Coal blending combustion control system and method

Abstract

According to the fire coal blending combustion control system and method provided by the invention, the fire coal blending combustion control system realizes intelligent and accurate control of the fire coal blending combustion process by constructing a cooperative control system of data acquisition, modeling analysis, decision control and man-machine interaction. The technical limitation of traditional manual coal blending is broken through, real-time accurate modeling of the coal yard form is achieved through the three-dimensional laser scanning and multi-source data fusion technology, and the coal diffusion prediction precision is remarkably improved by combining dynamic repose angle parameter correction; genetic algorithm optimization and a closed-loop compensation mechanism are adopted for synergistic effect, coal blending calorific value fluctuation is effectively controlled, and the risk that sulfur content exceeds the standard is greatly reduced; an intelligent operation and maintenance interface is constructed through a three-dimensional visualization and data tracing technology, and the abnormal working condition handling efficiency is remarkably improved. And obvious economic benefits and environmental benefits are realized while the stability of the calorific value of the fire coal is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of coal blending control, and in particular to a coal blending control system and method. Background Art

[0002] Traditional coal-fired power plant co-firing control relies primarily on manual experience and offline laboratory data, and suffers from inherent defects such as delayed coal quality analysis, low coal blending accuracy, and poor dynamic response. Specifically, (1) 3D coal field morphology monitoring relies on manual coal counting, making it impossible to obtain real-time data on coal pile volume and spatial distribution; (2) Silos use fixed empirical values ​​for coal flow characteristic parameters (such as the angle of repose), resulting in high errors in coal quality diffusion simulation, for example, exceeding 15%. This results in high fluctuations in the thermal efficiency of co-firing in coal-fired power plants and insufficient co-firing efficiency. Summary of the Invention

[0003] In view of this, the present invention provides a coal blending control system. One or more embodiments of this specification also relate to a coal blending control method to address the technical deficiencies in the prior art.

[0004] According to a first aspect of the present invention, a coal blending control system is provided, comprising: a data acquisition module configured to collect three-dimensional point cloud data of the coal yard, equipment positioning data, and coal quality parameters, and generate and output a coal flow feature vector based on the collected data; A modeling and analysis module is configured to generate and output simulation prediction data including coal quality diffusion distribution based on a pre-built three-dimensional coal yard model, pre-calculated silo coal pile repose angle parameters, and received coal flow characteristic vectors; The decision control module is configured to iterate based on the simulation prediction data and a preset genetic algorithm to generate optimal coal blending parameters, and generate control instructions including equipment action sequences and compensation coal loading tasks based on the optimal coal blending parameters; The human-computer interaction module is configured to generate instruction execution results, coal quality deviation warnings and inventory turnover information based on the control instructions received in real time and display them on the target display screen.

[0005] In some embodiments, the data acquisition module includes: A laser scanning unit is used to scan the surface of the coal yard at a preset period to generate and output three-dimensional point cloud data with a time stamp; Positioning monitoring unit, used to collect bucket wheel excavator boom angle and spatial coordinate data in real time, generate equipment positioning data and output; Coal quality detection unit, used for online analysis of ash, sulfur and calorific value parameters of the belt coal flow, and obtain coal quality parameters with time stamp; The data fusion unit is used to map the coal quality parameters to the three-dimensional space corresponding to the bucket wheel excavator, establish the corresponding relationship between the coal quality parameters and the coal quality in the three-dimensional space based on the timestamp; align the three-dimensional point cloud data with the equipment positioning data in time and space; and generate a standardized coal flow feature vector based on the coal quality corresponding relationship and the three-dimensional point cloud data and equipment positioning data that have been aligned in time and space.

[0006] In some embodiments, the modeling and analysis module includes: A space division unit is used to establish a three-dimensional voxel grid model with a mapping relationship of coal quality parameters by receiving coal yard space data transmitted by an external acquisition device; The silo modeling unit is used to analyze and generate the angle of repose parameter that characterizes the material properties based on the echo signal of the radar level meter array targeting the coal pile, and dynamically associate this parameter with the physical properties of the voxel grid established by the spatial division unit; The simulation prediction unit is used to simulate the material diffusion behavior in the coal blending process under preset boundary conditions by integrating the voxel model output by the space division unit and the material characteristic parameters provided by the silo modeling unit, and output simulation prediction data including the coal quality diffusion distribution.

[0007] In some embodiments, the formula for calculating the coal quality diffusion distribution is:

[0008] Among them, D t is the coal diffusion value at time t, Q i is the coal amount of the ith grid in the three-dimensional voxelized grid model; is the three-dimensional value of the spatial position corresponding to the center coordinate of the i-th grid, is the coal flow velocity correction factor of the jth silo; represents the temperature difference in the repose angle parameter between the surface and core of the coal pile in the jth silo, is the coal quality deviation sensitivity coefficient of the kth belt, obtained based on historical data statistics, C k represents the actual coal quality parameters of the kth belt, C k0 Represents the target coal quality parameters of the kth belt.

[0009] In some embodiments, the decision control module includes: The coal blending parameter determination unit is used to perform multi-generation iterative calculations using a preset genetic algorithm based on the coal quality diffusion distribution data output by the simulation prediction unit. Each iteration includes selecting the individual parameter combination with the highest fitness, performing crossover recombination operations, and implementing mutation operations, ultimately converging to generate the optimal coal blending parameters. The task decomposition unit is used to parse the coal loading instruction containing the optimal coal blending parameters into a sequence of equipment actions with a time sequence relationship; The conflict detection unit is used to monitor the operating status of the equipment in real time and generate a corresponding conflict mark when a conflict in resource occupation by multiple devices is detected; A scheduling optimization unit is used to receive conflict marker information and dynamically adjust the operating timing of the belt transmission system; The compensation control unit is used to continuously compare the actual coal quality parameters with the preset thresholds. When the deviation exceeds the allowable range, it automatically generates a compensation coal loading task instruction and feeds the instruction back to the task decomposition unit to restart the control process.

[0010] In some embodiments, the compensation control unit includes a genetic algorithm optimizer, a real-time compensator, and an effect verifier, wherein: The genetic algorithm optimizer receives coal quality parameter constraints and simulates the biological evolution process to perform population initialization, fitness evaluation, selection, recombination, and mutation operations in a multi-dimensional parameter space to generate an optimal coal blending solution that meets the multi-objective balance of calorific value, sulfur content, and cost. The real-time compensator continuously monitors the deviation between the actual coal quality parameters and the target values. When the deviation exceeds the preset tolerance, the dynamic adjustment mechanism is triggered, and the correction instructions are fed back to the genetic algorithm optimizer to restart the iterative calculation; The effect verifier collects feedback data after coal blending is executed, verifies the control accuracy by comparing the actual coal quality parameters with the expected targets, and synchronously transmits the verification results to the genetic algorithm optimizer for optimizing parameter weights; The genetic algorithm optimizer, real-time compensator and effect verifier form a closed-loop control circuit, which ensures that the coal blending plan meets both static constraints and dynamic compensation requirements through continuous iterative optimization.

[0011] In some embodiments, the human-computer interaction module includes: A three-dimensional visualization unit receives the coal quality parameters of the voxelized grid model and generates a visual rendering of the calorific value distribution. The coal quality parameters include calorific value and sulfur content. The data traceability unit is interconnected with the 3D visualization unit data to form a full-chain coal blending operation record by associating the bucket wheel excavator's operating trajectory, coal flow characteristics, and coal blending plan, and establish a traceable coal quality data lineage relationship; The intelligent early warning unit dynamically generates purchase recommendations and triggers inventory threshold alarms based on the inventory turnover analysis results output by the data tracing unit; Among them, the visual rendering of the three-dimensional visualization unit is fed back to the intelligent early warning unit to correct the priority ranking of procurement recommendations; the full-chain coal blending operation records of the data traceability unit are synchronously updated to the three-dimensional visualization unit to realize multi-dimensional dynamic labeling and display of the coal blending process.

