Glass fiber chopped strand total waste heat drying control system based on AI

Abstract

The invention relates to the technical field of glass fiber chopped strand production and processing, in particular to an AI-based glass fiber chopped strand total waste heat drying control system, which comprises a big data AI intelligent sub-module based on an intelligent SCADA system by adopting a PLC as a basis, adjusting the waste heat supply amount through an electric adjusting valve, and controlling the total waste heat supply amount through the electric adjusting valve. The operation state data is synchronized to the intelligent SCADA system for AI analysis; a temperature measuring sensor, a flow sensor and a pressure sensor are used for collecting multi-dimensional data, and the multi-dimensional data are transmitted to a PLC and an intelligent SCADA system big data storage server in real time; the data of the load and the operation state are monitored and AI analyzed in real time through the intelligent SCADA system; and the data is synchronized to the intelligent SCADA system to provide core data input for an AI energy efficiency optimization algorithm. The problems that in the glass fiber chopped strand drying process, the waste heat utilization rate is low, the temperature control precision is poor, system response lags behind, energy consumption is high, and intelligent early warning and active regulation and control capacity is lacked are solved.

Description

Technical Field

[0001] This invention belongs to the field of glass fiber chopped strand production and processing technology, specifically involving an AI-based glass fiber chopped strand full waste heat drying control system. Background Technology

[0002] The utilization of waste heat during the drying process of chopped glass fibers faces numerous challenges. Waste heat sources are highly susceptible to external factors; for example, weather changes can cause fluctuations in ambient temperature, which in turn affects the temperature of the waste heat. Fluctuations in the operating status of other waste heat-using equipment can also interfere with the supply of waste heat, resulting in significant fluctuations in temperature and flow rate over a day or even a year. These unstable factors severely restrict the utilization rate and effectiveness of waste heat, making it difficult for it to fully play its role in the drying process. For a long time, in order to ensure stable drying temperature, the industry has either used independent natural gas combustion units (which have high energy consumption and large carbon emissions) or relied on auxiliary heat sources to compensate for waste heat fluctuations (which cannot achieve true full utilization of waste heat). Even if some solutions use PLC control, they are still limited to passive parameter feedback adjustment - lacking the ability to predict temperature fluctuations and unable to intervene in advance based on historical operating data and environmental change patterns (such as seasonal temperature differences and diurnal temperature fluctuations). This results in a delayed response of the control system when waste heat fluctuates, making it difficult to maintain stable drying temperature, which is incompatible with the requirements of deep energy conservation, environmental protection and high-precision production.

[0003] Furthermore, while PLC control possesses basic adjustment, interlocking, and data acquisition functions, it lacks integration with big data analysis and AI learning. This prevents it from uncovering potential patterns in temperature fluctuations and optimizing control logic. In scenarios with full waste heat supply, there is still room for improvement in temperature control accuracy and system adaptability. Traditional SCADA systems, on the other hand, lack mature AI integration solutions, making it difficult to meet in-depth management needs such as intelligent early warning and energy efficiency optimization. The emergence of SCADA system software platforms integrating artificial intelligence provides a full-stack solution for AI empowerment in industrial scenarios.

[0004] Existing technologies suffer from drawbacks such as unstable waste heat utilization, low temperature control accuracy, slow system response, high energy consumption, and a lack of intelligent predictive capabilities. Traditional control methods struggle to cope with waste heat fluctuations, resulting in poor drying quality, high energy consumption, and a lack of AI integration for in-depth data analysis and proactive regulation. Summary of the Invention

[0005] To address the aforementioned issues, this invention provides an AI-based waste heat drying control system for chopped glass fibers. This system solves problems such as low waste heat utilization, poor temperature control accuracy, sluggish system response, high energy consumption, and lack of intelligent early warning and proactive control capabilities during the chopped glass fiber drying process. To achieve the above objectives, this invention adopts the following technical solution: The AI-based glass fiber chopped strand waste heat drying control system includes: a control core module, which uses a PLC as the foundation and links with a big data AI intelligent sub-module based on an intelligent SCADA system to form a dual control capability of passive adjustment and active prediction, constructing a full-stack AI control scheme; an energy supply module, which uses a waste heat supply device connected in parallel with a waste heat intake pipe, and regulates the waste heat supply through an electric regulating valve, synchronizing the operating status data to the intelligent SCADA system for AI analysis; a sensing module, which uses temperature, flow, and pressure sensors to collect multi-dimensional data and transmits it in real time to the PLC and the intelligent SCADA system's big data storage server; an execution module, which uses an electric regulating valve, hot air blower, exhaust fan, and waste heat air venting device to receive PLC control commands and execute actions, and monitors and analyzes load and operating status data in real time through the intelligent SCADA system; and a protection module, which uses a UPS to ensure normal operation and data security during power outages, and uses an electricity meter and flow meter to monitor power consumption and waste heat flow, synchronizing the data to the intelligent SCADA system to provide core data input for the AI ​​energy efficiency optimization algorithm.

