Muck recycling production line intelligent monitoring and regulation system based on Internet of Things

Abstract

The invention belongs to the technical field of Internet of Things, and particularly relates to an Internet of Things-based muck recycling production line intelligent monitoring and regulation and control system, which comprises an Internet of Things sensing layer arranged in each process link of a muck recycling production line in a distributed manner and used for collecting material attribute data, equipment operation state data and environmental parameters in real time; the edge calculation layer is in communication connection with the Internet of Things sensing layer and is used for carrying out preprocessing, local logic judgment and real-time equipment control on the collected original data; according to the system, the production efficiency and the intelligent level are improved, real-time and accurate data collection and localized quick response of the whole process (feeding, crushing, screening and forming / aging) of the production line are achieved through cooperation of the Internet of Things sensing layer and the edge calculation layer, and the real-time and accurate data collection and localized quick response of the whole process of the production line are achieved in combination with a process model and big data analysis of a cloud platform. The system can automatically generate and execute a global optimization regulation and control instruction, dynamically adjust production parameters, and reduce manual intervention dependence.

Description

Technical Field

[0001] This invention belongs to the field of Internet of Things (IoT) technology, specifically an intelligent monitoring and control system for waste soil resource utilization production lines based on IoT. Background Technology

[0002] By leveraging IoT technology, a "smart brain" for the resource-based production of construction waste is created, enabling a fundamental shift from experience-driven to data-driven approaches, from passive response to proactive intervention, and from single-machine control to global collaboration. Through intelligent monitoring and adaptive regulation of the entire process and all elements, comprehensive benefits are ultimately achieved, including improved production efficiency, enhanced safety, energy conservation and consumption reduction, and optimal resource utilization.

[0003] Existing technologies disclose several invention patents in the field of Internet of Things (IoT) technology. Among them, patent CN105045239A discloses a construction waste dumping monitoring and control system based on a fully enclosed dump truck. The fully enclosed dump truck is equipped with a construction waste dumping controller. The monitoring and control system includes a host monitoring and control terminal and various vehicle-mounted control terminals installed on each fully enclosed dump truck. Each vehicle-mounted control terminal includes a control box, a control module, and a first power module, an unlocking control button, an electric lock, a first wireless communication module, and a satellite positioning module, all connected to the control module. The corresponding construction waste dumping controller on the fully enclosed dump truck is housed in the control box. The host monitoring and control terminal includes a monitoring and processing terminal and various vehicle-mounted control terminals. The second power module and the second wireless communication module, connected to the monitoring and processing terminal, can ensure the dumping location of the slag and effectively prevent the random dumping of slag. However, this technical solution still has some shortcomings in its application. Existing production lines mostly rely on manual inspection and single, isolated sensors, lacking real-time, multi-dimensional, and distributed perception of the material's own properties and the status of the entire process equipment. The monitoring of visual information such as the condition of the raw material stockpile and the material processing form relies on manual visual inspection, which cannot be quantified and has a slow response. Moreover, it is mostly based on fixed logic or simple local and single-point control. Production parameters are usually preset based on experience and cannot be dynamically adjusted and globally coordinated and optimized according to raw material fluctuations, equipment performance degradation, and environmental changes, resulting in unstable quality and high energy consumption.

[0004] Based on this, the present invention designs an intelligent monitoring and control system for a waste soil recycling production line based on the Internet of Things to solve the above problems. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, this invention proposes an intelligent monitoring and control system for waste soil recycling production lines based on the Internet of Things (IoT). This invention primarily addresses the problems of existing production lines relying heavily on manual inspections and single, isolated sensors, lacking real-time, multi-dimensional, and distributed perception of material properties and the status of equipment throughout the entire process. Monitoring of visual information such as raw material stockpile conditions and material processing methods depends on manual observation, which is neither quantifiable nor responsive. Furthermore, existing systems are often based on fixed logic or simple local, single-point control, with production parameters typically preset based on experience. This lack of dynamic adjustment and global collaborative optimization in response to raw material fluctuations, equipment performance degradation, and environmental changes leads to unstable quality and high energy consumption.

[0006] The technical solution adopted by this invention to solve its technical problem is: an intelligent monitoring and control system for a waste soil resource utilization production line based on the Internet of Things, comprising:

[0007] The Internet of Things (IoT) sensing layer is distributed across various process stages of the waste recycling production line to collect material attribute data, equipment operating status data, and environmental parameters in real time.

[0008] The edge computing layer, which is communicatively connected to the IoT sensing layer, is used for preprocessing the collected raw data, performing local logic judgments, and controlling devices in real time.