[0012] According to a second aspect of the present invention, a method for controlling coal blending is provided, the method being applied to the system of the preceding claim, the method comprising: Collect 3D point cloud data, equipment positioning data, and coal quality parameters of the coal yard, and generate and output coal flow feature vectors based on the collected data; Generate and output simulation prediction data including coal quality diffusion distribution based on the pre-built 3D coal yard model, pre-calculated silo coal pile repose angle parameters, and received coal flow characteristic vectors; Based on the simulation prediction data and the preset genetic algorithm, it is iterated to generate the optimal coal blending parameters, and based on the optimal coal blending parameters, it generates control instructions including the equipment action sequence and the compensation coal loading task; Based on the control instructions received in real time, the instruction execution results, coal quality deviation warnings and inventory turnover information are generated and displayed on the target display screen.

[0013] In some embodiments, three-dimensional point cloud data, equipment positioning data, and coal quality parameters of the coal yard are collected, and a coal flow feature vector is generated and output based on the collected data, including: Obtain three-dimensional point cloud data with time stamps generated by scanning the coal yard surface at a preset period; Obtain real-time bucket wheel machine boom angle and spatial coordinate data, generate and output equipment positioning data; Obtain coal quality parameters obtained by online analysis of belt coal flow, including ash content, sulfur content and calorific value parameters; Map the coal quality parameters to the three-dimensional space corresponding to the bucket wheel excavator, and establish the corresponding relationship between the coal quality parameters and the coal quality in the three-dimensional space based on the timestamp; Temporally and spatially align 3D point cloud data with device positioning data; Based on the coal quality correspondence, as well as the time-space aligned three-dimensional point cloud data and equipment positioning data, a standardized coal flow feature vector is generated.

[0014] In some embodiments, based on a pre-built three-dimensional coal yard model, pre-calculated silo coal pile angle of repose parameters, and received coal flow characteristic vectors, simulation prediction data including coal quality diffusion distribution is generated and output, including: Receive coal yard spatial data transmitted by external acquisition devices and establish a three-dimensional voxel grid model with coal quality parameter mapping relationship; Based on the echo signal of the radar level meter array targeting the coal pile, the repose angle parameter that characterizes the material characteristics is analyzed and generated, and this parameter is dynamically associated with the physical properties of the voxel grid established by the spatial division unit; By integrating the voxelized model output by the space division unit and the material characteristic parameters provided by the silo modeling unit, the material diffusion behavior during the coal blending process is simulated under preset boundary conditions, and simulation prediction data including coal quality diffusion distribution is output.

[0015] In some embodiments, the formula for calculating the coal quality diffusion distribution is:

[0016] Among them, D t is the coal diffusion value at time t, Q i is the coal amount of the ith grid in the three-dimensional voxelized grid model; is the three-dimensional value of the spatial position corresponding to the center coordinate of the i-th grid, is the coal flow velocity correction factor of the jth silo; represents the temperature difference in the repose angle parameter between the surface and core of the coal pile in the jth silo, is the coal quality deviation sensitivity coefficient of the kth belt, obtained based on historical data statistics, C k represents the actual coal quality parameters of the kth belt, C k0 Represents the target coal quality parameters of the kth belt.

[0017] In some embodiments, based on the simulation prediction data and the preset genetic algorithm, an optimal coal blending parameter is generated through iteration, and a control instruction including an equipment action sequence and a compensation coal loading task is generated based on the optimal coal blending parameter, including: Based on the coal quality diffusion distribution data output by the simulation prediction unit, a preset genetic algorithm is used to perform multiple generations of iterative calculations. Each iteration includes selecting the individual parameter combination with the highest fitness, performing crossover recombination operations, and implementing mutation operations, ultimately converging to generate the optimal coal blending parameters. Parse the coal loading instruction containing the optimal coal blending parameters into a sequence of equipment actions with a time sequence relationship; Monitor the device's operating status in real time and generate corresponding conflict markers when conflicts in resource usage among multiple devices are detected; Receive conflict mark information and dynamically adjust the operating timing of the belt transmission system; Continuously compare the actual coal quality parameters with the preset thresholds. When the deviation exceeds the allowable range, a compensation coal loading task instruction is automatically generated and fed back to the task decomposition unit to restart the control process.

[0018] In some embodiments, the actual coal quality parameters are continuously compared with preset thresholds. When the deviation exceeds the allowable range, a compensation coal loading task instruction is automatically generated and fed back to the task decomposition unit to restart the control process, including: Accepting coal quality parameter constraints, the system simulates the biological evolution process to perform population initialization, fitness evaluation, selection, recombination, and mutation operations in a multi-dimensional parameter space to generate an optimal coal blending solution that meets the multi-objective balance of calorific value, sulfur content, and cost. Continuously monitor the deviation between actual coal quality parameters and target values. When the deviation exceeds the preset tolerance, a dynamic adjustment mechanism is triggered, and correction instructions are fed back to the genetic algorithm optimizer to restart iterative calculations. Collect feedback data after coal blending is executed, verify the control accuracy by comparing the actual coal quality parameters with the expected targets, and transmit the verification results synchronously to the genetic algorithm optimizer for optimizing parameter weights; Continuous iterative optimization ensures that the coal blending plan meets both static constraints and dynamic compensation requirements.

[0019] In some embodiments, based on the control instructions received in real time, the command execution results, coal quality deviation warnings, and inventory turnover information are generated and displayed on the target display screen, including: Receive coal quality parameters and generate a visual rendering of calorific value distribution. Coal quality parameters include calorific value and sulfur content index; By associating the bucket wheel excavator's operating trajectory, coal flow characteristics, and coal blending plan, a full-chain coal blending operation record is formed, and a traceable coal quality data lineage relationship is established; Based on the inventory turnover analysis results output by the data traceability unit, purchase recommendations are dynamically generated and inventory threshold alarms are triggered; Among them, the visual rendering is used to correct the priority ranking of procurement recommendations; the full-chain coal blending operation record is used to realize the multi-dimensional dynamic labeling display of the coal blending process.

[0020] At least one embodiment of the present invention achieves intelligent and precise control of the coal blending process by constructing a collaborative control system for data collection, modeling analysis, decision-making control, and human-computer interaction. This breaks through the technical limitations of traditional manual coal blending, achieves real-time and precise modeling of coal yard morphology through three-dimensional laser scanning and multi-source data fusion technology, and significantly improves the accuracy of coal quality diffusion prediction by combining dynamic repose angle parameter correction; uses genetic algorithm optimization and closed-loop compensation mechanism to effectively control the fluctuation of coal blending calorific value and significantly reduce the risk of excessive sulfur content; and constructs an intelligent operation and maintenance interface through three-dimensional visualization and data traceability technology, significantly improving the efficiency of handling abnormal working conditions. While ensuring the stability of the coal calorific value, it achieves significant economic and environmental benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a simplified structural diagram of a coal blending control system provided by the present invention; Figure 2 The present invention provides a flow chart of a coal blending control method. DETAILED DESCRIPTION

[0022] The following description sets forth many specific details to facilitate a thorough understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the scope of this specification. Therefore, this specification is not limited to the specific implementations disclosed below.

[0023] The terms used in one or more embodiments of this specification are for the purpose of describing specific embodiments only and are not intended to limit one or more embodiments of this specification. The singular forms of "a" and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items. The modifications of "one" and "a plurality" mentioned in this disclosure are illustrative and not restrictive, and those skilled in the art should understand that unless the context clearly indicates otherwise, it should be understood as "one or more".

[0024] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of this specification, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0025] See also Figure 1 , Figure 1 The following is a simplified structural diagram of a coal blending control system provided according to some embodiments of this specification, specifically including: The data acquisition module is configured to collect three-dimensional point cloud data, equipment positioning data and coal quality parameters of the coal yard, and generate and output a coal flow feature vector based on the collected data.

