Fabric spectrum modeling method and system oriented to MOF spectrum regulation and control

Abstract

The invention relates to the technical field of spectral signal processing, in particular to a fabric spectrum modeling method and system oriented to MOF spectrum regulation and control, and the method comprises the following steps: recombining structure parameters and spectrum data to generate a reference parameter set, analyzing load capacity and reflectivity correlation, recognizing abnormal fluctuation, and generating a diagnosis data set; and extracting a spectral distribution state to identify key regulation and control nodes to generate a distribution result, and identifying spectral offset of adjacent sections of environmental fluctuation to obtain an evolution model. According to the method, digital quantization of spectral response characteristics is achieved through structured recombination, energy exchange deviation is recognized through correlation analysis to eliminate interference, the radiation refrigeration mode recognition capacity is enhanced in combination with position coding, and the wave spectrum change trend is tracked in real time by building a dynamic evolution model; the regulation and control sensitivity under the complex microclimate is obviously improved, the self-adaptive cooperation of the refrigeration efficiency and the environmental fluctuation is ensured, and the efficient prediction of the spectrum behavior is realized.

Description

Technical Field

[0001] This invention relates to the field of spectral signal processing technology, and in particular to a fabric spectral modeling method and system for MOF spectral modulation. Background Technology

[0002] The field of spectral signal processing technology refers to the technical field of optimizing and utilizing spectral signals in various application scenarios through the acquisition, analysis, processing, and interpretation of spectral data. The core aspects of this technology include the acquisition and detection of spectral signals, methods for processing and analyzing spectral data, and the application of spectral information. Specifically, spectral signal processing technology focuses on how to acquire spectral information using advanced instruments and equipment, extract effective information using mathematical algorithms and processing techniques, and apply it to various fields such as environmental monitoring and materials analysis. Furthermore, spectral signal processing technology also includes techniques such as spectral modeling, data filtering, and signal enhancement, aiming to improve the accuracy and practicality of spectral data.

[0003] Traditional fabric spectral modeling methods for MOF spectral modulation involve using the spectral properties of metal-organic frameworks (MOFs) combined with the structural characteristics of the fabric to perform spectral modulation and modeling. The main challenge of this method is how to optimize the spectral properties of the fabric by adjusting the spectral response of the MOF material to achieve specific functional requirements, such as daytime radiative cooling. Traditional methods employ experimental techniques, adjusting the type, structure, and surface properties of the MOF material in conjunction with the fabric weaving process to perform spectral modulation and modeling.

[0004] Traditional methods mainly rely on physical experiments to adjust material types and weaving operations to try to optimize spectral characteristics. This experience-based static debugging mode is difficult to quantify and analyze the complex coupling relationship between the multi-layer structure of the fabric and the spectral response. The manual trial and error process cannot capture the spectral shift trajectory under dynamic environmental changes. The lack of multi-dimensional data correlation makes it impossible to accurately identify abnormal energy exchange areas and the source of spectral fluctuations. The static parameter setting of a single dimension is difficult to adapt to the continuous fluctuation of the microclimate environment. As a result, the optimization of radiation cooling performance lacks the support of an accurate evolution model and cannot effectively predict the dynamic response trend under complex working conditions, resulting in low control accuracy and sluggish response. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and propose a fabric spectral modeling method and system for MOF spectral modulation.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a fabric spectral modeling method for MOF spectral modulation, comprising the following steps: S1: Based on the initial input of the multi-layer structure of fabric and the spectral response of metal-organic framework materials, the structural parameters and spectral data of the differential sources in the fabric spectral modulation model are restructured and combined with time labels and spatial identifiers to generate a set of spectral response benchmark parameters. S2: Based on the spectral response reference parameter set, analyze the correlation between metal-organic framework loading and spectral reflectance distribution, identify the set of abnormal fluctuation intervals and energy exchange deviation range, filter spectral sampling units that exceed the preset threshold, and generate a spectral anomaly diagnostic dataset. S3: Based on the spectral anomaly diagnostic dataset, extract the spectral distribution state and radiation cooling change value, identify key spectral control nodes and material structure combinations, screen spectral units that conform to the preset standard radiation characteristic model, and superimpose position codes to mark the corresponding regions to generate fabric spectral distribution results. S4: Based on the continuous change information of multiple environmental parameters in the fabric spectral distribution results, identify the spectral shift of adjacent segments of environmental fluctuations, extract the shift direction and response trajectory, screen the spectral segments with consistent continuous shift directions, and obtain the fabric spectral dynamic response and regulation evolution model.

[0007] As a further embodiment of the present invention, the spectral response reference parameter set includes a spectral distribution feature value set, a spectral attribute grid, and a structural operation parameter surface; the spectral anomaly diagnostic dataset includes spectral fluctuation identification results, energy deviation markers, radiation energy data, and interference data removal masks; the fabric spectral distribution results include a target spectral candidate set, key node response templates, radiation cooling change values, and regional location codes; and the fabric spectral dynamic response and regulation evolution model includes a continuous offset trend trajectory, a path direction vector set, and regional offset nodes.

[0008] As a further aspect of the present invention, the step of obtaining the spectral response reference parameter set specifically includes: S111: Based on the initial input of metal-organic framework materials and fabric matrix, the raw data from the differential detection channel in the fabric spectral modulation model are collected, and the frame sequence is matched with the time label. The spectral intensity difference between adjacent frames is calculated, and the sampling units in the spectral distribution sequence whose difference between adjacent frames exceeds the fluctuation critical threshold are removed to generate a multidimensional spectral matrix. S112: Based on the multidimensional spectral matrix, extract the reflection intensity and transmission distribution data under the same time and space identifier, perform data integrity verification, filter out records with missing items, and generate a valid spectral running factor data group. S113: Call the effective spectral operating factor data group, identify the standard deviation of the reflection intensity in the sampling unit group with the same spectral operating mode and normalize it, calculate the normalized spectral difference intensity value, reclassify the sampling units in the region, and establish a spectral response benchmark parameter set.

[0009] As a further aspect of the present invention, the steps for obtaining the spectral anomaly diagnostic dataset are as follows: S211: Based on the spectral response reference parameter set, extract the spectral distribution state and radiation energy data in the target layer, match and compare the fluctuation value of the spectral sampling unit with the preset radiation range, analyze the correspondence, and obtain the spectral fluctuation range data set. S212: Call the spectral fluctuation range data set, combine it with the distribution status of the sampling units in the target layer, determine whether the value of the sampling unit falls into the specified range, identify the difference of the sampling units that are not in the range, determine the objects to be removed based on the difference, and obtain a list of removed sampling unit numbers. S213: Call the list of sampled units to be removed, delete the interference data associated with the corresponding number in the target layer, reorganize the spatial distribution of the remaining sampled units, and generate a spectral anomaly diagnostic dataset.

