Hollow core fiber drawing process and device for diameter adaptive control

Abstract

The application discloses a kind of hollow optical fiber drawing process wire diameter self-adaptive control method and device, it is related to optical fiber technical field.Process includes the following steps: real-time acquisition of different process position wire diameter data and process parameters in optical fiber drawing process;According to the wire diameter data and process parameters collected, confirm the current process mode in which drawing process is located;According to the deviation between the preset coupling perception model and the current wire diameter data and target wire diameter data, determine the influence weight of each parameter in the current process parameter on the deviation;According to the wire diameter control strategy corresponding to current process mode and the influence weight of each parameter in the current process parameter on the deviation, adjust current process parameter, so that optical fiber wire diameter is maintained in preset range;Wherein, the process parameter at least includes two different parameters;At least there are two kinds of process modes corresponding to different wire diameter control strategies.

Description

Technical Field

[0001] This application relates to the field of optical fiber technology, and in particular to a method and apparatus for adaptive control of fiber diameter during the drawing process of hollow optical fiber. Background Technology

[0002] Hollow-core optical fiber, with its superior properties such as low latency, low dispersion, and resistance to electromagnetic interference, has broad application prospects in high-speed communication, laser transmission, and sensing. Wire diameter uniformity is a core indicator determining the transmission performance, structural stability, and product yield of hollow-core optical fiber, and high-precision wire diameter control is a key technology for the mass production of hollow-core optical fiber. Currently, in the production of hollow-core optical fiber, the mass production yield has long been below 60%, and production costs remain high, hindering the development and application of hollow-core optical fiber. Summary of the Invention

[0003] This application provides a method and apparatus for adaptive control of fiber diameter during the hollow fiber drawing process, so as to improve the mass production yield of hollow fiber and reduce costs.

[0004] In a first aspect, this application provides a method for adaptive control of fiber diameter during the drawing process of hollow optical fiber, comprising the following steps: Real-time acquisition of fiber diameter data and process parameters at different process positions during optical fiber drawing; Based on the collected wire diameter data and process parameters, the current process mode of the wire drawing process is determined; Based on the preset coupled sensing model and the deviation between the current wire diameter data and the target wire diameter data, the influence weight of each parameter in the current process parameters on the deviation is determined; Based on the fiber diameter control strategy corresponding to the current process mode and the influence weight of each parameter in the current process parameters on the deviation, the current process parameters are adjusted to keep the fiber diameter within a preset range. The process parameters include at least two different parameters; There are at least two different wire diameter control strategies under different process modes.

[0005] This application adjusts the current process parameters by using the wire diameter control strategy corresponding to the current process mode and the influence weight of each parameter in the current process parameters on the deviation, so that the fiber wire diameter is maintained within a preset range. This allows different process parameters to be adjusted by different wire diameter control strategies under different process modes, so that the fiber wire diameter is maintained within a preset range, thereby stabilizing the fluctuation of the hollow fiber wire diameter within a preset range, reducing wire diameter deviation, and improving yield.

[0006] The pre-defined coupled sensing model refers to the coupled sensing model of wire diameter deviation and multiple process parameters established by calibrating the benchmark values ​​of five core parameters: heating power, feeding speed, traction tension, internal air pressure, and traction speed. ΔD=Σ(k_i ΔX_i)+Σ(k_ij ΔX_i ΔX_j)+Σ(k_i' d(ΔX_i) / dt)+(KpD ΔD+KiD ∫ΔDdt+KdD d (ΔD) / dt)+ε; where: Σ(k_i ΔX_i) is a linear term, representing the first-order linear contribution of the five process parameters individually to the wire diameter deviation. Σ(k_ij) ΔX_i ΔX_j) is a nonlinear interaction term, representing the nonlinear contribution of wire diameter deviation caused by the coupling and cross-interference between any two process parameters. Σ(k_i' d(ΔX_i) / dt) is the dynamic differential term, representing the dynamic hysteresis correction of the instantaneous rate of change of the process parameter to the wire diameter deviation. KpD ΔD+KiD ∫ΔDdt+KdD d(ΔD) / dt is the PID (Proportional-Integral-Derivative) closed-loop control term, representing the closed-loop negative feedback correction compensation based on the current wire diameter deviation. ε is the error term, representing the random disturbance residual term that the system cannot model. The preset coupled sensing model is used to characterize the mapping relationship between process parameter changes and wire diameter deviation.

[0007] In some embodiments, the filament diameter data includes at least one of the following: initial filament diameter, core roundness, cured filament diameter, microstructure spacing uniformity, final filament diameter, and core coaxiality. By collecting data on the initial filament diameter, core roundness, cured filament diameter, microstructure spacing uniformity, final filament diameter, and core coaxiality, filament diameter information can be monitored comprehensively, anomalies can be detected promptly, and corresponding adjustments can be made. The initial filament diameter can be measured non-contactly and in real-time online using a laser diameter gauge. Core roundness can be measured using a high-speed coaxial vision inspection instrument combined with image processing algorithms. The cured filament diameter can be measured online using a biaxial orthogonal laser diameter gauge. Microstructure spacing uniformity can be measured using a fiber optic end-face interference microscope (interferometer). The final filament diameter can be measured using a combination of offline multi-point re-measurement with a high-precision micrometer and online laser diameter measurement for full traceability. Core coaxiality can be measured using an eccentricity tester (light transmission method).

[0008] In some embodiments, the process parameters include at least two of the following: heating power, internal air pressure of the fiber core, traction speed, traction tension, and feed rate. Heating power and internal air pressure of the fiber core correspond to the internal environment of fiber drawing, while traction speed, traction tension, and feed rate correspond to the external steady-state environment of fiber drawing. This allows for the construction of an internal fast-response closed loop and an external steady-state optimization closed loop. The internal closed loop uses near-furnace monitoring point data as its core, linking the internal air pressure of the fiber core and the heating power, with a response time ≤25ms. The external closed loop uses shaping monitoring point and pre-coating monitoring point data as its core, linking the feed rate and the traction speed, with a response time ≤100ms. By collecting furnace temperature fluctuations, preform eccentricity, and environmental airflow interference signals, parameter correction commands can be output 10ms~50ms in advance, improving the yield rate.

[0009] Four levels of linked monitoring points can be set up in the fiber drawing tower: near furnace monitoring point, shaping monitoring point, pre-coating monitoring point, and process sensing point. High-frequency laser diameter gauge, high-speed camera, coaxiality diameter gauge, infrared thermometer, and air pressure / torque / tension sensor are used to obtain parameters such as initial fiber diameter, core roundness, fiber diameter after curing, microstructure spacing uniformity, final fiber diameter, core coaxiality, furnace temperature, internal air pressure of the core, traction tension, and feed rate in real time at a sampling frequency of ≥1kHz.

[0010] In some embodiments, the sampling frequency of the heating power is ≥500Hz; and / or, The sampling frequency of the internal air pressure of the fiber core is ≥1kHz; and / or, The sampling frequency of the traction speed is ≥1kHz; and / or, The sampling frequency of the feed rate is ≥1kHz.

[0011] The aforementioned sampling frequency can match the dynamic response characteristics of different controlled objects, enabling high-speed capture of instantaneous parameter disturbances caused by thermal hysteresis, pipeline damping, and mechanical vibration. By using high-frequency data iteration to drive the aforementioned multi-parameter coupled sensing model to complete advanced prediction calculations, the response time can be improved, and errors in the wire drawing process can be corrected in a timely manner.

[0012] In some embodiments, the current process mode includes any one of the following: initial forming mode, steady-state drawing mode, and final finishing mode. Based on the actual drawing situation, the process modes are divided into initial forming mode, steady-state drawing mode, and final finishing mode. For the material state, thermal field distribution, and microstructure forming stress differences of the three stages of hollow fiber preform drawing, the gain coefficient, PID parameters, and five-parameter linkage strategy within the coupled sensing model can be calibrated separately. This allows for targeted adjustment of process parameters according to each different mode, achieving differentiated control.