[0006] Furthermore, the core control module includes: a data acquisition submodule, which uses a PLC as the basic control core and links with a big data AI intelligent submodule based on an intelligent SCADA system to collect ambient temperature, furnace temperature, waste heat flow, and pressure data in real time through sensor modules; a temperature fluctuation submodule, which transmits the collected data to a big data storage server, where the AI ​​learning unit extracts historical operating patterns to obtain temperature fluctuation prediction results; and a control scheme submodule, which generates adjustment commands through intelligent algorithms, and the PLC actively adjusts the opening degree of the electric valve and the frequency of the fan based on the prediction results, forming a dual control capability of passive adjustment and active prediction, and constructing a full-stack AI control scheme.

[0007] Furthermore, the energy supply module includes: a waste heat transmission submodule, used to transport waste heat through a parallel layout of waste heat intake pipes using a full waste heat supply device, without connecting any auxiliary heat source; a data transmission submodule, used to precisely adjust the waste heat supply through an electric regulating valve to ensure that the drying process relies entirely on waste heat, and to extract and transmit the operating status data to the intelligent SCADA system in real time; and a data analysis submodule, used by the AI ​​learning unit to obtain pipe temperature, flow rate, and valve opening parameters, and to obtain the waste heat supply efficiency and fluctuation characteristics through data analysis.

[0008] Furthermore, the sensing module includes: a pressure information submodule, used to simultaneously collect ambient temperature and furnace temperature data using a temperature sensor, monitor waste heat flow using a flow sensor, and obtain pipeline and furnace pressure information using a pressure sensor; a storage service submodule, used to extract multi-dimensional data in real time and send it to the PLC for basic control via a data transmission line, while simultaneously uploading the raw data stream to the intelligent SCADA system's big data storage server; and an anomaly warning submodule, used by the intelligent SCADA system to acquire full operational data and obtain temperature fluctuation characteristics, flow change trends, and pressure anomaly warnings through built-in algorithms.

[0009] Furthermore, the execution module includes: an instruction execution submodule, used to adjust the waste heat supply using an electric regulating valve, drive air circulation using a hot air blower, discharge exhaust fan, and exhaust waste heat venting device to vent waste heat in case of abnormality; all four receive control instructions sent by the PLC to execute actions; a fault early warning submodule, used to extract equipment load data and operating status parameters in real time through the intelligent SCADA system, and the AI ​​learning unit analyzes the data stream to obtain equipment efficiency evaluation results and potential fault early warnings; and a result feedback submodule, used by the intelligent SCADA system to feed back the analysis results to the PLC, forming a closed-loop management of equipment status monitoring, AI analysis, and control optimization.

[0010] Furthermore, the protection module includes: a secure storage submodule, used to ensure the normal operation of the control module, sensing module, and AI subsystem during power outages by using a UPS uninterruptible power supply and ensuring secure data storage; a synchronous transmission submodule, used to continuously monitor power consumption data through an energy meter and synchronously acquire waste heat flow information through a flow meter, extracting power consumption data and waste heat flow information in real time and transmitting them to the intelligent SCADA system; and an AI optimization submodule, used for AI energy efficiency optimization algorithms to acquire power consumption data and waste heat flow information, and analyze the energy consumption distribution characteristics and waste heat utilization efficiency.

