Continuous production method of high-modulus low-smoke halogen-free polycarbonate material

Abstract

The invention discloses a continuous production method of a high-modulus low-smoke halogen-free polycarbonate material, and relates to the technical field of polycarbonate material production, and the continuous production method comprises the following steps: under the condition of determining that interfacial stress release of a flame retardant and resin is delayed, adding a flame retardant, extracting three-dimensional path data, flow velocity tension characteristics, distribution density gradient and shear rotation direction changes of the flame retardant particles in the continuous flow area to determine flow field aggregation characteristics under the condition of flame retardant and resin interface stress release delay, and outputting standardized aggregation characteristic indexes; according to the flow field aggregation characteristics under the condition of flame retardant and resin interface stress release delay, the aggregation rate change amplitude, the particle separation acceleration and the coating reconstruction trend change are extracted to be used for judging whether the flame retardant is subjected to layered desorption or not, and a desorption risk score is generated. According to the method, the problem that desorption of the flame retardant is difficult to judge is solved, and accurate recognition and dynamic regulation and control of the interface fusion state are achieved.

Description

Technical Field

[0001] This invention relates to the field of polycarbonate material production technology, specifically to a continuous production method for high-modulus, low-smoke, halogen-free polycarbonate materials. Background Technology

[0002] Continuous production of high-modulus, low-smoke, halogen-free polycarbonate materials refers to the continuous manufacturing of polycarbonate materials with high mechanical strength (high modulus), low smoke toxicity, and halogen-free flame retardant properties through multiple stages in an uninterrupted industrial process, including continuous raw material transportation, mixing, reaction, extrusion, cooling, and granulation. Current continuous production technologies typically employ twin-screw extrusion as the core process. First, the polycarbonate matrix resin is precisely proportioned with specific high-modulus reinforcing agents (such as glass fiber, carbon nanotubes, and aramid fiber) and halogen-free flame retardant systems (such as phosphorus-nitrogen flame retardants and silicone-based flame retardants). This mixture is then fed into a high-shear mixing zone via a continuous feeding system. During this process, the materials are melt-mixed under controlled temperature and pressure, ensuring thorough dispersion and fusion of the reinforcing phase and matrix, while simultaneously achieving uniform distribution of the flame retardant, thereby constructing a material system with excellent comprehensive performance. Compared to conventional polycarbonate materials, whose flexural modulus is typically around 2400 MPa, the modified polycarbonate material produced by this technology can significantly increase its flexural modulus to over 3000 MPa, exhibiting superior structural load-bearing capacity. Simultaneously, regarding flame retardancy, the oxygen index of traditional PC is approximately 26, while this technology can achieve an oxygen index exceeding 35, significantly reducing the release of smoke and toxic gases during combustion, meeting high safety and environmental regulations. Next, after extrusion through a die, the material enters a cooling system for rapid solidification via water or air cooling. Finally, it is guided by a traction system to a pelletizer for continuous granulation, resulting in stable, highly consistent polycarbonate granules. The entire production process relies on an automated control system to monitor and adjust key parameters such as temperature, pressure, torque, and feed rate in real time, ensuring that the product maintains excellent physical and mechanical properties, low smoke emissions, and halogen-free environmental characteristics even in mass production.

[0003] The existing technology has the following shortcomings:

[0004] In the continuous production of high-modulus, low-smoke, halogen-free polycarbonate materials, when the extrusion speed changes abruptly due to adjustments in the feeding system rhythm or batch switching of raw materials, the melt flow rate, temperature, and stress field under high temperature and high shear conditions will simultaneously undergo drastic changes. During this process, the physical coating structure established between the halogen-free flame retardant particles, which were originally stable within the polycarbonate matrix, and the resin will become unstable due to delayed local stress release. This causes the flame retardant particles to partially detach from the matrix and float in the flow field, forming a delamination phenomenon. Because this phenomenon occurs in the instantaneous rheological transition zone and is not accompanied by significant melt pressure fluctuations, torque changes, or energy consumption changes, existing continuous production technologies for high-modulus, low-smoke, halogen-free polycarbonate materials cannot accurately determine whether the flame retardant has undergone delamination based on the flow field aggregation characteristics under delayed stress release at the flame retardant-resin interface. The existence of this problem will lead to an imbalance in the distribution of flame retardants in the pellets, resulting in a "shell-core hollow" structure. This not only destroys the flame retardant uniformity inside the material and affects the oxygen index and UL-94 flammability rating, but may also cause the release of flame retardant components during injection molding, resulting in surface contamination and poor adhesion of the product, which seriously reduces the consistency of product performance and end-product reliability.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a continuous production method for high-modulus, low-smoke, halogen-free polycarbonate materials to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a continuous production method for high-modulus, low-smoke, halogen-free polycarbonate materials, specifically comprising the following steps:

[0008] S1. Obtain flow behavior data of flame retardant particles and polycarbonate resin in the shear section, including tensile response trajectory, viscoelastic properties and thermal response changes before and after shear rate change, and construct interfacial stress release time series curve to determine whether there is a delay in the release of interfacial stress between flame retardant and resin.

[0009] S2. When it is determined that there is a delay in the release of stress at the interface between the flame retardant and the resin, extract the three-dimensional path data, flow velocity tension characteristics, distribution density gradient and shear rotation direction changes of the flame retardant particles in the continuous flow area to determine the flow field aggregation characteristics under the condition of delayed release of stress at the interface between the flame retardant and the resin, and output the standardized aggregation characteristic index.

[0010] S3. Based on the flow field aggregation characteristics under the delayed stress release at the flame retardant-resin interface, extract the changes in aggregation rate, particle detachment acceleration, and coating reconstruction trend to determine whether the flame retardant has undergone delamination and generate a desorption risk score.

[0011] S4. Generate a desorption evaluation vector based on the desorption risk score and historical formation assessment curve, and divide the fusion stable state, desorption critical state and explicit stratification state accordingly.

[0012] S5. Adjust the material feeding rhythm, shear rate change slope and melt path distribution based on the desorption evaluation vector to dynamically control the interfacial fusion state between flame retardant particles and polycarbonate resin.

[0013] Preferably, S1 is as follows:

[0014] A multi-parameter synchronous acquisition device is set up in the shear section. The tensile response trajectory of flame retardant particles and polycarbonate resin is collected by the particle displacement tracking unit, the viscoelastic properties before and after the shear rate change are recorded by the dynamic viscoelastic response monitoring unit, and the thermal response change is recorded by the micro-area heat flux sensing unit.

[0015] The tensile response trajectory is specifically the continuous displacement-time curve of the particle along the shear direction; the viscoelastic properties before and after the shear rate change are specifically the change in the ratio of storage modulus to loss modulus in the two intervals before and after shear fluctuation; and the thermal response change is specifically the curve of heat flux density per unit area of ​​particle interface changing with time.

[0016] The tensile response trajectory, viscoelastic properties before and after shear rate change, and thermal response change are aligned on the time axis and fused to construct an interfacial stress release time series curve. Based on the time difference between the delayed inflection point of the tensile response trajectory in the constructed interfacial stress release time series curve and the starting point of viscoelastic property recovery, it is determined whether the time difference exceeds a preset hysteresis threshold. If it exceeds the threshold, it is determined that there is a situation of delayed stress release at the interface between the flame retardant and the resin. If it does not exceed the threshold, it is determined that there is no situation of delayed stress release at the interface between the flame retardant and the resin.

[0017] Preferably, S2 specifically includes the following steps:

[0018] S201. When it is determined that there is a delay in stress release at the interface between the flame retardant and the resin, the three-dimensional path data of the flame retardant particles in the continuous flow area is extracted based on the particle three-dimensional tracking technology. The flow velocity tension characteristics are calculated by using the path velocity distribution change rate. The distribution density gradient is constructed by the difference in the unit space ratio of the particles. The change in shear rotation direction is extracted by tracking the rotation trajectory of the particles around the main shear axis.

[0019] S202. The three-dimensional path data, flow velocity tension characteristics, distribution density gradient and shear rotation direction change are fused in a unified spatial grid to establish a flow field response coupling map. Based on the flow field response coupling map, the density overlap coefficient and directional concentration of the aggregation region are identified, which are used as the flow field aggregation characteristics under the stress release delay of the flame retardant and resin interface.

[0020] S203. Based on the determined flow field aggregation characteristics, the aggregation intensity, aggregation trend directionality, and regional coupling density are vectorized and encoded using a standard feature mapping function. A normalization algorithm is then used to output standardized aggregation characteristic indicators to characterize the severity of aggregation behavior.

[0021] Preferably, S201 specifically refers to:

[0022] When it is determined that there is a delay in stress release at the interface between the flame retardant and the resin, the particle three-dimensional tracking technology is used to acquire multi-angle synchronous images of the flame retardant particles. The two-dimensional plane projection is converted into trajectory data in the three-dimensional coordinate system through a stereo reconstruction algorithm, thereby obtaining the three-dimensional path data of the flame retardant particles in the continuous flow area, and marking each trajectory point in the form of a time series.

