Automatic control method of carbon black master batch production equipment based on PCL

Abstract

The invention relates to the technical field of carbon black master batch production data processing, and particularly discloses a PCL-based carbon black master batch production equipment automatic control method, which is used for solving the problems of poor carbon black dispersion uniformity, low conductive threshold regulation and control precision and the like caused by insufficient multi-source heterogeneous data fusion and cross-process parameter dynamic coupling deficiency. According to the method, heterogeneous data streams of mixing equipment, extrusion equipment and detection equipment are collected in real time, a sliding window mechanism is adopted to carry out time domain alignment processing and extract frequency domain features, a knowledge base is triggered to be iteratively updated through real-time quality data feedback, and a buffer material level closed-loop feedback mechanism is established to guarantee material transmission continuity. Through a technical path of combining data fusion, virtual simulation and dynamic optimization, the carbon black dispersion uniformity and the conductivity regulation and control precision are improved, meanwhile, the process debugging period and the abnormal shutdown risk are reduced, and the method is suitable for intelligent production control of the high-performance conductive carbon black master batch.

Description

Technical Field

[0001] The invention relates to the technical field of carbon black masterbatch production data processing, and in particular to an automatic control method for carbon black masterbatch production equipment based on PCL. Background Art

[0002] Carbon black masterbatch is a pre-dispersed particle formed by coating carbon black with a carrier resin and dispersing additives. It is mainly used to prepare conductive or antistatic polymer composite materials. Its performance indicators depend on the dispersion of carbon black in the resin matrix and the interfacial bonding strength. Existing carbon black masterbatch production equipment usually consists of a three-roll grinder, a mixer, an internal mixer, a twin-screw extruder, and a pelletizer. The three-roll grinder is a grinding machine that squeezes the surface area of ​​three rollers arranged horizontally to break up the pigment particle agglomerates in the color paste and produce primary particles; the mixer can preliminarily mix the pigment processed by the three-roll grinder with the resin carrier; the pigment and carrier preliminarily mixed by the mixer are mixed and plasticized by the internal mixer to make semi-finished masterbatch. The twin-screw extruder completes melt blending and dispersion enhancement, and the pelletizer performs granulation molding.

[0003] The production process of carbon black masterbatch involves the dynamic coupling of multi-dimensional process parameters such as mixing temperature, extrusion pressure, feeding rate and screw speed. Although the existing equipment control system has the function of data acquisition, it has the technical defect of insufficient fusion of multi-source heterogeneous data. Specifically, the thermodynamic parameters of the internal mixer (temperature gradient, torque value) and the rheological parameters of the twin-screw extruder (melt pressure, shear rate) belong to independent databases with different sampling frequencies and data structures, which makes it difficult to build a dynamic correlation model between process parameters. Taking the control of carbon black dispersion uniformity as an example, the particle size distribution data of carbon black agglomerates in the internal mixing stage and the melt rheological curve data in the extrusion stage failed to establish a real-time mapping relationship, resulting in the inability to adaptively adjust the rotor speed of the internal mixer according to the feedback of the extruder melt pressure. This data processing breakpoint directly leads to an increase in the probability of secondary agglomeration of carbon black, which expands the fluctuation range of the material volume resistivity to 1-2 orders of magnitude. The existing system lacks a time series alignment and feature parameter extraction mechanism for cross-process data, which seriously restricts the closed-loop control accuracy of the carbon black dispersion process. Summary of the invention

[0004] In view of the shortcomings of the existing technology, the present invention provides an automatic control method for carbon black masterbatch production equipment based on PCL, which is used to solve the technical pain points of the existing carbon black masterbatch production equipment, such as poor carbon black dispersion uniformity, low conductivity threshold control accuracy and limited process energy efficiency optimization due to insufficient multi-source heterogeneous data fusion and lack of dynamic coupling of cross-process parameters.

[0005] In order to solve the above technical problems, the specific technical solutions of the present invention are as follows: The present invention provides an automatic control method for a PCL-based carbon black masterbatch production device, comprising: Collect heterogeneous data streams of mixing equipment, extrusion equipment and detection equipment in real time, perform time domain alignment processing on the heterogeneous data streams to generate a synchronous process parameter sequence, and extract frequency domain characteristics and waveform characteristics of equipment operation parameters; A multi-dimensional production task simulator including raw material ratio parameters is constructed based on the equipment operation parameters, simulation data sets of different ratio scenarios are generated through digital twin technology, and a hierarchical model knowledge base is established according to the hierarchical relationship between raw material ratios, process parameters and quality indicators; When receiving a new production task, parsing the task parameters and matching the candidate control model from the hierarchical model knowledge base through a dynamic time warping algorithm based on the current equipment state characteristics; Perform multi-objective optimization sorting on the candidate control models according to the quality indicator priorities, and convert key parameters of the selected models into cross-device collaborative control instructions; During the execution of the control instructions, production quality data is collected in real time, and iterative updating of the hierarchical model knowledge base is triggered based on the quality deviation detection results.

[0006] Furthermore, in the automatic control method of the PCL-based carbon black masterbatch production equipment of the present invention, the time domain alignment processing includes: A sliding window mechanism is used to compensate the sampling frequency of the asynchronous data streams of the mixing device and the extrusion device; Outlier detection is performed on the compensated data sequence to generate a set of process parameters with continuously aligned timestamps.

[0007] Furthermore, the PCL-based carbon black masterbatch production equipment automatic control method of the present invention, the construction of a multi-dimensional production task simulator comprises: Establish the constraint relationship model between raw material ratio parameters and physical boundaries of extrusion equipment; The constraint relationship model is used to drive the digital twin system to simulate the material dispersion process and generate a training data set including a mixing temperature and torque correlation curve.

[0008] Furthermore, in the PCL-based carbon black masterbatch production equipment automatic control method of the present invention, the establishment of the hierarchical model knowledge base includes: Divide the basic scenario model by raw material ratio type as the first-level index; Storing the corresponding process parameter optimization combination under the basic scenario model as a secondary index; The process parameter combinations are associated with quality detection thresholds to form a three-level quality assessment standard.

[0009] Furthermore, in the automatic control method of the PCL-based carbon black masterbatch production equipment of the present invention, the dynamic time warping algorithm includes: Perform Fourier transform encoding on the real-time collected equipment preheating status data; Calculate the dynamic time warping distance between the encoded feature vector and the feature matrix of the knowledge base model; Screening models whose bending distance is less than a preset threshold as a set of candidate control models; Verify the compatibility of equipment for the selected candidate control models; When it is detected that the current device hardware version does not match the model requirements, the knowledge base physical boundary parameter update is automatically triggered.

[0010] Furthermore, in the PCL-based carbon black masterbatch production equipment automatic control method of the present invention, the multi-objective optimization sorting includes: Construct a weighted evaluation function including quality qualification rate, energy consumption coefficient and equipment load rate; Dynamically adjust the weight distribution ratio of the evaluation function according to the production task type; The foreign body detection enhancement model is forced to load for high carbon black ratio scenarios.

[0011] The technical necessity of forcibly loading the foreign body detection enhancement model in the high carbon black ratio scenario stems from the nonlinear amplification effect of the carbon black high filling system on the sensitivity of foreign matter. When the mass fraction of carbon black exceeds 25%, its surface can adsorb resin molecular chains to form a dense coating layer, causing the melt transmittance to drop below the signal-to-noise ratio critical value of the conventional visual inspection system, and the existing grayscale threshold segmentation algorithm fails to extract the morphological features of metal chips and gel groups. At this time, it is necessary to introduce the joint feature analysis of melt rheological parameters and dielectric constants: by high-frequency sampling of the power spectral density function of the melt pressure fluctuation signal, the characteristic frequency band corresponding to the local flow field distortion caused by foreign matter is captured (usually distributed in the range of 10-50Hz), combined with the abnormal offset of the dielectric loss factor detected by the broadband dielectric spectrometer, a multimodal foreign body classifier based on a support vector machine is constructed. At the same time, the formation of carbon black percolation network makes the resistivity response of the system to conductive foreign matter show exponential change characteristics. Metal particles with a volume fraction of 0.1% can cause percolation threshold drift. Dynamic differential resistance monitoring technology is required to locate the local conductive channel variation area caused by foreign matter through real-time spatial electric field distribution reconstruction of the four-probe array. The enhanced model further integrates the torque harmonic component analysis module of the internal mixer. When foreign matter causes a sudden change in the rotor instantaneous load, the third harmonic amplitude phase characteristics of the torque signal are extracted and dynamically time-regulated with the foreign matter mechanical response map pre-stored in the knowledge base, triggering the collaborative control logic of the feeding system emergency stop and reverse clearing program, thereby maintaining process stability and product performance consistency.