[0011] Furthermore, the big data AI intelligent submodule based on the intelligent SCADA system includes: a temperature fluctuation prediction subunit, which uses the intelligent SCADA system's big data storage server to store historical operating data in real time, extracts temperature fluctuation patterns through the LSTM algorithm of the AI ​​learning unit, and drives the PLC to adjust the electric valve in advance based on the prediction results, thus achieving a pre-prediction effect; a temperature curve adjustment subunit, which uses the sensor module to generate ambient and furnace temperature curves, trains a correlation model through the intelligent SCADA system's AI learning unit, and outputs correction commands, which drive the PLC to adjust in a coordinated manner, thus achieving dynamic temperature compensation and energy consumption reduction; and an AI learning optimization subunit, which uses the intelligent SCADA system's AI learning unit to periodically update the model and algorithm based on new data, and optimizes the adjustment parameters by supporting user-embedded custom models and constructing decision tree models, thereby improving control performance.

[0012] Furthermore, the big data AI intelligent submodule based on the intelligent SCADA system also includes: an intelligent alarm management subunit, which is used to automatically adjust the alarm threshold using the dynamic threshold optimization algorithm of the intelligent SCADA system, locate the root cause of the fault through multi-parameter joint alarm and cluster analysis, and combine early warning technology to obtain accurate alarms and handling suggestions.

[0013] Furthermore, the big data AI intelligent submodule based on the intelligent SCADA system also includes a health assessment subunit, which uses the intelligent SCADA system's AI subsystem to combine equipment sensor data and deep learning algorithms to assess the equipment's health status and predict the probability of failure, identify potential problems in advance and provide maintenance suggestions, thereby reducing the equipment failure rate.

[0014] In the technical solution provided by this invention, the control core module is used to form a dual control capability of passive adjustment and active prediction based on a PLC and linked with a big data AI intelligent sub-module based on an intelligent SCADA system, thus constructing a full-stack AI control solution; the energy supply module is used to use a full waste heat supply device connected in parallel with a full waste heat intake pipeline, and to regulate the waste heat supply through an electric regulating valve, with the operating status data synchronized to the intelligent SCADA system for AI analysis; the sensing module is used to collect multi-dimensional data using temperature, flow, and pressure sensors, and transmit it in real time to the PLC and the big data storage server of the intelligent SCADA system; the execution module is used to receive control commands from the PLC and execute actions for the electric regulating valve, hot air blower, exhaust fan, and waste heat air venting device, and to monitor and analyze the load and operating status data in real time through the intelligent SCADA system; the protection module is used to use a UPS to ensure normal operation and data security during power outages, and to use an energy meter and flow meter to monitor power consumption and waste heat flow, with the data synchronized to the intelligent SCADA system to provide core data input for the AI ​​energy efficiency optimization algorithm. This invention solves the problems of low waste heat utilization, poor temperature control accuracy, slow system response, high energy consumption, and lack of intelligent early warning and active control capabilities during the drying process of chopped glass fibers. Attached Figure Description

[0015] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention.

[0016] Figure 1 This is a schematic diagram of the process flow of an AI-based glass fiber chopped strand drying control system according to an embodiment of the present invention.

[0017] Figure 2 This is a schematic diagram of the control flow of an AI-based glass fiber chopped strand full waste heat drying control system in an embodiment of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0019] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this specification means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0020] An AI-based control system for drying chopped glass fiber using all residual heat, such as... Figure 1-2 As shown, it includes: a control core module, which uses a PLC as the foundation and links with a big data AI intelligent sub-module based on the intelligent SCADA system to form a dual control capability of passive adjustment and active prediction, constructing a full-stack AI control solution; an energy supply module, which uses a full waste heat supply device connected in parallel with a full waste heat intake pipeline, and regulates the waste heat supply through an electric regulating valve, synchronizing the operating status data to the intelligent SCADA system for AI analysis; a sensing module, which uses temperature, flow, and pressure sensors to collect multi-dimensional data and transmits it in real time to the PLC and the big data storage server of the intelligent SCADA system; an execution module, which uses electric regulating valves, hot air blowers, exhaust fans, and waste heat air venting devices to receive PLC control commands and execute actions, and monitors and analyzes load and operating status data in real time through the intelligent SCADA system; and a protection module, which uses a UPS to ensure normal operation and data security during power outages, and uses electricity meters and flow meters to monitor power consumption and waste heat flow, synchronizing the data to the intelligent SCADA system to provide core data input for AI energy efficiency optimization algorithms.

