An industrial robot automation steady-state control method for continuous production of recycled ABS plastic

Abstract

This invention belongs to the field of smart city information technology, specifically relating to an automated steady-state control method for industrial robots in the continuous production of recycled ABS plastics. It covers the entire continuous operation of recycled ABS from material conveying and reaction plasticization to finished product output. Employing core algorithms such as dynamic coordination of material conveying and reaction demand, steady-state control of multi-parameter coupling in the reaction process, and full-process quality data chain traceability, it overcomes the technical bottlenecks of traditional continuous recycled ABS production, including "disconnected conveying, process fluctuations, and lack of traceability." This achieves precise coordination of continuous material conveying, stable and controllable reaction processes, and traceable quality control of finished products, significantly improving the stability, production efficiency, and quality reliability of continuous recycled ABS production, and adapting to the automated production needs of large-scale continuous recycled ABS products.

Description

Technical Field

[0001] This invention belongs to the field of information technology, specifically relating to an automated steady-state control method for industrial robots in the continuous production of recycled ABS plastic. Background Technology

[0002] In the current field of continuous production of recycled ABS plastics, existing technologies have achieved basic continuous production operations. However, in practical industrial applications involving large-scale continuous collaboration (conveyance-reaction-detection), high-precision steady-state control (parameter fluctuation ≤ ±0.5%), and full-process quality traceability, specific and unresolved practical problems remain in three sub-scenarios: poor material conveying coordination, insufficient steady-state reaction process, and difficulty in traceability of finished product quality. These are all specific process-related issues, not macro-level technical challenges, as detailed below: Poor material conveying coordination and lack of dynamic coordination and matching algorithms: In the continuous production of recycled ABS, material conveying and reaction consumption need to be precisely synchronized. Existing technologies adopt a "fixed rate conveying - passive adaptation" mode, which lacks a dynamic coordination algorithm for material conveying and reaction demand. The conveying rate and reaction consumption are mismatched, which easily leads to overflow when the material level is too high or material interruption when the material level is too low. At the same time, there is no material level-rate adaptive matching algorithm, resulting in large fluctuations in material level, which leads to poor continuity of material supply and affects the stability of subsequent reactions.

[0003] The reaction process lacks steady-state stability and lacks coupled control and fluctuation prediction algorithms: Reaction plasticization is the core link of continuous production. Existing technology adopts the mode of "single parameter independent control - post-correction", which lacks multi-parameter coupled steady-state control algorithms for the reaction process. The coordination of temperature, pressure and speed is poor, and the reaction system is prone to fluctuation. Moreover, there are no reaction fluctuation prediction and suppression algorithms. It can only correct after the fluctuation occurs and cannot intervene in advance, resulting in large fluctuations in product performance.

[0004] Finished product quality traceability is difficult due to the lack of data chain traceability and anomaly tracing algorithms: In continuous production, finished product quality needs to be traceable. Existing technologies adopt a "batch general record - no precise correlation" model, which lacks a full-process quality data chain traceability algorithm. Production data and finished product quality are not uniquely related, making it impossible to accurately locate the quality anomaly link. Furthermore, there is no finished product quality anomaly tracing and correction algorithm, which makes it impossible to quickly trace the cause and correct it after an anomaly occurs, leading to batch quality problems.

[0005] Existing methods for continuous production of recycled ABS lack core algorithmic innovation in areas such as continuous material transport coordination, multi-parameter steady-state control of the reaction process, and quality traceability management throughout the entire production process. Significant technological gaps exist, particularly in modeling and solving dynamic coordination of transport and reaction, coupled steady-state control of multiple parameters, and end-to-end data chain traceability, failing to address the aforementioned specific problems. There is an urgent need for an automated steady-state control method for industrial robots in the continuous production of recycled ABS, centered on algorithmic innovation. This method should focus on three new technological perspectives: transport coordination, process steady-state control, and traceability management, to achieve stable, precise, and traceable continuous production of recycled ABS, filling the technological gap in large-scale continuous automated production of recycled ABS. Summary of the Invention

[0006] Addressing the three specific problems raised in the background art, the present invention aims to provide an automated steady-state control method for industrial robots in the continuous production of recycled ABS plastics. This method achieves continuous and coordinated conveying of recycled ABS materials, steady-state control of multiple parameters in the reaction process, and quality traceability and control throughout the entire production process. It solves the problems of poor material conveying coordination, insufficient steady-state control of the reaction process, and difficulty in tracing the quality of finished products. The method emphasizes algorithm innovation and modeling and solving processes throughout, without involving rules for intelligent activities, thereby improving the stability, production efficiency, and quality reliability of continuous recycled ABS production and further perfecting the technical system for continuous recycled ABS production.

