Method and system for intelligently regulating and controlling production and processing parameters of polycarbonate material

Abstract

The invention discloses an intelligent regulation and control method and system for polycarbonate material production and processing parameters, and relates to the technical field of polycarbonate processing regulation and control, and by performing interval mapping judgment on a heat supply state, the system not only can realize automatic increase and decrease regulation on the feeding amount, but also establishes a state fault-tolerant belt based on shear heat balance. And particularly, when the feeding correction factor is in an overload critical zone, the system can be automatically lowered for feeding, so that the risks of mechanical wear, energy consumption increase and final product defect caused by semi-molten particle clusters are avoided. And meanwhile, the feeding rate is increased in time when heat supply is surplus, and energy waste and low-efficiency operation are avoided. The regulation and control mode has an elastic interval regulation capability and a fault defense mechanism, so that the system can maintain an optimal operation window under the conditions of performance and load of different batches of granules, and by performing interval mapping judgment on the heat supply state, the system not only can realize automatic increase and decrease regulation on the feeding amount, but also can realize automatic control on the feeding amount. A state fault-tolerant band based on shear heat balance is also established.

Description

Technical Field

[0001] The present invention relates to the technical field of polycarbonate processing control, and in particular to a method and system for intelligently controlling polycarbonate material production processing parameters. Background Art

[0002] The widespread use of polycarbonate in high-precision applications such as optical lenses, safety sheets, and medical products places higher demands on thermal state control, melt uniformity management, and raw material adaptability during processing. In specific processing techniques, polycarbonate materials are typically fed in pellet form, sheared and melted by a screw, and then formed. The dynamic matching between the screw's shear heat generation capacity and the pellet state response capability is the key to achieving consistent finished product performance and process stability.

[0003] In traditional polycarbonate processing control systems, parameters such as feed rate, screw speed, and temperature zone temperature are often controlled based on preset curves or static empirical models, failing to account for the dynamic thermal response of the pellets in the shear zone. In particular, when there are slight fluctuations in raw material particle size, bulk density, or thermal conductivity, the system cannot immediately identify changes in the pellets' melting state. This can cause some pellets to fail to fully absorb heat within the designated area, resulting in a "half-melt, half-cold" mass. This phenomenon not only leads to unstable shear loads but can also cause problems such as uneven melt flow and residual stress within the product.

[0004] The core of the above-mentioned uneven melting problem is that the processing system cannot effectively perceive the impact of fluctuations in the physical properties of the pellets on the actual melting rate, and thus cannot convert it into a fine-tuning signal for the feeding rhythm. Typically, when the pellet size is too large, the stacking angle increases, or the heat absorption capacity of the pellets decreases, the shear heat per unit time cannot be completely converted into melting energy, and some particles will enter the downstream molding area in an incompletely melted state, causing an instantaneous increase in shear resistance. This not only affects production stability, but also limits the high-level performance of material properties. Therefore, there is an urgent need for a microparticle-scale melting behavior control mechanism that integrates the shear heat index and pellet state perception to achieve dynamic adaptive optimization of the feeding rhythm. Summary of the Invention

[0005] In view of the deficiencies in the prior art, the present invention provides a method and system for intelligently controlling the production and processing parameters of polycarbonate materials, which solves the problems mentioned in the background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: an intelligent control system for polycarbonate material production and processing parameters, including a raw material attribute identification and processing module, a pellet shearing and melting assessment module, a stirring zone torque dynamic feedback module, a feed control instruction generation module, a raw material stability learning module, and a risk index judgment and feedback module;

[0007] The raw material attribute recognition and processing module collects the production parameters of polycarbonate materials through sensors, fits them into the original data set YM, and performs preprocessing to obtain the processing parameter set GW;

[0008] The pellet shear and melting evaluation module constructs a shear heat generation model based on the obtained processing parameter set GW, calculates the shear heat generation rate Qsh of the pellet per unit time, and evaluates the pellet melting rate LRm;

[0009] The stirring zone torque dynamic feedback module monitors the mechanical load state of the melting zone in real time, collects the torque signal fluctuation in the stirring zone, obtains the torque value TV, and constructs the melting resistance disturbance intensity Tres;

[0010] The feed control instruction generation module obtains the feed correction factor ΔFin based on the obtained shear heat generation rate Qsh and pellet melting rate LRm, combined with the melting resistance disturbance intensity Tres;

[0011] The raw material stability learning module corrects the shear heat generation rate Qsh by the feed correction factor ΔFin to obtain the new shear heat generation rate Qtar after dynamic learning;

[0012] The risk index judgment and feedback module calculates the risk index Risk through the new shear heat generation rate Qtar and the torque value TV to judge the abnormality of the feed melting state.

[0013] Preferably, the raw material attribute identification and processing module includes a multi-parameter acquisition and calibration unit and a feature pre-processing unit;

[0014] The multi-parameter acquisition and calibration unit collects the production parameters of polycarbonate materials through sensors and detection devices, including pellet bulk density Ldp, pellet average particle size Ldm, pellet thermal conductivity Drp, pellet specific heat capacity Dbc, pellet initial surface temperature To, stacking angle θ, screw speed Vs and melt temperature Tm, and fits them into the original data set YM;

[0015] Among them, the bulk density Ldp of the pellets is calculated and obtained by the hopper bottom loading weighing system and the laser volume measurement device;

[0016] The bulk density of the pellets, Ldp, is obtained by the following formula:

[0017] ;

[0018] Where MB represents the total mass of the hopper after filling, MO represents the mass of the empty hopper, and VB represents the volume of the pile constructed by the laser point cloud;

[0019] The average particle size Ldm of the granular material is collected and obtained using an image recognition particle size analyzer;

[0020] The average particle size Ldm of the granules is obtained by the following formula:

[0021] ;

[0022] Where di represents the center particle size value of the i-th particle size interval, which represents the representative diameter of the particles in this interval; fi represents the frequency of the particles in the i-th particle size interval, that is, the proportion of particles in this particle size interval; n represents the total number of intervals into which the particle size distribution is divided. For example, if it is divided into 10 particle size segments, then n=10;

[0023] The thermal conductivity Drp and specific heat capacity Dbc of the pellets are derived and obtained through the online thermal response detection device;

[0024] The thermal conductivity of the pellets Drp is obtained by the following formula:

[0025] ;

[0026] Where Rq represents the applied heat flux, RT represents the measured temperature gradient, and Rh represents the measured thickness of the granular accumulation layer;

[0027] The specific heat capacity Dbc of the pellets is obtained by the following formula:

[0028] ;

[0029] Where Qin represents the short-term heat application, QM represents the mass of the heated pellets, and ΔT represents the temperature rise per unit time;

[0030] The initial surface temperature To of the pellets and the melt temperature Tm were acquired by infrared thermal imaging array and non-contact infrared point thermometer;

[0031] The stacking angle θ is obtained by laser profile measurement and image fitting;

[0032] The screw speed Vs is directly collected through the encoder of the host system;

[0033] The feature preprocessing unit cleans and normalizes the acquired original data set YM to obtain the processing parameter set GW;

[0034] Cleaning includes outlier removal, which is done by using the triple standard deviation method to remove outliers from the original dataset YM and filling them with linear interpolation;

[0035] Normalization processing is performed by using the Max-Min normalization method to process the original data set YM to obtain the processing parameter set GW;

[0036] The processing parameter set GW is obtained by the following formula:

[0037] ;

[0038] Where GWo represents the oth data item in the processing parameter set GW, YMo represents the oth data item in the original data set YM, minYMo represents the valley value of the oth data item in the original data set YM, and maxYMo represents the peak value of the oth data item in the original data set YM.

