Method and system for controlling fiber uniformity in spandex spinning process

Abstract

The invention discloses a spandex spinning fiber uniformity control method, and relates to the technical field of textile fabric fiber control. Through strict raw material screening, vacuum drying and stirring and mixing pretreatment, process parameters are optimized by combining a response surface with a genetic algorithm, real-time monitoring is performed by means of multiple technologies, feedback adjustment is performed according to a deviation formula, and the fiber uniformity is guaranteed. Through precise raw material pretreatment, intelligent process parameter optimization, real-time comprehensive monitoring and intelligent feedback adjustment, the spandex fiber uniformity can be greatly improved, defective products are reduced, the production efficiency is improved, the energy consumption is reduced, and the market competitiveness of products is enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of spinning fiber control, and particularly to a method and system for controlling fiber uniformity during the spandex spinning process. Background Art

[0002] As a high-elastic fiber, spandex is widely used in the textile field. In the early stage of spandex spinning, the raw material pretreatment only relied on simple screening, making it difficult to accurately remove fine impurities, and the moisture control lacked accuracy, resulting in unstable raw material quality and laying a hidden danger for fiber uniformity. For example, in some small spinning factories, due to the mixing of impurities, the fibers often had problems such as uneven thickness and breakage.

[0003] The traditional spinning process parameters mostly relied on empirical settings and were difficult to fit the complex and changeable raw material characteristics and equipment conditions. Facing different batches of raw materials, using fixed parameters could not take into account the fiber strength, elasticity and uniformity. Taking the spinning temperature as an example, an inappropriate temperature would cause internal structural defects in the fiber and affect the uniformity.

[0004] The monitoring means during the spinning process were limited. The accuracy of ordinary pressure and temperature sensors was insufficient to capture subtle changes in real time. Manually sampling and inspecting the fiber morphology had low efficiency and was difficult to cover comprehensively. In the current situation of high-speed production, traditional monitoring could not detect problems in time, delaying the adjustment opportunity and resulting in a large number of defective products. Therefore, it is urgent to develop an advanced method and system for controlling the fiber uniformity in spandex spinning. Summary of the Invention

[0005] The method and system for controlling fiber uniformity during the spandex spinning process proposed by the present invention are used to solve the problems mentioned in the above prior art.

[0006] To achieve the above object, the present invention adopts the following technical solutions: A method for controlling fiber uniformity during the spandex spinning process, comprising:

[0007] The step of pretreating the spinning raw materials: screening the spandex spinning raw materials, using optical sorting technology, combining multi-spectral imaging analysis to identify and remove impurities, placing the raw materials in a temperature and humidity controlled vacuum drying equipment, adopting a fuzzy control algorithm, and dynamically adjusting the drying temperature and vacuum degree according to the real-time humidity and temperature feedback of the raw materials to remove the moisture in the raw materials and prevent the moisture from causing fiber defects during the spinning process; using a stirring device to stir and mix the dried raw materials, the stirring device is equipped with variable frequency speed regulation and stirring mode switching functions, and at the same time, an inert gas is introduced during the stirring process to make the raw materials evenly mixed;

[0008] Steps for optimizing spinning process parameters: According to the characteristics of the spinning equipment and raw material parameters, the response surface optimization method combined with the genetic algorithm is used to determine the spinning process parameters; taking the spinning temperature, spinning pressure, screw speed, and spinning speed as independent variables and the fiber uniformity as the response value, a mathematical model is constructed; through experimental design, the fiber uniformity data under different parameter combinations are obtained, and the relationship between the fiber uniformity U and the spinning temperature T, spinning pressure P, screw speed n, and spinning speed v is obtained through regression analysis as U = aT 2 + bP 2 + cn 2 + dv 2 + eT + fP + gn + hv + i, where a, b, c, d, e, f, g, h, and i are regression coefficients; at the same time, considering the differences in raw materials of different batches, an adaptive parameter correction mechanism is introduced to fine-tune the process parameters according to the initial characteristics of the raw materials;

[0009] Steps for real-time monitoring of the spinning process: Install a pressure sensor at the spinning nozzle, adopt the fiber Bragg grating sensing technology to monitor the spinning pressure in real time, and perform high-frequency response to the dynamic changes of the pressure; set a temperature sensor in the spinning channel, adopt the distributed fiber optic temperature sensing technology to realize the measurement of the continuous space temperature during the spinning process, obtain the morphological images of the fibers during the spinning process through an online image acquisition device, and use the convolutional neural network CNN algorithm of deep learning for analysis, and then fine-tune in the actual spinning scenario;