[0012] Furthermore, the PCL-based carbon black masterbatch production equipment automatic control method of the present invention, the conversion into cross-device collaborative control instructions includes: Generate a dispersion disk linear speed adjustment instruction according to the torque spectrum characteristics of the mixing device; Calculate the screw speed compensation amount based on the extruder pressure balance model and generate a speed correction instruction; The speed gradient parameters of the pelletizing equipment are dynamically set according to the particle size distribution characteristics.

[0013] Furthermore, in the PCL-based carbon black masterbatch production equipment automatic control method of the present invention, the real-time collection of production quality data includes: A multidimensional quality assessment matrix including specific surface area, DBP oil absorption value, volume resistivity and specific energy consumption was constructed; When consecutive batches of data in the evaluation matrix exceed an associated quality threshold, retraining of the hybrid temperature compensation model is triggered.

[0014] Furthermore, in the automatic control method of the PCL-based carbon black masterbatch production equipment of the present invention, the iterative update includes: When unrecorded foreign body features are detected, the foreign body detection graph dimension in the knowledge base is expanded; Initiate the digital twin re-simulation process for scenario models whose matching failure rate exceeds the threshold; Establish a traceable process database including historical optimal model versions; The traceable process database includes: Store the equipment firmware information and raw material batch data corresponding to each historical version model; Provides comparative analysis of production process parameters based on the time dimension.

[0015] Furthermore, the PCL-based carbon black masterbatch production equipment automatic control method of the present invention further includes: Establish a closed-loop feedback mechanism for buffering material level between the discharge valve of the mixing equipment and the feed port of the extrusion equipment; When abnormal material transmission continuity is detected, the timing relationship between the feeding rate and the opening of the discharge valve is adaptively adjusted.

[0016] Beneficial effects of the present invention: The present invention realizes the time domain alignment and feature fusion of asynchronous data streams of equipment such as internal mixers and extruders through the sliding window mechanism and frequency domain feature extraction technology, eliminating the problem of missing cross-process parameter coupling caused by data delay in existing independent PLC control systems. By building a unified data characterization framework, the system can accurately capture the dynamic correlation between the torque fluctuation of the internal mixer and the melt pressure of the extruder, and then generate cross-device collaborative control instructions such as the linear speed of the dispersion disk and the screw speed. This technology chain effectively suppresses the secondary agglomeration of carbon black, significantly improves the dispersion uniformity of carbon black, and lays the foundation for the stable regulation of conductive properties.

[0017] The multi-dimensional production task simulator based on digital twin technology, combined with the dynamic matching mechanism of the hierarchical model knowledge base, breaks through the limitations of existing processes that rely on manual experience and trial and error. The system predicts the safety boundaries of process parameters through constraint modeling of raw material ratios and equipment physical boundaries; at the same time, it uses the rapid call of historical optimal parameter combinations and multi-objective optimization sorting to achieve rapid production of new formulas and precise parameter adaptation. For example, for high carbon black ratio scenarios, the system automatically loads the foreign matter detection enhancement model and melt pressure compensation strategy, significantly shortening the process debugging cycle and improving product consistency.

[0018] Through the closed-loop feedback mechanism of the real-time quality assessment matrix and the traceable process database, the system has dynamic iteration and self-optimization capabilities. When quality deviation or abnormal equipment status is detected, the hybrid temperature compensation model retraining and digital twin re-simulation process is triggered to enable the process parameters to continuously adapt to raw material fluctuations and equipment aging. At the same time, the buffer level closed-loop feedback mechanism suppresses the risk of material breakage or overflow through dynamic adjustment of the material transmission sequence to ensure production continuity. This technical solution greatly reduces the probability of abnormal downtime, extends the service life of equipment, and provides reliable guarantee for the efficient and stable production of carbon black masterbatch. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, for ordinary technicians in this field, other drawings can be obtained based on the drawings without paying any creative labor.

[0020] Figure 1 A flow chart of an automatic control method for a PCL-based carbon black masterbatch production equipment provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described in conjunction with the specific embodiments of the present invention and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. The technical solutions provided by the embodiments of the present invention are described in detail below in conjunction with the drawings.

[0022] In order to better understand the purpose of the present invention, the present invention is described in further detail below.

[0023] like Figure 1 As shown, the present invention provides an automatic control method for a carbon black masterbatch production device based on PCL, comprising: Step S101, collecting heterogeneous data streams of a mixing device, an extrusion device, and a detection device in real time, performing time domain alignment processing on the heterogeneous data streams to generate a synchronous process parameter sequence, and extracting frequency domain characteristics and waveform characteristics of the equipment operation parameters; In order to solve the problem of data sampling frequency difference between mixing equipment (such as internal mixer) and extrusion equipment (such as twin-screw extruder), a sliding window mechanism is used to resample and compensate for the asynchronous data stream. For example, the torque data collected once per second by the internal mixer and the melt pressure data collected 10 times per second by the extruder are aligned in time axis through window overlap interpolation.

[0024] Perform outlier detection based on statistical distribution (such as the 3σ principle) on the synchronized data sequence to eliminate abnormal values ​​caused by sensor noise or transient fluctuations of equipment, and generate a continuous and stable process parameter sequence.

[0025] The frequency domain energy distribution characteristics of the equipment parameters are extracted through fast Fourier transform (FFT), and the waveform decomposition technology is used to capture the morphological characteristics of the parameter curve, such as the steep change points and periodicity, to provide a multi-dimensional feature vector for subsequent model matching.

[0026] This step solves the problem of time axis misalignment and missing feature dimensions of multi-source data in existing independent PLC control systems through time domain alignment and feature extraction. For example, the dynamic correlation between the torque fluctuation of the internal mixer and the melt pressure of the extruder requires precise synchronization of time series data to establish an effective model, thereby avoiding the control lag caused by data delay during the carbon black dispersion process.

[0027] Step S102, constructing a multi-dimensional production task simulator including raw material ratio parameters based on the equipment operation parameters, generating simulation data sets of different ratio scenarios through digital twin technology, and establishing a hierarchical model knowledge base according to the hierarchical relationship between raw material ratios, process parameters and quality indicators; The physical boundary constraint model of the extrusion equipment is established based on the raw material ratio parameters (such as carbon black content and resin type). For example, in the scenario of high carbon black ratio, the upper limit of the screw speed needs to be limited to avoid exceeding the melt pressure limit.

[0028] When simulating the material dispersion process in the digital twin system, the mixing temperature gradient and the torque change curve are correlated and modeled to generate a training data set including the predicted value of carbon black dispersion.

[0029] The hierarchical model knowledge base is organized in a three-level structure: the first-level index is classified by raw material ratio (such as PA6-based / PS-based), the second-level index stores the corresponding process parameter combinations (such as mixing temperature range, screw speed ratio), and the third-level index is associated with the quality detection threshold (such as volume resistivity ≤1×10^3 Ω·cm).

[0030] Through the hierarchical modeling of raw materials, process and quality, the limitations of the existing empirical setting of process parameters have been broken through. For example, when the production task is switched to a new conductive masterbatch, the system can automatically match the process parameter combination of historical similar ratio scenarios according to the carbon black content, avoiding repeated trial and error and significantly shortening the process debugging cycle.

[0031] Step S103, when receiving a new production task, parsing the task parameters and matching the candidate control model from the hierarchical model knowledge base through a dynamic time warping algorithm based on the current equipment state characteristics; Dynamic time warping (DTW) matching: Perform Fourier encoding on the temperature rise curve of the equipment preheating phase, calculate its dynamic bending distance with the feature matrix of the knowledge base model, and screen out candidate models with similar morphology.

[0032] Step S104, performing multi-objective optimization sorting on the candidate control models according to the quality index priorities, and converting the key parameters of the selected models into cross-device collaborative control instructions; Multi-objective optimization: When constructing the weighted evaluation function, a higher weight is given to the quality pass rate of high-precision conductive masterbatch tasks, while the focus is on optimizing the energy consumption coefficient for cost reduction tasks.

[0033] Step S105 , collecting production quality data in real time during the execution of the control instruction, and triggering iterative update of the hierarchical model knowledge base based on the quality deviation detection result.

[0034] Instruction conversion and iteration: The optimized screw speed compensation is converted into PID control parameters that can be executed by the equipment; when the online detection of carbon black dispersion deviation exceeding ±5% is detected, the re-simulation and update of the corresponding process parameter combination in the knowledge base is triggered.

[0035] This closed-loop control mechanism achieves autonomous optimization from data to decision-making. For example, when the extruder feeding rate suddenly changes, causing melt pressure fluctuations, the system quickly identifies the anomaly through real-time pressure feedback, dynamically adjusts the synergistic relationship between the mixer rotor speed and the extruder screw speed, suppresses the secondary agglomeration of carbon black, and ensures the stability of dispersion uniformity.

