Method and device for detecting uniformity of non-woven fabric by flash evaporation method

Abstract

The invention discloses a method and device for detecting the uniformity of non-woven fabric through a flash evaporation method, relates to the field of fabric production and detection, and can solve the technical problem that an existing method for detecting the uniformity of the non-woven fabric through the flash evaporation method is large in delay performance. Performing time sequence analysis processing on the rolling pressure data and the light transmission intensity data to obtain a pressure change curve and a light transmission change curve; obtaining a uniformity curve of the detection sample based on an interpolation algorithm according to the pressure change curve and the light transmission change curve; marking outliers of the uniformity curve of the detection sample based on a maximum likelihood estimation method to obtain an outlier set; and according to a density-based clustering algorithm, obtaining a clustering analysis result of the outlier set, and according to the clustering analysis result, obtaining a uniformity detection result of the detection sample so as to adjust process parameters according to the uniformity detection result.

Description

Technical Field

[0001] The present application relates to the field of fabric production detection, and in particular to a method and device for detecting uniformity of flash-evaporated nonwoven fabrics. Background Art

[0002] Flash-steamed nonwoven fabric is a high-performance material that combines the advantages of paper, film and cloth, and has excellent physical properties such as high strength, tear resistance, good waterproof and breathable properties. During its production process, the uniformity of flash-steamed nonwoven fabric is affected by many factors, and the uniformity directly affects its physical properties. Therefore, it is necessary to test the uniformity of flash-steamed nonwoven fabrics to ensure stable product quality.

[0003] The existing flash steamed nonwoven fabric uniformity test method usually uses various measuring tools to measure and calculate the physical parameters such as thickness, mass and density of different parts of the flash steamed nonwoven fabric to evaluate its uniformity. The disadvantage of this type of method is that the detection means are single, and sampling and testing are required after the product is completed. There is a certain delay, and the test results cannot be obtained in time. Timely data is crucial for decision-making in the production process. Therefore, how to reduce the delay of flash steamed nonwoven fabric uniformity testing and improve the timeliness of testing is still a problem that needs to be solved urgently in the existing technology. Summary of the invention

[0004] The present application provides a flash-steamed nonwoven fabric uniformity detection method and device to solve the technical problem of large delay in existing flash-steamed nonwoven fabric uniformity detection methods.

[0005] According to a first aspect of the embodiment of the present application, a method for detecting uniformity of a flash-steamed nonwoven fabric is provided, comprising:

[0006] The rolling pressure data and light transmittance intensity data of the test sample are obtained; wherein the rolling pressure data is the surface pressure collected when the test sample is rolled.

[0007] The rolling pressure data and light transmittance intensity data are subjected to time series analysis and processing to obtain a pressure variation curve and a light transmittance variation curve.

[0008] According to the pressure change curve and the light transmittance change curve, based on the interpolation algorithm, the uniformity curve of the test sample is obtained.

[0009] Based on the maximum likelihood estimation method, outlier points of the uniformity curve of the detected sample are marked to obtain an outlier point set; wherein the outlier point is a point whose uniformity outlier probability is greater than a preset threshold.

[0010] According to the density-based clustering algorithm, the clustering analysis result of the outlier set is obtained, and the uniformity detection result of the detection sample is obtained according to the clustering analysis result, so as to adjust the process parameters according to the uniformity detection result.

[0011] In certain embodiments of the present application, the rolling pressure data and the light transmission intensity data are subjected to time series analysis and processing to obtain a pressure change curve and a light transmission change curve, specifically including:

[0012] The rolling pressure data is subjected to Fourier transformation to obtain first pressure frequency domain data, and the first pressure frequency domain data is filtered out according to a preset vibration frequency threshold to obtain second pressure frequency domain data.

[0013] The light transmission intensity data are decomposed by wavelet transform to obtain a plurality of first light transmission frequency domain data, and the plurality of first light transmission frequency domain data are filtered out based on a median filtering algorithm to obtain a plurality of second light transmission frequency domain data.

[0014] The second pressure frequency domain data is subjected to inverse Fourier transformation to obtain a pressure variation curve, and a plurality of second light transmission frequency domain data are subjected to inverse wavelet transformation synthesis to obtain a light transmission variation curve.