[0007] The present invention is implemented through the following specific technical solution: (a) Continuous conveying and coordination module for recycled ABS materials The core of this module is to achieve precise coordination between material conveying and reaction requirements, dynamic matching of material level and conveying rate, and to build a continuous conveying coordination model for recycled ABS materials. This solves the problems of poor material conveying coordination and unstable supply, and provides a continuous and stable material basis for the reaction process.

[0008] Modeling Approach: Abandoning the traditional "fixed rate - passive adaptation" conveying approach, we construct an integrated modeling logic of "material data acquisition - collaborative modeling - rate adjustment - material level matching - stable supply". Combining the characteristics of recycled ABS materials and reaction consumption patterns, we establish a conveying-reaction correlation model, a material level-rate matching model, and a collaborative optimization model. We design a dynamic collaborative algorithm for material conveying-reaction demand and a material level-rate adaptive matching algorithm to achieve continuous material conveying collaboration.

[0009] Solution Process: First, deploy flow sensors, ultrasonic level gauges, and humidity sensors to collect multi-dimensional data such as material conveying rate, silo level, and material humidity. Then, design a dynamic collaborative algorithm for material conveying and reaction demand, establish a correlation model between conveying rate and reaction consumption rate, and use a PID incremental algorithm to dynamically adjust the conveyor motor speed to ensure precise matching between conveying volume and reaction consumption. Simultaneously, design an adaptive matching algorithm between silo level and rate to monitor the silo level in real time, establish a correlation model between silo level and conveying rate, and automatically adjust the conveying rate when the silo level fluctuates to maintain silo level stability. Finally, construct a verification model for conveying collaboration effect, quantify conveying fluctuations, silo level stability, and supply continuity, and dynamically optimize algorithm parameters to ensure continuous and stable material supply.

[0010] (II) Multi-parameter steady-state control module for the reaction process The core of this module is to achieve the coupling and coordination of multiple parameters such as temperature, pressure, and rotation speed in the reaction process, as well as the early prediction and real-time suppression of reaction fluctuations. It constructs a multi-parameter steady-state control model for the reaction process, solves the problems of insufficient steady-state stability and large parameter fluctuations in the reaction process, and maintains the stability of the reaction system.

[0011] Modeling Approach: Abandoning the traditional reaction approach of "single-parameter independent control - post-correction", we construct an integrated modeling logic of "reaction data acquisition - coupled modeling - steady-state control - fluctuation prediction - early suppression". Combining the plasticizing characteristics of recycled ABS reaction with the multi-parameter coupling law, we establish a multi-parameter coupled correlation model, a steady-state control model, and a fluctuation prediction model. We design a multi-parameter coupled steady-state control algorithm for the reaction process and a reaction fluctuation prediction and suppression algorithm to achieve multi-parameter steady-state control of the reaction process.

[0012] Solution Process: First, temperature, pressure, and speed sensors are deployed to collect real-time data such as temperature, pressure, and screw speed during the reaction process. Then, a multi-parameter coupled steady-state control algorithm for the reaction process is designed. The control effect is quantified through a steady-state control accuracy calculation formula, and a multi-parameter coupled correlation model is established to coordinately adjust heating power, pressure compensation, and motor speed to maintain the steady state of the reaction system. Simultaneously, a reaction fluctuation prediction and suppression algorithm is designed. Based on historical data, an LSTM neural network is used to establish a fluctuation prediction model to predict fluctuation trends in advance and initiate targeted suppression strategies. Finally, a steady-state control effect verification model is constructed to quantify parameter fluctuation amplitude, steady-state maintenance rate, and reaction efficiency, and the algorithm parameters are dynamically optimized to ensure that the reaction process is stable and controllable.

[0013] (III) Production Process Quality Traceability and Control Module This module enables full-chain recording of production data, accurate tracing of finished product quality anomalies, and process correction. It constructs a full-process quality traceability and control model to solve the problems of difficulty in tracing finished product quality and inability to accurately correct anomalies, thus ensuring the traceability and consistency of finished product quality.

