodiment first approaches the problem from a physical spatial perspective, continuously acquiring the point cloud set of the target coal bunker's solid surface, and simultaneously collecting the stator current time series and active power time series of the coal mill drive motor on the bottom discharge side. To accurately reconstruct the asymmetric material volume, this embodiment performs Delaunay triangulation with preset constraint edges on the solid surface point cloud set, followed by Simpson double integral for volume integral accumulation calculation, thereby accurately extracting the true effective coal volume inside. Based on this, this embodiment extrapolates the static spatial reconstruction to the dynamic time domain, performing first-order finite difference calculation in the time domain on the effective coal volume, and outputting the time reduction rate of the effective coal volume, representing the macroscopic discharge velocity.

[0075] To correct theoretical biases arising from purely spatial calculations, this embodiment further explores the electromagnetic physical feedback characteristics of the underlying drive equipment. A transient current gradient is output by performing a first-order finite-difference calculation on the stator current time series of the coal mill drive motor, thus characterizing the real-time evolution of the material's adhesion resistance during the discharge and compression process. This embodiment substitutes the transient current gradient into a preset resistance viscosity mapping function to calculate a viscosity correction coefficient, which is then used to correct the baseline consumption ratio generated from the ratio of the time loss rate to the active power time series of the coal mill drive motor. This cross-dimensional feature fusion ultimately outputs a stock depletion time series gradient that closely approximates the physical reality. Subsequently, based on the current effective coal volume, this embodiment performs a first-order linear division extrapolation calculation using the stock depletion time series gradient to directly deduce and anchor the absolute depletion time characterizing the complete exhaustion of the material, i.e., the safe available time. When the safe available time reaches the preset constraint time limit, a start-up replenishment command is actively issued to drive the equipment to start via frequency conversion.

[0076] Furthermore, this embodiment addresses complex operating conditions induced by the evolution of underlying algorithms. It constructs an adaptive anti-disturbance closed loop with layered constraints to tackle extreme environments arising from asymmetric local collapse in wet, sticky coal, high-frequency numerical jitter during grid reconstruction, phase prediction lag caused by low-pass filtering, and extreme high-frequency peak-shaving oscillations in the power grid. Faults are captured by spatial height gradient vectors, and preset constraint edges are dynamically rewritten using topology. Low-pass Butterworth filtering truncation is triggered by sliding window statistical variance judgment. Then, smoothing lag feedforward compensation is performed using load change rate eigenvalues. Finally, the time observation step size is adaptively scaled based on transient power variance to ensure absolute closure and accurate parsing of the entire data flow under harsh conditions.

[0077] This embodiment completely breaks through the isolated barrier of traditional workshop material monitoring that relies solely on a single liquid level probe or static power threshold. It innovatively couples the volume loss field of the three-dimensional top morphology with the electromagnetic adhesion resistance field of the bottom discharge equipment in all time and space, constructing an absolute exhaustion time series prediction mechanism that can adapt to severe sudden changes in working conditions such as cliff collapse and high-frequency load changes. It accurately solves the deep-seated technical pain point of workshop prediction lag and cascade equipment shutdown due to the coupling interference of complex material characteristics and dynamic discharge.

[0078] Example 2 according to Figure 3 As shown, this embodiment also provides a workshop equipment intelligent operation control system based on load status, including: Data acquisition module 1 acquires the point cloud set of the physical surface of the target coal bunker, the stator current time sequence of the coal mill drive motor, and the active power time sequence of the coal mill drive motor; The coal bunker determination module 2 calculates the average transient elevation difference based on the point cloud set of the entity surface, and determines whether the target coal bunker is in a coal bunker state based on the relationship between the average transient elevation difference and the first preset threshold. Volume calculation module 3 performs Delaunay triangulation processing on the point cloud set of the entity surface with preset constraint edges to obtain the point cloud mesh of the entity connected domain, and then calculates the effective coal volume of the target coal bunker by accumulating the volume integrals of the Simpson double integral. Feature calculation module 4 performs time-domain first-order finite difference calculation on the effective coal volume to obtain the time loss rate of the effective coal volume; performs first-order finite difference calculation on the stator current time series of the coal mill drive motor to obtain the transient current gradient; and determines the basic consumption ratio based on the ratio of the time loss rate to the active power time series of the coal mill drive motor. The gradient correction module 5 inputs the transient current gradient into a preset resistance viscosity mapping function to obtain a viscosity correction coefficient. The viscosity correction coefficient is then used to perform a product correction on the basic consumption ratio to obtain the stock depletion time-series gradient. Time calculation module 6 inputs the effective coal volume and the time gradient of stock depletion into a first-order linear extrapolation operator for division extrapolation calculation to obtain the safe available time characterizing the absolute depletion of coal volume; If the safe available time is less than or equal to the preset constraint time, the equipment control module 7 outputs a start feeding command to drive the cascade frequency converter of the coal conveyor belt of the target coal bunker to start, thereby completing the intelligent operation control of the workshop equipment.