[0009] The cloud platform layer is connected to the edge computing layer via a communication network. It is used to receive, store, and analyze data from the edge computing layer, and generate global optimization and control instructions based on the pre-built process model and data analysis results.

[0010] The human-computer interaction layer, connected to the cloud platform layer, is used to visually display full-line monitoring information and analysis reports to users, and to receive advanced control commands from users.

[0011] Preferably, the IoT sensing layer includes:

[0012] The first sensing unit, installed in the raw material stockpile and loading process, includes a moisture sensor, an image recognition device, and a near-infrared spectrometer, used to identify the composition, moisture content, and impurity content of the slag.

[0013] The second sensing unit, installed in the crushing, screening, and sorting stages, includes a vibration sensor, a current sensor, an image recognition device, and an online particle size analyzer, used to monitor equipment load, material particle size distribution, and sorting purity.

[0014] The third sensing unit, which is installed in the molding or aging process, includes a temperature and humidity sensor, a pressure sensor, and a displacement sensor, and is used to monitor molding quality or aging environment parameters.

[0015] Preferably, the IoT sensing layer is also equipped with a video surveillance system. The overall architecture of the video surveillance system is constructed in three layers, including the monitoring front end, the sub-control center, and the central control center.

[0016] The video surveillance system in the raw material storage yard, combined with the AI ​​analysis system, uses cameras to monitor the dumping of slag and soil, the internal condition of the slag and soil, and the status of equipment in real time based on image information.

[0017] Preferably, the edge computing layer includes edge controllers located at each key process node, and the edge controllers are configured to:

[0018] Perform preprocessing operations, including data filtering, outlier removal, and data format standardization;

[0019] Perform local closed-loop control and start / stop, speed adjustment or parameter fine-tuning of associated equipment in real time according to preset thresholds;

[0020] The processed data and local event logs are uploaded to the cloud platform layer.

[0021] Preferably, the cloud platform layer includes:

[0022] The data warehouse module is used to store time-series data, process parameter libraries, equipment files, and historical control records;

[0023] The data analysis and modeling module includes a material ratio optimization model, an equipment energy efficiency analysis model, and a quality prediction model, which are used for big data correlation analysis, process simulation, and fault early warning.

[0024] The intelligent decision-making module is used to automatically generate control strategies for production line material flow, equipment coordination parameters, process formulas, or energy allocation based on the output of the data analysis and model module and in combination with preset optimization objectives.

[0025] Preferably, the optimization objectives of the intelligent decision-making module include at least one of the following: maximizing the final product qualification rate, minimizing unit product energy consumption, and maximizing the overall utilization rate of equipment.

[0026] Preferably, the cloud platform layer also adds anomaly alarm functions, including alarms for abnormal device operation, abnormal information transmission, abnormal model decision-making, and abnormal data.

[0027] The cloud platform layer pushes alarm information to user terminal devices.

[0028] Preferably, the alarm module includes four processes: sensor threshold acquisition, information push channel acquisition, real-time monitoring data acquisition, and alarm decision-making. The alarm module uses the device category, information push channel, and monitoring indicator category as the main identifiers and provides a multi-target alarm method based on the main identifiers. The system pushes abnormal information to the corresponding managers or experts according to the alarm level and device usage permissions.

[0029] Preferably, the intelligent decision-making module is also equipped with an AI analysis system, which uses deep learning technology to analyze the image data of the construction waste, identify the stage of the material in the current full production cycle, and adjust the production management strategy based on the processing degree information to achieve refined management of facility resource-based production.

[0030] Preferably, the intelligent decision-making module is also equipped with an information fusion system, which is used to couple multi-dimensional information into an evaluation strategy, and use multi-sensor data fusion technology to perform fusion analysis on the production line environment. The sensor fusion module fuses and calculates the information collected by multiple sensors and outputs a series of evaluation indicators. The AI ​​analysis system simultaneously analyzes the material processing situation. Within the comprehensive decision-making platform, the system combines the indicator system with the material processing information to trigger a set of environmental control strategies and material operation management strategies.

[0031] Multi-sensor data fusion, analysis, and judgment process and methods:

[0032]

[0033] The magnitude and phase angle of the gradient vector are then:

[0034]

[0035]

[0036] From the above two equations, we can see that: the smoothed image The point of drastic change is the gradient vector. Gradient magnitude in direction By finding the local maxima and applying thresholding to images at various scales, peripheral scratches can be detected.

[0037] The system also includes:

[0038] The feedback execution layer includes a variable frequency feeder for raw materials located upstream of the production line, adjustable process parameter equipment in the middle, and a product diversion device downstream. The feedback execution layer receives and executes control commands from the edge computing layer or the cloud platform layer.