[0010] As a further aspect of the present invention, the step of obtaining the fabric spectral distribution result specifically includes: S311: Based on the spectral anomaly diagnostic dataset, extract the distribution status and radiation cooling change value of the layer sampling units, combine the differentiated spectral nodes with the corresponding structural parameter data, identify typical spectral mode regions, and obtain the spectral mode characterization combination. S312: Call the spectral mode characterization combination, detect the fluctuation difference of sampling units and the degree of coordinated change of structural parameters in the combined region, aggregate and compare the response values ​​of sampling units at key nodes, identify the region with potential spectral behavior, and obtain the index set of spectral influence sensitive regions. S313: Call the set of spectral influence sensitive area indicators, match the sensitive area sampling unit with the spectral mode characterization combination, filter the sampling units that meet the feature requirements, and mark the location coding information of the corresponding area to generate the fabric spectral distribution result.

[0011] As a further aspect of the present invention, the steps for obtaining the fabric spectral dynamic response and regulation evolution model are specifically as follows: S411: Based on the spectral distribution results of the fabric, extract the continuous change information between multiple environmental time periods in the spectrum, identify the spectral offset of the geometric center of adjacent regions in the environmental sequence, extract the offset direction vector and displacement length, analyze the angle difference of the consistency of the offset direction between adjacent regions, and obtain a dataset of continuous spectral offset directions. S412: Call the aforementioned continuous spectral offset direction dataset, and based on the changing trend of the angle difference, use the formula: ; Calculate the angular fluctuation dispersion value, extract the region segment with stable offset direction, filter the continuous segment with angular change below the preset threshold, and obtain stable offset segment information; in, Represents the angular fluctuation dispersion value. Representing the The angle value of each data point. Representing the The angle value of each data point. Represents the total number of data points. Represents microclimate data associated with changes in angle. Represents the local rate of change of angle; S413: Call the stable offset segment information, combine the offset direction angle in each segment, the spectral distribution intensity between regions and the environmental sequence continuity index, identify trajectory segments with strong continuity and concentrated spectral signals, fit the evolution characteristics, and obtain the fabric spectral dynamic response and regulation evolution model.

[0012] As a further aspect of the present invention, the method further includes step S5: S5: Based on the fabric spectral dynamic response and regulation evolution model, identify the radiation intensity change trend and spectral response level, construct the radiation cooling efficiency distribution, compare the time synchronization change characteristics between different bands, identify the overlapping positions of the segment coordinates and spectral distributions that occur at the same time, and output the spectral behavior linkage response control instruction set. The spectral behavior linkage response control instruction set includes radiation linkage change zone, spectral response intersection point, and abnormal performance index group.

[0013] As a further aspect of the present invention, the step of obtaining the spectral behavior linkage response control instruction set specifically includes: S511: Based on the fabric spectral dynamic response and regulation evolution model, extract the radiation intensity change trend and spectral response level in the section, identify the daily change value of reflectance and the daily average value of emissivity in the section, analyze the radiation change amplitude and spectral fluctuation degree, and generate a synchronous spectral change characteristic value group. S512: Call the synchronous spectral change feature value group, and extract the inflection points of continuous time periods based on the efficiency sequence of the radiation cooling efficiency layer grid, combined with the radiation change amplitude, spectral fluctuation degree and cooling efficiency. Filter the segment numbers of the turning point features and summarize them to obtain the abrupt change synchronous segment index list. S513: Based on the mutation synchronization segment index list, match the spectral distribution layer coordinate units corresponding to the segment number, filter overlapping grids with consistent indices and mark them, and output the spectral behavior linkage response control instruction set.

[0014] The fabric spectral modeling system for MOF spectral manipulation is used to execute the aforementioned fabric spectral modeling method for MOF spectral manipulation. The system includes: The spectral benchmark construction module collects data on the sources of differentiation in the fabric spectral control model, extracts spectral operation data, and combines time and spatial identifiers to format the spectral data and structural factors to generate a spectral response benchmark parameter set. The spectral anomaly detection module, based on the spectral response reference parameter set, filters sampling units that exceed a preset threshold according to the spectral distribution state and radiation data, removes interference data associated with anomalies, integrates the remaining spectral distribution state and operational information, marks abnormal areas, and generates a spectral anomaly diagnostic dataset. Based on the spectral anomaly diagnostic dataset, the spectral behavior extraction module extracts the spectral distribution state, radiative cooling change value, and spectral intensity, identifies sampling units that conform to the spectral pattern, matches the spectral behavior template, overlays the location code and time label, performs spatial positioning, and generates the fabric spectral distribution result. Based on the fabric spectral distribution results, the spectral path tracking module extracts time tags and spectral operation information, identifies the spatial distance between spectral behavior regions in adjacent time periods, determines the offset direction and continuous trajectory, integrates path segments with consistent offset directions, and obtains a dynamic response and regulation evolution model of the fabric spectrum. The spectral linkage analysis module, based on the fabric spectral dynamic response and regulation evolution model, compares time series data according to the radiation intensity change trend, spectral response level and radiation cooling efficiency of the path segment, identifies synchronous turning points, locates the overlapping positions of turning areas and spectral behavior, and outputs a set of spectral behavior linkage response control instructions.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, by structurally reorganizing the multi-layered structure and spectral data of the fabric and combining it with spatiotemporal identifiers to generate benchmark parameters, the digital and precise quantification of spectral response characteristics is achieved. By using the correlation analysis between load and reflectivity to identify abnormal fluctuation ranges and energy exchange deviations, the influence of interference data on model accuracy is effectively eliminated. By combining position coding to extract key control nodes and material structure combinations, the ability to identify radiative cooling modes is enhanced. By identifying the continuous spectral shift direction and response trajectory under multiple environmental parameters, a dynamic evolution model is constructed to track spectral change trends in real time. This significantly improves the control sensitivity of the fabric under complex microclimates, ensures the adaptive coordination between radiative cooling efficiency and environmental fluctuations, and achieves efficient prediction and precise management of spectral behavior. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the workflow of the present invention; Figure 2This is a flowchart illustrating the process of obtaining the spectral response reference parameter set in this invention. Figure 3 This is a flowchart illustrating the acquisition process of the spectral anomaly diagnostic dataset in this invention. Figure 4 This is a flowchart illustrating the process of obtaining the spectral distribution results of the fabric in this invention. Figure 5 This is a flowchart illustrating the acquisition of the fabric spectral dynamic response and regulation evolution model in this invention. Figure 6 This is a flowchart illustrating the acquisition of the spectral behavior linkage response control instruction set in this invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0018] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0019] Example 1 Please see Figure 1 This invention provides a technical solution: a fabric spectral modeling method for MOF spectral modulation, comprising the following steps: S1: Based on the initial input of the multi-layer structure of fabric and the spectral response of metal-organic framework materials, the structural parameters and spectral data of the differential sources in the fabric spectral modulation model are restructured and combined with time labels and spatial identifiers to generate a set of spectral response benchmark parameters. S2: Based on the spectral response reference parameter set, analyze the correlation between metal-organic framework loading and spectral reflectance distribution, identify the abnormal spectral fluctuation range data set and energy exchange deviation range, screen spectral sampling units that exceed the preset threshold, remove related interference data, and generate a spectral anomaly diagnostic dataset. S3: Based on the spectral anomaly diagnosis dataset, extract the spectral distribution status and radiation cooling change value, identify key spectral control nodes and material structure combinations, screen spectral units that conform to the preset standard radiation characteristic model, and overlay position codes to mark the corresponding regions to generate fabric spectral distribution results. S4: Based on the continuous change information of multiple environmental parameters in the fabric spectral distribution results, identify the spectral shift of adjacent segments of environmental fluctuations, extract the shift direction and response trajectory, screen the spectral segments with the same continuous shift direction, and obtain the dynamic response and regulation evolution model of fabric spectrum. S5: Based on the dynamic response and regulation evolution model of fabric spectrum, identify the trend of radiation intensity change and spectral response level, construct the radiation cooling efficiency distribution, compare the time synchronization change characteristics between different bands, identify the overlapping positions of the segment coordinates and spectral distributions that occur at the same time, and output the spectral behavior linkage response control command set.