[0023] Based on the three-dimensional path data, the rate of change of the velocity vector of flame retardant particles within a continuous time period is calculated, and a tension distribution curve reflecting the dynamic changes of tensile tension and shear tension is constructed, thereby obtaining the velocity tension characteristics corresponding to the rate of change of the path velocity distribution.

[0024] The continuous flow region is divided into spatial grid cells, and the particle proportion per unit volume in each grid is statistically analyzed. A distribution density gradient is constructed based on the difference in particle proportion between adjacent grids. At the same time, the rotation trajectory of flame retardant particles around the main shear axis is tracked, and the continuous offset of the rotation vector direction over time is analyzed to extract the change in shear rotation direction.

[0025] Preferably, S202 specifically refers to:

[0026] A three-dimensional spatial grid of equal volume is constructed in the continuous flow region. The three-dimensional path data of flame retardant particles, the flow velocity and tension characteristics of corresponding path points, the distribution density gradient values ​​in the local grid, and the shear rotation direction change data are uniformly merged according to the grid index to generate a multi-parameter fused data matrix.

[0027] Based on the fused data matrix, the average cosine of the angle between the number of intersections of flame retardant particle paths and the direction vector in each spatial grid is calculated, which are represented as the density overlap coefficient and the direction concentration, respectively. A density-direction joint distribution function is constructed to evaluate the local aggregation intensity of particles in the flow field.

[0028] Regions with a density overlap coefficient higher than twice the standard deviation of the overall average density overlap coefficient of the particles and a directional concentration degree exceeding a preset directional concentration threshold are marked as high-aggregation areas. The joint characteristic parameters of these high-aggregation areas are then output as the flow field aggregation characteristics under the stress release delay condition at the flame retardant and resin interface.

[0029] Preferably, S3 specifically includes the following steps:

[0030] S301. Based on the flow field aggregation characteristics under the delayed stress release at the flame retardant and resin interface, extract the variation curve of the number of flame retardant particles in the aggregation region on the continuous time axis, calculate the derivative of the particle density change per unit time as the aggregation rate change amplitude; simultaneously identify the particle path that deviates from the aggregation center trajectory, and determine the particle deviating acceleration based on the path acceleration change.

[0031] S302. Select boundary samples of particle-coated structures within the aggregation region, and conduct joint analysis on their rotation direction vector, stress centripetal trend and interface contact area changes to construct a coating reconstruction trend change function, which is used to indicate whether the particles have a reconstructed and regressive motion tendency after instability.

[0032] S303. Based on the feature vectors of three dimensions—the magnitude of the aggregation rate change, the particle detachment acceleration, and the trend of coating reconstruction—the desorption scoring model is input, and feature weighted fusion and quantitative scoring are performed to generate a desorption risk score in the form of continuous output. When the desorption risk score exceeds the preset threshold, it is judged that the flame retardant has undergone stratification desorption; if it does not exceed the threshold, it is judged that stratification desorption has not occurred.

[0033] Preferably, S302 is as follows:

[0034] Flame retardant particles that are in contact with the matrix interface are extracted from the edge of the aggregation area. A rotation direction vector sequence is constructed to represent the rotation direction and angular displacement evolution trajectory of the particles around their center of mass, and the rotation speed change curve on the time axis is recorded simultaneously.

[0035] Based on the rotation direction vector sequence, the stress vector projection distribution of particles toward the aggregation center is calculated, the stress centripetal trend consistent with the particle rotation trend is extracted, and a centripetal coupling index is established; at the same time, the small changes in the contact area between the particle and resin interface over time are tracked to obtain the interface contact area change curve.

[0036] The rotation direction vector sequence, stress centripetal trend distribution and interface contact area change curve are coupled and fitted with multiple parameters to construct a coating reconstruction trend change function. Based on the consistency between the trend rising segment in the function and the path of particles returning to the aggregation center, it is determined whether there is a reconstruction regression motion trend.

[0037] Preferably, S4 is as follows:

[0038] A mapping function is constructed based on the desorption risk score and the historical molding assessment curve. The desorption risk score of the current flame retardant particles is input into the fitting model corresponding to the historical molding assessment curve. The score position and response stratification probability of the desorption risk score in the historical assessment data are extracted, and a two-dimensional coordinate point set containing the risk score and stratification probability is constructed.

[0039] The desorption risk score and the stratified probability corresponding to the historical molding evaluation curve are constructed into a two-dimensional feature vector. The score slope change rate and historical stability window offset are introduced as supplementary dimensions to generate a desorption evaluation vector containing score intensity, evolution trend and historical offset information, which is used to describe the multidimensional numerical state of the stability of the flame retardant and resin interface.

[0040] The state is divided based on the desorption evaluation vector. A first scoring threshold and a second scoring threshold are set as classification boundaries. If the desorption risk score is lower than the first scoring threshold and the stratification probability is lower than the first probability threshold, it is classified as a fusion stable state. If the desorption risk score is between the first scoring threshold and the second scoring threshold, or the stratification probability is close to the mean range, it is classified as a desorption critical state. If the desorption risk score is higher than the second scoring threshold and the stratification probability exceeds the second probability threshold, it is classified as an explicit stratification state.

[0041] Preferably, S5 is as follows:

[0042] Based on the desorption risk score, stratification probability, score slope change rate and historical stable window offset contained in the desorption evaluation vector, a desorption state control mapping relationship is constructed. The current desorption evaluation vector is input into the mapping model, and the target control parameter set is output, which includes the set values ​​of material feeding rhythm, shear rate change slope and melt path distribution mode that match the current state.

[0043] According to the feeding rhythm of the target material, the synchronous conveying unit of flame retardant particles and polycarbonate resin is rhythmically coordinated, including adjusting the particle feeding speed, the polycarbonate melt feeding rate and the phase matching control of the two, so that the components entering the mixing and shearing section maintain a stable ratio and suppress the fusion imbalance caused by sudden changes in feeding in local areas.

[0044] Based on the set values ​​of the target shear rate change slope and melt path distribution, the speed control module of the screw shear section and the flow channel control system are synchronously corrected. This includes adjusting the slope curve of the screw acceleration section to match the flame retardant response window and switching the internal guide configuration of the melt channel to distribute the particle streamline density. This enables dynamic control of the interface fusion state between the flame retardant particles and polycarbonate resin, and ensures the continuity and uniformity of the fusion process.

[0045] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0046] 1. This invention constructs a multi-dimensional monitoring and control mechanism based on interfacial stress release behavior as the core criterion. This mechanism can accurately identify delamination and desorption phenomena caused by interfacial fusion instability between flame retardant particles and polycarbonate resin during the continuous extrusion production of high-modulus, low-smoke, halogen-free polycarbonate materials. Compared to traditional methods relying on macroscopic parameters (such as pressure fluctuations and torque changes), this approach introduces microscopic characteristic indicators such as three-dimensional particle tracking, path tension distribution, distribution density gradient, and shear rotation direction. It establishes a fusion delay identification mechanism and a flow field aggregation feature extraction model, effectively capturing local anomalies at the particle scale. Furthermore, through multi-parameter fusion analysis, it achieves early prediction of desorption trends, thereby significantly improving the monitoring sensitivity and identification accuracy of material interfacial stability under complex transient rheological behavior.

[0047] 2. This invention constructs a closed-loop dynamic control path based on desorption evaluation vectors. This path adaptively adjusts the material feeding rhythm, shear rate change slope, and melt path distribution according to the real-time fusion state, achieving precise control over the fusion state of the flame retardant particles and polycarbonate resin interface. This control mechanism not only solves the problem of local interface instability caused by sudden changes in the feeding rhythm or raw material batch switching, significantly reducing the probability of the "shell-polymerized core-hollow" structure, but also ensures the uniformity of component distribution and flame retardant performance of the pellets, improving the final material's combustion rating, structural stability, and molding quality. It has good engineering application value and promising prospects for industrial promotion. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0049] Figure 1 This is a schematic flowchart of a continuous production method for a high-modulus, low-smoke, halogen-free polycarbonate material according to the present invention. Detailed Implementation

[0050] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0051] This invention provides, for example Figure 1The continuous production method of a high-modulus, low-smoke, halogen-free polycarbonate material shown includes the following steps:

[0052] S1. Obtain flow behavior data of flame retardant particles and polycarbonate resin in the shear section, including tensile response trajectory, viscoelastic properties and thermal response changes before and after shear rate change, and construct interfacial stress release time series curve to determine whether there is a delay in the release of interfacial stress between flame retardant and resin.

[0053] In this embodiment, S1 specifically refers to:

[0054] A multi-parameter synchronous acquisition device is set up in the shear section. The tensile response trajectory of flame retardant particles and polycarbonate resin is collected by the particle displacement tracking unit, the viscoelastic properties before and after the shear rate change are recorded by the dynamic viscoelastic response monitoring unit, and the thermal response change is recorded by the micro-area heat flux sensing unit.