[0010] Steps for feedback regulation: The spinning pressure, temperature, and fiber morphology data obtained by real-time monitoring are transmitted to the control system; when the spinning pressure deviates from the set value, the control system adjusts the metering pump speed n according to the pressure deviation ΔP p for pressure adjustment, and the adjustment formula is n p = n p0 + k1ΔP + k 11 ΔP 2 , where n p0 is the initial metering pump speed, k1 is the primary pressure adjustment coefficient, and k 11 is the secondary pressure adjustment coefficient, which is determined through experiments; when the temperature deviates from the set value, the power P of the heating or cooling device is adjusted hc , and the adjustment formula is P hc = P hc0 + k2ΔT + k 22 ΔT 2 , where P hc0 is the initial power, k2 is the primary temperature adjustment coefficient, and k 22 is the secondary temperature adjustment coefficient, which is determined through experiments; if the fiber diameter uniformity is lower than the set standard, it is optimized by adjusting the draw ratio R of the spinning nozzle, and the adjustment formula is R = R0 + k3ΔU + k 33 ΔU2 , where R0 is the initial draw ratio, k3 is the primary uniformity adjustment coefficient, and k 33 is the secondary uniformity adjustment coefficient, which is determined through experiments; meanwhile, a predictive feedback mechanism is introduced to perform fine-tuning in advance according to the prediction of process parameter changes within a future time by a machine learning model.

[0011] Further, it also includes:

[0012] Real-time detection step of raw material characteristics: Install a near-infrared spectroscopy analyzer in the raw material conveying pipeline to detect the chemical composition and molecular weight distribution of the raw materials in real time; adopt spectral fusion technology to combine near-infrared spectroscopy with Raman spectroscopy to obtain the molecular structure information of the raw materials; by establishing a correlation model between raw material characteristics and multi-spectral data, use the support vector machine (SVM) algorithm to analyze the spectral data to obtain the characteristic parameters of the raw materials.

[0013] Spinning environment control step: Install a temperature and humidity control system in the spinning workshop, adopt a distributed temperature and humidity sensor network based on the Internet of Things to achieve real-time monitoring and control of the temperature and humidity in the workshop; by installing air purification equipment and adopting the composite technology of electrostatic adsorption and filtration, remove dust impurities in the workshop air. At the same time, adopt active noise reduction technology, through the principle of anti-phase sound wave cancellation, to reduce the workshop noise and ensure that the fiber uniformity is not affected by environmental fluctuations.

[0014] Further, in the spinning raw material pretreatment step, perform surface modification treatment on the selected raw materials; adopt the composite treatment technology of low-temperature plasma and nano-coating to introduce active groups on the raw material surface while depositing a uniform nano-coating; through surface modification, not only improve the fluidity of the raw materials and their compatibility with additives, but also endow the raw materials with certain antistatic and antibacterial properties.

[0015] Further, in the spinning process parameter optimization step, consider the factors of aging and wear of the spinning equipment; adopt the fusion technology of acoustic emission detection and infrared thermal imaging sensors, and by establishing a fuzzy relationship model between equipment wear and process parameter correction, when the wear of equipment components reaches the threshold, automatically correct the spinning process parameters to maintain the fiber uniformity; at the same time, according to the prediction of the remaining service life of the equipment, arrange the equipment maintenance and replacement plan in advance.

[0016] Further, in the feedback adjustment step, introduce an adaptive control algorithm combined with a reinforcement learning algorithm; the control system automatically adjusts the adjustment coefficients k1, k2, k3, k 11 、k 22 、k 33; The reinforcement learning algorithm interacts with the spinning process and continuously tries different adjustment strategies to optimize the reward value of fiber uniformity; at the same time, a target optimization mechanism is introduced to balance the spinning efficiency and energy consumption while ensuring fiber uniformity.

[0017] Furthermore, in the real-time monitoring step of the spinning process, a machine learning algorithm is used to perform predictive analysis on the monitoring data. The long short-term memory network (LSTM) model based on the attention mechanism is adopted, taking historical monitoring data as input, automatically focusing on the key features and time series information in the data, and predicting the change trends of spinning pressure, temperature, and fiber morphology in the future time.