[0079] In summary, this invention provides a method and system for intelligent operation control of workshop equipment based on load status. By acquiring the point cloud set of the physical surface of the target coal bunker, and using the Delaunay triangulation of preset constraint edges and Simpson double integral for volume integral accumulation calculation, it overcomes the shortcomings of traditional single-point ranging that cannot characterize complex and distorted internal morphology, achieves high-precision reconstruction of the actual effective coal volume, and solves the problem of distortion in the calculation of storage volume under the asymmetric distribution of irregular wet and sticky materials.

[0080] By performing first-order finite difference calculations on the effective coal volume to output the time loss rate, and by performing first-order finite difference calculations on the stator current time series of the coal mill drive motor to output the transient current gradient, and then using a preset resistance viscosity mapping function to generate a viscosity correction coefficient to multiply and correct the basic consumption ratio, the macroscopic discharge velocity in the spatial dimension and the microscopic adhesion resistance in the electromagnetic physical dimension were successfully integrated across domains mathematically. This resulted in the calculation of an extremely accurate stock depletion time series gradient, solving the technical problem of existing technologies relying solely on load monitoring, which leads to deviations in material emptying predictions from reality.

[0081] The corrected and calibrated effective coal volume and the time gradient of stock depletion are input into a first-order linear extrapolation operator for division extrapolation calculation. The safe available time representing the absolute depletion of coal volume is directly output, and a feeding start command is output when the condition is met. This realizes a fundamental leap from passive post-event alarm to active feedforward frequency conversion control of workshop equipment, and greatly improves the continuity of material supply and system robustness under extreme high-frequency variable load scenarios.

[0082] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for intelligent operation control of workshop equipment based on load status, characterized in that, include: Obtain the point cloud set of the solid surface of the target coal bunker, the stator current time sequence of the coal mill drive motor, and the active power time sequence of the coal mill drive motor; The average transient elevation difference is calculated based on the point cloud set of the entity surface. Based on the relationship between the average transient elevation difference and the first preset threshold, it is determined whether the target coal bunker is in a coal-covered state. The point cloud set of the entity surface is subjected to Delaunay triangulation with preset constraint edges to obtain the entity connected domain point cloud mesh, and then the effective coal volume of the target coal bunker is obtained by Simpson double integral volume integral accumulation calculation. The effective coal volume is calculated using a first-order finite difference time-domain method to obtain the time reduction rate of the effective coal volume; the transient current gradient is obtained by calculating the stator current time sequence of the coal mill drive motor using a first-order finite difference method. The basic consumption ratio is determined based on the ratio of the time loss rate to the time sequence of the active power of the coal mill drive motor. The transient current gradient is input into a preset resistance viscosity mapping function to obtain a viscosity correction coefficient. The viscosity correction coefficient is then used to perform a product correction on the base consumption ratio to obtain the stock depletion time-series gradient. The effective coal volume and the time gradient of stock depletion are input into a first-order linear extrapolation operator for division extrapolation to obtain the safe available time characterizing the absolute depletion of coal volume. If the safe available time is less than or equal to the preset constraint time, a start feeding command is output to drive the cascade frequency converter of the coal conveyor belt in the target coal bunker to start, thus completing the intelligent operation control of the workshop equipment.

2. The intelligent operation control method for workshop equipment based on load status according to claim 1, characterized in that, The specific processing steps for performing Delaunay triangulation on the point cloud set of the entity surface with preset constraints include: Obtain the preset fixed elevation constant at the physical boundary between the upper cylindrical and lower conical parts of the target coal bunker, which characterizes the coal pile. Input the point cloud set of the entity surface into a preset three-dimensional coordinate filter, traverse and extract all boundary sampling points whose spatial Z-axis coordinates are equal to a preset fixed elevation constant, and output the boundary sampling point set. The set of boundary sampling points is input into the least squares circle fitting operator to fit the radius and center spatial coordinates, output the structural fracture edge of the spatial closed loop and lock it, and the structural fracture edge is recorded as the preset constraint edge.

