A data processing system and method for textile auxiliary production

Abstract

The application provides a data processing system and method for textile auxiliary production, which generates a dynamic gradient data set by monitoring the solution flow state of the textile auxiliary; acquires a reflection signal parameter set of a surface of a textile substrate to be processed under electromagnetic excitation of a preset frequency band, constructs an energy distribution map based on resonance frequency offset data associated with plasma resonance effect in the reflection signal parameter set, and dynamically couples the energy distribution map with the dynamic gradient data set to generate a time-space evolution data set, extracts infiltration dynamic fluctuation data, and generates a latent variable set representing the molecular structure characteristics of the textile auxiliary; in combination with a preset infiltration rate constraint condition, a target parameter data set is generated for data processing of textile auxiliary production; the application breaks through the limitation of the disconnection between static surface energy detection and dynamic auxiliary flow analysis in the prior art, realizes real-time dynamic adaptation of the substrate surface energy and the auxiliary molecular characteristics, and improves the infiltration uniformity and process control precision.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a data processing system and method for the production of textile auxiliaries. Background Technology

[0002] In the modern textile industry, the performance of textile auxiliaries plays a crucial role in improving the quality and functionality of textiles. Especially in the manufacturing of high-end textiles, to meet the growing market demand for waterproof, stain-resistant, and breathable properties, precise control of the molecular structure and wetting behavior of textile auxiliaries has become paramount. This not only improves product quality consistency but also reduces production costs and minimizes resource waste.

[0003] The current mainstream approach is a data processing method based on machine learning algorithms. This method predicts the optimal formulation and processing conditions for textile auxiliaries by learning from a large amount of historical data. It collects various physical parameters, including solution flowability, temperature, and pressure, and uses experimental data from successful past cases as a training set. Advanced machine learning models are then used to analyze this data to identify key factors associated with the ideal performance of textile auxiliaries. However, existing methods have some inherent limitations. These include their reliance on static historical data, resulting in poor adaptability and flexibility in the face of rapidly changing production environments or new materials; and their inability to accurately capture the full picture of the complex interactions between the molecular structure of textile auxiliaries and the substrate, leading to insufficient precision and dynamic response in the data processing results. Summary of the Invention

[0004] This invention provides a data processing system and method for textile auxiliary production, which solves the problems of existing technologies based on static data analysis, which have poor adaptability and flexibility to rapidly changing production environments or new materials, and lack of sufficient accuracy and dynamic response capability in data processing results.

[0005] In a first aspect, the present invention provides a data processing method for the production of textile auxiliaries, comprising:

[0006] Monitor the solution flow state of textile auxiliaries to generate a dynamic gradient data set;

[0007] A set of reflected signal parameters of the surface of the textile substrate to be treated under electromagnetic excitation in a preset frequency band is obtained. Based on the resonant frequency shift data associated with the plasma resonance effect in the set of reflected signal parameters, an energy distribution map is constructed.

[0008] The dynamic gradient data set is dynamically coupled with the surface energy characteristics of each partition of the energy distribution map to generate a spatiotemporal evolution data set associated with the infiltration behavior.

[0009] Extract the infiltration dynamic fluctuation data from the spatiotemporal evolution dataset, and generate a set of latent variables characterizing the molecular structure properties of textile auxiliaries based on the infiltration dynamic fluctuation data.

[0010] Based on the matching relationship between the set of latent variables and the preset wetting rate constraints, a target parameter dataset is generated for data processing of textile auxiliary production.

[0011] Optionally, the dynamic gradient data set is dynamically coupled with the surface energy characteristics of each partition of the energy distribution map to generate a spatiotemporal evolution data set associated with infiltration behavior, including:

[0012] The surface energy intensity value of each partition in the energy distribution map is multiplied by the plasma resonance frequency offset value of the corresponding partition to generate a set of correlation factors corresponding to the spatial location of each partition.

[0013] Monitor the temporal fluctuation amplitude of the dynamic gradient data set, and calculate the difference of the temporal fluctuation amplitude based on the preset tension value of adjacent time nodes to generate a set of weight coefficients bound to the spatial location of each partition.

[0014] The set of correlation factors and the set of weight coefficients are superimposed point by point according to the spatial location corresponding to each partition to generate a coupling parameter matrix;

[0015] Within a preset time slice sequence, the matrix units in the coupling parameter matrix that satisfy the phase synchronization condition are energy redistributed to generate the redistributed coupling parameter matrix.

[0016] Based on the allocated coupling parameter matrix, a target matrix unit is selected, and based on the target matrix unit, a spatiotemporal evolution data set associated with the infiltration behavior is generated.

[0017] Optionally, based on the allocated coupling parameter matrix, a target matrix element is selected, and based on the target matrix element, a spatiotemporal evolution dataset associated with infiltration behavior is generated, including:

[0018] Mark the target matrix units in the allocated coupling parameter matrix that exceed the preset gradient value, use the target matrix units as candidate units, and extract the spatial coordinate points corresponding to the candidate units to generate an initial coordinate point set;

[0019] The spatial coordinate points in the initial coordinate point set are used to perform bidirectional path verification with the first target point in the adjacent time slice in the preset time slice sequence to generate a verified coordinate point set.

[0020] Based on the liquid-solid interfacial tension value corresponding to each spatial coordinate point in the verified coordinate point set, a processing priority is assigned to the spatial coordinate points in the verified coordinate point set to obtain the target spatial coordinate point priority sequence.

[0021] Based on the priority sequence of the target spatial coordinate points, the verified coordinate points in the set of verified coordinate points within the same time slice are connected with the second target point in the adjacent time slice to generate an initial spatiotemporal connection network.

[0022] The branch nodes in the initial spatiotemporal connection network are detected to generate an optimized spatiotemporal connection network;

[0023] The spatial coordinates of the verified first target point and the verified second target point in the optimized spatiotemporal connection network are concatenated to generate a spatiotemporal evolution data set associated with the infiltration behavior.

[0024] Optionally, based on the liquid-solid interfacial tension value corresponding to each spatial coordinate point in the verified coordinate point set, a processing priority is assigned to the spatial coordinate points in the verified coordinate point set, resulting in a target spatial coordinate point priority sequence including:

[0025] Extract the liquid-solid interfacial tension values ​​of each spatial coordinate point in the verified coordinate point set;

[0026] Based on the liquid-solid interfacial tension value, sort the spatial coordinate points in the verified coordinate point set to generate a sorting result;

[0027] Based on the sorting results, priority weights are assigned to the spatial coordinate points in the set of verified coordinate points to generate a priority sequence corresponding to the set of verified coordinate points.

[0028] The priority sequence is matched with the spatial distribution density of the verified coordinate point set to generate a spatial coordinate point priority sequence. The priority of spatial coordinate points in the spatial coordinate point priority sequence that have a spatial distribution density higher than a preset threshold is enhanced to generate a target spatial coordinate point priority sequence.

[0029] Optionally, branch nodes in the initial spatiotemporal connectivity network are detected to generate an optimized spatiotemporal connectivity network, including:

[0030] Extract the liquid-solid interfacial tension values ​​on all connection paths of each branch node in the initial spatiotemporal connection network, and calculate the variation range of the liquid-solid interfacial tension values ​​between adjacent time slices. The branch node is defined as a spatial coordinate point where at least two different connection paths intersect.

[0031] Select a first target branch node from the branch nodes that has a connection path with a change amplitude exceeding a preset change threshold, select a second target branch node from the first target branch node that meets a preset condition, and degenerate the first target branch node that does not meet the preset condition into an ordinary node. The preset condition is that the direction of the liquid-solid interface tension gradient is consistent with the direction of the surface energy distribution of the textile substrate to be treated, and that there are at least two different connection paths.

[0032] The connection relationships between the second target branch node and the ordinary nodes associated with the second target branch node are reconstructed to generate the optimized spatiotemporal connection network.