[0013] In some embodiments, if the current process mode is the initial forming mode, and the absolute value of the deviation between the current fiber diameter data and the target fiber diameter data is greater than a first preset value, then the first fiber diameter control strategy is adopted to adjust the current process parameters so that the fiber diameter is maintained within the preset range. In the initial drawing mode, a dual-leader strategy of air pressure and heating power is adopted, supplemented by a traction tension buffer strategy. When the fiber diameter deviation is > ±1.5μm, the internal air pressure is adjusted first to stabilize the fiber core roundness to above 99%, and the heating power is simultaneously fine-tuned, while the traction tension is maintained at the reference value. This is because in the initial forming mode of fiber drawing, the end of the hollow fiber preform has just entered the high-temperature heating zone, the degree of melting of the quartz glass is uneven, the melt viscosity is in a state of dynamic and drastic change, the anti-resonant thin-walled microstructure support is extremely weak, and irreversible defects such as hole collapse, elliptical distortion, and fiber core out-of-roundness are very likely to occur; or If the current process mode is a steady-state drawing mode, and the absolute value of the deviation between the current fiber diameter data and the target fiber diameter data is greater than the second preset value, then the second fiber diameter control strategy is adopted to adjust the current process parameters so that the fiber diameter is maintained within the preset range. In the steady-state drawing mode, the feed rate and traction speed are coordinated with a pneumatic buffer strategy. When the cumulative fiber diameter deviation is ≥ ±0.8μm, the feed rate and traction speed are dynamically adjusted according to the drawing ratio. The coupled sensing model is calibrated every 5km. This is because the preform heating state, melt viscosity, and thermal field distribution enter a long-term stable range during the steady-state drawing stage, and the hollow core anti-resonance... The microstructure possesses stable self-supporting capabilities, and the core air pressure serves only as a buffer against minor disturbances to maintain the cross-sectional shape. The wire diameter deviation primarily originates from the draw ratio drift caused by the mismatch between the feed rate and the traction speed. By coordinating the feed and traction speeds, the outer diameter deviation can be precisely corrected from the source of material flow, resulting in high linearity and minimal long-term drift. Simultaneously, slight aging of the furnace body and minor attenuation of the air source during ultra-long-distance wire drawing cause a slow shift in model gain. Therefore, the coupled sensing model undergoes parameter self-calibration every 5km to ensure the consistency of the axial wire diameter throughout the ultra-long fiber. The ±0.8μm deviation threshold also meets the high-precision control requirements of steady-state mass production. If the current process mode is the end-stage drawing mode, and the absolute value of the deviation between the current fiber diameter data and the target fiber diameter data is greater than the third preset value, then the third fiber diameter control strategy is adopted to adjust the current process parameters so that the fiber diameter is maintained within the preset range. In the final stage of fiber drawing, a power buffering strategy is adopted, which combines traction tension and feed rate. When the absolute value of the wire diameter deviation is ≥ ±1.0 μm and the remaining length of the preform is less than 800 mm, traction tension is used to compensate for feed attenuation, while the feed rate is slightly adjusted and the heating power is slightly modified. This is because the remaining section of the preform is short at the end of the drawing process, resulting in insufficient axial heat conduction distance and easy tail gradient instability in the furnace thermal field. The preform's own weight causes the feed pushing resistance to increase, and the feed rate to naturally decrease. If the feed rate is significantly adjusted, it can easily cause violent fluctuations in the melt flow rate, directly leading to the collapse of the hollow thin-walled structure and wire breakage. The traction tension has a direct and mild response to the stretching of the molten optical fiber. Therefore, tension is used to compensate for feed attenuation, with only slight adjustment of the feed rate and slight modification of the melt viscosity by heating power. This ensures a smooth finish, maximizes the qualified length of the tail fiber, and avoids a decrease in the overall yield of the preform.

[0014] In some embodiments, the computational model of the first wire diameter control strategy satisfies: ΔD=(kwΔW+kfΔF+ktΔT+kpΔP+kcΔC)+(kwpΔWΔP+kwtΔWΔT+kptΔPΔT)+(kw′d( ΔW) / dt+kp′d(ΔP) / dt+kt′d(ΔT) / dt)+(KpDΔD+KiD∫ΔDdt+KdDd(ΔD) / dt)+ε; Where ΔD is the fiber diameter deviation, in μm; kW is the first-order gain coefficient of heating power with respect to diameter deviation, in μm / kW; ΔW represents the heating power fluctuation, in kW. kf is the first-order gain coefficient of feed rate on diameter deviation, in μm / (mm / min). ΔF represents the feed rate fluctuation, in mm / min; kt is the first-order gain coefficient of tension on diameter deviation, in μm / N; ΔT is the tension fluctuation, in N; kp is the first-order gain coefficient of the internal air pressure of the fiber core to the diameter deviation, in μm / Pa; ΔP is the pressure fluctuation inside the fiber core, in Pa; kc is the first-order gain coefficient of traction speed with respect to diameter deviation, in μm / (m / min). ΔC represents the fluctuation of the traction linear velocity, in m / min; kwp is the heating power-gas pressure cross-coupling gain coefficient, in μm / (kw*Pa). kwt is the heating power-tension cross-coupling gain coefficient, in μm / (kw*N). kpt is the pressure-tension cross-coupling gain coefficient, in μm / (Pa*N). kw′ is the differential adjustment coefficient of the power change rate, with units of μm*s / kw; kp′ is the differential adjustment coefficient of the rate of change of air pressure, in μm*s / Pa; kt′ is the differential adjustment coefficient of the rate of change of tension, in μm*s / N; KpD is the proportional closed-loop adjustment coefficient for diameter deviation, in units of... ; KiD is the integral closed-loop adjustment coefficient for diameter deviation, in units of... ; KdD is the differential closed-loop adjustment coefficient for diameter deviation, in units of... ; ε represents the environmental random disturbance noise term, in μm.

[0015] The aforementioned first filament diameter control strategy relies on a multi-parameter coupled sensing model, using a step-by-step linkage closed-loop control method with the internal air pressure of the fiber core as the dominant variable, heating power as the auxiliary correction variable, and constant traction tension as a bottom-line buffer to correct the filament diameter.

[0016] In some embodiments, the computational model of the second wire diameter control strategy satisfies: ΔD=(kwΔW+kfΔF+ktΔT+kpΔP+kcΔC)+(kfcΔFΔC+kfpΔFΔP+kcpΔCΔP)+(kf′d( ΔF) / dt+kc′d(ΔC) / dt+kp′d(ΔP) / dt)+(KpDΔD+KiD∫ΔDdt+KdDd(ΔD) / dt)+ε; Where ΔD is the fiber diameter deviation, in μm; kW is the first-order gain coefficient of heating power with respect to diameter deviation, in μm / kW; ΔW represents the heating power fluctuation, in kW. kf is the first-order gain coefficient of feed rate on diameter deviation, in μm / (mm / min). ΔF represents the feed rate fluctuation, in mm / min; kt is the first-order gain coefficient of tension on diameter deviation, in μm / N; ΔT is the tension fluctuation, in N; kp is the first-order gain coefficient of the internal air pressure of the fiber core to the diameter deviation, in μm / Pa; ΔP is the pressure fluctuation inside the fiber core, in Pa; kc is the first-order gain coefficient of traction speed with respect to diameter deviation, in μm / (m / min). ΔC represents the fluctuation of the traction linear velocity, in m / min; KFC is the cross-coupling gain coefficient of feed rate and traction rate, in μm / (mm / min*m / min). kfp is the feed rate-pressure cross-coupling gain coefficient, in μm / (mm / min*Pa). kcp is the traction speed-air pressure cross-coupling gain coefficient, in μm / (m / min*Pa). kf′ is the differential adjustment coefficient of the feed rate change rate, in μm*s / (mm / min). kc′ is the differential adjustment coefficient of the traction speed change rate, in μm*s / (m / min). kp′ is the differential adjustment coefficient of the rate of change of air pressure, in μm*s / Pa; KpD is the proportional closed-loop adjustment coefficient for diameter deviation, in units of... ; KiD is the integral closed-loop adjustment coefficient for diameter deviation, in units of... ; KdD is the differential closed-loop adjustment coefficient for diameter deviation, in units of... ; ε represents the environmental random disturbance noise term, in μm.