[0039] The beneficial effects of this invention are as follows:

[0040] 1. In this invention, production efficiency and intelligence are improved through the collaboration of the Internet of Things sensing layer and the edge computing layer. Real-time and accurate data collection and localized rapid response are achieved for the entire production line process (feeding, crushing, screening, forming / aging). Combined with the process model and big data analysis of the cloud platform, the system can automatically generate and execute global optimization and control instructions, dynamically adjust production parameters, reduce reliance on manual intervention, and significantly improve the automation and intelligence of the production line.

[0041] 2. In this invention, to ensure and improve product quality consistency, the system achieves closed-loop monitoring and predictive control of raw material characteristics, intermediate product status and final product quality through multi-source sensing (composition, particle size, temperature and humidity, pressure, etc.) and intelligent decision-making modules (AI analysis, quality prediction models). It can dynamically optimize process formulas and parameters based on real-time data to ensure the quality stability and pass rate of final products (such as recycled aggregates, brick blanks, etc.) and reduce batch differences.

[0042] 3. In this invention, the system for optimizing production energy consumption and cost integrates an equipment energy efficiency analysis model, which can monitor equipment load and energy consumption in real time, identify high energy consumption links, and the intelligent decision-making module aims at "minimizing energy consumption per unit product" and can automatically generate and execute the optimal energy allocation and equipment collaborative scheduling strategy, thereby effectively reducing the overall energy consumption and production cost per unit product.

[0043] 4. In this invention, the system reliability and operation and maintenance capabilities are enhanced. The local logic judgment and real-time control functions of the edge computing layer ensure the basic stable operation of key equipment when cloud command delays or network interruptions occur. At the same time, the cloud platform's abnormal alarm function (covering device, data, transmission, and decision anomalies) and multi-level push mechanism can help operation and maintenance personnel quickly locate and respond to faults, shorten downtime, and improve the reliability and maintainability of the entire system.

[0044] 5. In this invention, the human-machine interaction layer and video monitoring system for realizing the visualization and refined management of the production process provide a panoramic visualization interface of the production line, real-time data reports and historical trend analysis. Managers can remotely and intuitively grasp information such as production status, equipment efficiency and raw material consumption. Combined with the AI ​​analysis system's intelligent identification of material stages, the system can support more refined production scheduling and management decisions, improving management efficiency from both macro and micro levels.

[0045] 6. In this invention, accurate perception and fusion decision-making under complex working conditions are achieved by deploying multi-dimensional sensors (images, spectra, vibration, etc.) and using multi-sensor data fusion technology and AI image analysis. The system can overcome the limitations of a single sensor and achieve more comprehensive and accurate integrated perception and evaluation of material properties, equipment status and production environment under working conditions with complex soil composition and variable environment, providing a solid and reliable data foundation for intelligent decision-making.

[0046] 7. In this invention, the flexible and scalable monitoring and control architecture system adopts a layered decoupled architecture of "sensing-edge-cloud-interaction". Each layer has a clear function and a high degree of modularity. This design not only facilitates flexible configuration and deployment according to different production line scales and process requirements, but also provides good scalability for future access to new sensors, algorithm models or execution devices, protecting investment and adapting to technological development. Attached Figure Description

[0047] The invention will now be further described with reference to the accompanying drawings.

[0048] Figure 1 This is a schematic diagram of the system architecture in this invention;

[0049] Figure 2 This is a schematic diagram of the IoT sensing layer architecture in this invention;

[0050] Figure 3 This is a schematic diagram of the cloud platform layer architecture in this invention. Detailed Implementation

[0051] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0052] Example 1: An intelligent monitoring and control system for a waste soil recycling production line based on the Internet of Things and its working method

[0053] This embodiment provides a specific implementation of an intelligent monitoring and control system for a construction waste resource utilization production line based on the Internet of Things. The system is applied to a construction waste resource utilization production line, which mainly includes multiple process links such as raw material stockpile, primary crushing, screening, sorting, molding / aging and finished product stockpiling.

[0054] The specific system deployment in this embodiment is as follows:

[0055] IoT Sensing Layer:

[0056] The first sensing unit is deployed at the raw material stockpile and the feeding conveyor belt. A moisture sensor is embedded in the slag to be fed to measure the moisture content in real time. A near-infrared spectrometer is installed above the feeding conveyor belt to perform online spectral scanning on the passing slag and quickly analyze its composition (such as the proportion of clay, sand, and organic matter) and the approximate content of impurities (such as plastic and wood blocks). A high-definition camera (as an image recognition device) is positioned directly in front of the raw material stockpile and the feeding port to capture macroscopic images of the slag and images of the feeding flow.