[0033] Optionally, based on the wetting dynamic fluctuation data, a set of latent variables characterizing the molecular structure properties of textile auxiliaries is generated, including:

[0034] The dynamic fluctuation data of the infiltration is analyzed to generate a set of fluctuation components at different time scales;

[0035] Phase synchronization processing is performed on each wave component in the wave component set to generate a synchronized wave component sequence.

[0036] The synchronized fluctuation component sequence is mapped to a preset multi-dimensional feature space to generate a target feature space;

[0037] The weight distribution of the projected coordinates in the target feature space is adjusted according to the order of magnitude of the energy characteristics of the partitioned surface to obtain the adjusted multidimensional feature space;

[0038] The weight parameters of each projected coordinate in the adjusted feature space are extracted to generate a set of latent variables characterizing the molecular structure properties of textile auxiliaries.

[0039] Optionally, based on the matching relationship between the set of latent variables and the preset wetting rate constraints, a target parameter dataset is generated for data processing of textile auxiliary production based on the target parameter dataset, including:

[0040] Each latent variable in the set of latent variables is matched with the benchmark parameter range in the preset infiltration rate constraint at multiple levels to generate an intermediate parameter sequence with hierarchical correlation.

[0041] Select a target parameter sequence that meets the preset conflict resolution threshold condition from the intermediate parameter sequence, and generate a parameter optimization rule set based on the target parameter sequence;

[0042] Based on the parameter optimization rule set, the rule-driven parameters of the latent variable set are transformed to generate the target parameter dataset;

[0043] Data processing for textile auxiliary production is performed based on the target parameter dataset.

[0044] Secondly, the present invention provides a data processing system for the production of textile auxiliaries, comprising:

[0045] The monitoring module is used to monitor the solution flow state of textile auxiliaries and generate a dynamic gradient data set.

[0046] The construction module is used to acquire the set of reflection signal parameters of the surface of the textile substrate to be treated under electromagnetic excitation in a preset frequency band, and to construct an energy distribution map based on the resonance frequency offset data associated with the plasma resonance effect in the set of reflection signal parameters.

[0047] The coupling module is used to dynamically couple the dynamic gradient data set with the surface energy characteristics of each partition of the energy distribution map to generate a spatiotemporal evolution data set associated with the infiltration behavior.

[0048] The extraction module is used to extract the infiltration dynamic fluctuation data from the spatiotemporal evolution data set, and generate a set of latent variables characterizing the molecular structure properties of textile auxiliaries based on the infiltration dynamic fluctuation data.

[0049] The processing module is used to generate a target parameter dataset based on the matching relationship between the set of latent variables and the preset wetting rate constraints, so as to perform data processing for textile auxiliary production based on the target parameter dataset.

[0050] Thirdly, the present invention provides a computing device including a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform a data processing method for textile auxiliary production as described in any of the first aspects.

[0051] Fourthly, the present invention provides a computer storage medium storing computer program instructions thereon, wherein the computer program instructions, when executed by a processor, implement a data processing method for textile auxiliary production as described in any one of the first aspects.

[0052] In this invention, the solution flow state of textile auxiliaries is monitored to generate a dynamic gradient data set; a set of reflection signal parameters of the surface of the textile substrate to be treated under electromagnetic excitation in a preset frequency band is obtained; based on the resonance frequency shift data associated with the plasma resonance effect in the set of reflection signal parameters, an energy distribution map is constructed; the dynamic gradient data set is dynamically coupled with the surface energy characteristics of each partition of the energy distribution map to generate a spatiotemporal evolution data set associated with wetting behavior; wetting dynamic fluctuation data in the spatiotemporal evolution data set is extracted; based on the wetting dynamic fluctuation data, a set of latent variables characterizing the molecular structure properties of textile auxiliaries is generated; based on the matching relationship between the set of latent variables and preset wetting rate constraints, a target parameter dataset is generated for data processing of textile auxiliary production based on the target parameter dataset. The technical solution provided by this invention captures the changes in the flow characteristics of the additive solution on the substrate surface in real time, providing a dynamic data foundation for wetting behavior analysis and solving the problem that traditional static detection cannot reflect transient flow. It quantifies the dynamic changes in the energy distribution on the substrate surface through the plasma resonance effect, overcoming the technical limitation of existing optical detection technologies being insensitive to real-time surface energy response. It integrates flow state and substrate surface characteristic data to establish a spatiotemporal correlation model of wetting behavior, overcoming the modeling defects of separating fluid dynamics and surface energy data in existing solutions. It analyzes the molecular structure characteristics of the additive from dynamic fluctuations, achieving correlation optimization between process parameters and molecular-level characteristics, solving the shortcomings of traditional methods that rely solely on macroscopic parameter adjustments. It outputs optimized parameters adapted to substrate characteristics and additive molecular structure, improving wetting uniformity and production efficiency, filling the gap in existing technologies that cannot achieve real-time closed-loop control. Specifically, a set of correlation factors is generated by multiplying the energy intensity of the substrate surface with the offset of the plasma resonance frequency, and a set of weight coefficients is generated by combining the temporal fluctuation amplitude of dynamic gradient data to construct a coupling parameter matrix. Within a preset time slice, the matrix units that meet the phase synchronization condition are redistributed with energy to finally generate a set of spatiotemporal evolution data. The substrate surface characteristics and the flow state of the auxiliaries are dynamically correlated, and the accuracy of the wetting path prediction is optimized by energy redistribution. This solves the problem of wetting behavior modeling deviation caused by ignoring the dynamic effect of plasma resonance and the temporal fluctuation characteristics in the existing technical solutions, and improves the compatibility between textile auxiliaries and substrates.

[0053] These or other aspects of the invention will become more apparent from the following description of the embodiments. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 A flowchart of a data processing method for textile auxiliary agent production provided in an embodiment of the present invention;

[0056] Figure 2 This is a schematic diagram of the structure of a data processing system for textile auxiliary agent production provided in an embodiment of the present invention;

[0057] Figure 3 This is a schematic diagram of the structure of a computing device provided in an embodiment of the present invention. Detailed Implementation

[0058] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0059] In some of the processes described in the specification, claims, and accompanying drawings of this invention, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to different types.

[0060] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0061] To address the industry pain point that traditional methods in textile auxiliary production cannot accurately control auxiliary wetting behavior in real time, existing technical solutions based on hyperspectral imaging and fluid dynamics simulation have three key drawbacks: first, they can only acquire static surface energy distribution and cannot capture the dynamic changes in substrate surface energy caused by plasma resonance effects; second, they lack dynamic coupling analysis of auxiliary flow state and substrate surface properties; and third, the characterization of auxiliary molecular structure characteristics is disconnected from process parameter optimization. To address these issues, this invention provides a data processing method for textile auxiliary production. Figure 1 A flowchart of a data processing method for textile auxiliary agent production is provided as an embodiment of the present invention, such as... Figure 1As shown, the method includes:

[0062] Step 101: Monitor the solution flow state of textile auxiliaries and generate a dynamic gradient data set.

[0063] In this step, the dynamic gradient data set refers to the set of flow state parameters of the textile auxiliary agent solution collected in real time by the fluid sensor, including three-dimensional spatiotemporal distribution data such as velocity gradient, pressure gradient and temperature gradient.

[0064] In this embodiment of the invention, a high-precision fluid sensor is used to monitor in real time the flow velocity, pressure distribution, and liquid-solid interfacial tension changes of the textile auxiliary agent solution on the surface of the substrate to be treated, and collect multi-dimensional physical parameters to form an original monitoring data set; a sliding time window Fourier transform algorithm is used to perform time-frequency analysis on the original data to extract the flow state change gradient of each monitoring point within a preset time period, including the velocity gradient, pressure gradient, and tension gradient; the three types of gradient data are normalized and fused according to spatial coordinate position to finally generate a dynamic gradient data set containing spatiotemporal dimension information.