[0026] The modeling and analysis module is configured to generate and output simulation prediction data including coal quality diffusion distribution based on a pre-built three-dimensional coal yard model, pre-calculated silo coal pile repose angle parameters, and received coal flow characteristic vectors.

[0027] The decision control module is configured to iterate based on simulation prediction data and a preset genetic algorithm to generate optimal coal blending parameters, and generate control instructions including equipment action sequences and compensation coal loading tasks based on the optimal coal blending parameters.

[0028] The human-computer interaction module is configured to generate instruction execution results, coal quality deviation warnings and inventory turnover information based on the control instructions received in real time and display them on the target display screen.

[0029] The data acquisition module may refer to a hardware system that integrates multi-source sensing equipment, and is a core functional unit that acquires real-time information on the physical state of the coal yard through laser scanning, positioning reception, and coal quality testing devices. Three-dimensional point cloud data may refer to a set of spatial coordinates on the surface of the coal yard acquired through lidar scanning technology, containing geometric feature information of millions of discrete points, capable of reconstructing the three-dimensional shape and volume distribution of the coal pile. Equipment positioning data may refer to the real-time spatial coordinates and motion trajectory information of mechanical devices such as conveying equipment and stackers / reclaimers acquired based on GNSS or UWB technology. Coal quality parameters may refer to a set of key indicators reflecting the combustion characteristics of coal, including but not limited to chemical composition data such as calorific value, sulfur content, ash content, and volatile matter. Coal flow characteristic vectors may refer to a multidimensional data matrix that characterizes the dynamic characteristics of the coal transportation process, containing the fused calculation results of parameters such as coal quality distribution, flow rate changes, and equipment status.

[0030] The modeling and analysis module can refer to a computational unit that uses numerical simulation technology to perform predictive analysis of coal yard dynamics. It integrates geometric models, physical parameters, and real-time data streams to achieve visual simulation of coal quality diffusion processes. A three-dimensional coal yard model can refer to a digital spatial structure of a coal pile constructed using laser scanning or photogrammetry, including a computer-readable representation of its volume, shape, and position. The silo coal pile repose angle parameter can refer to the physical property value that reflects the maximum stable slope angle of coal when naturally deposited. This parameter directly affects the repose slope and stability calculations of the coal pile. Coal quality diffusion distribution can refer to the dynamic spatial variations in quality parameters (such as calorific value and sulfur content) caused by the physical movement of coal during transportation and stacking. Simulation prediction data can refer to the spatiotemporal evolution of coal flow trajectories and quality parameters derived through computational fluid dynamics or discrete element methods.

[0031] The decision-making control module may refer to the core processor that implements intelligent decision-making for the coal blending process, using an optimization algorithm to convert simulation results into executable control strategies. A genetic algorithm may refer to an intelligent optimization method that simulates the biological evolution process, searching for the optimal coal blending solution in the solution space through selection, crossover, and mutation operations. The optimal coal blending parameters may refer to the coal blending ratio determined through multi-objective optimization calculations, which must simultaneously meet constraints such as calorific value requirements, sulfur content limits, and cost control. The equipment action sequence may refer to a set of mechanical operation instructions such as the bucket wheel excavator reclaiming path and the belt conveyor start and stop sequence generated according to the coal blending plan. The compensatory coal feeding task may refer to a supplementary coal supply strategy designed to correct for coal quality deviations, achieving closed-loop quality control by dynamically adjusting the coal feeding amount.

[0032] The human-computer interaction module may refer to a visual terminal that connects the automation system and the operator, enabling two-way interaction between the issuance of control instructions and the monitoring of operating status. The result of instruction execution may refer to the actual action feedback data after the equipment responds to the control instruction, including operation log information such as completion status and execution duration. The coal quality deviation warning may refer to the abnormal alarm signal triggered when the difference between the detection value and the target value exceeds the threshold, which is used to indicate quality control risks. Inventory turnover information may refer to statistical indicators that reflect the inventory changes and turnover efficiency of each area of ​​the coal yard, including management data such as inventory volume and turnover rate. The target display screen may refer to an industrial-grade display screen deployed in the central control room, which supports multi-window split-screen display of key system operating parameters.

[0033] As a specific example, the coal blending control system may specifically include: The data acquisition module is configured as follows: The laser scanning unit uses a 32-line laser radar, scanning the coal pile surface once per minute, generating 3D point cloud data with a resolution of 5 cm. Each data point includes X, Y, and Z coordinates and reflection intensity values. The positioning monitoring unit integrates a GNSS positioner and an inclination sensor to collect the bucket wheel excavator's boom pitch angle (measurement range ±30°, accuracy 0.1°) and rotation angle (measurement range 0-360°, accuracy 0.5°) in real time, while also acquiring the equipment's UTM coordinates. The coal quality detection unit uses an online near-infrared analyzer, collecting ash content (range 5-40%), sulfur content (range 0.2-5%), and calorific value (range 3000-6500 kcal / kg) of the conveyor coal flow every 30 seconds. The data fusion unit aligns the sensor data via a time synchronization server, establishing a mapping between coal quality parameters and 3D coordinates, ultimately outputting a standardized coal flow feature vector containing 128 features.

[0034] The modeling and analysis module operates as follows: A 3D model of the coal yard is constructed, and the angle of repose parameters for the coal silo (typically 38° for bituminous coal) are imported. After receiving the feature vectors from the data fusion module, a discrete element simulation algorithm is used to predict the coal quality distribution over the next two hours. The module then outputs thermal map data (spatial resolution 0.5 m × 0.5 m) that includes the sulfur gradient.

[0035] The decision-making control module's execution logic uses a pre-set genetic algorithm population size of 200 and 50 iterations, with optimization objectives set at calorific value deviation (target ±200 kcal / kg) and sulfur content control (target ≤1.2%). The resulting optimal coal blending parameters include a feed rate of 85 tons / hour for Type 3 coal and 120 tons / hour for Type 5 coal. The corresponding equipment action sequence includes: rotating the bucket wheel excavator to an azimuth angle of 142° and lowering the pitch angle to -12°; and increasing the speed of the No. 3 conveyor belt to 2.8 m / s.

[0036] Implementation method of the human-computer interaction module: Control instructions are transmitted to the central control room via industrial Ethernet and displayed in three areas on a 55-inch LED screen: the left area shows the real-time operation status of the equipment (including the position coordinates of the bucket wheel excavator, belt speed, etc.); the middle area is the coal quality deviation warning (a red alarm is triggered when the ash content fluctuates by more than ±3%); the right area displays the inventory turnover dashboard (including data such as the inventory of each type of coal and the estimated number of days of availability).

[0037] The beneficial effects of one of the embodiments of this specification include at least: constructing a collaborative control system of data collection, modeling analysis, decision-making control and human-computer interaction, and realizing intelligent and precise control of the coal blending process. Breaking through the technical limitations of traditional manual coal blending, the real-time and precise modeling of the coal yard morphology is achieved through three-dimensional laser scanning and multi-source data fusion technology, and the dynamic repose angle parameter correction is combined to significantly improve the accuracy of coal quality diffusion prediction; using genetic algorithm optimization and closed-loop compensation mechanism to work together to effectively control the fluctuation of coal blending calorific value and greatly reduce the risk of excessive sulfur content; building an intelligent operation and maintenance interface through three-dimensional visualization and data traceability technology, and significantly improving the efficiency of handling abnormal working conditions. While ensuring the stability of the calorific value of coal, significant economic and environmental benefits are achieved.