[0036] Specifically, in the PCL-based carbon black masterbatch production equipment automatic control method of the present invention, the time domain alignment processing includes: A sliding window mechanism is used to compensate the sampling frequency of the asynchronous data streams of the mixing device and the extrusion device; Outlier detection is performed on the compensated data sequence to generate a set of process parameters with continuously aligned timestamps.

[0037] The use of sliding window mechanism for sampling frequency compensation includes the following technical contents: Asynchronous data stream processing: In view of the difference in sampling frequency between mixing equipment (such as internal mixer) and extrusion equipment (such as twin-screw extruder) (for example, 1 time per second for internal mixer and 10 times per second for extruder), a sliding window mechanism is used to interpolate and compensate for the low-frequency data stream. For example, through linear interpolation or cubic spline interpolation, the torque data of the internal mixer is expanded on the time axis to the same sampling density as the melt pressure data of the extruder.

[0038] Window overlap strategy: Set the window overlap ratio (such as 50%) to achieve smooth data transition between adjacent windows and avoid the loss of key features due to data truncation. For example, in the critical stage of torque mutation in the internal mixer, the complete waveform features can be retained by overlapping windows.

[0039] Timestamp alignment: The interpolated data stream is timestamped to associate the process parameters of the mixing equipment and the extrusion equipment under a unified time base. For example, the torque data of the 5th second of the internal mixer is accurately matched with the melt pressure data of the 5th second of the extruder.

[0040] In existing independent PLC control systems, the data of mixing equipment and extrusion equipment are misaligned in time axis due to sampling frequency differences, making it impossible to establish a dynamic correlation model. For example, the coordinated adjustment of the torque change of the internal mixer and the melt pressure of the extruder requires synchronous time series data. The sliding window mechanism eliminates data delays between devices through interpolation compensation and time calibration, providing a unified time benchmark for subsequent process parameter correlation analysis, thereby improving the real-time performance of cross-device control.

[0041] Performing outlier detection to generate an alignment parameter set includes the following technical contents: Outlier detection method: The 3σ principle (triple standard deviation method) based on statistical distribution is used to identify abnormal data. For example, the mean and standard deviation of the torque series of the internal mixer are calculated, and data points that are beyond the mean ±3σ range are eliminated.

[0042] Dynamic threshold adjustment: Dynamically adjust the outlier threshold according to the equipment operation stage (such as preheating and steady-state production). For example, a larger fluctuation range is allowed during the equipment startup phase to avoid misjudging transient data as outliers.

[0043] Data repair mechanism: The removed outliers are filled with linear interpolation of the previous and next data or the average value of adjacent windows to improve the continuity of the time series. For example, the torque anomaly of the internal mixer is replaced by the sliding average of the previous and next 1 second data.

[0044] Outlier detection solves the problem of abnormal data interference caused by sensor noise or transient fluctuations of equipment (such as voltage mutations and mechanical vibrations). For example, if the melt pressure sensor of the extruder produces abnormal peaks due to instantaneous blockage, directly using such data will mislead the optimization of process parameters. Through dynamic thresholds and data repair, a continuous and stable set of process parameters is generated, providing high-quality input for subsequent feature extraction and model matching, thereby improving the reliability of control instructions.

[0045] Specifically, the PCL-based carbon black masterbatch production equipment automatic control method of the present invention, the construction of a multi-dimensional production task simulator includes: Establish the constraint relationship model between raw material ratio parameters and physical boundaries of extrusion equipment; The constraint relationship model is used to drive the digital twin system to simulate the material dispersion process and generate a training data set including a mixing temperature and torque correlation curve.

[0046] Raw material parameter mapping: Establish quantitative correlation between the ratio parameters such as carbon black content and resin melt index and the physical boundaries of the extrusion equipment (such as the maximum screw speed and the upper limit of the melt pressure). For example, at high carbon black content (>30%), the melt viscosity increment is derived through the rheological model, and the screw speed is dynamically limited to avoid melt fracture.

[0047] Material property modeling: Based on the rheological property curve of the resin type (such as PA6, PS), the shear rate-viscosity relationship model at different temperatures is established to define the temperature setting range of each section of the extruder. For example, high-viscosity resins need to reduce the temperature of the melt conveying section to prevent local overheating.

[0048] Equipment capacity matching: Based on the screw geometry parameters (aspect ratio, screw groove depth) and drive power, a physical constraint model of screw speed-torque-capacity is constructed to ensure that the simulation parameters are within the actual load range of the equipment.

[0049] In the existing process, the matching of raw material ratio and equipment parameters depends on manual experience, which can easily lead to parameter over-limit (such as melt pressure overpressure alarm). The present invention realizes automatic prediction of parameter safety boundaries by quantifying the relationship between raw material characteristics and equipment physical boundaries. For example, when the carbon black content in the formula increases, the system automatically calculates the increase in melt viscosity and dynamically adjusts the upper limit of the extruder screw speed to avoid a decrease in dispersion or equipment overload due to insufficient shear force.

[0050] Multi-physics coupling simulation: In the digital twin system, the thermodynamic model (internal mixer temperature field distribution) and the fluid dynamics model (extruder melt flow) are coupled to simulate the dispersion process of carbon black particles in the resin. For example, the influence of the internal mixer rotor speed on the efficiency of carbon black agglomerate crushing is simulated to generate a torque-dispersion correlation curve.

[0051] Dynamic boundary condition loading: The equipment parameter limits (such as maximum screw speed) output by the constraint relationship model are used as boundary conditions for digital twin simulation to make the simulation scenario consistent with the actual equipment capabilities. For example, when simulating a high carbon black ratio, the screw speed is forced to not exceed the equipment safety threshold.

[0052] Feature data extraction: Extract key parameters such as mixing temperature gradient and torque fluctuation frequency from the simulation results to build a related data set including process parameters and quality indicators. For example, record the torque steady-state fluctuation range when the mixing temperature reaches the set value as a reference feature for the dispersion to meet the standard.

[0053] The existing trial and error method requires a lot of materials and time to verify process parameters, while the present invention rehearses the production process through digital twin technology, significantly shortening the process debugging cycle. For example, for the new conductive masterbatch formula, the system automatically loads the corresponding raw material rheological parameters, simulates the dispersion state of carbon black at different mixing temperatures, and outputs the optimal temperature-speed combination. This process not only avoids the risk of parameter overruns in actual production, but also can predict in advance whether the dispersion uniformity meets the standard, providing high-confidence training data for the knowledge base.

[0054] Specifically, in the PCL-based carbon black masterbatch production equipment automatic control method of the present invention, the establishment of the hierarchical model knowledge base includes: Divide the basic scenario model by raw material ratio type as the first-level index; Storing the corresponding process parameter optimization combination under the basic scenario model as a secondary index; The process parameter combinations are associated with quality detection thresholds to form a three-level quality assessment standard.

[0055] Basic scenario models divided by raw material ratio (first-level index): Raw material classification rules: Divide the basic scenarios according to the resin matrix type (such as PA6, PS) and the carbon black content range (such as 5%-10%, 10%-20%). For example, define "PA6-based-carbon black 15%" as an independent scenario model.

[0056] Scene feature extraction: Extract key features for each raw material ratio scene, including resin melt index, carbon black specific surface area, additive type, etc., to form a unique identification code for the scene.

[0057] Index storage structure: Use hash tables or tree structures to store scenario models, and use matching feature codes to achieve fast retrieval. For example, when entering a PA6-based formula with a carbon black content of 18%, the system automatically matches the "PA6-based-carbon black 15%-20%" scenario branch.

[0058] In the existing process, the empirical correspondence between raw material ratios and process parameters is scattered and unstructured, resulting in a long debugging cycle for new formulas. Through the scientific classification and feature coding of raw material ratios, the system can quickly locate similar historical scenarios. For example, when the production task is switched to "PS-based-carbon black 25%", the optimized parameter combination under the corresponding scenario is directly called to avoid repeated trial and error.

[0059] Storage process parameter optimization combination (secondary index): Parameter combination generation: Based on digital twin simulation data and historical production data, select parameter combinations such as mixing temperature range, screw speed ratio, and feeding rate that meet quality indicators. For example, for the "PA6-based-carbon black 15%" scenario, store the optimized combination of mixing temperature 180-190°C and screw speed 60-70rpm.

[0060] Parameter association constraints: Establish logical association rules between parameters, such as the linkage between the torque upper limit of the internal mixer and the melt pressure threshold of the extruder, to avoid storing invalid or conflicting parameter combinations.

[0061] Compression storage technology: Use parameter matrix or feature vector to compress and store process parameters. For example, normalize parameters such as temperature, speed, and pressure and encode them into multi-dimensional vectors to reduce storage redundancy.