[0020] The spectral response benchmark parameter set includes a spectral distribution feature value set, a spectral attribute grid, and a structural operation parameter surface. The spectral anomaly diagnostic dataset includes spectral fluctuation identification results, energy deviation markers, radiation energy data, and interference data removal masks. The fabric spectral distribution results include a target spectral candidate set, key node response templates, radiation cooling change values, and regional location codes. The fabric spectral dynamic response and regulation evolution model includes a continuous offset trend trajectory, a path direction vector set, and regional offset nodes. The spectral behavior linkage response control instruction set includes radiation linkage change zones, spectral response intersection points, and abnormal performance index groups.

[0021] Please see Figure 2 The specific steps for obtaining the spectral response reference parameter set are as follows: S111: Based on the initial input of metal-organic framework materials and fabric matrix, the raw data from the differential detection channel in the fabric spectral modulation model are collected, and the frame sequence is matched with the time label. The spectral intensity difference between adjacent frames is calculated, and the sampling units in the spectral distribution sequence whose difference between adjacent frames exceeds the fluctuation critical threshold are removed to generate a multidimensional spectral matrix. Based on the thickness parameters of the metal-organic framework material and the areal density parameters of the fabric matrix, the voltage signals of the detection channels are read. The signal at a wavelength of 500 nm is recorded as the parameter of channel 1, and the signal at a wavelength of 10 μm is recorded as the parameter of channel 2. The millisecond timestamp values ​​are extracted and subtracted. If the absolute value is within the range of 0 to 5 milliseconds, the parameter and timestamp are assigned to the matching data frame. The frame sequence is read and the frame number is arranged by time. The corresponding channel parameters of the current frame and the previous frame are extracted and subtracted to obtain the absolute value, thus obtaining the spectral intensity difference. The fluctuation critical threshold is read. The threshold setting is based on the dark current benchmark test record. The arithmetic mean of 100 test values ​​is calculated. The threshold is set to the arithmetic mean plus three times the standard deviation. Substituting the mean of 0.02 mV and the standard deviation of 0.005 mV, the threshold is calculated to be 0.035 mV. The difference is compared with 0.035 mV. When there is a channel difference greater than 0.035 mV, an abnormal label is assigned and a release command is called to clear the corresponding memory data. In frame 10, when the difference between two channels is less than the threshold, a normal label is assigned and retained. In frame 11, when there is a channel difference greater than the threshold, a clearing operation is performed. The normal label data is renumbered, and the wavelength coordinates, timestamps, and intensity parameters are extracted and filled into the matrix to generate a multidimensional spectral matrix.

[0022] S112: Based on the multidimensional spectral matrix, extract the reflection intensity and transmission distribution data under the same time and space identifier, perform data integrity verification, filter out records with missing items, and generate a valid spectral running factor data set. Read the value of each voxel in the multidimensional spectral matrix, separate the reflection intensity value R and transmission distribution value T at a specific time point t and spatial coordinates, and start the data integrity verification program; check the data type of all data points to ensure that they are in double-precision floating-point format, and scan for non-numeric NaN or infinite Inf identifiers, and count and mark about 0.5% of the packet loss or overflow anomalies in the original data; execute the physical meaning verification logic, and for the reflection intensity R and transmission distribution value T, force their effective physical value range to be between 0 and 1, that is, corresponding to 0% to 100% spectral responsivity; filter out records with values ​​less than 0 or greater than 1, and judge. This data was found to be erroneous and violated physical laws. All rows containing records marked as formatting errors, missing values, or physical out-of-bounds were permanently removed from the matrix to ensure the remaining data sets possessed complete reflection and transmission properties and conformed to physical laws. Statistically, the amount of effective data after cleaning accounted for 98.2% of the original data. The rigorously cleaned reflection intensity and transmission distribution data were re-indexed according to their original temporal and spatial order, generating an effective spectral running factor data set containing approximately 50,000 records. This data set served as a standardized input source, eliminating the influence of outliers in the original acquisition process and ensuring the accuracy of subsequent baseline parameter calculations.