[0018] A system applying the method for controlling fiber uniformity in the spandex spinning process includes:

[0019] Raw material pretreatment module: Equipped with a raw material screening device, using an optical sorting device combined with a multispectral imaging analysis system to identify and remove impurities in the raw material. The vacuum drying equipment is equipped with temperature and humidity control functions, and a fuzzy control algorithm is used to dynamically adjust the drying parameters according to the real-time state of the raw material; the stirring equipment adopts variable frequency speed regulation and stirring mode switching technology, and an inert gas is introduced during the stirring process. At the same time, it is also equipped with an additive dosing device to add according to the raw material characteristics and process requirements;

[0020] Process parameter optimization module: Built-in response surface optimization software combined with a genetic algorithm solver, constructing a mathematical model of fiber uniformity and spinning process parameters according to the input spinning equipment parameters, raw material parameters, and experimental data; using the genetic algorithm to search for an optimization solution in the solution space to obtain the spinning process parameters and transmit the parameters to the spinning equipment control system; it also has an adaptive parameter correction function to fine-tune the process parameters according to the initial characteristics of the raw material; at the same time, it is connected to the equipment status monitoring module to correct the process parameters in real time according to the equipment wear condition.

[0021] Real-time monitoring module: Includes a pressure sensor, using fiber Bragg grating sensing technology to real-time monitor the spinning pressure and respond to the dynamic changes of the pressure at a high frequency. The temperature sensor uses distributed fiber optic temperature sensing technology to measure the continuous space temperature of the spinning channel. The online image acquisition device uses a camera and a supporting light source, combined with the convolutional neural network (CNN) algorithm of deep learning to detect the fiber diameter change, distribution, and surface defects; it is also equipped with an acoustic emission sensor and an infrared thermal imager to monitor the operating status of the spinning equipment components;

[0022] Feedback regulation module: Composed of a control system, it receives data transmitted by the real-time monitoring module; when the data deviates from the set value, according to the adjustment formula, it adjusts the spinning process by controlling the rotational speed of the metering pump, the power of the heating or cooling device, and the stretching ratio of the spinning nozzle; it adopts an adaptive control algorithm combined with a reinforcement learning algorithm to automatically adjust the adjustment coefficient, and at the same time introduces a target optimization mechanism to balance fiber uniformity, spinning efficiency, and energy consumption; it also has a predictive feedback function to perform fine-tuning in advance according to the prediction results of the machine learning model.

[0023] Raw material property detection module: Install a near-infrared spectroscopy analyzer and a Raman spectrometer in the raw material conveying pipeline, and use spectral fusion technology to obtain the molecular structure information of the raw material; the data processing unit uses the support vector machine (SVM) algorithm to analyze the spectral data, obtains the raw material property parameters, and feeds back the results to the raw material pretreatment module to realize the adjustment of the raw material treatment method and the dosage of additives.

[0024] Environment control module: The temperature and humidity control system adopts an Internet of Things-based distributed temperature and humidity sensor network to realize real-time monitoring and control of the temperature and humidity in the workshop; the air purification equipment uses an electrostatic adsorption and filtration composite technology, and the active noise reduction equipment uses the principle of anti-phase sound wave cancellation to reduce the noise in the workshop; it also has a laminar air supply system to optimize the air flow distribution in the workshop and realize uniform and stable air flow in the spinning area.

[0025] Learning and prediction module: Constructed based on the long short-term memory network (LSTM) model with an attention mechanism, and trained using historical monitoring data; it predicts the future spinning pressure, temperature, and fiber morphology changes, and transmits the prediction results to the feedback regulation module and the process parameter optimization module to adjust the process parameters in advance.

[0026] Furthermore, it also includes:

[0027] Equipment status monitoring module: Through vibration sensors, displacement sensors, acoustic emission sensors, and infrared thermal imagers installed on the spinning equipment components, it monitors the operating status of the equipment in real time; uses sensor fusion technology and fault diagnosis algorithms to judge whether there are potential faults in the equipment according to the sensor data; issues an alarm when detecting equipment abnormalities, and transmits the equipment status data to the process parameter optimization module to realize the adjustment of process parameters; at the same time, the module is equipped with a function for predicting the remaining service life of the equipment to arrange equipment maintenance and replacement plans in advance.

[0028] Human-machine interaction module: Provides a visual operation interface, and uses virtual reality and augmented reality technologies. Operators can input spinning task parameters, view real-time monitoring data, historical data, and the system operating status; at the same time, set system parameters, including adjustment coefficients and alarm thresholds, to realize convenient management and monitoring of the system; it also has a voice interaction function to facilitate operators to operate in the environment.

