A polyester staple fiber production line process operation parameter optimization control system

Abstract

The application discloses a polyester staple fiber production line process operation parameter optimization control system, which comprises the following steps: constructing a digital twin model corresponding to a physical entity of a polyester staple fiber production line; performing single-factor variable process tests and multi-factor coupling process tests based on the digital twin model, and extracting process linkage rules under different test scene conditions; constructing a multi-variety process knowledge base based on the process linkage rules under different test scene conditions, and configuring an initial parameter combination for the polyester staple fiber production line in an initialization stage or a variety switching stage based on the multi-variety process knowledge base; generating simulation operation data of the polyester staple fiber production line in a production operation stage in real time based on the digital twin model, monitoring the simulation operation data in real time, performing parameter combination updating operation or feedforward disturbance determination according to the real-time monitoring result, and significantly improving product quality stability and production efficiency and reducing overall process energy consumption cost.

Description

Technical Field

[0001] This invention relates to the field of automatic control technology for polyester staple fiber production, specifically a process operation parameter optimization and control system for a polyester staple fiber production line. Background Technology

[0002] Polyester staple fiber is a core raw material for industries such as textiles and nonwovens. Its production process is characterized by a long process flow, strong process linkage, and high parameter coupling. The core processes include melt conveying, spinning, cooling, stretching, heat setting, cutting, and packaging. The operating parameters of each process (such as melt temperature, spinning speed, cooling air temperature and speed, stretching ratio, heat setting temperature, etc.) affect each other and are easily affected by external disturbances such as raw material characteristics (melt viscosity, moisture content), ambient temperature and humidity, and equipment conditions (metering pump wear, stretching roller aging).

[0003] Existing polyester staple fiber production line operating parameter control systems mostly adopt traditional setpoint control modes, which have the following technical defects: Each process uses independent control units, forming information silos and lacking a global coordination mechanism. Optimization of local parameters in a single process can easily lead to suboptimal results throughout the entire process. For example, increasing the temperature in the spinning process to achieve filament uniformity may cause a surge in energy consumption in the subsequent heat setting process. Furthermore, asynchronous parameter adjustments between upstream and downstream processes can easily lead to fiber breakage, quality fluctuations, and other problems. Moreover, relying on manual parameter adjustments based on operator experience makes it difficult to quickly find the optimal combination for multi-parameter coupled scenarios. Additionally, the optimization objective is singular, focusing only on product quality indicators and ignoring energy costs, resulting in low production efficiency and high operating costs. The use of a single feedback control strategy only allows for passive adjustments after disturbances affect production, failing to anticipate disturbances such as raw material fluctuations and environmental changes. Furthermore, the lack of a tiered handling mechanism for different levels and types of disturbances easily leads to poor product quality stability.

[0004] To address the aforementioned issues, there is an urgent need for a process operation parameter optimization and control system for polyester staple fiber production lines that features global coordination, precise perception, intelligent optimization, anti-disturbance and adaptive capabilities, and controllable energy consumption. This system would break through traditional technological bottlenecks and achieve a synergistic improvement in production quality, efficiency, and energy consumption. Summary of the Invention

[0005] The purpose of this invention is to provide a process operation parameter optimization and control system for a polyester staple fiber production line, including a monitoring center, which is communicatively connected to a data acquisition module, a model building module, a correlation analysis module, a parameter configuration module, and a parameter update module. The data acquisition module is used to acquire process information of the polyester staple fiber production line, divide the process into sub-sequences according to the process information, set up operation data monitoring nodes, and acquire operation data at each point; The model building module is used to build digital twin models of the physical entities of a polyester staple fiber production line; The correlation analysis module is used to conduct single-factor variable process experiments based on digital twin models, obtain the single-factor coupling coefficient between single-factor variables and quality indicators, and screen out the principal factors from the single-factor variables; based on the single-factor coupling coefficient of the principal factors, it conducts multi-factor coupled process experiments and extracts the process linkage rules under different test scenarios. The parameter configuration module is used to build a multi-variety process knowledge base based on the process linkage rules under different test scenarios, and to configure the initial parameter combination for the polyester staple fiber production line in the initialization stage or the variety switching stage based on the multi-variety process knowledge base. The parameter update module is used to generate simulated operation data of the polyester staple fiber production line in real time based on the digital twin model, monitor the simulated operation data in real time, and perform parameter combination update operations or feedforward disturbance judgment based on the real-time monitoring results.

[0006] Furthermore, the process of acquiring operational data at each location includes: Obtain the process characteristics of the current polyester staple fiber production equipment, break down the polyester staple fiber production line process into several process sub-sequences according to the process characteristics, and set up operation data monitoring nodes in each process sub-sequence. Based on the functional characteristics of the process flow characteristics of the corresponding process subsequence, the production monitoring indicators of each operation data monitoring node are obtained by data retrieval. The operation data monitoring node obtains operation data in real time according to the production monitoring indicators, marks the monitoring time, and sets the monitoring cycle.

[0007] Furthermore, the process of constructing a digital twin model corresponding to the physical entity of the polyester staple fiber production line includes: The process involves acquiring the physical entities of polyester staple fiber production equipment in the physical space of the current polyester staple fiber production line, constructing a three-dimensional digital space, mapping the physical entities of the polyester staple fiber production equipment into the three-dimensional digital space to generate a three-dimensional model, setting up API interfaces on the three-dimensional model, and connecting the assembly and connection relationships between the physical entities in the physical space to the three-dimensional model in the three-dimensional digital space. It also involves acquiring the three-dimensional models corresponding to each polyester staple fiber production equipment in each process subsequence, as well as the operational data of the operational data monitoring nodes, and mapping the operational data of each process subsequence to the corresponding three-dimensional model of the polyester staple fiber production equipment to generate a digital twin model.