[0021] By leveraging the LSTM time series prediction algorithm and dynamic threshold optimization capabilities of the intelligent SCADA system, the system can predict fluctuations over time periods and adjust them in real-time. In a scenario with no auxiliary heat source and full waste heat, the drying temperature control accuracy is maintained stably at ±0.5℃. The number of temperature overshoots is reduced by 60% compared to traditional PLC control, avoiding product quality defects caused by temperature fluctuations (such as uneven strength of chopped shreds and surface condensation). The product qualification rate is increased to over 99.5%.

[0022] By relying on the energy efficiency optimization algorithm of the intelligent SCADA system to build an energy consumption model for the drying system, 100% waste heat utilization is achieved, and auxiliary heat sources and natural gas combustion units are completely eliminated. This can reduce natural gas consumption by about 5,000 m³ per year (calculated based on a single unit), reduce energy costs by 30%-40%, and further reduce energy consumption by more than 8.3% compared to traditional solutions. At the same time, it reduces carbon dioxide emissions by about 11 tons per year, resulting in significant social benefits.

[0023] In this embodiment, the data acquisition submodule uses a PLC as the basic control core and links with the big data AI intelligent submodule based on the intelligent SCADA system to collect ambient temperature, furnace temperature, waste heat flow, and pressure data in real time through the sensor module; the temperature fluctuation submodule transmits the collected data to the big data storage server, and the AI ​​learning unit extracts historical operating patterns to obtain temperature fluctuation prediction results; the control scheme submodule generates adjustment commands through intelligent algorithms, and the PLC actively adjusts the opening degree of the electric valve and the frequency of the fan based on the prediction results, forming a dual control capability of passive adjustment and active prediction, and constructing a full-stack AI control scheme.

[0024] Based on a PLC, and linked with a big data AI intelligent subsystem (including a big data storage server and AI learning units) based on an intelligent SCADA system, a dual control capability of "passive adjustment + active prediction" is formed. The intelligent SCADA system provides a built-in AI algorithm development environment, supporting the entire process of data preprocessing (normalization, feature engineering), model training (supervised / unsupervised learning), and inference deployment, while also being compatible with importing models from mainstream AI frameworks such as TensorFlow and PyTorch. The PLC receives data from the sensor modules and prediction commands from the AI ​​subsystem, sending control signals to the actuators. High-precision temperature controllers display the furnace temperature in real time, assisting in monitoring control accuracy. This control core is deeply compatible with CPUs (Kunpeng, Phytium), operating systems (Tongxin UOS, Galaxy Kylin), and databases (DM, Kingbase), constructing a full-stack AI control solution.

[0025] In this embodiment, the waste heat delivery submodule is used to deliver waste heat through a parallel layout of waste heat intake pipes using a full waste heat supply device, without connecting any auxiliary heat source; the data transmission submodule is used to precisely adjust the waste heat supply through an electric regulating valve to ensure that the drying process relies entirely on waste heat, and the operating status data is extracted and transmitted to the intelligent SCADA system in real time; the data analysis submodule is used by the AI ​​learning unit to obtain pipe temperature, flow rate and valve opening parameters, and to obtain the waste heat supply efficiency and fluctuation characteristics through data analysis.

[0026] As a complete waste heat supply device, it adopts multiple parallel waste heat intake pipes (without any auxiliary heat source input), and adjusts the waste heat supply through electric regulating valves to ensure that the drying process relies entirely on waste heat. Its operating status data is synchronized to the intelligent SCADA system in real time for AI analysis.

[0027] In this embodiment, the pressure information submodule is used to simultaneously collect ambient temperature and furnace temperature data using a temperature sensor, monitor waste heat flow using a flow sensor, and obtain pipeline and furnace pressure information using a pressure sensor; the storage service submodule is used to extract multi-dimensional data in real time and send it to the PLC for basic control via a data transmission line, while simultaneously uploading the raw data stream to the big data storage server of the intelligent SCADA system; the anomaly warning submodule is used by the intelligent SCADA system to obtain full operating data and obtain temperature fluctuation characteristics, flow change trends, and pressure anomaly warnings through built-in algorithms.

[0028] Temperature sensors (simultaneously collecting ambient temperature and furnace temperature, with sampling frequencies of 1 minute / collection and 1 collection / 30 seconds respectively), flow sensors (monitoring waste heat flow), and pressure sensors (monitoring pipeline and furnace pressure) are used to transmit the collected data to the PLC and intelligent SCADA system big data storage server in real time, providing multi-dimensional data support for AI algorithms. The data acquisition efficiency is more than 10 times higher than that of traditional solutions.

