Real-time monitoring, regulating and controlling method for melt state of extruder

Abstract

The invention relates to a real-time monitoring, regulating and controlling method for the melt state of an extruder in the technical field of information, which comprises the following steps: acquiring melt related data through extruder equipment to obtain an initial data sequence; processing the initial data sequence to generate a smoothed basic data set; calculating a melt apparent viscosity approximate value sequence according to the basic data set, and extracting fluctuation characteristics; determining a melt viscosity fluctuation coefficient by adopting a statistical method according to the approximate value sequence; melt performance parameters are obtained through a measuring device, and a melt elastic component characterization value is calculated; fusing the melt viscosity fluctuation coefficient and the melt elastic component characterization value to generate a comprehensive melt state index; generating a regulation and control instruction set according to the deviation condition of the comprehensive melt state index, and transmitting the regulation and control instruction set to a control system for adjustment; and circularly updating the data through feedback data, recalculating the comprehensive melt state index, and judging whether to return to a preset processing state interval or not.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method for real-time monitoring and control of the melt state in an extruder. Background Technology

[0002] On plastic extrusion production lines, the real-time data collected by extruder die pressure sensors and melt pump speed monitoring devices are often affected by process fluctuations, generating high-frequency noise. This results in an uneven pressure and speed sequence, failing to accurately reflect the true flow state of the melt. Consequently, even after initial filtering, deviations remain, affecting the reliability of the melt apparent viscosity approximation sequence based on fluid dynamics calculations, particularly in the extraction of fluctuation characteristics, leading to inaccurate deviation direction identification. Subsequently, the viscosity fluctuation coefficient calculated by the sliding window is easily amplified by this deviation, causing the deviation amplitude quantification evaluation logic to fail. This fails to provide a stable basis for the elastic modulus to viscous modulus ratio obtained by the melt performance measurement device, resulting in errors in the calculation of the melt elastic component characterization value. The fused state index frequently falls into the range of decreased flow stability, but due to the lack of precise mapping of direction and amplitude, it cannot trigger targeted control commands. A deeper problem is that when the side feeder speed adjustment changes the additive ratio, without real-time feedback on melt strength, the dynamic additive ratio and heating temperature zone correction cannot work together, resulting in uneven temperature gradients, which further aggravates pressure fluctuations, forming a vicious cycle. Ultimately, this leads to more surface defects, unstable dimensions, and low output in the products. The entire process control lags behind the dynamic changes in melt strength that deviate from the target value, and a multi-parameter feedback mechanism is urgently needed to break this chain of constraints. Summary of the Invention

[0003] This invention provides a method for real-time monitoring and control of the melt state in an extruder, mainly comprising:

[0004] The initial data sequence is obtained by acquiring melt-related data through an extruder. This initial data sequence is then processed to generate a smoothed base dataset. An approximate melt viscosity sequence is calculated based on the base dataset, and fluctuation characteristics are extracted. A statistical method is used to determine the melt viscosity fluctuation coefficient for this approximate sequence. Melt performance parameters are acquired through a measuring device, and the melt elastic component characterization value is calculated. The melt viscosity fluctuation coefficient and the melt elastic component characterization value are fused to generate a comprehensive melt state index. Based on the deviation of the comprehensive melt state index, a set of control commands is generated and transmitted to the control system for adjustment. The comprehensive melt state index is recalculated by cyclically updating the data through feedback, and it is determined whether the melt has returned to the preset processing state range. Furthermore, the step of acquiring melt-related data through the extruder equipment to obtain an initial data sequence includes: real-time acquisition of pressure data and speed parameters using a pressure sensor and speed monitoring device built into the extruder die to generate an initial data sequence; for the initial data sequence, a low-pass filtering method is used to cut off high-frequency noise, and interference signals are removed by setting a cutoff frequency to obtain a filtered data sequence; fluctuation characteristics are extracted from the filtered data sequence, and the difference between adjacent data points is calculated as an amplitude index; if the amplitude index exceeds a preset threshold, anomalies are marked, and a labeled data sequence is generated; the speed parameters are adjusted according to the labeled data sequence, and the pressure value is simultaneously optimized through proportional-integral-derivative control to obtain a smoothed pressure and speed basic dataset; the smoothed pressure and speed basic dataset is used as the basis for subsequent calculations to ensure data accuracy and stability. Furthermore, the step of calculating the approximate sequence of apparent melt viscosity based on the basic dataset and extracting fluctuation features includes: extracting melt flow parameters from the smoothed basic dataset; processing the melt flow parameters using fluid dynamics calculation methods to obtain a preliminary viscosity estimate through the ratio of pressure to rotational speed; adjusting the preliminary viscosity estimate in conjunction with the real-time monitoring characteristics of melt strength to determine the approximate sequence of apparent melt viscosity; analyzing the fluctuation pattern for the approximate sequence and determining the deviation direction through the magnitude of sequence value changes; synchronously recording the fluctuation features according to the deviation direction and using the fluctuation features as input for subsequent fluctuation coefficient calculation to ensure the integrity and reliability of the fluctuation features.Furthermore, the step of determining the melt viscosity fluctuation coefficient using statistical methods for the approximate value sequence includes: obtaining continuous data segments from the melt apparent viscosity approximate value sequence; dividing the data segments using a sliding time window to obtain a data subset within the window; calculating the mean and standard deviation for the data subset within the window to obtain fluctuation characteristic values; determining a preliminary deviation index by using the absolute difference between the fluctuation characteristic value and the mean of the overall sequence as a quantitative assessment of the deviation magnitude; if the preliminary deviation index exceeds a preset threshold, adjusting the window size and recalculating the fluctuation characteristic value to obtain an optimized deviation index; generating the melt viscosity fluctuation coefficient by integrating the optimized deviation index with preset quantitative logic; and determining real-time control parameters by matching the fluctuation coefficient with historical control data in a preset database. Furthermore, the step of acquiring melt performance parameters through a measuring device and calculating the melt elastic component characterization value includes: acquiring elastic modulus and viscous modulus data at specific locations in the extrusion channel using a melt performance measuring device; calculating the ratio between the elastic modulus and viscous modulus data to determine the ratio; obtaining temperature influence correction parameters from a preset temperature curve using melt state index calculation rules combined with the ratio to obtain preliminary values ​​of the elastic component; applying component characterization rules to the preliminary values ​​of the elastic component to determine the melt elastic component characterization value; and using the melt elastic component characterization value as an important basis for comprehensively evaluating the melt state to ensure the scientificity and accuracy of the evaluation results. Furthermore, the step of fusing the melt viscosity fluctuation coefficient and the melt elastic component characterization value to generate a comprehensive melt state index includes: acquiring the melt viscosity fluctuation coefficient and the melt elastic component characterization value; applying a first weight to the fluctuation coefficient and a second weight to the characterization value using a preset weighting rule, and then performing a weighted summation to obtain the comprehensive melt state index; for the comprehensive melt state index, using intelligent deviation direction discrimination to extract a deviation vector from the directional difference between the index and the standard value; calculating the magnitude of the deviation vector through deviation amplitude quantification evaluation to determine the deviation result; if the deviation result falls into the flow stability decrease range, generating a trigger signal based on the distribution of the deviation vector within the range; and extracting a compensation factor based on the trigger signal to generate control parameters. Furthermore, the step of generating a set of control instructions based on the deviation of the comprehensive melt state index and transmitting it to the control system for adjustment includes: acquiring the comprehensive melt state index through a sensor, comparing it with a preset threshold to determine the deviation; generating a set of adjustment instructions for adjusting the side feeder speed based on the deviation, and simultaneously generating a set of adjustment instructions for correcting the heating temperature zones; transmitting the set of adjustment instructions to the extruder control system through a hierarchical distribution mechanism to synchronously execute precise temperature zone control; dynamically proportioning the additive ratio during the execution of precise temperature zone control to obtain real-time calibration results of the material viscosity; and if the calibration results meet the requirements of the comprehensive melt state index, then the synchronous operation is completed to ensure the stability of the control effect.Furthermore, the step of updating the data through feedback data loop and recalculating the comprehensive melt state index includes: obtaining new pressure and speed update data and integrated temperature variable data after the adjustment command is executed through feedback data loop; using the pressure and speed update data and the integrated temperature variable data, fitting the data to historical target values, and combining abnormal deviation correction to deduct noise offset, to recalculate the comprehensive melt state index; based on the comprehensive melt state index, fitting a trend line through a linear regression model to determine the regression trend; determining whether it has regressed to a preset processing state range by comparing it to the preset optimal range boundary; and storing the determination result and historical data for subsequent optimization reference. Furthermore, the process of processing the initial data sequence to generate a smoothed basic dataset includes: using a filtering method to remove high-frequency interference signals from the initial data sequence to obtain a filtered data sequence; extracting fluctuation features from the filtered data sequence and calculating the difference between adjacent data points as an amplitude index; if the amplitude index exceeds a preset threshold, marking outliers and generating a labeled data sequence; adjusting relevant parameters according to the labeled data sequence and synchronously optimizing data values ​​through a control algorithm to obtain a smoothed basic dataset; and using the smoothed basic dataset as input for subsequent calculations of the melt apparent viscosity approximation sequence to ensure the continuity and consistency of data processing. Furthermore, the step of generating a set of control instructions based on the deviation of the comprehensive melt state index includes: acquiring the comprehensive melt state index, comparing it with a preset threshold, and determining the direction and magnitude of the deviation; generating a set of adjustment instructions for equipment operating parameters based on the direction and magnitude of the deviation; determining temperature zone control parameters and additive ratio parameters through the set of adjustment instructions; generating specific control instructions based on the temperature zone control parameters and additive ratio parameters; transmitting the control instructions to the control system through a hierarchical distribution mechanism and synchronously executing the adjustment operation; acquiring the adjusted feedback data to verify whether the comprehensive melt state index meets the preset requirements, ensuring the closed-loop nature of the control process.