[0014] Modeling Approach: Abandoning the traditional traceability approach of "general recording without precise correlation", we construct an integrated modeling logic of "traceability coding allocation - real-time data correlation - data chain construction - anomaly tracing - process correction". Combining the batch characteristics and quality standards of continuous production, we establish a data chain correlation model, anomaly-process correlation model, and traceability verification model. We design a full-process quality data chain traceability algorithm and a finished product quality anomaly tracing and correction algorithm to achieve full-process quality traceability control in production.

[0015] Solution Process: First, a unique traceability code is assigned to each batch of recycled ABS material entering the production process. Then, a full-process quality data chain traceability algorithm is designed to link material conveying parameters, reaction parameters, and finished product testing data in real time, constructing a full-chain data association from raw materials to process to finished products to ensure data integrity. Simultaneously, a finished product quality anomaly traceability and correction algorithm is designed to establish a quality anomaly-process parameter association model. When a finished product quality anomaly is detected, the traceability code is used to locate the process link where the anomaly occurred, and the corresponding process parameters are dynamically adjusted. Finally, a traceability control effectiveness verification model is constructed to quantify traceability data integrity, traceability response time, and anomaly correction success rate, dynamically optimizing algorithm parameters to ensure the traceability and consistency of finished product quality.

[0016] Beneficial effects Material conveying-reaction demand dynamic coordination algorithm: Construct a conveying-reaction correlation model to achieve precise coordination between material conveying and reaction. Compared with traditional fixed-rate conveying, the supply fluctuation is ≤3% and the reaction interruption rate is reduced to below 0.1%, filling the technical gap in material conveying-reaction coordination in continuous production of recycled ABS. Material level-rate adaptive matching algorithm: Establish a material level-rate matching model to achieve stable material level control. Compared with no matching algorithm, the material level fluctuation range is reduced, which solves the problem of large material level fluctuation. Multi-parameter coupled steady-state control algorithm for reaction process: Design a formula for calculating the accuracy of steady-state control to achieve multi-parameter coupling and synergy. Compared with single-parameter control, the parameter fluctuation amplitude is reduced, focusing on multi-parameter steady-state innovation in reaction process; Reaction fluctuation prediction and suppression algorithm: It realizes the early prediction and suppression of reaction fluctuations. Compared with the post-correction mode, the fluctuation amplitude is reduced by 70% and the reaction stability is significantly improved, filling the technical gap in the prediction of reaction fluctuations of recycled ABS. End-to-end quality data chain traceability algorithm: Establishes a full-chain data association model to achieve accurate quality traceability, which solves the problem of difficult quality traceability compared with the general recording mode; Finished Product Quality Anomaly Traceability and Correction Algorithm: Enables accurate anomaly traceability and process correction, focusing on closed-loop management innovation for quality anomalies compared to the non-traceability correction mode. Attached Figure Description

[0017] Figure 1 Workflow diagram of the continuous conveying and coordination module for recycled ABS materials Detailed Implementation

[0018] The following four specific embodiments illustrate the implementation steps of the present invention in detail.

[0019] Example 1: Steady-state control of multi-parameter coupling in the reaction process Implementation steps Step 1: Data Acquisition and Modeling: Select the continuous extrusion pipe production scenario of recycled ABS, deploy temperature, pressure, and speed sensors to collect reaction process data, and determine the optimal reaction temperature. Optimal reaction pressure Optimal screw speed Establish a multi-parameter coupled correlation model.

[0020] Step 2: Steady-state control accuracy calculation: A multi-parameter coupled steady-state control algorithm for the reaction process is adopted to set a steady-state accuracy threshold for high-precision continuous production scenarios. Through the steady-state control accuracy calculation formula The comprehensive steady-state control accuracy under the initial process parameters is measured in real time. The initial calculation found that temperature fluctuations caused the accuracy to exceed the threshold.

[0021] Step 3: Multi-parameter coordinated adjustment: Based on the multi-parameter coupling correlation model, implement a multi-parameter coordinated adjustment strategy: optimize the heating zone power to correct temperature deviation, and simultaneously adjust the pressure compensation value and screw speed to ensure that temperature, pressure, and speed are coupled and coordinated to maintain the steady state of the reaction system.