[0065] Step 102: Obtain the set of reflection signal parameters of the surface of the textile substrate to be treated under electromagnetic excitation in a preset frequency band, and construct an energy distribution map based on the resonance frequency offset data associated with the plasma resonance effect in the set of reflection signal parameters.

[0066] In this step, the textile substrate to be treated specifically refers to the textile material to be treated, such as fibers and fabrics. Its surface characteristics, such as roughness, polarity, and porosity, directly affect the wettability and effectiveness of the auxiliaries, including fibers, fabrics, or combinations thereof. The set of reflected signal parameters refers to the set of parameters such as the intensity and phase difference of the reflected signal returned by the substrate surface under electromagnetic wave irradiation in a specific frequency band; the resonant frequency offset data refers to the characteristic frequency offset in the reflected signal caused by the plasma resonance effect, reflecting the surface energy state of the substrate; the energy distribution map refers to a two-dimensional distribution map characterizing the energy intensity of different regions on the substrate surface.

[0067] In this embodiment of the invention, a frequency-tunable microwave transmitter projects scanning electromagnetic waves in the 0.5-3 THz frequency band onto the surface of the textile substrate to be treated. The reflected signals are received by a quantum well detector and demodulated to obtain the amplitude-phase parameter set at each frequency point. Using a plasma resonance feature extraction algorithm, the resonance frequency shift data at the resonance frequency point is identified from the reflected signal parameter set. The shift amount is nonlinearly positively correlated with the surface free energy of the substrate to be treated. Based on a spatial interpolation algorithm, the resonance frequency shift data at each detection point is mapped and calculated with the surface energy calibration curve to generate a partitioned surface energy intensity numerical matrix with a resolution unit of 50 μm. Finally, an energy distribution map reflecting the dynamic energy distribution of the surface of the substrate to be treated is constructed.

[0068] Step 103: Dynamically couple the dynamic gradient data set with the surface energy characteristics of each partition of the energy distribution map to generate a spatiotemporal evolution data set associated with the infiltration behavior.

[0069] In this step, the spatiotemporal evolution dataset refers to a multidimensional dataset that characterizes the dynamic changes in the adjuvant wetting process, including parameters such as spatiotemporal coordinates, wetting depth, and diffusion rate.

[0070] In this embodiment of the invention, the surface energy intensity value of each partition in the energy distribution spectrum is multiplied with the plasma resonance frequency offset value of the corresponding partition. The dynamic correlation factor corresponding to each partition is generated by tensor multiplication, resulting in a correlation factor set. Simultaneously, based on the temporal fluctuation amplitude of the dynamic gradient data set, the difference data of the temporal fluctuation amplitude is calculated to generate a weight coefficient set. An adaptive weighted fusion algorithm is used to superimpose the correlation factor set and the weight coefficient set point by point to form a coupling parameter matrix reflecting the interaction strength between the additive and the substrate. Finally, a three-dimensional convolutional neural network is used to extract spatiotemporal features from the coupling parameter matrix, outputting a spatiotemporal evolution data set containing multi-dimensional parameters such as the wetting front position and the penetration rate.

[0071] Step 104: Extract the infiltration dynamic fluctuation data from the spatiotemporal evolution data set, and generate a set of latent variables characterizing the molecular structure properties of textile auxiliaries based on the infiltration dynamic fluctuation data.

[0072] In this step, the wetting dynamic fluctuation data refers to the high-frequency variation components extracted from the spatiotemporal evolution data, which reflect the microscopic interaction between the additive molecules and the substrate; the set of latent variables refers to the set of latent variables that characterize the polarity of the additive molecules or the surface energy compatibility of the substrate.

[0073] In this embodiment of the invention, a set of wave components characterizing the unsteady-state characteristics of the infiltration process is separated from the spatiotemporal evolution data set. These components include parameters such as the amplitude of local flow velocity oscillations and the frequency of interfacial tension fluctuations. Each wave component in the set of wave components is subjected to phase synchronization processing, and the synchronized wave component sequence is mapped to a preset multidimensional feature space to generate a target feature space. The weight distribution of the projected coordinates in the target feature space is adjusted according to the order of magnitude of the energy characteristics of the partitioned surface to generate a set of latent variables characterizing the molecular structure characteristics.

[0074] Step 105: Generate a target parameter dataset based on the matching relationship between the set of latent variables and the preset wetting rate constraint, so as to perform data processing for textile auxiliary production based on the target parameter dataset.

[0075] In this step, the preset wetting rate constraint refers to the allowable range and fluctuation tolerance of the wetting rate set in advance according to the production process requirements, such as a maximum rate of 50 mm / s and a minimum rate of 20 mm / s; the target parameter dataset refers to the optimized set of process control parameters.

[0076] In this embodiment of the invention, the features of each dimension in the latent variable set are matched and analyzed with the preset infiltration rate constraint to generate an intermediate parameter sequence; a target parameter sequence that meets the preset conflict resolution threshold condition is selected from the intermediate parameter sequence to generate a parameter optimization rule set; based on the parameter optimization rule set, the rule-driven parameters of the latent variable set are transformed to generate a target parameter dataset, and data processing for textile auxiliary agent production is performed accordingly.

[0077] For example, in an experiment using polyester fabric treated with auxiliaries as the test sample, a sensor array monitored the solution flow state of the textile auxiliaries at a sampling frequency of 100 Hz. The measured pressure gradient at the inlet was 5 Pa / mm, the velocity gradient in the central region was 0.2 m / s / cm, and the temperature gradient at the edge was 0.5 ℃ / mm. After processing, a dynamic gradient data set containing 200×200 grid points was generated, providing a basis for subsequent analysis. An electromagnetic excitation of 1.2 THz was applied to the polyester fabric, and a set of reflected signal parameters was acquired. This showed that the measured reflected signal exhibited a resonance peak at 1.18 THz, offset by 20 GHz. After processing, an energy distribution spectrum with a resolution of 500 dpi was generated, showing that the surface energy was higher at the textured areas of the polyester fabric (e.g., 120 mJ / m²) and lower at the pores (e.g., 80 mJ / m²). The dynamic gradient data set was then compared with the energy distribution spectrum. The surface energy characteristics of the partitioned spectral regions are dynamically coupled to generate a spatiotemporal evolution dataset, which shows high surface energy regions of polyester fabrics, such as the 150 mJ / m² additive wetting rate reaching 0.3 mm / s, which is twice that of low energy regions, such as 90 mJ / m². In addition, when processing silicone oil additives, the wetting dynamic fluctuation data in the spatiotemporal evolution dataset is extracted to generate a latent variable set, which shows that the third dimension value is relatively high, such as 0.82, corresponding to the characteristic vibration mode of silicon-oxygen bonds. Based on the matching relationship between the latent variable set and the preset wetting rate constraint, the target parameter dataset is generated. For example, for silicone oil additives with a latent variable set of [0.82, 0.15, ...], the system will recommend using processing conditions of 50℃ and 0.2 MPa to stabilize the wetting rate within the range of 0.25 ± 0.03 mm / s.

[0078] This invention overcomes the limitations of traditional static detection and isolated analysis by dynamically monitoring the coupling changes between the flow state of the additive and the surface energy distribution of the substrate; it utilizes the plasma resonance effect to capture the dynamic characteristics of surface energy in real time, solving the response lag problem of existing optical detection technologies; it analyzes molecular structure characteristics through wetting fluctuation data to achieve precise matching between process parameters and molecular-level characteristics; and the target parameter dataset improves the wetting uniformity of different batches of additives, reducing the defect rate.

[0079] This invention provides a specific embodiment. Step 103 involves dynamically coupling the dynamic gradient data set with the surface energy characteristics of each partition of the energy distribution map to generate a spatiotemporal evolution data set associated with infiltration behavior. This specifically includes the following steps:

[0080] Step 301: Multiply the surface energy intensity value of each partition in the energy distribution map with the plasma resonance frequency offset value of the corresponding partition to generate a set of correlation factors corresponding to the spatial location of each partition.