[0029] In this embodiment, the instruction execution submodule is used to regulate the waste heat supply using an electric regulating valve, the hot air fan drives air circulation, the exhaust fan is responsible for exhausting the exhaust gas, and the waste heat exhaust device empties the waste heat in case of abnormality. All four receive control instructions sent by the PLC and execute actions. The fault early warning submodule is used to extract equipment load data and operating status parameters in real time through the intelligent SCADA system. After the AI ​​learning unit obtains the data stream, it performs analysis to obtain equipment efficiency evaluation results and potential fault early warnings. The result feedback submodule is used for the intelligent SCADA system to feed back the analysis results to the PLC, forming a closed-loop management of equipment status monitoring, AI analysis, and control optimization.

[0030] Electric regulating valve (regulating waste heat supply), hot air blower (air circulation), exhaust fan (exhaust gas emission), and waste heat exhaust device (exhausting waste heat in case of failure / maintenance / predicted abnormality) receive PLC control commands to execute actions. Their load, operating status and other data are monitored and analyzed in real time by the intelligent SCADA system and AI.

[0031] In this embodiment, the secure storage submodule is used to ensure the normal operation of the control module, sensing module, and AI subsystem during power outages by using a UPS uninterruptible power supply, thus ensuring secure data storage; the synchronous transmission submodule is used to continuously monitor power consumption data through the energy meter and synchronously acquire waste heat flow information through the flow meter, extracting the power consumption data and waste heat flow information in real time and transmitting them to the intelligent SCADA system; the AI ​​optimization submodule is used to obtain power consumption data and waste heat flow information through AI energy efficiency optimization algorithms, and analyze the energy consumption distribution characteristics and waste heat utilization efficiency.

[0032] The UPS ensures the normal operation and data security of the control module, sensing module, and AI subsystem during power outages; the energy meter and flow meter monitor power consumption and waste heat flow respectively, and the data is synchronously transmitted to the intelligent SCADA system to provide core data input for the AI ​​energy efficiency optimization algorithm.

[0033] During a power outage, the UPS ensures the operation of the PLC and the AI ​​subsystem of the intelligent SCADA system. The PLC automatically shuts off the electric regulating valve and hot air blower to prevent abnormal equipment damage. The intelligent SCADA system synchronously stores the operating data before the power outage to ensure data security.

[0034] When the hot air blower malfunctions, the PLC immediately shuts off the waste heat electric regulating valve and starts the waste heat exhaust device to remove residual waste heat from the oven. The intelligent SCADA system and AI subsystem predict abnormal loads of the hot air blower based on historical current data, provide early maintenance reminders, and reduce the failure rate.

[0035] It supports seamless switching between manual and automatic modes. During the switching process, the PLC, in conjunction with the AI ​​subsystem of the intelligent SCADA system, provides real-time correction instructions to ensure that the temperature fluctuation is ≤±0.2℃, thus guaranteeing the continuity of drying. The switching instructions are responded to quickly through the open API interface of the intelligent SCADA system.

[0036] When temperature / pressure / flow parameters are abnormal, the PLC triggers an alarm signal and activates the waste heat exhaust device. The AI ​​subsystem of the intelligent SCADA system simultaneously uses root cause analysis algorithms to determine the cause of the abnormality (such as waste heat source fluctuations, sensor failures, and pipe blockages) and outputs optimization suggestions, improving the response speed by 30%.

[0037] During maintenance, the waste heat exhaust device is manually triggered. After the PLC confirms that the waste heat has been completely exhausted (the furnace temperature drops to ambient temperature ±2℃) by combining data from the sensor module, maintenance personnel are allowed to operate. The intelligent SCADA system records the maintenance process data simultaneously, providing a basis for subsequent AI model optimization and ensuring personnel safety.

[0038] When the big data algorithm of the intelligent SCADA system predicts that the waste heat parameters will exceed the normal range (such as a 15% decrease in flow rate or a 5°C decrease in temperature), the PLC will trigger the electric valve opening adjustment and fan frequency linkage in advance, without waiting for the parameter abnormality alarm, to realize the active control closed loop of "prediction-adjustment-stabilization". This function relies on edge AI inference technology to ensure that the control latency is less than 100ms.