[0022] Step 4: Steady-state accuracy iterative verification: After each adjustment, re-acquire reaction parameters and calculate steady-state accuracy. If the target is still not met, continue to optimize parameters (such as adjusting the heating response rate and optimizing the pressure compensation coefficient) until... This ensures the stability of the reaction process.

[0023] Step 5: Verification of Control Effect: Perform continuous extrusion pipe production, test the pipe wall thickness uniformity, dimensional accuracy and mechanical properties to ensure stable product quality without significant fluctuations, and meet the production requirements of recycled ABS continuous extrusion pipes.

[0024] Modeling Innovation Principles Abandoning the traditional crude control approach of "single-parameter modeling - uncoupled calculation," this paper constructs an integrated closed-loop model of "data acquisition - coupled modeling - precision calculation - collaborative adjustment - iterative verification." It uses the reaction characteristics and high-precision steady-state requirements of continuously extruded pipes as core inputs, overcoming the limitations of multi-parameter coupled steady-state control. Multi-parameter coupled correlation modeling achieves quantitative correlation between temperature, pressure, and rotational speed, providing a scientific basis for collaborative adjustment. Steady-state control precision calculation formula modeling enables comprehensive quantitative judgment of steady-state effects, which is more comprehensive than single-parameter evaluation. Multi-parameter collaborative adjustment modeling achieves dynamic adaptation of the three parameters, avoiding secondary fluctuations caused by single-parameter adjustment, filling the modeling gap in multi-parameter coupled steady-state control of continuously reacting recycled ABS. The modeling process focuses on continuous extrusion scenarios, completely different from existing modeling approaches and technical directions, representing a novel modeling direction.

[0025] Algorithm efficiency enhancement principle The multi-parameter coupled steady-state control algorithm for the reaction process accurately captures the interaction patterns of temperature, pressure, and rotation speed by establishing a multi-parameter coupled correlation model. Compared with independent control without correlation, it can formulate optimization strategies from a global perspective, avoiding "attention to one aspect at the expense of another." The steady-state control accuracy calculation formula normalizes and comprehensively quantifies the deviations of multiple parameters. Compared with evaluating a single parameter individually, it can scientifically judge the overall steady-state state and accurately locate the core of fluctuations. The multi-parameter collaborative adjustment mechanism ensures the dynamic adaptation of each parameter. Compared with step-by-step adjustment, it significantly reduces the risk of secondary fluctuations and significantly improves the steady-state maintenance rate of the reaction system. The iterative verification and optimization process ensures that the reaction process is always in the optimal steady state. Compared with the fixed parameter mode, the product quality stability is greatly improved. This algorithm completely solves the industry pain point of insufficient steady-state performance in the continuous extrusion production of recycled ABS.

[0026] Compared with existing technologies Existing technologies in the continuous extrusion production of recycled ABS employ independent control of single parameters, lacking multi-parameter coupled modeling and steady-state accuracy calculation formulas. This results in poor parameter coordination and large fluctuations, leading to uneven pipe wall thickness and significant dimensional deviations, failing to meet the demands of continuous large-scale production. This embodiment, through algorithmic innovation and modeling optimization, achieves coupled steady-state control of multiple parameters in the reaction process, significantly improving product quality stability. It completely resolves the pain points of existing technologies and achieves a novel breakthrough without any overlap in technical direction or modeling approach with existing technologies, perfectly adapting to the production requirements of continuously extruded recycled ABS pipes.

[0027] Example 2: Dynamic Coordination of Material Transportation and Reaction Demand (Adapted to the Continuous Injection Molding Shell Production Scenario of Recycled ABS) Implementation steps Step 1: Data Acquisition and Modeling: Select the continuous injection molding shell production scenario of recycled ABS, collect data on material conveying rate, reaction consumption rate, and silo level, establish a material conveying-reaction demand correlation model, and clarify the material consumption patterns at different reaction stages.

[0028] Step 2: Dynamic Coordination Adjustment: The material conveying-reaction demand dynamic coordination algorithm is adopted. Based on the correlation model, the PID incremental algorithm is used to dynamically adjust the conveyor motor speed: reduce the conveying rate in the early stage of the reaction, maintain a constant rate in the stable period of the reaction, and appropriately reduce the rate in the final stage of the reaction to ensure that the conveying volume and the reaction consumption are accurately matched.