[0017] The aforementioned second filament diameter control strategy relies on a multi-parameter coupled sensing model. It uses the feed rate and traction rate in coordination with the drawing ratio as the core adjustment, the internal air pressure of the fiber core as a buffer for small disturbances, and a long-term steady-state closed-loop control method that performs self-calibration iteration of the coupled sensing model at fixed intervals to correct the filament diameter.

[0018] In some embodiments, if the current process mode is the end-stage drawing mode, the calculation model of the third wire diameter control strategy satisfies: ΔD=(kwΔW+kfΔF+ktΔT+kpΔP+kcΔC)+(kftΔFΔT+kfwΔFΔW+kwtΔWΔT)+(kf′d( ΔF) / dt+kt′d(ΔT) / dt+kw′d(ΔW) / dt)+(KpDΔD+KiD∫ΔDdt+KdDd(ΔD) / dt)+ε; Where: ΔD is the fiber diameter deviation, in μm; kW is the first-order gain coefficient of heating power with respect to diameter deviation, in μm / kW; ΔW represents the heating power fluctuation, in kW. kf is the first-order gain coefficient of feed rate on diameter deviation, in μm / (mm / min). ΔF represents the feed rate fluctuation, in mm / min; kt is the first-order gain coefficient of tension on diameter deviation, in μm / N; ΔT is the tension fluctuation, in N; kp is the first-order gain coefficient of the internal air pressure of the fiber core to the diameter deviation, in μm / Pa; ΔP is the pressure fluctuation inside the fiber core, in Pa; kc is the first-order gain coefficient of traction speed with respect to diameter deviation, in μm / (m / min). ΔC represents the fluctuation of the traction linear velocity, in m / min; kft is the feed rate-tension cross-coupling gain coefficient, in μm / (mm / min*N). kfw is the cross-coupling gain coefficient of feed rate and heating power, in μm / (mm / min*kw). kwt is the heating power-tension cross-coupling gain coefficient, in μm / (kw*N). kf′ is the differential adjustment coefficient of the feed rate change rate, in μm*s / (mm / min). kt′ is the differential adjustment coefficient of the rate of change of tension, in μm*s / N; kw′ is the differential adjustment coefficient of the power change rate, with units of μm*s / kw; KpD is the proportional closed-loop adjustment coefficient for diameter deviation, in units of... ; KiD is the integral closed-loop adjustment coefficient for diameter deviation, in units of... ; KdD is the differential closed-loop adjustment coefficient for diameter deviation, in units of... ; ε represents the environmental random disturbance noise term, in μm.

[0019] The aforementioned third wire diameter control strategy relies on a multi-parameter coupled sensing model, using traction tension as the dominant compensation variable, slight adjustment of the feed rate, and minor correction of heating power as a buffer to correct the wire diameter in a tail-end steady-state progressive closed-loop control method.

[0020] Secondly, this application provides a wire diameter adaptive control device for the hollow fiber drawing process, comprising: The first execution unit is used to collect fiber diameter data and process parameters at different process positions during the fiber drawing process in real time. The first judgment unit is used to determine the current process mode of the wire drawing process based on the collected wire diameter data and process parameters. The second judgment unit is used to determine the influence weight of each parameter in the current process parameters on the deviation based on the preset coupling sensing model and the deviation between the current wire diameter data and the target wire diameter data. The second execution unit is used to adjust the current process parameters according to the wire diameter control strategy corresponding to the current process mode and the influence weight of each parameter in the current process parameters on the deviation, so as to keep the fiber wire diameter within a preset range. The process parameters include at least two different parameters; There are at least two different wire diameter control strategies under different process modes.

[0021] This application adjusts the current process parameters by using the wire diameter control strategy corresponding to the current process mode and the influence weight of each parameter in the current process parameters on the deviation, so that the fiber wire diameter is maintained within a preset range. This allows different process parameters to be adjusted by different wire diameter control strategies under different process modes, so that the fiber wire diameter is maintained within a preset range, thereby stabilizing the fluctuation of the hollow fiber wire diameter within a preset range, reducing wire diameter deviation, and improving yield. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart of a method for adaptive control of fiber diameter during the hollow fiber drawing process according to an embodiment of this application.

[0024] Figure 2 This is a schematic diagram of a wire diameter adaptive control device for the hollow fiber drawing process according to an embodiment of this application.

[0025] Figure 3 This is a diagram showing the equipment layout in the wire diameter adaptive control method for the hollow fiber drawing process in Embodiments 1 to 3 of this application. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with the embodiments of this application. Obviously, the described embodiments are only some, not all, of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0027] Hollow-core optical fiber, with its superior properties such as low latency, low dispersion, and resistance to electromagnetic interference, has broad application prospects in high-speed communication, laser transmission, and sensing. Wire diameter uniformity is a core indicator determining the transmission performance, structural stability, and product yield of hollow-core optical fiber, and high-precision wire diameter control is a key technology for the mass production of hollow-core optical fiber. Currently, in the production of hollow-core optical fiber, the mass production yield has long been below 60%, and production costs remain high, hindering the development and application of hollow-core optical fiber.

[0028] Currently, in hollow fiber drawing production, wire diameter control generally adopts the traditional control mode of single-parameter closed-loop and end-point detection. It corrects wire diameter deviation solely by adjusting a single parameter such as traction speed or heating power, which has significant technical defects: 1) Strong control lag and poor disturbance response: The response time to instantaneous disturbances such as furnace temperature fluctuations, airflow disturbances, and preform eccentricity is greater than 100ms, and the wire diameter fluctuation range reaches ±1~3μm, which cannot meet the high-precision control requirement of ±0.8μm for hollow fiber; 2) Conflict between wire diameter control and core conformation: [The text abruptly ends here, likely due to an incomplete sentence or missing information.] Linking the internal air pressure of the fiber core with the wire diameter control parameters can easily lead to high-temperature collapse and eccentricity of the air core when correcting the wire diameter, thus damaging the integrity of the microstructure cladding; 3) Insufficient long-distance drawing stability: Under a fixed drawing ratio, long-distance (≥20km) drawing is prone to the accumulation of wire diameter deviation, which can cause stretching distortion of the microstructure cladding and affect the consistency of fiber performance; 4) Lack of multi-mode adaptability: It is impossible to dynamically adjust the control strategy according to different production stages such as the initial stage, steady state, and final stage of drawing, resulting in a mass production yield that is consistently below 60% and high production costs.

[0029] In view of this, this application provides a method and apparatus for adaptive control of wire diameter in the hollow fiber drawing process, so as to improve the mass production yield of hollow fiber and reduce costs.

[0030] Firstly, such as Figure 1 As shown, this application provides a method for adaptive control of fiber diameter during the drawing process of hollow optical fiber, including the following steps: S10. Real-time acquisition of fiber diameter data and process parameters at different process positions during the fiber drawing process; S20. Based on the collected wire diameter data and process parameters, confirm the current process mode of the wire drawing process; S30. Based on the preset coupled sensing model and the deviation between the current wire diameter data and the target wire diameter data, determine the influence weight of each parameter in the current process parameters on the deviation; S40. Based on the fiber diameter control strategy corresponding to the current process mode and the influence weight of each parameter in the current process parameters on the deviation, adjust the current process parameters so that the fiber diameter is maintained within the preset range. The process parameters include at least two different parameters; There are at least two different wire diameter control strategies under different process modes.

[0031] This application adjusts the current process parameters by using the wire diameter control strategy corresponding to the current process mode and the influence weight of each parameter in the current process parameters on the deviation, so that the fiber wire diameter is maintained within a preset range. This allows different process parameters to be adjusted by different wire diameter control strategies under different process modes, so that the fiber wire diameter is maintained within a preset range, thereby stabilizing the fluctuation of the hollow fiber wire diameter within a preset range, reducing wire diameter deviation, and improving yield.

[0032] Based on the collected wire diameter data and process parameters, the current process mode of the wire drawing process can be determined by a combination of three threshold logic operations: real-time detection of the remaining length of the preform, determination of the wire diameter fluctuation range, and determination of the steady-state fluctuation amplitude of the heating power.