[0057] The second sensing unit is deployed at several key equipment locations, including the crusher, vibrating screen, and air classifier. Vibration sensors are installed on the crusher bearing housing to monitor vibration intensity, current sensors are installed in the equipment drive motor circuit to monitor real-time load current, high-definition cameras are positioned directly opposite the screen outlet and the sorting equipment outlet to capture material flow images, and online particle size analyzers (such as laser particle size analyzers) are installed on the material conveyor belt after key screening to analyze the material particle size distribution in real time.

[0058] The third sensing unit is deployed in the brick forming press or aging chamber. Pressure sensors and displacement sensors are deployed in the mold cavity and finished product outlet of the brick forming press to monitor the forming pressure and brick size. A network of temperature and humidity sensors is deployed at multiple points inside the aging chamber to monitor the aging environment.

[0059] In addition, network cameras are installed throughout the plant area (especially the raw material storage yard, production workshop, and finished product area) to form a video surveillance system, and the video streams converge to the plant's branch control center and the group's central control center;

[0060] Edge computing layer:

[0061] Edge controllers (such as industrial gateways or industrial computers) are deployed in the raw material feeding area, crushing area, and forming area respectively. Each edge controller is connected to the sensing unit device in its area via RS-485, Ethernet or IO interface. For example, the edge controller in the crushing area is connected to vibration sensors, current sensors and cameras.

[0062] Cloud platform layer:

[0063] Deployed on enterprise private cloud or industrial internet platform, it connects to edge controllers in each factory area via 4G / 5G or fiber optic network;

[0064] Feedback Execution Layer:

[0065] The frequency converter of the variable frequency feeder (such as the plate feeder) in the raw material stockpile is connected to the edge controller. The hydraulic system regulating valve of the crusher, the excitation force regulator of the vibrating screen, the pressure setting unit of the brick press and other process parameter adjustable equipment are connected to the corresponding edge controller or cloud platform.

[0066] The actuator of the three-way diverter at the end of the finished product conveyor belt (used to distinguish between qualified and unqualified products) is connected to the system;

[0067] Human-computer interaction layer: Deploy a large-screen monitoring system in the factory's central control room, and install client software on the computers, mobile phones / tablets of managers and engineers or access it through a web browser;

[0068] System Workflow Example

[0069] The following example, "Adaptive adjustment of feeding and crushing process based on material characteristics," illustrates the workflow of this system:

[0070] Step 1: Data Acquisition and Edge Preprocessing

[0071] When raw materials are fed, the first sensing unit is activated. The spectral data collected by the near-infrared spectrometer and the moisture content data from the moisture sensor are uploaded to the edge controller of the feeding area. The edge controller performs a moving average filter on the raw spectral data to remove outliers caused by material discontinuity and standardizes the data format into JSON messages. The raw material images captured by the image recognition device are initially compressed and labeled in the edge controller.

[0072] Step 2: Local Rapid Response and Control

[0073] The edge controller has a built-in simple logic judgment module that compares real-time moisture content data with preset thresholds (e.g., crushing efficiency will decrease significantly when moisture content is >20%). If the moisture content is detected to be momentarily exceeded, the edge controller will immediately reduce the frequency of the variable frequency feeder through control commands to reduce the amount of high-moisture material fed in, thus preventing subsequent crusher blockage. This is a local closed-loop control with a response time in the millisecond range.

[0074] Step 3: Data Upload and In-Depth Cloud Analysis

[0075] Meanwhile, the preprocessed component data, moisture content data, processed images, and local event logs (such as "frequency reduction triggered due to high moisture content at a certain time and minute") are packaged and uploaded to the cloud platform layer by the edge controller.

[0076] Step 4: Cloud-based intelligent decision-making and global optimization

[0077] After receiving the data, the cloud platform's data analysis and modeling module starts the material proportioning optimization model and the equipment energy efficiency analysis model.

[0078] Based on the near-infrared analysis of the slag composition (such as "today's raw material has a high clay content and a low sand ratio"), combined with the historical successful formula data stored in the data warehouse module, the model calculates the recommended proportion of auxiliary materials (such as cement and admixtures) that need to be added to achieve the target product strength.

[0079] Meanwhile, the model analyzes the historical relationship between current raw material characteristics and crusher energy consumption, and predicts the optimal spindle speed and feed rate of the crusher under current material characteristics in order to achieve the lowest unit energy consumption.

[0080] Based on the above analysis results, the intelligent decision-making module generates a global optimization and control command: "Adjust the raw material ratio scheme A, set the main shaft speed of the crusher to X rpm, and adjust the stable frequency of the upstream feeder to Y Hz;

[0081] Step 5: Issuance and Execution of Commands

[0082] The optimization command is sent from the cloud platform to the corresponding edge controller, which then parses the command into specific device control signals.