[0029] Compared with the existing technology, the beneficial effects of the present invention are as follows:

[0030] In terms of improving the fiber quality, through precise control methods and intelligent systems, key parameters in the spinning process can be monitored and adjusted in real time, effectively reducing the unevenness of fiber thickness. The diameter deviation of spandex fibers is controlled within a very small range, significantly improving the uniformity of the fibers. The spandex fibers produced are more stable and consistent in key performance indicators such as strength and elasticity, significantly enhancing the overall quality of spandex products and meeting the stringent requirements of the high-end market for fiber quality.

[0031] From the perspective of production efficiency, the system has the ability of rapid response and automatic regulation, can quickly respond to and correct abnormal conditions in the spinning process, avoid frequent shutdown adjustments caused by uneven fibers, effectively reduce the production interruption time, enable the spandex spinning production line to operate stably for a long time, significantly increase the fiber output per unit time, reduce the production cost, and enhance the competitiveness of the enterprise in the market. In terms of raw material utilization, due to the effective control of fiber uniformity, the waste generated due to fiber quality problems is reduced, the utilization rate of raw materials is increased, resource waste is reduced, which conforms to the concept of sustainable development and saves a large amount of raw material procurement costs for the enterprise.

[0032] In addition, the control method and system have good compatibility and scalability, can be effectively integrated with existing spandex spinning equipment, without large-scale equipment replacement, reducing the cost and difficulty of enterprise technology upgrading. At the same time, it can be functionally extended and optimized according to the production needs and technological development of the enterprise, providing strong support for the technological progress and industrial upgrading of the spandex spinning industry, and promoting the entire industry to develop in the direction of high quality and high efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a schematic block diagram of the fiber uniformity control system in the spandex spinning process proposed by the present invention;

[0034] Figure 2 It is a schematic block diagram of the fiber uniformity control method in the spandex spinning process proposed by the present invention;

[0035] Figure 3 It is a schematic block diagram of the fiber uniformity comparison of the fiber uniformity control method in the spandex spinning process proposed by the present invention;

[0036] Figure 4 It is a schematic block diagram of the spinning efficiency comparison of the fiber uniformity control method in the spandex spinning process proposed by the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0037] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0038] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the present invention.

[0039] In addition, the terms "first" and "second" are only used for descriptive purposes and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present invention, "a plurality" means two or more unless otherwise specifically defined. In addition, the terms "installation", "connection", and "connection" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. The present invention will be further described in detail below with reference to the accompanying drawings.

[0040] Refer to Figures 1 - 4 : Specific implementation of a method and system for controlling fiber uniformity during spandex spinning

[0041] I. System overall framework

[0042] The method and system for controlling fiber uniformity during spandex spinning cover multiple key modules such as raw material pretreatment, process parameter optimization, real-time monitoring, feedback regulation, raw material property detection, environmental control, machine learning prediction, equipment status monitoring, and human-computer interaction. Each module works closely together, starting from the raw material preparation stage, to precisely control the entire spinning process and ensure that the uniformity of spandex fibers reaches the optimal level.

[0043] II. Detailed implementation steps of each module

[0044] (1) Spinning raw material pretreatment module

[0045] 1. Raw material screening: A high-precision optical sorting device is paired with a multispectral imaging analysis system. Based on the differences in light absorption and reflection of substances, impurities are identified by a multispectral camera and removed by high-speed air blowing with a precision reaching the micron level. Additionally, a magnetic separator is set up to remove metal particles.

[0046] 2. Vacuum drying: The vacuum drying equipment has an intelligent temperature and humidity control system. With the help of temperature and humidity sensors and a fuzzy control algorithm, the heating power and pumping rate are adjusted in real time according to the humidity of the raw materials, maintaining a temperature of 60 - 80°C and a pressure of 10 - 20 Pa for 4 - 6 hours of drying.

[0047] 3. Stirring and mixing: The high-speed stirring equipment is equipped with a variable-frequency motor. The rotation speed is adjusted to 800 - 1200 revolutions per minute according to the properties of the raw materials. Special impellers are used to enhance the effect. Argon is introduced at a rate of 0.1 - 0.5 L / min during stirring to prevent oxidation. Stirring is completed in 15 - 25 minutes. Additives are added as required by a high-precision metering pump.

[0048] (2) Spinning process parameter optimization module

[0049] 1. Model construction and solution: The module is built-in with response surface optimization software and a genetic algorithm solver. Operators input the parameters of the spinning equipment and raw materials. Through multiple groups of experiments, the fiber uniformity data under different parameter combinations are obtained, and a model of U = aT 2 +bP 2 +cn 2 +dv 2 +eT + fP + gn + hv + i is constructed, and the genetic algorithm searches for the optimal parameters.