[0033] Based on the preset coupled sensing model and the deviation between the current wire diameter data and the target wire diameter data, the influence weight of each parameter in the current process parameters on the deviation can be determined. Specifically, this can be based on the decomposition of the coefficients of the linear and cross-nonlinear terms of the coupled sensing model, combined with the static weight assignment of modal segmentation and the dynamic weight correction operation of the deviation magnitude.

[0034] The pre-defined coupled sensing model refers to the coupled sensing model of wire diameter deviation and multiple process parameters established by calibrating the benchmark values ​​of five core parameters: heating power, feeding speed, traction tension, internal air pressure, and traction speed. ΔD=Σ(k_i ΔX_i)+Σ(k_ij ΔX_i ΔX_j)+Σ(k_i' d(ΔX_i) / dt)+(KpD ΔD+KiD ∫ΔDdt+KdD d (ΔD) / dt)+ε; where: Σ(k_i ΔX_i) is a linear term, representing the first-order linear contribution of the five process parameters individually to the wire diameter deviation. Σ(k_ij) ΔX_i ΔX_j) is a nonlinear interaction term, representing the nonlinear contribution of wire diameter deviation caused by the coupling and cross-interference between any two process parameters. Σ(k_i' d(ΔX_i) / dt) is the dynamic differential term, representing the dynamic hysteresis correction of the instantaneous rate of change of the process parameter to the wire diameter deviation. KpD ΔD+KiD ∫ΔDdt+KdD d(ΔD) / dt is the PID closed-loop control term, representing the closed-loop negative feedback correction compensation based on the current wire diameter deviation. ε is the error term, representing the random disturbance residual term that the system cannot model.

[0035] In conjunction with the first aspect, in some embodiments provided in this application, the wire diameter data includes at least one of the following: initial wire diameter, core roundness, cured wire diameter, microstructure spacing uniformity, final wire diameter, and core coaxiality. By collecting the initial wire diameter, core roundness, cured wire diameter, microstructure spacing uniformity, final wire diameter, and core coaxiality, wire diameter information can be monitored comprehensively, anomalies can be detected in a timely manner, and corresponding adjustments can be made. The initial wire diameter can be measured non-contactly and in real-time online using a laser diameter gauge. Core roundness can be measured using a high-speed coaxial vision inspection instrument combined with image processing algorithms. The cured wire diameter can be measured online using a biaxial orthogonal laser diameter gauge. Microstructure spacing uniformity can be measured by scanning with a fiber optic end-face interference microscope (interferometer). The final wire diameter can be measured using a combination of offline multi-point re-measurement with a high-precision micrometer and online laser diameter measurement with full traceability. Core coaxiality can be measured using an eccentricity tester (light transmission method).

[0036] In conjunction with the first aspect, in some embodiments provided in this application, the process parameters include at least two of the following: heating power, internal air pressure of the fiber core, traction speed, traction tension, and feeding speed. Heating power and internal air pressure of the fiber core correspond to the internal environment of fiber drawing, while traction speed, traction tension, and feeding speed correspond to the external steady-state environment of fiber drawing. This allows for the construction of an internal rapid response closed loop and an external steady-state optimization closed loop. The internal closed loop uses near-furnace monitoring point data as its core, linking the internal air pressure of the fiber core and the heating power, with a response time ≤25ms. The external closed loop uses shaping monitoring point and pre-coating monitoring point data as its core, linking the feeding speed and the traction speed, with a response time ≤100ms. By collecting furnace temperature fluctuations, preform eccentricity, and environmental airflow interference signals, parameter correction commands can be output 10ms~50ms in advance, improving the yield rate.

[0037] A four-level linkage monitoring system can be deployed in the fiber drawing tower, including near-furnace monitoring points, shaping monitoring points, pre-coating monitoring points, and process sensing points. High-frequency laser diameter gauges, high-speed cameras, coaxiality diameter gauges, infrared thermometers, and pressure / torque / tension sensors are used to acquire real-time parameters such as initial fiber diameter, core roundness, post-curing fiber diameter, microstructure spacing uniformity, final fiber diameter, core coaxiality, furnace temperature, internal core pressure, traction tension, and feed rate at a sampling frequency of ≥1kHz. Specifically, the near-furnace monitoring point is installed approximately 5cm downstream of the furnace outlet, using a high-frequency laser diameter gauge and high-speed camera to collect the initial fiber diameter and core roundness at a sampling frequency of ≥3kHz. The shaping monitoring point is installed downstream of the cooling and curing section, using a high-precision laser diameter gauge to collect the post-curing fiber diameter and microstructure spacing uniformity at a sampling frequency of ≥2.5kHz. The pre-coating monitoring point is installed at the front of the UV (ultraviolet) coating mold, using a coaxiality diameter gauge to collect the final fiber diameter and core coaxiality at a sampling frequency of ≥2kHz. Process sensing points are installed in the heating furnace, gas supply mechanism, and traction mechanism. Infrared thermometers, air pressure sensors, and torque sensors are used to collect furnace temperature, internal air pressure of fiber core, internal traction tension, and feeding speed at a sampling frequency of ≥1kHz.

[0038] In conjunction with the first aspect, in some embodiments provided in this application, the sampling frequency of the heating power is ≥500Hz.

[0039] In conjunction with the first aspect, in some embodiments provided in this application, the sampling frequency of the internal air pressure of the fiber core is ≥1kHz.

[0040] In conjunction with the first aspect, in some embodiments provided in this application, the sampling frequency of the traction speed is ≥1kHz.

[0041] In conjunction with the first aspect, in some embodiments provided in this application, the sampling frequency of the feed rate is ≥1kHz.

[0042] The aforementioned sampling frequency can match the dynamic response characteristics of different controlled objects, enabling high-speed capture of instantaneous parameter disturbances caused by thermal hysteresis, pipeline damping, and mechanical vibration. By using high-frequency data iteration to drive the aforementioned multi-parameter coupled sensing model to complete advanced prediction calculations, the response time can be improved, and errors in the wire drawing process can be corrected in a timely manner.

[0043] In conjunction with the first aspect, in some embodiments provided in this application, the current process mode includes any one of the following: initial forming mode, steady-state drawing mode, and final finishing mode. Based on the actual drawing situation, the process modes are divided into initial forming mode, steady-state drawing mode, and final finishing mode. For the material state, thermal field distribution, and microstructure forming stress differences of the three stages of hollow fiber preform drawing, the gain coefficient, PID parameters, and five-parameter linkage strategy within the coupled sensing model can be calibrated respectively. This allows for targeted adjustment of process parameters according to each different mode, achieving differentiated control.

[0044] In conjunction with the first aspect, in some embodiments provided in this application, if the current process mode is the initial forming mode, and the absolute value of the deviation between the current wire diameter data and the target wire diameter data is greater than a first preset value, then the first wire diameter control strategy is adopted to adjust the current process parameters so that the fiber wire diameter is maintained within the preset range. In the initial forming mode of wire drawing, a dual-leader strategy of air pressure and heating power is adopted, supplemented by a traction tension buffer strategy. When the wire diameter deviation is > ±1.5μm, the internal air pressure is adjusted first to stabilize the fiber core roundness to above 99%, and the heating power is adjusted simultaneously. The traction tension is maintained at the reference value. This is because in the initial forming mode of wire drawing, the end of the hollow fiber preform has just entered the high-temperature heating zone. The degree of melting of the quartz glass is uneven and the viscosity of the melt is in a state of dynamic and drastic change. The support force of the anti-resonant thin-walled microstructure is extremely weak, and irreversible defects such as hole collapse, elliptical distortion, and fiber core out-of-roundness are very likely to occur.