[0029] Step 3: Material level matching optimization: Through the material level-rate adaptive matching algorithm, the material level in the silo is monitored in real time. When the material level is lower than the set threshold, the conveying rate is automatically increased; when the material level is higher than the threshold, the conveying rate is reduced to maintain the material level in a stable range.

[0030] Step 4: Validate the synergistic effect: Detect the fluctuation range of material supply and the number of reaction interruptions, and determine whether the supply fluctuation is ≤3% and the reaction interruption rate is ≤0.1%. If the standards are not met, adjust the PID parameters and material level thresholds and re-optimize.

[0031] Step 5: Verification of the synergistic effect of the conveying system: Apply the stable supply of materials to continuous injection molding production, and test the filling integrity and surface quality of the injection molded shell to ensure that the product is free of missing materials and air bubbles, thus meeting the requirements of continuous injection molding production.

[0032] Modeling Innovation Principles Abandoning the traditional extensive conveying approach of "fixed rate - no collaborative modeling," this paper constructs an integrated closed-loop model of "data acquisition - correlation modeling - rate adjustment - material level matching - effect verification." It uses the reaction consumption characteristics and stable supply demand of continuous injection molding as core inputs, overcoming the limitation of difficulty in dynamically coordinating material conveying and reaction demand. The correlation modeling of conveying and reaction demand achieves a quantitative correspondence between conveying rate and reaction consumption, providing a scientific basis for dynamic adjustment. The PID incremental algorithm modeling achieves precise control of the conveying rate, with faster response and less fluctuation compared to traditional PID algorithms. The material level-rate matching modeling achieves dynamic adaptation of material level and rate, avoiding supply problems caused by abnormal material levels and filling the modeling gap in the collaborative conveying of recycled ABS materials in continuous injection molding. The modeling process focuses on the continuous injection molding scenario, completely different from the modeling ideas and technical directions of existing technologies, representing a completely new modeling direction.

[0033] Algorithm efficiency enhancement principle The dynamic coordination algorithm for material conveying and reaction demand establishes a correlation model to accurately capture the real-time correspondence between conveying rate and reaction consumption. Compared with fixed-rate conveying, it can dynamically adapt to changes in consumption during the reaction stage, significantly improving supply stability. The PID incremental algorithm can quickly respond to subtle changes in reaction consumption. Compared with traditional control algorithms, the conveying rate adjustment is more precise, and the fluctuation range is reduced to below 3%. The material level-rate adaptive matching mechanism can correct material level deviations in real time. Compared with no material level control, the material level stability is significantly improved, avoiding the risk of overflow or material shortage. The collaborative optimization process ensures uninterrupted coordination between conveying and reaction. Compared with traditional conveying modes, the reaction interruption rate is significantly reduced, and the production efficiency is significantly improved. This algorithm realizes a stable material supply for continuous injection molding production of recycled ABS, completely solving the problems of conveying disconnection and supply fluctuation in traditional production.

[0034] Compared with existing technologies Existing technologies in continuous injection molding of recycled ABS employ fixed-rate material delivery without dynamic collaborative algorithms and correlation models. This leads to a mismatch between the delivery rate and reaction consumption, resulting in large material level fluctuations and a tendency for overflows, material shortages, or insufficient material supply. Consequently, these issues result in high product defect rates, low production efficiency, and an inability to meet the demands of continuous production. This embodiment, through algorithmic innovation and model optimization, achieves dynamic coordination between material delivery and reaction requirements. This significantly improves supply stability and production efficiency, completely resolving the pain points of existing technologies. Furthermore, it avoids any overlap with existing technologies in terms of technical direction and modeling approach. Its innovative points are prominent and highly practical, effectively improving the quality and efficiency of continuous injection molding of recycled ABS.

[0035] Example 3: End-to-end quality data chain traceability (adapted to the scenario of continuous production of recycled ABS sheets) Implementation steps Step 1: Traceability Coding Assignment and Modeling: Select the scenario of continuous production of recycled ABS sheets, assign a unique traceability code to each batch of raw materials entering the production process, establish a data chain association model of "raw materials-process-finished products", and clarify the key parameters that need to be associated (transport rate, reaction temperature, sheet thickness, mechanical properties).