[0039] Preferably, the pellet shear and melting assessment module includes a shear heat generation calculation unit and a melting rate assessment and conversion unit;

[0040] The shear heat generation calculation unit constructs a shear heat generation model through the processing parameter set GW and calculates the apparent viscosity Napp;

[0041] The apparent viscosity Napp is obtained by the following formula:

[0042] ;

[0043] Where No represents zero shear viscosity, λ0 represents time relaxation constant, Zn represents shear index, and 0<Zn<1, reflecting shear thinning characteristics, Sve represents shear rate, and Sve=k1*Vs, k1 represents the proportional coefficient related to screw structure and extrusion cavity, exp represents exponential function, α T represents the thermal sensitivity factor, the inhibition of the reaction temperature on the viscosity, and Tre represents the reference temperature;

[0044] According to the obtained apparent viscosity Napp, the shear heat generation rate Qsh is calculated;

[0045] The shear heat generation rate Qsh is obtained by the following formula:

[0046] ;

[0047] Where α1 represents the empirical heat transfer factor, which is obtained by the ratio of the thermal conductivity Drp of the pellet to the specific heat capacity Dbc of the pellet, and Veff represents the equivalent shear volume;

[0048] The equivalent shear volume is obtained by the following formula:

[0049] ;

[0050] Where LC represents the flow channel structural constant, represents the approximate volume of the screw shear chamber, and tan() represents the sine function.

[0051] Preferably, the melting rate evaluation and conversion unit evaluates the pellet melting rate LRm based on the obtained shear heat generation rate Qsh in combination with the pellet specific heat capacity Dbc;

[0052] The unit mass heat melting requirement Hreq is obtained by the specific heat capacity Dbc of the pellets;

[0053] The unit mass hot melt requirement Hreq is obtained by the following formula:

[0054] Hreq=Dbc*(Tm-To);

[0055] The pellet melting rate LRm is obtained by the following formula:

[0056] ;

[0057] Where sp represents the melting heat conversion efficiency;

[0058] The melting heat conversion efficiency sp is obtained by the following formula:

[0059] ;

[0060] Where Rh represents the measured thickness of the pellet stacking layer, Φ1 represents the unit heat capacity density, which is obtained by multiplying the bulk density Ldp and the pellet specific heat capacity Dbc, and Φ2 represents the unit thermal conductivity path coefficient, which is obtained by the ratio of the pellet thermal conductivity Drp to the average pellet size Ldm.

[0061] Preferably, the stirring zone torque dynamic feedback module includes a torque signal acquisition and decoding unit and a disturbance intensity index calculation unit;

[0062] The torque signal acquisition and decoding unit collects the torque signal data of the main shaft through the dynamic torque sensor set in the main shaft of the stirring zone, and collects the torque value TV of the torque sampling point Td in each unit time band;

[0063] The disturbance intensity index calculation unit constructs the first-order derivative, variation rate and normalized index based on the torque value TV to extract the melting resistance disturbance intensity Tres;

[0064] The torque change rate ΔTV is obtained by constructing the first-order derivative;

[0065] The torque change rate ΔTV is obtained by the following formula:

[0066] ;

[0067] Where TV(t) represents the torque value at time t, TV(t-Δt) represents the torque value at time t-Δt, and Δt represents the time interval;

[0068] The melting resistance disturbance intensity Tres is obtained by the following formula:

[0069] ;

[0070] In the formula, β1 represents the rate change adjustment factor, β2 represents the fluctuation amplitude adjustment factor, represents the sliding average of the torque, and σTV represents the standard deviation of the torque signal.

[0071] Preferably, the feed control instruction generation module includes a state deviation calculation unit and a multi-factor fusion calculation unit;

[0072] The state deviation calculation unit compares the obtained shear heat generation rate Qsh with the theoretical generation rate Qget, calculates the melting heat deviation index Phea, and determines the heating state of the current working condition;

[0073] The melting heat deviation index Phea is obtained by the following formula:

[0074] ;

[0075] The heating status of the working condition is obtained by matching in the following ways:

[0076] When the melting heat deviation index Phea ≥ 0, it means insufficient heat supply and the feeding rhythm is slowed down;

[0077] When the melting heat deviation index Phea is less than 0, it means that there is excess heat supply and the feeding rhythm should be adjusted upward;

[0078] Compare the obtained pellet melting rate LRm with the actual feed rate Lar to obtain the melting limit index Lme;

[0079] The melting limit index Lme is obtained by the following formula:

[0080] .

[0081] When the melting limit index Lme>0, it means that the feeding is too fast and exceeds the upper limit of melting processing, and the feeding needs to be slowed down.

[0082] Preferably, the multi-factor fusion calculation unit combines the obtained melting heat deviation index Phea and melting limit index Lme with the melting resistance disturbance intensity Tres to calculate the feed correction factor ΔFin;

[0083] The feed correction factor ΔFin is obtained by the following formula:

[0084] ;

[0085] Where, Respectively represent the preset weight values ​​of melting heat deviation index Phea, melting limit index Lme and melting resistance disturbance intensity Tres, and ;

[0086] Adjust the output strategy through the feed correction factor ΔFin;

[0087] When 0<feed correction factor ΔFin<0.3, it means that the current heat supply is sufficient and production is stable, so increase the feed rate and improve production capacity;

[0088] When 0.3≤feed correction factor ΔFin<0.7, it means that the heating is normal and the current feeding rhythm is maintained;

[0089] When 0.7≤feed correction factor ΔFin<1.0, it indicates abnormal heating and is approaching the upper limit of load. Reduce the feed rate to relieve the load and prevent unmelted agglomeration.

[0090] Preferably, the raw material stability learning module aggregates and analyzes the feed correction factor ΔFin of multiple historical cycles with the corresponding shear heat generation rate Qsh and the melting resistance disturbance intensity Tres, constructs a heat supply deviation response model for the current pellets, and derives the shear heat correction factor θco to generate a new shear heat generation rate Qtar;

[0091] The shear heat correction factor θco is obtained by the following formula:

[0092] ;

[0093] Where c1 represents the response adjustment coefficient of disturbance to feed deviation, c2 represents the response adjustment coefficient of disturbance to melting resistance, ΔFin(t) represents the feed correction factor ΔFin at time t, and Tres(t) represents the disturbance intensity of melting resistance at time t.