[0045] In conjunction with the first aspect, in some embodiments provided in this application, if the current process mode is a steady-state drawing mode, and the absolute value of the deviation between the current fiber diameter data and the target fiber diameter data is greater than a second preset value, then a second fiber diameter control strategy is adopted to adjust the current process parameters so that the fiber diameter is maintained within the preset range. In the steady-state drawing mode, the feed rate and traction speed are coordinated with a pneumatic buffering strategy. When the cumulative fiber diameter deviation is ≥ ±0.8 μm, the feed rate and traction speed are dynamically adjusted according to the drawing ratio. The coupled sensing model is calibrated every 5 km. This is because the preform heating state, melt viscosity, and thermal field distribution enter a long-term state during the steady-state drawing stage. In the stable range, the hollow anti-resonant microstructure has a stable self-supporting capability, and the fiber core air pressure only serves as a buffer for small disturbances to maintain the cross-sectional shape. The wire diameter deviation mainly comes from the wire drawing ratio drift caused by the mismatch between the feed rate and the traction drawing speed. By coordinating the feed rate and the traction speed, the outer diameter deviation can be accurately corrected from the source of material flow, resulting in high linearity and small long-term drift. At the same time, slight aging of the furnace body and slight attenuation of the air source during ultra-long-distance wire drawing will cause the model gain to shift slowly. Therefore, the parameters of the coupled sensing model are self-calibrated every 5km to ensure the consistency of the axial wire diameter of the entire ultra-long optical fiber. The deviation threshold of ±0.8μm also meets the high-precision control requirements of steady-state mass production.

[0046] In conjunction with the first aspect, in some embodiments provided in this application, if the current process mode is the end-stage drawing mode and the absolute value of the deviation between the current fiber diameter data and the target fiber diameter data is greater than a third preset value, then a third fiber diameter control strategy is adopted to adjust the current process parameters so that the fiber diameter is maintained within the preset range. In the final stage of fiber drawing, a power buffering strategy is adopted, which combines traction tension and feed rate. When the absolute value of the wire diameter deviation is ≥ ±1.0 μm and the remaining length of the preform is less than 800 mm, traction tension is used to compensate for feed attenuation, while the feed rate is slightly adjusted and the heating power is slightly modified. This is because the remaining section of the preform is short at the end of the drawing process, resulting in insufficient axial heat conduction distance and easy tail gradient instability in the furnace thermal field. The preform's own weight causes the feed pushing resistance to increase, and the feed rate to naturally decrease. If the feed rate is significantly adjusted, it can easily cause violent fluctuations in the melt flow rate, directly leading to the collapse of the hollow thin-walled structure and wire breakage. The traction tension has a direct and mild response to the stretching of the molten optical fiber. Therefore, tension is used to compensate for feed attenuation, with only slight adjustment of the feed rate and slight modification of the melt viscosity by heating power. This ensures a smooth finish, maximizes the qualified length of the tail fiber, and avoids a decrease in the overall yield of the preform.

[0047] In conjunction with the first aspect, in some embodiments provided in this application, the calculation model of the first wire diameter control strategy satisfies: ΔD=(kwΔW+kfΔF+ktΔT+kpΔP+kcΔC)+(kwpΔWΔP+kwtΔWΔT+kptΔPΔT)+(kw′d( ΔW) / dt+kp′d(ΔP) / dt+kt′d(ΔT) / dt)+(KpDΔD+KiD∫ΔDdt+KdDd(ΔD) / dt)+ε; Where ΔD is the fiber diameter deviation, in μm; kW is the first-order gain coefficient of heating power to diameter deviation, in μm / kW; ΔW represents the heating power fluctuation, in kW. kf is the first-order gain coefficient of feed rate on diameter deviation, in μm / (mm / min). ΔF represents the feed rate fluctuation, in mm / min; kt is the first-order gain coefficient of tension on diameter deviation, in μm / N; ΔT is the tension fluctuation, in N; kp is the first-order gain coefficient of the internal air pressure of the fiber core to the diameter deviation, in μm / Pa; ΔP is the pressure fluctuation inside the fiber core, in Pa; kc is the first-order gain coefficient of traction speed with respect to diameter deviation, in μm / (m / min). ΔC represents the fluctuation of the traction linear velocity, in m / min; kwp is the heating power-gas pressure cross-coupling gain coefficient, in μm / (kw*Pa). kwt is the heating power-tension cross-coupling gain coefficient, in μm / (kw*N). kpt is the pressure-tension cross-coupling gain coefficient, in μm / (Pa*N). kw′ is the differential adjustment coefficient of the power change rate, with units of μm*s / kw; kp′ is the differential adjustment coefficient of the rate of change of air pressure, in μm*s / Pa; kt′ is the differential adjustment coefficient of the rate of change of tension, in μm*s / N; KpD is the proportional closed-loop adjustment coefficient for diameter deviation, in units of... ; KiD is the integral closed-loop adjustment coefficient for diameter deviation, in units of... ; KdD is the differential closed-loop adjustment coefficient for diameter deviation, in units of... ; ε represents the environmental random disturbance noise term, in μm.

[0048] The aforementioned first filament diameter control strategy relies on a multi-parameter coupled sensing model, using a step-by-step linkage closed-loop control method with the internal air pressure of the fiber core as the dominant variable, heating power as the auxiliary correction variable, and constant traction tension as a bottom-line buffer to correct the filament diameter.

[0049] In conjunction with the first aspect, in some embodiments provided in this application, the calculation model of the second wire diameter control strategy satisfies: ΔD=(kwΔW+kfΔF+ktΔT+kpΔP+kcΔC)+(kfcΔFΔC+kfpΔFΔP+kcpΔCΔP)+(kf′d( ΔF) / dt+kc′d(ΔC) / dt+kp′d(ΔP) / dt)+(KpDΔD+KiD∫ΔDdt+KdDd(ΔD) / dt)+ε; Where ΔD is the fiber diameter deviation, in μm; kW is the first-order gain coefficient of heating power to diameter deviation, in μm / kW; ΔW represents the heating power fluctuation, in kW. kf is the first-order gain coefficient of feed rate on diameter deviation, in μm / (mm / min). ΔF represents the feed rate fluctuation, in mm / min; kt is the first-order gain coefficient of tension on diameter deviation, in μm / N; ΔT is the tension fluctuation, in N; kp is the first-order gain coefficient of the internal air pressure of the fiber core to the diameter deviation, in μm / Pa; ΔP is the pressure fluctuation inside the fiber core, in Pa; kc is the first-order gain coefficient of traction speed with respect to diameter deviation, in μm / (m / min). ΔC represents the fluctuation of the traction linear velocity, in m / min; KFC is the cross-coupling gain coefficient of feed rate and traction rate, in μm / (mm / min*m / min). kfp is the feed rate-pressure cross-coupling gain coefficient, in μm / (mm / min*Pa). kcp is the traction speed-air pressure cross-coupling gain coefficient, in μm / (m / min*Pa). kf′ is the differential adjustment coefficient of the feed rate change rate, in μm*s / (mm / min). kc′ is the differential adjustment coefficient of the traction speed change rate, in μm*s / (m / min). kp′ is the differential adjustment coefficient of the rate of change of air pressure, in μm*s / Pa; KpD is the proportional closed-loop adjustment coefficient for diameter deviation, in units of... ; KiD is the integral closed-loop adjustment coefficient for diameter deviation, in units of... ; KdD is the differential closed-loop adjustment coefficient for diameter deviation, in units of... ; ε represents the environmental random disturbance noise term, in μm.

[0050] The aforementioned second filament diameter control strategy relies on a multi-parameter coupled sensing model. It uses the feed rate and traction rate in coordination with the drawing ratio as the core adjustment, the internal air pressure of the fiber core as a buffer for small disturbances, and a long-term steady-state closed-loop control method that performs self-calibration iteration of the coupled sensing model at fixed intervals to correct the filament diameter.