[0081] In this step, the surface energy intensity value refers to the quantified value of the surface energy of the partition extracted from the energy distribution spectrum, reflecting the adsorption capacity of the substrate surface for the auxiliary agent; the plasma resonance frequency offset value refers to the offset of the plasma resonance frequency of the substrate surface relative to the reference value, characterizing the dynamic change of surface energy; the correlation factor set refers to the set of parameters generated by the product of surface energy intensity and frequency offset, quantifying the interaction strength between the textile substrate and textile auxiliary agent to be treated.

[0082] In this embodiment of the invention, the surface energy intensity value and frequency offset value are extracted for each partition. For example, the surface energy intensity value is extracted by quantization of pixel values ​​in the energy distribution map, and the frequency offset value is calculated by reflection signal spectrum analysis. The two are multiplied to obtain the correlation factor of the partition. The correlation factors of each partition are combined to finally form a set of correlation factors bound to spatial location.

[0083] Step 302: Monitor the temporal fluctuation amplitude of the dynamic gradient data set, and calculate the difference of the temporal fluctuation amplitude according to the preset tension value of adjacent time nodes, so as to generate a set of weight coefficients bound to the spatial location of each partition.

[0084] In this step, the temporal fluctuation amplitude refers to the fluctuation of the liquid-solid interfacial tension in the dynamic gradient data over time; the preset tension value of adjacent time nodes refers to the liquid-solid interfacial tension measurement value collected at fixed time intervals; the weight coefficient set refers to the parameter set generated based on the difference of the temporal fluctuation amplitude, reflecting the dynamic weight of the flow state.

[0085] In this embodiment of the invention, the temporal fluctuation amplitude of the dynamic gradient data set is monitored. By calculating the difference data between the tension values ​​at preset adjacent time nodes (such as t1 and t2), for example, Δt = t2 tension value - t1 tension value, a set of weight coefficients bound to the spatial location of each partition is generated. Specifically, the weight coefficients are adjusted according to the proportional relationship between the difference and the preset benchmark value, and the weighted coefficients are associated with the corresponding partition coordinates to generate a set of weight coefficients bound to the spatial location of each partition.

[0086] Step 303: Overlay the set of correlation factors and the set of weight coefficients point by point according to the spatial position corresponding to each partition to generate a coupling parameter matrix.

[0087] In this step, the spatial location corresponding to each partition refers to the coordinate position of the region divided in the energy distribution map; the coupling parameter matrix characterizes the interaction intensity between flow and surface energy.

[0088] In this embodiment of the invention, the correlation factor and weight coefficient of each spatial location are weighted and summed to obtain the coupling parameter corresponding to each spatial location. Finally, the coupling parameters corresponding to all spatial locations are combined to form a coupling parameter matrix that reflects the interaction between the flow of the additive and the energy of the substrate surface.

[0089] Step 304: Within a preset time slice sequence, perform energy redistribution on the matrix units in the coupling parameter matrix that satisfy the phase synchronization condition to generate the redistributed coupling parameter matrix.

[0090] In this step, the preset time slice sequence refers to the sequence of data processing units divided according to fixed time intervals; the phase synchronization condition refers to the matching rule between the pulse trigger interval and the infiltration fluctuation period in the temporal coding mechanism of spiking neurons; and the energy redistribution refers to the operation of adjusting the energy value of the matrix unit according to the gradient direction.

[0091] In this embodiment of the invention, within a preset time slice sequence, the energy values ​​of the units that meet the phase condition in the coupling parameter matrix are redistributed according to the gradient direction, for example, according to the diffusion from the high energy region to the low energy region, to generate the redistributed coupling parameter matrix.

[0092] Step 305: Based on the allocated coupling parameter matrix, select a target matrix unit, and based on the target matrix unit, generate a spatiotemporal evolution data set associated with the infiltration behavior.

[0093] In this step, the target matrix element refers to the element in the coupling parameter matrix where the intensity of the liquid-solid interface interaction exceeds the threshold; the spatiotemporal evolution dataset refers to the additive permeation path trajectory data concatenated along the time dimension.

[0094] In this embodiment of the invention, based on the allocated coupling parameter matrix, a target matrix unit, namely the unit where the liquid-solid interface interaction intensity exceeds a threshold, is selected. According to the spatial coordinates of the target matrix unit and the preset time slice sequence, a spatiotemporal evolution data set associated with the infiltration behavior is generated by concatenating them along the time dimension.

[0095] This invention improves the accuracy of wetting path prediction by dynamically coupling the surface energy characteristics of the substrate with the flow state of the auxiliaries, and by combining temporal fluctuation weights and phase synchronization optimization. It solves the modeling deviation problem caused by neglecting the dynamic effect of plasma resonance and temporal fluctuation characteristics in the prior art, and achieves precise control of the compatibility between textile auxiliaries and substrates.

[0096] This invention provides a specific embodiment, step 305, which involves selecting a target matrix unit based on the allocated coupling parameter matrix, and generating a spatiotemporal evolution data set associated with infiltration behavior based on the target matrix unit, specifically including the following steps:

[0097] Step 351: Mark the target matrix units in the allocated coupling parameter matrix that exceed the preset gradient value, take the target matrix units as candidate units, and extract the spatial coordinate points corresponding to the candidate units to generate an initial coordinate point set.

[0098] In this step, the allocated coupling parameter matrix refers to the matrix after energy redistribution, and each cell value reflects the intensity of the liquid-solid interface interaction; the preset gradient value refers to the threshold used to screen the interaction region.

[0099] In this embodiment of the invention, by traversing the gradient values ​​of all matrix units in the allocated coupling parameter matrix, matrix units exceeding a preset gradient value are marked as target matrix units, and these target matrix units are used as candidate units. For example, when the preset gradient value is 5.0, all target matrix units with gradient values ​​≥ 5.0 are selected as candidate units. Subsequently, the three-dimensional spatial coordinate points (x, y, z) corresponding to each candidate unit are extracted to generate an initial coordinate point set. The core of this step is to filter out the regions of liquid-solid interface interaction intensity through gradient thresholding, providing a data foundation for subsequent path analysis.

[0100] Step 352: Perform bidirectional path verification between the spatial coordinate points in the initial coordinate point set and the first target point in the adjacent time slice in the preset time slice sequence to generate a verified coordinate point set.

[0101] In this step, the first target point refers to the candidate point to be verified in the adjacent time slice; bidirectional path verification means that the current point and the adjacent points must satisfy the connection condition; the set of coordinate points after verification refers to the set of spatial coordinate points that are retained through bidirectional path verification and satisfy the continuity of gradient direction, representing the starting point of the path.

[0102] In this embodiment of the invention, for coordinate point A of the current time slice, a point B that is spatially closest and has the same direction of liquid-solid interface tension gradient is searched in the adjacent slices. If it passes bidirectional verification, it is retained, such as A to B and B to A both pass verification; otherwise, it is discarded. Finally, a set of verified coordinate points is generated to ensure path continuity.

[0103] Step 353: Based on the liquid-solid interfacial tension value corresponding to each spatial coordinate point in the verified coordinate point set, assign processing priorities to the spatial coordinate points in the verified coordinate point set to obtain the target spatial coordinate point priority sequence.

[0104] In this step, processing priority refers to the path processing order weight assigned based on the liquid-solid interfacial tension value; target spatial coordinate point priority sequence refers to the list of target spatial coordinate points sorted by priority, such as [point A(1), point B(2)].

[0105] In this embodiment of the invention, based on the liquid-solid interfacial tension values ​​corresponding to each spatial coordinate point in the verified coordinate point set, processing priorities are assigned to each spatial coordinate point in descending order. Specifically, the higher the tension value, the higher the priority. For example, if point A has a tension value of 8.5 and point B has a tension value of 6.2, then point A has a priority of 1 and point B has a priority of 2, forming a priority sequence of target spatial coordinate points to guide the path connection order.