[0008] Furthermore, the process of conducting single-factor variable process experiments based on digital twin models, obtaining the single-factor coupling coefficient between single-factor variables and quality indicators, and screening out principal factors from single-factor variables includes: Based on the production monitoring indicators of each operational data monitoring node, experimental variables, fixed indicators, and quality indicators are set, and information on different processing scenarios is obtained based on the target product specifications and historical operational data of the polyester staple fiber production line in several historical monitoring cycles. One processing scenario is randomly selected from different processing scenario information as the experimental scenario. One experimental variable is randomly selected as the single-factor variable. Other experimental variables are converted into normal values ​​under the experimental scenario. The gradient of the single-factor variable is set. The gradient of the single-factor variable and the normal value are input into the digital twin model to conduct multi-gradient experiments. The single-factor coupling coefficient between the single-factor variable and the quality indicator is obtained. The above process is repeated to obtain the single-factor coupling coefficient between different single-factor variables and the quality indicator under different experimental scenario conditions. The coupling coefficient threshold is preset. The single-factor variables with a single-factor coupling coefficient with the quality indicator that is greater than the coupling coefficient threshold are marked as principal factors.

[0009] Furthermore, the process of extracting process linkage rules under different experimental scenarios by conducting multi-factor coupled process experiments based on the single-factor coupling coefficients of the principal factors includes: Define the coupling gradient combination of the principal factors under the current experimental scenario conditions, convert the experimental variables other than the principal factors into regular values ​​under the current experimental scenario, input the coupling gradient combination of the principal factors and the regular values ​​into the digital twin model to conduct multi-gradient experiments, obtain the coupling effect coefficient between the principal factors and the linkage coupling coefficient between the principal factors and the quality indicators, extract the process linkage rules structured from the linkage coupling coefficient between the principal factors and the quality indicators, obtain the process linkage rules under the current experimental scenario conditions, repeat the above process to obtain the process linkage rules under different experimental scenario conditions.

[0010] Furthermore, a multi-product process knowledge base is constructed based on the process linkage rules under different experimental scenarios. The process of configuring initial parameter combinations for polyester staple fiber production lines in the initialization or product switching stages based on the multi-product process knowledge base includes: Based on the process linkage rules of the polyester staple fiber production line under different test scenarios, a multi-product process knowledge base is constructed. The multi-product process knowledge base is linked with the digital twin model through an API interface to determine the current operating stage of the polyester staple fiber production line. When the polyester staple fiber production line is in the initialization stage, the target product specifications and environmental parameters in the operating data of the current polyester staple fiber production line are input into the multi-product process knowledge base for matching to obtain the initial parameter combination of the current polyester staple fiber production line under the current target product specifications and environmental parameter conditions. The initial parameter combination includes the core process parameter range of each process subsequence and the process linkage rules between each process subsequence.

[0011] When the polyester staple fiber production line is in the product switching stage, the target product specifications after the switch and the environmental parameters in the operating data are input into the multi-product process knowledge base for matching to obtain the initial parameter combination of the current polyester staple fiber production line.

[0012] Furthermore, the process of generating simulated operating data of the polyester staple fiber production line in real time based on the digital twin model, monitoring the simulated operating data in real time, and performing parameter combination update operations or feedforward disturbance determination based on the real-time monitoring results includes: When the polyester staple fiber production line is in the production operation phase, the operation data collected by each operation data monitoring node is input into the digital twin model. Based on the output of the digital twin model, the simulated operation data of each process subsequence within the current monitoring cycle is obtained, and the threshold range of the production monitoring indicators of each process subsequence is obtained. The numerical time series of each production monitoring indicator in the simulated operation data is compared with the corresponding threshold range to obtain the cumulative time of each production monitoring indicator that is not within the corresponding threshold range. A preset time error threshold is set. If the cumulative time corresponding to a production monitoring indicator is greater than the time error threshold, a process abnormality warning is generated and parameter combination update operation is performed. If the cumulative time corresponding to each production monitoring indicator is less than or equal to the time error threshold, normal process information is generated and feedforward disturbance judgment is performed.

[0013] Furthermore, the process of performing parameter combination update operations includes: When the polyester staple fiber production line is in the production operation stage and generates a process abnormality warning, constraints are set according to the threshold range of the production monitoring indicators of each process sub-sequence. Several parameter combinations are randomly generated under the current target product specifications and environmental parameters. Chromosome encoding and initialization of the parameter combinations are performed to generate an initial population. An adaptive product quality objective function and an energy objective function are constructed. A fitness function is constructed based on the adaptive product quality objective function and the energy objective function. Based on the initial population, fitness function and constraints, experiments are conducted through a digital twin model to obtain the optimal parameter combination. The initial parameter combination is then updated based on the optimal parameter combination.

[0014] Furthermore, the process of constructing an adaptive product quality objective function includes: Statistical analysis is performed on the single-factor coupling coefficients between each single-factor variable and quality indicator under the current target product specifications and environmental parameters to obtain the coupling sensitivity coefficient of each quality indicator. The specification requirements of the current target product are obtained. Based on the coupling sensitivity coefficients and specification requirements, the process weights of each quality indicator are set. An adaptive product quality objective function is constructed based on each quality indicator and its process weight.