[0094] The shear heat generation rate Qsh is corrected by the obtained shear heat correction factor θco to generate a new shear heat generation rate Qtar;

[0095] The new shear heat generation rate Qtar is obtained by the following formula:

[0096] ;

[0097] Where Qtar(t) represents the new shear heat generation rate at time t, Qsh(t) represents the shear heat generation rate at time t, and c3 represents the thermal response sensitivity coefficient.

[0098] Preferably, the risk index judgment and feedback module constructs a thermal power deviation index RpQ and a torque anomaly index RTV through the new shear heat generation rate Qtar and the torque value TV;

[0099] The thermal power deviation index RpQ is obtained by the following formula:

[0100] ;

[0101] The torque abnormality index RTV is obtained by the following formula:

[0102] ;

[0103] Where D1 represents the instantaneous torque change adjustment coefficient, D2 represents the fluctuation amplitude adjustment coefficient, d represents the integral sign, TV(t) represents the torque value at time t, and σTV represents the standard deviation of the torque signal;

[0104] The risk index Risk is calculated by obtaining the thermal power deviation index RpQ and the torque abnormality index RTV, and compared with the preset risk threshold Rth to determine the risk status;

[0105] The risk index Risk is obtained by the following formula:

[0106] ;

[0107] Where B1 and B2 represent the preset weight values ​​of thermal power deviation index RpQ and torque abnormality index RTV respectively;

[0108] The risk status is obtained by matching:

[0109] When the risk index Risk ≤ risk threshold Rth, it means that there is no risk and it remains within a safe range;

[0110] When the risk index Risk>risk threshold Rth, it means there is risk and anomaly occurs, and the screw speed Vs and melt temperature Tm are adjusted to obtain new screw speed nVs and new melt temperature nTm;

[0111] The new screw speed nVs is obtained by the following formula:

[0112] ;

[0113] Where Rsv represents the speed reduction factor, which is manually controlled and has a value of 0.05;

[0114] The new melt temperature nTm is obtained by the following formula:

[0115] ;

[0116] Where ΔTM represents the heating amount in the temperature zone, which is controlled manually.

[0117] A method for intelligently controlling production parameters of polycarbonate materials comprises the following steps:

[0118] Step 1: The raw material attribute recognition and processing module collects the production parameters of the polycarbonate material through sensors, fits them into the original data set YM, and performs preprocessing to obtain the processing parameter set GW;

[0119] Step 2: The pellet shear and melting evaluation module constructs a shear heat generation model based on the obtained processing parameter set GW, calculates the shear heat generation rate Qsh per unit time of the pellet, and evaluates the pellet melting rate LRm;

[0120] Step 3: The stirring zone torque dynamic feedback module monitors the mechanical load state of the melting zone in real time, collects the torque signal fluctuations in the stirring zone, obtains the torque value TV, and constructs the melting resistance disturbance intensity Tres;

[0121] Step 4: The feed control instruction generation module combines the obtained shear heat generation rate Qsh and pellet melting rate LRm with the melting resistance disturbance intensity Tres to obtain the feed correction factor ΔFin;

[0122] Step 5: The raw material stability learning module corrects the shear heat generation rate Qsh by the feed correction factor ΔFin to obtain the new shear heat generation rate Qtar after dynamic learning;

[0123] Step 6: The risk index judgment and feedback module calculates the risk index Risk through the new shear heat generation rate Qtar and the torque value TV to judge the abnormality of the feed melting state.

[0124] The present invention provides a method and system for intelligently controlling the production and processing parameters of polycarbonate materials, which has the following beneficial effects:

[0125] (1) During system operation, the raw material property identification and processing module collects basic physical parameters of polycarbonate pellets, such as pellet bulk density Ldp, average pellet size Ldm, pellet thermal conductivity Drp, pellet specific heat capacity Dbc, pellet initial surface temperature To, stacking angle θ, screw speed Vs, and melt temperature Tm, online for the first time and converts them into a unified input processing parameter set GW. This preprocessing mechanism provides a pellet-level thermal response modeling foundation for downstream shear heat calculation and melting rate assessment, thereby enabling interpretable analysis of pellet behavior and process modeling, significantly improving the system's perception accuracy of raw material fluctuations and modeling flexibility.

[0126] (2) A dynamic apparent viscosity model driven by shear rate was constructed by integrating the screw speed and extrusion structure parameters through the shear heat generation calculation unit. The melt heat deviation index Phea and the reference temperature were incorporated into the feed correction factor ΔFin, achieving dynamic adaptation of viscosity to the local thermal state of the pellets. Unlike traditional rough estimates based solely on processing temperature, this model can adjust the prediction of energy conversion resistance in real time based on raw material characteristics and operating conditions. It has a stronger ability to depict the authenticity of pellet response and can provide accurate support for subsequent shear heat power calculations.

[0127] (3) By performing interval mapping judgment on the heating status, the system can not only realize automatic increase or decrease adjustment of the feed rate, but also establish a state fault tolerance band based on shear heat balance. In particular, when the feed correction factor is in the overload critical zone, the system will automatically reduce the feed rate to avoid mechanical wear, increased energy consumption and the risk of final product defects caused by semi-molten pellets. At the same time, the feed rate is increased in time when there is excess heat to avoid energy waste and inefficient operation. This control method has flexible interval adjustment capabilities and fault protection mechanisms, enabling the system to maintain the optimal operating window under different batches of pellet performance and load conditions.

[0128] (4) By comprehensively quantifying the thermophysical properties of polycarbonate pellets before processing, using several uncommon parameters such as pellet bulk density Ldp, average pellet size Ldm, pellet thermal conductivity Drp, and pellet specific heat capacity Dbc, the system established a real-time calculation model for shear heat generation rate and melting rate. Compared with the traditional method that relies on "experience + single-point temperature" to estimate processing status, this method establishes a complete chain from raw material properties to processing behavior to feedback strategy, ensuring that the system has the ability to truly characterize, accurately predict, and quantify parameters of pellet status, significantly improving the scientific nature and response accuracy of the control process. BRIEF DESCRIPTION OF THE DRAWINGS

[0129] Figure 1 This is a block diagram and flow chart of an intelligent control system for polycarbonate material production and processing parameters according to the present invention;

[0130] Figure 2 This is a schematic diagram of the steps of a method for intelligently controlling the production and processing parameters of polycarbonate materials according to the present invention;

[0131] Figure 3 A schematic diagram of the risk index acquisition process of the present invention;

[0132] Figure 4 It is a line graph of the feed correction factor of the present invention. DETAILED DESCRIPTION

[0133] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0134] Example 1

[0135] The present invention provides a method and system for intelligently controlling the production and processing parameters of polycarbonate materials. Figures 1 to 4 , including raw material attribute identification and processing module, pellet shear and melting evaluation module, stirring zone torque dynamic feedback module, feed control instruction generation module, raw material stability learning module and risk index judgment and feedback module;

[0136] The raw material attribute recognition and processing module collects the production parameters of polycarbonate materials through sensors, fits them into the original data set YM, and performs preprocessing to obtain the processing parameter set GW;

[0137] The pellet shear and melting evaluation module constructs a shear heat generation model based on the obtained processing parameter set GW, calculates the shear heat generation rate Qsh of the pellet per unit time, and evaluates the pellet melting rate LRm;

[0138] The stirring zone torque dynamic feedback module monitors the mechanical load state of the melting zone in real time, collects the torque signal fluctuation in the stirring zone, obtains the torque value TV, and constructs the melting resistance disturbance intensity Tres;

[0139] The feed control instruction generation module obtains the feed correction factor ΔFin based on the obtained shear heat generation rate Qsh and pellet melting rate LRm, combined with the melting resistance disturbance intensity Tres;

[0140] The raw material stability learning module corrects the shear heat generation rate Qsh by the feed correction factor ΔFin to obtain the new shear heat generation rate Qtar after dynamic learning;

[0141] The risk index judgment and feedback module calculates the risk index Risk through the new shear heat generation rate Qtar and the torque value TV to judge the abnormality of the feed melting state.