[0062] The existing parameter settings rely on manual experience, and it is difficult to ensure the coordination of multi-device parameters. By storing verified optimized parameter combinations, the system can directly call cross-device collaborative control instructions. For example, for high carbon black ratio scenarios, the system automatically loads the "high temperature mixing + low screw speed" combination to inhibit the secondary agglomeration of carbon black and improve the uniformity of dispersion.

[0063] Association quality detection threshold (three-level evaluation criteria): Quality index mapping: Combine process parameters with specific surface area (NSA ≥ 200 10 3 cm3 / g), volume resistivity (≤1×10³ Ω·cm), and other quality detection thresholds are dynamically bound. For example, when the volume resistivity detection value corresponding to parameter combination A exceeds the threshold, it is automatically marked as "needs optimization".

[0064] Dynamic threshold adjustment: Adaptively update the quality threshold according to raw material batch fluctuations (such as changes in carbon black particle size distribution). For example, when the carbon black D50 particle size increases by 5%, the blackness value threshold is automatically relaxed to NSA ≥ 180 10 3 cm 3 / g.

[0065] Quality assessment model: Build a decision tree or logistic regression model to determine whether the current parameter combination meets the quality requirements. For example, when the mixing temperature is lower than the set range, the "insufficient dispersion" warning is triggered and a temperature increase compensation strategy is recommended.

[0066] Existing quality control relies on post-testing and cannot correlate process parameters with quality results in real time. Through the three-level quality assessment standard, the system can predict quality risks during the production process. For example, when it is detected that the melt pressure of the extruder exceeds the threshold associated with the current parameter combination, the screw speed is immediately adjusted to avoid substandard conductivity, thus achieving "parameter-quality" closed-loop control.

[0067] Specifically, the PCL-based carbon black masterbatch production equipment automatic control method of the present invention, the dynamic time warping algorithm includes: Perform Fourier transform encoding on the real-time collected equipment preheating status data; Calculate the dynamic time warping distance between the encoded feature vector and the feature matrix of the knowledge base model; Screening models whose bending distance is less than a preset threshold as a set of candidate control models; Verify the compatibility of equipment for the selected candidate control models; When it is detected that the current device hardware version does not match the model requirements, the knowledge base physical boundary parameter update is automatically triggered.

[0068] Fourier transform encoding of equipment preheating data: Segmented feature extraction: The time series data of the equipment preheating stage (such as internal mixer heating and extruder heating) is divided into sub-stages such as initial heating and steady-state maintenance according to the temperature change trend. Fourier transform is performed on the data of each stage separately to extract the frequency domain energy distribution characteristics. For example, the proportion of low-frequency components of torque fluctuations when the internal mixer rotor is started is analyzed to characterize the mechanical load characteristics.

[0069] Feature dimensionality reduction encoding: Sort the spectrum data after Fourier transformation by energy intensity, retain the first N principal components (such as the frequency band with energy accounting for more than 90%), and generate a compact feature vector. For example, the high-frequency noise in the preheating stage of the internal mixer is filtered out, and only the low-frequency features reflecting the friction characteristics of the bearing are retained.

[0070] Existing methods directly use time domain data to match models, which are susceptible to time expansion of the equipment preheating curve (such as slower temperature rise in winter) or transient noise interference. Fourier transform effectively captures the essential state of the equipment preheating process (such as enhanced low-frequency vibration caused by bearing wear) through frequency domain energy feature extraction, improving the robustness of subsequent model matching.

[0071] Dynamic time warping distance calculation: Non-aligned sequence processing: In view of the difference in equipment preheating time (such as the different heating efficiencies of new and old equipment), the dynamic time warping (DTW) algorithm is used to nonlinearly align the feature vectors to eliminate the deviation caused by time axis expansion and contraction. For example, the 30-second preheating curve of the current internal mixer is matched with the 25-second historical curve in the knowledge base.

[0072] Similarity quantification: Calculate the cumulative distance of the aligned sequences, integrate the differences in frequency domain features (such as the main frequency offset) and time series morphology (such as the peak position offset), and generate a multi-dimensional similarity score.

[0073] The DTW algorithm solves the existing Euclidean distance's dependence on strict alignment of time series. For example, the change in heating rate caused by the difference in ambient temperature during the extruder preheating process will not affect the accurate judgment of the "preheating completion status", thereby improving the accuracy of candidate model screening.

[0074] Candidate control model screening: Dynamic threshold setting: Dynamically adjust the bending distance threshold according to the type of production task (such as high-precision conductive masterbatch or general masterbatch). For example, a more stringent threshold is used for conductive masterbatch tasks to prioritize quality, while a more relaxed threshold can be used for general masterbatch tasks to improve efficiency.

[0075] Priority sorting: Arrange candidate models in ascending order by bending distance, and add weights of additional dimensions such as quality pass rate and equipment load rate to generate a comprehensive sorting list.

[0076] Existing screening methods rely only on a single distance metric, which easily ignores the multi-objective trade-offs in actual production. Through dynamic thresholds and multi-dimensional sorting, the system can take into account quality, efficiency, and equipment life. For example, when the bending distances of multiple models are similar, the parameter combination with the lowest historical energy consumption is preferred.

[0077] Device compatibility verification: Hardware parameter verification: Compare the compatibility of the device firmware version, sensor type (such as pressure sensor range), drive motor power and other parameters stored in the candidate model with the current device. For example, when filtering out models that are only suitable for servo motor drives, if the current device is an asynchronous motor, it will be automatically eliminated.

[0078] Physical boundary constraints: Verify whether the model parameters (such as the upper limit of the screw speed and the torque threshold of the internal mixer) exceed the mechanical load capacity of the current equipment. For example, when the model requires the extruder speed to be 120rpm and the current equipment supports a maximum of 100rpm, an alarm is triggered and the model is excluded.

[0079] Compatibility verification prevents the risk of "paper optimization". For example, old extruders cannot withstand high-speed models due to screw material limitations, and directly loading such models may cause equipment damage. The feasibility of control instructions is improved through dual verification of hardware parameters and physical boundaries.

[0080] Knowledge Base Physical Boundary Updates: Boundary parameter recalculation: When an equipment hardware upgrade is detected (such as replacement of a high-power motor), the digital twin system is called to re-simulate the physical boundaries (such as maximum screw speed, upper limit of melt pressure) and update the constraints of the associated model in the knowledge base.

[0081] Incremental model iteration: Only the scenario models affected by hardware changes (such as high-speed related models) are retrained, and the original parameter combinations of other scenarios are retained to reduce computing resource consumption.

[0082] The existing system requires manual maintenance of the matching relationship between equipment parameters and models, and the response is delayed. The present invention automatically triggers boundary updates to synchronize the knowledge base with the equipment status in real time. For example, after the extruder is upgraded, the system automatically removes the original speed limit and loads the optimized model adapted to the new hardware without stopping production for debugging.

[0083] The dynamic time warping algorithm solves the contradiction between device state drift and model solidification through the technical chain of state encoding → flexible matching → compatibility verification → closed-loop update: Anti-interference matching: frequency domain feature extraction and DTW algorithm eliminate timing deviation and improve model screening accuracy; Safety control guarantee: compatibility check prevents equipment from running beyond the limit; Enhanced adaptability: Dynamic updates of physical boundaries enable the knowledge base to continuously adapt to device upgrades.

[0084] Specifically, the PCL-based carbon black masterbatch production equipment automatic control method of the present invention, the multi-objective optimization sorting includes: Construct a weighted evaluation function including quality qualification rate, energy consumption coefficient and equipment load rate; Dynamically adjust the weight distribution ratio of the evaluation function according to the production task type; The foreign body detection enhancement model is forced to load for high carbon black ratio scenarios.

[0085] Construct a weighted evaluation function: Quantification of quality pass rate: Real-time calculation of pass rate weight based on online detection data (such as specific surface area, DBP oil absorption value, volume resistivity), for example, mapping the volume resistivity deviation range (±10%) to the quality score.

[0086] Energy consumption coefficient modeling: Associate equipment power data with unit output energy consumption to establish a dynamic energy consumption model. For example, for every 10% increase in the extruder melt pressure, the system automatically calculates the corresponding energy consumption increment and converts it into an energy consumption coefficient.

[0087] Equipment load rate assessment: The equipment load status is estimated in real time through parameters such as motor current and bearing temperature. For example, the ratio of screw speed to rated power is used as a load rate indicator.

[0088] Existing process optimization often only pursues quality or energy consumption indicators, resulting in equipment overload or efficiency loss. Through the weighted evaluation function, the system can dynamically balance quality, energy efficiency and equipment life. For example, when the quality score is close to the threshold, the system automatically reduces the energy consumption weight to prioritize product qualification rate; when the equipment load rate exceeds the limit, it focuses on reducing the speed to extend the equipment life.

[0089] Dynamically adjust weight distribution: Task type identification: parse the task labels in the production order (such as "high-precision conductive grade" and "general cost reduction type") and call the preset weight template. For example, the quality weight in the conductive grade task is set to 70%, while the energy consumption weight in the cost reduction task is increased to 60%.