[0055] In the continuous production process of high-modulus, low-smoke, halogen-free polycarbonate materials, to achieve real-time and accurate capture of the flow behavior of flame retardant particles and polycarbonate resin within the shear section, a multi-parameter synchronous acquisition device can be installed in the shear section to achieve high-frequency, zero-delay acquisition of physical response data throughout the entire process. This multi-parameter synchronous acquisition device is constructed as an integrated array structure, capable of synchronously reading and time-locking data from different physical dimensions within the same spatial segment, ensuring that the acquired data possesses high temporal consistency and physical comparability. In this device, the particle displacement tracking unit, based on high-frame-rate microscopic imaging and fluorescent marker tracking technology, can capture the continuous displacement curves of a single flame retardant particle along the shear direction and perpendicular direction in a high-shear flow field, used to generate tensile response trajectories. The dynamic viscoelastic response monitoring unit uses micro-inductive impedance spectroscopy analysis technology to dynamically analyze the changes in the storage modulus and loss modulus of the polycarbonate melt before and after a sudden change in shear rate, thereby characterizing the changing trends of its shear recovery and elongation capabilities. The micro-area heat flux sensing unit integrates an embedded micro-thermal lattice array, possessing high spatial resolution heat flux response detection capabilities, and can acquire the dynamic process of heat flux density per unit area on the melt interface changing over time, used to analyze whether local energy transfer is abnormal. Through the coordinated operation of these units, not only can key physical behaviors at the particle and interface levels be comprehensively collected, but also a precise data foundation is provided for subsequent construction of interface stress release time-series curves.

[0056] The tensile response trajectory is specifically the continuous displacement-time curve of the particle along the shear direction; the viscoelastic properties before and after the shear rate change are specifically the change in the ratio of storage modulus to loss modulus in the two intervals before and after shear fluctuation; and the thermal response change is specifically the curve of heat flux density per unit area of ​​particle interface changing with time.

[0057] In the continuous production process of high-modulus, low-smoke, halogen-free polycarbonate materials, in order to achieve multi-dimensional quantitative analysis of the physical behavior of flame retardant particles and polycarbonate resin under shear disturbance, it is necessary to extract three types of parameters: tensile response trajectory, viscoelastic property change, and thermal response change, as the basis for subsequent construction of interface stress release time series curves. Among them, the continuous displacement-time curve of particles along the shear direction refers to the instantaneous position change of a single flame retardant particle along the main shear direction in the shear flow field, recorded by high frame rate image tracking or laser particle imaging technology, and its continuous trajectory is reconstructed according to the time sequence, reflecting whether there is abrupt displacement or velocity lag of the particles; the change amplitude of the ratio of storage modulus to loss modulus in the two intervals before and after shear fluctuation is obtained by acquiring stress-strain response data before and after shear through a micro inductive impedance spectroscopy system, calculating the ratio of storage modulus (elastic energy storage capacity) and loss modulus (energy dissipation capacity) in each time period and then taking the difference, reflecting the structural recovery and dissipation characteristics of the material under shear abrupt change; the curve of heat flux density per unit area at the particle interface with time refers to the real-time acquisition of the heat flux value per unit area at the interface between the flame retardant particles and polycarbonate resin using a high-sensitivity micro-area heat flux array, and outputting the dynamic change trend of heat flux density according to the time axis, thereby monitoring whether there is abnormal thermal imbalance or hysteresis. The dynamic parameter system formed by the interaction of these three elements can accurately characterize the motion, structure, and heat conduction behavior of particles during rheological impact, providing comprehensive and quantifiable basic data for subsequent judgment on whether interfacial stress release hysteresis occurs.

[0058] The tensile response trajectory, viscoelastic properties before and after shear rate change, and thermal response change are aligned on the time axis and fused to construct an interfacial stress release time series curve. Based on the time difference between the delayed inflection point of the tensile response trajectory in the constructed interfacial stress release time series curve and the starting point of viscoelastic property recovery, it is determined whether the time difference exceeds a preset hysteresis threshold. If it exceeds the threshold, it is determined that there is a situation of delayed stress release at the interface between the flame retardant and the resin. If it does not exceed the threshold, it is determined that there is no situation of delayed stress release at the interface between the flame retardant and the resin.

[0059] To quantitatively identify the interfacial stress release behavior between flame retardant particles and polycarbonate resin, it is first necessary to map data from different sources—tensile response trajectory, viscoelastic properties before and after shear rate changes, and thermal response changes—to a unified time axis system. A time-resolved synchronization algorithm is then used for time reference alignment to eliminate sampling time differences and synchronization offsets between physical responses. After alignment, a three-dimensional fusion matrix is ​​constructed to encode the change characteristics of the three types of data in a time-series combination, forming an interfacial stress release time-series curve. This time-series curve not only reflects the motion behavior of flame retardant particles in the flow field but also integrates the temporal evolution of the melt structure recovery state and heat exchange state. By performing gradient fitting analysis on the inflection points of the tensile response curve in the time-series curve, the key inflection points where particle motion lags can be identified. Furthermore, the starting points of the recovery and stabilization of loss modulus and storage modulus in the viscoelastic property curve are extracted as time markers for interfacial structure recovery. Calculating the time difference between these two points reflects the degree of temporal coupling between the interfacial physical behavior and the melt structure response, which can then be used to determine whether there is a hysteresis in the interfacial coupling.

[0060] The delayed inflection point of the tensile response trajectory refers to the critical time point at which the motion trajectory of flame retardant particles shows a significant slowdown or stagnation after a shear mutation. This time point represents the instantaneous response of the particle's degrees of freedom after interfacial instability. The viscoelastic property recovery starting point refers to the first time point at which the ratio of storage modulus to loss modulus begins to stabilize again in the melt system after shear disturbance. This point reflects the initial instant when the polycarbonate matrix returns from a heterogeneous structure to a relatively homogeneous state. The time difference between the two reflects the degree of coordination between the behavior of flame retardant particles and the viscoelastic recovery of the matrix. If the particle hysteresis time exceeds the starting point of this structural recovery, it indicates that it has failed to re-establish a stable coating within the matrix recovery period, posing a risk of interfacial coupling fracture. The preset hysteresis threshold is a judgment standard set based on historical experience data, material combination characteristics, and the stability window boundary under production conditions. It is used to delineate whether the delay time exceeds the critical line of the normal fusion hysteresis range. This judgment method can achieve quantitative identification of interfacial stress release delay and avoid misjudgment caused by relying on macroscopic variables such as energy consumption and pressure.

[0061] S2. When it is determined that there is a delay in the release of stress at the interface between the flame retardant and the resin, extract the three-dimensional path data, flow velocity tension characteristics, distribution density gradient and shear rotation direction changes of the flame retardant particles in the continuous flow area to determine the flow field aggregation characteristics under the condition of delayed release of stress at the interface between the flame retardant and the resin, and output the standardized aggregation characteristic index.

[0062] In this embodiment, S2 specifically includes the following steps:

[0063] S201. When it is determined that there is a delay in stress release at the interface between the flame retardant and the resin, the three-dimensional path data of the flame retardant particles in the continuous flow area is extracted based on the particle three-dimensional tracking technology. The flow velocity tension characteristics are calculated by using the path velocity distribution change rate. The distribution density gradient is constructed by the difference in the unit space ratio of the particles. The change in shear rotation direction is extracted by tracking the rotation trajectory of the particles around the main shear axis.

[0064] S202. The three-dimensional path data, flow velocity tension characteristics, distribution density gradient and shear rotation direction change are fused in a unified spatial grid to establish a flow field response coupling map. Based on the flow field response coupling map, the density overlap coefficient and directional concentration of the aggregation region are identified, which are used as the flow field aggregation characteristics under the stress release delay of the flame retardant and resin interface.

[0065] S203. Based on the determined flow field aggregation characteristics, the aggregation intensity, aggregation trend directionality, and regional coupling density are vectorized and encoded using a standard feature mapping function. A normalization algorithm is then used to output standardized aggregation characteristic indicators to characterize the severity of aggregation behavior.

[0066] Based on the construction of flow field aggregation characteristics, it is necessary to uniformly encode various aggregation parameters through standard feature mapping functions to achieve a quantitative characterization of the severity of aggregation behavior. First, three types of feature parameters from high-aggregation areas are processed separately: aggregation intensity, derived from the density overlap coefficient, describes the convergence density of particle paths per unit volume; aggregation trend directionality, determined by directional concentration, indicates the consistency of particle motion directions; and regional coupling density reflects the continuity and coupling degree of particle aggregation characteristics in spatially adjacent regions. For these three types of physical quantities, a one-to-one standard feature mapping function is established, converting their values ​​into standard vector elements to maintain dimensional consistency and comparability between data, ensuring that each parameter has a reasonable weight in the overall index. Subsequently, the three-dimensional feature vector is input into a normalization algorithm, and through min-max transformation or Z-score standardization, the output results are limited to a unified range. This standardized aggregation feature index characterizes the aggregation severity of each aggregation region in continuous numerical form, providing a high-resolution, adjustable quantitative basis for subsequent desorption risk scoring and interface control, thereby enabling effective perception and handling of potential interface instability areas in continuous production.