3. The intelligent operation control method for workshop equipment based on load status according to claim 1, characterized in that, The specific process of obtaining the transient current gradient by performing first-order finite difference calculation on the stator current time series of the coal mill drive motor includes: The stator current timing sequence of the coal mill drive motor is input into a first-order discrete difference time-domain converter. The first-order numerical difference processing is performed on two adjacent current timing sampling points according to a preset fixed sampling period, and the transient current gradient is output.

4. The intelligent operation control method for workshop equipment based on load status according to claim 1, characterized in that, The specific process of obtaining the viscosity correction coefficient by inputting the transient current gradient into a preset resistance viscosity mapping function includes: The preset drag viscosity mapping function is a first-order polynomial calibration equation for drag viscosity with transient current gradient as independent variable and viscosity correction coefficient as dependent variable. The transient current gradient of the current time series is substituted as an independent variable into the first-order polynomial equation. After performing algebraic multiplication with the preset viscosity sensitivity proportionality coefficient, and then superimposed with the preset reference zero-point bias constant, the calculated dependent variable value is the viscosity correction coefficient.

5. The intelligent operation control method for workshop equipment based on load status according to claim 1, characterized in that, The specific processing steps for obtaining the safe available time characterizing the absolute depletion of coal volume by inputting the effective coal volume and the time gradient of stock depletion into a first-order linear extrapolation operator for division extrapolation calculation include: Based on the principle of first-order linear extrapolation, a first-order linear evolution equation is constructed with the future prediction time as the independent variable and the coal volume at the future time as the dependent variable. Set the current effective coal volume as the initial intercept constant of the equation, and set the negative value of the time gradient of stock depletion as the slope parameter of the first derivative of the equation. Setting the boundary condition of the dependent variable of the equation to zero characterizes the ultimate state of complete depletion of the target coal bunker material. By solving the equation through algebraic rearrangement, the effective coal volume is divided by the absolute value of the stock depletion time gradient, and the resulting value is the safe available time.

6. The intelligent operation control method for workshop equipment based on load status according to claim 1, characterized in that, Before outputting the start material replenishment command, a status check is also included, the specific process of which is as follows: Extract the power consumption token state vector and the operating status feature bits of the coal conveyor belt in the distributed workshop network of the target coal bunker; When the power token state vector is determined to be in a locked waiting state and the running state feature bit is in a standby pause state, the output start feeding command is executed to drive the cascade frequency converter start of the coal feeding belt of the target coal bunker.

7. The intelligent operation control method for workshop equipment based on load status according to claim 1, characterized in that, After outputting the start material replenishment command, the process also includes dynamic updating of constraint edges, as detailed below: The point cloud set of the entity surface is used to calculate the partial derivative of the tangent plane using a local surface least squares plane fitting algorithm based on k-nearest neighbors, and the spatial height gradient vector corresponding to each sampling point is output. When the magnitude of the spatial height gradient vector is greater than the preset cliff shape determination threshold, the current working condition is determined to be an asymmetric local collapse working condition, and the following steps are performed: Sampling points larger than the cliff morphology determination threshold are selected to form a set of feature edge points of the slip surface; The set of feature edge points of the slip surface is reconstructed by concatenation using a nearest neighbor edge tracking algorithm based on Euclidean distance, and the dynamic constraint edge of the slip fault in the form of a polyline segment is output. The point cloud mesh of the entity connected domain in the current time series is traversed by a line segment intersection detection algorithm based on vector outer product, and the set of intersecting topological edges that geometrically intersect with each polyline segment of the dynamic constraint edge of the slip fault is retrieved and extracted. The set of intersecting topological edges is removed from the point cloud mesh of the entity connected domain by the mesh topology adjacency deletion algorithm, exposing the polygon hole domain; The dynamic constraint edge of the slip fault is used as a rigid, impenetrable boundary and fixedly embedded in the polygonal cavity domain, dividing the polygonal cavity domain into a left closed polygonal domain and a right closed polygonal domain. For the closed polygonal domains on the left and right sides respectively, the local topological mesh reconstruction is carried out by the mesh retriangulation algorithm based on the constrained empty circle criterion, thus completing the in-situ topological rewriting of the point cloud mesh of the entity connected domain. By using a one-dimensional dynamic array appending cascade algorithm, the polyline data matrix of the dynamic constraint edge of the slip fault is merged and written into the tail of the original structural fracture edge storage matrix, and the updated preset constraint edge is output for the Deloitte triangulation processing in the next time series.