[0005] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0006] This invention discloses a method for real-time monitoring and control of the melt state in extruders, addressing the problem of uneven processing quality caused by decreased melt flow stability. By collecting and filtering pressure and speed data in real time, and combining this with fluid dynamics calculations to approximate the melt's apparent viscosity, a melt viscosity fluctuation coefficient is generated using a sliding time window to statistically analyze fluctuation characteristics. Simultaneously, the elastic component of the melt is calculated using the ratio of elastic modulus to viscous modulus, and these two factors are combined to generate a comprehensive melt state index. If the index deviates from the stable range, this invention automatically triggers control commands, generating a set of adjustment commands for the side feeder speed and heating temperature zones, and precisely controlling the temperature and additive ratio through a hierarchical distribution mechanism. After adjustment, data is continuously updated and the return to the optimal state is monitored; historical data is stored for optimization. This invention ensures melt state stability and improves processing quality and production efficiency through intelligent judgment and dynamic control. Attached Figure Description

[0007] Figure 1 This is a flowchart of a method for real-time monitoring and control of the melt state in an extruder according to the present invention. Detailed Embodiments

[0008] The technical solutions of the embodiments of the present invention will be clearly and thoroughly described below with reference to the accompanying drawings. The described embodiments are merely some embodiments of the present invention.

[0009] like Figure 1 This embodiment of a method for real-time monitoring and control of the melt state in an extruder may specifically include:

[0010] Step S101: Through the pressure sensor and melt pump speed monitoring device built into the extruder die head, real-time data sequences of pressure data dynamic acquisition and speed parameter synchronous adjustment are collected. The collected data is preliminarily filtered to obtain a smoothed basic dataset of pressure and speed.

[0011] Pressure data and speed parameter sequences are collected using a pressure sensor built into the extruder die and a melt pump speed monitoring device to obtain an initial data sequence. For this initial data sequence, a low-pass filter is used to cut off high-frequency noise. By setting a cutoff frequency to remove interference signals, a filtered data sequence is obtained. Fluctuation characteristics are extracted from the filtered data sequence, and the difference between adjacent data points is calculated as an amplitude index. If the amplitude index exceeds a preset threshold, anomalies are marked, resulting in a labeled data sequence. The speed parameters are adjusted based on the labeled data sequence, and the pressure value is simultaneously optimized using proportional-integral-derivative control to obtain a smoothed pressure and speed baseline dataset.

[0012] In one implementation, pressure data and speed parameters are collected in real time using a pressure sensor and a melt pump speed monitoring device built into the extruder die. The extruder die is a key component of plastic extrusion equipment, used to shape molten material, while the melt pump is responsible for stably delivering the melt, and its speed directly affects the material flow. The pressure sensor, embedded inside the die, captures pressure fluctuations within the die, while the speed monitoring device is connected to the melt pump's drive system to record the pump's rotational speed. These devices work together to ensure the synchronization of data acquisition.

[0013] For example, when the pressure sensor detects an increase in die pressure, the speed monitoring device simultaneously adjusts the pump speed to maintain stable output. This real-time acquisition forms a data sequence encompassing pressure values ​​and corresponding speed parameters for subsequent analysis. In plastic pipe extrusion scenarios, this method can monitor the continuous operation of the production line and prevent material blockage. Furthermore, the acquired data undergoes preliminary filtering to obtain a smoothed baseline dataset of pressure and speed. The filtering process aims to remove noise interference, such as random fluctuations caused by equipment vibration or environmental factors.

[0014] Specifically, a moving average filtering method is used to apply an average calculation with a window size of 10 to the pressure data sequence, that is, to take the average of several points before and after each data point, thereby smoothing the curve. Similarly, a similar processing is applied to the speed parameter sequence to ensure the continuity and reliability of the speed data. After this filtering, the dataset is more suitable for the optimized control of the extrusion process.