[0038] In some embodiments, the data acquisition module includes: a laser scanning unit, which is used to scan the surface of the coal yard at a preset period to generate and output three-dimensional point cloud data with timestamps; a positioning monitoring unit, which is used to collect the bucket wheel cantilever angle and spatial coordinate data in real time, generate and output equipment positioning data; a coal quality detection unit, which is used to analyze the ash content, sulfur content and calorific value parameters of the belt coal flow online to obtain coal quality parameters with timestamps; a data fusion unit, which is used to map the coal quality parameters to the three-dimensional space corresponding to the bucket wheel excavator, and establish a corresponding relationship between the coal quality parameters and the coal quality in the three-dimensional space according to the timestamp; align the three-dimensional point cloud data with the equipment positioning data in time and space; and generate a standardized coal flow feature vector based on the coal quality correspondence, and the three-dimensional point cloud data and equipment positioning data that have been aligned in time and space.

[0039] A laser scanning unit may refer to a device that uses laser ranging technology to periodically scan the surface of a coal pile, obtaining a set of spatial coordinates of the geometric features of the coal pile surface by emitting a laser beam and receiving reflected signals. A preset period may refer to a scanning time interval parameter set according to the intensity of coal yard operations, ensuring that the frequency of data updates matches the dynamic changes in the material. A timestamp may refer to the time-series mark information that records the time of data collection, including time dimension data such as year, month, day, hour, minute, and second. Three-dimensional point cloud data may refer to a set of discrete spatial coordinates generated by a laser scanning unit, where each point contains a three-dimensional coordinate value and reflection intensity characteristics.

[0040] The positioning monitoring unit may refer to a measurement system that integrates an angle sensor and a spatial positioning device, used to capture the bucket wheel crane boom's pitch angle, rotation angle, and overall machine spatial coordinates in real time. The bucket wheel crane boom angle may refer to the pitch and rotation angle parameters that reflect the reclaimer's spatial posture and determine the contact position between the scraper and the coal pile. Spatial coordinate data may refer to the absolute position of the equipment, including longitude, latitude, and altitude, obtained using navigation and positioning technology. Equipment positioning data may refer to a composite dataset that integrates boom angle and spatial coordinates, representing the real-time spatial state of the bucket wheel crane during operation.

[0041] The coal quality testing unit refers to an online analyzer installed on the coal conveyor belt section, which uses spectroscopy or X-ray technology to detect coal quality parameters in real time. The belt coal flow refers to the continuously moving coal material on the conveyor. Its cross-sectional shape and speed determine the representativeness of the sample. Ash content refers to the mass percentage of incombustible matter remaining after complete combustion of the coal, which affects the thermal efficiency and emission indicators of the boiler. Sulfur content refers to the mass fraction of sulfur compounds in the coal and is directly related to the sulfur dioxide emission concentration. The calorific value parameter refers to the heat energy released per unit mass of coal after complete combustion and is a core control indicator for coal blending.

[0042] The data fusion unit can refer to a processor that performs spatiotemporal correlation and feature extraction of multi-source data, using coordinate system conversion and time series matching algorithms. The coal quality correspondence can refer to the binding relationship between coal quality parameters and three-dimensional spatial positions established through spatiotemporal mapping, realizing a three-dimensional representation of the internal characteristics of the coal pile. Spatiotemporal alignment can refer to the process of performing time interpolation and spatial registration on asynchronously collected data to eliminate timing and coordinate differences between sensors. Standardization can refer to the implementation of unified dimensional conversion and normalization processing on heterogeneous data to ensure the comparability of feature vectors and model compatibility. The coal flow feature vector can refer to a multidimensional data matrix that integrates coal quality distribution, equipment status, and spatial topology, which serves as the input feature of the intelligent coal blending system.

[0043] This module builds a digital control system for the coal blending process through the deep collaboration of multimodal sensor data. A closed-loop verification mechanism using laser scanning and positioning monitoring is used to achieve millimeter-level reconstruction of coal pile morphology and centimeter-level tracking of equipment trajectory, significantly improving the accuracy of spatial modeling; through three-dimensional spatial mapping and timestamp alignment of coal quality parameters, a three-dimensional database of the internal quality of the coal pile is established, breaking through the limitations of traditional surface sampling; based on the dynamic generation of standardized coal flow feature vectors, a full-factor input including material characteristics, equipment status and spatial topology is provided to the intelligent coal blending system, upgrading the blending control from empirical decision-making to a data-driven mode. While ensuring the stability of the calorific value of coal, this solution achieves precise pre-control of sulfur distribution, providing core technical support for clean combustion and environmental protection standards.

[0044] In some embodiments, the modeling and analysis module includes: a spatial division unit, which is used to establish a three-dimensional voxel grid model with a coal quality parameter mapping relationship by receiving coal yard spatial data transmitted by an external acquisition device; a silo modeling unit, which is used to analyze and generate an angle of repose parameter that characterizes the material properties based on the echo signal of the radar level meter array for the coal pile, and dynamically associate the parameter with the voxel grid physical properties established by the spatial division unit; a simulation prediction unit, which is used to simulate the material diffusion behavior in the coal blending process under preset boundary conditions by integrating the voxel model output by the spatial division unit and the material characteristic parameters provided by the silo modeling unit, and output simulation prediction data including coal quality diffusion distribution.

[0045] A spatial partitioning unit can refer to a computational module that uses spatial discretization technology to decompose the three-dimensional space of a coal yard into regular geometric units. This module then establishes a structured grid to enable spatial indexing and rapid query of coal quality parameters.16 Coal yard spatial data can refer to a dataset of coal pile surface geometric features acquired through laser scanning or photogrammetry, including three-dimensional attributes such as coordinate information, contour features, and surface texture.13 A three-dimensional voxelized grid model can refer to a digital model that divides a continuous space into cubic units of equal size, with each voxel storing coal quality parameters and physical property data corresponding to a spatial location.6

[0046] The silo modeling unit may refer to a dedicated processor that analyzes the internal structure of a coal pile based on the principle of radar wave reflection, and extracts the characteristic parameters of material accumulation through signal processing algorithms. The radar level meter array may refer to a multi-probe ranging system installed on the top of the silo, which realizes non-contact measurement of the surface morphology of the coal pile by emitting millimeter waves and receiving echo signals. The echo signal may refer to the reflected waveform data generated after the radar wave encounters the surface of the coal pile, and its time delay and amplitude characteristics reflect the dielectric constant and surface roughness of the material. The angle of repose parameter may refer to the physical quantity that characterizes the maximum stable slope angle when bulk materials are naturally accumulated, and its value depends on the coal particle size distribution and friction coefficient. The physical properties of the voxel grid may refer to the mechanical characteristic parameters assigned to the three-dimensional voxel unit, including the friction coefficient, stacking density, and other constitutive relationships that affect the flow behavior of the material.

[0047] The simulation prediction unit can refer to a numerical calculation engine that performs discrete element or computational fluid dynamics simulations, predicting the coal-to-fuel blending process by solving material transport equations. Preset boundary conditions can refer to the material motion constraints set during simulation calculations, including process parameters such as belt speed and drop height. Material diffusion behavior can refer to the physical phenomenon of interpenetration of materials of different qualities during coal blending, which is affected by factors such as particle size distribution and humidity. Coal quality diffusion distribution can refer to the spatial variation of quality parameters output by the simulation, reflecting the dynamic evolution of indicators such as sulfur content and calorific value during the blending process.

[0048] Through the deep integration of multi-level modeling and high-precision simulation, digital preview and optimization decision-making of the coal blending process are realized. The voxelized model constructed by the spatial division unit upgrades the traditional two-dimensional coal yard management to three-dimensional stereoscopic management and control. By accurately binding coal quality parameters with spatial coordinates, it solves the pain point of fuzzy spatial positioning of historical data. The innovative radar echo analysis algorithm of the silo modeling unit realizes dynamic measurement and automatic calibration of the repose angle parameters, overcomes the lag defects of manual sampling, and enables physical property parameters to reflect the actual state of the material in real time. The simulation prediction unit adopts a multi-physical field coupling calculation method to accurately simulate the spatiotemporal evolution of coal quality under different blending schemes, providing a visual decision-making basis for process optimization. This solution significantly improves the predictability and control accuracy of the coal blending process, ensuring boiler combustion stability while reducing environmental protection management costs.