[0142] In this example, the raw material property identification and processing module, for the first time, collects online the fundamental physical parameters of polycarbonate pellets, including pellet bulk density Ldp, average pellet size Ldm, pellet thermal conductivity Drp, pellet specific heat capacity Dbc, pellet initial surface temperature To, stacking angle θ, screw speed Vs, and melt temperature Tm. These parameters are then converted into a unified set of inputtable processing parameters GW. This preprocessing mechanism provides a pellet-level thermal response modeling foundation for downstream shear heat calculations and melt rate assessments, enabling interpretable analysis of pellet behavior and process modeling, significantly improving the system's perception of raw material fluctuations and modeling flexibility.

[0143] By building a shear heat generation model and a melting rate assessment mechanism, combined with the dynamic variation of the stirring zone torque, the system can capture the load disturbances caused by the semi-molten, semi-cold state of the pellets without relying on explicit temperature control, thereby indirectly diagnosing the pellets' melting behavior. Feed correction instructions generated based on these results dynamically adapt the feeding strategy to the pellets' thermal response, forming an intelligent feed control mechanism from an energy-material balance perspective, avoiding the melting deviations and product quality fluctuations caused by the traditional fixed feeding rhythm.

[0144] Building on feed corrections, the system further utilizes a raw material stability learning module to analyze historical disturbances and dynamically update target heat input requirements based on shear heat performance trends. This allows for intelligent migration of shear heat supply strategies and batch adaptability in multi-batch, variable raw material environments. When the system reaches a critical state, the risk index assessment and feedback module proactively triggers compensatory controls in non-feeding dimensions, such as speed reduction, temperature zone heating, and time-delay buffering strategies. This significantly enhances the system's resilience and fault avoidance capabilities under abnormal fluctuations.

[0145] Example 2

[0146] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the raw material attribute recognition and processing module includes a ,multi-parameter acquisition and calibration unit and a feature pre-processing unit;

[0147] The multi-parameter acquisition and calibration unit collects the production parameters of polycarbonate materials through sensors and detection devices, including pellet bulk density Ldp, pellet average particle size Ldm, pellet thermal conductivity Drp, pellet specific heat capacity Dbc, pellet initial surface temperature To, stacking angle θ, screw speed Vs and melt temperature Tm, and fits them into the original data set YM;

[0148] Among them, the bulk density Ldp of the pellets is calculated and obtained by the hopper bottom loading weighing system and the laser volume measurement device;

[0149] The average particle size Ldm of the granular material is collected and obtained using an image recognition particle size analyzer;

[0150] The thermal conductivity Drp and specific heat capacity Dbc of the pellets are derived and obtained through the online thermal response detection device;

[0151] The initial surface temperature To of the pellets and the melt temperature Tm were acquired by infrared thermal imaging array and non-contact infrared point thermometer;

[0152] The stacking angle θ is obtained by laser profile measurement and image fitting;

[0153] The screw speed Vs is directly collected through the encoder of the host system;

[0154] The feature preprocessing unit cleans and normalizes the acquired original data set YM to obtain the processing parameter set GW;

[0155] Cleaning includes outlier removal, which is done by using the triple standard deviation method to remove outliers from the original dataset YM and filling them with linear interpolation;

[0156] Normalization processing is performed by using the Max-Min normalization method to process the original data set YM to obtain the processing parameter set GW;

[0157] The processing parameter set GW is obtained by the following formula:

[0158] ;

[0159] Where GWo represents the oth data item in the processing parameter set GW, YMo represents the oth data item in the original data set YM, minYMo represents the valley value of the oth data item in the original data set YM, and maxYMo represents the peak value of the oth data item in the original data set YM.

[0160] In this embodiment, a multi-parameter acquisition and calibration unit is provided to monitor not only traditional acquisition parameters such as temperature and rotational speed, but also real-time acquisition of non-explicit process influencing factors such as pellet bulk density, thermal conductivity, specific heat capacity, stacking angle, and average particle size. This design breaks away from the traditional process approach that relies solely on external temperature control and host parameter adjustment, and implements an active sampling mechanism that addresses the physical properties of the pellets. In particular, the introduction of stacking angle and particle size recognition methods enables the system to identify physical structural differences such as the tightness of raw material stacking, heat transfer efficiency, and heat absorption capacity, establishing an underlying architecture that "has the ability to identify raw material behavior," providing a foundational guarantee for the adaptability of subsequent control models.

[0161] By setting up an outlier elimination mechanism and a linear interpolation compensation algorithm, this embodiment effectively avoids data distortion problems caused by factors such as fluctuations in detection equipment, sudden changes in raw material input, and environmental temperature drift, and avoids the introduction of deviations when key process parameters enter the model calculation. At the same time, the Max-Min normalization mechanism is used to unify the dimensions and adjust the magnitudes of various physical parameters, so that different physical properties have a unified influence scale in subsequent models, thereby achieving balanced modeling of various raw material batches. This mechanism effectively solves the problems of inconsistent dimensions, unbalanced influences, and unstable inputs in traditional processing parameter systems, ensuring the basic quality of downstream shear heat calculations and melting rate assessments.

[0162] The acquisition and preprocessing logic designed in this embodiment not only provides an accurate and usable parameter set for subsequent modules such as shear heat modeling, feed pacing, and abnormal risk identification, but also forms the first perception chain node in the system's intelligent closed-loop control. Compared to traditional systems that rely on empirical parameter curves for set speed and temperature, this embodiment dynamically establishes a control rhythm based on the measured characteristics of the raw materials, significantly enhancing its adaptability to fluctuations in multiple raw material batches and its ability to flexibly adapt to processing strategies.

[0163] Example 3

[0164] This embodiment is explained in Example 2, please refer to Figure 1 and Figure 3 ,Specifically: the granular shear and melting evaluation module includes a shear heat generation ,calculation unit and a melting rate evaluation and conversion unit;

[0165] The shear heat generation calculation unit constructs a shear heat generation model through the processing parameter set GW and calculates the apparent viscosity Napp;

[0166] The apparent viscosity Napp is obtained by the following formula:

[0167] ;

[0168] Where No represents zero shear viscosity, λ0 represents time relaxation constant, Zn represents shear index, Sve represents shear rate, and Sve=k1*Vs, k1 represents the proportional coefficient related to screw structure and extrusion cavity, exp represents exponential function, α T represents the thermal sensitivity factor, Tre represents the reference temperature;

[0169] According to the obtained apparent viscosity Napp, the shear heat generation rate Qsh is calculated;

[0170] The shear heat generation rate Qsh is obtained by the following formula:

[0171] ;

[0172] Where α1 represents the empirical heat transfer factor (obtained by the ratio of the thermal conductivity Drp of the pellet to the specific heat capacity Dbc of the pellet), and Veff represents the equivalent shear volume;

[0173] The equivalent shear volume is obtained by the following formula:

[0174] ;

[0175] Where LC represents the flow channel structure constant, and tan() represents the sine function.