[0083] Send the new mixing ratio to the automatic batching system;

[0084] Adjust the frequency of the crusher drive motor to the corresponding speed X;

[0085] Provided that the moisture content has stabilized, the feeder frequency will be adjusted from the temporary reduced frequency value to the optimized YHz.

[0086] The corresponding devices in the feedback execution layer receive and execute these instructions;

[0087] Step 6: Example of abnormal alarm linkage

[0088] During the crushing process, the vibration sensor of the second sensing unit detected that the vibration value of the crusher bearing continued to rise and exceeded the secondary warning threshold. This data was uploaded by the crushing zone edge controller.

[0089] The alarm module at the cloud platform layer monitors this data in real time, and the alarm decision function triggers "abnormal equipment operation (bearing wear warning)";

[0090] The system pushes alarm information via SMS and APP according to preset rules (equipment category: crusher; monitoring index: vibration; alarm level: level 2), accurately sending it to equipment maintenance engineers and workshop directors, rather than broadcasting it to all employees;

[0091] Meanwhile, the cloud platform's data analysis module correlates current current and production data to perform fault diagnosis and may provide auxiliary suggestions on the management interface, such as "it is recommended to arrange a shutdown inspection after the next batch of production is completed."

[0092] Step 7: Visual and Multi-Information Fusion Analysis

[0093] For the management of raw material stockpiles, the video surveillance system's cameras, combined with the cloud platform's AI analysis system, perform real-time analysis of stockpile images.

[0094] The outline and volume of the waste soil pile are identified by a deep learning model to estimate the stock.

[0095] Identify whether there is any illegal stacking or abnormal stagnation of equipment (such as loaders) in the storage yard;

[0096] By combining the information fusion system, the general material type identified by visual recognition (such as "slag heap with more bricks and tiles"), the composition data of the first sensing unit, and the production plan data are integrated to generate a comprehensive evaluation of the available raw materials in the stockpile, providing production schedulers with decision support for "prioritizing the use of raw materials in a certain area".

[0097] Step 8: Human-Computer Interaction and Monitoring

[0098] Throughout the process, the human-computer interaction layer on the central control room's large screen and the user terminal provides real-time visual display:

[0099] The entire production line process flow diagram, with key data (pressure, temperature, current, output) updated in real time;

[0100] Raw material composition analysis report, equipment health status radar chart. List of triggered alarms and their processing status;

[0101] The system automatically generates daily energy consumption analysis reports and product qualification rate trend reports.

[0102] Senior engineers can use this interface to manually issue advanced control commands such as adjusting model parameters and temporarily modifying control strategies to intervene in or optimize the automatic operation process;

[0103] Specific applications of multi-sensor data fusion analysis

[0104] When monitoring the surface quality of finished brick blanks (such as detecting peripheral scratch defects), the system uses data from the third sensing unit (pressure and displacement sensors) in the forming process, and combines it with image data from a high-definition industrial camera installed at the brick outlet to make a fusion judgment.

[0105] Image data is fed into a pre-trained AI analysis system (convolutional neural network model) for preliminary defect region localization.

[0106] At the same time, the system calls the multi-sensor data fusion algorithm in the information fusion system to perform multi-scale analysis such as wavelet transform on the high-resolution image. The specific process is as follows:

[0107] Let the acquired image of the brick blank surface be... After two-dimensional smoothing function After smoothing, At this scale, its wavelet transform components in two directions and The calculation conforms to the method described above:

[0108]

[0109] Then the magnitude of the gradient vector is calculated. and phase angle ;

[0110]

[0111]

[0112] In phase angle Find the gradient magnitude in the direction By thresholding images at various scales, the local maxima of the brick blank can be accurately detected, revealing the fine peripheral scratch contours and depth information on the brick blank surface.

[0113] Ultimately, the intelligent decision-making module combines the defect categories identified by AI with the defect severity quantified by the multi-sensor fusion algorithm to form a comprehensive quality evaluation. If the defect is determined to be unacceptable, an instruction is generated to control the downstream product diversion device to remove the brick blank to the scrap line, and parameters such as molding pressure (feedback execution layer) are adjusted in reverse.

[0114] Through the above embodiments, this system has achieved closed-loop intelligent control from perception, analysis, decision-making to execution, significantly improving the automation level, product quality, resource utilization rate and operation management efficiency of the waste soil recycling production line.

[0115] Example 2

[0116] Video surveillance and AI analysis subsystem:

[0117] Monitoring front end: Deploy AI-enabled network cameras (IPCs) at key workstations in the raw material yard and various workshops.