[0049] In some embodiments, the formula for calculating the coal quality diffusion distribution is:

[0050] Among them, D t is the coal diffusion value at time t, Q i is the coal amount of the ith grid in the three-dimensional voxelized grid model; is the three-dimensional value of the spatial position corresponding to the center coordinate of the i-th grid, is the coal flow velocity correction factor of the jth silo; represents the temperature difference in the repose angle parameter between the surface and core of the coal pile in the jth silo, is the coal quality deviation sensitivity coefficient of the kth belt, obtained based on historical data statistics, C k represents the actual coal quality parameters of the kth belt, C k0 Represents the target coal quality parameters of the kth belt.

[0051] The coal flow velocity correction factor may refer to a material flow rate compensation coefficient adjusted according to the structural characteristics of the silo, and is used to correct for deviations in the discharge rate caused by factors such as the silo inclination and inner wall friction. The coal quality deviation sensitivity coefficient may refer to an indicator of the belt conveyor system's response intensity to coal quality fluctuations derived from historical data analysis, reflecting the weight of the impact of a specific conveying line on the uniformity of the mixed coal. Actual coal quality parameters may refer to coal quality data measured in real time by online detection devices, including actual detection values ​​of key process indicators such as calorific value and sulfur content. Target coal quality parameters may refer to quality standard values ​​preset based on boiler combustion requirements, which serve as a benchmark for coal blending process control.

[0052] This calculation formula achieves precise control of the coal blending process and active compensation for quality deviations by establishing a multi-parameter coupled coal diffusion calculation model. Its innovation lies in: using voxelized grids and dynamic parameter mapping technology to upgrade traditional empirical coal blending to precise regulation based on spatial quantification, effectively solving the problem of combustion instability caused by uneven coal quality distribution; by introducing a temperature gradient correction factor, the calculation accuracy of the angle of repose parameters under different working conditions is significantly improved, making the material flow simulation closer to actual physical behavior; the innovatively designed deviation sensitivity coefficient system can automatically identify key influencing links and make targeted adjustments, forming an intelligent compensation mechanism with self-learning capabilities. This solution significantly reduces the frequency of manual intervention, reducing the risk of exceeding environmental emissions standards while ensuring the thermal efficiency of the boiler.

[0053] In some embodiments, the decision control module includes: a coal blending parameter determination unit, which is used to perform multi-generation iterative calculations based on the coal quality diffusion distribution data output by the simulation prediction unit through a preset genetic algorithm, wherein each round of iteration includes selecting the individual parameter combination with the highest fitness, performing a cross-recombination operation, and implementing a mutation operation, and finally converges to generate the optimal coal blending parameters; a task decomposition unit, which is used to parse the coal loading instructions containing the optimal coal blending parameters into a device action sequence with a time relationship; a conflict detection unit, which is used to monitor the equipment operation status in real time, and generate a corresponding conflict mark when a conflict in the resource occupation of multiple devices is detected; a scheduling optimization unit, which is used to receive conflict mark information and dynamically adjust the operation timing of the belt transmission system; a compensation control unit, which is used to continuously compare the actual coal quality parameters with the preset threshold, and automatically generate a compensation coal loading task instruction when the deviation exceeds the allowable range, and feed the instruction back to the task decomposition unit to restart the control process.

[0054] The coal blending parameter determination unit may refer to a core processor that uses an intelligent optimization algorithm to calculate the coal blending ratio, and determines the optimal raw material ratio scheme that meets constraints such as calorific value and sulfur content through multi-objective optimization. A genetic algorithm may refer to a global optimization method that simulates the mechanism of biological evolution, iteratively improving the quality of the solution through operations such as selection, crossover, and mutation. An individual parameter combination may refer to a set of candidate solutions generated during the algorithm iteration process, each solution containing the material ratio of each coal type and equipment operating parameters. A crossover recombination operation may refer to the operation in a genetic algorithm that exchanges the gene fragments of two high-quality solutions to produce a new solution with parental advantages. A mutation operation may refer to an artificially introduced random perturbation mechanism that prevents the algorithm from falling into a local optimum by slightly adjusting the solution parameters. The optimal coal blending parameters may refer to the global optimal solution output after the algorithm converges, which includes process control values ​​such as coal type ratio and equipment operating speed.

[0055] A task decomposition unit can refer to a translator that converts high-level control instructions into low-level execution actions, enabling multi-device collaborative operations through timeline orchestration. A device action sequence can refer to a chronological set of mechanical operation instructions, including the timing relationships for specific actions such as bucket wheel excavator positioning and belt start and stop. A conflict detection unit can refer to a monitoring module that diagnoses system resource contention in real time, identifying operational conflicts by analyzing device status data. A conflict marker can refer to a system-generated abnormal status identifier that contains metadata such as the conflict type, the devices involved, and the severity.

[0056] The scheduling optimization unit can refer to the decision-making engine that dynamically adjusts logistics routes, eliminating equipment usage conflicts by rescheduling tasks. The compensation control unit can refer to the feedback module that implements closed-loop quality control, automatically triggering process corrections through deviation analysis. The compensation coal loading task instruction can refer to the system-generated correction control command, which includes parameters such as the type of coal to be replenished and the compensation amount.

[0057] This module realizes fully automated and precise management of the coal blending process by constructing a closed-loop intelligent control system. The coal blending parameter determination unit uses an improved genetic algorithm for multi-objective optimization, breaking through the limitations of traditional manual trial and error methods, so that the coal blending plan can simultaneously meet the dual needs of maximizing thermal efficiency and minimizing pollutant emissions; the innovative timing scheduling mechanism of the task decomposition unit converts complex control instructions into executable action chains to ensure the timing accuracy of multi-device collaborative operations; the linkage mechanism formed by the conflict detection unit and the scheduling optimization unit can avoid equipment resource competition in real time and improve the overall operating efficiency of the system; the dynamic feedback loop established by the compensation control unit eliminates the impact of coal quality fluctuations through continuous monitoring and automatic compensation, which significantly improves the stability of the final coal quality entering the furnace. This solution significantly reduces the frequency of manual intervention, ensuring the combustion efficiency of the boiler while achieving stable compliance with environmental protection indicators.

[0058] In some embodiments, the compensation control unit includes a genetic algorithm optimizer, a real-time compensator and an effect verifier, wherein the genetic algorithm optimizer receives coal quality parameter constraints, performs population initialization, fitness evaluation, selection recombination and mutation operations in a multi-dimensional parameter space by simulating the biological evolution process, and generates an optimal coal blending scheme that meets the multi-objective balance of calorific value-sulfur content-cost; the real-time compensator continuously monitors the deviation between the actual coal quality parameters and the target value, and when the deviation exceeds the preset tolerance, the dynamic adjustment mechanism is triggered, and the correction instruction is fed back to the genetic algorithm optimizer to restart the iterative calculation; the effect verifier collects feedback data after coal blending is executed, verifies the control accuracy by comparing the actual coal quality parameters with the expected targets, and synchronously transmits the verification results to the genetic algorithm optimizer for optimizing parameter weights; the genetic algorithm optimizer, the real-time compensator and the effect verifier constitute a closed-loop control loop, and ensures that the coal blending scheme meets both static constraints and dynamic compensation requirements through continuous iterative optimization.