[0015] Furthermore, the process of determining the feedforward disturbance includes: Extract the linkage coupling coefficients between the main factors and quality indicators from the simulated operation data of each process subsequence within the current monitoring period. Extract the linkage coupling coefficients using process linkage rules in a structured manner to generate process linkage rules. Perform consistency verification between the process linkage rules and the process linkage rules in the initial parameter combination of the current polyester staple fiber production line. If the consistency verification fails, perform a parameter combination update operation.

[0016] Compared with the prior art, the beneficial effects of the present invention are: 1. Based on the digital twin model, standardized single-factor variable process experiments are conducted to accurately obtain the single-factor coupling coefficients and screen out the main factors that play a key role in production quality. Then, multi-factor coupling process experiments are carried out on the main factors to deeply explore the coupling effect between main factors and the linkage coupling relationship between main factors and quality indicators. In this way, the process linkage rules are extracted in a structured manner, so that the parameter linkage between each process of the production line has a clear process basis. This realizes the transformation from "experience-based parameter linkage" to "scientific parameter linkage", and makes the parameter coordination between processes more in line with the production process rules.

[0017] 2. By incorporating the process linkage rules under different test scenarios into a multi-product process knowledge base and realizing the linkage between the knowledge base and the digital twin model, the corresponding initial parameter combination can be accurately matched from the knowledge base according to the target product specifications and real-time environmental parameters during production line initialization or product switching. This parameter combination includes the core process parameter range of each process subsequence and is equipped with corresponding process linkage rules, making the initial parameter configuration more in line with the actual needs of current production. This eliminates the need for repeated manual debugging, enabling rapid and accurate adaptation of multi-product production and improving the production flexibility and product switching efficiency of the production line.

[0018] 3. By statistically analyzing the single-factor coupling coefficient, the coupling sensitivity coefficient of quality indicators is obtained. Differentiated process weights for quality indicators are then set based on the target product specifications, thereby constructing an adaptive product quality objective function. This function aligns the focus of quality optimization with the actual specifications of the product and the controllability of quality indicators under current production conditions. It ensures that core product quality indicators meet standards while also achieving balanced optimization of various quality indicators, making the direction of product quality optimization more aligned with actual market and production needs. This ensures the stability and adaptability of product quality from a process design perspective. Attached Figure Description

[0019] Figure 1 This is a flowchart of a process operation parameter optimization control system for a polyester staple fiber production line according to an embodiment of this application. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. 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.

[0021] like Figure 1 As shown, a process operation parameter optimization and control system for a polyester staple fiber production line includes a monitoring center, which is communicatively connected to a data acquisition module, a model building module, a correlation analysis module, a parameter configuration module, and a parameter update module. The data acquisition module is used to acquire process information of the polyester staple fiber production line, divide the process into sub-sequences according to the process information, set up operation data monitoring nodes, and acquire operation data at each point; The model building module is used to build digital twin models of the physical entities of a polyester staple fiber production line; The correlation analysis module is used to conduct single-factor variable process experiments based on digital twin models, obtain the single-factor coupling coefficient between single-factor variables and quality indicators, and screen out the principal factors from the single-factor variables; based on the single-factor coupling coefficient of the principal factors, it conducts multi-factor coupled process experiments and extracts the process linkage rules under different test scenarios. The parameter configuration module is used to build a multi-variety process knowledge base based on the process linkage rules under different test scenarios, and to configure the initial parameter combination for the polyester staple fiber production line in the initialization stage or the variety switching stage based on the multi-variety process knowledge base. The parameter update module is used to generate simulated operation data of the polyester staple fiber production line in real time based on the digital twin model, monitor the simulated operation data in real time, and perform parameter combination update operations or feedforward disturbance judgment based on the real-time monitoring results.

[0022] It should be further explained that, in the specific implementation process, the process of obtaining operational data at each location includes: Obtain the process characteristics of the current polyester staple fiber production equipment, break down the polyester staple fiber production line process into several process sub-sequences according to the process characteristics, and set up operation data monitoring nodes in each process sub-sequence. Based on the functional characteristics of the process flow characteristics of the corresponding process subsequence, data retrieval is used to obtain the production monitoring indicators (including raw material characteristic indicators such as melt viscosity, moisture content, and end carboxyl group content, environmental parameters (temperature, humidity, dust, air pressure, etc.), product intermediate parameters (using non-contact detection technologies such as machine vision and laser diameter measurement to detect intermediate indicators such as the diameter uniformity of the spinning filament, the surface morphology of the fiber after stretching, and the crystallinity of the fiber after heat setting), and equipment operating parameters (temperature, pressure, speed, and other conventional parameters)) of each operation data monitoring node. The operation data monitoring node obtains the operation data in real time according to the production monitoring indicators, marks the monitoring time, and sets the monitoring cycle.

[0023] It should be further explained that, in the specific implementation process, the operational data monitoring nodes include: Raw material characteristic sensing unit: such as deploying online viscosity sensors, moisture content detectors, and end carboxyl content analyzers on melt conveying pipelines to collect key indicators such as melt viscosity, moisture content, and end carboxyl content in real time. The sampling frequency is 1 time / minute, and the data accuracy is ±0.01dL / g (viscosity) and ±0.1% (moisture content). The data is then transmitted to the data fusion module in real time.

[0024] Environmental disturbance sensing unit: Temperature and humidity sensors, air pressure sensors, and dust sensors are deployed in the spinning workshop, cooling channel, and heat setting area to cover key areas of the entire process. They collect environmental temperature and humidity (accuracy ±0.1℃, ±1%RH), air pressure (accuracy ±0.01MPa), and dust concentration (accuracy ±0.1mg / m³) in real time to identify the impact of environmental disturbances on production parameters.