[0015] For example, in film extrusion production, smoothing datasets can help identify problems where pressure peaks and rotational speeds are mismatched, allowing for adjustments to equipment parameters to improve product uniformity.

[0016] It should be noted that the filtering process is executed in real time in the embedded controller, and the processing time is controlled in milliseconds to match the production rhythm.

[0017] Preferably, in another embodiment, the pressure sensor is a piezoresistive sensor, installed near the die outlet, directly contacting the melt to obtain accurate readings. The speed monitoring device uses an optical encoder, fixed on the melt pump shaft, and outputs pulse signals converted into speed values. Real-time data acquisition is achieved through a data acquisition card, with a sampling frequency set to 100Hz to ensure the capture of dynamic changes. Synchronous adjustment is accomplished through a feedback loop; when the pressure exceeds a threshold, the speed is automatically fine-tuned to balance the system. This setup is widely used in cable sheath extrusion and can handle pressure fluctuations in high-viscosity materials. The acquired data sequence includes timestamps, pressure values, and speed values, forming a multi-dimensional array for easy storage before filtering.

[0018] For example, in profile extrusion scenarios, the initial filtering process can be expanded into median filtering, replacing outlier data points by selecting the median value of the sequence to replace noise points, thus obtaining a more robust and smooth dataset. This method is particularly suitable for noisy environments, such as factory workshops. The pressure curve of the filtered dataset shows a smooth trend, while the rotational speed parameter reflects stable adjustment, improving the overall monitoring accuracy of the extrusion process.

[0019] Understandably, the combination of the aforementioned acquisition and filtering processes enables the reliable construction of a data foundation in the field of plastic extrusion. Through these steps, the system not only captures real-time dynamics but also provides smoothed data to support subsequent optimizations, such as reducing scrap rates and improving energy efficiency.

[0020] In one possible implementation, the dataset is further used for trend analysis, but the focus remains on the initial processing stage.

[0021] Specifically, in the sheet extrusion embodiment, the integration of the pressure sensor and rotation speed device allows for customized sampling intervals, adjusted according to material type. For example, for polyethylene, the sampling frequency is increased to 200Hz to capture rapid changes. Filtering is combined with a low-pass filter with a cutoff frequency set to 5Hz to remove high-frequency noise, resulting in a clear baseline dataset. This versatility demonstrates the versatility of the technology within the same field, ensuring continuous and stable production processes.

[0022] Step S102: Based on the smoothed pressure and rotation speed dataset, the fluid dynamics calculation method is used, combined with the real-time monitoring of melt strength, to determine the approximate value sequence of the melt apparent viscosity, and simultaneously record the fluctuation characteristics required for intelligent discrimination of deviation direction.

[0023] A smoothed baseline dataset of pressure and rotational speed is obtained, and melt flow parameters are extracted from this dataset. These parameters are processed using fundamental fluid dynamics calculation methods, and a preliminary viscosity estimate is obtained based on the pressure-rotational speed ratio. The preliminary viscosity estimate is adjusted in conjunction with real-time melt strength monitoring to determine an approximate sequence of apparent melt viscosity. Fluctuation patterns are analyzed based on this approximate sequence, and the direction of deviation is determined by the magnitude of changes in the sequence values. The desired fluctuation characteristics are then synchronously recorded based on these deviation directions.

[0024] In one implementation, the collected pressure and speed baseline dataset is first smoothed to reduce noise interference.

[0025] Specifically, a window function, for example, with a window size of 5 to 10 data points, is applied to the original data sequence using a moving average filtering method to obtain a smoothed sequence of pressure and rotational speed values. This processing helps improve the accuracy of subsequent calculations, ensuring that the data reflects the true melt flow state. Based on the smoothed dataset, fundamental fluid dynamics calculation methods are used to determine the apparent viscosity of the melt. These methods utilize the relationship between shear stress and shear rate, for example, using the formula η = τ / γ, where η is the apparent viscosity, τ is the shear stress, and γ is the shear rate. Pressure data can be converted to shear stress, and rotational speed data corresponds to shear rate, thus calculating the viscosity value at each time point, forming an approximate value sequence. The determination of this sequence involves iterative calculations, such as averaging over 10 consecutive data points to obtain stable approximate values. Furthermore, the viscosity calculation is refined by incorporating the characteristics of real-time melt strength monitoring. Real-time melt strength monitoring refers to acquiring strength indicators of the melt during the flow process in real time, such as tensile modulus or breaking strength, through sensors installed on extruders or injection molding machines, such as tensile strength testers. These characteristics are integrated into the calculations. For example, when the intensity value exceeds a threshold, the correction factor in the viscosity calculation is adjusted to reflect the change in flow resistance of the melt at high intensity, thus making the approximate value sequence more consistent with actual production scenarios. Simultaneously, the fluctuation characteristics required for intelligent discrimination of deviation direction are recorded.

[0026] Specifically, while determining the viscosity sequence, fluctuations within the sequence are monitored, such as by calculating the absolute value and direction of the difference between adjacent values. A positive difference indicates an increase in viscosity, while a negative difference indicates a decrease. These fluctuation characteristics are recorded as a vector sequence for subsequent intelligent discrimination.

[0027] For example, in the plastic extrusion process, if the fluctuation characteristics show a continuous positive deviation, it may indicate that the melt is overheated and the speed needs to be adjusted.

[0028] In one possible implementation, the method is applied to an injection molding machine.

[0029] For example, a pressure sensor is installed at the mold inlet, and a speed sensor is placed in the screw drive system. After smoothing, the viscosity sequence is calculated, and the parameters are adjusted in conjunction with strength monitoring.

[0030] For example, when melt strength monitoring shows a decrease in strength, a negative correction is introduced into the viscosity calculation to ensure the accuracy of the sequence. Fluctuation characteristic recording is implemented through a software module, extracting peak and trough values ​​to determine the direction of deviation; for example, an upward deviation prompts the need for increased cooling.

[0031] Preferably, in another embodiment, multi-point monitoring is added for continuous extrusion processes. A smoothed dataset covers pressure and rotational speed at multiple locations, and fluid dynamics calculations are extended to consider the effects of pipe geometry, for example, by introducing a simplified form of the Hagen-Poiseuille equation to estimate viscosity. Melt strength characteristics are acquired in real time via optical sensors and combined with calculations to generate a sequence. Fluctuation characteristics are recorded in terms of frequency and amplitude; for example, fluctuation frequencies above 5 Hz are marked as abnormal deviations.

[0032] It should be noted that these implementation methods ensure the versatility of the technical solution in the field of melt processing. Through the above steps, effective monitoring and fluctuation analysis of the apparent viscosity of the melt can be achieved, improving process stability in production, such as reducing the occurrence of defective products.

