Intelligent preparation process of calcined sand based on distributed internet of things

Abstract

The application discloses a kind of based on distributed internet of things and is roasted sand intelligent preparation process, including the following steps: in each process node of roasting sand production line, distributed intelligent sensing terminal is deployed, and whole-process process parameters are collected;Real-time processing is carried out to the data collected by each node edge computing terminal, and closed-loop control is executed according to local decision;Cloud collaborative platform gathers the data of all distributed terminals, establishes quality prediction model and generates global optimization instruction;The global optimization instruction is issued to each edge node to execute parameter adjustment, and forms full-closed-loop intelligent control;Through distributed intelligent temperature field control and online quality prediction closed-loop feedback, the high-temperature linear expansion rate of roasting sand is reduced, and batch-to-batch quality fluctuation is greatly reduced.

Description

Technical Field

[0001] This invention belongs to the field of calcined sand preparation technology, specifically referring to a smart calcined sand preparation process based on a distributed Internet of Things. Background Technology

[0002] Calcined sand is a high-quality foundry base material obtained by modifying raw sand through high-temperature heat treatment. After high-temperature calcination, the silica sand undergoes a crystal transformation, which significantly reduces its thermal expansion rate, eliminates organic matter and water of crystallization, and alters the physical and chemical properties of the sand grain surface. Coated sand produced from calcined sand therefore possesses excellent properties such as low expansion, low gas generation, and high strength, and is widely used in foundry coated sand, glass sand, and ceramsite sand, making it a key basic material in the high-end foundry industry. Research shows that 80%–90% of casting production is sand casting, and the vast majority of castings still rely on sand molding processes for production. Therefore, the quality of calcined sand directly affects the quality of castings and production costs.

[0003] According to the definition in GB / T 9442-2024 "Silica Sand for Foundry", calcined sand refers to natural silica sand for foundry that has been calcined at high temperature (600℃~900℃). The main performance indicators of calcined sand include high-temperature linear expansion rate (the percentage of linear change in particle size after free expansion at 1000℃ to the original size at room temperature), loss on ignition (the mass loss rate after calcination at 850℃±25℃ for 30 min, with an allowable deviation of ≤0.3%), gas generation (the gas generation per unit mass of sand mold at 600℃ is determined by the quartz tube method, with a resolution of 0.1 mL), and tensile strength, etc.

[0004] Currently, the field of calcined sand production technology has achieved a certain level of patent portfolio and technological accumulation, mainly reflected in the following aspects: (a) Heat recovery and energy-saving technology. Existing patents disclose a general-purpose heat recovery system for drying sand and roasting sand. Through the integrated application of boiling device and heat recovery system, heat energy is recovered in the production of roasting sand for casting, so as to reduce production costs and energy consumption.

[0005] (II) Improvements to the Structure of the Roasting Furnace. Existing technologies have disclosed high-temperature roasting furnaces for rapidly preheating coated sand. By incorporating a ventilated and insulated barrel, a main roasting furnace body, and a sealed top cover, these technologies address the issues of cold air entering the preheating chamber during sand feeding and the impact of low heating air temperature on the combustion chamber temperature. Furthermore, high-capacity sand thermal regeneration roasting furnaces utilize partitions within the furnace cavity to divide it into several working chambers, fully leveraging high-temperature gas in a layered manner to increase heat treatment output under the same heat source conditions. Regarding the control system, existing patents have proposed control systems and methods for gas-suspended roasting furnaces. These systems collect raw operating data in real-time using sensors and transmit this data to a standardized module based on a distributed control system. The roasting furnace is then optimized through a control optimization model. Other research combines distributed control systems (DCS) with edge computing for temperature control of sintering furnaces, employing programmable logic controllers (PLCs) to reduce electrical wiring and switch contacts, thereby enhancing the intelligence level of the control system.