[0025] Equipment Condition Sensing Unit: Vibration sensors, noise sensors, temperature sensors, and current sensors are deployed on core equipment such as metering pumps, stretching rollers, spinning assemblies, and cutting machines to monitor equipment vibration amplitude (accuracy ±0.01mm / s), noise decibels (accuracy ±1dB), bearing temperature (accuracy ±0.1℃), and operating current (accuracy ±0.1A) in real time, and to identify hidden faults and changes in operating conditions such as equipment wear, jamming, and overload.

[0026] The intermediate and final state sensing unit of the product uses a machine vision inspection system, a laser diameter gauge, and an online crystallinity detector to non-contactly detect intermediate state indicators such as the uniformity of the diameter of the nascent filament (accuracy ±0.001mm), the surface morphology of the fiber after stretching, and the crystallinity of the fiber after heat setting (accuracy ±1%).

[0027] It should be further explained that, in the specific implementation process, the process of constructing a digital twin model corresponding to the physical entity of the polyester staple fiber production line includes: The process involves acquiring the physical entities of polyester staple fiber production equipment in the physical space of the current polyester staple fiber production line, constructing a three-dimensional digital space, mapping the physical entities of the polyester staple fiber production equipment into the three-dimensional digital space to generate a three-dimensional model, setting up API interfaces on the three-dimensional model, and connecting the assembly and connection relationships between the physical entities in the physical space to the three-dimensional model in the three-dimensional digital space. It also involves acquiring the three-dimensional models corresponding to each polyester staple fiber production equipment in each process subsequence, as well as the operational data of the operational data monitoring nodes, and mapping the operational data of each process subsequence to the corresponding three-dimensional model of the polyester staple fiber production equipment to generate a digital twin model.

[0028] It should be further explained that, in the specific implementation process, the generation of the digital twin model is based on the recent stable production process data of the polyester staple fiber production line (production indicators, equipment operating parameters, environmental parameters, and product quality data of each operational data monitoring node). The digital twin model is subjected to static and dynamic dual calibration to ensure that the deviation between the model's parameter response, quality transfer, equipment linkage and the physical production line is ≤±2%, meeting the accuracy requirements of the process test. Static calibration: Match the 3D model parameters of the core equipment in each process subsequence (such as metering pump speed-flow rate, stretching roller speed ratio-stretching ratio, cutting machine cutter spacing-length accuracy) to the physical equipment; Dynamic calibration: Simulate the normal production state of the physical production line to verify that the linkage response of the parameters in the whole process of "melt preparation → spinning → cooling → stretching → heat setting → cutting" and the transmission law of intermediate quality indicators are consistent with the physical production line. For example, after adjusting the spinning temperature, the adaptation changes of the crystallinity of the nascent filament and the cooling wind speed in the model need to match the physical production line.

[0029] It should be further explained that, in the specific implementation process, the process of conducting single-factor variable process experiments based on the digital twin model, obtaining the single-factor coupling coefficient between the single-factor variable and the quality indicator, and screening out the principal factors from the single-factor variable includes: Based on the production monitoring indicators of each operational data monitoring node, experimental variables (core controllable parameters of each subsequence (such as melt temperature for melt preparation, spinning box temperature for spinning, side-blowing air temperature / speed for cooling, total stretching ratio for stretching, and heat setting temperature / time), environmental parameters (temperature and humidity fluctuations), and raw material characteristic parameters (melt viscosity deviation), all of which are parameters that can be precisely controlled in the digital twin model), fixed indicators (intermediate quality baseline indicators of each subsequence (such as melt viscosity deviation ≤ ±0.03dL / g, nascent filament diameter uniformity ≥97%, and stretching orientation ≥75%) are set. If these indicators exceed the standard during the test, the experimental parameter combination is deemed invalid, and the corresponding data is discarded) and quality indicators (finished product quality, such as nascent filament crystallinity, stretching orientation, and finished product heat shrinkage rate) are set. Based on the target product specifications (such as 1.56dtex×38mm ordinary polyester staple fiber) and historical operational data of the polyester staple fiber production line in several historical monitoring periods, information on different processing scenarios is obtained (different target product specifications and environmental parameters correspond to different processing scenarios). One processing scenario is randomly selected from different processing scenario information as the experimental scenario. One experimental variable is randomly selected as the single-factor variable from the experimental variables. Other experimental variables are converted to their normal values ​​under the experimental scenario. A gradient for the single-factor variable is set (e.g., spinning temperature is adjusted in gradients of 275℃, 278℃, 280℃, 282℃, and 285℃, with a step size of 3℃ / 2℃). The gradient and normal values ​​of the single-factor variable are input into a digital twin model for multi-gradient experiments to obtain the single-factor coupling coefficient between the single-factor variable and the quality indicators. For example, cooling, stretching, and heat setting parameters are fixed at normal values, and the melt preparation parameters are adjusted... The melt viscosity (0.65, 0.68, 0.70, 0.72, 0.75 dL / g) and melt delivery temperature (270, 275, 280, 285℃) of the sequence were monitored. The uniformity of filament output, the breakage rate of filament, and the crystallinity of the nascent filament were monitored. The correspondence between the "melt parameter adjustment range - the change range of spinning quality index" was recorded. The above process was repeated to obtain the single-factor coupling coefficient between different single-factor variables and quality indexes under different experimental scenarios. The coupling coefficient threshold was preset (0.75). Single-factor variables with single-factor coupling coefficients greater than the coupling coefficient threshold were marked as principal factors.