[0015] The present application first performs Fourier transform on the rolling pressure data, and then filters out the second pressure frequency domain data based on a preset vibration frequency threshold, which can filter out the noise caused by mechanical vibration within a specific frequency range; at the same time, the light transmittance intensity data is subjected to wavelet decomposition transform, and then filtered out based on a median filtering algorithm to obtain multiple second light transmittance frequency domain data, which can filter out the peak noise caused by the optical signal within a specific frequency range; and then the second pressure frequency domain data and multiple second light transmittance frequency domain data are subjected to Fourier inverse transform and wavelet inverse transform to synthesize and obtain a pressure change curve and a light transmittance change curve, which can improve data accuracy and reduce data errors.

[0016] In certain embodiments of the present application, obtaining the uniformity curve of the detected sample based on the pressure change curve and the light transmittance change curve based on an interpolation algorithm specifically includes:

[0017] The pressure change curve and the transmittance change curve are aligned according to the timestamps, and the pressure change curve and the transmittance change curve are divided and combined according to a plurality of preset partitions according to a first transmission speed corresponding to the pressure change curve and a second transmission speed corresponding to the transmittance change curve, to obtain a plurality of regional continuous point sets.

[0018] Based on the interpolation algorithm, multiple uniformity curves corresponding to multiple preset partitions are obtained according to the pressure and transmittance of each point in the continuous point set of each area.

[0019] Multiple uniformity curves are spliced ​​in the order of timestamps to obtain the uniformity curve of the test sample.

[0020] The present application first aligns the pressure change curve and the transmittance change curve according to the timestamp, and combines the pressure change curve and the transmittance change curve according to the first transmission speed, the second transmission speed and the preset multiple partition divisions to obtain multiple regional continuous point sets, which can divide the detection area into multiple sub-areas, improve the detection efficiency, and reduce the detection error. Then, based on the interpolation algorithm, multiple uniformity curves corresponding to the preset multiple partitions are obtained, thereby obtaining the uniformity curve of the detection sample. The interpolation algorithm can be used to calculate and smooth the uniformity curve, reduce curve burrs, and then reduce data errors, thereby ensuring data accuracy.

[0021] In certain embodiments of the present application, the method of marking the outliers of the uniformity curve of the detected sample based on the maximum likelihood estimation method to obtain the outlier set specifically includes:

[0022] Based on the preset sampling interval, the uniformity curve of the detection sample is sampled to obtain a sampling point set.

[0023] Based on the maximum likelihood estimation method, the outlier probability of each point in the sampling point set is calculated, and the outlier points are marked according to the outlier probability threshold to obtain the outlier point set.

[0024] The present application first samples the uniformity curve of the test sample based on a preset sampling interval to obtain a sampling point set, and calculates the outlier probability of each point in the sampling point set based on the maximum likelihood estimation method, and then marks and obtains the outlier point set. The calculation of the outlier probability by the maximum likelihood estimation method is more accurate and simple, and can improve the calculation speed while ensuring accuracy, thereby improving the detection efficiency.

[0025] In certain embodiments of the present application, the clustering analysis results of the outlier set are obtained according to the density-based clustering algorithm, and the uniformity test results of the test sample are obtained according to the clustering analysis results, specifically including:

[0026] According to the density-based clustering algorithm, all outlier points in the outlier point set are clustered to obtain multiple clusters.

[0027] According to the number of outliers in each of the multiple clusters and a preset outlier number threshold, cluster analysis results of the multiple clusters are determined, and cluster analysis results of the outlier set are obtained according to the cluster analysis results of the multiple clusters.

[0028] According to the cluster analysis results, the uniformity test results of the test samples are determined.

[0029] The present application first obtains multiple clusters by clustering through a density-based clustering algorithm, and can classify all outliers according to the distribution of the fabric area, and then determine the clustering analysis results based on a preset outlier number threshold. It can intuitively reflect the uniformity change through the number of outliers and determine the uniformity analysis monitoring results. Compared with the complex calculations of the prior art, it is simpler, thereby improving the detection efficiency.