[0067] In this embodiment, S201 specifically refers to:

[0068] When it is determined that there is a delay in stress release at the interface between the flame retardant and the resin, the particle three-dimensional tracking technology is used to acquire multi-angle synchronous images of the flame retardant particles. The two-dimensional plane projection is converted into trajectory data in the three-dimensional coordinate system through a stereo reconstruction algorithm, thereby obtaining the three-dimensional path data of the flame retardant particles in the continuous flow area, and marking each trajectory point in the form of a time series.

[0069] When a delay in stress release at the flame retardant-resin interface is identified, particle 3D tracking technology uses multiple high-speed visible light or infrared cameras with different angles positioned around the continuous flow region to simultaneously acquire images of the flow behavior of flame retardant particles within the shear section. During acquisition, the system records the two-dimensional spatial projection trajectory of the particles within the field of view of each camera angle and precisely timestamps each frame. Subsequently, a stereo reconstruction algorithm based on stereo geometric projection inversion is used to match and reconstruct the three-dimensional coordinates of the two-dimensional trajectory projections from the multi-angle images, constructing a continuous displacement path of the flame retardant particles in three-dimensional space. This path data is labeled in time series form, ensuring that each trajectory point not only has a clear three-dimensional coordinate position but also specific time information, facilitating subsequent analysis of dynamic parameters such as velocity, tension, and rotation. The core of particle 3D tracking technology lies in achieving image clarity and real-time trajectory analysis of particles in non-transparent high-temperature melts. The key challenges are controlling image synchronization accuracy and correcting imaging distortion under high shear conditions. This technology enables the acquisition of dynamic trajectory data of flame retardant particles in the entire time domain and space during the flow process, providing a data foundation for subsequent construction of aggregation features.

[0070] Based on the three-dimensional path data, the rate of change of the velocity vector of flame retardant particles within a continuous time period is calculated, and a tension distribution curve reflecting the dynamic changes of tensile tension and shear tension is constructed, thereby obtaining the velocity tension characteristics corresponding to the rate of change of the path velocity distribution.

[0071] Calculating the rate of change of flame retardant particle velocity vector over a continuous time period based on three-dimensional path data requires first sorting the three-dimensional positions of each particle at different time points to form a time-stamped trajectory sequence. By analyzing the positional changes between adjacent time points, the instantaneous velocity vector of each trajectory segment is calculated. Then, the trend of velocity vector change over time is analyzed at equal time intervals to obtain the rate of change of particle velocity vector. This rate of change reflects whether the particle has experienced abrupt acceleration, deceleration, or directional change in the flow field, thus deriving the dynamic changes in tensile and shear tension. Tensile tension is mainly manifested as a sudden change in particle velocity along the mainstream direction, while shear tension is reflected as a lateral shift in the velocity vector direction. For example, when a particle's velocity continuously increases from the mainstream direction for a period of time, and then suddenly decreases with a directional shift, it indicates that its location has experienced flow instability and shear disturbance. Aggregating the rate of change of velocity over different time periods into a time series can form a tension distribution curve, describing the dynamic evolution of the particle's stress environment. The flow velocity tension characteristics are extracted from the slope range, amplitude gradient, and trend continuity of this curve, serving as the core basis for judging the strength of the influence of the flow field on the particles. This technology can accurately capture the tension response of microscale particles under complex flow conditions, reflecting whether the interfacial structure of flame retardant particles is in a stable or disturbed state, and is a key foundation for achieving quantitative evaluation of interfacial behavior.

[0072] The continuous flow region is divided into spatial grid cells, and the particle proportion per unit volume in each grid is statistically analyzed. A distribution density gradient is constructed based on the difference in particle proportion between adjacent grids. At the same time, the rotation trajectory of flame retardant particles around the main shear axis is tracked, and the continuous offset of the rotation vector direction over time is analyzed to extract the change in shear rotation direction.

[0073] When analyzing a continuous flow region, the entire region needs to be divided into multiple equal-volume spatial grid cells to construct a three-dimensional monitoring framework with spatial resolution. The number of flame retardant particles in each spatial grid cell is obtained through particle tracking statistics. Then, based on the volume of the grid, the proportion of particles per unit volume is calculated to obtain the density distribution of particles at various spatial locations. Next, the differences in particle proportions between adjacent grids are compared one by one to calculate the density gradient change in space, thereby constructing a complete distribution density gradient field that reflects the aggregation or sparseness trend of flame retardant particles in the flow region. Based on this, to analyze shear rotation behavior, it is necessary to continuously track the trajectory of flame retardant particles around the principal shear axis, extract the rotation vector of each particle in the time series, and arrange these rotation vectors in time order to form a continuous rotation direction evolution curve. This evolution curve can reveal whether the particles experience abrupt changes or shifts in rotation mode during flow shearing, reflecting problems such as vortex disturbances or uneven rotation direction in the flow field. The continuous shift of the rotation vector direction over time can be calculated through the angle change trend to extract quantitative characteristics of the shear rotation direction change. This process integrates spatial density distribution and time-series rotation behavior, enabling it to capture key microscopic changes in flame retardant particles, such as aggregation and loss of rotational consistency, when subjected to shearing. It serves as a fundamental data source for identifying fluctuations in the stability of interface fusion.

[0074] In this embodiment, S202 specifically refers to:

[0075] A three-dimensional spatial grid of equal volume is constructed in the continuous flow region. The three-dimensional path data of flame retardant particles, the flow velocity and tension characteristics of corresponding path points, the distribution density gradient values ​​in the local grid, and the shear rotation direction change data are uniformly merged according to the grid index to generate a multi-parameter fused data matrix.

[0076] To construct a uniform-volume three-dimensional spatial grid in the continuous flow region, the resolution of the spatial division must first be set based on the actual geometric dimensions and particle distribution density of the melt in the shear section, ensuring that each grid cell has sufficient data carrying capacity and statistical significance. Subsequently, the three-dimensional path data of the flame retardant particles is analyzed frame by frame, and the three-dimensional coordinates of the particles at each time point are mapped to grid numbers to determine the particle's spatial location. For each grid cell, the system simultaneously summarizes the velocity change information of all particles in that cell during path tracking, extracts the velocity-tension characteristics representing changes in tensile and shear tension, and loads the distribution density gradient value corresponding to that cell. This value is calculated based on the difference in particle unit volume ratio between the cell and adjacent grids, reflecting the local aggregation trend of particles. Furthermore, the rotation direction change data formed by the particles rotating around the principal shear axis must be integrated. This data, by tracking the angular shift of the rotation vector over time, characterizes the directional consistency of the shear disturbance experienced by the particles in that region. All the above data are aggregated into a unified multi-parameter fusion data matrix using a grid index as the primary key. Each spatial unit is composed of behavioral data from all particles within it, forming a high-dimensional information unit. The completed data matrix serves as the foundational structure for subsequent flow field aggregation feature modeling. The completeness and accuracy of this matrix determine the spatial visualization and computational controllability of the interface fusion behavior, providing core support for achieving high-precision desorption prediction.

[0077] Based on the fused data matrix, the average cosine of the angle between the number of intersections of flame retardant particle paths and the direction vector in each spatial grid is calculated, which are represented as the density overlap coefficient and the direction concentration, respectively. A density-direction joint distribution function is constructed to evaluate the local aggregation intensity of particles in the flow field.

[0078] In the fused data matrix, to evaluate the aggregation behavior of flame retardant particles in the continuous flow region, it is necessary to perform intersection and orientation analysis on the particle paths within each spatial grid. First, all three-dimensional path data are traversed to identify particle pairs with overlapping trajectories within the same grid cell, and the number of path intersections within that grid cell is counted as a density overlap coefficient reflecting the degree of local particle aggregation. A higher number of path intersections indicates that the particle trajectories in that region tend to overlap, and a more pronounced aggregation trend exists in the flow field. Simultaneously, the angles between each pair of particle motion direction vectors within the grid are calculated, and the average of the cosine values ​​is taken to obtain the consistency of particle motion directions in that region, i.e., directional concentration. A higher directional concentration indicates that the particle shear response tends to be uniform, and the aggregation exhibits high synergy. Using the density overlap coefficient and directional concentration as bivariate parameters, a density-direction joint distribution function is constructed to quantify the aggregation intensity of particles in the local space. For example, if the number of particle path intersections in a certain grid is significantly higher than the average level across the entire field, and the directional concentration is close to its maximum value, it means that the flame retardant particles in this region are not only highly dense but also have a uniform shear response direction, indicating that it is a typical aggregation hotspot. This approach can effectively reveal the aggregation behavior of flame retardants under stress release hysteresis, providing a high-precision criterion for desorption prediction and interface fusion control.