[0176] The melting rate evaluation and conversion unit evaluates the pellet melting rate LRm based on the obtained shear heat generation rate Qsh and the pellet specific heat capacity Dbc;

[0177] The unit mass heat melting requirement Hreq is obtained by the specific heat capacity Dbc of the pellets;

[0178] The unit mass hot melt requirement Hreq is obtained by the following formula:

[0179] Hreq=Dbc*(Tm-To);

[0180] The pellet melting rate LRm is obtained by the following formula:

[0181] ;

[0182] Where sp represents the melting heat conversion efficiency;

[0183] The melting heat conversion efficiency sp is obtained by the following formula:

[0184] ;

[0185] Where Rh represents the measured thickness of the granular stacking layer, Φ1 represents the unit heat capacity density, and Φ2 represents the unit thermal conductivity path coefficient.

[0186] In this example, a shear-rate-driven dynamic apparent viscosity model was constructed using a shear heat generation calculation unit, integrating screw speed and extrusion structure parameters. The melt heat deviation index Phea and the reference temperature were incorporated into the feed correction factor ΔFin, enabling dynamic viscosity adaptation to the pellet's local thermal state. Unlike traditional rough estimates based solely on processing temperature, this model can adjust its prediction of energy conversion resistance in real time based on raw material characteristics and operating conditions. This model provides a more realistic depiction of pellet response and provides precise support for subsequent shear heat power calculations.

[0187] By introducing an equivalent shear volume parameter and correlating it with the pellet packing density Ldp, the packing angle θ, and the channel geometry, this embodiment incorporates the coupling factor between structural geometry and pellet packing behavior into the shear heat calculation, breaking through the limitations of the traditional "heat power = velocity × viscosity" simple product model. This design can correct the "effective heat carrier volume" involved in shear in real time based on different particle size distributions, dynamic changes in packing looseness, and other conditions, significantly improving the fit of the thermal power calculation results to the actual behavior of the raw materials, ultimately contributing to the robustness and accuracy of the system in evaluating actual heating capacity.

[0188] The melt rate assessment and conversion unit not only accurately derives the energy required to melt per unit mass based on the pellet's specific heat capacity and temperature difference, but also introduces the key intermediary variable, melt heat conversion efficiency, and exponentially couples it with microstructural parameters such as pellet stacking thickness, unit heat density, and unit thermal conductivity path. This design, for the first time, implements a dynamic modeling mechanism that feeds back microscopic thermal structural characteristics into macroscopic melt rate prediction. It can identify abnormal behaviors such as insufficient thermal conductivity, pellet agglomeration, and batch-to-batch variations in heat transfer capacity, providing a scientific and reliable basis for the system's subsequent feed control decisions.

[0189] Example 4

[0190] This embodiment is explained in Example 3, please refer to Figure 1 and Figure 4 ,Specifically: the stirring zone torque dynamic feedback module includes a torque signal ,acquisition and decoding unit and a disturbance intensity index calculation unit;

[0191] The torque signal acquisition and decoding unit collects the torque signal data of the main shaft through the dynamic torque sensor set in the main shaft of the stirring zone, and collects the torque value TV of the torque sampling point Td in each unit time band;

[0192] The disturbance intensity index calculation unit constructs the first-order derivative, variation rate and normalized index based on the torque value TV to extract the melting resistance disturbance intensity Tres;

[0193] The torque change rate ΔTV is obtained by constructing the first-order derivative;

[0194] The torque change rate ΔTV is obtained by the following formula:

[0195] ;

[0196] Where TV(t) represents the torque value at time t, TV(t-Δt) represents the torque value at time t-Δt, and Δt represents the time interval;

[0197] The melting resistance disturbance intensity Tres is obtained by the following formula:

[0198] ;

[0199] In the formula, β1 represents the rate change adjustment factor, β2 represents the fluctuation amplitude adjustment factor, represents the sliding average of the torque, and σTV represents the standard deviation of the torque signal.

[0200] The feed control instruction generation module includes a state deviation calculation unit and a multi-factor fusion calculation unit;

[0201] The state deviation calculation unit compares the obtained shear heat generation rate Qsh with the theoretical generation rate Qget, calculates the melting heat deviation index Phea, and determines the heating state of the current working condition;

[0202] The melting heat deviation index Phea is obtained by the following formula:

[0203] ;

[0204] The heating status of the working condition is obtained by matching in the following ways:

[0205] When the melting heat deviation index Phea ≥ 0, it means insufficient heat supply and the feeding rhythm is slowed down;

[0206] When the melting heat deviation index Phea is less than 0, it means that there is excess heat supply and the feeding rhythm should be adjusted upward;

[0207] Compare the obtained pellet melting rate LRm with the actual feed rate Lar to obtain the melting limit index Lme;

[0208] The melting limit index Lme is obtained by the following formula:

[0209] .

[0210] The multi-factor fusion calculation unit combines the obtained melting heat deviation index Phea and melting limit index Lme with the melting resistance disturbance intensity Tres to calculate the feed correction factor ΔFin;

[0211] The feed correction factor ΔFin is obtained by the following formula:

[0212] ;

[0213] Where, Respectively represent the preset weight values ​​of melting heat deviation index Phea, melting limit index Lme and melting resistance disturbance intensity Tres, and ;

[0214] Adjust the output strategy through the feed correction factor ΔFin;

[0215] When 0<feed correction factor ΔFin<0.3, it means that the current heat supply is sufficient and production is stable, so increase the feed rate and improve production capacity;

[0216] When 0.3≤feed correction factor ΔFin<0.7, it means that the heating is normal and the current feeding rhythm is maintained;

[0217] When 0.7≤feed correction factor ΔFin<1.0, it indicates that the heating is abnormal and the load limit is approaching, so the feed rate should be reduced.

[0218] Specific examples:

[0219] Table 1: Multi-factor fusion calculation results table;

[0220] Group number Melting heat deviation index Phea Melting limit index Lme Melting resistance disturbance strength Tres Feed correction factor ΔFin Group 1 0.70 0.80 0.50 0.685 Group 2 0.55 0.60 0.35 0.517 Group 3 0.40 0.50 0.25 0.397 Group 4 0.25 0.30 0.15 0.242 Group 5 0.10 0.20 0.05 0.125

[0221] In this embodiment, by setting up a dynamic torque sensor and introducing the first-order derivative, standard deviation and normalization processing indicators, this embodiment opens up a path to infer the stability of the melting state from the mechanical signal. Traditional methods mostly rely on temperature zone feedback, which has a time lag and is easily affected by thermal inertia, while this module can instantly capture the rising trend of local resistance caused by uneven melting or semi-molten particles in the stirring zone. This disturbance intensity indicator can serve as the first perception source of the "mechanical-melting anomaly signal" in the system, providing highly sensitive and dynamic predictive input for subsequent decision-making, significantly improving the system's abnormal response speed and feed rhythm pre-adjustment capabilities.