[0090] Real-time working condition adaptation: fine-tune the weight ratio according to the equipment status (such as the degree of wear of the extruder screw). For example, when the screw wear exceeds the limit, the load rate weight is reduced to avoid further wear.

[0091] Priority arbitration mechanism: When multiple objectives conflict (such as increasing energy consumption to improve quality), the Pareto optimal solution is selected based on the optimization results of historical data.

[0092] Dynamic weight allocation solves the problem that fixed weight models cannot adapt to diversified production needs. For example, when producing at night when electricity prices are low, the system can temporarily increase the energy consumption weight to give priority to low-energy consumption processes, while when rushing to produce high-value orders, it will give priority to ensuring quality stability to achieve the optimal allocation of resources.

[0093] Force loading of foreign body detection enhancement model: High carbon black scene recognition: automatically trigger the enhanced detection mode through the raw material ratio parameters (carbon black content ≥ 25%).

[0094] Extension of detection dimensions: On the basis of conventional visual inspection, the analysis of abnormal melt rheological parameters (such as pressure mutation detection) and infrared spectral foreign body feature matching are added. For example, when non-periodic fluctuations in the melt pressure curve are detected, the linked visual system focuses on the suspected contaminated area.

[0095] Dynamic model loading: Call pre-trained foreign body detection models (such as metal shavings and gel particle feature libraries) from the knowledge base to improve detection sensitivity and response speed.

[0096] In scenarios with high carbon black ratios, carbon black particles can easily cover up foreign matter or form false agglomerates, which are easily missed by existing detection. By strengthening the multi-dimensional analysis of the model, the system can distinguish between real dispersion defects and foreign matter interference. For example, when the difference in rheological properties between carbon black agglomerates and metal chips is identified, the system automatically adjusts the speed of the internal mixer to break up the agglomerates, and at the same time triggers the foreign matter removal mechanism to avoid abnormal conductivity.

[0097] Multi-objective optimization sorting systematically solves the problem of goal conflicts in process optimization through dynamic trade-offs and scenario enhancement: Comprehensive decision-making ability: weighted evaluation function realizes the coordinated optimization of quality, energy consumption and equipment status, avoiding suboptimal solutions caused by a single goal; Flexible adaptation: The dynamic weight adjustment mechanism adapts to diverse production needs and improves process generalization capabilities; Risk prevention: The enhanced foreign body detection model in high carbon black scenarios proactively intercepts quality risks and ensures product consistency.

[0098] Specifically, the PCL-based carbon black masterbatch production equipment automatic control method of the present invention, the conversion into cross-device collaborative control instructions includes: Generate a dispersion disk linear speed adjustment instruction according to the torque spectrum characteristics of the mixing device; Calculate the screw speed compensation amount based on the extruder pressure balance model and generate a speed correction instruction; The speed gradient parameters of the pelletizing equipment are dynamically set according to the particle size distribution characteristics.

[0099] Disperser disc linear speed regulation based on torque spectrum: Spectral feature analysis: Perform fast Fourier transform (FFT) on the torque time series data of mixing equipment (such as internal mixer) to extract the energy distribution of low frequency band (0-5Hz) and high frequency band (5-20Hz). When the proportion of low frequency energy is high, it indicates that the carbon black agglomerates are not fully broken, and the linear speed of the dispersion disk needs to be increased to enhance the shear strength.

[0100] Dynamic mapping strategy: Establish the association rules between the energy proportion of torque frequency band and line speed. For example, when the low-frequency energy proportion exceeds 50%, the line speed is increased by 5%-10% according to the gradient; if the high-frequency energy proportion continues to rise, the line speed is fine-tuned to maintain the dispersion stability.

[0101] Closed-loop feedback control: According to the online dispersion test results of the extruder inlet material (such as the specific surface area fluctuation range), the linear speed adjustment range is corrected in the reverse direction to suppress the risk of resin degradation caused by over-shearing.

[0102] The existing dispersion disk control relies on fixed speed parameters and cannot adapt to the dynamic changes of carbon black agglomerates. Through torque spectrum analysis, the system can accurately identify the dispersion state: low-frequency energy reflects large-sized agglomerates, and high-frequency energy represents the degree of dispersion refinement. For example, when the low-frequency torque of the internal mixer increases suddenly, the system automatically increases the line speed to enhance the shearing effect and synchronously links the extruder feeding rate to avoid abnormal fluctuations in melt pressure due to uneven dispersion.

[0103] Extruder pressure balance and speed compensation: Pressure dynamic modeling: Based on the extruder melt pressure sensor data, a differential relationship model between the pressure change rate and the screw speed is constructed. For example, when the pressure rise rate exceeds the threshold, the compensation amount that the screw speed needs to be reduced to balance the feeding rate is derived.

[0104] Compensation calculation: Based on the pressure fluctuation direction (up / down) and historical trend, the proportional-integral (PI) control algorithm is used to generate the speed correction instruction. For example, when the melt pressure is detected to exceed the limit for 3 consecutive seconds, the speed compensation is dynamically calculated according to the integral value of the pressure deviation (such as reducing 2-3rpm per minute).

[0105] Multi-section collaboration: The independent pressure data of different sections of the extruder (melting section, mixing section) are modeled separately to generate segmented speed compensation instructions to optimize the uniformity of melt flow.

[0106] The existing screw speed regulation lags behind the change in melt pressure, which can easily cause fluctuations in carbon black dispersion. Through the pressure balance model, the system can predict the pressure trend and compensate in advance. For example, when the feed rate suddenly increases and causes the pressure in the melting section to rise, the system simultaneously reduces the speed of the mixing section to expand the melt residence time window, avoiding secondary agglomeration of carbon black caused by excessive shear rate, thereby stabilizing the volume resistivity index.

[0107] Dynamic setting of the speed gradient of pelletizing equipment: Particle size feature mapping: Obtain masterbatch particle size distribution data (such as D50, D90) through an online particle size analyzer, and establish a correlation model between particle size concentration and pelletizer speed gradient. For example, when the D90 value exceeds 800μm, increase the speed ratio of the high-speed blade group to refine the particles.

[0108] Gradient parameter generation: Set the speed gradient band according to the target particle size range (such as 300-500μm) and dynamically allocate the operating time ratio of the low-speed, medium-speed and high-speed knife groups. For example, when the particle size distribution is detected to be right-biased, extend the working time of the high-speed knife group to reduce the proportion of small particle size.

[0109] Ensure the consistency of material flow: Link the extruder outlet die head pressure monitoring data to adjust the matching relationship between the pelletizer feed rate and the knife speed to prevent abnormal particle morphology caused by sudden changes in material flow rate.

[0110] Existing pelletizing equipment uses fixed speed parameters, which makes it difficult to adapt to the difference in melt viscosity caused by changes in carbon black content. Through particle size distribution feedback, the system can dynamically adjust the knife group speed gradient. For example, when the melt viscosity is high and the fluidity is poor with a high carbon black ratio, the system reduces the proportion of the low-speed knife group to avoid material accumulation, while increasing the speed of the high-speed knife group to make the particles uniform and reduce the energy consumption of the subsequent screening process.

[0111] Cross-device collaborative control instructions systematically solve the process collaboration problem of carbon black masterbatch production through the technical chain of state perception → dynamic modeling → instruction linkage: Improved dispersion uniformity: Torque spectrum analysis enables precise control of dispersion intensity and reduces the probability of secondary agglomeration of carbon black; Melt stability guarantee: The pressure balance model predictively adjusts the screw speed to suppress fluctuations in conductive properties; Optimization of particle size consistency: The pelletizing speed gradient dynamically matches the melt characteristics to reduce the proportion of over-fine or over-coarse particles. The present invention provides a closed-loop optimization framework for the full process automation control of carbon black masterbatch production.

[0112] Specifically, the PCL-based carbon black masterbatch production equipment automatic control method of the present invention, the real-time collection of production quality data includes: Construct a multi-dimensional quality evaluation matrix including black value, volume resistivity and unit energy consumption; When consecutive batches of data in the evaluation matrix exceed an associated quality threshold, retraining of the hybrid temperature compensation model is triggered.

[0113] Construct a multidimensional quality assessment matrix: Multi-source data fusion: real-time collection of volume resistivity and unit energy consumption data through online testing equipment. For example, a four-probe resistivity tester is used to dynamically measure the conductive properties of the masterbatch, and the energy consumption data (kW·h / kg) of the internal mixer and extruder are collected simultaneously.

[0114] Dynamic weight allocation: adjust the weight of quality indicators according to the type of production task. For example, the volume resistivity weight in the conductive masterbatch task is set to 60%.

[0115] Data normalization: Standardize indicators of different dimensions (such as resistivity in Ω·cm and energy consumption in kW·h / kg) to eliminate the impact of magnitude differences on the evaluation results.