[0033] For example, in actual operation, assuming a smoothing pressure of 200 Pa and a rotation speed of 50 rpm, the initial viscosity is calculated to be 100 Pa·s using fluid dynamics, and corrected to 95 Pa·s by intensity monitoring, forming part of the sequence. Fluctuation characteristic recordings show the sequence deviating downwards, intelligently identifying insufficient rotation speed.

[0034] Understandably, the scalability of this method is reflected in different melt materials, such as polyethylene or polypropylene, while maintaining a consistent calculation process, only adjusting the strength monitoring threshold.

[0035] In one embodiment, the entire process is integrated into the control system, forming a closed-loop control from data acquisition to feature recording, thereby enabling real-time optimization in an industrial environment.

[0036] Step S103: For the approximate value sequence of the apparent viscosity of the melt, the sliding time window statistical method is used to calculate the fluctuation characteristics of the data within the window. Combined with the logic of quantitative evaluation of deviation amplitude, the melt viscosity fluctuation coefficient is obtained as the basis for subsequent regulation.

[0037] Continuous data segments are obtained from the apparent viscosity approximation sequence of the melt. These segments are then divided using a sliding time window to obtain a subset of data within the window. The mean and standard deviation of each subset are calculated to obtain fluctuation characteristic values. The absolute difference between these fluctuation characteristic values ​​and the overall sequence mean is used as a quantitative assessment of the deviation magnitude to determine a preliminary deviation index. This preliminary deviation index is compared with a preset threshold. If the preliminary deviation index exceeds the threshold, the window size is adjusted, and the fluctuation characteristic values ​​are recalculated to obtain an optimized deviation index. Based on the optimized deviation index and a preset quantification logic, a melt viscosity fluctuation coefficient is generated. This fluctuation coefficient is then matched with historical control data obtained from a preset database to determine real-time control parameters.

[0038] In one implementation, the approximate sequence of the melt's apparent viscosity is processed. First, viscosity data sequences of the melt during processing are collected, for example, in a plastic extruder. Then, the melt flow resistance is monitored in real time by sensors and converted into approximate viscosity values, forming a time-ordered numerical sequence. This sequence reflects the dynamic changes of the melt under the influence of factors such as temperature and pressure.

[0039] It should be noted that the apparent viscosity of the melt refers to the effective viscosity of the melt in shear flow, which is greatly affected by processing conditions. Excessive fluctuations may lead to product defects, such as uneven surface or insufficient strength. This data collection method ensures the continuity and representativeness of the data sequence, providing a foundation for subsequent analysis. Furthermore, a sliding time window statistical method is used to analyze this sequence.

[0040] Specifically, a sliding time window is a technique for dynamically segmenting a sequence, dividing the entire sequence into multiple overlapping sub-segments. For example, the window size is set to 10 time points with a step size of 1, moving forward one point at a time to cover different parts of the sequence. Within the window, the fluctuation characteristics of the data, such as the mean, standard deviation, or range, are calculated. These characteristics quantify the degree of variation in viscosity values.

[0041] For example, in the context of plastic injection molding, a large standard deviation within the window indicates unstable melt viscosity, which may be caused by uneven raw material distribution or temperature fluctuations. This method helps to capture local fluctuations rather than global averages, avoiding the overlooking of short-term anomalies.

[0042] Preferably, the fluctuation characteristics within the window are further processed by combining the logic of deviation magnitude quantification assessment. Deviation magnitude refers to the deviation of each data point from the window average, and quantification assessment obtains a numerical index by calculating the absolute value or sum of squares of the deviation.

[0043] For example, summing all deviations and dividing by the window size yields an amplitude coefficient. The principle behind this is that a larger deviation indicates a greater deviation from the expected steady state of viscosity. In melt processing, this might correspond to misalignment of process parameters, such as unstable screw speed. This quantification transforms fluctuation characteristics into comparable values, supporting subsequent decision-making.

[0044] In one possible implementation, the melt viscosity fluctuation coefficient is calculated based on the aforementioned fluctuation characteristics and deviation magnitude. Specifically, this involves multiplying or weighting the standard deviation within the window with the deviation magnitude coefficient to obtain a comprehensive coefficient.

[0045] For example, if the standard deviation is 5 and the amplitude coefficient is 2, the fluctuation coefficient might be 10. This calculation reflects the intensity and persistence of the fluctuation. In the plastic extrusion process, this coefficient is used to assess melt stability; if the coefficient exceeds a threshold, such as 8, it indicates that regulation is needed.

[0046] It should be noted that the calculation process of this coefficient emphasizes the combination of local and overall factors to ensure that the coefficient is sensitive to real-time changes rather than static indicators.

[0047] For example, this method can be applied at different stages of melt processing. In the initial melting stage, the window size can be adjusted to 5 to capture rapid fluctuations, while in the stable extrusion stage, the window size increases to 15 to assess long-term trends. This flexibility demonstrates the versatility of the technology, applicable within the same field, such as the processing of different plastic types (e.g., polyethylene or polypropylene). Furthermore, this melt viscosity fluctuation coefficient serves as the basis for subsequent control.

[0048] In one embodiment, if the coefficient is higher than a preset value, the system automatically adjusts the heating temperature or pressure, for example, by lowering the temperature to reduce viscosity fluctuations, thereby improving melt flow uniformity. This regulation helps improve product consistency and, in actual plastic molding, can reduce scrap rates and optimize energy consumption.

[0049] Understandably, in another implementation, combining multi-window analysis to enhance accuracy, such as using nested windows, first calculates local fluctuations in a smaller window and then integrates the deviation magnitude in a larger window to obtain more robust coefficients. This method performs well in complex melt environments, such as in plastics processing with added fillers, because it can distinguish noise from true fluctuations.

[0050] Specifically, the calculation of the fluctuation coefficient can also incorporate a weighting factor, assigning higher weight to recent data based on processing experience. For example, the deviation at the most recent time point could be multiplied by a weight of 1.2, which enhances the real-time responsiveness of the coefficient. In melt extruder monitoring, this weighted logic ensures timely adjustments and avoids production interruptions.

[0051] In one embodiment, the effectiveness of the coefficient is verified using historical data, for example, by comparing the decrease in the coefficient before and after adjustment. If it decreases from 12 to 4, it indicates improved stability. This objective evaluation supports the application and promotion of the method. Finally, in continuous monitoring of melt processing, this coefficient is integrated into the control system to form a closed-loop feedback, such as by linking with a PLC system to achieve automated adjustment, thereby maintaining process optimization.

[0052] Step S104: The elastic modulus and viscous modulus data are obtained at a specific location in the extrusion channel using a melt performance measuring device. The ratio of the two is calculated, and the melt elastic component characterization value is determined by combining the rules for calculating the melt state index, which is used to comprehensively evaluate the melt state.

[0053] Elastic modulus and viscous modulus data are acquired at specific locations in the extrusion channel using a melt performance measuring device, thus obtaining the modulus data. The ratio of elastic modulus to viscous modulus is calculated based on this modulus data, and the ratio is determined. Temperature influence correction parameters are obtained from a preset temperature curve using melt state index calculation rules and this ratio, resulting in preliminary values ​​for the elastic component. Component characterization rules are then applied to these preliminary values ​​to determine the characterized values ​​of the melt elastic component.