[0079] Regions with a density overlap coefficient higher than twice the standard deviation of the overall average density overlap coefficient of the particles and a directional concentration degree exceeding a preset directional concentration threshold are marked as high-aggregation areas. The joint characteristic parameters of these high-aggregation areas are then output as the flow field aggregation characteristics under the stress release delay condition at the flame retardant and resin interface.

[0080] Identifying high-aggregation areas in continuous flow requires a dual-threshold screening based on the density overlap coefficient and directional concentration calculated in the previous stage. First, the density overlap coefficients of all spatial grids across the entire region are statistically analyzed, and their average and standard deviations are calculated. Grid regions with density overlap coefficients exceeding twice the standard deviation of the overall average are marked as density-abnormal aggregation regions. This standard is used to identify spatial units with significant particle trajectory overlap. Next, the directional concentration is assessed, and regions with directional concentrations exceeding a preset threshold are selected. This threshold, determined through historical flow behavior samples or calibration experiments, is used to identify particle groups with highly consistent motion directions. When a spatial unit simultaneously meets both the conditions of exceeding the density overlap coefficient and the directional concentration threshold, it is marked as a high-aggregation region, indicating abnormal aggregation behavior of flame retardant particles under shear response delay. The joint characteristic parameters of these high-aggregation regions, including density overlap, directional concentration, and spatial location index, are extracted and uniformly grouped into flow field aggregation characteristics under the delayed stress release condition at the flame retardant-polycarbonate resin interface. This feature output not only enables spatial localization of aggregation hotspots but also provides specific physical quantity support for subsequent desorption risk assessment and interface control strategies.

[0081] S3. Based on the flow field aggregation characteristics under the delayed stress release at the flame retardant-resin interface, extract the changes in aggregation rate, particle detachment acceleration, and coating reconstruction trend to determine whether the flame retardant has undergone delamination and generate a desorption risk score.

[0082] In this embodiment, S3 specifically includes the following steps:

[0083] S301. Based on the flow field aggregation characteristics under the delayed stress release at the flame retardant and resin interface, extract the variation curve of the number of flame retardant particles in the aggregation region on the continuous time axis, calculate the derivative of the particle density change per unit time as the aggregation rate change amplitude; simultaneously identify the particle path that deviates from the aggregation center trajectory, and determine the particle deviating acceleration based on the path acceleration change.

[0084] After determining the flow field aggregation characteristics under the delayed stress release condition at the flame retardant-resin interface, a continuous time axis data recording system was established within the aggregation region. The number of flame retardant particles in the aggregation region within each time unit was statistically analyzed, and a curve showing the change in the number of flame retardant particles over time was constructed to obtain the trend of particle density variation at different time points. The difference in particle number between adjacent time periods was divided by the corresponding time interval to obtain the derivative of the particle density change per unit time, which was used as a measure of the aggregation rate variation. A large variation in the aggregation rate indicates a severe disturbance in the aggregation state, reflecting the possibility of potential desorption or re-aggregation. Simultaneously, starting from the aggregation center, the three-dimensional paths of all flame retardant particles were traced, identifying particle trajectories that deviated from the aggregation center trend and whose path curvature continuously increased. These trajectories were classified as suspected detachment behavior paths. Further kinematic analysis was performed on these particle paths to calculate the magnitude of velocity change within continuous time periods. The velocity increment was divided by the time interval to obtain the acceleration of each suspected detached particle. The mean and fluctuation range of these particle accelerations were taken as quantitative indicators of particle detachment acceleration. For example, within a certain time window after a high-shear disturbance, if the number of particles in the aggregation region suddenly drops from 1200 per cubic millimeter to 800 per cubic millimeter over a 2-second interval, the density change rate is 200 particles per second, and the density derivative is negative, indicating a significant decrease in the aggregation rate, possibly corresponding to large-scale particle detachment. Simultaneously, if multiple particles are detected escaping radially outward, with their velocity increasing from 2 millimeters per second to 5 millimeters per second and maintaining an acceleration growth trend of more than 0.5 seconds, it can be considered that genuine particle desorption behavior exists. Combining these two technical indicators on the time axis can be used to determine whether there is a significant stratification and desorption trend in the current region.

[0085] S302. Select boundary samples of particle-coated structures within the aggregation region, and conduct joint analysis on their rotation direction vector, stress centripetal trend and interface contact area changes to construct a coating reconstruction trend change function, which is used to indicate whether the particles have a reconstructed and regressive motion tendency after instability.

[0086] S303. Based on the feature vectors of three dimensions—the magnitude of the aggregation rate change, the particle detachment acceleration, and the trend of coating reconstruction—the desorption scoring model is input, and feature weighted fusion and quantitative scoring are performed to generate a desorption risk score in the form of continuous output. When the desorption risk score exceeds the preset threshold, it is judged that the flame retardant has undergone stratification desorption; if it does not exceed the threshold, it is judged that stratification desorption has not occurred.

[0087] In the process of determining stratified detachment, constructing a detachment scoring model based on multi-dimensional feature fusion is the core technical path to achieve accurate identification. This model takes data from three dimensions as input: the magnitude of aggregation rate change, particle detachment acceleration, and the trend of encapsulation remodeling. First, the raw values ​​of the three dimensions are normalized and mapped to a unified feature space to eliminate the influence of dimensional differences. Then, a feature weighting mechanism is introduced, assigning weight coefficients based on the sensitivity and discriminative contribution of each indicator to stratified detachment in historical sample data, constructing a weighted fusion function, and forming a comprehensive score output reflecting the stability of particle behavior. The model is trained using algorithms such as logistic regression or support vector regression, and its parameters are fitted using known detachment labels in the training set, thereby achieving quantitative prediction of the detachment risk of new samples.

[0088] The desorption risk score, as the model's output, is presented as a continuous real number, reflecting the similarity between the particle's current behavior and typical desorption paths. To achieve clear judgment, a preset threshold for desorption determination needs to be set during the model training phase based on a large amount of actual operational data. This threshold serves as a dividing line, typically determined by the point corresponding to the maximum F1 score or the optimal compromise point under the ROC curve. When the desorption risk score of a particle or particle group exceeds this preset threshold, it indicates that its behavior significantly deviates from the normal coating state, tending towards irreversible stratification desorption; therefore, the system determines that stratification desorption has occurred. If the score is below the threshold, it is considered that the interface structure is still in a stable or recoverable adhesion state. Through this judgment mechanism based on feature fusion and quantitative scoring, real-time early warning of desorption risk can be achieved during continuous flow, providing a basis for subsequent process control decisions.

[0089] In this embodiment, S302 specifically refers to:

[0090] Flame retardant particles that are in contact with the matrix interface are extracted from the edge of the aggregation area. A rotation direction vector sequence is constructed to represent the rotation direction and angular displacement evolution trajectory of the particles around their center of mass, and the rotation speed change curve on the time axis is recorded simultaneously.

[0091] When determining whether flame retardant particles exhibit a tendency for coating remodeling, it is first necessary to screen for flame retardant particles with contact relationships with the polycarbonate matrix interface at the edge of the flow field aggregation region. These particles are located at the most vulnerable positions of the interfacial physical coating structure and can reflect the stress response sensitivity characteristics of the coating system. A high-resolution three-dimensional particle tracking system is used to acquire the positional change data of each particle over a continuous time series. Combined with optical speckle characteristics or subtle differences in particle morphology, the rotational behavior of the particle relative to its center of mass is determined. Vector information of the rotation direction of each particle per unit time is constructed in a three-dimensional coordinate system and continuously arranged along the time axis to generate a rotational direction vector sequence, which characterizes the evolution of the angular displacement direction and amplitude of the particle around its center of mass. Simultaneously, the change in the rotational angular velocity of each particle on the time axis is acquired, forming a rotational velocity variation curve. This rotational direction vector sequence reflects the dynamic orientation changes of the particles under the disturbance of local shear stress and interfacial tension, while the rotational velocity variation curve reveals the trend of acceleration or deceleration. Together, they characterize whether the particles are in different dynamic states under the boundary conditions of the coating layer, such as stable rotation, disturbance-induced displacement, or a tendency to realign with the matrix. This information constitutes the basic kinematic data for determining whether particles have the potential for encapsulation deconstruction or reconstruction, and plays a crucial role in subsequent stress centripetal analysis and trend function modeling.

[0092] Based on the rotation direction vector sequence, the stress vector projection distribution of particles toward the aggregation center is calculated, the stress centripetal trend consistent with the particle rotation trend is extracted, and a centripetal coupling index is established; at the same time, the small changes in the contact area between the particle and resin interface over time are tracked to obtain the interface contact area change curve.