[0030] It should be further explained that, in the specific implementation process, the process of obtaining the single-factor coupling coefficient between the single-factor variable and the quality indicator includes: ; in, is the single-factor coupling coefficient, and n is the number of gradient test groups for the single-factor variable (n≥5). Set values ​​for the single-factor variables in the i-th group of experiments. , where represents the measured value of the quality index of the i-th group of experiments (the result of the digital twin model simulation).

[0031] It should be further explained that, in the specific implementation process, the process of conducting multi-factor coupled process experiments based on the single-factor coupling coefficient of the principal factor and extracting the process linkage rules under different experimental scenarios includes: The coupling gradient combination of principal factors under the current experimental scenario is set according to the orthogonal experimental design. The experimental variables other than the principal factors are converted into normal values ​​under the current experimental scenario. The coupling gradient combination of principal factors and normal values ​​are input into the digital twin model for multi-gradient experiments to obtain the coupling effect coefficient between principal factors and the linkage coupling coefficient between principal factors and quality indicators. The linkage coupling coefficient between principal factors and quality indicators is extracted by process linkage rule structure to obtain the process linkage rules under the current experimental scenario. For example, spinning temperature (principal factor 1) and cooling wind speed (principal factor 2) are selected as coupling variables. The orthogonal experimental design is used with a 4×4 gradient combination (e.g., spinning temperature 275 / 278 / 280 / 285℃, cooling wind speed 3.0 / 3.2 / 3.5 / 4.0m / s). The crystallinity and diameter uniformity of the nascent filament are monitored, and the influence of the coupling effect of the two factors on the quality indicators of the cooling subsequence is analyzed. The process linkage rule of "spinning temperature + cooling wind speed" is extracted. The above process is repeated to obtain the process linkage rules under different experimental scenario conditions.

[0032] It should be further explained that, in the specific implementation process, the process of obtaining the coupling effect coefficients between principal factors and the linkage coupling coefficients between principal factors and quality indicators includes: ; in, The coupling effect coefficient is... Main factor , Simultaneously, when adjusting, the actual relative rate of change (experimental measured value) of the quality indicator Q, / Principal factors , When adjusted individually, the relative change rate of quality indicator Q by a single factor (measured value of single factor experiment). ; in, The magnitude of the change in the quality index. This is the coupling effect correction coefficient, derived from the coupling effect coefficient. The fitting yielded: ( >0), ( <0), (No interaction) Main factor The single-factor coupling coefficient, Main factor The single-factor coupling coefficient, , Principal factors , The magnitude of the change, This is the linkage coupling coefficient.

[0033] It should be further explained that, in the specific implementation process, the structured extraction of process linkage rules is based on the quantitative results of data correlation analysis. The linkage rules are extracted in a structured manner according to four core elements (triggering condition, adjustment object, adjustment direction, and adjustment magnitude), forming a rule expression that is machine-recognizable and executable by the control system. For example: Example of rule expression: Triggering conditions: Spinning temperature increases by 2℃ + cooling air velocity decreases by 0.3m / s (two-factor coupling, ordinary spinning varieties); Adjustment targets: Cooling air temperature of the cooling subsequence, stretching heat box temperature of the stretching subsequence; Adjust direction: air temperature decreases, hot box temperature increases; Adjustment range: Cooling air temperature -3℃ (synergistic effect, the range is greater than the sum of single factors), stretching box temperature +5℃.

[0034] It should be further explained that, in the specific implementation process, the process of constructing a multi-variety process knowledge base based on the process linkage rules under different test scenarios, and configuring the initial parameter combination for the polyester staple fiber production line in the initialization stage or variety switching stage based on the multi-variety process knowledge base includes: Based on the process linkage rules of polyester staple fiber production lines under different test scenarios, a multi-variety process knowledge base is constructed. The multi-variety process knowledge base covers the core process parameter ranges and process linkage rules of different varieties of polyester staple fibers, such as ordinary spinning, high-speed spinning, high-strength, low-shrinkage, and antistatic, under different environmental parameters. The knowledge base supports online updates, and the process parameters and production experience of newly developed varieties can be quickly entered to realize the accumulation and reuse of knowledge. The multi-variety process knowledge base is linked with the digital twin model through the API interface to determine the current operating stage of the polyester staple fiber production line. When the polyester staple fiber production line is in the initialization stage, the target product specifications and environmental parameters in the operating data of the current polyester staple fiber production line are input into the multi-variety process knowledge base for matching to obtain the initial parameter combination of the current polyester staple fiber production line under the current target product specifications and environmental parameter conditions. The initial parameter combination includes the core process parameter range of each process subsequence and the process linkage rules between each process subsequence. When the polyester staple fiber production line is in the product switching stage, the target product specifications after the switch and the environmental parameters in the operating data are input into the multi-product process knowledge base for matching to obtain the initial parameter combination of the current polyester staple fiber production line.

[0035] It should be further explained that, in the specific implementation process, the process of generating simulated operating data of the polyester staple fiber production line in real time based on the digital twin model, monitoring the simulated operating data in real time, and performing parameter combination update operations or feedforward disturbance judgment based on the real-time monitoring results includes: When the polyester staple fiber production line is in the production operation phase, the operation data collected by each operation data monitoring node is input into the digital twin model. Based on the output of the digital twin model, the simulated operation data of each process subsequence within the current monitoring cycle is obtained, and the threshold range of the production monitoring indicators of each process subsequence is obtained. The numerical time series of each production monitoring indicator in the simulated operation data is compared with the corresponding threshold range to obtain the cumulative time of each production monitoring indicator that is not within the corresponding threshold range. A preset time error threshold is set. If the cumulative time corresponding to a production monitoring indicator is greater than the time error threshold, a process abnormality warning is generated and parameter combination update operation is performed. If the cumulative time corresponding to each production monitoring indicator is less than or equal to the time error threshold, normal process information is generated and feedforward disturbance judgment is performed.