[0036] Step 2: Real-time data association and collection: Using a full-process quality data chain traceability algorithm, material conveying parameters, reaction process parameters, and finished product testing data are collected in real time. Data from each stage is associated and stored through traceability coding to ensure data integrity and relevance.

[0037] Step 3: Data Chain Construction and Verification: Construct a full-process quality data chain, check data integrity and correlation accuracy, determine whether the integrity of traceability data is ≥99.9%, and if there are missing data or correlation errors, optimize the data collection and correlation strategy and rebuild it.

[0038] Step 4: Quality Anomaly Traceability Drill: Simulate a finished product quality anomaly scenario (such as uneven board thickness), locate the process link where the anomaly occurred (such as reaction temperature fluctuation) through traceability coding, and verify whether the traceability response time is ≤1s.

[0039] Step 5: Verify the traceability effect: Apply the established data chain to actual continuous production, detect the quality consistency of batch sheets, ensure that the cause can be quickly traced when an anomaly occurs, and meet the quality traceability requirements of recycled ABS continuous production sheets.

[0040] Modeling Innovation Principles Abandoning the traditional, crude approach of "general recording and unrelated modeling," this approach constructs an integrated closed-loop model encompassing "code allocation, data collection, data chain construction, traceability verification, and effect optimization." It uses the batch characteristics and traceability requirements of continuously produced ABS sheets as core inputs, overcoming the limitations of tracing the entire process's quality data chain. Unique traceability coding modeling achieves precise binding between batches and data, providing a unique identifier for traceability. The "raw material-process-finished product" data chain association modeling achieves systematic data correlation across all stages, offering greater logic compared to scattered recording. Real-time data collection and association modeling enable dynamic data updates, ensuring the timeliness of traceability data and filling the modeling gap in the continuous production data chain traceability of recycled ABS. The modeling process focuses on continuous ABS sheet production scenarios, representing a completely different modeling approach and direction from existing technologies, thus offering a novel modeling direction.

[0041] Algorithm efficiency enhancement principle The end-to-end quality data chain traceability algorithm, by assigning unique traceability codes, enables full-process tracking of each batch of materials from raw materials to finished products. Compared to general records without codes, the accuracy of traceability is significantly improved. The data chain association model systematically links key parameters of each link, enabling rapid location of quality anomalies compared to scattered data storage, reducing traceability response time to less than 1 second. Real-time data acquisition ensures the integrity and timeliness of traceability data, providing stronger data reliability and avoiding data loss or errors compared to post-event data entry. The data chain construction and verification process ensures the stable operation of the traceability system, significantly improving the traceability success rate compared to traceability modes without verification. This algorithm realizes end-to-end quality traceability for continuously produced recycled ABS sheets, completely solving the problem of inaccurate traceability of quality anomalies in traditional production, and providing core technical support for quality control.

[0042] Compared with existing technologies Existing technologies in the continuous production of recycled ABS use batch-based, general data recording without unique traceability codes or data chain associations. This makes it impossible to accurately pinpoint specific stages of quality anomalies, resulting in low traceability efficiency and accuracy, and failing to meet the quality control requirements of large-scale continuous production. This embodiment, through algorithmic innovation and model optimization, achieves end-to-end quality data chain traceability, significantly improving traceability accuracy and efficiency. It completely solves the pain points of existing technologies and has no overlap with their technical direction or modeling approach, effectively adapting to the quality control requirements of continuously produced recycled ABS sheets.

[0043] Example 4: Prediction and Suppression of Reaction Fluctuations (Adapted to the Production Scenarios of Continuous Molding Profiles from Recycled ABS) Implementation steps Step 1: Data Acquisition and Modeling: Select the production scenario of continuously molded profiles of recycled ABS, collect historical reaction data (temperature and pressure fluctuation data), establish a reaction fluctuation prediction model based on LSTM neural network, and clarify the correspondence between the fluctuation prediction lead time (≥50ms) and the suppression strategy.

[0044] Step 2: Real-time monitoring and prediction of fluctuations: Using a reaction fluctuation prediction and suppression algorithm, temperature and pressure data of the reaction process are collected in real time, input into the prediction model, and fluctuation trends are predicted in advance (such as temperature is about to rise or pressure is about to fall).