[0093] After extracting the characteristics of particle rotation behavior, it is necessary to further establish the stress coupling relationship between the particle and the aggregation center to determine whether the particle exhibits a centripetal reconstruction trend. Specifically, based on the rotation direction vector sequence, the rotation orientation of each particle in three-dimensional space can be vectorized, and the displacement vector of the particle's centroid pointing towards the aggregation center can be calculated by combining its position coordinates. On this basis, the projection of the rotation direction vector onto this displacement vector is analyzed at each time point to obtain the degree of directional consistency between the particle rotation direction and the line connecting the aggregation center, thereby constructing the coupling relationship between the particle rotation trend and the centripetal stress distribution. The stress vector projection distribution extracted in this process is used to measure whether the particle is affected by the shear guiding force facing the aggregation core, and the trend consistency at multiple time points constitutes the stress centripetal trend. Based on this trend, a centripetal coupling degree index is formed to characterize the inherent potential of the particle's reconstruction and encapsulation capability. At the same time, in order to capture the actual change trajectory of the interface structure micromorphology, a high frame rate thermal field response acquisition system is used to track the subtle fluctuations of the contact area between the flame retardant particles and polycarbonate resin in a continuous time series, and the interface contact area change curve is recorded and plotted. This curve reflects whether particles continuously embed, peel off, or re-adhere to the resin matrix, and is a key parameter in the analysis of interfacial structural stability. When the particle rotation trend points towards the aggregation center and the contact area shows an increasing trend, it is often accompanied by active repair of the coating structure or secondary film formation. However, if the two trends are opposite or fluctuate violently, it indicates that the coating relationship is unstable or even in a desorption state. This data provides quantitative input for constructing a complete coating reconstruction trend change function.

[0094] The rotation direction vector sequence, stress centripetal trend distribution and interface contact area change curve are coupled and fitted with multiple parameters to construct a coating reconstruction trend change function. Based on the consistency between the trend rising segment in the function and the path of particles returning to the aggregation center, it is determined whether there is a reconstruction regression motion trend.

[0095] The core of constructing the encapsulation and remodeling trend change function lies in fusing multi-dimensional features of the particle rotation direction vector sequence, stress centripetal trend distribution, and interface contact area change curve to form a continuously evolving fitting model, dynamically characterizing the particle's remodeling behavior tendency after instability. In the specific implementation, firstly, the rotation direction vector sequence is time-normalized to extract the evolution trend of the angle between the rotation orientation and the aggregation center direction at each moment; then, the stress centripetal trend distribution is mapped onto the rotation trajectory according to time points, forming a centripetal stress driving force spectrum synchronized with the rotation behavior; simultaneously, the time points corresponding to each peak and trough in the interface contact area change curve are matched with the above two dimensions, and the change patterns among the three are jointly fitted. The fitting method employs a temporal collaborative modeling algorithm to identify trend-consistent segments, fluctuation correspondences, and the degree of overlap of inflection points, thereby generating an encapsulation and remodeling trend change function that expresses the dynamic behavior of the particles.

[0096] This function not only presents the reconstructed structures of different particles under varying shear conditions, but also quantifies the clarity and persistence of their reconstructive intent based on the steepness of the rising segment, the duration of the stable segment, and the overall curve trend. When determining whether a reconstructive regression trend exists, the slope between the starting point and peak of the rising segment in the reconstructive trend function is compared with the spatial overlap of the particle's actual return path to the aggregation center. If the two are highly consistent and the path converges within the historical reconstructive location region, the particle is considered to have a clear reconstructive regression trend. Conversely, if the curve shows multiple fluctuations or no obvious rising segment, and the particle path is divergent, it can be determined that it lacks a reconstructive trend and may have entered an irreversible desorption stage. The entire analysis process provides precise evidence for subsequent risk quantification and interface control strategy formulation.

[0097] S4. Generate a desorption evaluation vector based on the desorption risk score and historical formation assessment curve, and divide the fusion stable state, desorption critical state and explicit stratification state accordingly.

[0098] In this embodiment, S4 specifically refers to:

[0099] A mapping function is constructed based on the desorption risk score and the historical molding assessment curve. The desorption risk score of the current flame retardant particles is input into the fitting model corresponding to the historical molding assessment curve. The score position and response stratification probability of the desorption risk score in the historical assessment data are extracted, and a two-dimensional coordinate point set containing the risk score and stratification probability is constructed.

[0100] When constructing the mapping function, a systematic sampling and statistical analysis of flame retardant stratification and desorption during historical molding processes is first performed to establish a historical assessment database containing molding parameters, desorption risk scores, and actual stratification results. This database forms a nonlinear mapping relationship between the score and the stratification probability by labeling the risk score of each sample with the corresponding desorption probability. A fitting model is trained using this database, typically employing methods such as Gaussian regression, spline interpolation, or multinomial regression to generate a continuous response curve from the desorption risk score to the stratification probability. The real-time desorption risk score of the current flame retardant particles in the molding flow field is input into this fitting model, and the model outputs its corresponding stratification probability, thus locating the current state in the historical assessment space within a two-dimensional coordinate system. The horizontal axis of the two-dimensional coordinate system represents the real-time desorption risk score, and the vertical axis represents the corresponding stratification probability, constituting a joint expression of risk intensity and instability probability. This forms a score-probability point set for subsequent judgment, providing a basic numerical basis for classifying the interface stability state. The “desorption risk score” in this process reflects the degree of microstructural instability, while the “historical formation assessment curve” carries statistical empirical data. The point set generated by mapping the two is used to quantitatively characterize whether the current particle group is close to the historical instability critical range.

[0101] The desorption risk score and the stratified probability corresponding to the historical molding evaluation curve are constructed into a two-dimensional feature vector. The score slope change rate and historical stability window offset are introduced as supplementary dimensions to generate a desorption evaluation vector containing score intensity, evolution trend and historical offset information, which is used to describe the multidimensional numerical state of the stability of the flame retardant and resin interface.

[0102] To comprehensively characterize the stability of the flame retardant-polycarbonate resin interface during the molding process of high-modulus, low-smoke, halogen-free materials, a two-dimensional feature vector needs to be constructed, combining the desorption risk score with its corresponding stratification probability in historical molding evaluation curves. The risk score reflects the current instability intensity of the microstructure, while the stratification probability represents the likelihood of instability under that risk score in the statistical history. To enhance the dynamic response capability of the desorption trend, the rate of change of the score slope, i.e., the rate of increase or decrease of the risk score within a continuous time window, is introduced to reflect the evolution direction and acceleration of the stratification trend. Simultaneously, to determine whether the current interface state deviates from the normal molding stability region, a historical stability window offset is introduced. This parameter is calculated by comparing the current score position with the center position of the historical stability score interval, thus reflecting the degree to which the system deviates from the ideal process area. Combining the above three types of information—score intensity, trend change, and offset degree—into a multi-dimensional vector constitutes the desorption evaluation vector, forming a multi-source index basis for stability determination. This vector is used to characterize whether the interfacial coupling behavior of flame retardant particle groups is on an evolutionary path of stable fusion, critical loosening, or explicit stratification, thus providing a structured numerical basis for intelligent discrimination.

[0103] The state is divided based on the desorption evaluation vector. A first scoring threshold and a second scoring threshold are set as classification boundaries. If the desorption risk score is lower than the first scoring threshold and the stratification probability is lower than the first probability threshold, it is classified as a fusion stable state. If the desorption risk score is between the first scoring threshold and the second scoring threshold, or the stratification probability is close to the mean range, it is classified as a desorption critical state. If the desorption risk score is higher than the second scoring threshold and the stratification probability exceeds the second probability threshold, it is classified as an explicit stratification state.

[0104] To achieve intelligent classification of the coupling state of the flame retardant-polycarbonate resin interface, multi-level state division is required based on the previously constructed desorption evaluation vector. Specifically, two risk scoring thresholds are set as scoring boundaries defining the stable and unstable intervals, while two probability thresholds are set as references for defining historical stratification risks. The current desorption risk score of the flame retardant particles and the corresponding historical stratification probability together constitute a two-dimensional discrimination space. If the current score is significantly lower than the first scoring threshold, and the stratification probability in the historical curve is lower than the first probability threshold, the system interface is considered tightly coupled with extremely low instability risk, and is classified as a fused stable state. If the score is between the first and second scoring thresholds, or the stratification probability is close to the historical average level, it indicates that the system has a potential risk of short-term disturbance or instability, and is judged as a critical desorption state. If the score is significantly higher than the second scoring threshold, and the stratification probability also exceeds the second probability threshold, it indicates that the flame retardant particles have significantly deviated from their original aggregation center and exhibit a clear interface decoupling trend, and is classified as an explicit stratification state, thus achieving feedforward identification and classification of interface instability states.

[0105] The first scoring threshold is typically set as the mean of desorption scores within the historical stable scoring interval minus one standard deviation, representing the upper limit of the stable state. The second scoring threshold is set as the mean plus one standard deviation, representing the critical point where the instability trend significantly increases. The first probability threshold is the 25th percentile of the stratification probability distribution within the low desorption rate interval during historical molding processes, used to identify extremely low-probability stable states. The second probability threshold is the 75th percentile of the historical stratification probability, used to define the boundary region of high stratification risk. Both the scoring threshold and the probability threshold can be adaptively updated based on sample statistics, ensuring stable and reliable classification capabilities under different material batches or process conditions. These thresholds, serving as classification boundaries, combined with multi-dimensional desorption evaluation vectors, can effectively quantify and classify the interfacial fusion state of flame retardants and polycarbonate resins, facilitating real-time monitoring and risk control during the engineering process.