[0118] Sub-control centers: located in the central control rooms of each workshop, responsible for the storage and preliminary analysis of video streams in their respective areas;

[0119] Central Control Center: Located on the cloud platform, it aggregates all video streams;

[0120] AI analytics applications (for example, in a raw material stockpile):

[0121] Construction waste dumping compliance monitoring: Through real-time camera footage, AI models (such as the YOLO algorithm) automatically identify whether the construction waste dumping exceeds the designated area and whether the dumping height exceeds the standard, and generate alarms.

[0122] Internal condition prediction: By combining multiple days of video and meteorological data, AI can help predict the risk of seepage and compaction inside the reactor body;

[0123] Equipment status auxiliary monitoring: Identify whether the loading machinery (such as excavators) is in working condition and whether the personnel are safely on duty;

[0124] All sensing unit data is uploaded to the edge computing layer via industrial Ethernet or hybrid networks such as 5G / LoRa;

[0125] The edge computing layer deploys industrial edge computing gateways (edge ​​controllers) at key process nodes such as crushing workshops, screening workshops, and forming workshops.

[0126] Edge controller core functions:

[0127] Data preprocessing: The raw sensor data is filtered by moving average (to remove noise), obvious outliers are removed using the 3σ criterion, and all data is converted into a unified JSON format and tagged with timestamps and device IDs.

[0128] Local logic judgment and closed-loop control:

[0129] Example 1 (Crusher Protection): Analyze vibration sensor data in real time. If the vibration amplitude exceeds the safety threshold for 5 consecutive seconds, immediately send an emergency stop command to the crusher PLC and upload the "Vibration Exceeds Limit - Emergency Stop" event log to the cloud platform.

[0130] Example 2 (Feeding Flow Control): Receives the feeding rate setpoint from the cloud platform and adjusts the frequency of the variable frequency feeder in the raw material silo in real time through a PID algorithm to achieve precise quantitative feeding;

[0131] Example 3 (Screening Efficiency Optimization): Based on the real-time feedback from the particle size analyzer, fine-tune the frequency and amplitude of the vibrating screen (achieved by adjusting the frequency converter) to keep the particle size distribution close to the set range.

[0132] Data Upload: Preprocessed high-quality data packets, device status snapshots, and local event logs are uploaded to the cloud platform layer periodically / in real-time via the MQTT protocol;

[0133] The cloud platform layer, deployed on public or private cloud servers, adopts a microservice architecture and mainly includes the following modules:

[0134] Data warehouse module: Uses a time-series database (such as InfluxDB) to store all sensor time-series data; uses a relational database (such as MySQL) to store process parameter formulas, equipment files (model, maintenance records), and historical control records;

[0135] Data Analysis and Modeling Module: Material Proportioning Optimization Model: Based on raw material composition (NIR data), moisture content, and target product (e.g., MU15 standard brick) requirements, combined with a cost database, a linear programming algorithm is used to recommend the optimal raw material proportions and admixture (e.g., cement) addition amount daily.

[0136] Equipment energy efficiency analysis model: Correlate and analyze equipment current, output, and operating time data to calculate the "energy consumption per ton of material" for each piece of equipment, identify inefficient operating periods (such as no-load and light-load), and generate energy-saving reports;

[0137] Quality prediction model: Using molding pressure, curing temperature and humidity, and raw material composition as inputs, a machine learning model (such as random forest) is trained to predict the 28-day compressive strength of finished bricks, thereby achieving quality feedforward control;

[0138] Intelligent decision-making module:

[0139] Optimize target settings: Users can set "maximizing product qualification rate" or "minimizing unit product energy consumption" as the primary target in the human-computer interaction layer;

[0140] Decision-making process: The module calls the above model for analysis. For example, if the goal is to minimize energy consumption, the module may generate a set of control strategies such as "appropriately reducing the main shaft speed of the crusher and optimizing the start-up and shutdown time of the screening machine while ensuring that the particle size is qualified."

[0141] AI Image Deep Analysis: Deep learning (CNN network) is applied to the images of construction waste uploaded by the first sensing unit to not only identify impurities, but also to determine the "processing stage" of the material (such as "fresh construction waste", "preliminary weathering", "containing high plastic clay" etc.), providing more refined material characteristic input for the proportioning model;

[0142] Information fusion and comprehensive evaluation:

[0143] Multi-sensor fusion analysis: For "production line environment" assessment, data from multiple sensors such as vibration, noise, dust, temperature, and humidity are fused. A wavelet transform fusion algorithm (as described in the formula) is employed to enhance the detection capability of abnormal features (such as scratching noises from equipment). The calculated local maxima of the gradient modulus can be used to accurately identify abnormal vibrations or image edges (scratches).