[0054] In one implementation, elastic modulus and viscous modulus data are acquired at specific locations in the extrusion runner using a melt performance measurement device. This device is typically installed in the extruder runner, such as near the die inlet, to monitor melt flow characteristics in real time. The elastic modulus reflects the melt's elastic recovery under stress, while the viscous modulus represents the melt's resistance to flow. These modulus data can be obtained through oscillatory rheological testing or capillary rheometers, for example, by applying a sinusoidal stress wave and measuring the strain response, thereby calculating the storage modulus (elastic portion) and loss modulus (viscous portion). This acquisition method ensures that the data accurately reflects the dynamic behavior of the melt during the actual extrusion process for subsequent calculations. Furthermore, the ratio of the elastic modulus to the viscous modulus is calculated.

[0055] Specifically, this ratio is derived using the formula tanδ = viscous modulus / elastic modulus, where tanδ is called the loss factor, quantifying the viscoelastic balance of the melt. In the field of plastic extrusion, for example, when processing polyethylene melt, a ratio greater than 1 indicates that the melt tends towards viscous flow, facilitating molding; conversely, a ratio less than 1 indicates stronger elasticity, which may lead to extrusion instability. This calculation allows for a preliminary assessment of the melt's viscosity and elasticity without introducing complex numerical simulations.

[0056] Preferably, the melt elastic component characterization value is determined by combining the rules for calculating the melt state index. The melt state index, typically referred to as the melt flow index (MFI), is measured by standard tests such as ASTM D1238 and reflects the flow rate of the melt at a specific temperature and load. The calculation rules involve weighting the modulus ratio with the MFI, for example using the empirical formula: Elastic component characterization value = k * (tanδ) * (1 / MFI), where k is a correction factor adjusted according to the material type. This combination allows for the quantification of the melt's elastic contribution; for example, in polypropylene extrusion, a high MFI corresponds to low viscosity, and combining this with a low tanδ yields a lower elastic component, indicating that the melt is easy to process.

[0057] In one possible implementation, the method is applied to comprehensively evaluate the melt state.

[0058] For example, on a continuous extrusion production line, after acquiring data, ratios and characterization values ​​are calculated. If the characterization value exceeds a threshold, extrusion parameters such as temperature or screw speed are adjusted to optimize melt stability. This evaluation covers a variety of scenarios, such as film extrusion or pipe extrusion, demonstrating the versatility of the method.

[0059] It should be noted that the specific location of the melt performance measuring device affects the data accuracy. Installing the device in the middle of the extrusion runner can capture the state of the melt after it has been fully mixed, while the device at the end reflects the characteristics close to molding.

[0060] For example, in cable sheath extrusion, selecting the middle position helps to detect elasticity abnormalities early and avoid product defects.

[0061] Specifically, the calculation of the ratio of the two, combined with the rules of the melt state index, needs to take into account material variations.

[0062] For example, for recycled plastic melt, the index calculation rules can incorporate a temperature compensation factor to ensure that the characterization value accurately reflects the elastic component. In practice, the modulus data is measured first, and then the index value is input for fusion calculation; this sequence of steps ensures the reliability of the assessment.

[0063] For example, in a preform extrusion scenario, after acquiring modulus data, the device calculates a ratio of 0.5. Combined with the rule that MFI is 10 g / 10 min, a specific numerical value is derived to determine whether the melt is suitable for injection molding. This implementation expands the applicability of the method through different parameter combinations. Furthermore, the method improves the accuracy of quantitative evaluation of the melt state.

[0064] For example, in fiber extrusion, determined elastic component characterization values ​​can guide process optimization and reduce fiber breakage without the need for subjective judgment.

[0065] In one embodiment, for high-viscosity melts such as polyamide, the device acquires data at specific locations in the flow channel, such as curved sections, calculates ratios, and combines them with exponential rules. The characterization values ​​are used for real-time monitoring to ensure production consistency.

[0066] Understandably, through the above steps, this method provides a systematic melt evaluation approach in the field of plastic extrusion, applicable to various polymer types.

[0067] Step S105: The melt viscosity fluctuation coefficient and the melt elastic component characterization value are integrated to generate a comprehensive melt state index based on a preset weighting rule. If the index falls into the range of decreased flow stability after intelligent discrimination of deviation direction and quantitative evaluation of deviation magnitude, the process of automatically generating control instructions is triggered.

[0068] The melt viscosity fluctuation coefficient and the melt elastic component characterization value are obtained. A first weight is applied to the coefficient and a second weight to the characterization value using a preset weighting rule, and then weighted summation is performed to obtain a comprehensive melt state index. For the melt state index, intelligent deviation direction discrimination is used to extract a deviation vector from the directional difference between the index and the standard value. The magnitude of the vector is calculated through quantitative evaluation of the deviation amplitude to determine the deviation result. If the deviation result falls into the range of decreased flow stability, a trigger signal is generated based on the distribution of the deviation vector within the range. Using the trigger signal, melt temperature compensation adjustment is used to extract a compensation factor from the temperature deviation corresponding to the signal, and the control parameters are determined based on the matching of the factor with a preset temperature curve. From the control parameters, control instructions are generated, realizing an automatic generation process.

[0069] In one embodiment, the melt viscosity fluctuation coefficient is characterized by monitoring the viscosity change of the melt during the flow process.

[0070] Specifically, this coefficient reflects the degree of fluctuation in melt viscosity over time or shear rate. For example, in plastic injection molding, sensors installed inside the extruder collect melt viscosity data in real time, and then the ratio of the standard deviation to the average viscosity value is calculated to obtain the fluctuation coefficient. This calculation process helps identify instabilities in melt flow; for example, when batch variations in raw materials lead to increased viscosity fluctuations, this coefficient will increase accordingly. In this way, the accurate quantitative description of the melt state is ensured, providing a foundation for subsequent fusion. Furthermore, the melt elastic component characterization value is used to capture the elastic behavior of the melt.

[0071] In one possible implementation, this value is generated based on rheological test data, such as measuring the storage modulus and loss modulus of the melt using a rotational rheometer, and then characterizing the elastic component by the ratio of the elastic modulus to the total modulus. In plastics processing scenarios, this characterization reflects the melt's resilience under high pressure; if the elastic component is too high, it may lead to uneven flow. The process of obtaining this characterization includes data acquisition, modulus calculation, and proportional quantization, ensuring its clear role in melt state assessment. Generating a comprehensive melt state index based on preset weighting rules is the core step.