[0006] (III) Calcination Quality Improvement Technology. Existing patents disclose calcination methods for controlling porosity defects in quartz sand and increasing its bulk density. These methods utilize alternating vacuum and oxygen-containing atmospheres to improve the compactness of quartz sand. Studies show that high-temperature treatment reduces the gas evolution of quartz sand, significantly increases the tensile strength of quartz resin sand, and results in a more uniform particle size distribution. The thermal expansion capacity of calcined quartz sand is also lower than that of raw sand. Calcinated sand is characterized by a clean surface, low expansion rate, low mud content, and low loss on ignition. In resin sand processes, it can significantly improve bonding efficiency, reduce resin addition, and decrease gas evolution in the sand core. Furthermore, due to the phase transformation of silica sand, it can reduce expansion-related porosity defects in castings.

[0007] (IV) Preliminary Application of Distributed Control Technology in Sand Processing. In the field of sand processing control, existing literature introduces a distributed microcomputer monitoring system for molding sand processing in foundry production lines. This system employs a two-level distributed control system with a programmable logic controller (PLC) and an intelligent molding sand moisture controller for real-time control, and a microcomputer for monitoring and management. Furthermore, research has developed a remote operation and maintenance system for sand processing equipment based on an intelligent gateway, solving some problems related to remote operation monitoring, remote data acquisition, and remote alarm management of sand processing equipment. Regarding combustion control in roasting furnaces, existing technologies utilize fieldbuses to form a local area network, enabling each field controller to act as the core of the real-time control system. In addition, existing patents disclose a digital twin system and control method for roasting kilns. By constructing a 3D model of the kiln body, real-time monitoring of kiln temperature, pressure, and kiln lining temperature is achieved, and a mathematical model is established to simulate the process parameters within the kiln.

[0008] However, considering existing technologies, the following significant technical shortcomings still exist in the current field of roasted sand production technology: First, existing technologies mainly focus on localized improvements to single equipment, such as optimizing the structure of the roasting furnace and adding heat recovery devices, lacking integrated and intelligent control solutions for the entire roasted sand production system. Information is isolated between various process units (raw material pretreatment, preheating, roasting, cooling, screening, and packaging), and parameters at each node are not interconnected, making global collaborative optimization difficult. Particularly in the temperature control of the roasting furnace, although distributed control systems (DCS) have been applied, they are mostly limited to data acquisition and local PID control, failing to achieve coordinated regulation between multiple combustion zones and integrated end-edge-cloud intelligent control throughout the entire process.

[0009] Secondly, the detection of calcination quality still relies on offline sampling and testing, which has a long testing cycle (usually several hours to half a day), making it impossible to achieve online real-time monitoring and closed-loop feedback control. When quality abnormalities are detected, a large number of substandard products have often already been produced, resulting in serious resource waste and economic losses. Currently, there is no intelligent solution in the technology that can predict the quality of calcined sand in real time based on multi-source process parameters and automatically adjust the process parameters.

[0010] Third, in terms of temperature field control in the roasting furnace, existing technologies typically use single-point or a few temperature measurement points, which cannot obtain three-dimensional temperature distribution information inside the furnace, resulting in blind spots in temperature control. The combustion control of each combustion zone is independent of each other and lacks coordination, making it difficult to ensure the uniformity of temperature distribution across the furnace cross-section, and easily leading to regional differences in over-burning and under-burning.

[0011] Fourth, existing technologies lack real-time monitoring and intelligent optimization methods for carbon emissions throughout the entire roasted sand production process. Reducing energy consumption and carbon emissions during roasted sand production has become a crucial issue facing the industry.

[0012] Fifth, the traceability capability of calcined sand products is weak. Existing processes cannot provide complete production process data records from raw materials to finished products, making it difficult to meet the stringent requirements of the high-end casting industry for full-process traceability of material quality.

[0013] To address the shortcomings of the existing technologies, this invention proposes a smart preparation process for calcined sand based on a distributed Internet of Things. Summary of the Invention

[0014] To address the needs and problems mentioned in the background above, this invention provides a smart preparation process for calcined sand based on a distributed Internet of Things, which at least partially solves the aforementioned problems.

[0015] According to the technical solution of the present invention, a smart preparation process for calcined sand based on a distributed Internet of Things is provided, comprising the following steps: Distributed intelligent sensing terminals are deployed at each process node of the calcined sand production line to collect process parameters throughout the entire process; Each node's edge computing terminal processes the collected data in real time and executes closed-loop control based on local decisions; The cloud-based collaborative platform aggregates data from all distributed terminals, establishes a quality prediction model, and generates global optimization instructions. The global optimization command is sent to each edge node to adjust the execution parameters, forming a fully closed-loop intelligent control.