[0039] In this embodiment, the temperature fluctuation prediction subunit uses the big data storage server of the intelligent SCADA system to store historical operating data in real time, extracts temperature fluctuation patterns through the LSTM algorithm of the AI ​​learning unit, and drives the PLC to adjust the electric valve in advance based on the prediction results, thus achieving a pre-prediction effect. The temperature curve adjustment subunit uses the sensor module to generate ambient and furnace temperature curves, trains the correlation model through the AI ​​learning unit of the intelligent SCADA system, and outputs correction commands, which drive the PLC to adjust in a coordinated manner, thus achieving dynamic temperature compensation and energy consumption reduction. The AI ​​learning optimization subunit uses the AI ​​learning unit of the intelligent SCADA system to periodically update the model and algorithm based on new data, and optimizes the adjustment parameters by supporting user-embedded custom models and building decision tree models, thereby improving control performance.

[0040] The intelligent SCADA system's big data storage server stores historical operating data in real time (including ambient temperature, furnace temperature, waste heat flow / pressure, electric valve opening, fan frequency, etc.). The AI ​​learning unit uses its supported LSTM time series prediction algorithm to mine temperature fluctuation patterns over different time periods (e.g., a 10%-15% decrease in waste heat flow between 8:00-10:00 AM daily, and a 3°C lag in furnace temperature decrease for every 5°C drop in ambient temperature during winter). Combined with early warning technology, the system outputs prediction results 10-30 minutes in advance, driving the PLC to intervene in advance to adjust the electric valve opening (e.g., increasing the electric valve opening by 12%-15% in advance if a 5°C drop in waste heat temperature is predicted), thus preventing significant temperature fluctuations. This function enables a shift from "post-event alarm" to "pre-event prediction."

[0041] The sensing module synchronously generates ambient temperature curves and furnace temperature curves. The AI ​​learning unit of the intelligent SCADA system trains a correlation model between the two through supervised learning (e.g., for every 3°C drop in ambient temperature, the furnace temperature will drop by 2°C after 5 minutes), and outputs correction commands in real time. When the ambient temperature fluctuates, the PLC adjusts the fan frequency and electric valve opening in conjunction (e.g., when the ambient temperature drops by 3°C, the hot air fan frequency increases by 5Hz, and the electric valve opening increases by 8% simultaneously), achieving dynamic temperature compensation. Simultaneously, relying on the energy efficiency optimization algorithm of the intelligent SCADA system, the parameters are dynamically optimized based on the drying system's energy consumption model, reducing energy consumption by more than 8.3%.

[0042] The intelligent SCADA system's AI learning unit regularly updates the temperature correlation model and prediction algorithm based on newly added operational data (cumulative data volume ≥1000 sets / week), optimizing adjustment parameters. It supports users embedding custom AI models via C++, Python, or other code, or quickly building decision tree models through a drag-and-drop interface. It automatically corrects the electric valve opening adjustment coefficient for the residual heat characteristics of low-temperature winter environments, improving control response speed by 20% and reducing temperature overshoot to within ±0.3℃. Furthermore, its multimodal data fusion capability can combine environmental text data (such as weather forecasts) and equipment operating parameters to further improve model prediction accuracy.

[0043] (1) Intelligentization: The AI ​​subsystem of the intelligent SCADA system realizes "self-learning and self-optimization", supports the integration of custom algorithms and multimodal data fusion. After running for 6 months, the accuracy of the control model is improved by 30%, and there is no need for frequent manual adjustment of parameters; the core code has an autonomy rate of over 90%, ensuring the system's autonomy and controllability; (2) Digitalization: The big data storage server has accumulated ≥100,000 sets of historical data. The intelligent SCADA system provides digital reports such as “waste heat utilization efficiency analysis”, “temperature fluctuation root cause location”, and “equipment health assessment” to help optimize the production process. The second-level processing capability of millions of data points ensures real-time data analysis and decision-making.

[0044] In this embodiment, the intelligent alarm management subunit is used to automatically adjust the alarm threshold using the dynamic threshold optimization algorithm of the intelligent SCADA system. By combining multi-parameter joint alarm and cluster analysis to locate the root cause of the fault, it combines early warning technology to obtain accurate alarms and handling suggestions.