[0059] A genetic algorithm optimizer can refer to a computational module that uses biological evolutionary principles to achieve self-optimization of parameters. It simulates natural selection mechanisms to search within the solution space for the optimal coal blending solution that satisfies multiple objective constraints. Coal quality parameter constraints can refer to a set of boundary limits for coal quality control, including key process indicators such as the lower limit on calorific value, the upper limit on sulfur content, and the cost threshold. A multidimensional parameter space can refer to a high-dimensional search domain composed of variables such as coal type ratios and equipment parameters. The number of dimensions is positively correlated with the complexity of the optimization problem. Population initialization can refer to the random generation of a first-generation set of candidate solutions within a genetic algorithm. The diversity of these solutions directly impacts the overall optimization performance. Fitness evaluation can refer to the computational process of quantifying the quality of candidate solutions, with the objective function value reflecting the degree to which a solution satisfies the constraints. Selective recombination can refer to the algorithm's selection mechanism for retaining high-quality genes, using roulette or tournament strategies to determine parent individuals. Mutation can refer to the use of random perturbations to disrupt population homogeneity and prevent premature convergence. The optimal coal blending solution can refer to the Pareto frontier solution output by the algorithm, achieving the optimal balance between calorific value, environmental performance, and economic efficiency.

[0060] A real-time compensator can refer to a feedback controller that performs online deviation correction, maintaining the stability of process indicators through dynamic adjustment. The deviation can refer to the difference between the actual detection value and the target set value, usually expressed as an absolute value or percentage. The preset tolerance can refer to the maximum fluctuation range allowed by the process, exceeding which triggers the compensation mechanism. The dynamic adjustment mechanism can refer to a rule base that automatically generates correction strategies based on real-time errors, including algorithms such as proportional-integral control. The correction instruction can refer to the parameter adjustment command output by the compensator, including the increase or decrease in coal type and the change in equipment operating parameters.

[0061] The effect verifier can refer to the post-evaluation module of coal quality, which analyzes the actual effectiveness of the control system through data backtracking. Feedback data can refer to traceability information such as coal quality test results and equipment operation logs collected during the production process. Control accuracy can refer to the degree of deviation between actual coal quality parameters and theoretical expected values, usually quantified by indicators such as root mean square error. Parameter weight can refer to the relative importance coefficient of each indicator in the objective function, which determines the preferred direction of the optimization process. The closed-loop control loop can refer to a self-regulating system composed of feedforward optimization and feedback correction to achieve continuous performance improvement. Static constraints can refer to fixed limiting factors in the production process, such as hard indicators such as the upper limit of equipment capacity. Dynamic compensation requirements can refer to real-time adjustment requirements generated by changes in operating conditions, such as temporary ratio adjustments caused by coal quality fluctuations.

[0062] This module realizes fully automated and precise management of the coal blending process by constructing a closed-loop intelligent control system. The coal blending parameter determination unit uses an improved genetic algorithm for multi-objective optimization, breaking through the limitations of traditional manual trial and error methods, so that the coal blending plan can simultaneously meet the dual needs of maximizing thermal efficiency and minimizing pollutant emissions; the innovative timing scheduling mechanism of the task decomposition unit converts complex control instructions into executable action chains to ensure the timing accuracy of multi-device collaborative operations; the linkage mechanism formed by the conflict detection unit and the scheduling optimization unit can avoid equipment resource competition in real time and improve the overall operating efficiency of the system; the dynamic feedback loop established by the compensation control unit eliminates the impact of coal quality fluctuations through continuous monitoring and automatic compensation, which significantly improves the stability of the final coal quality entering the furnace. This solution significantly reduces the frequency of manual intervention, ensuring the combustion efficiency of the boiler while achieving stable compliance with environmental protection indicators.

[0063] In some embodiments, the human-computer interaction module includes: a three-dimensional visualization unit, which receives the coal quality parameters of the voxelized grid model and generates a visual rendering of the calorific value distribution, where the coal quality parameters include calorific value and sulfur content indicators; a data tracing unit, which is interconnected with the data of the three-dimensional visualization unit, and forms a full-chain operation record of coal blending by associating the bucket wheel excavator's operating trajectory, coal flow characteristics and coal blending plan, and establishes a traceable coal quality data lineage relationship; an intelligent early warning unit, which dynamically generates procurement recommendations and triggers inventory threshold alarms based on the inventory turnover rate analysis results output by the data tracing unit; wherein the visual rendering of the three-dimensional visualization unit is fed back to the intelligent early warning unit for correcting the priority ranking of the procurement recommendations; the full-chain operation record of coal blending of the data tracing unit is synchronously updated to the three-dimensional visualization unit to realize multi-dimensional dynamic labeling and display of the coal blending process.

[0064] A three-dimensional visualization unit may refer to a graphics processing module that renders the spatial distribution of coal quality parameters based on a voxelized grid, visually displaying the three-dimensional distribution characteristics of indicators such as calorific value and sulfur content through color gradients and three-dimensional projection technology. A voxelized grid model may refer to a digital representation that discretizes the three-dimensional space of a coal pile into regular cubic units, with each voxel storing the coal quality parameters and physical property data for the corresponding area. Calorific value distribution may refer to the quantitative distribution of coal combustion calorific value parameters in three-dimensional space, reflecting the energy output potential of coal quality in different regions. A visual rendering may refer to a three-dimensional color mapping image generated by a computer graphics algorithm, using pseudo-color technology to convert numerical parameters into visually discernible color information. The sulfur content index may refer to the percentage parameter of sulfur content in coal, a key environmental indicator that directly affects the concentration of sulfur dioxide emissions after combustion.

[0065] The data traceability unit can refer to an analysis system that realizes data association throughout the entire life cycle of the coal blending process, and establishes a causal chain of operational behaviors through timestamps and spatial coordinates. The bucket wheel excavator operation trajectory can refer to the movement path data of the material-reclaiming equipment in the coal yard space, including spatiotemporal information such as position coordinates, operating speed, and material-reclaiming depth. Coal flow characteristics can refer to a set of physical state parameters of coal during transportation, including dynamic properties such as flow rate, particle size distribution, and humidity changes. The full-chain operation record of coal blending can refer to a complete process log from raw material extraction to mixed output, covering structured data such as equipment parameters, process settings, and quality inspections. The blood relationship of coal quality data can refer to a parameter association map established through data traceability technology, which reveals the transmission path between coal quality changes and operational behaviors.

[0066] An intelligent early warning unit can refer to a decision support module based on data analysis, which predicts inventory anomalies through pattern recognition and generates control recommendations. Inventory turnover rate can refer to the ratio parameter of raw material consumption to average inventory, which is a key performance indicator reflecting the efficiency of supply chain operations. Procurement recommendations can refer to system-generated raw material replenishment plans, including structured recommendations such as coal type, purchase quantity, and arrival time. Inventory threshold alarms can refer to early warning signals triggered when actual inventory levels exceed safety boundaries, including level classification and emergency response guidelines. Prioritization can refer to a decision-making algorithm that performs weighted evaluations on task sequences based on urgency and scope of impact. Multi-dimensional dynamic labeling can refer to a visual labeling technology that overlays multiple information such as time and quality in a three-dimensional scene.

[0067] This module realizes fully automated and precise management of the coal blending process by constructing a closed-loop intelligent control system. The coal blending parameter determination unit uses an improved genetic algorithm for multi-objective optimization, breaking through the limitations of traditional manual trial and error methods, so that the coal blending plan can simultaneously meet the dual needs of maximizing thermal efficiency and minimizing pollutant emissions; the innovative timing scheduling mechanism of the task decomposition unit converts complex control instructions into executable action chains to ensure the timing accuracy of multi-device collaborative operations; the linkage mechanism formed by the conflict detection unit and the scheduling optimization unit can avoid equipment resource competition in real time and improve the overall operating efficiency of the system; the dynamic feedback loop established by the compensation control unit eliminates the impact of coal quality fluctuations through continuous monitoring and automatic compensation, which significantly improves the stability of the final coal quality entering the furnace. This solution significantly reduces the frequency of manual intervention, ensuring the combustion efficiency of the boiler while achieving stable compliance with environmental protection indicators.