8. The intelligent operation control method for workshop equipment based on load status according to claim 7, characterized in that, After outputting the updated preset constraint edges, the process also includes volume data smoothing, as detailed below: The effective coal volume output continuously in time is input into a preset sliding short window. The time-domain variance of the effective coal volume is calculated using a sliding window statistical variance algorithm. If the time-domain variance is greater than the preset grid jitter tolerance, the current operating condition is determined to be an unbalanced high-frequency numerical jitter condition, and the following steps are executed: The digital low-pass Butterworth filtering algorithm using the bilinear transform method is used to filter and truncate all effective coal volumes within the sliding short window, eliminating high-frequency discrete numerical components caused by local switching of the grid topology, and outputting a smooth effective average coal volume. The effective coal volume of the current time series is replaced by the smoothed effective average coal volume.

9. The intelligent operation control method for workshop equipment based on load status according to claim 8, characterized in that, After replacing the effective coal volume in the current time series, the process also includes safety time compensation control, the specific process of which is as follows: The rate of change of the active power time series of the coal mill drive motor is extracted by the first-order backward finite difference algorithm, and the load change rate feature value is output. If the characteristic value of the load change rate is greater than the preset lower threshold for rapid material discharge, the current operating condition is determined to be a high-lag, undercompensated operating condition, and the following steps are executed: Multiply the characteristic value of the load variation rate by the preset phase lag constant corresponding to the digital low-pass Butterworth filter algorithm of the bilinear transform method to obtain the safe time deviation compensation amount; The corrected safe available time is obtained by subtracting the safe time deviation compensation from the safe available time output by the first-order linear extrapolation operator; If the corrected safe available time is less than or equal to the preset constraint time, a start-up feeding command carrying the target coal bunker identification code and the preset frequency converter start control word is generated through the preset industrial control network instruction code, and sent to the coal conveyor frequency converter controller to drive the coal conveyor cascade frequency converter start. If the revised safe available time is greater than the preset constraint time, the coal conveyor belt will remain in standby and suspended state.

10. A workshop equipment intelligent operation control system based on load status, characterized in that, include: The data acquisition module acquires the point cloud set of the physical surface of the target coal bunker, the stator current time sequence of the coal mill drive motor, and the active power time sequence of the coal mill drive motor. The coal bunker determination module calculates the average transient elevation difference based on the point cloud set of the entity surface, and determines whether the target coal bunker is in a coal bunker state based on the relationship between the average transient elevation difference and a first preset threshold. The volume calculation module performs Delaunay triangulation processing on the point cloud set of the entity surface with preset constraint edges to obtain the point cloud mesh of the entity connected domain, and then calculates the effective coal volume of the target coal bunker by accumulating the volume integrals of the Simpson double integral. The feature calculation module performs a first-order finite difference calculation in the time domain on the effective coal volume to obtain the time loss rate of the effective coal volume; performs a first-order finite difference calculation on the stator current time series of the coal mill drive motor to obtain the transient current gradient; and determines the basic consumption ratio based on the ratio of the time loss rate to the active power time series of the coal mill drive motor. The gradient correction module inputs the transient current gradient into a preset resistance viscosity mapping function to obtain a viscosity correction coefficient. The viscosity correction coefficient is then used to perform a product correction on the base consumption ratio to obtain the stock depletion time-series gradient. The time calculation module inputs the effective coal volume and the time gradient of stock depletion into a first-order linear extrapolation operator for division extrapolation calculation to obtain the safe available time characterizing the absolute depletion of coal volume. If the safe available time is less than or equal to the preset constraint time, the equipment control module outputs a start feeding command to drive the cascade frequency converter of the coal conveyor belt in the target coal bunker to start, thus completing the intelligent operation control of the workshop equipment.

Images