[0030] According to a second aspect of the embodiment of the present application, a flash-evaporation nonwoven fabric uniformity detection device is provided, comprising a data acquisition module, a time series processing module, an interpolation processing module, a likelihood estimation module and a cluster analysis module.

[0031] The data acquisition module is used to acquire the rolling pressure data and light transmittance intensity data of the test sample; wherein the rolling pressure data is the surface pressure collected when the test sample is rolled.

[0032] The time series processing module is used to perform time series analysis and processing on the rolling pressure data and the light transmittance intensity data to obtain a pressure change curve and a light transmittance change curve.

[0033] The interpolation processing module is used to obtain the uniformity curve of the test sample based on the pressure change curve and the light transmittance change curve and the interpolation algorithm.

[0034] The likelihood estimation module is used to mark the outliers of the uniformity curve of the detection sample based on the maximum likelihood estimation method to obtain an outlier set; wherein the outlier is a point whose uniformity outlier probability is greater than a preset threshold.

[0035] The cluster analysis module is used to obtain the cluster analysis results of the outlier set according to the density-based clustering algorithm, and obtain the uniformity test results of the test samples according to the cluster analysis results, so as to adjust the process parameters according to the uniformity test results.

[0036] In certain embodiments of the present application, the timing processing module includes a pressure processing unit, a light intensity processing unit, and a transformation processing unit.

[0037] The pressure processing unit is used to perform Fourier transform on the rolling pressure data to obtain first pressure frequency domain data, and filter the first pressure frequency domain data according to a preset vibration frequency threshold to obtain second pressure frequency domain data.

[0038] The light intensity processing unit is used to perform wavelet transform decomposition on the light transmission intensity data to obtain a plurality of first light transmission frequency domain data, and filter the plurality of first light transmission frequency domain data based on a median filtering algorithm to obtain a plurality of second light transmission frequency domain data.

[0039] The transformation processing unit is used to perform inverse Fourier transformation on the second pressure frequency domain data to obtain a pressure change curve, and perform inverse wavelet transformation synthesis on multiple second light transmission frequency domain data to obtain a light transmission change curve.

[0040] In some embodiments of the present application, the interpolation processing module includes an alignment processing unit, an interpolation processing unit and a splicing processing unit.

[0041] An alignment processing unit is used to align the pressure change curve and the transmittance change curve according to the timestamp, and divide and combine the pressure change curve and the transmittance change curve according to a plurality of preset partitions according to a first transmission speed corresponding to the pressure change curve and a second transmission speed corresponding to the transmittance change curve, so as to obtain a plurality of regional continuous point sets.

[0042] The interpolation processing unit is used to obtain a plurality of uniformity curves corresponding to a plurality of preset partitions based on the pressure and transmittance of each point in the continuous point set of each region based on the interpolation algorithm.

[0043] The splicing processing unit is used to splice multiple uniformity curves in the order of timestamps to obtain the uniformity curve of the test sample.

[0044] In certain embodiments of the present application, the likelihood estimation module includes a curve sampling unit and a likelihood estimation unit.

[0045] The curve sampling unit is used to sample the uniformity curve of the detection sample based on a preset sampling interval to obtain a sampling point set.

[0046] The likelihood estimation unit is used to calculate the outlier probability of each point in the sampling point set based on the maximum likelihood estimation method, and mark the outlier points according to the outlier probability threshold to obtain the outlier point set.

[0047] In certain embodiments of the present application, the clustering analysis module includes a density clustering unit, a clustering analysis unit and a uniformity detection unit.

[0048] The density clustering unit is used to cluster all outlier points in the outlier point set according to a density-based clustering algorithm to obtain multiple cluster clusters.

[0049] The cluster analysis unit is used to determine the cluster analysis results of multiple clusters according to the number of outliers in each of the multiple clusters and a preset outlier number threshold, and obtain the cluster analysis results of the outlier set according to the cluster analysis results of the multiple clusters.

[0050] The uniformity detection unit is used to determine the uniformity detection result of the detection sample according to the cluster analysis result.