[0072] In one embodiment, the melt viscosity fluctuation coefficient and the melt elastic component characterization value are first used as input parameters, and then a weighted formula is applied for fusion, for example, assigning a higher weight to the fluctuation coefficient to emphasize its influence on stability. Specifically, this involves setting preset weights, such as 0.6 for the fluctuation coefficient and 0.4 for the elastic component, and then calculating the weighted sum to obtain the exponential value. This fusion mechanism is common in injection molding; for example, the weights can be adjusted for different polymer materials to achieve a comprehensive assessment of the overall melt state. The innovation of this step lies in integrating multi-dimensional parameters through weighted rules, avoiding the limitations of a single indicator, thereby improving the accuracy of flow monitoring. In practical applications, this exponential generation can effectively capture the transition of the melt from stable to unstable states, providing a reliable basis for subsequent judgment.

[0073] For example, the index analyzes its changing trend through intelligent deviation direction determination. In one implementation, deviation direction determination employs a trend analysis algorithm, such as comparing the difference between the current index and the historical average. If the difference is positive, it is determined to be an upward deviation, indicating a potential decrease in stability. Through this intelligent determination, the direction of index deviation can be identified in real time during the plastic melting process. For example, when temperature fluctuations cause the index to deviate upward, the system will mark it as a risk signal. Details of this determination process include data sequence analysis and threshold comparison to ensure the accuracy of the determination results.

[0074] It should be noted that the deviation magnitude quantitative assessment further quantifies the degree of deviation.

[0075] Specifically, the assessment calculates the absolute magnitude of the deviation and compares it to a preset threshold; for example, a deviation exceeding 10% is considered significant. In injection molding scenarios, this quantification helps assess the risk level; for instance, small deviations may only require monitoring, while large deviations trigger alarms. The detailed process description includes a magnitude calculation formula such as (current index - benchmark index) / benchmark index, and outputs a quantified score to support subsequent range determination.

[0076] Preferably, if the index falls into the range where flow stability decreases, a control command is automatically generated.

[0077] In one embodiment, the flow stability decline range is predefined as the range where the exponent value exceeds 1.2. After the aforementioned discrimination and evaluation, if the condition is met, the system automatically generates instructions such as adjusting the heating temperature or shear rate. In plastics processing, this triggering process ensures timely intervention; for example, when the exponent enters the decline range, instructions are generated to reduce the extrusion speed to restore stability. The technical advantage of this mechanism is that it achieves automated control and reduces delays caused by human intervention.

[0078] For example, in another implementation, for high-viscosity polymer melts, the fusion process can adjust the weighting rules to increase the weight of the elastic component, thereby adapting to specific material properties. This variant demonstrates the versatility of the technology, covering different melt types within the same plastic injection molding field.

[0079] Understandably, the entire process achieves closed-loop control through the integration of sensors and computing modules. In actual injection molding production lines, the logical sequence from parameter acquisition to instruction generation ensures continuous monitoring of the melt state.

[0080] Step S106: Based on the deviation of the melt state index, a set of adjustment instructions for adjusting the side feeder speed and correcting the heating temperature zones is generated. The adjustment instructions are transmitted to the extruder control system through a hierarchical mechanism, and the precise temperature zone control and dynamic ratio of additives are executed simultaneously.

[0081] Melt state index (MSI) is acquired from the extruder using sensors. The MSI is compared to a preset threshold to determine deviations. Based on these deviations, a set of adjustment commands is generated for adjusting the side feeder speed, and simultaneously, a set of adjustment commands is generated for correcting heating temperature zones. These adjustment command sets are transmitted to the extruder control system via a hierarchical distribution mechanism, enabling synchronized execution of precise temperature zone control. During the execution of precise temperature zone control, the additive ratio is dynamically adjusted, and real-time material viscosity calibration results are obtained for optimizing feed uniformity. If the results meet the MSI requirements, the synchronization operation is complete.

[0082] In one implementation, the melt state index refers to the quantified value of the fluidity of a plastic melt at a specific temperature and pressure, typically measured using standard testing equipment to assess the viscosity and uniformity of the material during the extrusion process.

[0083] Specifically, when the extruder is running, the system collects melt samples in real time and calculates their state index. If the actual index deviates from the preset target value—for example, if the target value is 10g / 10min as required by standard production, and the actual value is 12g / 10min—the deviation is quantified as a positive deviation, indicating that the melt flowability is too strong, which may lead to uneven product thickness. Based on this deviation assessment, the system generates instructions for adjusting the side feeder speed, such as reducing the speed to decrease the material input and thus balance the melt state. Furthermore, the generation of the adjustment instruction set is based on a threshold grading of the degree of deviation.

[0084] In one possible implementation, if the deviation is less than 5%, the instruction only involves fine-tuning the side feeder speed, such as reducing it from an initial 60 rpm to 58 rpm; if the deviation exceeds 10%, it is combined with heating temperature zone correction, for example, adjusting the temperature of the first heating zone from 200°C to 195°C to reduce the melt temperature and control flow. This set includes multiple subsets of instructions to ensure coverage of different zones of the extruder.

[0085] For example, the hierarchical transmission mechanism transmits adjustment commands to the extruder control system in a hierarchical manner according to priority and module. Specifically, high-priority commands, such as temperature correction commands, are first sent to the heating module via the main control layer, and then speed adjustment commands are transmitted hierarchically to the feeding layer. This mechanism avoids command conflicts and ensures system stability during transmission. For instance, in a plastic pipe extrusion scenario, temperature zone adjustments are performed first, followed by synchronized speed changes.

[0086] It should be noted that synchronous execution of precise temperature zone control involves dividing the extruder heating cylinder into multiple independent zones, each equipped with a sensor to monitor temperature deviation in real time and independently adjust the heating power according to instructions.

[0087] For example, in an embodiment of plastic film production, the temperature of the first zone is set to 180°C and the temperature of the second zone is 220°C. The system achieves precise control through a PID controller, with the deviation controlled within ±1°C, thereby ensuring uniform melt flow.

[0088] In one embodiment, the dynamic proportioning of additives is performed simultaneously with temperature control.

[0089] Specifically, the proportion of additives such as plasticizers or stabilizers is dynamically adjusted based on deviations from the melt state index. For example, when the index is high, the stabilizer proportion is increased from 2% to 3%, precisely metered in via a side feeder. This formulation mechanism relies on a feedback loop, first analyzing the chemical composition of the melt sample, then calculating the required additive dosage to ensure real-time optimization of material properties during extrusion.

[0090] Preferably, in the scenario of cable sheath material extrusion, the system generates a set of instructions after monitoring the melt index deviation. For example, in the case of high deviation, the side feeder speed is adjusted to 50 rpm and the temperature of the second heating zone is corrected to 210°C. Layered distribution ensures orderly execution of instructions. The temperature instruction is first transmitted to the main module of the control system, and then the proportioning instruction is sent to the additive mixing module to achieve synchronous operation.

[0091] Understandably, this adjustment mechanism also applies to preform extrusion production. For example, when a low melt index indicates excessively high viscosity, the instruction set includes increasing the rotation speed to 70 rpm and raising the temperature zone value to enhance flow. This method improves the stability of the extrusion process through deviation analysis and instruction generation. Furthermore...