[0051] In conjunction with the first aspect, in some embodiments provided in this application, if the current process mode is the end-stage drawing mode, the calculation model of the third wire diameter control strategy satisfies: ΔD=(kwΔW+kfΔF+ktΔT+kpΔP+kcΔC)+(kftΔFΔT+kfwΔFΔW+kwtΔWΔT)+(kf′d( ΔF) / dt+kt′d(ΔT) / dt+kw′d(ΔW) / dt)+(KpDΔD+KiD∫ΔDdt+KdDd(ΔD) / dt)+ε; Where: ΔD is the fiber diameter deviation, in μm; kW is the first-order gain coefficient of heating power to diameter deviation, in μm / kW; ΔW represents the heating power fluctuation, in kW. kf is the first-order gain coefficient of feed rate on diameter deviation, in μm / (mm / min). ΔF represents the feed rate fluctuation, in mm / min; kt is the first-order gain coefficient of tension on diameter deviation, in μm / N; ΔT is the tension fluctuation, in N; kp is the first-order gain coefficient of the internal air pressure of the fiber core to the diameter deviation, in μm / Pa; ΔP is the pressure fluctuation inside the fiber core, in Pa; kc is the first-order gain coefficient of traction speed with respect to diameter deviation, in μm / (m / min). ΔC represents the fluctuation of the traction linear velocity, in m / min; kft is the feed rate-tension cross-coupling gain coefficient, in μm / (mm / min*N). kfw is the cross-coupling gain coefficient of feed rate and heating power, in μm / (mm / min*kw). kwt is the heating power-tension cross-coupling gain coefficient, in μm / (kw*N). kf′ is the differential adjustment coefficient of the feed rate change rate, in μm*s / (mm / min). kt′ is the differential adjustment coefficient of the rate of change of tension, in μm*s / N; kw′ is the differential adjustment coefficient of the power change rate, with units of μm*s / kw; KpD is the proportional closed-loop adjustment coefficient for diameter deviation, in units of... ; KiD is the integral closed-loop adjustment coefficient for diameter deviation, in units of... ; KdD is the differential closed-loop adjustment coefficient for diameter deviation, in units of... ; ε represents the environmental random disturbance noise term, in μm.

[0052] The aforementioned third wire diameter control strategy relies on a multi-parameter coupled sensing model, using traction tension as the dominant compensation variable, slight adjustment of the feed rate, and minor correction of heating power as a buffer to correct the wire diameter in a tail-end steady-state progressive closed-loop control method.

[0053] Secondly, such as Figure 2 As shown, this application provides a wire diameter adaptive control device for the hollow fiber drawing process, comprising: The first execution unit is used to collect fiber diameter data and process parameters at different process positions during the fiber drawing process in real time. The first judgment unit is used to determine the current process mode of the wire drawing process based on the collected wire diameter data and process parameters. The second judgment unit is used to determine the influence weight of each parameter in the current process parameters on the deviation based on the preset coupling sensing model and the deviation between the current wire diameter data and the target wire diameter data. The second execution unit is used to adjust the current process parameters according to the wire diameter control strategy corresponding to the current process mode and the influence weight of each parameter in the current process parameters on the deviation, so as to keep the fiber wire diameter within a preset range. The process parameters include at least two different parameters; There are at least two different wire diameter control strategies under different process modes.

[0054] This application adjusts the current process parameters by using the wire diameter control strategy corresponding to the current process mode and the influence weight of each parameter in the current process parameters on the deviation, so that the fiber wire diameter is maintained within a preset range. This allows different process parameters to be adjusted by different wire diameter control strategies under different process modes, so that the fiber wire diameter is maintained within a preset range, thereby stabilizing the fluctuation of the hollow fiber wire diameter within a preset range, reducing wire diameter deviation, and improving yield.

[0055] The technical solutions provided in this application will be described in detail below with reference to the embodiments.

[0056] Example 1 Hollow fiber diameter control in the initial drawing mode: This method is designed for the initial drawing stage of hollow fiber, when the preform has just entered the heating zone and the fiber has not yet been stably formed, to achieve rapid diameter stabilization and core shape preservation.

[0057] exist Figure 3 Monitoring setup: Real-time acquisition of initial filament diameter, core roundness, furnace temperature, internal air pressure, heating power, and traction tension using near-furnace monitoring points (3kHz) and process sensing points (1kHz); Benchmark calibration was established: Historical data regression analysis was used to determine benchmark values ​​for heating power, internal air pressure, and traction tension. A coupled model was established between filament diameter and air pressure, heating power, and traction tension. The heating power was set at 23 kW, the internal air pressure at 2 mbar, and the traction tension at 300 N. The coupled sensing model was established as a dynamic equation for filament diameter deviation, constructed based on linear terms of air pressure, heating power, and traction tension, pairwise cross-coupled nonlinear terms, differential terms of parameter change rate, and a closed-loop PID correction term for filament diameter. The influence weights of air pressure on filament diameter and core roundness were set at 40%, heating power on filament diameter and core roundness at 40%, and traction tension on filament diameter and core roundness at 20%. In the initial stage of fiber drawing, the measured filament diameter was larger than expected, reaching 232.8 μm. The industrial control computer responded quickly, increasing the internal air pressure to 2.1 mbar (20 ms response) and simultaneously increasing the heating power to 24.25 kW (30 ms response). Within 10 seconds, the filament diameter dropped back to 231.0 ± 0.4 μm, and the core roundness stabilized at 99.6%. Disturbance response: When encountering a furnace temperature fluctuation of +6℃ during the wire drawing process, the feedforward module fine-tunes the heating power 15ms in advance, and there is no significant fluctuation in wire diameter and no accumulation of deviation. Results: The finished optical fiber diameter fluctuation was ±0.5μm, the core roundness was 99.6%, the 50km drawing steady-state rate was 99.2%, the yield was 98.8%, and the transmission loss in the 1.55μm band was 0.05dB / km, meeting the requirements for large-scale production and engineering applications.

[0058] Example 2 Hollow fiber diameter control in steady-state drawing mode, targeting long-distance steady-state drawing conditions, aiming for continuous drawing of 50km, suppressing deviation accumulation, and ensuring microstructure uniformity.

[0059] exist Figure 3 Under these circumstances, the monitoring deployment is as follows: a four-level full-link monitoring system is activated, specifically: a first-level monitoring system using high-frequency laser diameter measurement at the furnace opening; a second-level monitoring system using a high-speed camera on the end face of the bare fiber; a third-level monitoring system using high-precision laser diameter measurement before coating; and a fourth-level monitoring system for coaxiality and roundness before winding. Throughout the entire process, the system synchronously collects filament diameter data, core roundness, coaxiality, and a complete set of process parameters, including feed rate, traction speed, internal air pressure of the core, and furnace temperature, at a sampling frequency of ≥2.5kHz. The system focuses on collecting filament diameter data and process parameters at the shaping monitoring point and the pre-coating monitoring point. Determine the benchmark calibration: Using the drawing ratio as the core, determine the benchmark parameters of feed rate, traction speed, and internal air pressure through historical data regression analysis, and establish a long-distance coupled sensing model; set the feed rate to 1.2 mm / min, the traction speed to 55 m / min, and the internal air pressure to 2.2 mbar; establish the coupled sensing model as a dynamic equation for wire diameter deviation based on linear terms of feed rate, traction speed, and gas pressure, pairwise cross-coupled nonlinear terms, differential terms of parameter change rate, and wire diameter closed-loop PID correction terms; set the influence weight of traction speed on wire diameter to 40%, the influence weight of feed rate on wire diameter to 40%, and the influence weight of internal air pressure on wire diameter to 20%. A coordinated air pressure buffering strategy is implemented to coordinate the feeding speed and traction speed; when the cumulative filament diameter deviation is ≥±0.8μm, the feeding and traction speeds are dynamically and synchronously adjusted according to the drawing ratio to ensure proportional stretching of the microstructure cladding; model parameters are automatically calibrated every 5km; and internal air pressure is linked to maintain filament diameter stability. Results: The diameter fluctuation of the filament throughout the 50km drawing process was ≤±0.5μm, the microstructure was undistorted, and the continuous steady-state rate was ≥99%.

[0060] Example 3 Hollow fiber diameter control in the final drawing mode is designed to prevent sudden changes in fiber diameter and fiber breakage, especially in cases of insufficient preform margin and feed attenuation at the end of the drawing process.