[0116] Existing quality inspections rely on offline sampling, which results in data lag. The multidimensional matrix dynamically reflects the production quality status through real-time data fusion. For example, when the melt pressure fluctuations of the extruder cause the dispersion of carbon black to decrease, the correlation between the increase in blackness value and the decrease in resistivity can be captured by the matrix, and the root cause of the process abnormality (such as insufficient mixing temperature or screw speed imbalance) can be quickly located.

[0117] Trigger hybrid temperature compensation retraining: Dynamic threshold setting: The quality threshold is set based on the historical optimal batch data (such as the average value ± 3σ of the first 100 batches), and dynamically adjusted according to the raw material batch characteristics (such as carbon black particle size distribution). For example, when the carbon black D50 particle size increases, the specific surface area ≤ 180 10 3 g / cm 3 .

[0118] Continuous abnormality judgment: When the same quality indicator (such as volume resistivity) exceeds the limit for three consecutive batches, it is judged as a systematic deviation, triggering the retraining of the mixed temperature compensation model.

[0119] Incremental retraining: Combining the current abnormal batch data with the historical optimal data, the digital twin system simulates the temperature compensation strategy (such as increasing the mixing temperature by 2-5°C) to generate an optimization model that adapts to the current raw material characteristics.

[0120] In the existing process, temperature compensation relies on manual experience adjustment, which has a slow response speed and is easy to introduce new errors. The present invention realizes precise closed-loop control through a real-time data-driven retraining mechanism. For example, when the dispersion decreases due to the replacement of carbon black batches, the system automatically identifies the synergistic degradation trend of specific surface area and resistivity, triggers the update of the internal mixing temperature compensation model, and dynamically increases the heating power of the internal mixer to compensate for the changes in raw material characteristics.

[0121] The real-time quality data collection and retraining trigger mechanism solves the pain points of existing quality control through the technical path of multi-dimensional monitoring → dynamic threshold determination → model self-optimization: Closed-loop quality control: The multi-dimensional matrix captures process anomalies in real time to avoid batch quality accidents; Improved adaptability: Incremental retraining enables the temperature compensation model to continuously adapt to raw material fluctuations; Energy efficiency synergistic optimization: unit energy consumption index participates in the evaluation to prevent energy consumption from being out of control due to quality improvement. The present invention provides dual guarantees for the quality stability and process economy of carbon black masterbatch production.

[0122] Specifically, the automatic control method of the PCL-based carbon black masterbatch production equipment of the present invention, the iterative update includes: When unrecorded foreign body features are detected, the foreign body detection graph dimension in the knowledge base is expanded; Initiate the digital twin re-simulation process for scenario models whose matching failure rate exceeds the threshold; Establish a traceable process database including historical optimal model versions; The traceable process database includes: Store the equipment firmware information and raw material batch data corresponding to each historical version model; Provides comparative analysis of production process parameters based on the time dimension.

[0123] Foreign body detection map dimensions: Multimodal feature fusion: When the online detection system (such as visual cameras, metal detectors) captures unrecorded foreign body signals, the morphological features (contour, size), material features (spectral reflectivity) and rheological features (abnormal fluctuation pattern of melt pressure) of the foreign body are extracted to construct a composite feature vector.

[0124] Dynamic update of the spectrum: The new foreign body features are compared with the existing spectrum in the knowledge base for similarity. If the match is lower than the threshold, a new spectrum category is automatically created and associated with a disposal strategy (such as shutdown alarm, pneumatic rejection). For example, when an unknown gel particle is detected, its infrared spectrum features are recorded and updated to the "organic foreign body" spectrum branch.

[0125] Cross-device linkage: The foreign matter feature library is linked with the control parameters of the internal mixer and extruder. When the same foreign matter appears repeatedly, the feeding rate or screw speed is automatically optimized to reduce the probability of foreign matter mixing.

[0126] Existing foreign body detection relies on a fixed feature library and cannot cope with new pollutants (such as new additive residues). By dynamically expanding the atlas dimension, the system can autonomously identify unrecorded foreign body types. For example, when a trace amount of metal chips is mixed into carbon black, the system can quickly locate the source of contamination and update the disposal strategy through the temporal correlation between the sudden change in melt pressure and the metal detection signal, thereby reducing batch quality risks.

[0127] Digital twin re-simulation process trigger: Dynamic monitoring of failure rate: Count the number of matching failures of the knowledge base model in continuous production tasks (for example, 3 out of 10 tasks fail to meet the quality threshold), calculate the failure rate and compare it with the preset threshold (such as 25%).

[0128] Re-simulation parameter reconstruction: Resample the raw material ratio and equipment status data of the failure scenario, adjust the boundary conditions of the digital twin model (such as melt viscosity correction coefficient and screw wear compensation parameter), and generate a training data set suitable for the current working conditions.

[0129] Model incremental training: Only the sub-models associated with the failed scenario are retrained, and the optimized parameters of other scenarios are retained to reduce computing resource consumption. For example, for the matching failure of the "PP-based-carbon black 30%" scenario, only the material dispersion process under this ratio is re-simulated.

[0130] Existing model updates require full retraining, which is inefficient. By triggering local re-simulation based on the failure rate, the system can quickly repair model deviations in specific scenarios. For example, when the original speed model fails due to extruder screw wear, the re-simulation process generates a compensation model based on the current screw gap parameters to restore the stability of the melt pressure.

[0131] Construction of traceable process database: Versioned storage: Store the historical optimal model version and its associated metadata by timestamp, including device firmware version (such as PLC program version number), sensor calibration records, and raw material batch information (carbon black D50 particle size, resin melt index).

[0132] Index optimization: Establish a multi-dimensional index structure to support rapid retrieval of historical process data by time range, raw material type, equipment model, etc. For example, when searching for the task "PA6-based conductive masterbatch in 2023", the optimal mixing temperature parameters for the corresponding period are automatically associated.

[0133] Comparative analysis tool: Provides a timeline sliding comparison function to visualize the correlation and change trends between process parameters and quality indicators in different periods. For example, compare the volume resistivity distribution under the new and old screw speed models to assist in process optimization decisions.

[0134] Existing process data is stored in a scattered manner, making it difficult to trace the source of abnormalities. With a traceable database, technicians can quickly locate historical problems. For example, when the resistivity of a batch of products is abnormal, tracing back to the raw material batch data reveals fluctuations in the specific surface area of ​​carbon black, thereby optimizing the incoming material inspection standards. In addition, the comparative analysis function supports historical verification of process parameters to avoid repetitive trial and error.

[0135] The iterative update mechanism systematically improves the process adaptability through a technical closed loop of dynamic learning → local optimization → data backtracking: Foreign matter prevention and control upgrade: Dynamic expansion of the map enhances the efficiency of identifying and handling unknown pollutants; Model quick repair: Targeted re-simulation of failure scenarios shortens the model iteration cycle by more than 50%; Knowledge accumulation and reuse: The traceable database supports the structured accumulation and rapid call of process optimization experience. The present invention provides closed-loop self-optimization capabilities for continuous process improvement of carbon black masterbatch production.

[0136] Specifically, the PCL-based carbon black masterbatch production equipment automatic control method of the present invention also includes: Establish a closed-loop feedback mechanism for buffering material level between the discharge valve of the mixing equipment and the feed port of the extrusion equipment; When abnormal material transmission continuity is detected, the timing relationship between the feeding rate and the opening of the discharge valve is adaptively adjusted.

[0137] Establishment of closed-loop feedback mechanism for buffer material level: Multi-sensor fusion monitoring: Deploy laser level meters and weighing sensors in the buffer bin to collect material stacking height and quality data in real time, and combine the vibration sensor signal (such as amplitude frequency) at the extruder feed port to comprehensively judge the material level status.

[0138] Dynamic threshold setting: dynamically adjust the material level safety range according to material characteristics (carbon black content, resin fluidity). For example, when the material with high carbon black ratio has poor fluidity, the lower limit of the material level is increased by 10%-15% to prevent material shortage.

[0139] Anti-interference filtering: Perform sliding average filtering and mutation suppression on the level sensor signal to eliminate instantaneous interference caused by material collapse or mechanical vibration.

[0140] In the existing open-loop control, the discharge valve and the feeding equipment operate independently, which can easily lead to empty buffer bins or overflow due to timing deviations. Closed-loop feedback improves the continuity of material transmission through real-time material level monitoring and dynamic threshold adjustment. For example, when it is detected that the buffer bin level is continuously below the lower limit, the system automatically delays the closing time of the discharge valve and increases the extruder feeding rate to maintain the stability of material supply.

[0141] Abnormal detection and adaptive adjustment of material transmission: Abnormal pattern recognition: Analyze the correlation between the material level change rate and the equipment operating parameters. For example, when the discharge valve opening reaches 100% but the material level continues to drop, it is determined that the mixer is under-fed or the feeding rate is too high.