[0016] Preferably, the distributed intelligent sensing terminal includes one or more of the following sensors deployed at at least three different nodes in the raw material warehouse, preheating section, roasting section, cooling section, screening section, and finished product warehouse: temperature sensor, oxygen content sensor, sand level sensor, vibration sensor, gas composition sensor, moisture sensor, particle size analyzer, and impurity detection sensor; each sensor is interconnected through an industrial Internet of Things protocol.

[0017] Preferably, the edge computing terminal includes intelligent combustion control terminals set in each combustion zone of the calcining furnace. Each intelligent combustion control terminal integrates a gas flow sensor and a proportional regulating valve, a combustion air flow sensor and a variable frequency fan, an oxygen content sensor and an edge computing chip. The edge computing chip independently adjusts the air-fuel ratio in its area and coordinates with the intelligent combustion control terminals in adjacent combustion zones to control the temperature distribution of the furnace cross section through an Internet of Things bus.

[0018] Preferably, a distributed thermocouple array is deployed along the axial and radial directions inside the roasting furnace, and an infrared thermal imaging sensor is set at the observation hole of the roasting furnace. The edge computing terminal fuses the thermocouple array data and the infrared thermal imaging data to construct a three-dimensional temperature field digital model inside the furnace, and automatically adjusts the combustion parameters of each combustion zone based on the model.

[0019] Preferably, an edge computing terminal integrating temperature sensor, vibration sensor and sand level sensor is deployed at the discharge port of the roasting furnace. This edge computing terminal establishes a dynamic discrimination model of roasting endpoint by real-time analysis of sand temperature change rate and flow characteristics, and adaptively adjusts the discharge speed according to the discrimination result to control the residence time of sand in the roasting zone.

[0020] Preferably, the cloud-based collaborative platform is equipped with a calcined sand quality prediction model based on machine learning algorithms. This model takes multi-source process parameters collected in real time as input and predicts at least one of the following quality indicators of the product: high-temperature linear expansion rate, gas generation, loss on ignition, and tensile strength. When the predicted value deviates from the target range, the system automatically generates the optimal combination of process parameters through an optimization algorithm and sends it to each edge node for adjustment.

[0021] Preferably, it also includes real-time carbon emission monitoring and intelligent optimization steps: deploying CO2, CO, and NO monitoring at fuel inlets, exhaust outlets, and key equipment nodes.x With particulate matter gas composition sensors, carbon emission data is collected in real time through a distributed Internet of Things platform. The cloud-based carbon emission calculation model automatically calculates the carbon footprint of each product based on fuel consumption, product output, and emission concentration. Through multi-objective optimization algorithms, it seeks the optimal balance point between energy consumption, carbon emissions, and production capacity while ensuring product quality.

[0022] Preferably, it also includes a full-process digital twin and traceability step: a digital twin model of the calcined sand production line is constructed based on the full-process data collected by the distributed Internet of Things, and the operating status of the physical production line is mapped in real time in the virtual space, supporting the backtracking of historical production processes and virtual debugging of process parameters; a traceability QR code containing information such as the source of raw materials, key parameters of each process section, quality inspection data, production time and production line number is generated on the finished product packaging.

[0023] Preferably, the roasted sand production line includes a raw material pretreatment unit, a preheating unit, a roasting unit, a cooling unit, a screening unit, and a packaging unit connected in sequence; wherein the preheating unit uses the high-temperature exhaust gas emitted from the roasting unit to preheat the raw sand in a gradient manner, and the cooling unit adopts a multi-stage air-cooling and water-cooling series structure, and sends the recovered hot air into the preheating unit. Beneficial effects

[0024] 1. Significantly improved product quality. Through distributed intelligent temperature field control and online quality prediction closed-loop feedback, the high-temperature linear expansion rate of calcined sand is reduced, and the quality fluctuation between batches is greatly reduced.

[0025] 2. Energy consumption is significantly reduced. Through precise air-fuel ratio control and waste heat recovery, the energy consumption per unit product is reduced compared to traditional processes.