[0092] In one possible implementation, the system integrates a sensor network to support deviation monitoring, for example, by using an online rheometer to acquire melt data in real time and calculating the deviation value using an algorithm. This process ensures the accuracy of command generation and reduces material waste in actual operation.

[0093] For example, in plastic granule extrusion scenarios, after synchronous execution, temperature zoning control ensures a uniform distribution of melt temperature, while dynamic adjustment of additive ratios optimizes product toughness, thereby improving efficiency without altering the overall production process. In another embodiment, for multi-layer co-extruders, the adjustment instruction set is extended to multiple feeders, deviation analysis covers the melt index of each layer, and a layered distribution mechanism transmits instructions hierarchically, ensuring synchronous control of the temperature and ratio of different materials to achieve precise production of complex products.

[0094] Step S107: After executing the adjustment command, continuously update and obtain new pressure and speed data through feedback data, and recalculate the melt state index in combination with the target value regression detection logic to determine whether it has returned to the optimal processing state range. At the same time, store historical data for reference for subsequent optimization.

[0095] Through feedback data looping, new pressure and speed updates and temperature variable integration are obtained after the execution of the adjustment command. Using the updated pressure and speed and integrated temperature variables, combined with target value regression (fitting historical target values) and anomaly correction (subtracting noise offsets), the melt state index is recalculated. The melt state index is obtained by a weighted average of pressure, speed, and temperature variables. Based on the melt state index, regression detection logic is applied to fit a trend line using a linear regression model, and state interval judgment is performed by comparing with the preset optimal interval boundary to determine whether it has regressed to the optimal processing interval. The state interval judgment result is obtained, and historical data is stored for subsequent optimization reference to obtain the optimized processing state.

[0096] In one implementation, after the adjustment command is executed, the system continuously updates the pressure and speed data through a feedback data loop mechanism.

[0097] Specifically, the cycle first obtains real-time pressure and rotation speed values ​​from the sensors of the injection molding equipment, which are derived from melt flow monitoring during the injection process.

[0098] For example, pressure data can be acquired via a pressure sensor installed inside the injection cylinder, while rotational speed data is provided by a screw drive motor. The system compares this new data with preset target values, ensuring that the data is updated at a frequency of no less than once per second to maintain the continuity of the processing. In this way, dynamic tracking of the melt state is achieved, avoiding processing deviations caused by sudden changes. Furthermore, the system processes the acquired data in conjunction with the logic of target value regression detection.

[0099] It should be noted that the target value regression detection logic refers to a comparison mechanism based on historical target values ​​and current data, used to assess whether the data is approaching a predefined ideal state.

[0100] In one possible implementation, the logic employs a linear regression model, using new pressure and speed data as input variables to calculate a vector of deviations from the target value.

[0101] For example, if the target pressure is 100 MPa and the current pressure is 95 MPa, the deviation is -5 MPa. The system generates a regression curve based on this to predict whether the data will revert to the target range in subsequent cycles. This detection helps identify potential problems early, such as pressure fluctuations caused by abnormal melt viscosity, thus providing a basis for adjustments. The core of this process lies in the iterative calculation of the deviation vector, ensuring that each cycle optimizes the model parameters based on the previous result, achieving higher prediction accuracy. In injection molding, this logic can be applied to the melting process of different materials, such as polyethylene or polypropylene, demonstrating its versatility within the same field. Through this detection, the system can achieve precise control of melt flow, reducing scrap rates.

[0102] Preferably, when recalculating the melt state index, the system integrates pressure, rotational speed, and regression test results. The melt state index is a comprehensive indicator that reflects the fluidity and stability of the melt. Its calculation process includes multiplying the pressure value by a weighted sum of rotational speed coefficients and then dividing by a standardization factor.

[0103] Specifically, assuming the pressure value is P and the rotational speed is R, the exponent can be expressed as (P * k1 + R * k2) / S, where k1 and k2 are empirical coefficients, and S is a standardized constant. This calculation is based on feedback data to ensure that the exponent is updated in real time.

[0104] In one embodiment, for high-temperature melt processing scenarios, the system first filters the data to remove noise, and then applies the above formula to calculate the index value. This method not only covers standard injection molding processes but is also applicable to variable-speed injection scenarios, demonstrating the flexibility of the technology. Through detailed calculation steps, the system can accurately quantify changes in the melt state, providing a reliable basis for subsequent judgments. In another embodiment, determining whether the melt state has returned to the optimal processing state range involves comparing the calculated melt state index with a preset range. The optimal processing state range refers to an index value within a specific range, such as between 0.8 and 1.2, which is determined based on historical processing experience. The system uses threshold checking logic to make a judgment; if the index value falls within the range, the return is confirmed to be successful; otherwise, an alarm is triggered or further adjustments are made.

[0105] For example, on a continuous injection molding production line, when the index returns from 1.3 to 1.0, the system records this change to confirm that the processing status is stable. This judgment process enhances the robustness of the system and can be applied to mass production or prototype testing scenarios in the injection molding field.

[0106] Understandably, historical data is also stored for future optimization. This historical data includes past pressure, speed, and index values, and is stored in a database for easy retrieval.

[0107] In one embodiment, the system employs a time-series storage method, with each batch of processing data timestamped for easy subsequent analysis. By referencing this data, the optimization process can adjust empirical coefficients, such as updating the k1 value in the next calculation to improve the accuracy of the exponent. This storage mechanism supports long-term optimization, ensuring that the system's performance gradually improves within the same domain of injection molding.

[0108] The above are only some preferred embodiments of the present invention, but the present invention is not limited thereto, and many improvements and modifications can be made. Any improvements and modifications made based on the basic principles of the present invention should be considered to fall within the protection scope of the present invention.

Claims

1. A method for real-time monitoring and control of the melt state in an extruder, characterized in that, include: The initial data sequence is obtained by acquiring melt-related data through an extruder. The initial data sequence is processed to generate a smoothed base dataset; For the initial data sequence, a filtering method is used to remove high-frequency interference signals, resulting in a filtered data sequence. Fluctuation features are extracted from the filtered data sequence, and the difference between adjacent data points is calculated as an amplitude index. If the amplitude index exceeds a preset threshold, outliers are marked, generating a labeled data sequence. Relevant parameters are adjusted based on the labeled data sequence, and data values ​​are synchronously optimized through a control algorithm to obtain a smoothed base dataset. The smoothed base dataset is used as input for subsequent calculations of the melt apparent viscosity approximation sequence, ensuring the continuity and consistency of data processing. The melt apparent viscosity approximation sequence is calculated based on the base dataset, and fluctuation features are extracted. The melt viscosity fluctuation coefficient is determined using statistical methods based on the approximate value sequence; melt performance parameters are acquired through a measuring device, and the melt elastic component characterization value is calculated; the melt viscosity fluctuation coefficient and the melt elastic component characterization value are fused to generate a comprehensive melt state index; based on the deviation of the comprehensive melt state index, a set of control commands is generated and transmitted to the control system for adjustment. The data is updated cyclically by feedback data, and the comprehensive melt state index is recalculated to determine whether it has returned to the preset processing state range.