[0023] S113: Call the effective spectral operating factor data set, identify the standard deviation of the reflection intensity in sampling unit groups with consistent spectral operating modes, and normalize it using the following formula: ; Calculate the normalized spectral difference intensity values, reclassify the sampling units in the region, and establish a set of spectral response reference parameters. in, This represents the normalized spectral difference intensity value. Representing the first under the same spectral operating mode The reflection intensity value of each sampling unit. This represents the arithmetic mean of the reflection intensity values ​​of all sampling units within the sampling unit group. This represents the maximum reflection intensity result within the set of reflection intensity values ​​in the sampling unit group. This represents the minimum reflection intensity result within the set of reflection intensity values ​​in the sampling unit group. This represents the number of sampling units in the same spectral operating mode participating in the calculation. This represents the reference reflection intensity value corresponding to the current sampling unit group; The program calls upon the effective spectral operating factor data set, which originates from the cleaned results of hyperspectral imaging data within a 50 mm × 50 mm area on the fabric surface. This data set includes latitude and longitude coordinates, timestamps, and corresponding reflectance intensity sequences. The program first groups sampling units with the same mode label into the same analysis group based on the characteristic labels of the spectral operating mode (e.g., "high reflectance-low emission" mode, "selective radiance" mode). The number of sampling units included in the sliding window for a single calculation is then set. 100 (this number) The selection is based on the central limit theorem and the load balancing experiment. The experiment shows that when When the confidence level of the statistical features reaches over 98%, the standard deviation of the reflection intensity is identified and normalized for the sampling units in this group, in order to quantify the dispersion of the spectral response and the impact of extreme value fluctuations in the current region. In this process, the system does not directly use the original physical quantities for simple statistics, but instead introduces environmental benchmark parameters for weighted calculations. First, by using a standard blackbody radiation source deployed in the fabric testing environment and a white board reference, the current benchmark reflection intensity value is calibrated in real time. Specifically, The value is obtained by calibrating a high-precision reflectivity sensor (accuracy 0.1%) by aligning it with a standard white board (reflectivity known to be 0.99) under current lighting, temperature, and humidity conditions. If the current ambient light intensity is 1000 watts per square meter, the baseline reflectivity value is derived by measuring the reflected energy of the white board. Set to 0.95 (this value is used as the denominator for normalization to eliminate interference from different ambient light intensity bases); Next, the normalized spectral difference intensity values ​​are calculated using the following formula: ; To clearly explain the calculation logic and physical meaning of this formula, the following example uses a set of actually collected numerical values: Assume the number of sampling units currently involved in the calculation is... The number is 100; the reflection intensity values ​​of these 100 sampling units are iterated over. Calculate its arithmetic mean. Assuming the read reflection intensity values ​​are between 0.88 and 0.94, the average value is obtained by summing them and dividing by 100. ; Then calculate the first term in the numerator of the formula (the standard deviation term): for each Calculate the difference between this and the average of 0.91, and square the difference; sum these 100 squared values, assuming the summation result is... The sum is 0.0025 (representing the dispersion of the data); divide this sum by... (100), resulting in 0.000025; finally, taking the square root of the quotient yields... ; Simultaneously calculate the second term (range average term) in the numerator of the formula: scan the data set and extract the result of the maximum reflection intensity. and minimum reflection intensity results The absolute value of the difference between the two is calculated as follows: Divide the range value by the number of sampling units. (100), obtained (This term is used to introduce the perturbation weight of extreme values ​​on overall stability); Add the two terms in the numerator: Finally, divide the molecular result by the reference reflectance value. (0.95), that is The final normalized spectral difference intensity value was calculated. ; The calculation result (Approximately 0.00589) is a dimensionless value that comprehensively reflects the uniformity of the fabric's spectral response within this region. A smaller value indicates more consistent and stable spectral behavior among sampling units; a larger value indicates significant local defects or uneven dynamic response. Based on statistical analysis of a large amount of experimental data, a reclassification threshold standard is set: when... When, it is determined to be in the "absolute steady state region"; when When, it is determined to be a "metastable adjustment region"; when When the value is 0.00589 (less than 0.01), it is identified as a "dynamic mutation zone". Based on the above example result of 0.00589, the sampling units in this region are reclassified as "absolute steady-state zones" and labeled with corresponding reference labels. Finally, by traversing the calculation results of all regions, a spectral response reference parameter set containing location coordinates, classification labels, and ΔQ values ​​is established. The advantage of this formula is that it not only considers the overall dispersion (standard deviation) of the data, but also introduces sensitivity to individual outliers (such as fabric breakage points) through the range term, which significantly enhances the sensitivity of the reference parameter set to detecting minor defects.

[0024] Please see Figure 3 The specific steps for obtaining the spectral anomaly diagnostic dataset are as follows: S211: Based on the spectral response reference parameter set, extract the spectral distribution state and radiation energy data in the target layer, match and compare the fluctuation value of the spectral sampling unit with the preset radiation range, analyze the correspondence, and obtain the spectral fluctuation range data set. Data records identified as high-dynamic-range fluctuation areas in the spectral response benchmark parameter set are read, and an infrared thermal imaging layer of the fabric surface is loaded. Radiation energy data at corresponding coordinate points is extracted, in watts per square meter (W / m²). A preset radiation range of 300 W / m² to 450 W / m² is set, covering the radiation power frequency band corresponding to human comfort temperature. An Intersection over Union (IoU) algorithm is used to geometrically match the spatial coverage of the spectral sampling units with abnormal spots in the infrared thermal image whose radiation energy exceeds the preset range. The overlap area between pixel regions with a spectral difference intensity value (Delta_Q) greater than 0.15 and pixel regions with radiation energy greater than 450 W / m² is calculated. If the overlap rate exceeds 80%, a strong correspondence is confirmed. The fluctuation amplitude and radiation energy values ​​of all sampling units that satisfy the correspondence are recorded, constructing a spectral fluctuation range data set containing a triplet of coordinates, fluctuation values, and radiation values. This set accurately identifies the regions on the fabric surface where the spectral response is unstable due to thermal and moisture overload. This multimodal data matching effectively identifies thermal and moisture comfort failure points, providing precise target coordinates for subsequent anomaly removal.

[0025] S212: Call the spectral fluctuation range data set, combine it with the distribution status of the sampling units in the target layer, determine whether the value of the sampling unit falls into the specified range, identify the difference of the sampling units that are not in the range, determine the objects to be removed based on the difference, and obtain the list of removed sampling unit numbers. The algorithm iterates through each triplet record in the spectral fluctuation range dataset, checking whether its recorded radiant energy value falls within the specified effective control range, which is set to 320 watts per square meter to 420 watts per square meter. For sampling units with radiant values ​​outside this range, such as a detected value of 460 watts per square meter, a difference calculation is performed. The detected value of 460 is subtracted from the upper limit of the range of 420, resulting in a difference of 40 watts per square meter. Based on this difference, objects are removed, with a difference tolerance threshold set at 10 watts per square meter. If the calculated difference... If the value of 40 is greater than the threshold of 10, it is determined that the data of the sampling unit is severely interfered with by external strong heat sources such as direct sunlight or high-temperature equipment, and its value has deviated from the material's inherent characteristics. The unique ID of the sampling unit is extracted and added to the list of excluded sampling units. In a detection task containing 1000 sampling points, 50 points were identified as having radiation values ​​that exceeded the standard and differences that exceeded the tolerance threshold. The IDs of these 50 points were written into the list to remove extreme data caused by non-material factors from the analysis queue and prevent them from having a biased impact on the overall model.