[0045] By using the dynamic threshold optimization algorithm of the intelligent SCADA system, historical temperature, pressure, and flow data are analyzed to automatically adjust alarm thresholds, reducing invalid alarms by 90%. It supports multi-parameter joint alarms and cluster analysis. When abnormal parameters occur in the drying system, the algorithm locates the root cause of the fault (such as whether abnormal waste heat flow is caused by pipe blockage or heat source fluctuation) and generates an alarm summary. At the same time, combined with early warning technology, potential faults such as abnormal hot air blower load and abnormal pipe pressure are identified in advance, and handling suggestions are output.

[0046] In this embodiment, the health assessment subunit is used to combine the intelligent SCADA system AI subsystem with equipment sensor data and use deep learning algorithms to assess the health status of the equipment and predict the probability of failure, identify potential problems in advance and provide maintenance suggestions, thereby reducing the equipment failure rate.

[0047] The AI ​​subsystem of the intelligent SCADA system combines sensor data such as vibration, temperature, and current from equipment like hot air blowers and ovens, and uses deep learning algorithms to assess the health status of the equipment and predict the probability of failure. Referring to the equipment monitoring logic of the EMS system of the energy storage power station, it can identify potential problems such as wear on the bearings of hot air blowers and aging of oven seals in advance, and provide maintenance suggestions to reduce the equipment failure rate by more than 25%.

[0048] (1) Equipment safety: The intelligent SCADA system’s intelligent prediction and interlocking function and predictive maintenance capability intervene in waste heat fluctuations 20-30 seconds in advance, identify potential equipment hazards in advance, reduce the risk of equipment damage caused by sudden temperature rise / fall and component aging, and reduce equipment failure rate by 25%; (2) Abnormal safety: When parameters are abnormal, the root cause analysis algorithm is used to quickly locate the problem, and the linkage response speed of the waste heat exhaust device is increased by 30%, avoiding problems such as pipe bursts and furnace damage caused by heat accumulation; (3) Personnel safety: In the maintenance mode, the risk of personnel operation is reduced to zero by using sensor data and AI to confirm the discharge of residual heat.

[0049] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. An AI-based control system for drying chopped glass fiber using waste heat, characterized in that, The AI-based glass fiber chopped strand full waste heat drying control system includes: The core control module is used to link the PLC-based big data AI intelligent sub-module based on the intelligent SCADA system to form a dual control capability of passive adjustment and active prediction, and to build a full-stack AI control solution. The energy supply module is used to adopt a full waste heat supply device, connect the full waste heat intake pipe in parallel, regulate the waste heat supply through an electric regulating valve, and synchronize the operating status data to the intelligent SCADA system for AI analysis. The sensing module is used to collect multi-dimensional data using temperature, flow, and pressure sensors and transmit it in real time to the PLC and the big data storage server of the intelligent SCADA system. The execution module is used to receive PLC control commands from electric regulating valves, hot air blowers, exhaust fans, and waste heat air venting devices to execute actions, and to monitor and analyze load and operating status data in real time through an intelligent SCADA system and AI analysis. The protection module is used to ensure normal operation and data security during power outages using a UPS. The energy meter and flow meter monitor power consumption and waste heat flow, and the data is synchronized to the intelligent SCADA system to provide core data input for AI energy efficiency optimization algorithms.

2. The AI-based glass fiber chopped strand full waste heat drying control system according to claim 1, characterized in that, The control core module includes: The data acquisition submodule uses a PLC as the basic control core and links with the big data AI intelligent submodule based on the intelligent SCADA system to collect ambient temperature, furnace temperature, waste heat flow and pressure data in real time through the sensor module. The temperature fluctuation submodule is used to transmit the collected data to the big data storage server. The AI ​​learning unit extracts historical operating patterns and obtains temperature fluctuation prediction results. The control scheme submodule is used to generate adjustment commands through intelligent algorithms. The PLC actively adjusts the opening degree of the electric valve and the frequency of the fan based on the prediction results, forming a dual control capability of passive adjustment and active prediction, and building a full-stack AI control scheme.

3. The AI-based glass fiber chopped strand full waste heat drying control system according to claim 1, characterized in that, The energy supply module includes: The waste heat transfer submodule is used to transfer waste heat through parallel-connected waste heat intake pipes using a full waste heat supply device, without connecting any auxiliary heat source. The data transmission submodule is used to precisely adjust the waste heat supply through an electric regulating valve to ensure that the drying process relies entirely on waste heat, and to extract and transmit the operating status data to the intelligent SCADA system in real time. The data analysis submodule is used by the AI ​​learning unit to acquire parameters such as pipeline temperature, flow rate, and valve opening, and to obtain the waste heat supply efficiency and fluctuation characteristics through data analysis.