[0106] Step 354: Based on the priority sequence of the target spatial coordinate points, connect the verified coordinate points in the set of verified coordinate points within the same time slice with the second target point in the adjacent time slice to generate an initial spatiotemporal connection network.

[0107] In this step, the same time slice refers to the data set within the same time period, such as slice t1 containing data from 0 to 1 second; the second target point refers to the verified connection endpoint in the adjacent slice; the initial spatiotemporal connection network refers to the preliminary constructed path topology graph.

[0108] In this embodiment of the invention, according to the priority sequence of target spatial coordinate points, the verified coordinate points within the same time slice (e.g., t1) are sequentially connected to the second target point in the adjacent time slice (e.g., t2) that has the closest spatial distance and the same gradient direction. For example, point A with priority 1 is preferentially connected to the nearest matching point C in the adjacent slice, generating an initial spatiotemporal connection network that contains all connection paths.

[0109] Step 355: Detect the branch nodes in the initial spatiotemporal connection network to generate an optimized spatiotemporal connection network.

[0110] In this step, a branch node refers to a coordinate point that connects at least two paths; the optimized spatiotemporal connection network refers to the final path topology after removing conflicting paths and reconstructing the connection relationships.

[0111] In this embodiment of the invention, conflict detection is performed on each branch node in the initial spatiotemporal connection network. Specifically, this includes: calculating the change range of the liquid-solid interfacial tension value of the associated path of the branch node; if the change range of a certain path exceeds a preset change threshold, such as 30%, then the path is removed; if the branch node is left with only a single path after removal, it is degenerated into an ordinary node, and finally an optimized spatiotemporal connection network is generated to ensure that the path conforms to physical laws.

[0112] Step 356: Connect the spatial coordinates of the verified first target point and the verified second target point in the optimized spatiotemporal connection network to generate a spatiotemporal evolution data set associated with the infiltration behavior.

[0113] In this step, the first target point after verification refers to the adjacent slice coordinate points that are filtered and retained through step 352; the second target point after verification refers to the adjacent slice coordinate points connected through step 354; the spatiotemporal evolution dataset refers to the structured dataset that represents the spatiotemporal trajectory of infiltration behavior.

[0114] In this embodiment of the invention, the coordinate points in the optimized spatiotemporal connection network are connected in chronological order to form a spatiotemporal evolution data set, which includes the spatial coordinates, timestamps, and intensity of the infiltration path.

[0115] The embodiments of the present invention solve the problem of prediction deviation of the wetting path caused by ignoring the dynamic effect of plasma resonance and the temporal fluctuation characteristics in the prior art; improve the compatibility of textile auxiliaries with the substrate to be treated, and realize precise closed-loop control of process parameters.

[0116] This invention provides a specific embodiment. Step 353 involves assigning processing priorities to the spatial coordinate points in the verified coordinate point set based on the liquid-solid interfacial tension values ​​corresponding to each spatial coordinate point, thereby obtaining a priority sequence of target spatial coordinate points. This specifically includes the following steps:

[0117] Step 531: Extract the liquid-solid interfacial tension values ​​of each spatial coordinate point in the verified coordinate point set.

[0118] In this step, the liquid-solid interfacial tension value refers to a quantitative parameter that reflects the intensity of the interaction between the additive solution and the substrate interface, and is derived from the unit values ​​of the coupling parameter matrix.

[0119] In this embodiment of the invention, each spatial coordinate point in the set of verified coordinate points is traversed, and the liquid-solid interfacial tension value stored in the corresponding matrix unit is directly read according to its position index in the allocated coupling parameter matrix.

[0120] Step 532: Based on the liquid-solid interfacial tension value, sort the spatial coordinate points in the verified coordinate point set to generate a sorting result.

[0121] In this step, the sorting result refers to the sequential list generated by arranging all spatial coordinate points in the verified coordinate point set from high to low according to the liquid-solid interfacial tension value.

[0122] In this embodiment of the invention, a quick sorting algorithm is used to sort the liquid-solid interfacial tension values ​​in descending order. The sorting position of each spatial coordinate point is directly related to the magnitude of its tension value. The sorting result is generated as follows: the coordinate point with the highest liquid-solid interfacial tension value is ranked first, and the rest are ranked in descending order.

[0123] Step 533: Based on the sorting results, assign priority weights to the spatial coordinate points in the verified coordinate point set, and generate a priority sequence corresponding to the verified coordinate point set.

[0124] In this step, priority weight refers to the weight coefficient assigned based on the position of the spatial coordinate points in the sorting result, which is used to characterize the processing priority; the priority sequence corresponding to the set of verified coordinate points refers to an ordered list containing spatial coordinate points and their priority weights.

[0125] In this embodiment of the invention, a linear weight allocation rule is adopted to assign priority weights to spatial coordinate points in the set of verified coordinate points. Specifically, the weight coefficient of the first spatial coordinate point is set to 1, and the weight coefficients of subsequent points are calculated according to the formula weight = 1 - (current position / total number of points). The final generated priority sequence is a list containing all spatial coordinate points and their corresponding weight coefficients.

[0126] Step 534: Match the priority sequence with the spatial distribution density of the verified coordinate point set to generate a spatial coordinate point priority sequence, enhance the priority of spatial coordinate points in the spatial coordinate point priority sequence whose spatial distribution density is higher than a preset threshold, and generate a target spatial coordinate point priority sequence.

[0127] In this step, spatial distribution density refers to the number of verified coordinate points per unit area; preset threshold refers to the density threshold that triggers priority enhancement, such as 5 points / cm²; priority refers to the priority level of the coordinate points in the processing flow; target spatial coordinate point priority sequence refers to the final priority list after density correction.

[0128] In this embodiment of the invention, the specific matching process is as follows: the weight coefficients in the priority sequence are combined with the spatial distribution density, and a weighted superposition rule is adopted: if the density value of a certain region exceeds a preset threshold, such as 5 points / cm², then the weight coefficients of all spatial coordinate points in that region are multiplied by the density enhancement coefficient, such as 1.2, and the final target spatial coordinate point priority sequence is generated.

[0129] The embodiments of the present invention solve the problem of neglecting local high-density areas in the existing technology by dynamically allocating priority weights and integrating spatial distribution density, thereby improving the penetration efficiency of textile auxiliaries in high-energy areas on the substrate surface and avoiding resource waste.

[0130] This invention provides a specific embodiment, step 355, which involves detecting branch nodes in the initial spatiotemporal connectivity network to generate an optimized spatiotemporal connectivity network, specifically including the following steps:

[0131] Step 551: Extract the liquid-solid interfacial tension values ​​on all connection paths of each branch node in the initial spatiotemporal connection network, and calculate the variation range of the liquid-solid interfacial tension values ​​between adjacent time slices. The branch node is defined as a spatial coordinate point where at least two different connection paths intersect.

[0132] In this step, all connection paths refer to the multiple spatiotemporal trajectory segments associated with each branch node in the spatiotemporal connection network; the change amplitude of the liquid-solid interface tension value refers to the relative change rate of the liquid-solid interface tension value between adjacent time slices, which is used to quantify the intensity of dynamic fluctuations; branch nodes refer to nodes that connect at least two paths in the spatiotemporal network, representing the bifurcation or intersection of the immersion path.

[0133] In this embodiment of the invention, all branch nodes in the initial spatiotemporal connection network are first traversed. For each branch node, the liquid-solid interfacial tension values ​​recorded on all its connection paths are extracted. Subsequently, the change range of liquid-solid interfacial tension values ​​between adjacent time slices is calculated. The specific formula is: Change range = (liquid-solid interfacial tension value of the current time slice - liquid-solid interfacial tension value of the previous time slice) / liquid-solid interfacial tension value of the previous time slice × 100%. This step identifies conflict paths that may be caused by substrate surface defects or abnormal flow of additives by quantifying the dynamic fluctuation of tension.