[0026] S213: Call the list of sampled unit numbers to be removed, delete the interference data associated with the corresponding numbers in the target layer, reorganize the spatial distribution of the remaining sampled units, and generate a spectral anomaly diagnostic dataset. The original target layer data matrix is ​​loaded, and the list of culled sampling unit numbers is read. Masking technology is used to zero out or mark the corresponding grid data in the data matrix as null. For holes created in the data matrix by the culling operation, Kriging interpolation is used to fill in the holes using the numerical values ​​and spatial correlations of surrounding valid sampling units, reorganizing the spatial distribution structure of the remaining sampling units to ensure the continuity and smoothness of the data in geometric space. After interpolation, the entire layer is smoothed to eliminate the data step effect caused by the culling boundary, ultimately generating a clean spectral anomaly diagnostic dataset that reflects the inherent regulatory performance of the fabric. Compared to the original data, this dataset has a significantly improved signal-to-noise ratio, effectively removing environmental noise and interference sources, providing high-fidelity data input for subsequent deep learning pattern recognition, and ensuring the accuracy of obstacle diagnosis.

[0027] Please see Figure 4 The specific steps for obtaining the spectral distribution results of the fabric are as follows: S311: Based on the spectral anomaly diagnosis dataset, extract the distribution status and radiation cooling change value of the layer sampling unit, combine the differential spectral nodes with the corresponding structural parameter data, identify typical spectral mode regions, and obtain the spectral mode characterization combination. A deep convolutional neural network (CNN) was used to extract features from a spectral anomaly diagnostic dataset. A network structure consisting of three convolutional layers and two pooling layers was constructed. The first layer was configured with 32 convolutional kernels of size 3x3 to capture the texture details and spectral distribution features of the fabric surface, using ReLU as the activation function. The input consisted of fabric structural parameters, including porosity percentage, fiber diameter in micrometers, and radiative cooling variation in watts per square meter. The extracted image features were then concatenated and fused with the structural parameter values ​​in a fully connected layer. Through network inference, regions exhibiting specific spectral behaviors, such as high reflectance and low transmittance or selective emission, were identified. For example, a region with a porosity of 30% and a 20% increase in radiative cooling was identified as an enhanced radiative cooling mode. The output consisted of a spectral pattern representation combination containing the mode category, center coordinates, and coverage area. This combination transformed abstract spectral data into a concrete description of functional regions, achieving a digital representation of the photothermal regulation mechanism of the fabric surface.

[0028] S312: Call the spectral mode to characterize the combination, detect the fluctuation difference of the sampling unit in the combination area and the degree of coordinated change of the structural parameters, aggregate and compare the response values ​​of the sampling unit at key nodes, identify the region with potential spectral behavior, and obtain the index set of spectral influence sensitive areas. For each identified spectral pattern characterization region, the fluctuation differences of sampling units within the region and the synergy of structural parameters are analyzed in depth. The Pearson correlation coefficient is used to calculate the correlation between the spectral response value sequence of the sampling unit and local structural parameters such as fiber arrangement density sequences. If the correlation coefficient is greater than 0.85, it is determined that there is a high degree of synergistic change, indicating that changes in structural parameters significantly dominate the spectral response. The response values ​​at key nodes such as peaks and troughs are aggregated, and their coefficient of variation (CV) is calculated. If the CV value is less than 0.1, it indicates that the spectral behavior of that region is stable. Regions with a correlation coefficient greater than 0.85 and a CV value less than 0.1 are labeled as potential regions for spectral behavior, representing potential points for material performance optimization. A set of spectral influence sensitive region indicators, including region ID, correlation coefficient, and sensitivity score, is generated. This indicator set quantifies the sensitivity of different regions to structural regulation, providing data support for subsequent targeted structural optimization.

[0029] S313: Call the spectral influence sensitive area index set, match the sensitive area sampling unit with the spectral mode characterization combination, filter the sampling units that meet the feature requirements, and mark the location coding information of the corresponding area to generate the fabric spectral distribution results; A spectral pattern template library containing standard curves for ideal broadband reflectance and narrowband selective emission was loaded. The sampling unit data from the spectral influence sensitive region index set were matched with the templates. The Dynamic Time Warping (DTW) algorithm was used to calculate the similarity distance between the spectral curves of the sampling units and the template curves, with a similarity threshold of 0.9. Sampling units with similarity greater than 0.9 were selected to confirm they met the ideal feature requirements. The physical coordinates (x, y) of the sampling units on the fabric surface were obtained, and combined with their corresponding spectral pattern codes, a fabric spectral distribution result file was generated. The file clearly records the distribution information of broadband reflectance patterns corresponding to coordinates (50, 120) and possessing high sensitivity characteristics. As shown in Table 1, some distribution result data are presented. Sampling number 1024 is located at coordinates (50, 120), matching a broadband reflectance pattern with a similarity score of 0.94 and a structural synergy of 0.88. This distribution result intuitively demonstrates the functional attributes of each region of the fabric, providing a spatial navigation map for subsequent dynamic control. Table 1: Fabric Spectral Distribution and Feature Matching Results

[0030] Table 1 shows the spectral distribution of each sampling unit of the fabric and its matching results with different spectral modes, including the coordinate position of each sampling unit, the type of matching mode, the similarity score and structural synergy, to intuitively present the functional attributes of each region of the fabric.

[0031] Please see Figure 5 The specific steps for obtaining the dynamic response and regulation evolution model of fabric spectra are as follows: S411: Based on the spectral distribution results of the fabric, extract the continuous change information between multiple environmental time periods in the spectrum, identify the spectral offset of the geometric center of adjacent regions in the environmental sequence, extract the offset direction vector and displacement length, analyze the angle difference of the consistency of the offset direction between adjacent regions, and obtain the dataset of continuous spectral offset directions. The fabric spectral distribution data were continuously read over 24 hours with a time resolution of 1 hour. Optical flow was used to track the movement of high-response regions (hotspots) in the spectral data. The geometric center coordinates of the same characteristic region were identified between two adjacent times (e.g., 10 AM to 11 AM), and the displacement vector between the two points was calculated. The offset direction vector and displacement length were extracted. The angular difference Delta_Theta between adjacent regions was calculated; for example, if the offset angle is 30 degrees at time t1 and 35 degrees at time t2, the difference is 5 degrees. The displacement and angle data for the entire day were summarized to construct a continuous spectral offset direction dataset. This dataset reveals the dynamic drift pattern of the fabric spectral response with changes in solar altitude angle and ambient temperature and humidity, providing a quantitative basis for understanding the time-varying influence of environmental factors on spectral performance and forming the foundation for establishing a dynamic evolution model.