[0222] When generating feed control instructions, this embodiment integrates the heat supply melt heat deviation index Phea, the difference between the pellet melt rate LRm and the actual feed rhythm, and the melt resistance disturbance intensity Tres. Together, these three factors form the basis for calculating the feed correction factor. This strategy overcomes the limitations of traditional feeding strategies that rely on a single temperature or speed signal to control feeding. It can comprehensively assess feeding trends based on the system's current actual heating capacity, material consumption rate, and load risk. By integrating these heterogeneous factors into a single control model, the system achieves real-time unified judgment and decision-making capabilities for multi-dimensional energy and physical states.

[0223] By mapping the heating status to intervals, the system not only automatically adjusts the feed rate but also establishes a state tolerance band based on shear heat balance. In particular, when the feed correction factor reaches the critical overload zone, the system automatically adjusts the feed rate downward to avoid mechanical wear, increased energy consumption, and the risk of final product defects caused by semi-molten pellets. Furthermore, when excess heat is available, the feed rate is promptly increased to avoid energy waste and inefficient operation. This control method offers flexible range adjustment capabilities and fault protection mechanisms, enabling the system to maintain an optimal operating window under varying batch properties and load conditions.

[0224] Example 5

[0225] This embodiment is explained in Example 4. Please refer to Figure 3 Specifically: the raw material stability learning module aggregates and analyzes the feed correction factor ΔFin of multiple historical cycles with the corresponding shear heat generation rate Qsh and melting resistance disturbance intensity Tres, constructs a heat supply deviation response model for the current pellets, and derives the shear heat correction factor θco to generate a new shear heat generation rate Qtar;

[0226] The shear heat correction factor θco is obtained by the following formula:

[0227] ;

[0228] Where c1 represents the response adjustment coefficient of disturbance to feed deviation, c2 represents the response adjustment coefficient of disturbance to melting resistance, ΔFin(t) represents the feed correction factor ΔFin at time t, and Tres(t) represents the disturbance intensity of melting resistance at time t.

[0229] The shear heat generation rate Qsh is corrected by the obtained shear heat correction factor θco to generate a new shear heat generation rate Qtar;

[0230] The new shear heat generation rate Qtar is obtained by the following formula:

[0231] ;

[0232] Where Qtar(t) represents the new shear heat generation rate at time t, Qsh(t) represents the shear heat generation rate at time t, and c3 represents the thermal response sensitivity coefficient.

[0233] The risk index judgment and feedback module constructs the thermal power deviation index RpQ and torque anomaly index RTV through the new shear heat generation rate Qtar and torque value TV;

[0234] The thermal power deviation index RpQ is obtained by the following formula:

[0235] ;

[0236] The torque abnormality index RTV is obtained by the following formula:

[0237] ;

[0238] Where D1 represents the instantaneous torque change adjustment coefficient, D2 represents the fluctuation amplitude adjustment coefficient, d represents the integral sign, TV(t) represents the torque value at time t, and σTV represents the standard deviation of the torque signal;

[0239] The risk index Risk is calculated by obtaining the thermal power deviation index RpQ and the torque abnormality index RTV, and compared with the preset risk threshold Rth to determine the risk status;

[0240] The risk index Risk is obtained by the following formula:

[0241] ;

[0242] Where B1 and B2 represent the preset weight values ​​of thermal power deviation index RpQ and torque abnormality index RTV respectively;

[0243] The risk status is obtained by matching:

[0244] When the risk index Risk ≤ risk threshold Rth, it means that there is no risk and it remains within a safe range;

[0245] When the risk index Risk>risk threshold Rth, it means there is risk and anomaly occurs, and the screw speed Vs and melt temperature Tm are adjusted to obtain new screw speed nVs and new melt temperature nTm;

[0246] The new screw speed nVs is obtained by the following formula:

[0247] ;

[0248] Where Rsv represents the speed reduction factor;

[0249] The new melt temperature nTm is obtained by the following formula:

[0250] ;

[0251] Where ΔTM represents the heating amount in the temperature zone.

[0252] In this embodiment, a raw material stability learning module aggregates and models the feed correction factors, shear heat generation rates, and disturbance intensities across multiple historical cycles. This module establishes an evolutionary response mechanism for pellet thermal behavior. This mechanism extracts and derives shear heat correction factors to proactively correct current shear heat generation capacity, exhibiting triple functional characteristics: retrospective, responsive, and proactive. Unlike traditional systems that passively execute control commands based on pre-set models, this module derives control trends in real time based on historical operating trajectories, enabling it to learn and memorize fluctuations in raw material thermal conductivity characteristics. This significantly improves the system's adaptability to raw material batch variations and its parameter migration capabilities.

[0253] This embodiment innovatively applies a sliding learning mechanism to the gradual correction process of the shear heat generation rate, allowing the system to smooth thermal power output fluctuations in real time based on the operating performance over a time series, filtering out instantaneous deviations while retaining trend changes. Combined with the joint evaluation of the thermal power deviation index and the torque anomaly index in the risk module, the system can accurately determine whether the current processing state has deviated from the thermal supply and demand balance state, and identify potential critical areas of processing risk in advance. This effectively solves the problems of slow thermal compensation strategies, feedback lag, and imbalanced control amplitudes in traditional systems, and improves the system's sensitivity and control flexibility to complex heating disturbances.

[0254] The risk index judgment and feedback module not only accurately identifies abnormal conditions during processing but also establishes a dual-channel intervention mechanism based on thermal risk sources. This mechanism reduces the screw speed and reduces shear power input; it also increases heating capacity in the temperature zone to enhance local thermal conductivity, thereby mitigating overload risks caused by insufficient heat or unmelted material. Compared to traditional approaches that only adjust feed or temperature, this dual-variable redundant control significantly improves the accuracy, timeliness, and redundancy of abnormality handling, ensuring stable system operation under complex processing conditions.

[0255] Example 6

[0256] A method for intelligently controlling the production and processing parameters of polycarbonate materials, please refer to Figure 2 , specifically: including the following steps:

[0257] Step 1: The raw material attribute recognition and processing module collects the production parameters of the polycarbonate material through sensors, fits them into the original data set YM, and performs preprocessing to obtain the processing parameter set GW;

[0258] Step 2: The pellet shear and melting evaluation module constructs a shear heat generation model based on the obtained processing parameter set GW, calculates the shear heat generation rate Qsh per unit time of the pellet, and evaluates the pellet melting rate LRm;

[0259] Step 3: The stirring zone torque dynamic feedback module monitors the mechanical load state of the melting zone in real time, collects the torque signal fluctuations in the stirring zone, obtains the torque value TV, and constructs the melting resistance disturbance intensity Tres;

[0260] Step 4: The feed control instruction generation module combines the obtained shear heat generation rate Qsh and pellet melting rate LRm with the melting resistance disturbance intensity Tres to obtain the feed correction factor ΔFin;

[0261] Step 5: The raw material stability learning module corrects the shear heat generation rate Qsh by the feed correction factor ΔFin to obtain the new shear heat generation rate Qtar after dynamic learning;

[0262] Step 6: The risk index judgment and feedback module calculates the risk index Risk through the new shear heat generation rate Qtar and the torque value TV to judge the abnormality of the feed melting state.