[0051] The present application first obtains the rolling pressure data and light transmittance data of the test sample through the pressure sensing array and the light intensity detection array, which can improve the data acquisition speed and reduce the detection delay, and then performs time series analysis and processing on the rolling pressure data and the light transmittance data to obtain the pressure change curve and the light transmittance change curve, and then obtains the uniformity curve of the test sample based on the interpolation algorithm, while considering the pressure exerted on the flash steamed nonwoven fabric during rolling and the light transmittance after rolling. Compared with the prior art that only considers one parameter, it is more comprehensive and can improve the detection confidence and accuracy, and then the outlier point set is marked based on the maximum likelihood estimation method, and the clustering analysis result is obtained based on the density clustering algorithm to obtain the uniformity detection result of the test sample. The direct analysis and statistics are performed by mathematical statistical methods, which can further reduce the detection delay, thereby improving the efficiency of uniformity detection while ensuring the uniformity detection accuracy of the flash steamed nonwoven fabric. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 : A schematic flow chart of a flash steam nonwoven fabric uniformity detection method shown in certain embodiments of the present application;

[0053] Figure 2 : A module structure diagram of a flash nonwoven fabric uniformity detection device shown in certain embodiments of the present application. DETAILED DESCRIPTION

[0054] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions from beginning to end. The embodiments described below in conjunction with the accompanying drawings are exemplary and are only used to explain some embodiments of the present application, and should not be construed as limitations on the embodiments of the present application. Based on the embodiments shown in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.

[0055] In the description of this application, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, unless otherwise clearly and specifically defined, "multiple" and "several" mean two or more.

[0056] It should be understood that the "uniformity" mentioned in the present application refers specifically to the uniformity of the nonwoven fabric, and more specifically refers to the uniformity of the fiber web distribution of the nonwoven fabric.

[0057] See also Figure 1The embodiment of the present application provides a flash nonwoven fabric uniformity detection method, including steps S101 to S105, each step is as follows:

[0058] Step S101: Obtain rolling pressure data and light transmittance data of the test sample; wherein the rolling pressure data is the surface pressure collected when the test sample is rolled.

[0059] In certain embodiments of the present application, the rolling pressure data is acquired by collecting the surface pressure of the test sample during rolling through the pressure sensing array, and the light intensity data is acquired by collecting the light intensity that can be transmitted by the test sample through the light intensity detection array. The number of rows, columns and arrangement of the pressure sensing array are the same as those of the light intensity detection array. Specifically, the number of rows and columns of the pressure sensing array and the light intensity detection array can be set to 10 rows and 10 columns. The test sample can be divided into multiple areas of at least 10×10 according to a preset frequency and a preset speed through the pressure sensing array and the light intensity detection array, and the time series rolling pressure data and light intensity data can be acquired according to a preset time interval.

[0060] Step S102: performing time series analysis on the rolling pressure data and the light transmittance data to obtain a pressure variation curve and a light transmittance variation curve.

[0061] In certain embodiments of the present application, the rolling pressure data and the light transmission intensity data are subjected to time series analysis and processing to obtain a pressure change curve and a light transmission change curve, specifically including:

[0062] The rolling pressure data is subjected to Fourier transformation to obtain first pressure frequency domain data, and the first pressure frequency domain data is filtered out according to a preset vibration frequency threshold to obtain second pressure frequency domain data.

[0063] The light transmission intensity data are decomposed by wavelet transform to obtain a plurality of first light transmission frequency domain data, and the plurality of first light transmission frequency domain data are filtered out based on a median filtering algorithm to obtain a plurality of second light transmission frequency domain data.

[0064] The second pressure frequency domain data is subjected to inverse Fourier transformation to obtain a pressure variation curve, and a plurality of second light transmission frequency domain data are subjected to inverse wavelet transformation synthesis to obtain a light transmission variation curve.

[0065] In certain embodiments of the present application, the preferred scheme for the preset vibration frequency threshold is 60 Hz. According to the preset vibration frequency threshold, the first pressure frequency domain data is filtered out to obtain the second pressure frequency domain data, specifically:

[0066] The data greater than a preset vibration frequency threshold in the first pressure frequency domain data is filtered out to obtain the second pressure frequency domain data.

[0067] In certain embodiments of the present application, the wavelet transform decomposition is based on wavelet basis functions, and the preferred solution is db4 wavelet with 5 decomposition levels.