[0036] It should be further explained that, in the specific implementation process, the parameter combination update operation includes: When the polyester staple fiber production line is in the production operation stage and generates a process abnormality warning, constraints are set according to the threshold range of the production monitoring indicators of each process sub-sequence. These constraints include fully considering the process constraints of polyester staple fiber production (such as the spinning temperature must not exceed the melt thermal degradation temperature, the stretch ratio must match the crystallinity of the nascent fiber, and the matching range of heat setting temperature and time), equipment constraints (such as the rated range of the metering pump speed and the stretching roller speed ratio, and the load limit of the equipment operation), raw material constraints (such as the adaptation range of spinning parameters for melts of different viscosities), and environmental constraints (such as the lower limit of cooling wind speed in high humidity environments). All constraints are quantified and embedded into a digital twin model to ensure that the optimized parameter combination fully conforms to the actual production and has no execution risk. Several parameter combinations are randomly generated under the current target product specifications and environmental parameters. Chromosome encoding and initialization of several parameter combinations are performed on these combinations to generate an initial population. An adaptive product quality objective function and an energy objective function are constructed. A fitness function is constructed based on the adaptive product quality objective function and the energy objective function. Based on the initial population, the fitness function, and the constraints, experiments are conducted through the digital twin model to obtain the optimal parameter combination. The initial parameter combination is then updated based on the optimal parameter combination.

[0037] It should be further explained that, in the specific implementation process, the fitness function is: ; in, Let p be the fitness value of the parameter combination. The product quality coefficient is the parameter combination p. The total energy consumption of the entire process for parameter combination p. =0.6, =0.4; The adaptive product quality objective function is: ; ; ; in, Let i be the process weight of the i-th quality indicator. Let i be the single-index quality of the i-th quality indicator. The optimal target value for the i-th quality indicator (set based on customer product quality requirements), Let i be the actual experimental value of the i-th quality indicator. Let be the lower limit of the threshold for the i-th quality indicator. is the upper limit of the threshold for the i-th quality indicator.

[0038] The energy objective function is: ; in, The total energy consumption of the parameter combination p throughout the entire process is represented by a lower value, which indicates better energy efficiency. The core energy types are represented by 3 (electricity, steam, and cooling water, covering more than 95% of the energy consumption in polyester staple fiber production). The total consumption of the j-th type of energy during the production process (calculated by statistical period, such as 1 hour / 1 batch). The standard coal conversion factor (kgce / unit energy) for the j-th energy source is taken according to the national "General Rules for Calculation of Comprehensive Energy Consumption" (GB / T2589-2020) and remains fixed. To monitor the waste heat recovery and convert it into standard coal equivalent (kgce) during the monitoring period, the main waste heat recovery is from spinning / heat setting and cooling water. To monitor the production (t) of qualified polyester staple fiber during the monitoring period, only qualified products are counted, and unqualified products are excluded (to avoid distortion of energy consumption allocation).

[0039] It should be further explained that, in the specific implementation process, the process of obtaining the optimal parameter combination through experiments using a digital twin model based on the initial population, fitness function, and constraint formula includes: The parameter combinations are encoded as chromosomes. For example, using the core controllable parameters of each subsequence (such as melt temperature for melt preparation, spinning box temperature for spinning, side-blowing air temperature / speed for cooling, total stretching ratio for stretching, and setting temperature / time for heat setting), environmental parameters (temperature and humidity fluctuations), and raw material characteristic parameters, a certain number of chromosomes are randomly generated to form an initial population. Each chromosome represents a possible parameter combination. Each parameter combination is input into a digital twin model for experimentation to obtain a fitness function. Then, a tournament selection method is used to select chromosomes with higher fitness from the current population as parents. The parent chromosomes are cross-crossed, exchanging some genes to generate new offspring chromosomes, simulating the exchange of genes in biological genetics to generate new parameter combinations. The offspring chromosomes are mutated, randomly changing some genes to increase population diversity and avoid getting trapped in local optima. The above steps are repeated iteratively until the termination condition is met, such as reaching the maximum number of iterations or the fitness value no longer significantly improving, at which point the optimal parameter combination is output.

[0040] It should be further explained that, in the specific implementation process, the process of constructing the adaptive product quality objective function includes: Statistical analysis is performed on the single-factor coupling coefficients between each single-factor variable and quality indicator under the current target product specifications and environmental parameters to obtain the coupling sensitivity coefficient of each quality indicator. The specification requirements of the current target product are obtained. Based on the coupling sensitivity coefficients and specification requirements, the process weights of each quality indicator are set. An adaptive product quality objective function is constructed based on each quality indicator and its process weight.

[0041] It should be further explained that, in the specific implementation process, the process of obtaining the coupling sensitivity coefficients of each quality indicator includes: ; in, It is the absolute value of the single-factor coupling coefficient of the k-th single-factor variable with respect to the i-th quality index (considering only the degree of coupling, not the direction of coupling). The maximum value of the weighted sum of the single-factor coupling coefficients of all quality indicators is used for normalization to make the maximum value of the coupling sensitivity coefficient 1, where m is the total number of single-factor variables and n is the total number of quality indicators.