[0106] S5. Adjust the material feeding rhythm, shear rate change slope and melt path distribution based on the desorption evaluation vector to dynamically control the interfacial fusion state between flame retardant particles and polycarbonate resin.

[0107] In this embodiment, S5 specifically refers to:

[0108] Based on the desorption risk score, stratification probability, score slope change rate and historical stable window offset contained in the desorption evaluation vector, a desorption state control mapping relationship is constructed. The current desorption evaluation vector is input into the mapping model, and the target control parameter set is output, which includes the set values ​​of material feeding rhythm, shear rate change slope and melt path distribution mode that match the current state.

[0109] To achieve dynamic adjustment of the interfacial fusion state between flame retardant particles and polycarbonate resin, a control mapping relationship must first be constructed based on the multidimensional feature information contained in the desorption evaluation vector. This mapping relationship models different combinations of desorption risk scores, stratification probabilities, score slope change rates, and historical stable window offsets, extracting the optimal set of process parameters corresponding to each state from historical molding data. Inputting the current desorption evaluation vector into this mapping model outputs a set of target control parameters that best match the current state, including appropriate material feeding rhythm, shear rate change slope, and melt path distribution. This strategy realizes a closed-loop control mechanism from state identification to process response, effectively avoiding fusion imbalance and escalation of desorption risk caused by parameter solidification, thereby improving the system's response sensitivity and control accuracy to different desorption evolution trends.

[0110] Among them, the desorption risk score measures the strength of the current particle desorption trend, the stratification probability reflects the likelihood of this risk appearing in the historical assessment curve, the rate of change of the score slope describes the growth rate during the risk evolution process, revealing the urgency of the risk escalation, and the historical stability window offset measures the distance between the current state and the existing stable state, which can identify potential stability shift trends. The desorption evaluation vector formed by the coupled expression of these four indicators can fully capture the overall information of interface stability. The control mapping relationship serves as an input-output comparison model. By constructing a nonlinear mapping in a multi-parameter space, complex state characteristics are transformed into executable process parameter adjustment strategies, ensuring that the target control parameter set maintains dynamic consistency with the actual interface behavior, and achieving highly reliable closed-loop process control.

[0111] According to the feeding rhythm of the target material, the synchronous conveying unit of flame retardant particles and polycarbonate resin is rhythmically coordinated, including adjusting the particle feeding speed, the polycarbonate melt feeding rate and the phase matching control of the two, so that the components entering the mixing and shearing section maintain a stable ratio and suppress the fusion imbalance caused by sudden changes in feeding in local areas.

[0112] To achieve precise control of the interfacial fusion state between flame retardant particles and polycarbonate resin, the synchronous conveying unit needs to be rhythmically adjusted according to the target material feeding rhythm. This process includes coordinating and adjusting the particle feeding speed and the polycarbonate melt feeding rate to ensure that the two materials enter the mixing and shearing section synchronously in a set ratio. Furthermore, a phase-matching control mechanism needs to be introduced to monitor and dynamically correct the phase difference between the two material channels in real time, avoiding uneven mixing caused by feeding fluctuations or response lags. When the material feeding rhythm is too fast or uncoordinated, it can easily cause localized ratio shifts, resulting in fusion imbalance during mixing, leading to abnormal flame retardant particle distribution or an increased risk of desorption. By adjusting the rhythm parameters in real time, a stable supply of various components in both spatial and temporal dimensions can be effectively maintained, improving the consistency of interfacial fusion and the stability of the system process.

[0113] The pellet feed speed refers to the output of flame retardant pellets per unit time in the screw or gravity feeder, and its change directly affects the spatial density of the pellets in the mixing section. The polycarbonate melt feed rate is controlled by a melt pump or metering device to determine the resin's coating capacity relative to the pellets and the synchronization of transport. Phase matching control is a key regulation mechanism, which essentially involves real-time phase adjustment during dynamic conveying by comparing the relative positions of the pellet supply peak and the melt flow rate fluctuation cycle to form a synergistic coupling in material rhythm. The three mechanisms work together to create highly consistent input conditions, providing a structural foundation for stable dispersion and fusion in the subsequent mixing and shearing section, and also providing front-end protection for interface stress release and desorption risk control.

[0114] Based on the set values ​​of the target shear rate change slope and melt path distribution, the speed control module of the screw shear section and the flow channel control system are synchronously corrected. This includes adjusting the slope curve of the screw acceleration section to match the flame retardant response window and switching the internal guide configuration of the melt channel to distribute the particle streamline density. This enables dynamic control of the interface fusion state between the flame retardant particles and polycarbonate resin, and ensures the continuity and uniformity of the fusion process.

[0115] To achieve dynamic control of the interfacial fusion state between flame retardant particles and polycarbonate resin, the speed control module of the screw shearing section and the flow channel control system need to be synchronously modified based on the set values ​​of the target shear rate change slope and melt path distribution. In actual operation, firstly, the slope curve of the screw acceleration section is adjusted by the speed control module to make the shear rate change with time or space more smoothly match the response window of the flame retardant particles, that is, the sensitive range of particle shear energy transfer. Secondly, the melt path achieves the redistribution of particle streamline density by switching the internal flow channel guide configuration. By coupling and adjusting these two types of control units, the flame retardant particles can obtain sufficient energy and contact opportunities when entering the high-shear region, while avoiding local desorption or accumulation caused by uneven shear or disordered flow distribution, thereby improving the continuity and uniformity of the fusion process.

[0116] The slope curve of the screw acceleration section refers to the growth trend curve of the shear rate along the screw axis. Its shape determines the rhythm of shear intensity change experienced by the particles. This parameter needs to be adapted to the viscoelastic response window of the flame retardant particles to ensure that the particles achieve sufficient deformation and wetting coating in a plastic state. The speed control module includes a servo motor and control chip, which has real-time response and closed-loop feedback capabilities to precisely control the slope change of the acceleration section. The guide configuration in the flow channel control system is an embedded flow channel geometry design. The guide plate angle, position, or channel shape can be switched by mechanical or fluid drive to adjust the flow line cross-linking density of the resin and particles. Flow line density reflects the degree of aggregation of particle flow tracks per unit cross-section. Higher density means that more particles have the opportunity to contact the polycarbonate melt and form an effective fusion interface. The coordinated adjustment of the two can ensure continuous fusion of the molding area at the macro level and improve the density and uniformity of the interface bonding at the micro level.

[0117] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions according to the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. A computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0118] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0119] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0120] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0121] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0122] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0123] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A continuous production method for high-modulus, low-smoke, halogen-free polycarbonate material, characterized in that, Specifically, the following steps are included: S1. Obtain flow behavior data of flame retardant particles and polycarbonate resin in the shear section, including tensile response trajectory, viscoelastic properties and thermal response changes before and after shear rate change, and construct interfacial stress release time series curve to determine whether there is a delay in the release of interfacial stress between flame retardant and resin. S2. When it is determined that there is a delay in the release of stress at the interface between the flame retardant and the resin, extract the three-dimensional path data, flow velocity tension characteristics, distribution density gradient and shear rotation direction changes of the flame retardant particles in the continuous flow area to determine the flow field aggregation characteristics under the condition of delayed release of stress at the interface between the flame retardant and the resin, and output the standardized aggregation characteristic index. S3. Based on the flow field aggregation characteristics under the delayed stress release at the flame retardant-resin interface, extract the changes in aggregation rate, particle detachment acceleration, and coating reconstruction trend to determine whether the flame retardant has undergone delamination and generate a desorption risk score. S4. Generate a desorption evaluation vector based on the desorption risk score and historical formation assessment curve, and divide the fusion stable state, desorption critical state and explicit stratification state accordingly. S5. Adjust the material feeding rhythm, shear rate change slope and melt path distribution based on the desorption evaluation vector to dynamically control the interfacial fusion state between flame retardant particles and polycarbonate resin.

2. The continuous production method of a high-modulus, low-smoke, halogen-free polycarbonate material according to claim 1, characterized in that, S1 specifically refers to: A multi-parameter synchronous acquisition device is set up in the shear section. The tensile response trajectory of flame retardant particles and polycarbonate resin is collected by the particle displacement tracking unit, the viscoelastic properties before and after the shear rate change are recorded by the dynamic viscoelastic response monitoring unit, and the thermal response change is recorded by the micro-area heat flux sensing unit. The tensile response trajectory is specifically the continuous displacement-time curve of the particle along the shear direction; the viscoelastic properties before and after the shear rate change are specifically the change in the ratio of storage modulus to loss modulus in the two intervals before and after shear fluctuation; and the thermal response change is specifically the curve of heat flux density per unit area of ​​particle interface changing with time. The tensile response trajectory, viscoelastic properties before and after shear rate change, and thermal response change are aligned on the time axis and fused to construct an interfacial stress release time series curve. Based on the time difference between the delayed inflection point of the tensile response trajectory in the constructed interfacial stress release time series curve and the starting point of viscoelastic property recovery, it is determined whether the time difference exceeds a preset hysteresis threshold. If it exceeds the threshold, it is determined that there is a situation of delayed stress release at the interface between the flame retardant and the resin. If it does not exceed the threshold, it is determined that there is no situation of delayed stress release at the interface between the flame retardant and the resin.