[0068] In certain embodiments of the present application, the plurality of first light transmission frequency domain data are filtered out based on the median filtering algorithm to obtain the plurality of second light transmission frequency domain data, specifically:

[0069] Data greater than the median transmittance value in the plurality of first light transmission frequency domain data are filtered out to obtain a plurality of second light transmission frequency domain data; wherein the median transmittance value is the median value of a sequence formed by all data in the plurality of first light transmission frequency domain data in order of data size.

[0070] The present application first performs Fourier transform on the rolling pressure data, and then filters out the second pressure frequency domain data based on a preset vibration frequency threshold, which can filter out the noise caused by mechanical vibration within a specific frequency range; at the same time, the light transmittance intensity data is subjected to wavelet decomposition transform, and then filtered out based on a median filtering algorithm to obtain multiple second light transmittance frequency domain data, which can filter out the peak noise caused by the optical signal within a specific frequency range; and then the second pressure frequency domain data and multiple second light transmittance frequency domain data are subjected to Fourier inverse transform and wavelet inverse transform to synthesize and obtain a pressure change curve and a light transmittance change curve, which can improve data accuracy and reduce data errors.

[0071] Step S103: Obtaining a uniformity curve of the test sample according to the pressure change curve and the light transmittance change curve based on an interpolation algorithm.

[0072] In certain embodiments of the present application, obtaining the uniformity curve of the detected sample based on the pressure change curve and the light transmittance change curve based on an interpolation algorithm specifically includes:

[0073] The pressure change curve and the transmittance change curve are aligned according to the timestamps, and the pressure change curve and the transmittance change curve are divided and combined according to a plurality of preset partitions according to a first transmission speed corresponding to the pressure change curve and a second transmission speed corresponding to the transmittance change curve, to obtain a plurality of regional continuous point sets.

[0074] Based on the interpolation algorithm, multiple uniformity curves corresponding to multiple preset partitions are obtained according to the pressure and transmittance of each point in the continuous point set of each area.

[0075] Multiple uniformity curves are spliced ​​in the order of timestamps to obtain the uniformity curve of the test sample.

[0076] In certain embodiments of the present application, according to the first transmission speed corresponding to the pressure change curve and the second transmission speed corresponding to the light transmittance change curve, the pressure change curve and the light transmittance change curve are divided and combined according to a plurality of preset partitions to obtain a plurality of regional continuous point sets, specifically:

[0077] According to the first transmission speed corresponding to the pressure sensing array, the pressure change curve is divided into a plurality of preset partitions to obtain a plurality of regional continuous pressure point sets.

[0078] According to the second transmission speed corresponding to the light intensity detection array, the light transmission change curve is divided into a plurality of preset partitions to obtain a plurality of regional continuous light intensity point sets.

[0079] The points of the multiple regional continuous pressure point sets and the multiple regional continuous light intensity point sets are combined one by one to obtain multiple regional continuous point sets.

[0080] In certain embodiments of the present application, the interpolation algorithm includes but is not limited to Lagrange interpolation, Newton interpolation and cubic spline interpolation.

[0081] In certain embodiments of the present application, the plurality of uniformity curves corresponding to the plurality of preset partitions are obtained according to the pressure and transmittance of each point in the continuous point set of each region, specifically:

[0082] The uniformity of each point in the continuous point set of each region is obtained according to the weighted sum of the logarithm of the pressure and the logarithm of the transmittance of each point in the continuous point set of each region.

[0083] According to the uniformity of each point in the continuous point set of each region, a plurality of uniformity curves corresponding to the preset plurality of partitions are obtained.

[0084] In certain embodiments of the present application, a preferred implementation of the weighted sum of the logarithm of pressure and the logarithm of transmittance is: the weight of the logarithm of pressure is 0.5, and the weight of the logarithm of transmittance is 0.5.

[0085] The present application first aligns the pressure change curve and the transmittance change curve according to the timestamp, and combines the pressure change curve and the transmittance change curve according to the first transmission speed, the second transmission speed and the preset multiple partition divisions to obtain multiple regional continuous point sets, which can divide the detection area into multiple sub-areas, improve the detection efficiency, and reduce the detection error. Then, based on the interpolation algorithm, multiple uniformity curves corresponding to the preset multiple partitions are obtained, thereby obtaining the uniformity curve of the detection sample. The interpolation algorithm can be used to calculate and smooth the uniformity curve, reduce curve burrs, and then reduce data errors, thereby ensuring data accuracy.