[0050] 2. Adaptive parameter correction: It is connected to the raw material property detection and equipment status monitoring modules. When the raw material properties change, the parameters are adjusted according to the associated model. When the key components of the equipment are worn, they are adjusted according to the fuzzy relationship model. For example, when the nozzle aperture increases, the pressure is reduced, and the rotation speed and speed combination are adjusted.

[0051] (3) Real-time monitoring module for the spinning process

[0052] 1. Pressure monitoring: A pressure sensor using fiber Bragg grating sensing technology is installed at the spinning nozzle. The optical signal is transmitted based on the characteristic that the wavelength of the fiber grating changes with pressure, with a precision of ±0.005 MPa and a sampling frequency exceeding 100 Hz. The fiber has strong anti-interference ability.

[0053] 2. Temperature monitoring: A sensor using distributed fiber optic temperature sensing technology is used inside the spinning channel. The continuous space temperature is measured based on the characteristics of the backward scattered light, with a precision of ±0.1°C.

[0054] 3. Fiber morphology monitoring: A high-resolution camera is equipped with a special lens and light source to capture the fiber morphology at a frame rate of 50 - 100 fps. A pre-trained CNN algorithm is used to identify the diameter, distribution, and microscopic defects.

[0055] 4. Equipment status monitoring: Acoustic emission sensors are installed to determine equipment failures based on signal characteristics, and an infrared thermal imager is used to monitor equipment overheating warnings.

[0056] (IV) Feedback regulation module

[0057] 1. Parameter adjustment: The control system adjusts according to real-time monitoring data. When there is a pressure deviation ΔP, the metering pump speed is adjusted according to n p = n p0 + k1ΔP + k 11 ΔP 2 When there is a temperature deviation ΔT, the heating and cooling power is adjusted according to P hc = P hc0 + k2ΔT + k 22 ΔT 2 When there is a uniformity deviation ΔU, the nozzle draw ratio is adjusted according to R = R0 + k3ΔU + k 33 ΔU 2 The coefficients are determined through a large number of experiments.

[0058] 2. Application of intelligent algorithms: The combination of adaptive and reinforcement learning algorithms. The adaptive algorithm adjusts the coefficients, and the reinforcement learning optimizes the strategy according to the reward value, taking into account fiber uniformity, spinning efficiency, and energy consumption, and setting weights for comprehensive optimization.

[0059] 3. Predictive feedback: Fine-tune in advance according to the prediction results of machine learning. For example, if the pressure is predicted to rise, the metering pump speed is reduced in advance.

[0060] (V) Raw material property detection module

[0061] 1. Spectral detection: Near-infrared and Raman spectrometers are installed in the raw material delivery pipe. The former measures the vibration of hydrogen-containing groups, and the latter measures the molecular skeleton. Spectral fusion technology is used to integrate the data.

[0062] 2. Data processing and feedback: The data processing unit uses the SVM algorithm, collects the spectra and characteristic parameters of multiple batches of raw materials to build a training set, trains the correlation model, and after real-time detection, feeds back to the raw material pretreatment module to adjust the processing method and additive dosage.

[0063] (VI) Environment control module

[0064] 1. Temperature and humidity control: An IoT distributed temperature and humidity sensor network is used. The sensors transmit data wirelessly, and the central system controls air conditioners, humidifiers, and dehumidifiers to control the temperature at 22 - 25°C and the humidity at 50 - 60% RH, with an accuracy of ±0.5°C and ±1% RH.

[0065] 2. Air purification: The air purification equipment uses the composite technology of electrostatic adsorption and filtration. Electrostatic force makes dust charged and adsorbed, and multiple layers of filtration remove tiny particles, with a purification rate of 99.9%.

[0066] 3. Noise reduction and air flow optimization: The active noise reduction equipment emits anti-phase sound waves to reduce the noise to below 65 dB(A). The laminar air supply system sets up proper air supply and return air outlets to form a uniform laminar flow.

[0067] (VII) Machine learning prediction module

[0068] 1. Model construction and training: Use the LSTM-Attention model. Collect historical monitoring data and divide it into training and test sets. When training, the model learns the rules through the attention mechanism and adjusts the parameters.

[0069] 2. Prediction and decision-making: Input real-time data to predict the changes in the next 10 - 30 minutes. The results are transmitted to the feedback adjustment and process parameter optimization module, and comprehensive decisions are made in combination with the expert knowledge base.