3. The continuous production method of a high-modulus, low-smoke, halogen-free polycarbonate material according to claim 1, characterized in that, S2 specifically includes the following steps: S201. When it is determined that there is a delay in stress release at the interface between the flame retardant and the resin, the three-dimensional path data of the flame retardant particles in the continuous flow area is extracted based on the particle three-dimensional tracking technology. The flow velocity tension characteristics are calculated by using the path velocity distribution change rate. The distribution density gradient is constructed by the difference in the unit space ratio of the particles. The change in shear rotation direction is extracted by tracking the rotation trajectory of the particles around the main shear axis. S202. The three-dimensional path data, flow velocity tension characteristics, distribution density gradient and shear rotation direction change are fused in a unified spatial grid to establish a flow field response coupling map. Based on the flow field response coupling map, the density overlap coefficient and directional concentration of the aggregation region are identified, which are used as the flow field aggregation characteristics under the stress release delay of the flame retardant and resin interface. S203. Based on the determined flow field aggregation characteristics, the aggregation intensity, aggregation trend directionality, and regional coupling density are vectorized and encoded using a standard feature mapping function. A normalization algorithm is then used to output standardized aggregation characteristic indicators to characterize the severity of aggregation behavior.

4. The continuous production method of a high-modulus, low-smoke, halogen-free polycarbonate material according to claim 3, characterized in that, S201 specifically refers to: When it is determined that there is a delay in stress release at the interface between the flame retardant and the resin, the particle three-dimensional tracking technology is used to acquire multi-angle synchronous images of the flame retardant particles. The two-dimensional plane projection is converted into trajectory data in the three-dimensional coordinate system through a stereo reconstruction algorithm, thereby obtaining the three-dimensional path data of the flame retardant particles in the continuous flow area, and marking each trajectory point in the form of a time series. Based on the three-dimensional path data, the rate of change of the velocity vector of flame retardant particles within a continuous time period is calculated, and a tension distribution curve reflecting the dynamic changes of tensile tension and shear tension is constructed, thereby obtaining the velocity tension characteristics corresponding to the rate of change of the path velocity distribution. The continuous flow region is divided into spatial grid cells, and the particle proportion per unit volume in each grid is statistically analyzed. A distribution density gradient is constructed based on the difference in particle proportion between adjacent grids. At the same time, the rotation trajectory of flame retardant particles around the main shear axis is tracked, and the continuous offset of the rotation vector direction over time is analyzed to extract the change in shear rotation direction.

5. The continuous production method of a high-modulus, low-smoke, halogen-free polycarbonate material according to claim 3, characterized in that, S202 specifically refers to: A three-dimensional spatial grid of equal volume is constructed in the continuous flow region. The three-dimensional path data of flame retardant particles, the flow velocity and tension characteristics of corresponding path points, the distribution density gradient values ​​in the local grid, and the shear rotation direction change data are uniformly merged according to the grid index to generate a multi-parameter fused data matrix. Based on the fused data matrix, the average cosine of the angle between the number of intersections of flame retardant particle paths and the direction vector in each spatial grid is calculated, which are represented as the density overlap coefficient and the direction concentration, respectively. A density-direction joint distribution function is constructed to evaluate the local aggregation intensity of particles in the flow field. Regions with a density overlap coefficient higher than twice the standard deviation of the overall average density overlap coefficient of the particles and a directional concentration degree exceeding a preset directional concentration threshold are marked as high-aggregation areas. The joint characteristic parameters of these high-aggregation areas are then output as the flow field aggregation characteristics under the stress release delay condition at the flame retardant and resin interface.

6. The continuous production method of a high-modulus, low-smoke, halogen-free polycarbonate material according to claim 1, characterized in that, S3 specifically includes the following steps: S301. Based on the flow field aggregation characteristics under the delayed stress release at the flame retardant and resin interface, extract the variation curve of the number of flame retardant particles in the aggregation region on the continuous time axis, calculate the derivative of the particle density change per unit time as the aggregation rate change amplitude; simultaneously identify the particle path that deviates from the aggregation center trajectory, and determine the particle deviating acceleration based on the path acceleration change. S302. Select boundary samples of particle-coated structures within the aggregation region, and conduct joint analysis on their rotation direction vector, stress centripetal trend and interface contact area changes to construct a coating reconstruction trend change function, which is used to indicate whether the particles have a reconstructed and regressive motion tendency after instability. S303. Based on the feature vectors of three dimensions—the magnitude of the aggregation rate change, the particle detachment acceleration, and the trend of coating reconstruction—the desorption scoring model is input, and feature weighted fusion and quantitative scoring are performed to generate a desorption risk score in the form of continuous output. When the desorption risk score exceeds the preset threshold, it is judged that the flame retardant has undergone stratification desorption; if it does not exceed the threshold, it is judged that stratification desorption has not occurred.

7. The continuous production method of a high-modulus, low-smoke, halogen-free polycarbonate material according to claim 6, characterized in that, S302 specifically refers to: Flame retardant particles that are in contact with the matrix interface are extracted from the edge of the aggregation area. A rotation direction vector sequence is constructed to represent the rotation direction and angular displacement evolution trajectory of the particles around their center of mass, and the rotation speed change curve on the time axis is recorded simultaneously. Based on the rotation direction vector sequence, the stress vector projection distribution of particles toward the aggregation center is calculated, the stress centripetal trend consistent with the particle rotation trend is extracted, and a centripetal coupling index is established; at the same time, the small changes in the contact area between the particle and resin interface over time are tracked to obtain the interface contact area change curve. The rotation direction vector sequence, stress centripetal trend distribution and interface contact area change curve are coupled and fitted with multiple parameters to construct a coating reconstruction trend change function. Based on the consistency between the trend rising segment in the function and the path of particles returning to the aggregation center, it is determined whether there is a reconstruction regression motion trend.

8. The continuous production method of a high-modulus, low-smoke, halogen-free polycarbonate material according to claim 1, characterized in that, S4 specifically refers to: A mapping function is constructed based on the desorption risk score and the historical molding assessment curve. The desorption risk score of the current flame retardant particles is input into the fitting model corresponding to the historical molding assessment curve. The score position and response stratification probability of the desorption risk score in the historical assessment data are extracted, and a two-dimensional coordinate point set containing the risk score and stratification probability is constructed. The desorption risk score and the stratified probability corresponding to the historical molding evaluation curve are constructed into a two-dimensional feature vector. The score slope change rate and historical stability window offset are introduced as supplementary dimensions to generate a desorption evaluation vector containing score intensity, evolution trend and historical offset information, which is used to describe the multidimensional numerical state of the stability of the flame retardant and resin interface. The state is divided based on the desorption evaluation vector. A first scoring threshold and a second scoring threshold are set as classification boundaries. If the desorption risk score is lower than the first scoring threshold and the stratification probability is lower than the first probability threshold, it is classified as a fusion stable state. If the desorption risk score is between the first scoring threshold and the second scoring threshold, or the stratification probability is close to the mean range, it is classified as a desorption critical state. If the desorption risk score is higher than the second scoring threshold and the stratification probability exceeds the second probability threshold, it is classified as an explicit stratification state.

9. The continuous production method of a high-modulus, low-smoke, halogen-free polycarbonate material according to claim 1, characterized in that, S5 specifically refers to: Based on the desorption risk score, stratification probability, score slope change rate and historical stable window offset contained in the desorption evaluation vector, a desorption state control mapping relationship is constructed. The current desorption evaluation vector is input into the mapping model, and the target control parameter set is output, which includes the set values ​​of material feeding rhythm, shear rate change slope and melt path distribution mode that match the current state. According to the feeding rhythm of the target material, the synchronous conveying unit of flame retardant particles and polycarbonate resin is rhythmically coordinated, including adjusting the particle feeding speed, the polycarbonate melt feeding rate and the phase matching control of the two, so that the components entering the mixing and shearing section maintain a stable ratio and suppress the fusion imbalance caused by sudden changes in feeding in local areas. Based on the set values ​​of the target shear rate change slope and melt path distribution, the speed control module of the screw shear section and the flow channel control system are synchronously corrected. This includes adjusting the slope curve of the screw acceleration section to match the flame retardant response window and switching the internal guide configuration of the melt channel to distribute the particle streamline density. This enables dynamic control of the interface fusion state between the flame retardant particles and polycarbonate resin, and ensures the continuity and uniformity of the fusion process.

Images