[0026] 3. Significantly improved production efficiency. The automation rate of the entire process can reach over 95%, manual intervention is reduced by 80%, the overall efficiency (OEE) of the production line is improved by over 30%, and the annual capacity of a single production line can be increased by over 20%.

[0027] 4. Environmental emissions meet and are controllable. Real-time carbon emission monitoring and intelligent optimization systems reduce carbon emissions per unit of product, NO... x Emission concentration can be controlled at ≤100mg / Nm 3 Particulate matter emission concentration ≤20mg / Nm 3 All indicators of exhaust gas emissions meet the ultra-low emission standards.

[0028] 5. Strong traceability. The end-to-end data recording and QR code traceability system from raw materials to finished products can meet the stringent requirements of high-end casting enterprises for material quality traceability and enhance product market competitiveness.

[0029] 6. High applicability. The process scheme of this invention is compatible with various roasted sand production scenarios such as quartz sand, ceramsite sand, and coated sand regeneration. Process parameters and control strategies can be flexibly adjusted according to the characteristics of different raw materials, and it has broad application prospects. Detailed Implementation

[0030] The technical solutions in the embodiments will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection.

[0031] This invention overcomes the shortcomings of existing technologies by providing a smart calcined sand preparation process based on a distributed Internet of Things (IoT). With a distributed IoT architecture at its core, it organically integrates multiple intelligent sensing terminals, edge computing nodes, and execution control units distributed throughout the entire calcined sand production process through an industrial IoT platform, forming an intelligent production system with end-edge-cloud collaboration. Specifically, it includes the following process steps: Step 1: Raw material pretreatment and intelligent feeding After large particles of impurities are removed by a vibrating screen, the raw sand (preferably silica sand) enters the raw material buffer silo. Distributed sensor terminals are deployed at the top and bottom of the raw material silo, including: Material level sensor: Uses radar level gauge or ultrasonic level gauge to monitor the sand level in the silo in real time, with a measurement accuracy of ±1cm, to prevent empty silo or overflow. Moisture sensor: Employs near-infrared moisture detection technology or microwave moisture detection technology to detect the moisture content of raw sand online, with a measurement accuracy of ±0.1% and a detection range of 0%~10%; Particle size analyzer: It adopts online dynamic image analysis technology or laser diffraction technology to monitor the particle size distribution of the feed sand in real time. It can simultaneously calculate morphological parameters such as particle size, aspect ratio, and roundness. The detection range is 0.01~3500μm, and the detection frequency is ≥1 time / minute. Impurity detection sensor: Employs photoelectric color sorting technology or X-ray fluorescence spectroscopy to detect the impurity content in raw materials online.

[0032] The intelligent feeder at the bottom of the raw material silo adopts frequency conversion speed regulation control. Based on the target feeding rate instruction sent from the cloud (the feeding rate range is 5~50 tons / hour) and combined with the sand level data fed back by local sensors, it automatically adjusts the feeding amount through PID control algorithm to achieve precise material batching.

[0033] Step 2: Preheating Section – Waste Heat Recovery and Gradient Heating The raw sand is fed into the preheating unit via a closed conveyor system. The preheating unit utilizes the high-temperature exhaust gas (temperature range 150~250℃) from the roasting furnace to preheat the raw sand, raising its temperature from room temperature to 150~200℃. The preheating unit is equipped with: Distributed temperature sensor array: A temperature measuring point is arranged every 0.5~1.0m along the axial direction of the preheating device to form a temperature field distribution map, with a temperature measurement accuracy of ±1℃; Intelligent air valve control system: Automatically adjusts the exhaust gas flow rate in each zone based on temperature field data to achieve gradient preheating, with temperature control deviation in each zone ≤ ±5℃; Exhaust gas flow sensor: Real-time monitoring of exhaust gas flow in each area, with a measurement accuracy of ±2%FS.

[0034] After preheating, the sand material is judged by the edge computing terminal to see if it has reached the predetermined temperature threshold (150~200℃). Only if it is qualified can it enter the roasting section. If the sand material does not meet the standard, it will be automatically returned to the preheating device for recirculation and preheating.