[0068] Corresponding to the above system embodiment, this specification also provides an embodiment of a coal blending control method. Figure 2 FIG1 shows a flow chart of a coal blending control method provided in some embodiments of this specification. Figure 2 As shown, the specific steps include: Step 201 : Collect three-dimensional point cloud data, equipment positioning data, and coal quality parameters of the coal yard, and generate and output a coal flow feature vector based on the collected data.

[0069] Step 202 : Based on the pre-built three-dimensional coal yard model, the pre-calculated silo coal pile angle of repose parameters, and the received coal flow characteristic vector, generate and output simulation prediction data including coal quality diffusion distribution.

[0070] Step 203 : Based on the simulation prediction data and the preset genetic algorithm, it is iterated to generate the optimal coal blending parameters, and based on the optimal coal blending parameters, a control instruction including the equipment action sequence and the compensation coal loading task is generated.

[0071] Step 204 : Based on the control instructions received in real time, the instruction execution results, coal quality deviation warning and inventory turnover information are generated and displayed on the target display screen.

[0072] In some optional implementations, three-dimensional point cloud data, equipment positioning data, and coal quality parameters of the coal yard are collected, and a coal flow feature vector is generated and output based on the collected data, including: Obtain three-dimensional point cloud data with time stamps generated by scanning the coal yard surface at a preset period; Obtain real-time bucket wheel machine boom angle and spatial coordinate data, generate and output equipment positioning data; Obtain coal quality parameters obtained by online analysis of belt coal flow, including ash content, sulfur content and calorific value parameters; Map the coal quality parameters to the three-dimensional space corresponding to the bucket wheel excavator, and establish the corresponding relationship between the coal quality parameters and the coal quality in the three-dimensional space based on the timestamp; Temporally and spatially align 3D point cloud data with device positioning data; Based on the coal quality correspondence, as well as the time-space aligned three-dimensional point cloud data and equipment positioning data, a standardized coal flow feature vector is generated.

[0073] In some optional implementations, based on a pre-built three-dimensional coal yard model, pre-calculated silo coal pile angle of repose parameters, and received coal flow characteristic vectors, simulation prediction data including coal quality diffusion distribution is generated and output, including: Receive coal yard spatial data transmitted by external acquisition devices and establish a three-dimensional voxel grid model with coal quality parameter mapping relationship; Based on the echo signal of the radar level meter array targeting the coal pile, the repose angle parameter that characterizes the material characteristics is analyzed and generated, and this parameter is dynamically associated with the physical properties of the voxel grid established by the spatial division unit; By integrating the voxelized model output by the space division unit and the material characteristic parameters provided by the silo modeling unit, the material diffusion behavior during the coal blending process is simulated under preset boundary conditions, and simulation prediction data including coal quality diffusion distribution is output.

[0074] In some optional implementations, the formula for calculating the coal quality diffusion distribution is:

[0075] Among them, D t is the coal diffusion value at time t, Q i is the coal amount of the ith grid in the three-dimensional voxelized grid model; is the three-dimensional value of the spatial position corresponding to the center coordinate of the i-th grid, is the coal flow velocity correction factor of the jth silo; represents the temperature difference in the repose angle parameter between the surface and core of the coal pile in the jth silo, is the coal quality deviation sensitivity coefficient of the kth belt, obtained based on historical data statistics, C k represents the actual coal quality parameters of the kth belt, C k0 Represents the target coal quality parameters of the kth belt.

[0076] In some optional implementations, based on simulation prediction data and a preset genetic algorithm, an optimal coal blending parameter is generated through iteration, and based on the optimal coal blending parameter, a control instruction including an equipment action sequence and a compensation coal loading task is generated, including: Based on the coal quality diffusion distribution data output by the simulation prediction unit, a preset genetic algorithm is used to perform multiple generations of iterative calculations. Each iteration includes selecting the individual parameter combination with the highest fitness, performing crossover recombination operations, and implementing mutation operations, ultimately converging to generate the optimal coal blending parameters. Parse the coal loading instruction containing the optimal coal blending parameters into a sequence of equipment actions with a time sequence relationship; Monitor the device's operating status in real time and generate corresponding conflict markers when conflicts in resource usage among multiple devices are detected; Receive conflict mark information and dynamically adjust the operating timing of the belt transmission system; Continuously compare the actual coal quality parameters with the preset thresholds. When the deviation exceeds the allowable range, a compensation coal loading task instruction is automatically generated and fed back to the task decomposition unit to restart the control process.

[0077] In some optional implementations, actual coal quality parameters are continuously compared with preset thresholds. When the deviation exceeds the allowable range, a compensation coal loading task instruction is automatically generated and fed back to the task decomposition unit to restart the control process, including: Accepting coal quality parameter constraints, the system simulates the biological evolution process to perform population initialization, fitness evaluation, selection, recombination, and mutation operations in a multi-dimensional parameter space to generate an optimal coal blending solution that meets the multi-objective balance of calorific value, sulfur content, and cost. Continuously monitor the deviation between actual coal quality parameters and target values. When the deviation exceeds the preset tolerance, a dynamic adjustment mechanism is triggered, and correction instructions are fed back to the genetic algorithm optimizer to restart iterative calculations. Collect feedback data after coal blending is executed, verify the control accuracy by comparing the actual coal quality parameters with the expected targets, and transmit the verification results synchronously to the genetic algorithm optimizer for optimizing parameter weights; Continuous iterative optimization ensures that the coal blending plan meets both static constraints and dynamic compensation requirements.

[0078] In some optional implementations, based on the control instructions received in real time, the command execution results, coal quality deviation warnings, and inventory turnover information are generated and displayed on a target display screen, including: Receive coal quality parameters and generate a visual rendering of calorific value distribution. Coal quality parameters include calorific value and sulfur content index; By associating the bucket wheel excavator's operating trajectory, coal flow characteristics, and coal blending plan, a full-chain coal blending operation record is formed, and a traceable coal quality data lineage relationship is established; Based on the inventory turnover analysis results output by the data traceability unit, purchase recommendations are dynamically generated and inventory threshold alarms are triggered; Among them, the visual rendering is used to correct the priority ranking of procurement recommendations; the full-chain coal blending operation record is used to realize the multi-dimensional dynamic labeling display of the coal blending process.

[0079] The above is a schematic diagram of a coal blending control method according to this embodiment. It should be noted that the technical solution of this coal blending control method is based on the same concept as the technical solution of the aforementioned coal blending control system. For details not described in detail in the technical solution of the coal blending control method, please refer to the description of the technical solution of the aforementioned coal blending control system.

[0080] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0081] The preferred embodiments disclosed above are intended only to help illustrate this specification. The optional embodiments do not exhaustively describe all details, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the content of the present invention. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize this specification. This specification is limited only by the claims and their full scope and equivalents.

Claims

1. A coal blending control system, characterized in that: include: a data acquisition module configured to collect three-dimensional point cloud data, equipment positioning data, and coal quality parameters of the coal yard, and generate and output a coal flow feature vector based on the collected data; a modeling and analysis module configured to generate and output simulation prediction data including coal quality diffusion distribution based on a pre-built three-dimensional coal yard model, pre-calculated silo coal pile angle of repose parameters, and the received coal flow characteristic vector; a decision control module configured to iterate based on the simulation prediction data and a preset genetic algorithm to generate optimal coal blending parameters, and generate control instructions including equipment action sequences and compensation coal loading tasks based on the optimal coal blending parameters; The human-computer interaction module is configured to generate instruction execution results, coal quality deviation warnings and inventory turnover information based on the control instructions received in real time and display them on the target display screen.