[0263] In this example, the system comprehensively quantifies the thermophysical properties of polycarbonate pellets before processing by using several less common parameters, including pellet bulk density (Ldp), average pellet size (Ldm), pellet thermal conductivity (Drp), and pellet specific heat capacity (Dbc). This system then establishes a real-time calculation model for shear heat generation rate and melting rate. Compared to traditional methods that rely on "empirical knowledge" and single-point temperature estimation to estimate processing status, this method establishes a complete chain from raw material properties to processing behavior and then feedback strategy. This ensures the system's ability to accurately characterize, precisely predict, and quantify pellet status, significantly improving the scientific nature and response accuracy of the control process.

[0264] By monitoring the changing trends of the torque signal in the stirring zone in real time and constructing a disturbance intensity index, this method implements an early identification mechanism for hot melt states based on mechanical fluctuations. Compared to traditional strategies that rely solely on temperature fluctuations before responding, this module proactively detects mechanical load anomalies caused by issues like unmelted raw materials and particle agglomeration. This effectively prevents system overload and product performance fluctuations caused by the "half-melted, half-cold" state, significantly enhancing the system's interference immunity and operational robustness under variable operating conditions.

[0265] This method incorporates a raw material stability learning mechanism to model the dynamic coupling relationship between shear heat output, feed behavior, and torque fluctuations over historical cycles, forming an adaptive feedback mechanism that can be used to modify the current control strategy. Furthermore, a combined risk index of thermal power deviation and mechanical fluctuation is introduced into the risk assessment process. When abnormal trends are identified, screw speed and heating power are adjusted in a coordinated manner. This enables proactive risk-oriented response and multi-channel coordinated intervention, significantly enhancing the overall system security and strategic flexibility.

[0266] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent control system for polycarbonate material production and processing parameters, characterized by: It includes raw material attribute identification and processing module, pellet shear and melting evaluation module, stirring zone torque dynamic feedback module, feed control instruction generation module, raw material stability learning module and risk index judgment and feedback module; The raw material attribute recognition and processing module collects the production parameters of polycarbonate materials through sensors, fits them into the original data set YM, and performs preprocessing to obtain the processing parameter set GW; The pellet shear and melting evaluation module constructs a shear heat generation model based on the obtained processing parameter set GW, calculates the shear heat generation rate Qsh of the pellet per unit time, and evaluates the pellet melting rate LRm; The stirring zone torque dynamic feedback module monitors the mechanical load state of the melting zone in real time, collects the torque signal fluctuation in the stirring zone, obtains the torque value TV, and constructs the melting resistance disturbance intensity Tres; The feed control instruction generation module obtains the feed correction factor ΔFin based on the obtained shear heat generation rate Qsh and pellet melting rate LRm, combined with the melting resistance disturbance intensity Tres; The raw material stability learning module corrects the shear heat generation rate Qsh by the feed correction factor ΔFin to obtain the new shear heat generation rate Qtar after dynamic learning; The risk index judgment and feedback module calculates the risk index Risk through the new shear heat generation rate Qtar and the torque value TV to judge the abnormality of the feed melting state.

2. The intelligent control system for polycarbonate material production and processing parameters according to claim 1, characterized in that: The raw material attribute recognition and processing module includes a multi-parameter acquisition and calibration unit and a feature pre-processing unit; The multi-parameter acquisition and calibration unit collects the production parameters of polycarbonate materials through sensors and detection devices, including pellet bulk density Ldp, pellet average particle size Ldm, pellet thermal conductivity Drp, pellet specific heat capacity Dbc, pellet initial surface temperature To, stacking angle θ, screw speed Vs and melt temperature Tm, and fits them into the original data set YM; Among them, the bulk density Ldp of the pellets is calculated and obtained by the hopper bottom loading weighing system and the laser volume measurement device; The average particle size Ldm of the granular material is collected and obtained using an image recognition particle size analyzer; The thermal conductivity Drp and specific heat capacity Dbc of the pellets are derived and obtained through the online thermal response detection device; The initial surface temperature To of the pellets and the melt temperature Tm were acquired by infrared thermal imaging array and non-contact infrared point thermometer; The stacking angle θ is obtained by laser profile measurement and image fitting; The screw speed Vs is directly collected through the encoder of the host system; The feature preprocessing unit cleans and normalizes the acquired original data set YM to obtain the processing parameter set GW; Cleaning includes outlier removal, which is done by using the triple standard deviation method to remove outliers from the original dataset YM and filling them with linear interpolation; Normalization processing is performed by using the Max-Min normalization method to process the original data set YM to obtain the processing parameter set GW; The processing parameter set GW is obtained by the following formula: ； Where GWo represents the oth data item in the processing parameter set GW, YMo represents the oth data item in the original data set YM, minYMo represents the valley value of the oth data item in the original data set YM, and maxYMo represents the peak value of the oth data item in the original data set YM.

3. The intelligent control system for polycarbonate material production and processing parameters according to claim 2, characterized in that: The pellet shear and melting evaluation module includes a shear heat generation calculation unit and a melting rate evaluation and conversion unit; The shear heat generation calculation unit constructs a shear heat generation model through the processing parameter set GW and calculates the apparent viscosity Napp; The apparent viscosity Napp is obtained by the following formula: ； Where No represents zero shear viscosity, λ0 represents time relaxation constant, Zn represents shear index, Sve represents shear rate, and Sve=k1*Vs, k1 represents the proportional coefficient related to screw structure and extrusion cavity, exp represents exponential function, α T represents the thermal sensitivity factor, Tre represents the reference temperature; According to the obtained apparent viscosity Napp, the shear heat generation rate Qsh is calculated; The shear heat generation rate Qsh is obtained by the following formula: ； Where α1 represents the empirical heat transfer factor, Veff represents the equivalent shear volume; The equivalent shear volume is obtained by the following formula: ； Where LC represents the flow channel structure constant, and tan() represents the sine function.

4. The intelligent control system for polycarbonate material production and processing parameters according to claim 3, characterized in that: The melting rate evaluation and conversion unit evaluates the pellet melting rate LRm based on the obtained shear heat generation rate Qsh and the pellet specific heat capacity Dbc; The unit mass heat melting requirement Hreq is obtained by the specific heat capacity Dbc of the pellets; The unit mass hot melt requirement Hreq is obtained by the following formula: Hreq=Dbc*(Tm-To); The pellet melting rate LRm is obtained by the following formula: ； Where, sp represents the melting heat conversion efficiency; The melting heat conversion efficiency sp is obtained by the following formula: ； Where Rh represents the measured thickness of the granular stacking layer, Φ1 represents the unit heat capacity density, and Φ2 represents the unit thermal conductivity path coefficient.