[0035] Step 3: Firing Section – Intelligent Temperature Field Sensing and Dynamic Firing The preheated sand is fed into the roasting furnace through a sealed feeding device. The roasting furnace preferably employs a double-chamber vertical kiln structure, but a rotary kiln or fluidized bed roasting furnace structure can also be used. The control system of the roasting furnace is the core of the technical solution of this invention. (1) Intelligent temperature field sensing and control. A distributed thermocouple array (no less than 12 temperature measuring points, preferably 16-24 temperature measuring points) is deployed along the axial and radial directions inside the roasting furnace to collect three-dimensional temperature distribution data in real time. Each temperature measuring point uses K-type or S-type thermocouples with a temperature measurement accuracy of ±1℃ and a response time of ≤1s. At the same time, an infrared thermal imaging sensor (temperature measurement range 300~1500℃, temperature measurement accuracy ±2℃) is set at the observation hole of the roasting furnace. The sensor data is fused with the thermocouple array data to construct a digital model of the three-dimensional temperature field inside the furnace.

[0036] (2) Distributed combustion control. Independent intelligent combustion control terminals are installed in each combustion zone of the calcining furnace. Each terminal includes: Gas flow sensor and proportional control valve (gas flow detection accuracy ±1%FS, control valve opening accuracy ±1%). Combustion air flow sensor and variable frequency fan (air flow detection accuracy ±1.5%FS, variable frequency speed range 0~100%). The oxygen content sensor (λ sensor) uses a zirconia oxygen analyzer, with a measurement range of 0~21% and an accuracy of ±0.2%. The edge computing chip uses an ARM Cortex-A series processor or an industrial-grade processor with equivalent performance, a main frequency of ≥1GHz, memory of ≥1GB, and a built-in real-time operating system.

[0037] Each combustion control terminal independently adjusts the air-fuel ratio based on local sensor data (the air-fuel ratio adjustment range is 0.8 to 1.2 times the theoretical air-fuel ratio), and simultaneously collaborates with adjacent terminals via an industrial IoT bus (preferably Modbus TCP / IP or PROFINET protocol) to ensure uniform temperature distribution across the furnace cross-section. With independent control of each combustion zone, the uniformity of temperature distribution across the furnace cross-section is improved, and the temperature deviation can be reduced from ±30℃ in traditional processes to within ±10℃.

[0038] (3) Intelligent adjustment of residence time. An intelligent discharge valve is installed at the discharge port of the roasting furnace, driven by a servo motor, with a discharge speed control accuracy of ±1%. The control system of the discharge valve dynamically adjusts the discharge speed according to the sand level sensor in the furnace (detection accuracy ±1cm), temperature field data, and target roasting time (preferably 30~60 minutes, adjustable according to the type of raw sand and target product requirements) through the following formula:

[0039] In the formula: v out V represents the discharge speed (kg / s). furnace Effective volume of the roasting furnace (m³) 3 ρ is the bulk density of the sand (kg / m³) 3 ), η is the filling coefficient (usually taken as 0.6~0.8), t target S represents the target roasting time (s), and S represents the cross-sectional area of ​​the discharge port (m²). 2 ).

[0040] (4) Intelligent atmosphere control. The oxygen content sensor in the furnace monitors the combustion atmosphere in real time, and the edge computing terminal automatically adjusts the air-fuel ratio and combustion air volume according to the target oxygen content (usually controlled at 3%~6%) to achieve precise switching between reducing, neutral or oxidizing atmospheres. The switching response time is ≤30s.

[0041] The roasting temperature is typically controlled between 800 and 1100℃ (adjusted according to the type of raw sand and the requirements of the target product), with a temperature control accuracy of ±10℃. The temperature of the roasted sand can reach 700 to 900℃. When the roasting temperature exceeds 573℃, quartz undergoes a crystal transformation from α-quartz to β-quartz; within the range of 800 to 1100℃, a further transformation from β-quartz to β-tridymite occurs, while simultaneously removing organic matter and water of crystallization, achieving deep activation of the sand grain surface.