2. The system according to claim 1, wherein: The data acquisition module includes: A laser scanning unit, configured to scan the surface of the coal yard at a preset period to generate and output the three-dimensional point cloud data with a time stamp; A positioning monitoring unit, used to collect bucket wheel machine boom angle and spatial coordinate data in real time, generate and output the equipment positioning data; A coal quality detection unit, used for online analysis of the ash content, sulfur content and calorific value parameters of the belt coal flow, and obtaining the coal quality parameters with a time stamp; A data fusion unit is used to map the coal quality parameters to the three-dimensional space corresponding to the bucket wheel excavator, establish a corresponding relationship between the coal quality parameters and the coal quality in the three-dimensional space based on the timestamp; align the three-dimensional point cloud data with the equipment positioning data in time and space; and generate a standardized coal flow feature vector based on the coal quality corresponding relationship and the three-dimensional point cloud data and equipment positioning data that have been aligned in time and space.

3. The system according to claim 2, characterized in that The modeling and analysis module includes: A space division unit is used to establish a three-dimensional voxel grid model with a mapping relationship of coal quality parameters by receiving coal yard space data transmitted by an external acquisition device; The silo modeling unit is used to analyze and generate the angle of repose parameter that characterizes the material properties based on the echo signal of the radar level meter array targeting the coal pile, and dynamically associate this parameter with the physical properties of the voxel grid established by the spatial division unit; The simulation prediction unit is used to simulate the material diffusion behavior in the coal blending process under preset boundary conditions by integrating the voxel model output by the space division unit and the material characteristic parameters provided by the silo modeling unit, and output the simulation prediction data including the coal quality diffusion distribution.

4. The system according to claim 3, characterized in that The calculation formula for calculating the coal quality diffusion distribution is: Among them, D t is the coal diffusion value at time t, Q i is the coal amount of the i-th grid in the three-dimensional voxelized grid model; is the three-dimensional value of the spatial position corresponding to the center coordinate of the i-th grid, is the coal flow velocity correction factor of the jth silo; represents the temperature difference in the repose angle parameter between the surface and core of the coal pile in the jth silo, is the coal quality deviation sensitivity coefficient of the kth belt, obtained based on historical data statistics, C k represents the actual coal quality parameters of the kth belt, C k0 Represents the target coal quality parameters of the kth belt.

5. The system according to claim 3, wherein: The decision control module includes: The coal blending parameter determination unit is used to perform multi-generation iterative calculations using a preset genetic algorithm based on the coal quality diffusion distribution data output by the simulation prediction unit. Each iteration includes selecting the individual parameter combination with the highest fitness, performing crossover recombination operations, and implementing mutation operations, ultimately converging to generate the optimal coal blending parameters. The task decomposition unit is used to parse the coal loading instruction containing the optimal coal blending parameters into a sequence of equipment actions with a time sequence relationship; The conflict detection unit is used to monitor the operating status of the equipment in real time and generate a corresponding conflict mark when a conflict in resource occupation by multiple devices is detected; A scheduling optimization unit is used to receive conflict marker information and dynamically adjust the operating timing of the belt transmission system; The compensation control unit is used to continuously compare the actual coal quality parameters with the preset thresholds. When the deviation exceeds the allowable range, it automatically generates a compensation coal loading task instruction and feeds the instruction back to the task decomposition unit to restart the control process.

6. The system according to claim 5, characterized in that The compensation control unit includes a genetic algorithm optimizer, a real-time compensator and an effect verifier, wherein: The genetic algorithm optimizer receives coal quality parameter constraints and performs population initialization, fitness evaluation, selection, recombination and mutation operations in a multi-dimensional parameter space by simulating the biological evolution process to generate an optimal coal blending solution that meets the multi-objective balance of calorific value, sulfur content and cost; The real-time compensator continuously monitors the deviation between the actual coal quality parameters and the target values. When the deviation exceeds the preset tolerance, the dynamic adjustment mechanism is triggered, and the correction instruction is fed back to the genetic algorithm optimizer to restart the iterative calculation; The effect verifier collects feedback data after coal blending is executed, verifies the control accuracy by comparing the actual coal quality parameters with the expected targets, and synchronously transmits the verification results to the genetic algorithm optimizer for optimizing parameter weights; The genetic algorithm optimizer, real-time compensator and effect verifier form a closed-loop control circuit, which ensures that the coal blending plan meets both static constraints and dynamic compensation requirements through continuous iterative optimization.

7. The system according to claim 6, characterized in that The human-computer interaction module includes: a three-dimensional visualization unit, receiving the coal quality parameters of the voxelized grid model and generating a visualization rendering of the calorific value distribution, wherein the coal quality parameters include calorific value and sulfur content index; The data tracing unit is interconnected with the data of the three-dimensional visualization unit, and forms a full-chain operation record of coal blending by associating the bucket wheel excavator's running trajectory, coal flow characteristics and coal blending plan, and establishes a traceable blood relationship of coal quality data; An intelligent early warning unit, based on the inventory turnover analysis results output by the data tracing unit, dynamically generates procurement recommendations and triggers inventory threshold alarms; Among them, the visual rendering of the three-dimensional visualization unit is fed back to the intelligent early warning unit for correcting the priority ranking of procurement recommendations; the full-chain coal blending operation records of the data traceability unit are synchronously updated to the three-dimensional visualization unit to realize multi-dimensional dynamic labeling and display of the coal blending process.

8. A method for controlling coal blending, characterized in that: The method is applied to the system according to any one of claims 1 to 7, and the method comprises: Collecting three-dimensional point cloud data, equipment positioning data, and coal quality parameters of the coal yard, and generating and outputting a coal flow feature vector based on the collected data; Generate and output simulation prediction data including coal quality diffusion distribution based on a pre-built three-dimensional coal yard model, pre-calculated silo coal pile angle of repose parameters, and the received coal flow characteristic vector; Iterating based on the simulation prediction data and a preset genetic algorithm to generate optimal coal blending parameters, and generating control instructions including equipment action sequences and compensation coal loading tasks based on the optimal coal blending parameters; Based on the control instructions received in real time, instruction execution results, coal quality deviation warnings and inventory turnover information are generated and displayed on a target display screen.

9. The method according to claim 8, characterized in that The collecting of three-dimensional point cloud data, equipment positioning data, and coal quality parameters of the coal yard, and generating and outputting a coal flow feature vector based on the collected data, includes: Obtaining the three-dimensional point cloud data with time stamps generated by scanning the coal yard surface at a preset period; Acquire real-time bucket wheel machine boom angle and spatial coordinate data, and generate and output the equipment positioning data; Obtaining coal quality parameters obtained by online analysis of the belt coal flow, wherein the coal quality parameters include ash content, sulfur content and calorific value parameters; Mapping the coal quality parameters to the three-dimensional space corresponding to the bucket wheel excavator, and establishing a corresponding relationship between the coal quality parameters and the coal quality in the three-dimensional space according to the timestamp; Performing spatiotemporal alignment of the three-dimensional point cloud data with device positioning data; The standardized coal flow feature vector is generated based on the coal quality correspondence, the time-space aligned three-dimensional point cloud data and the equipment positioning data.

10. The method according to claim 9, characterized in that Based on the pre-built three-dimensional coal yard model, the pre-calculated silo coal pile angle of repose parameters, and the received coal flow characteristic vector, simulation prediction data including coal quality diffusion distribution is generated and output, including: Receive coal yard spatial data transmitted by external acquisition devices and establish a three-dimensional voxel grid model with coal quality parameter mapping relationship; Based on the echo signal of the radar level meter array targeting the coal pile, the repose angle parameter that characterizes the material characteristics is analyzed and generated, and this parameter is dynamically associated with the physical properties of the voxel grid established by the spatial division unit; The voxelized model output by the space division unit and the material characteristic parameters provided by the silo modeling unit are integrated to simulate the material diffusion behavior in the coal blending process under preset boundary conditions, and output the simulation prediction data including the coal quality diffusion distribution.

Images