[0045] Step 3: Execution of fluctuation suppression strategy: When the model predicts fluctuation risk, the suppression strategy is activated in advance: when temperature fluctuation is predicted, the heating power is adjusted; when pressure fluctuation is predicted, pressure compensation is activated; when speed fluctuation is predicted, the motor drive parameters are adjusted to suppress fluctuations.

[0046] Step 4: Validation of suppression effect: Detect the fluctuation range of the response parameters after suppression and determine whether it has been reduced to less than 30% of the traditional mode. If the suppression effect is not good, adjust the model parameters (such as increasing training data and optimizing the suppression strategy parameters) and re-optimize.

[0047] Step 5: Steady-state production verification: Perform continuous molding profile production, test the dimensional accuracy and mechanical property stability of the profiles, ensure that there are no significant fluctuations in the reaction process, and ensure consistent product quality to meet the production requirements of recycled ABS continuous molding profiles.

[0048] Modeling Innovation Principles Abandoning the traditional, crude suppression approach of "post-event detection - no predictive modeling," this paper constructs an integrated closed-loop model encompassing "data acquisition - model building - fluctuation prediction - suppression execution - effect verification." It uses the reaction fluctuation characteristics and steady-state requirements of continuously molded profiles as core inputs, overcoming the limitations of predicting and suppressing reaction fluctuations. LSTM neural network predictive modeling leverages the advantages of time-series data processing to achieve early prediction of fluctuation trends, significantly improving prediction accuracy and lead time compared to traditional prediction algorithms. Fluctuation-suppression strategy correlation modeling achieves precise correspondence between prediction results and suppression measures, providing greater targeting than empirical suppression. Real-time monitoring and early suppression modeling enable proactive fluctuation control, avoiding passive correction after fluctuations occur, filling the modeling gap in predicting and suppressing continuous reaction fluctuations in regenerated ABS. The modeling process focuses on continuous molding scenarios, representing a completely new modeling direction compared to existing technologies.

[0049] Algorithm efficiency enhancement principle The reaction fluctuation prediction and suppression algorithm, through an LSTM neural network model, can uncover the temporal variation patterns of reaction parameters, achieving advance prediction of ≥50ms. Compared to post-correction without prediction, this provides ample time to initiate suppression strategies, significantly improving fluctuation suppression effectiveness. A real-time fluctuation monitoring mechanism ensures the timeliness and accuracy of the prediction data. Compared to timed detection, it can quickly capture early signs of fluctuations, resulting in stronger prediction reliability. Targeted suppression strategies activate corresponding measures based on different fluctuation types, achieving higher suppression accuracy and significantly reducing fluctuation amplitude compared to general suppression methods. The model iterative optimization process continuously optimizes model parameters based on actual suppression effects, continuously improving prediction accuracy and suppression effectiveness compared to a fixed model. This algorithm achieves proactive control of reaction fluctuations in continuous molding production of recycled ABS, completely solving the problems of frequent fluctuations and unstable product quality in traditional production.

[0050] Compared with existing technologies Existing technologies in the continuous molding production of recycled ABS can only passively correct fluctuations after they occur, lacking algorithms for fluctuation prediction and suppression. This results in large fluctuation amplitudes and poor product quality stability, failing to meet the demands of continuous large-scale production. This embodiment, through algorithmic innovation and modeling optimization, achieves early prediction and suppression of reaction fluctuations, significantly improving reaction stability and product quality. It completely solves the pain points of existing technologies and has no overlap with existing technologies in terms of technical direction and modeling ideas. The innovation is clear and highly practical, effectively contributing to the stable production of continuously molded recycled ABS profiles.