[0042] Step 4: Cooling Section – High-Efficiency Cooling and Cascaded Heat Recovery The high-temperature calcined sand is fed into the cooling system via the discharge device. The cooling system adopts a multi-stage air-cooled + water-cooled series structure: Primary cooling: Fluidized bed air cooling, which utilizes the heat exchange between cooling air and high-temperature sand. The cooling air is heated to 300~400℃ and then recycled for use in the preheating section, with a heat recovery efficiency of ≥70%. Secondary cooling: Jacketed water cooling, with cooling water inlet temperature ≤25℃ and outlet temperature ≤45℃, reducing the sand temperature to below 80℃; Three-stage cooling: natural air cooling or continued cooling in a fluidized bed to reduce the sand temperature to below 40℃.

[0043] Each cooling section is equipped with a distributed temperature monitoring terminal (temperature measurement accuracy ±1℃). The intelligent control system automatically adjusts the cooling air volume (air volume adjustment range 0~100% of rated air volume) or cooling water flow rate (flow rate adjustment range 0~100% of rated flow rate) according to the sand temperature to ensure uniform cooling speed. The cooling speed is controlled within the range of 5~20℃ / min to prevent sand particles from cracking due to rapid cooling.

[0044] Step 5: Screening and Finished Product Quality Control The cooled calcined sand is then classified by particle size using a vibrating screen. Screening system configuration: Online particle size monitoring terminal: It adopts dynamic image analysis technology or laser diffraction technology to detect the distribution ratio of each particle size after sieving in real time (detection frequency ≥ 1 time / minute, detection accuracy ±0.5%), and is usually graded into specifications such as 20~40 mesh, 40~70 mesh, and 70~140 mesh. Intelligent image analysis terminal: performs real-time image analysis of sand particle morphology (resolution ≥ 5 million pixels, analysis frequency ≥ 1 time / minute) to identify abnormally large particles or excessively broken particles. Impurity detection sensor: It uses photoelectric color sorting technology to remove discolored particles, with a color sorting accuracy of ≥99%.

[0045] All quality inspection data are uploaded to the cloud-based quality management system in real time and compared with preset quality standards. Quality control indicators include: high-temperature linear expansion rate, loss on ignition, gas evolution, tensile strength, etc. If a quality abnormality is detected, the system automatically alarms and records the abnormal batch, while simultaneously tracing the source back to the corresponding process parameters for adjustment.

[0046] Step Six: Finished Product Packaging and Traceability Code Generation Qualified finished products are conveyed into the finished product silo via a pneumatic conveying system. The intelligent packaging system at the silo's outlet automatically weighs and packages the products according to order requirements (weighing accuracy ±0.1kg), and prints a unique traceability QR code on each bag. The QR code information includes: The source of raw materials for this batch of roasted sand (origin of raw sand, batch number); Records of key parameters for each process stage (calcination temperature curve, residence time, oxygen content, etc.); Quality inspection data (particle size distribution, moisture content, thermal expansion rate, gas generation, etc.); Production time, production line number, and operator information.

[0047] Users can scan a code to view complete production process data, achieving full traceability from raw materials to finished products.

[0048] Step 7: Distributed IoT Collaborative Optimization In each of the above steps, the distributed terminal data is uploaded to the cloud in real time via an industrial IoT platform, with an upload frequency of ≥1Hz and a data transmission latency of ≤100ms. A calcined sand quality prediction model based on machine learning algorithms is deployed in the cloud; the preferred machine learning algorithms are Support Vector Regression (SVR), Random Forest (RF), or Deep Neural Network (DNN).

[0049] This model uses historical production data as training samples to establish a mapping relationship between process parameters (calcination temperature, residence time, oxygen content, feed rate, preheating temperature, etc.) and product quality indicators (high-temperature linear expansion rate, gas evolution, loss on ignition, tensile strength, etc.). The training data sample size is no less than 1000 sets, and the model prediction accuracy (based on high-temperature linear expansion rate ±0.05×10⁻⁶) is [not specified in the original text]. -6 / ℃ (tolerance) ≥90%.