[0142] Timing relationship optimization: establish a timing relationship model between the discharge valve action and the feeding rate, and dynamically adjust the action interval through the proportional-differential (PD) control algorithm. For example, when the feeding rate increases by 20%, the discharge valve opening time is advanced by 0.5 seconds to match the feeding demand.

[0143] Cross-device collaborative control: Link the mixer discharge valve opening and extruder screw speed. When an abnormal level in the buffer bin is detected, the screw speed and discharge valve opening are reduced simultaneously to prevent material accumulation or excessive shearing.

[0144] In existing methods, abnormal material transmission often leads to drastic fluctuations in the melt pressure of the extruder (such as a sudden drop in pressure caused by material cutoff), which affects the uniformity of carbon black dispersion. Through adaptive timing adjustment, the system can quickly respond to abnormal working conditions. For example, when carbon black agglomerates block the discharge valve and cause feed interruption, the system immediately reduces the extruder speed and triggers the reverse clearing procedure, while adjusting the subsequent production rhythm to compensate for downtime and minimize the generation of defective products.

[0145] The buffer material level closed-loop feedback mechanism solves the problem of stability of cross-device material transmission through the technical path of real-time perception → dynamic decision-making → collaborative control: Continuous production guarantee: Dynamic material level monitoring and equipment coordination prevent material breakage or overflow, improving production line operation efficiency; Optimized dispersion uniformity: Stable material supply reduces extruder melt pressure fluctuations and inhibits secondary agglomeration of carbon black; Abnormal rapid response: The timing adaptive adjustment mechanism reduces the abnormal downtime of the equipment by more than 30%. The present invention provides key control guarantee for the efficient and continuous operation of carbon black masterbatch production.

[0146] The technical terms of the present invention are explained as follows: PCL (Programmable Control Logic): Definition: A programmable control architecture based on an industrial automation system, used to achieve logical control and coordinated scheduling of carbon black masterbatch production equipment.

[0147] Application scenario: In the present invention, PCL serves as the core control platform, responsible for multi-source heterogeneous data fusion, cross-device command issuance and real-time feedback processing, replacing the decentralized control mode of existing independent PLC units.

[0148] Dynamic Time Warping (DTW): Definition: An algorithm for matching morphological similarity of time series data, supporting alignment of sequences of different lengths or time axis expansion.

[0149] Application scenario: Used to match the current device preheating curve with the knowledge base historical model, eliminate timing deviations caused by device aging or environmental differences, and improve model screening accuracy.

[0150] Digital Twin Technology: Definition: A technology that maps the state of physical equipment in real time through virtual models, supporting multi-physics simulation and parameter optimization of the production process.

[0151] Application scenario: Construct a virtual simulation environment for the carbon black dispersion process, simulate the rheological behavior of the melt under different raw material ratios, and generate an optimized combination of process parameters.

[0152] Multi-source heterogeneous data fusion: Definition: The process of time alignment and feature extraction of asynchronous, multi-structured data from different equipment (internal mixer, extruder, testing equipment). Application scenario: Unify heterogeneous data such as internal mixer torque and extruder melt pressure into a synchronous process parameter sequence to support dynamic coupling analysis across processes.

[0153] Hierarchical Model Knowledge Base: Definition: A process knowledge storage system organized in a three-level structure of raw material ratio, process parameters, and quality indicators, supporting scenario-based model retrieval and call.

[0154] Application scenario: Quickly match historical optimization parameters according to production task type (such as PA6-based / PS-based) to shorten the process debugging cycle.

[0155] Buffer material level closed loop feedback mechanism: Definition: A closed-loop control method that uses a material level sensor to monitor the material transfer status in real time and dynamically adjust the equipment's action sequence.

[0156] Application scenario: Prevent material accumulation or interruption between mixer unloading and extruder feeding to ensure production continuity.

[0157] Fourier Transform Encoding: Definition: A data processing method that converts time domain signals into frequency domain energy distribution characteristics, which is used to extract the core characteristics of the equipment's operating status.

[0158] Application scenario: Perform frequency domain analysis on the torque fluctuation data of the internal mixer to identify the key frequency band characteristics of the carbon black dispersion state.

[0159] Melt pressure balance model: Definition: A mathematical model that describes the dynamic relationship between the extruder melt pressure, screw speed, and feed rate. Application scenario: Generate screw speed compensation instructions based on real-time pressure change trends to suppress fluctuations in carbon black dispersion.

[0160] Volume Resistivity: Definition: A key indicator that characterizes the electrical conductivity of a material. The unit is Ω·cm. The lower the value, the better the conductivity.

[0161] Application scenario: As the core parameter for carbon black masterbatch quality assessment, it is used for closed-loop optimization of mixing temperature and screw speed.

[0162] Foreign body detection enhanced model: Definition: A pollutant identification and treatment strategy model built based on multimodal data (visual, spectral, rheological).

[0163] Application scenarios: For scenarios with high carbon black ratio, the detection sensitivity and removal efficiency of foreign matter such as metal chips and gel particles can be enhanced.

[0164] Traceable process database: Definition: A structured database that stores historical optimal process models, equipment status, and raw material batch data, supporting comparative analysis in the time dimension.

[0165] Application scenario: Rapidly trace the production parameters of batches with abnormal quality to assist in root cause analysis and process optimization.

[0166] Weighted evaluation function of quality pass rate: Definition: A dynamic scoring model that integrates quality indicators such as blackness value and volume resistivity, and supports adaptive weight adjustment.

[0167] Application scenario: Dynamically optimize process target priority according to different production task types (such as high-precision conductive grade / general grade).

[0168] Description of the relevance of technical terms: The above technical terms jointly construct the "data perception → virtual simulation → dynamic matching → closed-loop control" technology chain for carbon black masterbatch production: The data layer (PCL, multi-source heterogeneous data fusion) realizes comprehensive perception of equipment status; The model layer (digital twin, hierarchical knowledge base) supports intelligent optimization of process parameters; The control layer (dynamic time regularization, melt pressure balance model) ensures cross-device coordination accuracy; Quality layer (volume resistivity, foreign matter detection) achieves product performance consistency.

[0169] Specific implementation steps of the present invention: Device startup and data synchronization: Equipment startup: The internal mixer, twin-screw extruder, and online detection equipment are powered on, and the PCL system automatically loads the equipment firmware version information (internal mixer V3.2, extruder V5.1).

[0170] Data collection and synchronization: Internal mixer: collect temperature (unit: °C) and torque (unit: N·m) data once per second; Extruder: collect melt pressure (unit: MPa) and temperature of each section (unit: °C) data 10 times per second; Online detection equipment: real-time collection of specific surface area and volume resistivity (unit: Ω·cm).

[0171] Time domain alignment: A sliding window mechanism (window length 1 second, overlap rate 50%) is used to interpolate and compensate the low-frequency data of the internal mixer to generate a synchronized timing parameter sequence.

[0172] Hierarchical model knowledge base matching: Scene Matching: Input the task parameters (resin, carbon black content 20%), and the system will match the "ABS-based-carbon black 15%-25%" scenario model from the first-level index of the knowledge base; The secondary index retrieves the corresponding process parameter combination: mixing temperature 180-190℃, screw speed 65-75rpm, feeding rate 80kg / h; Level 3 index loading quality threshold: volume resistivity ≤1.2×10³ Ω·cm (safety redundancy design).

[0173] Dynamic Time Warping (DTW) Verification: The heating curve of the internal mixer during the equipment preheating stage (0-5 minutes) was Fourier encoded, and its dynamic bending distance with the knowledge base model was calculated to screen out three candidate models; Verify device compatibility: exclude models that require extruder speed > 80 rpm (the current device supports a maximum of 75 rpm).

[0174] Multi-objective optimization control: Weighted evaluation function loading: Since the task type is "high-precision conductive level", the weight distribution is: quality qualification rate 60%, energy consumption coefficient 25%, equipment load rate 15%; The model with the highest comprehensive score was screened out: mixing temperature 185°C, screw speed 70rpm, and feeding rate 85kg / h.

[0175] Cross-device collaborative control instructions are issued: Internal mixer: Based on the torque spectrum analysis (low-frequency energy accounts for 45%), an instruction is generated to increase the linear speed of the dispersion disk to 25m / s; Extruder: Based on the melt pressure balance model, the screw speed compensation is calculated as -2 rpm (actual operation is 68 rpm); Pelletizer: According to the D50=420μm fed back by the online particle size analyzer, the high-speed knife group is set to account for 60% and the medium-speed knife group to account for 30%.