[0051] 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 principles of 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 claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. An automated steady-state control method for industrial robots in continuous production of recycled ABS plastic, characterized in that, Includes the following steps: S1: Continuous conveying and collaborative processing of recycled ABS materials. It collects multi-source data such as material conveying rate, material level, and material moisture. Through dynamic collaborative algorithms for material conveying and reaction demand and adaptive matching algorithms for material level and rate, it achieves precise coordination between material conveying and reaction demand and dynamic matching between material level and conveying rate, ensuring a continuous and stable supply of materials. S2: Multi-parameter steady-state control of the reaction process. Based on material characteristics and reaction standards, it achieves coupled and coordinated control of multiple parameters such as temperature, pressure, and rotation speed, as well as early prediction and real-time suppression of reaction fluctuations through multi-parameter coupled steady-state control algorithm and reaction fluctuation prediction and suppression algorithm, thereby maintaining the steady state of the reaction system. S3: Full-process quality traceability and control processing, real-time collection of full-process production parameters and finished product quality data, through full-process quality data chain traceability algorithm and finished product quality anomaly tracing and correction algorithm, to achieve full-chain recording of production data, accurate tracing of finished product quality anomalies and process correction, ensuring the traceability and consistency of finished product quality; The multi-parameter coupled steady-state control algorithm for the reaction process in step S2 includes a formula for calculating the steady-state control accuracy: The constraints are , To improve the accuracy of steady-state control, This is to account for the actual reaction temperature deviation. The optimal reaction temperature, This is to account for the actual pressure deviation. For optimal reaction pressure, This represents the actual screw speed deviation. To achieve the optimal screw speed, For steady-state accuracy thresholds, high-precision continuous production scenarios Conventional continuous production scenarios .

2. The method according to claim 1, characterized in that, The material conveying-reaction demand dynamic coordination algorithm in step S1 includes the following sub-steps: establishing a material conveying rate-reaction consumption rate correlation model, using a PID incremental algorithm to dynamically adjust the conveying motor speed, achieving uninterrupted coordination between conveying and reaction, with material supply fluctuation ≤3% and reaction interruption rate reduced to below 0.1%.

3. The method according to claim 1, characterized in that, The material level-rate adaptive matching algorithm in step S1 includes the following sub-steps: real-time monitoring of the material level in the silo using an ultrasonic level gauge, establishing a material level-conveying rate correlation model, automatically increasing the conveying rate when the material level is below the threshold, and decreasing the rate when the material level is above the threshold, thus reducing the material level fluctuation range to less than 20% of the traditional mode.

4. The method according to claim 1, characterized in that, The reaction fluctuation prediction and suppression algorithm in step S2 includes the following sub-steps: establishing a fluctuation prediction model based on historical reaction data, using an LSTM neural network to predict temperature and pressure fluctuation trends in advance, and starting a suppression strategy (adjusting heating power and pressure compensation) 50ms in advance, reducing the fluctuation amplitude to less than 30% of the traditional mode.

5. The method according to claim 1, characterized in that, The full-process quality data chain traceability algorithm in step S3 includes the following sub-steps: assigning a unique traceability code to each batch of materials, associating the transport parameters, reaction parameters, and test data in real time, establishing a "raw material-process-finished product" data chain, ensuring traceability data integrity ≥99.9%, and a traceability response time ≤1s.

6. The method according to claim 1, characterized in that, The finished product quality anomaly tracing and correction algorithm in step S3 includes the following sub-steps: establishing a quality anomaly-process parameter correlation model, locating the abnormal link through traceability coding, dynamically adjusting the corresponding process parameters (such as reaction temperature and conveying rate), and achieving an anomaly correction success rate of ≥95%.

7. The method according to any one of claims 1-6, characterized in that, The process parameters for continuous material conveying coordination are: conveying rate control accuracy ≤ 0.01m / s, material level monitoring response time ≤ 10ms, material supply continuity ≥ 99.9%, and adaptability to recycled ABS materials with different viscosities and humidity.

8. The method according to any one of claims 1-6, characterized in that, The process parameters for steady-state control of the reaction process are: reaction temperature control accuracy ≤ ±0.5℃, pressure control accuracy ≤ ±0.01MPa, rotation speed control accuracy ≤ ±1r / min, and steady-state maintenance rate of the reaction system ≥ 99.5%.

9. The method according to any one of claims 1-6, characterized in that, The method can be applied to 6-9 axis industrial robots, and is compatible with continuous production processes of recycled ABS such as continuous extrusion, continuous injection molding, and continuous compression molding, supporting the mass production of products such as pipes, sheets, and profiles.

10. An automated steady-state control system for industrial robots in continuous production of recycled ABS plastic, characterized in that, It includes a continuous conveying and coordination module for recycled ABS materials, a multi-parameter steady-state control module for the reaction process, a quality traceability and control module for the entire production process, and a steady-state control center. The steady-state control center communicates bidirectionally with the three functional modules, integrates six core algorithms, and executes the method described in any one of claims 1-9 to achieve algorithmic, automated, and steady-state control of the entire continuous production process of recycled ABS.

Images