[0032] S412: Call the continuous spectral offset direction dataset, and use the formula based on the changing trend of the angle difference: ; Calculate the angular fluctuation dispersion value, extract the region segment with stable offset direction, filter the continuous segment with angular change below the preset threshold, and obtain stable offset segment information; in, Represents the angular fluctuation dispersion value. Representing the The angle value of each data point. Representing the The angle value of each data point. Represents the total number of data points. Represents microclimate data associated with changes in angle. Represents the local rate of change of angle; The program calls upon a continuous spectral offset direction dataset, which records the angular change trajectory of the fabric's spectral feature center over time. To quantify the stability of this change, the program sets the sliding time window length N to 20 data points (corresponding to a sampling duration of 20 minutes and a frequency of once per minute). The selection of this window size is based on the determination of the fabric's thermal response hysteresis effect, ensuring that it covers a complete micro-thermal equilibrium cycle. Next, the program calls upon the formula for calculating angular fluctuation dispersion, aiming to combine environmental microclimate to perform a weighted evaluation of angular changes. Before applying the formula, it is necessary to clarify how each parameter is obtained in real time: Parameter (Microclimate data associated with angle changes) is a comprehensive environmental impact factor. It is not a direct measurement, but is calculated from real-time relative humidity (RH, in %) and surface temperature (T, in degrees Celsius) collected by miniature temperature and humidity sensors on the fabric surface. The calculation logic is as follows: ,in The standard humidity is 60%. The standard temperature is 25 degrees Celsius; for example, if the current sensor reading is 45% humidity and 30 degrees Celsius, then... This parameter reflects the potential driving force of environmental humid and hot conditions on fabric fiber expansion and spectral angle shift. The local rate of change of angle is obtained by calculating the average of the first derivatives of the angle changes at the three most recent time points, in radians per second; for example, if the rate of change of angle at the most recent moment is extremely slow, The value is 0.02 radians / second; the angular fluctuation dispersion value is calculated using the following formula: ; The following detailed numerical substitution calculations demonstrate how the algorithm determines the stability of the offset: setting the window size First, calculate the summation term. The program calculates the spectral shift angle between two adjacent time points within the window one by one. The absolute value of the difference. Assuming the angular jumps between these 20 data points are very small, such as {0.01, 0.02, 0.01…}, summing them all gives a total angular path length of 0.4 radians. Next, the average term in the formula is calculated: the total path length 0.4 is divided by the total number of data points, 20, which is… This step yields the unweighted average angular fluctuation value; then the environmental weight term is calculated. The aforementioned microclimate data With local rate of change Substitute; calculate the product Take the square root of the product. This parameter serves as a correction factor to amplify the impact of fluctuations under harsh environments or rapid changes; finally, the final dispersion value is calculated. Multiply the average fluctuation value of 0.02 by the environmental weighting coefficient of 0.134, i.e. ; Calculated This represents the comprehensive instability index of spectral shift within this time period. The preset stability threshold is 0.01 (this threshold was determined through benchmark testing of the fabric in an ideal constant temperature and humidity chamber, covering 99% of the steady-state noise range); the calculated result of 0.00268 is compared with the threshold of 0.01, because... The program determines that the current segment is in a "stable offset" state. It will then filter out all segments that meet the criteria. For a continuous time period defined by the condition, the start and end times (e.g., 10:00 to 10:20) and the corresponding spectral features are extracted to generate stable migration segment information. This formula introduces... This method effectively eliminates spurious stable signals caused by sudden environmental changes (such as gusts of wind or sudden changes in humidity). Experimental data show that compared with the method that only uses angle variance, this algorithm significantly improves the accuracy of identifying the true steady-state range. Table 2 below lists the calculation results and judgment conclusions after substituting different environmental parameters into the formula, verifying the robustness of the model in complex environments. Table 2: Example Table of Calculation Results for Stable Offset Sections

[0033] This result indicates that even when the cumulative angular difference is the same (e.g., A01 and B02), differences in environmental factors can fine-tune the final result. The values ​​reflect the model's sensitivity to environmental stress; however, under conditions of high change rates or high cumulative differences (C03, D04), The value significantly exceeded the threshold, ensuring that non-steady-state data was accurately removed.

[0034] S413: Call up the information of stable offset sections, combine the offset direction angle, spectral distribution intensity between regions and environmental sequence continuity index in each section, identify trajectory segments with strong continuity and concentrated spectral signals, fit the evolution characteristics, and obtain the fabric spectral dynamic response and regulation evolution model. Information on all identified stable offset segments was extracted. The average offset direction angle, mean spectral intensity across regions, and environmental sequence continuity index (ranging from 0 to 1, with 1 representing no abrupt environmental changes) were combined in each segment. Spline interpolation was applied to fit the trajectory of discrete stable segments, constructing a continuous fabric spectral response evolution curve. For segments identified as unstable, a weighted moving average method was used for smoothing correction. Trajectory segments with continuous first derivatives and a peak half-width at half-maximum (FWHM) of less than 20 nm were identified in the evolution curve, parameterized, and used to establish a dynamic response and control evolution model for the fabric's spectrum. This model can quantify the fabric's response path under different environmental sequences and predict the spectral response trend within the next hour with over 90% accuracy, providing predictive model support for feedforward control.

[0035] Please see Figure 6 The specific steps for obtaining the spectral behavior linkage response control instruction set are as follows: S511: Based on the dynamic response and regulation evolution model of fabric spectrum, extract the trend of radiation intensity change and spectral response level in the section, identify the daily change value of reflectance and the daily average value of emissivity of the section, analyze the radiation change amplitude and the degree of spectral fluctuation, and generate a synchronous spectral change characteristic value group. Based on the established dynamic response and regulation evolution model of fabric spectral data, the system inputs meteorological forecast data for the next 24 hours to predict the trend of radiation intensity changes in a given area. It extracts the daily variation values ​​of reflectance (e.g., decreasing from 0.85 in the morning to 0.70 at noon) and the daily average value of emissivity (e.g., 0.92) from the prediction model. It calculates the amplitude of radiation change (the difference between the maximum and minimum values) and the degree of spectral fluctuation (variance). Periods with radiation change amplitudes greater than 20% and spectral fluctuations greater than 0.05 are marked as key regulation periods, generating a set of synchronous spectral change characteristic values. This set includes timestamps, predicted reflectance, predicted emissivity, and fluctuation characteristic values. This process maps meteorological forecast data to expected changes in fabric performance parameters, identifies performance degradation or drastic fluctuation windows requiring close monitoring, and provides timing triggering conditions for the generation of control commands.