[0042] The process of setting the process weights for each quality indicator based on the coupling sensitivity coefficient and specification requirements includes: ; in, This is the product quality requirement correction coefficient for the i-th quality indicator, assigned a value based on the specifications of the target product, reflecting the importance of this quality indicator in product quality evaluation; for example, in the production of high-strength polyester staple fiber, the breaking strength... of =1.5, linear density deviation of =1.0, elongation at break of =0.8; When producing ordinary spun fibers, all indicators are... =1.0 (no special requirements); Correction coefficient selection principle: core quality indicators ≥1, secondary quality indicator ≤1;

[0043] This is the sum of the products of the coupling sensitivity coefficients and correction coefficients of all quality indicators, used for normalization. The larger the value, the higher the weight of this quality indicator in process parameter optimization and comprehensive quality evaluation. The control system will prioritize ensuring this indicator meets the target requirements during parameter optimization. The process weight varies with the target product specifications (…). ) and environmental / raw material conditions ( (Dynamically changing) and adjusted in real time to achieve adaptive setting of weights.

[0044] It should be further explained that, in the specific implementation process, the process of determining the feedforward disturbance includes: Extract the linkage coupling coefficients between principal factors and quality indicators from the simulated operation data of each process subsequence within the current monitoring period. Extract the linkage coupling coefficients using a structured process linkage rule extraction method to generate process linkage rules. Perform consistency verification between these process linkage rules and the process linkage rules in the initial parameter combination of the current polyester staple fiber production line. (This involves establishing qualitative comparison standards and quantitative calculation indicators for each element in the order of triggering condition → adjustment object → adjustment direction → adjustment magnitude, clarifying anomaly judgment thresholds, and calculating the quantitative deviation of the adjustment magnitude under the premise that the triggering condition, adjustment object, and adjustment direction are all qualitatively consistent. If the adjustment magnitude deviation meets the preset deviation judgment standard, the consistency verification passes; otherwise, the consistency verification fails.) If the consistency verification fails, perform a parameter combination update operation.

[0045] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A process operation parameter optimization and control system for a polyester staple fiber production line, characterized in that, This includes a monitoring center, which has communication connections with data acquisition modules, model building modules, correlation analysis modules, parameter configuration modules, and parameter update modules. The data acquisition module is used to acquire process information of the polyester staple fiber production line, divide the process into sub-sequences according to the process information, set up operation data monitoring nodes, and acquire operation data at each point; The model building module is used to build digital twin models of the physical entities of a polyester staple fiber production line; The correlation analysis module is used to conduct single-factor variable process experiments based on digital twin models, obtain the single-factor coupling coefficient between single-factor variables and quality indicators, and screen out the principal factors from the single-factor variables; based on the single-factor coupling coefficient of the principal factors, it conducts multi-factor coupled process experiments and extracts the process linkage rules under different test scenarios. The parameter configuration module is used to build a multi-variety process knowledge base based on the process linkage rules under different test scenarios, and to configure the initial parameter combination for the polyester staple fiber production line in the initialization stage or the variety switching stage based on the multi-variety process knowledge base. The parameter update module is used to generate simulated operation data of the polyester staple fiber production line in real time based on the digital twin model, monitor the simulated operation data in real time, and perform parameter combination update operations or feedforward disturbance judgment based on the real-time monitoring results.

2. The process operation parameter optimization and control system for a polyester staple fiber production line according to claim 1, characterized in that, The process of acquiring operational data at each location includes: Obtain the process characteristics of the current polyester staple fiber production equipment, break down the polyester staple fiber production line process into several process sub-sequences according to the process characteristics, and set up operation data monitoring nodes in each process sub-sequence. Based on the functional characteristics of the process flow characteristics of the corresponding process subsequence, the production monitoring indicators of each operation data monitoring node are obtained by data retrieval. The operation data monitoring node obtains operation data in real time according to the production monitoring indicators, marks the monitoring time, and sets the monitoring cycle.

3. The process operation parameter optimization and control system for a polyester staple fiber production line according to claim 2, characterized in that, The process of constructing a digital twin model of the physical entity of a polyester staple fiber production line includes: The process involves acquiring the physical entities of polyester staple fiber production equipment in the physical space of the current polyester staple fiber production line, constructing a three-dimensional digital space, mapping the physical entities of the polyester staple fiber production equipment into the three-dimensional digital space to generate a three-dimensional model, setting up API interfaces on the three-dimensional model, and connecting the assembly and connection relationships between the physical entities in the physical space to the three-dimensional model in the three-dimensional digital space. It also involves acquiring the three-dimensional models corresponding to each polyester staple fiber production equipment in each process subsequence, as well as the operational data of the operational data monitoring nodes, and mapping the operational data of each process subsequence to the corresponding three-dimensional model of the polyester staple fiber production equipment to generate a digital twin model.

4. The process operation parameter optimization and control system for a polyester staple fiber production line according to claim 3, characterized in that, The process of conducting single-factor variable process experiments based on digital twin models, obtaining the single-factor coupling coefficient between single-factor variables and quality indicators, and screening principal factors from single-factor variables includes: Based on the production monitoring indicators of each operational data monitoring node, experimental variables, fixed indicators, and quality indicators are set, and information on different processing scenarios is obtained based on the target product specifications and historical operational data of the polyester staple fiber production line in several historical monitoring cycles. One processing scenario is randomly selected from different processing scenario information as the experimental scenario. One experimental variable is randomly selected as the single-factor variable. Other experimental variables are converted into normal values ​​under the experimental scenario. The gradient of the single-factor variable is set. The gradient of the single-factor variable and the normal value are input into the digital twin model to conduct multi-gradient experiments. The single-factor coupling coefficient between the single-factor variable and the quality indicator is obtained. The above process is repeated to obtain the single-factor coupling coefficient between different single-factor variables and the quality indicator under different experimental scenario conditions. The coupling coefficient threshold is preset. The single-factor variables with a single-factor coupling coefficient with the quality indicator that is greater than the coupling coefficient threshold are marked as principal factors.