[0134] Step 552: Select a first target branch node from the branch nodes that has a connection path with a change amplitude exceeding a preset change threshold, select a second target branch node that meets preset conditions from the first target branch node, and degenerate the first target branch node that does not meet the preset conditions into a normal node. The preset conditions are that the direction of the liquid-solid interface tension gradient is consistent with the direction of the surface energy distribution of the textile substrate to be treated, and that there are at least two different connection paths.

[0135] In this step, the first target branch node refers to the branch node with conflicting paths (the change in liquid-solid interfacial tension value exceeds the preset change threshold) that has been initially screened; the second target branch node refers to the branch node that has been retained after directional consistency verification; the direction of the liquid-solid interfacial tension gradient refers to the trend of the liquid-solid interfacial tension value over time, such as increasing / decreasing; the direction of the surface energy distribution of the textile substrate to be treated refers to the spatial trend of the surface energy intensity of the substrate to be treated, such as the diffusion direction from the high energy region to the low energy region, which is determined by the increasing / decreasing trend of the surface energy intensity values ​​of adjacent partitions in the energy distribution map.

[0136] In this embodiment of the invention, firstly, branch nodes belonging to connection paths whose change amplitude exceeds a preset change threshold (e.g., 30%) are selected as first target branch nodes; further, second target branch nodes that meet preset conditions are selected from the first target branch nodes, the conditions including: 1. The direction of the liquid-solid interface tension gradient is consistent with the direction of energy distribution on the substrate surface; 2. The node still retains at least two different connection paths; first target branch nodes that do not meet the conditions (e.g., only a single path remains or the direction is inconsistent) are degenerated into ordinary nodes and lose their branch attributes. The consistency determination method between the direction of the liquid-solid interface tension gradient and the direction of energy distribution on the substrate surface is as follows: the direction of energy distribution on the substrate surface is determined by the increasing or decreasing trend of surface energy intensity values ​​of adjacent partitions in the energy distribution spectrum; the direction of the liquid-solid interface tension gradient is determined by the trend of tension value change at the same spatial coordinate point within adjacent time slices; when the angle between the two trend directions is less than a preset angle, they are determined to be consistent.

[0137] Step 553: Reconstruct the connection relationship between the second target branch node and the ordinary nodes associated with the second target branch node to generate the optimized spatiotemporal connection network.

[0138] In this step, the ordinary node associated with the second target branch node refers to the non-branch node that is directly connected to the filtered second target branch node.

[0139] In this embodiment of the invention, for the second target branch node, the spatiotemporal path connection relationship between it and the associated ordinary node is reconstructed; and the spatial distance is calculated and verified to be less than a preset angle with the energy direction of the substrate surface, retaining the connection path that conforms to the physical law, and finally generating an optimized spatiotemporal connection network. This network improves the accuracy of infiltration behavior prediction by pruning abnormal paths and strengthening paths.

[0140] The embodiments of the present invention solve the problem of prediction deviation of the wetting path caused by ignoring the sudden change of liquid-solid interfacial tension and the adaptability of substrate surface energy in the existing solution by dynamically screening conflict paths, verifying directional consistency and node degradation mechanism. This improves the modeling accuracy of the penetration path of textile auxiliaries on complex substrate surfaces and reduces the risk of production anomalies caused by surface defects or process fluctuations.

[0141] This invention provides a specific embodiment. Step 104 involves generating a set of latent variables characterizing the molecular structure properties of textile auxiliaries based on the wetting dynamic fluctuation data. This specifically includes the following steps:

[0142] Step 401: Analyze the infiltration dynamic fluctuation data to generate a set of fluctuation components at different time scales.

[0143] In this step, the set of fluctuation components refers to the set of fluctuation feature data with different time resolutions obtained through multi-scale decomposition.

[0144] In this embodiment of the invention, the infiltration dynamic fluctuation data is decomposed into multiple scales, and the fluctuation features at different time resolutions are extracted by time-frequency analysis algorithm to generate a set of fluctuation components containing high-frequency, mid-frequency and low-frequency components. The specific process is as follows: the infiltration dynamic fluctuation data is input into the wavelet transform module, and by adjusting the time window length and basis function parameters, the fluctuation components that characterize short-term abrupt changes, medium-term trends and long-term steady states are separated to form a set of fluctuation components with hierarchical characteristics.

[0145] Step 402: Perform phase synchronization processing on each wave component in the wave component set to generate a synchronized wave component sequence.

[0146] In this step, phase synchronization processing refers to the signal processing process of aligning the phases of wave components at different time scales to make their peaks and troughs consistent on the time axis; the synchronized wave component sequence refers to the set of wave components that have undergone phase alignment processing and have temporal continuity.

[0147] In this embodiment of the invention, phase synchronization processing is performed on each component in the set of wave components, and the phase offset of components at different time scales is adjusted by using a spiking neuron temporal coding mechanism. The specific process is as follows: each wave component is input into the spiking neural network layer, and according to the preset phase synchronization threshold, that is, the ratio of the period length of the previous wave component to the period length of the next wave component is less than the set threshold, the peak and trough positions of each wave component are aligned to generate a temporally coherent synchronized wave component sequence.

[0148] Step 403: Map the synchronized fluctuation component sequence to a preset multidimensional feature space to generate a target feature space.

[0149] In this step, the preset multidimensional feature space refers to a predefined mathematical space, with each dimension corresponding to a type of molecular bonding energy (such as hydrogen bond energy or van der Waals energy); the target feature space refers to the high-dimensional vector space generated by the wave component sequence after mapping synchronization.

[0150] In this embodiment of the invention, the synchronized wave component sequence is mapped to a preset multidimensional feature space, and the correlation between the wave component and molecular properties is established through nonlinear projection relationship. The specific process is as follows: the time-domain features of the synchronized wave component sequence are converted into vector coordinates in a high-dimensional space using a kernel function, with each coordinate axis corresponding to a type of molecular bonding energy (such as hydrogen bond energy or van der Waals energy), forming a target feature space.

[0151] Step 404: Adjust the weight distribution of the projected coordinates in the target feature space according to the order of magnitude of the energy features of the partitioned surface to obtain the adjusted multidimensional feature space.

[0152] In this step, the order of magnitude of the partitioned surface energy characteristics refers to the numerical range of the surface energy intensity of each partition in the energy distribution spectrum (e.g., 0~100 units); the weight distribution of the projected coordinates refers to the characterization weight of each coordinate axis in the multidimensional feature space for molecular properties; the adjusted multidimensional feature space refers to the feature space after adjusting the weights according to the surface energy intensity.

[0153] In this embodiment of the invention, the surface energy intensity values ​​(energy value per unit area) of each partition in the energy distribution map are normalized and multiplied with the coordinate axes of the target feature space (coordinate weight = original weight × surface energy intensity / reference energy value) to match the latent variable dimension with the surface energy space resolution of the substrate, thereby generating an adjusted multidimensional feature space.

[0154] Step 405: Extract the weight parameters of each projected coordinate in the adjusted feature space to generate a set of latent variables characterizing the molecular structure properties of textile auxiliaries.

[0155] In this step, the weight parameter of each projected coordinate refers to the weight value of each coordinate axis in the adjusted feature space.

[0156] In this embodiment of the invention, the weight parameters of each projected coordinate in the adjusted feature space are extracted, and redundant dimensions are compressed through principal component analysis. The specific process is as follows: calculate the variance contribution rate of the weight parameters of each coordinate axis, retain the coordinate axes whose contribution rates exceed the threshold, and aggregate their weight parameters according to spatial position to generate a set of latent variables characterizing the molecular structure of textile auxiliaries.