2. The method for real-time monitoring and control of the melt state in an extruder as described in claim 1, characterized in that, The process of acquiring melt-related data through the extruder to obtain an initial data sequence includes: Real-time acquisition of pressure data and speed parameters using a pressure sensor and speed monitoring device built into the extruder die to generate an initial data sequence; Low-pass filtering of the initial data sequence to cut off high-frequency noise by setting a cutoff frequency to remove interference signals, resulting in a filtered data sequence; Extracting fluctuation characteristics from the filtered data sequence and calculating the difference between adjacent data points as an amplitude index; Marking outliers if the amplitude index exceeds a preset threshold, generating a labeled data sequence; Adjusting the speed parameters based on the labeled data sequence, and simultaneously optimizing the pressure value using proportional-integral-derivative control to obtain a smoothed pressure and speed baseline dataset; Using the smoothed pressure and speed baseline dataset as the basis for subsequent calculations to ensure data accuracy and stability.

3. The method for real-time monitoring and control of the melt state in an extruder as described in claim 1, characterized in that, The step of calculating an approximate sequence of apparent melt viscosity based on the base dataset and extracting fluctuation features includes: extracting melt flow parameters from the smoothed base dataset; processing the melt flow parameters using fluid dynamics calculation methods to obtain a preliminary viscosity estimate through the ratio of pressure to rotational speed; adjusting the preliminary viscosity estimate in conjunction with real-time melt strength monitoring characteristics to determine an approximate sequence of apparent melt viscosity; analyzing fluctuation patterns for the approximate sequence and determining the direction of deviation through the magnitude of sequence value changes; synchronously recording fluctuation features based on the direction of deviation, and using the fluctuation features as input for subsequent fluctuation coefficient calculations to ensure the integrity and reliability of the fluctuation features.

4. The method for real-time monitoring and control of the melt state in an extruder as described in claim 1, characterized in that, The step of determining the melt viscosity fluctuation coefficient using statistical methods for the approximate value sequence includes: obtaining continuous data segments from the melt apparent viscosity approximate value sequence; dividing the data segments using a sliding time window to obtain a data subset within the window; calculating the mean and standard deviation for the data subset within the window to obtain fluctuation characteristic values; determining a preliminary deviation index by using the absolute difference between the fluctuation characteristic value and the mean of the overall sequence as a quantitative assessment of the deviation magnitude; if the preliminary deviation index exceeds a preset threshold, adjusting the window size and recalculating the fluctuation characteristic value to obtain an optimized deviation index; generating the melt viscosity fluctuation coefficient by integrating the optimized deviation index with preset quantitative logic; and determining real-time control parameters by matching the fluctuation coefficient with historical control data in a preset database.

5. The method for real-time monitoring and control of the melt state in an extruder as described in claim 1, characterized in that, The step of acquiring melt performance parameters and calculating the melt elastic component characterization value through a measuring device includes: acquiring elastic modulus and viscous modulus data at specific locations in the extrusion channel using a melt performance measuring device; calculating the ratio between the elastic modulus and viscous modulus data to determine the ratio; obtaining temperature influence correction parameters from a preset temperature curve using melt state index calculation rules and the ratio to obtain preliminary values ​​of the elastic component; applying component characterization rules to the preliminary values ​​of the elastic component to determine the melt elastic component characterization value; and using the melt elastic component characterization value as an important basis for comprehensively evaluating the melt state to ensure the scientificity and accuracy of the evaluation results.

6. The method for real-time monitoring and control of the melt state in an extruder as described in claim 1, characterized in that, The process of integrating the melt viscosity fluctuation coefficient and the melt elastic component characterization value to generate a comprehensive melt state index includes: acquiring the melt viscosity fluctuation coefficient and the melt elastic component characterization value; applying a first weight to the fluctuation coefficient and a second weight to the characterization value using a preset weighting rule, and then performing a weighted summation to obtain the comprehensive melt state index; extracting a deviation vector from the directional difference between the index and the standard value using intelligent deviation direction discrimination for the comprehensive melt state index; calculating the magnitude of the deviation vector through deviation amplitude quantification evaluation to determine the deviation result; if the deviation result falls into the flow stability decrease range, generating a trigger signal based on the distribution of the deviation vector within the range; and extracting a compensation factor based on the trigger signal to generate control parameters.

7. The method for real-time monitoring and control of the melt state in an extruder as described in claim 1, characterized in that, The step of generating a set of control commands based on the deviation of the comprehensive melt state index and transmitting them to the control system for adjustment includes: acquiring the comprehensive melt state index through a sensor, comparing it with a preset threshold to determine the deviation; generating a set of adjustment commands for adjusting the side feeder speed based on the deviation, and simultaneously generating a set of adjustment commands for correcting the heating temperature zones; transmitting the set of adjustment commands to the extruder control system through a hierarchical distribution mechanism to synchronously execute precise temperature zone control; dynamically proportioning the additive ratio during the execution of precise temperature zone control to obtain real-time calibration results of the material viscosity; and completing the synchronous operation if the calibration results meet the requirements of the comprehensive melt state index to ensure the stability of the control effect.

8. The method for real-time monitoring and control of the melt state in an extruder as described in claim 1, characterized in that, The step of updating data cyclically through feedback data and recalculating the comprehensive melt state index includes: obtaining new pressure and speed update data and integrated temperature variable data after the adjustment command is executed through feedback data cyclically; using the pressure and speed update data and the integrated temperature variable data, fitting the data to historical target values, and combining abnormal deviation correction to deduct noise offset, recalculating the comprehensive melt state index; based on the comprehensive melt state index, fitting a trend line through a linear regression model to determine the regression trend; determining whether it has regressed to a preset processing state range by comparing it to the preset optimal range boundary; and storing the determination result along with historical data for subsequent optimization reference.

9. The method for real-time monitoring and control of the melt state in an extruder as described in claim 1, characterized in that, The step of generating a set of control instructions based on the deviation of the comprehensive melt state index includes: acquiring the comprehensive melt state index, comparing it with a preset threshold, and determining the direction and magnitude of the deviation; generating a set of adjustment instructions for equipment operating parameters based on the direction and magnitude of the deviation; determining temperature zone control parameters and additive ratio parameters through the set of adjustment instructions; generating specific control instructions based on the temperature zone control parameters and additive ratio parameters; transmitting the control instructions to the control system through a hierarchical distribution mechanism and executing the adjustment operation synchronously; and acquiring the adjusted feedback data to verify whether the comprehensive melt state index meets the preset requirements, ensuring the closed-loop nature of the control process.

Images