[0036] S512: Call the synchronous spectral change feature value group, and based on the efficiency sequence of the radiation cooling efficiency layer grid, combine the radiation change amplitude, spectral fluctuation degree and cooling efficiency to extract the inflection point of the continuous time period, filter the segment number of the turning feature and summarize it to obtain the index list of the synchronous segment of the sudden change. The system calls upon the radiation cooling efficiency layer grid efficiency sequence, which records the cooling power of different grid cells under different spectral states, in watts per square meter. Combined with the generated synchronous spectral change characteristic value set, the first derivative of cooling efficiency over time is calculated, and the time point where the derivative sign changes is identified—the inflection point where cooling efficiency changes from increasing to decreasing or vice versa. Sections exhibiting drastic radiation changes near the inflection point, i.e., a change rate exceeding 5% per hour, are selected. These sections with inflection characteristics are then compiled into a list of synchronous abrupt change sections. This list indicates the critical nodes where fabric properties undergo qualitative changes under environmental forcing, prompting the control center to intervene in specific time periods and areas to prevent cooling failure or heat accumulation.

[0037] S513: Based on the list of mutation synchronization segment indexes, match the spectral distribution layer coordinate units corresponding to the segment number, filter and mark overlapping grids with the same index, and output the spectral behavior linkage response control instruction set. Based on the segment number in the mutation synchronization segment index list, it is back-mapped to the specific coordinate unit (x, y) of the spectral distribution layer; it checks whether the coordinate unit belongs to an overlapping grid, that is, a complex region simultaneously affected by multiple environmental factors. For overlapping grids with consistent indices, high-priority control instructions are generated; the instructions include the target coordinates, adjusting the target value such as increasing the reflectivity to 0.95, and the action type such as activating the piezoelectric response layer or changing the MOF aperture. The information is encoded into a binary control instruction stream, and a spectral behavior linkage response control instruction set is output; after executing this instruction set, the cooling efficiency attenuation rate of the fabric under the mutation environment is significantly reduced, realizing an adaptive and rapid response to environmental changes, and verifying the effectiveness of the full-link closed-loop regulation from monitoring, diagnosis, prediction to control.

[0038] A fabric spectral modeling system for MOF spectral manipulation is used to execute the aforementioned fabric spectral modeling method for MOF spectral manipulation. The system includes: The spectral benchmark construction module collects data on the sources of differentiation in the fabric spectral control model, extracts spectral operation data, and combines time and spatial identifiers to format the spectral data and structural factors to generate a spectral response benchmark parameter set. The spectral anomaly detection module is based on the spectral response reference parameter set. According to the spectral distribution state and radiation data, it filters sampling units that exceed the preset threshold, removes interference data associated with anomalies, integrates the remaining spectral distribution state and operation information, marks abnormal areas, and generates a spectral anomaly diagnostic dataset. The spectral behavior extraction module extracts spectral distribution status, radiative cooling change value, and spectral intensity based on the spectral anomaly diagnosis dataset. It identifies sampling units that conform to the spectral pattern, matches spectral behavior templates, overlays location codes and time labels, performs spatial positioning, and generates fabric spectral distribution results. The spectral path tracking module extracts time stamps and spectral operation information based on the fabric spectral distribution results, identifies the spatial distance between spectral behavior regions in adjacent time periods, determines the offset direction and continuous trajectory, integrates path segments with consistent offset directions, and obtains a dynamic response and regulation evolution model of the fabric spectrum. The spectral linkage analysis module is based on the dynamic response and regulation evolution model of fabric spectrum. It compares the time series according to the radiation intensity change trend, spectral response level and radiation cooling efficiency of the path segment, identifies the synchronous turning position, locates the overlapping position of the turning region and spectral behavior, and outputs the spectral behavior linkage response control command set.

[0039] The above embodiments illustrate preferred embodiments of the present invention. Any equivalent adjustments to the technical solution based on software engineering methods are within the scope of protection, including but not limited to: implementing algorithm logic using different programming languages, refactoring functional modules into services, adjusting data interaction protocols, and optimizing resource scheduling strategies. Any implementation scheme derived from reasonable modifications to the data processing flow, service call chain, or system architecture layer without departing from the core technology of the present invention should be considered within the protection scope defined by the technical solution of the present invention.

Claims

1. A fabric spectral modeling method for MOF spectral modulation, characterized in that, Includes the following steps: S1: Based on the initial input of the multi-layer structure of fabric and the spectral response of metal-organic framework materials, the structural parameters and spectral data of the differential sources in the fabric spectral modulation model are restructured and combined with time labels and spatial identifiers to generate a set of spectral response benchmark parameters. S2: Based on the spectral response reference parameter set, analyze the correlation between metal-organic framework loading and spectral reflectance distribution, identify the set of abnormal fluctuation intervals and energy exchange deviation range, filter spectral sampling units that exceed the preset threshold, and generate a spectral anomaly diagnostic dataset. S3: Based on the spectral anomaly diagnostic dataset, extract the spectral distribution state and radiation cooling change value, identify key spectral control nodes and material structure combinations, screen spectral units that conform to the preset standard radiation characteristic model, and superimpose position codes to mark the corresponding regions to generate fabric spectral distribution results. S4: Based on the continuous change information of multiple environmental parameters in the fabric spectral distribution results, identify the spectral shift of adjacent segments of environmental fluctuation, extract the shift direction and response trajectory, screen the spectral segments with the same continuous shift direction, and obtain the fabric spectral dynamic response and regulation evolution model. S5: Based on the fabric spectral dynamic response and regulation evolution model, identify the radiation intensity change trend and spectral response level, construct the radiation cooling efficiency distribution, compare the time synchronization change characteristics between different bands, identify the overlapping positions of the segment coordinates and spectral distributions that occur at the same time, and output the spectral behavior linkage response control instruction set. The spectral behavior linkage response control instruction set includes radiation linkage change zone, spectral response intersection point, and abnormal performance index group.

2. The fabric spectral modeling method for MOF spectral modulation according to claim 1, characterized in that, The spectral response reference parameter set includes a spectral distribution feature value set, a spectral attribute grid, and a structural operation parameter surface. The spectral anomaly diagnostic dataset includes spectral fluctuation identification results, energy deviation markers, radiation energy data, and interference data removal masks. The fabric spectral distribution results include a target spectral candidate set, key node response templates, radiation cooling change values, and regional location codes. The fabric spectral dynamic response and regulation evolution model includes a continuous offset trend trajectory, a path direction vector set, and regional offset nodes.