[0050] After real-time production data is input into the model, the system automatically predicts the product quality of the current batch. If the predicted value deviates from the target range, the system automatically optimizes and generates the optimal combination of process parameters, which is then distributed to each edge node for adjustment. The optimization algorithm preferably uses either Genetic Algorithm (GA) or Particle Swarm Optimization (PSO), with an optimization calculation time of ≤5s and a parameter adjustment response time of ≤30s. This forms a closed-loop quality control system encompassing perception, prediction, decision-making, and control.

[0051] It should be noted that the process principle of this invention is based on the basic physicochemical mechanism of high-temperature heat treatment of calcined sand, and combines distributed Internet of Things technology to build an intelligent process control system.

[0052] (I) Basic Physicochemical Principles of Calcined Sand. Silica sand (mainly composed of SiO2) in raw sand undergoes a crystal transformation during high-temperature calcination. Quartz undergoes a transformation from α-quartz to β-quartz around 573℃, accompanied by volume changes; a transformation from β-quartz to β-tridymite occurs around 870℃; and a transformation from β-tridymite to β-cristobalite occurs around 1470℃. By controlling the calcination temperature within a specific range (600~900℃, which can be finely adjusted to 1000℃ for special raw materials), the pores and microcracks inside the silica sand particles can be closed, organic impurities and water of crystallization can be eliminated, and the physicochemical activity of the sand particle surface can be altered, reducing the coefficient of thermal expansion. The three key control parameters for calcined sand quality are: calcination temperature, residence time, and atmospheric conditions. In traditional processes, these three parameters mainly rely on the operator's experience for adjustment, resulting in drawbacks such as strong subjectivity, slow response, and large fluctuations.

[0053] (II) Distributed IoT-based Intelligent Control Principle. This invention deploys distributed intelligent terminals at key nodes in the calcined sand production line. Each intelligent terminal integrates local sensors, edge computing chips, and execution control units, forming an independent perception-computation-execution closed loop. These distributed terminals communicate with each other via industrial IoT protocols and collaborate with the cloud platform to achieve real-time acquisition, edge preprocessing, collaborative optimization, and closed-loop control of process parameters throughout the entire process.

[0054] Specifically, the process control principle of this invention can be summarized into three levels: The first layer is the distributed sensing layer: various types of sensor terminals (temperature sensors, pressure sensors, oxygen content sensors, sand level sensors, vibration sensors, gas composition sensors, etc.) are deployed at each node of the calcined sand production line, including the raw material warehouse, preheating section, calcination section, cooling section, screening section, and finished product warehouse. The total number of sensors is no less than 30, and the process parameters and environmental data of each node are collected in real time.

[0055] The second layer—the edge computing layer—involves setting up an edge computing gateway at each process node to perform real-time preprocessing (including data cleaning, filtering, and normalization), feature extraction, and local decision-making on the collected data. For example, an edge computing terminal is set up at the outlet of the roasting furnace to determine the roasting status in real time based on the temperature change rate and sand particle flowability parameters, and automatically adjust the burner power or feed rate. The data processing latency of the edge computing gateway is ≤10ms, and the local decision-making response time is ≤100ms.

[0056] The third layer—the cloud-based collaboration layer—aggregates data from all distributed endpoints through an IoT platform to a cloud-based intelligent decision-making system. Utilizing big data analytics and AI algorithms, it establishes a predictive model for calcined sand quality, enabling self-optimization of global process parameters and distributing optimization commands to each edge node for execution. The cloud platform supports concurrent connections of ≥1000 terminals and a data processing capacity of ≥10,000 records / second.

[0057] Through the above three-layer collaboration, a fully closed-loop intelligent control system is formed, which integrates perception, calculation, decision-making, execution, and feedback. This fundamentally solves the inherent defects of traditional roasted sand processes, such as information silos, slow response, and large quality fluctuations.

[0058] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A distributed Internet of Things-based intelligent preparation process for calcined sand, characterized in that, Includes the following steps: Distributed intelligent sensing terminals are deployed at each process node of the calcined sand production line to collect process parameters throughout the entire process; Each node's edge computing terminal processes the collected data in real time and executes closed-loop control based on local decisions; The cloud-based collaborative platform aggregates data from all distributed terminals, establishes a quality prediction model, and generates global optimization instructions. The global optimization command is sent to each edge node to adjust the execution parameters, forming a fully closed-loop intelligent control.