[0176] Real-time quality monitoring and closed-loop updates: Multidimensional quality assessment: 5 consecutive batches of test data: volume resistivity (0.9-1.1×10³ Ω·cm), black value (L*=23-24), unit energy consumption (0.85kW·h / kg); The quality was judged to be up to standard and there were no abnormalities in the evaluation matrix.

[0177] Foreign object detection trigger: In the sixth batch, abnormal fluctuations in melt pressure (±15%) were detected, and the linked vision system identified 0.5 mm metal chips; Expand the knowledge base foreign body atlas, add the "metal chips - spectral reflectance > 90%" feature, and trigger the extruder reverse clearing procedure.

[0178] Model iterative update: The traceability database records the raw material batch number (carbon black Lot#2309) and equipment status (extruder screw wear 2%) corresponding to the abnormal batch; The digital twin system simulates the scenario of metal chips mixing in and updates the foreign object detection sensitivity parameters of the "ABS-based-carbon black 20%" model.

[0179] Buffer material level closed loop feedback: Exception handling: During the 7th batch production, the buffer bin material level continued to be lower than the lower limit (30%), and the system determined that the mixer discharge valve was not open enough; Adaptive adjustment: the opening of the discharge valve is increased from 80% to 95%, the feeding rate is reduced from 85kg / h to 80kg / h, and the material level balance is restored within 10 seconds.

[0180] Embodiment effect: Process efficiency: task switching time is reduced from 4 hours to 20 minutes; Quality indicators: Volume resistivity fluctuation range is reduced by 50%, blackness value standard deviation ≤ 0.5; Energy efficiency optimization: unit energy consumption decreased by 15%, and equipment load rate stabilized at 85%-90%.

[0181] Summary of embodiments: This embodiment verifies the entire process of data synchronization → model matching → multi-objective optimization → closed-loop control, which proves the practical application value of the technical solution of the present invention in the production of carbon black masterbatch, significantly improves process stability and product consistency, and provides reliable technical support for the large-scale production of high-performance conductive masterbatch.

[0182] The present invention effectively solves the problem of dispersion uniformity and process energy efficiency optimization in the production of carbon black masterbatch through efficient fusion and dynamic collaborative control of multi-source heterogeneous data. First, in view of the sampling frequency and structural differences of multi-device data, a sliding window mechanism is used to perform time domain alignment processing on the asynchronous data streams of the mixing device and the extrusion device, and a synchronous process parameter sequence is generated in combination with outlier detection. By extracting the frequency domain characteristics and waveform characteristics of the equipment operating parameters, a unified data characterization framework is constructed to eliminate the time axis deviation of cross-process parameters such as mixing temperature and extrusion pressure, laying the foundation for the establishment of a dynamic correlation model.

[0183] Secondly, a multi-dimensional production task simulator is constructed through digital twin technology, and the raw material ratio parameters are combined with the equipment physical boundary constraint model to generate simulation data sets under different ratio scenarios. A hierarchical model knowledge base is established based on the hierarchical relationship between raw material ratios, process parameters and quality indicators, and a dynamic time warping algorithm is used to match real-time tasks with historical models. This process uses Fourier transform encoding features and dynamic time bending distance calculation to screen out candidate control models that are compatible with the current equipment status, and then combines multi-objective optimization sorting such as quality qualification rate and energy consumption coefficient to generate cross-device collaborative control instructions to achieve precise linkage adjustment of parameters such as the linear speed of the internal mixer dispersion disk and the screw speed of the extruder.

[0184] Finally, a closed-loop optimization is formed through real-time quality data feedback and iterative updates of the knowledge base. A multi-dimensional quality assessment matrix is ​​constructed to monitor key indicators such as blackness value and volume resistivity. When continuous batch data is detected to exceed the threshold, the mixed temperature compensation model retraining is automatically triggered. At the same time, a traceable process database is established to store the historical optimal model version and equipment operation data, supporting rapid backtracking and digital twin re-simulation of abnormal scenarios. Combined with the buffer material level closed-loop feedback mechanism, the material transmission timing is dynamically adjusted to ensure production continuity and process stability, thereby continuously improving the carbon black dispersion uniformity and the conductivity threshold control accuracy.

Claims

1. The automatic control method of carbon black masterbatch production equipment based on PCL is characterized in that: include: Collect heterogeneous data streams of mixing equipment, extrusion equipment and detection equipment in real time, perform time domain alignment processing on the heterogeneous data streams to generate a synchronous process parameter sequence, and extract frequency domain characteristics and waveform characteristics of equipment operation parameters; A multi-dimensional production task simulator including raw material ratio parameters is constructed based on the equipment operation parameters, simulation data sets of different ratio scenarios are generated through digital twin technology, and a hierarchical model knowledge base is established according to the hierarchical relationship between raw material ratios, process parameters and quality indicators; When receiving a new production task, parsing the task parameters and matching the candidate control model from the hierarchical model knowledge base through a dynamic time warping algorithm based on the current equipment state characteristics; Perform multi-objective optimization sorting on the candidate control models according to the quality indicator priorities, and convert key parameters of the selected models into cross-device collaborative control instructions; During the execution of the control instructions, production quality data is collected in real time, and iterative updating of the hierarchical model knowledge base is triggered based on the quality deviation detection results.

2. The automatic control method for carbon black masterbatch production equipment based on PCL according to claim 1, characterized in that: The time domain alignment process includes: A sliding window mechanism is used to compensate the sampling frequency of the asynchronous data streams of the mixing device and the extrusion device; Outlier detection is performed on the compensated data sequence to generate a set of process parameters with continuously aligned timestamps.

3. The automatic control method for carbon black masterbatch production equipment based on PCL according to claim 1, characterized in that: The construction of a multi-dimensional production task simulator comprises: Establish the constraint relationship model between raw material ratio parameters and physical boundaries of extrusion equipment; The constraint relationship model is used to drive the digital twin system to simulate the material dispersion process and generate a training data set including a mixing temperature and torque correlation curve.

4. The automatic control method for carbon black masterbatch production equipment based on PCL according to claim 1, characterized in that: The establishment of the hierarchical model knowledge base includes: Divide the basic scenario model by raw material ratio type as the first-level index; Storing the corresponding process parameter optimization combination under the basic scenario model as a secondary index; The process parameter combinations are associated with quality detection thresholds to form a three-level quality assessment standard.

5. The automatic control method for carbon black masterbatch production equipment based on PCL according to claim 1, characterized in that: The dynamic time warping algorithm includes: Perform Fourier transform encoding on the real-time collected equipment preheating status data; Calculate the dynamic time warping distance between the encoded feature vector and the feature matrix of the knowledge base model; Screening models whose bending distance is less than a preset threshold as a set of candidate control models; Verify the compatibility of equipment for the selected candidate control models; When it is detected that the current device hardware version does not match the model requirements, the knowledge base physical boundary parameter update is automatically triggered.

6. The automatic control method for PCL-based carbon black masterbatch production equipment according to claim 1, characterized in that: The multi-objective optimization sorting includes: Construct a weighted evaluation function including quality qualification rate, energy consumption coefficient and equipment load rate; Dynamically adjust the weight distribution ratio of the evaluation function according to the production task type; The foreign body detection enhancement model is forced to load for high carbon black ratio scenarios.

7. The automatic control method for PCL-based carbon black masterbatch production equipment according to claim 1, characterized in that: The conversion into a cross-device collaborative control instruction includes: Generate a dispersion disk linear speed adjustment instruction according to the torque spectrum characteristics of the mixing device; Calculate the screw speed compensation amount based on the extruder pressure balance model and generate a speed correction instruction; The speed gradient parameters of the pelletizing equipment are dynamically set according to the particle size distribution characteristics.

8. The automatic control method for PCL-based carbon black masterbatch production equipment according to claim 1, characterized in that: The real-time collection of production quality data includes: A multidimensional quality assessment matrix including specific surface area, DBP oil absorption value, volume resistivity and specific energy consumption was constructed; When consecutive batches of data in the evaluation matrix exceed an associated quality threshold, retraining of the hybrid temperature compensation model is triggered.

9. The automatic control method for PCL-based carbon black masterbatch production equipment according to claim 1, characterized in that: The iterative update includes: When unrecorded foreign body features are detected, the foreign body detection graph dimension in the knowledge base is expanded; Initiate the digital twin re-simulation process for scenario models whose matching failure rate exceeds the threshold; Establish a traceable process database including historical optimal model versions; The traceable process database includes: Store the equipment firmware information and raw material batch data corresponding to each historical version model; Provides comparative analysis of production process parameters based on the time dimension.

10. The automatic control method for PCL-based carbon black masterbatch production equipment according to claim 7, characterized in that: Also includes: Establish a closed-loop feedback mechanism for buffering material level between the discharge valve of the mixing equipment and the feed port of the extrusion equipment; When abnormal material transmission continuity is detected, the timing relationship between the feeding rate and the opening of the discharge valve is adaptively adjusted.

Images