5. The intelligent control system for polycarbonate material production and processing parameters according to claim 3, characterized in that: The stirring zone torque dynamic feedback module includes a torque signal acquisition and decoding unit and a disturbance intensity index calculation unit; The torque signal acquisition and decoding unit collects the torque signal data of the main shaft through the dynamic torque sensor set in the main shaft of the stirring zone, and collects the torque value TV of the torque sampling point Td in each unit time band; The disturbance intensity index calculation unit constructs the first-order derivative, variation rate and normalized index based on the torque value TV to extract the melting resistance disturbance intensity Tres; The torque change rate ΔTV is obtained by constructing the first-order derivative; The torque change rate ΔTV is obtained by the following formula: ； Where TV(t) represents the torque value at time t, TV(t-Δt) represents the torque value at time t-Δt, and Δt represents the time interval; The melting resistance disturbance intensity Tres is obtained by the following formula: ； In the formula, β1 represents the rate change adjustment factor, β2 represents the fluctuation amplitude adjustment factor, represents the sliding average of the torque, and σTV represents the standard deviation of the torque signal.

6. The intelligent control system for polycarbonate material production and processing parameters according to claim 5, characterized in that: The feed control instruction generation module includes a state deviation calculation unit and a multi-factor fusion calculation unit; The state deviation calculation unit compares the obtained shear heat generation rate Qsh with the theoretical generation rate Qget, calculates the melting heat deviation index Phea, and determines the heating state of the current working condition; The melting heat deviation index Phea is obtained by the following formula: ； The heating status of the working condition is obtained by matching in the following ways: When the melting heat deviation index Phea ≥ 0, it means insufficient heat supply and the feeding rhythm is slowed down; When the melting heat deviation index Phea is less than 0, it means that there is excess heat supply and the feeding rhythm should be adjusted upward; Compare the obtained pellet melting rate LRm with the actual feed rate Lar to obtain the melting limit index Lme; The melting limit index Lme is obtained by the following formula: 。 7. The intelligent control system for polycarbonate material production and processing parameters according to claim 6, characterized in that: The multi-factor fusion calculation unit combines the obtained melting heat deviation index Phea and melting limit index Lme with the melting resistance disturbance intensity Tres to calculate the feed correction factor ΔFin; The feed correction factor ΔFin is obtained by the following formula: ； Where, Respectively represent the preset weight values ​​of melting heat deviation index Phea, melting limit index Lme and melting resistance disturbance intensity Tres, and ; Adjust the output strategy through the feed correction factor ΔFin; When 0<feed correction factor ΔFin<0.3, it means that the current heat supply is sufficient and production is stable, so increase the feed rate and improve production capacity; When 0.3≤feed correction factor ΔFin<0.7, it means that the heating is normal and the current feeding rhythm is maintained; When 0.7≤feed correction factor ΔFin<1.0, it indicates that the heating is abnormal and the load limit is approaching, so the feed rate should be reduced.

8. The intelligent control system for polycarbonate material production and processing parameters according to claim 7, characterized in that: The raw material stability learning module aggregates and analyzes the feed correction factor ΔFin of multiple historical cycles with the corresponding shear heat generation rate Qsh and melt resistance disturbance intensity Tres, constructs a heat supply deviation response model for the current pellets, and derives the shear heat correction factor θco to generate a new shear heat generation rate Qtar; The shear heat correction factor θco is obtained by the following formula: ； Where c1 represents the response adjustment coefficient of disturbance to feed deviation, c2 represents the response adjustment coefficient of disturbance to melting resistance, ΔFin(t) represents the feed correction factor ΔFin at time t, and Tres(t) represents the disturbance intensity of melting resistance at time t. The shear heat generation rate Qsh is corrected by the obtained shear heat correction factor θco to generate a new shear heat generation rate Qtar; The new shear heat generation rate Qtar is obtained by the following formula: ； Where Qtar(t) represents the new shear heat generation rate at time t, Qsh(t) represents the shear heat generation rate at time t, and c3 represents the thermal response sensitivity coefficient.

9. The intelligent control system for polycarbonate material production and processing parameters according to claim 8, characterized in that: The risk index judgment and feedback module constructs the thermal power deviation index RpQ and torque anomaly index RTV through the new shear heat generation rate Qtar and torque value TV; The thermal power deviation index RpQ is obtained by the following formula: ； The torque abnormality index RTV is obtained by the following formula: ； Where D1 represents the instantaneous torque change adjustment coefficient, D2 represents the fluctuation amplitude adjustment coefficient, d represents the integral sign, TV(t) represents the torque value at time t, and σTV represents the standard deviation of the torque signal; The risk index Risk is calculated by obtaining the thermal power deviation index RpQ and the torque abnormality index RTV, and compared with the preset risk threshold Rth to determine the risk status; The risk index Risk is obtained by the following formula: ； Where B1 and B2 represent the preset weight values ​​of thermal power deviation index RpQ and torque abnormality index RTV respectively; The risk status is obtained by matching: When the risk index Risk ≤ risk threshold Rth, it means that there is no risk and it remains within a safe range; When the risk index Risk>risk threshold Rth, it means there is risk and anomaly occurs, and the screw speed Vs and melt temperature Tm are adjusted to obtain new screw speed nVs and new melt temperature nTm; The new screw speed nVs is obtained by the following formula: ； Where Rsv represents the speed reduction factor; The new melt temperature nTm is obtained by the following formula: ； Where ΔTM represents the heating amount in the temperature zone.

10. A method for intelligently controlling the production and processing parameters of a polycarbonate material, applied to the intelligent control system for the production and processing parameters of a polycarbonate material according to any one of claims 1 to 9, characterized in that: The following steps are involved: Step 1: The raw material attribute recognition and processing module collects the production parameters of the polycarbonate material through sensors, fits them into the original data set YM, and performs preprocessing to obtain the processing parameter set GW; Step 2: The pellet shear and melting evaluation module constructs a shear heat generation model based on the obtained processing parameter set GW, calculates the shear heat generation rate Qsh per unit time of the pellet, and evaluates the pellet melting rate LRm; Step 3: The stirring zone torque dynamic feedback module monitors the mechanical load state of the melting zone in real time, collects the torque signal fluctuations in the stirring zone, obtains the torque value TV, and constructs the melting resistance disturbance intensity Tres; Step 4: The feed control instruction generation module combines the obtained shear heat generation rate Qsh and pellet melting rate LRm with the melting resistance disturbance intensity Tres to obtain the feed correction factor ΔFin; Step 5: The raw material stability learning module corrects the shear heat generation rate Qsh by the feed correction factor ΔFin to obtain the new shear heat generation rate Qtar after dynamic learning; Step 6: The risk index judgment and feedback module calculates the risk index Risk through the new shear heat generation rate Qtar and the torque value TV to judge the abnormality of the feed melting state.

Images