[0061] exist Figure 3 In this case, the focus is on monitoring the tension sensor, feed rate, heating power and final wire diameter to identify the preform balance status in real time; Benchmark calibration was established: Through regression analysis of historical data, a coupled model of traction tension, feed rate, and heating power was established, and a final buffer threshold was set when the absolute value of the wire diameter deviation was ≥ ±1.0 μm and the remaining length of the preform was less than 800 mm. The feed rate was set to 1.0 mm / min, the traction speed to 50 m / min, and the heating power to 20.5 KW. A coupled sensing model was established, which is a dynamic equation for wire diameter deviation constructed based on linear terms of feed rate, traction tension, and heating power, pairwise cross-coupled nonlinear terms, differential terms of parameter change rate, and wire diameter closed-loop PID correction terms. The influence weight of traction tension on wire diameter was set to 40%, the influence weight of feed rate on wire diameter was set to 40%, and the influence weight of heating power on wire diameter was set to 20%. The system employs a power buffering strategy, adjusting traction tension and feed rate. Traction tension is the primary adjustment mechanism to compensate for feed rate attenuation and prevent sudden thinning of the filament diameter. The feed rate is adjusted as the remaining length of the preform decreases. Heating power is slightly modified to maintain glass viscosity, specifically through a gradient power reduction adjustment as the remaining length of the preform decreases: power is reduced by 0.3kW each time in the 800mm~500mm length range, and by 0.5kW each time in the 500mm~200mm length range, with a minimum protection value of 18.5kW. By gradually lowering the furnace temperature, the excessively soft fluidity of the quartz melt is reduced, offsetting the tensile distortion caused by the weight of the preform at the tail end, preventing excessive thinning of the melt that could lead to sudden changes in filament diameter, collapse of the hollow microstructure, and fiber breakage. Only small-amplitude, step-by-step adjustments are made throughout the process, avoiding large-power jumps to prevent secondary deviations caused by drastic thermal fluctuations and fiber breakage. Results: There were no fiber breaks, no thinning, and no cladding distortion at the end of the fiber drawing process. The diameter of the tail section remained stable within ±0.7μm, and the fiber utilization rate was significantly improved.

[0062] Comparative Example 1 The traditional control mode, which uses single-parameter closed-loop and end-point detection, corrects the wire diameter deviation only by adjusting the traction speed.

[0063] Specifically, the end wire diameter is only detected by a single-point laser diameter measuring instrument at the back end of the fiber winding. The detection deviation is directly output to the traction motor through a single-loop PID controller. The heating power, internal air pressure, and feeding speed are locked at fixed reference values ​​throughout the process and do not participate in linkage correction. No multi-parameter coupled sensing model has been established, and no differentiated control has been made for the three drawing modes of initial, steady state, and final stage. There are no parameter cross-coupling terms or dynamic differential term advance compensation.

[0064] Results: Under the same 50km long-distance drawing conditions, the long-term cumulative fluctuation deviation of the filament diameter can reach ±2.2μm, the core roundness fluctuation is >5%, and the hollow anti-resonance microstructure is prone to local collapse and uneven spacing distortion. When the preform is insufficient at the end of the drawing process, the feeding attenuation is very likely to cause a sudden change in filament diameter and fiber breakage. The effective yield of the whole preform is only 76%, and the continuous steady-state drawing qualification rate is only 82%.

[0065] Comparative Example 2 The traditional control mode, which uses single-parameter closed-loop and end-point detection, corrects the wire diameter deviation only by adjusting the heating power.

[0066] Specifically, it also adopts the single-point detection method at the end of the back end, and only outputs the wire diameter deviation in a closed loop to the heating power actuator of the high-temperature furnace. It passively corrects the outer diameter by relying on the change of melt viscosity, while the traction speed, feeding speed and internal air pressure remain constant. It does not make dynamic prediction and compensation for the large hysteresis characteristics of thermal inertia, and has no segmented modal weight assignment and long-distance model self-calibration mechanism. It belongs to a pure hysteresis passive correction scheme.

[0067] Results: Due to the large inertial lag of the furnace body, the adjustment response lag is large, the wire diameter oscillation is obvious, and the wire diameter fluctuation range is ±2.5μm throughout the process; large power fluctuations will cause thermal deformation of the hollow thin-walled structure, and the coaxiality deviation exceeds the standard; long-distance wire drawing drift cannot be suppressed, the axial consistency is extremely poor at 50km, the probability of fiber breakage is high at the tail end stage, and the effective utilization rate of optical fiber is only 73%, which is far lower than the index of the three-mode coupled sensing control scheme of this invention.

[0068] As can be seen from Examples 1-3 and Comparative Examples 1 to 2, this application has the following advantages: Breakthrough in fiber diameter control precision: Through multi-mode coupling regulation, the fiber diameter fluctuation of hollow fiber is stabilized within ±0.5μm, which is far lower than the industry's traditional ±1μm level, and the consistency of the standard 125μm outer diameter is improved by more than 2 times. Core shape preservation and wire diameter coordination: The core roundness is ≥99.5% throughout the drawing process, with no collapse or eccentricity issues, and the microstructure cladding spacing deviation is ≤0.1μm, which improves the problem of conflict between wire diameter correction and core structure. Excellent long-distance fiber drawing stability: After drawing fibers over a distance of 50km, the fiber diameter deviation is still controlled within ±0.8μm, the microstructure cladding is undistorted, and the continuous fiber drawing stability rate is ≥99%; Mass production yield has been greatly improved: the yield rate has increased from below 60% in traditional processes to over 90%, a single preform can be drawn into more than 50km of optical fiber, and production costs have been reduced by 45%; Millisecond-level response speed: internal closed-loop response time ≤25ms, feedforward prediction outputs correction instructions 10ms~50ms in advance, quickly suppressing instantaneous disturbances and eliminating deviation accumulation.

[0069] In summary, by adjusting the fiber diameter control strategy corresponding to the current process mode and the influence weight of each parameter in the current process parameters on the deviation, the current process parameters can be adjusted to keep the fiber diameter within a preset range. This allows different process parameters to be adjusted through different fiber diameter control strategies under different process modes, thereby keeping the fiber diameter within a preset range. As a result, the fluctuation of the hollow fiber diameter is stabilized within a preset range, reducing the fiber diameter deviation and improving the yield.

[0070] In the description of this specification, the references to terms such as "one embodiment / mode," "some embodiments / modes," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment / mode or example is included in at least one embodiment / mode or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment / mode or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments / modes or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments / modes or examples described in this specification, as well as the features of different embodiments / modes or examples.

[0071] It should be noted that in this application, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. In this application, "a plurality of" means at least two, such as two, three, etc., unless otherwise expressly specified.

[0072] The above description is merely a specific embodiment of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for adaptive control of fiber diameter during hollow fiber drawing, characterized in that, Includes the following steps: Real-time acquisition of fiber diameter data and process parameters at different process positions during optical fiber drawing; Based on the collected wire diameter data and process parameters, the current process mode of the wire drawing process is determined; Based on the preset coupled sensing model and the deviation between the current wire diameter data and the target wire diameter data, the influence weight of each parameter in the current process parameters on the deviation is determined; Based on the fiber diameter control strategy corresponding to the current process mode and the influence weight of each parameter in the current process parameters on the deviation, the current process parameters are adjusted to keep the fiber diameter within a preset range. The process parameters include at least two different parameters; There are at least two different wire diameter control strategies under different process modes.

2. The method for adaptive control of fiber diameter in the hollow fiber drawing process as described in claim 1, characterized in that, The filament diameter data includes at least one of the following: initial filament diameter, core roundness, filament diameter after curing, microstructure spacing uniformity, final filament diameter, and core coaxiality.

3. The method for adaptive control of fiber diameter in the hollow fiber drawing process as described in claim 1, characterized in that, The process parameters include at least two of the following: heating power, internal air pressure of the fiber core, traction speed, traction tension, and feeding speed.

4. The fiber diameter adaptive control method for the hollow fiber drawing process as described in claim 3, characterized in that: The sampling frequency of the heating power is ≥500Hz; and / or, The sampling frequency of the internal air pressure of the fiber core is ≥1kHz; and / or, The sampling frequency of the traction speed is ≥1kHz; and / or, The sampling frequency of the feed rate is ≥1kHz.

5. The method for adaptive control of fiber diameter in the hollow fiber drawing process as described in claim 1, characterized in that, The current process mode includes any one of the following: initial forming mode, steady-state wire drawing mode, and final finishing mode.