[0157] The embodiments of the present invention solve the problem of poor process parameter adaptability caused by ignoring the correlation between the dynamic changes of substrate surface energy and molecular structure in the existing solutions; it realizes coupled modeling of wetting dynamic fluctuation data and molecular-level characteristics, improves the optimization accuracy of textile auxiliary formulations, and enables process parameters to achieve synergistic matching with substrate surface energy distribution and auxiliary molecular structure.

[0158] This invention provides a specific embodiment. Step 105 involves generating a target parameter dataset based on the matching relationship between the latent variable set and the preset wetting rate constraint, and then performing data processing for textile auxiliary agent production based on the target parameter dataset. This specifically includes the following steps:

[0159] Step 501: Perform multi-level matching between each latent variable in the set of latent variables and the benchmark parameter range in the preset infiltration rate constraint to generate an intermediate parameter sequence with hierarchical correlation.

[0160] In this step, the reference parameter range refers to the range of process parameters corresponding to the constraint conditions, such as the viscosity range [2.5 Pa·s, 4.0 Pa·s]; multi-level matching refers to the process of comparing latent variables with multiple reference ranges in order of priority; hierarchical association refers to the priority or dependency relationship between different matching levels, such as high-precision levels taking precedence over regular levels; intermediate parameter sequence refers to the temporary dataset generated during the matching process that contains latent variables, matching levels, and deviations.

[0161] In this embodiment of the invention, firstly, based on a preset wetting rate constraint, multiple reference parameter intervals are defined, each interval corresponding to a different process requirement level, such as high precision, normal, and relaxed. Then, each latent variable is matched with each reference parameter interval, and the ratio of the latent variable to the interval boundary is calculated, such as the latent variable divided by the upper limit of the interval. The matching level is determined based on whether the ratio falls within the interval range. Finally, an intermediate parameter sequence is generated according to the matching level order, such as from high to low or according to priority, wherein each sequence entry contains a latent variable identifier, a matching level, and a parameter deviation.

[0162] Step 502: Select a target parameter sequence that meets the preset conflict resolution threshold condition from the intermediate parameter sequence, and generate a parameter optimization rule set based on the target parameter sequence;

[0163] In this step, the preset conflict resolution threshold refers to the screening conditions used to resolve multi-parameter conflicts, such as parameter deviation <10% or priority ≥3; the parameter optimization rule set refers to the logical set containing the direction and magnitude of parameter adjustment, such as viscosity increase of 15% or frequency decrease of 5kHz.

[0164] In this embodiment of the invention, a conflict resolution threshold is first set, such as the parameter deviation being less than 10% or the hierarchical priority being higher than a preset level. The intermediate parameter sequences are traversed, and sequence entries that meet the threshold conditions are selected to generate the target parameter sequence. For target parameter sequences with conflicts, such as multiple latent variables matching the same baseline interval but with opposite parameter deviation directions, a weighted fusion strategy is adopted, such as latent variables with higher priority accounting for 60% of the weight and those with lower priority accounting for 40%, to generate parameter optimization rules. Finally, all the selected and fused rules are integrated into a parameter optimization rule set according to the process logic, which includes the parameter adjustment direction (such as increasing / decreasing the concentration of additives) and the adjustment range.

[0165] Step 503: Based on the parameter optimization rule set, transform the rule-driven parameters of the latent variable set to generate the target parameter dataset;

[0166] In this step, rule-driven parameters refer to process parameters that need to be dynamically adjusted according to optimization rules, such as additive viscosity, temperature, and electromagnetic excitation frequency.

[0167] In this embodiment of the invention, each rule in the parameter optimization rule set is first parsed to determine the rule-driven parameters that need to be adjusted; then, according to the adjustment direction and magnitude defined in the parameter optimization rule set, the corresponding rule-driven parameters in the latent variable set are numerically transformed (e.g., the original viscosity value is multiplied by the adjustment coefficient 1.2); finally, the transformed rule-driven parameters are classified and integrated according to process type (e.g., wetting rate optimization, molecular structure adaptation) to generate a target parameter dataset containing process parameter key-value pairs.

[0168] Step 504: Perform data processing for textile auxiliary production based on the target parameter dataset.

[0169] In this step, data processing refers to a closed-loop process that controls production equipment based on target parameters and provides real-time feedback for optimization.

[0170] In this embodiment of the invention, the target parameter dataset is input into the production control system, and the process parameter key-value pairs (such as electromagnetic excitation frequency = 120kHz, textile auxiliary agent concentration = 5.8%) are parsed. Based on the process parameter key-value pairs, the operating parameters of the production equipment are adjusted by the control module (such as adjusting the output frequency of the electromagnetic generator and modifying the flow rate of the injection pump). The wetting behavior data generated after the adjustment is monitored in real time and compared with the preset constraints. If the deviation exceeds the limit, the feedback mechanism is triggered to re-execute steps 501-503, forming a closed-loop optimization.

[0171] This invention achieves dynamic adaptation between the molecular properties of textile auxiliaries and the surface energy of the substrate through a multi-level matching and conflict resolution mechanism, solving the problem of uneven wetting caused by static parameter settings in the prior art; through rule-driven parameter conversion and closed-loop feedback, it improves the accuracy of process adjustment and production efficiency, filling the technical gap that traditional methods cannot optimize molecular-level properties in real time.

[0172] Figure 2 This invention provides a schematic diagram of a data processing system for textile auxiliary agent production, as shown in the embodiment of the invention. Figure 2 As shown, the system includes:

[0173] Monitoring module 21 is used to monitor the solution flow state of textile auxiliaries and generate a dynamic gradient data set;

[0174] The construction module 22 is used to acquire the set of reflection signal parameters of the surface of the textile substrate to be treated under electromagnetic excitation in a preset frequency band, and to construct an energy distribution map based on the resonance frequency shift data associated with the plasma resonance effect in the set of reflection signal parameters.

[0175] The coupling module 23 is used to dynamically couple the dynamic gradient data set with the surface energy characteristics of each partition of the energy distribution map to generate a spatiotemporal evolution data set associated with the infiltration behavior.

[0176] Extraction module 24 is used to extract the infiltration dynamic fluctuation data in the spatiotemporal evolution data set, and generate a set of latent variables characterizing the molecular structure properties of textile auxiliaries based on the infiltration dynamic fluctuation data.

[0177] The processing module 25 is used to generate a target parameter dataset based on the matching relationship between the set of latent variables and the preset wetting rate constraint, so as to perform data processing for textile auxiliary production based on the target parameter dataset.

[0178] Figure 2 The aforementioned data processing system for textile auxiliary production can execute... Figure 1 The implementation principle and technical effects of the data processing method for textile auxiliary agent production described in the illustrated embodiment will not be repeated here. The specific methods by which each module and unit of the data processing system for textile auxiliary agent production in the above embodiments perform operations have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0179] In one possible design, Figure 2 The data processing system for textile auxiliary production shown in the embodiment can be implemented as a computing device, such as... Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0180] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 32.

[0181] The processing component 32 is used to: monitor the solution flow state of textile auxiliaries and generate a dynamic gradient data set; acquire a set of reflection signal parameters of the surface of the textile substrate to be treated under electromagnetic excitation in a preset frequency band, and construct an energy distribution map based on the resonance frequency offset data associated with the plasma resonance effect in the set of reflection signal parameters; dynamically couple the dynamic gradient data set with the surface energy characteristics of each partition of the energy distribution map to generate a spatiotemporal evolution data set associated with wetting behavior; extract the wetting dynamic fluctuation data in the spatiotemporal evolution data set, and generate a set of latent variables characterizing the molecular structure properties of textile auxiliaries based on the wetting dynamic fluctuation data; and generate a target parameter dataset based on the matching relationship between the set of latent variables and the preset wetting rate constraint conditions, so as to perform data processing for textile auxiliary production based on the target parameter dataset.

[0182] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method.

[0183] Storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0184] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0185] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0186] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0187] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0188] This invention also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1 The illustrated embodiment is a data processing method for the production of textile auxiliaries.