[0144] Generate evaluation indicators: Output a series of indicators such as "Comprehensive Environmental Index" and "Equipment Health Index";

[0145] Comprehensive Decision Making: The AI ​​analysis system provides information on the material processing stage in real time. Within the comprehensive decision making platform, the "environmental comprehensive index" is combined with the information that "the current material is high-moisture viscous slag" to trigger a set of strategies: 1. Activate environmental control strategies (such as increasing the ventilation volume in the aging workshop), 2. Activate operation management strategies (such as suggesting to increase the drying time before crushing and automatically scheduling the turning machine).

[0146] Human-computer interaction layer

[0147] Format: Web-based large-screen visualization system + mobile APP (WeChat mini-program / dedicated APP);

[0148] Function:

[0149] Panoramic monitoring screen: Dynamically displays the status of all equipment (color-coded), material flow, and key parameters (such as current moisture content and median particle size) in the form of GIS map and 3D production line simulation.

[0150] Analysis reports: Automatically generate daily, weekly, and monthly reports, including KPI charts such as output, energy consumption, pass rate, and OEE (Overall Equipment Effectiveness);

[0151] Command issuance: In "Expert Mode", process engineers can modify cloud platform model parameters or directly issue advanced commands to the feedback execution layer, such as manually setting new ingredient formulas.

[0152] Feedback Execution Layer

[0153] This layer is the system's "hands and feet," receiving and executing instructions from the edge layer (for rapid response) or the cloud platform (for optimization strategies).

[0154] Upstream: Variable frequency feeder in the raw material silo receives instructions to adjust the feeding rate;

[0155] Midstream: Equipment with adjustable process parameters, such as frequency converters for crushers, solenoid valves for adding water to mixers, and pressure servo controllers for brick presses;

[0156] Downstream: Product diversion devices, such as vision-based robotic sorting arms or pneumatic push rods, automatically remove unqualified products identified by AI to the rework silo.

[0157] Example of system workflow (with the goal of "improving product qualification rate")

[0158] Sensing: The first sensing unit detected that the moisture content of the newly arrived slag was as high as 18% (exceeding the optimal 12%), and the NIR showed that the clay content was too high;

[0159] Edge processing: The edge controller slightly reduces the feed rate of the feeder and uploads the data;

[0160] Cloud Platform Analysis and Decision-Making:

[0161] The quality prediction model indicates that direct production may result in insufficient brick forming strength.

[0162] The intelligent decision-making module combines AI image recognition (the material is "high-viscosity wet slag") and the "high ambient humidity" index from multi-sensor fusion to activate the strategy set;

[0163] Strategy A: Issue a command to the edge layer, requiring an additional "pre-drying and stirring" process after the crushing stage (controlling the start of the hot air blower);

[0164] Strategy B: Adjust the subsequent batching model and fine-tune the cement dosage to compensate;

[0165] Strategy C: Feedback the instructions to the aging workshop to issue to the equipment, extend the maintenance cycle and adjust the humidity setting;

[0166] Execution and feedback: As each actuator takes action, new production process data is collected by the perception layer, forming a closed loop. The system continuously compares finished product inspection data with predicted data to optimize model parameters.

[0167] Through the above embodiments, the present invention system realizes the digitalization, networking, and intelligentization of waste soil resource production, significantly improving resource utilization, product quality stability, and production management efficiency.

[0168] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. An intelligent monitoring and control system for a waste soil recycling production line based on the Internet of Things, characterized in that, include: The Internet of Things (IoT) sensing layer is distributed across various process stages of the waste recycling production line to collect material attribute data, equipment operating status data, and environmental parameters in real time. The edge computing layer, which is communicatively connected to the IoT sensing layer, is used for preprocessing the collected raw data, performing local logic judgments, and controlling devices in real time. The cloud platform layer is connected to the edge computing layer via a communication network. It is used to receive, store, and analyze data from the edge computing layer, and generate global optimization and control instructions based on the pre-built process model and data analysis results. The human-computer interaction layer, connected to the cloud platform layer, is used to visually display full-line monitoring information and analysis reports to users, and to receive advanced control commands from users.

2. The intelligent monitoring and control system for the waste soil resource utilization production line based on the Internet of Things as described in claim 1, characterized in that: The IoT sensing layer includes: The first sensing unit, installed in the raw material stockpile and loading process, includes a moisture sensor, an image recognition device, and a near-infrared spectrometer, used to identify the composition, moisture content, and impurity content of the slag. The second sensing unit, installed in the crushing, screening, and sorting stages, includes a vibration sensor, a current sensor, an image recognition device, and an online particle size analyzer, used to monitor equipment load, material particle size distribution, and sorting purity. The third sensing unit, which is installed in the molding or aging process, includes a temperature and humidity sensor, a pressure sensor, and a displacement sensor, and is used to monitor molding quality or aging environment parameters.