[0070] (VIII) Equipment status monitoring module

[0071] 1. Multi-sensor monitoring: Install vibration, displacement, acoustic emission, infrared thermal imaging and other multi-sensors on the key components of the spinning equipment. The vibration and displacement sensors measure the operating status of the components, and the acoustic emission and infrared thermal imaging monitor potential faults.

[0072] 2. Fault diagnosis and early warning: Use the multi-sensor fusion technology and fault diagnosis algorithm to integrate and analyze the data, judge potential faults. When abnormal, an alarm is issued, and the equipment status data is transmitted to the process parameter optimization module to adjust the parameters, and the remaining service life is predicted to arrange maintenance and replacement.

[0073] (IX) Human-machine interaction module

[0074] 1. Visual operation interface: Provide a visual interface using VR and AR technologies. Operators can immerse themselves in inputting task parameters, viewing real-time and historical data, and the system status.

[0075] 2. Intelligent voice interaction: Have the function of intelligent voice interaction, which is convenient for operation in special environments. System parameters such as adjustment coefficients and alarm thresholds can also be set.

[0076] III. Data representation of beneficial effects

[0077] Evaluation index Traditional spinning system Spinning system of this patent Improvement ratio Fiber uniformity (CV value) 8％ 3％ Reduced by 62.5% Defect rate 15％ 5％ Reduced by 66.7% Spinning efficiency (output per day) 10 tons 13 tons Increased by 30% Energy consumption (kWh / ton of fiber) 300 250 Reduced by 16.7%

[0078] It can be seen from the data that the patent system significantly improves the uniformity of spandex spinning fibers, reduces the defective rate, improves efficiency, and reduces energy consumption, with obvious advantages.

[0079] The above are only the preferred specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention should cover within the protection scope of the present invention according to the technical solution and inventive concept of the present invention by means of equivalent replacement or modification.

Claims

1. A method for controlling fiber uniformity during the spandex spinning process, characterized in that, Including: Spinning raw material pretreatment step: Screen the spandex spinning raw materials, use optical sorting technology, combine multi-spectral imaging analysis to identify and remove impurities, place the raw materials in a temperature and humidity controlled vacuum drying equipment, adopt a fuzzy control algorithm, and dynamically adjust the drying temperature and vacuum degree according to the real-time humidity and temperature feedback of the raw materials to remove the moisture in the raw materials and prevent fiber defects caused by moisture during the spinning process; Use a stirring device to stir and mix the dried raw materials. The stirring device is equipped with variable frequency speed regulation and stirring mode switching functions, and at the same time, an inert gas is introduced during the stirring process to make the raw materials evenly mixed; Steps for optimizing spinning process parameters: Based on the characteristics of the spinning equipment and raw material parameters, use the response surface optimization method combined with the genetic algorithm to determine the spinning process parameters; construct a mathematical model with spinning temperature, spinning pressure, screw speed, and spinning speed as independent variables and fiber uniformity as the response value; through experimental design, obtain fiber uniformity data under different parameter combinations, and obtain the relationship between fiber uniformity U and spinning temperature T, spinning pressure P, screw speed n, and spinning speed v through regression analysis: U = aT 2 + bP 2 + cn 2 + dv 2 + eT + fP + gn + hv + i, where a, b, c, d, e, f, g, h, and i are regression coefficients; at the same time, considering the differences in raw materials of different batches, introduce an adaptive parameter correction mechanism to fine-tune the process parameters according to the initial characteristics of the raw materials; Real-time monitoring step of the spinning process: Install a pressure sensor at the spinning nozzle, use fiber Bragg grating sensing technology to monitor the spinning pressure in real time and perform high-frequency response to the dynamic changes of the pressure; Set temperature sensors in the spinning duct, use distributed fiber optic temperature sensing technology to achieve the measurement of the continuous space temperature during the spinning process, obtain the morphological images of the fibers during the spinning process through an online image acquisition device, and use the convolutional neural network CNN algorithm of deep learning for analysis, and then fine-tune in the actual spinning scenario.