[0189] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0190] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0191] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0192] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A data processing method for textile auxiliary production, characterized in that, include: Monitor the solution flow state of textile auxiliaries to generate a dynamic gradient data set; A set of reflected signal parameters of the surface of the textile substrate to be treated under electromagnetic excitation in a preset frequency band is obtained. Based on the resonant frequency shift data associated with the plasma resonance effect in the set of reflected signal parameters, an energy distribution map is constructed. The dynamic gradient data set is dynamically coupled with the surface energy characteristics of each partition of the energy distribution map to generate a spatiotemporal evolution data set associated with the infiltration behavior. Extract the infiltration dynamic fluctuation data from the spatiotemporal evolution dataset, and generate a set of latent variables characterizing the molecular structure properties of textile auxiliaries based on the infiltration dynamic fluctuation data. Based on the matching relationship between the set of latent variables and the preset wetting rate constraints, a target parameter dataset is generated for data processing of textile auxiliary agent production based on the target parameter dataset. The dynamic gradient data set is dynamically coupled with the surface energy characteristics of each partition of the energy distribution map to generate a spatiotemporal evolution data set associated with infiltration behavior, including: The surface energy intensity value of each partition in the energy distribution map is multiplied by the plasma resonance frequency offset value of the corresponding partition to generate a set of correlation factors corresponding to the spatial location of each partition. Monitor the temporal fluctuation amplitude of the dynamic gradient data set, and calculate the difference of the temporal fluctuation amplitude based on the preset tension value of adjacent time nodes to generate a set of weight coefficients bound to the spatial location of each partition. The set of correlation factors and the set of weight coefficients are superimposed point by point according to the spatial location corresponding to each partition to generate a coupling parameter matrix; Within a preset time slice sequence, the matrix units in the coupling parameter matrix that satisfy the phase synchronization condition are energy redistributed to generate the redistributed coupling parameter matrix. Based on the allocated coupling parameter matrix, a target matrix unit is selected, and based on the target matrix unit, a spatiotemporal evolution data set associated with the infiltration behavior is generated. Based on the aforementioned wetting dynamic fluctuation data, a set of latent variables characterizing the molecular structural properties of textile auxiliaries is generated, including: The dynamic fluctuation data of the infiltration is analyzed to generate a set of fluctuation components at different time scales; Phase synchronization processing is performed on each wave component in the wave component set to generate a synchronized wave component sequence. The synchronized fluctuation component sequence is mapped to a preset multi-dimensional feature space to generate a target feature space; The weight distribution of the projected coordinates in the target feature space is adjusted according to the order of magnitude of the surface energy characteristics of the partition to generate a set of latent variables characterizing the molecular structure properties of textile auxiliaries.

2. The method according to claim 1, characterized in that, Based on the allocated coupling parameter matrix, target matrix units are selected, and based on the target matrix units, a spatiotemporal evolution data set associated with infiltration behavior is generated, including: Mark the target matrix units in the allocated coupling parameter matrix that exceed the preset gradient value, use the target matrix units as candidate units, and extract the spatial coordinate points corresponding to the candidate units to generate an initial coordinate point set; The spatial coordinate points in the initial coordinate point set are used to perform bidirectional path verification with the first target point in the adjacent time slice in the preset time slice sequence to generate a verified coordinate point set. Based on the liquid-solid interfacial tension value corresponding to each spatial coordinate point in the verified coordinate point set, a processing priority is assigned to the spatial coordinate points in the verified coordinate point set to obtain the target spatial coordinate point priority sequence. Based on the priority sequence of the target spatial coordinate points, the verified coordinate points in the set of verified coordinate points within the same time slice are connected with the second target point in the adjacent time slice to generate an initial spatiotemporal connection network. The branch nodes in the initial spatiotemporal connection network are detected to generate an optimized spatiotemporal connection network; The spatial coordinates of the verified first target point and the verified second target point in the optimized spatiotemporal connection network are concatenated to generate a spatiotemporal evolution data set associated with the infiltration behavior.

3. The method according to claim 2, characterized in that, Based on the liquid-solid interfacial tension values ​​corresponding to each spatial coordinate point in the verified coordinate point set, processing priorities are assigned to the spatial coordinate points in the verified coordinate point set, resulting in a priority sequence of target spatial coordinate points, including: Extract the liquid-solid interfacial tension values ​​of each spatial coordinate point in the verified coordinate point set; Based on the liquid-solid interfacial tension value, sort the spatial coordinate points in the verified coordinate point set to generate a sorting result; Based on the sorting results, priority weights are assigned to the spatial coordinate points in the set of verified coordinate points to generate a priority sequence corresponding to the set of verified coordinate points. The priority sequence is matched with the spatial distribution density of the verified coordinate point set to generate a spatial coordinate point priority sequence. The priority of spatial coordinate points in the spatial coordinate point priority sequence that have a spatial distribution density higher than a preset threshold is enhanced to generate a target spatial coordinate point priority sequence.

4. The method according to claim 2, characterized in that, Detecting branch nodes in the initial spatiotemporal connectivity network to generate an optimized spatiotemporal connectivity network includes: Extract the liquid-solid interfacial tension values ​​on all connection paths of each branch node in the initial spatiotemporal connection network, and calculate the variation range of the liquid-solid interfacial tension values ​​between adjacent time slices. The branch node is defined as a spatial coordinate point where at least two different connection paths intersect. Select a first target branch node from the branch nodes that has a connection path with a change amplitude exceeding a preset change threshold, select a second target branch node from the first target branch node that meets a preset condition, and degenerate the first target branch node that does not meet the preset condition into an ordinary node. The preset condition is that the direction of the liquid-solid interface tension gradient is consistent with the direction of the surface energy distribution of the textile substrate to be treated, and that there are at least two different connection paths. The connection relationships between the second target branch node and the ordinary nodes associated with the second target branch node are reconstructed to generate the optimized spatiotemporal connection network.

5. The method according to claim 1, characterized in that, Based on the matching relationship between the set of latent variables and the preset wetting rate constraints, a target parameter dataset is generated for data processing in textile auxiliary production, including: Each latent variable in the set of latent variables is matched with the benchmark parameter range in the preset infiltration rate constraint at multiple levels to generate an intermediate parameter sequence with hierarchical correlation. Select a target parameter sequence that meets the preset conflict resolution threshold condition from the intermediate parameter sequence, and generate a parameter optimization rule set based on the target parameter sequence; Based on the parameter optimization rule set, the rule-driven parameters of the latent variable set are transformed to generate the target parameter dataset; Data processing for textile auxiliary production is performed based on the target parameter dataset.

6. A data processing system for textile auxiliary production, used to execute the data processing method for textile auxiliary production according to any one of claims 1 to 5, characterized in that, include: The monitoring module is used to monitor the solution flow state of textile auxiliaries and generate a dynamic gradient data set. The construction module is used to acquire the set of reflection signal parameters of the surface of the textile substrate to be treated under electromagnetic excitation in a preset frequency band, and to construct an energy distribution map based on the resonance frequency offset data associated with the plasma resonance effect in the set of reflection signal parameters. The coupling module is used to dynamically couple the dynamic gradient data set with the surface energy characteristics of each partition of the energy distribution map to generate a spatiotemporal evolution data set associated with the infiltration behavior. The extraction module is used to extract the infiltration dynamic fluctuation data from the spatiotemporal evolution data set, and generate a set of latent variables characterizing the molecular structure properties of textile auxiliaries based on the infiltration dynamic fluctuation data. The processing module is used to generate a target parameter dataset based on the matching relationship between the set of latent variables and the preset wetting rate constraints, so as to perform data processing for textile auxiliary production based on the target parameter dataset.

7. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement a data processing method for textile auxiliary production as described in any one of claims 1 to 5.

8. A computer storage medium, characterized in that, The device contains a computer program that, when executed by a computer, implements a data processing method for the production of textile auxiliaries as described in any one of claims 1 to 5.

Images