3. The intelligent monitoring and control system for the waste soil resource utilization production line based on the Internet of Things as described in claim 1 or 2, characterized in that: The IoT sensing layer is also equipped with a video surveillance system. The overall architecture of the video surveillance system is built in three layers, including the monitoring front end, the sub-control center and the central control center. The video surveillance system in the raw material storage yard, combined with the AI ​​analysis system, uses cameras to monitor the dumping of slag and soil, the internal condition of the slag and soil, and the status of equipment in real time based on image information.

4. The intelligent monitoring and control system for the waste soil resource utilization production line based on the Internet of Things as described in claim 3, characterized in that: The edge computing layer includes edge controllers located at each key process node, and the edge controllers are configured to: Perform preprocessing operations, including data filtering, outlier removal, and data format standardization; Perform local closed-loop control and start / stop, speed adjustment or parameter fine-tuning of associated equipment in real time according to preset thresholds; The processed data and local event logs are uploaded to the cloud platform layer.

5. The intelligent monitoring and control system for the waste soil resource utilization production line based on the Internet of Things as described in claim 4, characterized in that: The cloud platform layer includes: The data warehouse module is used to store time-series data, process parameter libraries, equipment files, and historical control records; The data analysis and modeling module includes a material ratio optimization model, an equipment energy efficiency analysis model, and a quality prediction model, which are used for big data correlation analysis, process simulation, and fault early warning. The intelligent decision-making module is used to automatically generate control strategies for production line material flow, equipment coordination parameters, process formulas, or energy allocation based on the output of the data analysis and model module and in combination with preset optimization objectives.

6. The intelligent monitoring and control system for the waste soil resource utilization production line based on the Internet of Things as described in claim 5, characterized in that: The optimization objectives of the intelligent decision-making module include at least one of the following: maximizing the final product qualification rate, minimizing unit product energy consumption, and maximizing the overall utilization rate of equipment.

7. The intelligent monitoring and control system for the waste soil resource utilization production line based on the Internet of Things as described in claim 6, characterized in that: The cloud platform layer also adds anomaly alarm functions, including alarms for abnormal device operation, abnormal information transmission, abnormal model decision-making, and abnormal data. The cloud platform layer pushes alarm information to user terminal devices.

8. The intelligent monitoring and control system for waste soil resource utilization production line based on the Internet of Things as described in claim 7, characterized in that: The alarm module includes four processes: sensor threshold acquisition, information push channel acquisition, real-time monitoring data acquisition, and alarm decision-making. The alarm module uses device category, information push channel, and monitoring indicator category as the main identifiers and provides a multi-target alarm method based on the main identifiers. The system pushes abnormal information to the corresponding managers or experts according to the alarm level and device usage permissions.

9. The intelligent monitoring and control system for the waste soil resource utilization production line based on the Internet of Things as described in claim 5, characterized in that: The intelligent decision-making module is also equipped with an AI analysis system, which uses deep learning technology to analyze slag image data, identify the stage of the material in the current full production cycle, and adjust production management strategies based on the processing degree information to achieve refined management of facility resource-based production.

10. The intelligent monitoring and control system for the waste soil resource utilization production line based on the Internet of Things as described in claim 5 or 9, characterized in that: The intelligent decision-making module is also equipped with an information fusion system, which is used to couple multi-dimensional information into an evaluation strategy. It uses multi-sensor data fusion technology to perform fusion analysis on the production line environment. The sensor fusion module fuses and calculates the information collected by multiple sensors and outputs a series of evaluation indicators. The AI ​​analysis system simultaneously analyzes the material processing situation. Within the comprehensive decision-making platform, the system combines the indicator system with the material processing information to trigger a set of environmental control strategies and material operation management strategies. Multi-sensor data fusion, analysis, and judgment process and methods: ； The magnitude and phase angle of the gradient vector are then: ； ； From the above two equations, we can see that: the smoothed image The point of drastic change is the gradient vector. Gradient magnitude in direction By finding the local maxima and applying thresholding to images at various scales, peripheral scratches can be detected. The system also includes: The feedback execution layer includes a variable frequency feeder for raw materials located upstream of the production line, adjustable process parameter equipment in the middle, and a product diversion device downstream. The feedback execution layer receives and executes control commands from the edge computing layer or the cloud platform layer.

Images