6. The method for adaptive control of fiber diameter in the hollow fiber drawing process as described in claim 5, characterized in that: If the current process mode is the initial forming mode, and the absolute value of the deviation between the current fiber diameter data and the target fiber diameter data is greater than a first preset value, then the first fiber diameter control strategy is used to adjust the current process parameters to maintain the fiber diameter within the preset range; or, If the current process mode is a steady-state drawing mode, and the absolute value of the deviation between the current fiber diameter data and the target fiber diameter data is greater than a second preset value, then the second fiber diameter control strategy is used to adjust the current process parameters to maintain the fiber diameter within the preset range; or, If the current process mode is the end-stage drawing mode, and the absolute value of the deviation between the current fiber diameter data and the target fiber diameter data is greater than the third preset value, then the third fiber diameter control strategy is adopted to adjust the current process parameters so that the fiber diameter is maintained within the preset range.

7. The method for adaptive control of fiber diameter in the hollow fiber drawing process as described in claim 6, characterized in that, The calculation model of the first wire diameter control strategy satisfies: ΔD=(kwΔW+kfΔF+ktΔT+kpΔP+kcΔC)+(kwpΔWΔP+kwtΔWΔT+kptΔPΔT)+(kw′d( ΔW) / dt+kp′d(ΔP) / dt+kt′d(ΔT) / dt)+(KpDΔD+KiD∫ΔDdt+KdDd(ΔD) / dt)+ε; Where ΔD is the fiber diameter deviation, in μm; kW is the first-order gain coefficient of heating power with respect to diameter deviation, in μm / kW; ΔW represents the heating power fluctuation, in kW. kf is the first-order gain coefficient of feed rate on diameter deviation, in μm / (mm / min). ΔF represents the feed rate fluctuation, in mm / min; kt is the first-order gain coefficient of tension on diameter deviation, in μm / N; ΔT is the tension fluctuation, in N; kp is the first-order gain coefficient of the internal air pressure of the fiber core to the diameter deviation, in μm / Pa; ΔP is the pressure fluctuation inside the fiber core, in Pa; kc is the first-order gain coefficient of traction speed with respect to diameter deviation, in μm / (m / min). ΔC represents the fluctuation of the traction linear velocity, in m / min; kwp is the heating power-gas pressure cross-coupling gain coefficient, in μm / (kw*Pa). kwt is the heating power-tension cross-coupling gain coefficient, in μm / (kw*N). kpt is the pressure-tension cross-coupling gain coefficient, in μm / (Pa*N). kw′ is the differential adjustment coefficient of the power change rate, with units of μm*s / kw; kp′ is the differential adjustment coefficient of the rate of change of air pressure, in μm*s / Pa; kt′ is the differential adjustment coefficient of the rate of change of tension, in μm*s / N; KpD is the proportional closed-loop adjustment coefficient for diameter deviation, in units of... ; KiD is the integral closed-loop adjustment coefficient for diameter deviation, in units of... ; KdD is the differential closed-loop adjustment coefficient for diameter deviation, in units of... ; ε represents the environmental random disturbance noise term, in μm.

8. The method for adaptive control of fiber diameter in the hollow fiber drawing process as described in claim 6, characterized in that, The calculation model of the second wire diameter control strategy satisfies: ΔD=(kwΔW+kfΔF+ktΔT+kpΔP+kcΔC)+(kfcΔFΔC+kfpΔFΔP+kcpΔCΔP)+(kf′d( ΔF) / dt+kc′d(ΔC) / dt+kp′d(ΔP) / dt)+(KpDΔD+KiD∫ΔDdt+KdDd(ΔD) / dt)+ε; Where ΔD is the fiber diameter deviation, in μm; kW is the first-order gain coefficient of heating power with respect to diameter deviation, in μm / kW; ΔW represents the heating power fluctuation, in kW. kf is the first-order gain coefficient of feed rate on diameter deviation, in μm / (mm / min). ΔF represents the feed rate fluctuation, in mm / min; kt is the first-order gain coefficient of tension on diameter deviation, in μm / N; ΔT is the tension fluctuation, in N; kp is the first-order gain coefficient of the internal air pressure of the fiber core to the diameter deviation, in μm / Pa; ΔP is the pressure fluctuation inside the fiber core, in Pa; kc is the first-order gain coefficient of traction speed with respect to diameter deviation, in μm / (m / min). ΔC represents the fluctuation of the traction linear velocity, in m / min; KFC is the cross-coupling gain coefficient of feed rate and traction rate, in μm / (mm / min*m / min). kfp is the feed rate-pressure cross-coupling gain coefficient, in μm / (mm / min*Pa). kcp is the traction speed-air pressure cross-coupling gain coefficient, in μm / (m / min*Pa). kf′ is the differential adjustment coefficient of the feed rate change rate, in μm*s / (mm / min). kc′ is the differential adjustment coefficient of the traction speed change rate, in μm*s / (m / min). kp′ is the differential adjustment coefficient of the rate of change of air pressure, in μm*s / Pa; KpD is the proportional closed-loop adjustment coefficient for diameter deviation, in units of... ; KiD is the integral closed-loop adjustment coefficient for diameter deviation, in units of... ; KdD is the differential closed-loop adjustment coefficient for diameter deviation, in units of... ; ε represents the environmental random disturbance noise term, in μm.

9. The method for adaptive control of fiber diameter in the hollow fiber drawing process as described in claim 6, characterized in that, If the current process mode is the end-stage drawing mode, the calculation model of the third wire diameter control strategy satisfies: ΔD=(kwΔW+kfΔF+ktΔT+kpΔP+kcΔC)+(kftΔFΔT+kfwΔFΔW+kwtΔWΔT)+(kf′d( ΔF) / dt+kt′d(ΔT) / dt+kw′d(ΔW) / dt)+(KpDΔD+KiD∫ΔDdt+KdDd(ΔD) / dt)+ε; Where: ΔD is the fiber diameter deviation, in μm; kW is the first-order gain coefficient of heating power with respect to diameter deviation, in μm / kW; ΔW represents the heating power fluctuation, in kW. kf is the first-order gain coefficient of feed rate on diameter deviation, in μm / (mm / min). ΔF represents the feed rate fluctuation, in mm / min; kt is the first-order gain coefficient of tension on diameter deviation, in μm / N; ΔT is the tension fluctuation, in N; kp is the first-order gain coefficient of the internal air pressure of the fiber core to the diameter deviation, in μm / Pa; ΔP is the pressure fluctuation inside the fiber core, in Pa; kc is the first-order gain coefficient of traction speed with respect to diameter deviation, in μm / (m / min). ΔC represents the fluctuation of the traction linear velocity, in m / min; kft is the feed rate-tension cross-coupling gain coefficient, in μm / (mm / min*N). kfw is the cross-coupling gain coefficient of feed rate and heating power, in μm / (mm / min*kw). kwt is the heating power-tension cross-coupling gain coefficient, in μm / (kw*N). kf′ is the differential adjustment coefficient of the feed rate change rate, in μm*s / (mm / min). kt′ is the differential adjustment coefficient of the rate of change of tension, in μm*s / N; kw′ is the differential adjustment coefficient of the power change rate, with units of μm*s / kw; KpD is the proportional closed-loop adjustment coefficient for diameter deviation, in units of... ; KiD is the integral closed-loop adjustment coefficient for diameter deviation, in units of... ; KdD is the differential closed-loop adjustment coefficient for diameter deviation, in units of... ; ε represents the environmental random disturbance noise term, in μm.

10. A wire diameter adaptive control device for the hollow fiber drawing process, characterized in that, include: The first execution unit is used to collect fiber diameter data and process parameters at different process positions during the fiber drawing process in real time. The first judgment unit is used to determine the current process mode of the wire drawing process based on the collected wire diameter data and process parameters. The second judgment unit is used to determine the influence weight of each parameter in the current process parameters on the deviation based on the preset coupling sensing model and the deviation between the current wire diameter data and the target wire diameter data. The second execution unit is used to adjust the current process parameters according to the wire diameter control strategy corresponding to the current process mode and the influence weight of each parameter in the current process parameters on the deviation, so as to keep the fiber wire diameter within a preset range. The process parameters include at least two different parameters; There are at least two different wire diameter control strategies under different process modes.

Images