5. The process operation parameter optimization and control system for a polyester staple fiber production line according to claim 4, characterized in that, The process of extracting process linkage rules under different experimental scenarios by conducting multi-factor coupled process experiments based on the single-factor coupling coefficient of the principal factors includes: Define the coupling gradient combination of the principal factors under the current experimental scenario conditions, convert the experimental variables other than the principal factors into regular values ​​under the current experimental scenario, input the coupling gradient combination of the principal factors and the regular values ​​into the digital twin model to conduct multi-gradient experiments, obtain the coupling effect coefficient between the principal factors and the linkage coupling coefficient between the principal factors and the quality indicators, extract the process linkage rules structured from the linkage coupling coefficient between the principal factors and the quality indicators, obtain the process linkage rules under the current experimental scenario conditions, repeat the above process to obtain the process linkage rules under different experimental scenario conditions.

6. The process operation parameter optimization and control system for a polyester staple fiber production line according to claim 5, characterized in that, Based on the process linkage rules under different test scenarios, a multi-variety process knowledge base is constructed. The process of configuring initial parameter combinations for polyester staple fiber production lines in the initialization or variety switching stages based on the multi-variety process knowledge base includes: Based on the process linkage rules of the polyester staple fiber production line under different test scenarios, a multi-product process knowledge base is constructed. The multi-product process knowledge base is linked with the digital twin model through an API interface to determine the current operating stage of the polyester staple fiber production line. When the polyester staple fiber production line is in the initialization stage, the target product specifications and environmental parameters in the operating data of the current polyester staple fiber production line are input into the multi-product process knowledge base for matching to obtain the initial parameter combination of the current polyester staple fiber production line under the current target product specifications and environmental parameter conditions. The initial parameter combination includes the core process parameter range of each process subsequence and the process linkage rules between each process subsequence. When the polyester staple fiber production line is in the product switching stage, the target product specifications after the switch and the environmental parameters in the operating data are input into the multi-product process knowledge base for matching to obtain the initial parameter combination of the current polyester staple fiber production line.

7. The process operation parameter optimization and control system for a polyester staple fiber production line according to claim 6, characterized in that, The process of generating simulated operating data of a polyester staple fiber production line in real time based on a digital twin model, monitoring the simulated operating data in real time, and performing parameter combination updates or feedforward disturbance determination based on the real-time monitoring results includes: When the polyester staple fiber production line is in the production operation phase, the operation data collected by each operation data monitoring node is input into the digital twin model. Based on the output of the digital twin model, the simulated operation data of each process subsequence within the current monitoring cycle is obtained, and the threshold range of the production monitoring indicators of each process subsequence is obtained. The numerical time series of each production monitoring indicator in the simulated operation data is compared with the corresponding threshold range to obtain the cumulative time of each production monitoring indicator that is not within the corresponding threshold range. A preset time error threshold is set. If the cumulative time corresponding to a production monitoring indicator is greater than the time error threshold, a process abnormality warning is generated and parameter combination update operation is performed. If the cumulative time corresponding to each production monitoring indicator is less than or equal to the time error threshold, normal process information is generated and feedforward disturbance judgment is performed.

8. The process operation parameter optimization and control system for a polyester staple fiber production line according to claim 7, characterized in that, The process of performing parameter combination update operations includes: When the polyester staple fiber production line is in the production operation stage and generates a process abnormality warning, constraints are set according to the threshold range of the production monitoring indicators of each process sub-sequence. Several parameter combinations are randomly generated under the current target product specifications and environmental parameters. Chromosome encoding and initialization of the parameter combinations are performed to generate an initial population. An adaptive product quality objective function and an energy objective function are constructed. A fitness function is constructed based on the adaptive product quality objective function and the energy objective function. Based on the initial population, fitness function and constraints, experiments are conducted through a digital twin model to obtain the optimal parameter combination. The initial parameter combination is then updated based on the optimal parameter combination.

9. The process operation parameter optimization and control system for a polyester staple fiber production line according to claim 8, characterized in that, The process of constructing an adaptive product quality objective function includes: Statistical analysis is performed on the single-factor coupling coefficients between each single-factor variable and quality indicator under the current target product specifications and environmental parameters to obtain the coupling sensitivity coefficient of each quality indicator. The specification requirements of the current target product are obtained. Based on the coupling sensitivity coefficients and specification requirements, the process weights of each quality indicator are set. An adaptive product quality objective function is constructed based on each quality indicator and its process weight.

10. The process operation parameter optimization and control system for a polyester staple fiber production line according to claim 9, characterized in that, The process of determining feedforward disturbances includes: Extract the linkage coupling coefficients between the main factors and quality indicators from the simulated operation data of each process subsequence within the current monitoring period. Extract the linkage coupling coefficients using process linkage rules in a structured manner to generate process linkage rules. Perform consistency verification between the process linkage rules and the process linkage rules in the initial parameter combination of the current polyester staple fiber production line. If the consistency verification fails, perform a parameter combination update operation.

Images