2. The method for controlling fiber uniformity during spandex spinning according to claim 1, characterized in that, Also including: Feedback adjustment step: Transmit the spinning pressure, temperature and fiber morphology data obtained by real-time monitoring to the control system; When the spinning pressure deviates from the set value, the control system adjusts the pressure according to the pressure deviation ΔP by regulating the rotational speed n of the metering pump p The pressure adjustment is carried out, and the adjustment formula is n p = n p0 + k1ΔP + k 11 ΔP 2 , where n p0 is the initial rotational speed of the metering pump, k1 is the primary pressure adjustment coefficient, and k 11 is the secondary pressure adjustment coefficient, which is determined through experiments; when the temperature deviates from the set value, the power P of the heating or cooling device is adjusted, and the adjustment formula is P hc = P hc + k2ΔTk hc0 ΔT 22 2 , where P hc0 is the initial power, k2 is the primary temperature adjustment coefficient, and k 22 is the secondary temperature adjustment coefficient, which is determined through experiments; if the fiber diameter uniformity is lower than the set standard, it is optimized by adjusting the draw ratio R of the spinning nozzle, and the adjustment formula is R = R0 + k3ΔU + k 33 2 ΔU 33 , where R0 is the initial draw ratio, k3 is the primary uniformity adjustment coefficient, and k 33 is the secondary uniformity adjustment coefficient, which is determined through experiments; meanwhile, a predictive feedback mechanism is introduced to make fine adjustments in advance according to the prediction of the changes in process parameters within the future time by the machine learning model; Real-time detection step of raw material characteristics: Install a near-infrared spectroscopy analyzer in the raw material conveying pipeline to detect the chemical composition and molecular weight distribution of the raw materials in real time; Adopt spectral fusion technology to combine near-infrared spectroscopy and Raman spectroscopy to obtain the molecular structure information of the raw materials; By establishing a correlation model between raw material characteristics and multi-spectral data, use the support vector machine SVM algorithm to analyze the spectral data to obtain the characteristic parameters of the raw materials.

3. The method for controlling fiber uniformity during spandex spinning according to claim 1, characterized in that Also including: Spinning environment control step: Install a temperature and humidity control system in the spinning workshop, use an Internet of Things-based distributed temperature and humidity sensor network to achieve real-time monitoring and control of the temperature and humidity in the workshop; By installing air purification equipment, adopt electrostatic adsorption and filtration composite technology to remove dust impurities in the workshop air. At the same time, adopt active noise reduction technology, through the principle of anti-phase sound wave cancellation, to reduce the workshop noise and ensure that the fiber uniformity is not affected by environmental fluctuations.

4. The method for controlling fiber uniformity during spandex spinning according to claim 1, wherein, In the spinning raw material pretreatment step, perform surface modification treatment on the screened raw materials; Adopt a low-temperature plasma and nano-coating composite treatment technology to introduce active groups on the surface of the raw materials while depositing a uniform nano-coating; Through surface modification, not only improve the fluidity of the raw materials and the compatibility with additives, but also endow the raw materials with certain antistatic and antibacterial properties.

5. The method for controlling fiber uniformity during spandex spinning according to claim 1, characterized in that, In the spinning process parameter optimization step, consider the factors of aging and wear of the spinning equipment; Adopt a fusion technology based on acoustic emission detection and infrared thermal imaging sensors. By establishing a fuzzy relationship model between equipment wear and process parameter correction, when the wear of equipment components reaches the threshold, automatically correct the spinning process parameters to maintain the fiber uniformity; At the same time, according to the prediction of the remaining service life of the equipment, arrange the equipment maintenance and replacement plan in advance.

6. The method for controlling fiber uniformity during spandex spinning according to claim 2, characterized in that, In the feedback regulation step, an adaptive control algorithm is introduced in combination with a reinforcement learning algorithm; the control system automatically adjusts the regulation coefficients k1, k2, k3, k 11 , k 22 , k 33 ; the reinforcement learning algorithm interacts with the spinning process and continuously tries different regulation strategies to optimize the reward value of fiber uniformity; at the same time, a target optimization mechanism is introduced to take into account the spinning efficiency and energy consumption while ensuring fiber uniformity.

7. The method for controlling the fiber uniformity during the spandex spinning process according to claim 1, characterized in that, In the real-time monitoring step of the spinning process, machine learning algorithms are used to perform predictive analysis on the monitoring data. The long short-term memory network (LSTM) model based on the attention mechanism is adopted. Taking historical monitoring data as input, it automatically focuses on the key features and time series information in the data to predict the changing trends of spinning pressure, temperature, and fiber morphology in the future.