3. The fabric spectral modeling method for MOF spectral modulation according to claim 1, characterized in that, The specific steps for obtaining the spectral response reference parameter set are as follows: S111: Based on the initial input of metal-organic framework materials and fabric matrix, the raw data from the differential detection channel in the fabric spectral modulation model are collected, and the frame sequence is matched with the time label. The spectral intensity difference between adjacent frames is calculated, and the sampling units in the spectral distribution sequence whose difference between adjacent frames exceeds the fluctuation critical threshold are removed to generate a multidimensional spectral matrix. S112: Based on the multidimensional spectral matrix, extract the reflection intensity and transmission distribution data under the same time and space identifier, perform data integrity verification, filter out records with missing items, and generate a valid spectral running factor data group. S113: Call the effective spectral operating factor data group, identify the standard deviation of the reflection intensity in the sampling unit group with the same spectral operating mode and normalize it, calculate the normalized spectral difference intensity value, reclassify the sampling units in the region, and establish a spectral response benchmark parameter set.

4. The fabric spectral modeling method for MOF spectral modulation according to claim 3, characterized in that, The specific steps for obtaining the spectral anomaly diagnostic dataset are as follows: S211: Based on the spectral response reference parameter set, extract the spectral distribution state and radiation energy data in the target layer, match and compare the fluctuation value of the spectral sampling unit with the preset radiation range, analyze the correspondence, and obtain the spectral fluctuation range data set. S212: Call the spectral fluctuation range data set, combine it with the distribution status of the sampling units in the target layer, determine whether the value of the sampling unit falls into the specified range, identify the difference of the sampling units that are not in the range, determine the objects to be removed based on the difference, and obtain a list of removed sampling unit numbers. S213: Call the list of sampled units to be removed, delete the interference data associated with the corresponding number in the target layer, reorganize the spatial distribution of the remaining sampled units, and generate a spectral anomaly diagnostic dataset.

5. The fabric spectral modeling method for MOF spectral modulation according to claim 4, characterized in that, The specific steps for obtaining the spectral distribution results of the fabric are as follows: S311: Based on the spectral anomaly diagnostic dataset, extract the distribution status and radiation cooling change value of the layer sampling units, combine the differentiated spectral nodes with the corresponding structural parameter data, identify typical spectral mode regions, and obtain the spectral mode characterization combination. S312: Call the spectral mode characterization combination, detect the fluctuation difference of sampling units and the degree of coordinated change of structural parameters in the combined region, aggregate and compare the response values ​​of sampling units at key nodes, identify the region with potential spectral behavior, and obtain the index set of spectral influence sensitive regions. S313: Call the set of spectral influence sensitive area indicators, match the sensitive area sampling unit with the spectral mode characterization combination, filter the sampling units that meet the feature requirements, and mark the location coding information of the corresponding area to generate the fabric spectral distribution result.

6. The fabric spectral modeling method for MOF spectral modulation according to claim 5, characterized in that, The specific steps for obtaining the fabric spectral dynamic response and regulation evolution model are as follows: S411: Based on the spectral distribution results of the fabric, extract the continuous change information between multiple environmental time periods in the spectrum, identify the spectral offset of the geometric center of adjacent regions in the environmental sequence, extract the offset direction vector and displacement length, analyze the angle difference of the consistency of the offset direction between adjacent regions, and obtain a dataset of continuous spectral offset directions. S412: Call the aforementioned continuous spectral offset direction dataset, and based on the changing trend of the angle difference, use the formula: ； Calculate the angular fluctuation dispersion value, extract the region segment with stable offset direction, filter the continuous segment with angular change below the preset threshold, and obtain stable offset segment information; in, Represents the angular fluctuation dispersion value. Representing the The angle value of each data point. Representing the The angle value of each data point. Represents the total number of data points. Represents microclimate data associated with changes in angle. Represents the local rate of change of angle; S413: Call the stable offset segment information, combine the offset direction angle in each segment, the spectral distribution intensity between regions and the environmental sequence continuity index, identify trajectory segments with strong continuity and concentrated spectral signals, fit the evolution characteristics, and obtain the fabric spectral dynamic response and regulation evolution model.

7. The fabric spectral modeling method for MOF spectral modulation according to claim 6, characterized in that, The specific steps for obtaining the spectral behavior linkage response control instruction set are as follows: S511: Based on the fabric spectral dynamic response and regulation evolution model, extract the radiation intensity change trend and spectral response level in the section, identify the daily change value of reflectance and the daily average value of emissivity in the section, analyze the radiation change amplitude and spectral fluctuation degree, and generate a synchronous spectral change characteristic value group. S512: Call the synchronous spectral change feature value group, and extract the inflection points of continuous time periods based on the efficiency sequence of the radiation cooling efficiency layer grid, combined with the radiation change amplitude, spectral fluctuation degree and cooling efficiency. Filter the segment numbers of the turning point features and summarize them to obtain the abrupt change synchronous segment index list. S513: Based on the mutation synchronization segment index list, match the spectral distribution layer coordinate units corresponding to the segment number, filter overlapping grids with consistent indices and mark them, and output the spectral behavior linkage response control instruction set.

8. A fabric spectral modeling system for MOF spectral modulation, characterized in that, The system is used to implement the fabric spectral modeling method for MOF spectral modulation as described in any one of claims 1-7, and the system comprises: The spectral benchmark construction module collects data on the sources of differentiation in the fabric spectral control model, extracts spectral operation data, and combines time and spatial identifiers to format the spectral data and structural factors to generate a spectral response benchmark parameter set. The spectral anomaly detection module, based on the spectral response reference parameter set, filters sampling units that exceed a preset threshold according to the spectral distribution state and radiation data, removes interference data associated with anomalies, integrates the remaining spectral distribution state and operational information, marks abnormal areas, and generates a spectral anomaly diagnostic dataset. Based on the spectral anomaly diagnostic dataset, the spectral behavior extraction module extracts the spectral distribution state, radiative cooling change value, and spectral intensity, identifies sampling units that conform to the spectral pattern, matches the spectral behavior template, overlays the location code and time label, performs spatial positioning, and generates the fabric spectral distribution result. The spectral path tracking module extracts time stamps and spectral operation information based on the fabric's spectral distribution results, identifies the spatial distance between spectral behavior regions in adjacent time periods, determines the offset direction and continuous trajectory, integrates path segments with consistent offset directions, and obtains a dynamic response and regulation evolution model for the fabric's spectral behavior. The spectral linkage analysis module, based on the fabric's dynamic response and regulation evolution model, performs time series comparison according to the radiation intensity change trend, spectral response level, and radiation cooling efficiency of the path segments, identifies synchronous turning points, locates the overlapping positions of turning areas and spectral behavior, and outputs a set of spectral behavior linkage response control instructions.

Images