4. The AI-based glass fiber chopped strand full waste heat drying control system according to claim 1, characterized in that, The sensing module includes: The pressure information submodule is used to simultaneously collect ambient temperature and furnace temperature data using temperature sensors, monitor waste heat flow through flow sensors, and obtain pipeline and furnace pressure information using pressure sensors. The storage service submodule is used to extract multi-dimensional data in real time and send it to the PLC for basic control via the data transmission line, while uploading the raw data stream to the big data storage server of the intelligent SCADA system. The anomaly warning submodule is used by the intelligent SCADA system to acquire full operational data and obtain early warnings of temperature fluctuation characteristics, flow rate change trends, and pressure anomalies through built-in algorithms.

5. The AI-based glass fiber chopped strand full waste heat drying control system according to claim 1, characterized in that, The execution module includes: The instruction execution submodule is used to regulate the waste heat supply by an electric regulating valve, drive the air circulation by a hot air blower, and be responsible for exhaust gas discharge by an exhaust fan. The waste heat exhaust device exhausts waste heat in case of abnormality. All four receive control instructions sent by the PLC to execute actions. The fault warning submodule is used to extract equipment load data and operating status parameters in real time through the intelligent SCADA system. The AI ​​learning unit analyzes the data stream to obtain equipment efficiency evaluation results and potential fault warnings. The results feedback submodule is used by the intelligent SCADA system to feed back the analysis results to the PLC, forming a closed-loop management of equipment status monitoring, AI analysis, and control optimization.

6. The AI-based glass fiber chopped strand full waste heat drying control system according to claim 1, characterized in that, The protection module includes: The secure storage submodule is used to ensure the normal operation of the control module, sensor module and AI subsystem during power outages by using UPS uninterruptible power supply and real-time power supply, thus ensuring secure data storage. The synchronous transmission submodule is used to continuously monitor power consumption data through the power meter and synchronously acquire waste heat flow information through the flow meter. It extracts power consumption data and waste heat flow information in real time and transmits them to the intelligent SCADA system. The AI ​​optimization submodule is used by AI energy efficiency optimization algorithms to obtain power consumption data and waste heat flow information, and analyze the energy consumption distribution characteristics and waste heat utilization efficiency.

7. The AI-based glass fiber chopped strand full waste heat drying control system according to claim 1, characterized in that, The big data AI intelligent submodule based on the intelligent SCADA system includes: The temperature fluctuation prediction subunit is used to store historical operating data in real time using the big data storage server of the intelligent SCADA system. It extracts the temperature fluctuation pattern through the LSTM algorithm of the AI ​​learning unit and drives the PLC to adjust the electric valve in advance to achieve the effect of advance prediction. The temperature curve adjustment subunit is used to generate environmental and furnace temperature curves using a sensor module, train the correlation model through the AI ​​learning unit of the intelligent SCADA system and output correction instructions, and drive the PLC to adjust in linkage to achieve dynamic temperature compensation and energy consumption reduction. The AI ​​learning optimization subunit is used to periodically update the model and algorithm based on new data using the AI ​​learning unit of the intelligent SCADA system. By supporting users to embed custom models and build decision tree models, the adjustment parameters are optimized to achieve improved control performance.

8. The AI-based glass fiber chopped strand full waste heat drying control system according to claim 1, characterized in that, The big data AI intelligent submodule based on the intelligent SCADA system also includes an intelligent alarm management subunit, which is used to automatically adjust the alarm threshold using the dynamic threshold optimization algorithm of the intelligent SCADA system, locate the root cause of the fault through multi-parameter joint alarm and cluster analysis, and combine early warning technology to obtain accurate alarms and handling suggestions.

9. The AI-based glass fiber chopped strand full waste heat drying control system according to claim 1, characterized in that, The big data AI intelligent submodule based on the intelligent SCADA system also includes a health assessment subunit, which uses the intelligent SCADA system's AI subsystem to combine equipment sensor data and deep learning algorithms to assess the equipment's health status and predict the probability of failure, identify potential problems in advance and provide maintenance suggestions, thereby reducing the equipment failure rate.

Images