8. A system for implementing the method for controlling fiber uniformity during the spandex spinning process according to any one of claims 1-7, characterized in that, It includes: Raw material pretreatment module: Equipped with a raw material screening device, an optical sorting device combined with a multispectral imaging analysis system is used to identify and remove impurities in the raw materials. The vacuum drying equipment is equipped with temperature and humidity control functions, and a fuzzy control algorithm is used to dynamically adjust the drying parameters according to the real-time state of the raw materials; the stirring equipment adopts variable frequency speed regulation and stirring mode switching technology, and an inert gas is introduced during the stirring process. At the same time, it is also equipped with an additive dosing device to add according to the raw material characteristics and process requirements; Process parameter optimization module: Built-in response surface optimization software combined with a genetic algorithm solver. According to the input spinning equipment parameters, raw material parameters, and experimental data, a mathematical model of fiber uniformity and spinning process parameters is constructed; the genetic algorithm is used to search for optimization solutions in the solution space to obtain spinning process parameters, and the parameters are transmitted to the spinning equipment control system; it also has an adaptive parameter correction function to fine-tune the process parameters according to the initial characteristics of the raw materials; at the same time, it is connected to the equipment status monitoring module to correct the process parameters in real time according to the equipment wear condition. Real-time monitoring module: It includes a pressure sensor, which uses fiber Bragg grating sensing technology to monitor the spinning pressure in real time and respond to the dynamic changes of the pressure with high frequency. The temperature sensor uses distributed fiber optic temperature sensing technology to measure the continuous spatial temperature of the spinning channel. The online image acquisition device uses a camera and a supporting light source, and combines the convolutional neural network (CNN) algorithm of deep learning to detect the fiber diameter changes, distribution, and surface defects; it is also equipped with an acoustic emission sensor and an infrared thermal imager to monitor the operating status of the spinning equipment components; Feedback adjustment module: Composed of a control system, it receives the data transmitted by the real-time monitoring module; when the data deviates from the set value, according to the adjustment formula, the spinning process is adjusted by controlling the rotational speed of the metering pump, the power of the heating or cooling device, and the stretching ratio of the spinning nozzle; an adaptive control algorithm combined with a reinforcement learning algorithm is used to automatically adjust the adjustment coefficient, and at the same time, a target optimization mechanism is introduced to take into account fiber uniformity, spinning efficiency, and energy consumption; it also has a predictive feedback function to perform fine-tuning in advance according to the prediction results of the machine learning model; Raw material property detection module: Near-infrared spectrometers and Raman spectrometers are installed in the raw material conveying pipeline, and spectral fusion technology is used to obtain the molecular structure information of the raw materials; the data processing unit uses the support vector machine (SVM) algorithm to analyze the spectral data, obtain the raw material property parameters, and feedback the results to the raw material pretreatment module to realize the adjustment of the raw material processing method and the additive dosing amount; Environmental control module: The temperature and humidity control system adopts a distributed temperature and humidity sensor network based on the Internet of Things to realize real-time monitoring and control of the temperature and humidity in the workshop; the air purification equipment adopts the composite technology of electrostatic adsorption and filtration, and the active noise reduction equipment adopts the principle of anti-phase sound wave cancellation to reduce the workshop noise; it also has a laminar air supply system to optimize the air flow distribution in the workshop and achieve uniform and stable air flow in the spinning area. Learning and prediction module: It is constructed based on the long short-term memory network (LSTM) model with an attention mechanism and trained using historical monitoring data; it predicts the future spinning pressure, temperature and fiber morphology changes, and transmits the prediction results to the feedback adjustment module and the process parameter optimization module to adjust the process parameters in advance.

9. The fiber uniformity control system in the spandex spinning process according to claim 8, characterized in that, It also includes: Equipment status monitoring module: Through vibration sensors, displacement sensors, acoustic emission sensors and infrared thermal imagers installed on the spinning equipment components, it monitors the running status of the equipment in real time; uses sensor fusion technology and fault diagnosis algorithms to judge whether there are potential equipment faults based on sensor data; issues an alarm when abnormal equipment is detected and transmits the equipment status data to the process parameter optimization module to adjust the process parameters; at the same time, the module is equipped with a function to predict the remaining service life of the equipment to arrange equipment maintenance and replacement plans in advance.

10. The fiber uniformity control system during the spandex spinning process according to claim 8, characterized in that, It also includes: Human-machine interaction module: It provides a visual operation interface and adopts virtual reality and augmented reality technologies. The operator inputs spinning task parameters, views real-time monitoring data, historical data and the system running status; at the same time, it sets system parameters, including adjustment coefficients and alarm thresholds, to achieve convenient management and monitoring of the system; it also has a voice interaction function to facilitate the operator's operation in the environment.

Images