2. The distributed Internet of Things based intelligent preparation process of calcined sand according to claim 1, characterized in that, The distributed intelligent sensing terminal includes one or more of the following sensors deployed at at least three different nodes in the raw material warehouse, preheating section, roasting section, cooling section, screening section, and finished product warehouse: temperature sensor, oxygen content sensor, sand level sensor, vibration sensor, gas composition sensor, moisture sensor, particle size analyzer, and impurity detection sensor; each sensor is interconnected through an industrial Internet of Things protocol.

3. The distributed Internet of Things based intelligent preparation process of calcined sand according to claim 1, characterized in that, The edge computing terminal includes intelligent combustion control terminals set in each combustion zone of the calcining furnace. Each intelligent combustion control terminal integrates a gas flow sensor and proportional regulating valve, a combustion air flow sensor and variable frequency fan, an oxygen content sensor and an edge computing chip. The edge computing chip independently adjusts the air-fuel ratio in its area and coordinates with the intelligent combustion control terminals in adjacent combustion zones to control the temperature distribution of the furnace cross section through an Internet of Things bus.

4. The distributed Internet of Things based intelligent preparation process of calcined sand as claimed in claim 1, wherein, A distributed thermocouple array is deployed along the axial and radial directions inside the roasting furnace, while an infrared thermal imaging sensor is set at the observation hole of the roasting furnace. The edge computing terminal fuses the thermocouple array data and infrared thermal imaging data to construct a three-dimensional temperature field digital model inside the furnace, and automatically adjusts the combustion parameters of each combustion zone based on the model.

5. The distributed Internet of Things based intelligent preparation process of calcined sand as claimed in claim 1, wherein, An edge computing terminal integrating temperature sensors, vibration sensors, and sand level sensors is deployed at the discharge port of the roasting furnace. This edge computing terminal establishes a dynamic discrimination model for the roasting endpoint by real-time analysis of the sand particle temperature change rate and flow characteristics, and adaptively adjusts the discharge speed according to the discrimination results to control the residence time of the sand in the roasting zone.

6. The distributed Internet of Things based intelligent preparation process of calcined sand as claimed in claim 1, wherein, The cloud-based collaborative platform is equipped with a calcined sand quality prediction model based on machine learning algorithms. This model takes multi-source process parameters collected in real time as input and predicts online at least one of the following quality indicators of the product: high-temperature linear expansion rate, gas evolution, loss on ignition, and tensile strength. When the predicted value deviates from the target range, the system automatically generates the optimal combination of process parameters through an optimization algorithm and sends it to each edge node for adjustment.

7. The distributed Internet of Things based intelligent preparation process of roasted sand as claimed in claim 1, wherein, Also includes carbon emission real-time monitoring and intelligent optimization steps: deploy CO2, CO, NO x And particulate matter gas composition sensor, real-time collection of carbon emission data through distributed Internet of Things platform, cloud carbon emission calculation model automatically accounts for the carbon footprint of unit product according to fuel consumption, product output and emission concentration, and finds the optimal balance point of energy consumption-carbon emission-production capacity under the premise of ensuring product quality through multi-objective optimization algorithm.

8. The distributed Internet of Things based intelligent preparation process of roasted sand as claimed in claim 1, wherein, It also includes a full-process digital twin and traceability process: a digital twin model of the calcined sand production line is built based on the full-process data collected by the distributed Internet of Things, which maps the operating status of the physical production line in real time in the virtual space, supports the backtracking of historical production processes and virtual debugging of process parameters; and a traceability QR code containing information such as the source of raw materials, key parameters of each process section, quality inspection data, production time and production line number is generated on the finished product packaging.

9. The distributed Internet of Things based intelligent preparation process of roasted sand as claimed in claim 1, wherein, The roasted sand production line includes a raw material pretreatment unit, a preheating unit, a roasting unit, a cooling unit, a screening unit, and a packaging unit connected in sequence. The preheating unit uses the high-temperature exhaust gas emitted from the roasting unit to preheat the raw sand in a gradient manner. The cooling unit adopts a multi-stage air-cooling and water-cooling series structure and sends the recovered hot air into the preheating unit.

Images