[0086] Step S104: Based on the maximum likelihood estimation method, outlier points of the uniformity curve of the test sample are marked to obtain an outlier point set; wherein the outlier point is a point whose uniformity outlier probability is greater than a preset threshold.

[0087] In certain embodiments of the present application, the method of marking the outliers of the uniformity curve of the detected sample based on the maximum likelihood estimation method to obtain the outlier set specifically includes:

[0088] Based on the preset sampling interval, the uniformity curve of the detection sample is sampled to obtain a sampling point set.

[0089] Based on the maximum likelihood estimation method, the outlier probability of each point in the sampling point set is calculated, and the outlier points are marked according to the outlier probability threshold to obtain the outlier point set.

[0090] In certain embodiments of the present application, a preferred implementation of the preset sampling interval is to sample once every 5 points.

[0091] In certain embodiments of the present application, the maximum likelihood estimation method is used to calculate the outlier probability of each point in the sampling point set, and the outlier points are marked according to the outlier probability threshold to obtain the outlier point set, specifically:

[0092] The mean and standard deviation of all sampling points in the sampling point set are calculated, and the outlier probability of each point in the sampling point set is calculated based on the maximum likelihood estimation method according to the mean and standard deviation.

[0093] All sampling points in the sampling point set whose outlier probability is greater than the outlier probability threshold are marked to obtain the outlier point set.

[0094] In certain embodiments of the present application, the preferred value of the outlier probability threshold is 0.6.

[0095] The present application first samples the uniformity curve of the test sample based on a preset sampling interval to obtain a sampling point set, and calculates the outlier probability of each point in the sampling point set based on the maximum likelihood estimation method, and then marks and obtains the outlier point set. The calculation of the outlier probability by the maximum likelihood estimation method is more accurate and simple, and can improve the calculation speed while ensuring accuracy, thereby improving the detection efficiency.

[0096] Step S105: Obtain a cluster analysis result of the outlier set according to a density-based clustering algorithm, and obtain a uniformity test result of the test sample according to the cluster analysis result, so as to adjust the process parameters according to the uniformity test result.

[0097] In certain embodiments of the present application, the clustering analysis results of the outlier set are obtained according to the density-based clustering algorithm, and the uniformity test results of the test sample are obtained according to the clustering analysis results, specifically including:

[0098] According to the density-based clustering algorithm, all outlier points in the outlier point set are clustered to obtain multiple clusters.

[0099] According to the number of outliers in each of the multiple clusters and a preset outlier number threshold, cluster analysis results of the multiple clusters are determined, and cluster analysis results of the outlier set are obtained according to the cluster analysis results of the multiple clusters.

[0100] According to the cluster analysis results, the uniformity test results of the test samples are determined.

[0101] In certain embodiments of the present application, the preset outlier number threshold is preferably set to 10.

[0102] In certain embodiments of the present application, the clustering analysis results of the multiple clusters are determined according to the number of outliers of each of the multiple clusters and the preset outlier number threshold, specifically:

[0103] If the number of outliers in each of the multiple clusters exceeds the preset outlier number threshold, the cluster analysis result of the corresponding cluster is "unqualified uniformity".

[0104] In certain embodiments of the present application, the clustering analysis results of the outlier set are obtained according to the clustering analysis results of the multiple cluster clusters, specifically:

[0105] If the cluster analysis result of any cluster among the multiple clusters is "unqualified uniformity", the cluster analysis result of the outlier set is "unqualified uniformity" and the area corresponding to the cluster whose cluster analysis result is "unqualified uniformity" among the multiple clusters.

[0106] In certain embodiments of the present application, the uniformity test result of the test sample is a cluster analysis result of an outlier set.

[0107] The present application first obtains multiple clusters by clustering through a density-based clustering algorithm, and can classify all outliers according to the distribution of the fabric area, and then determine the clustering analysis results based on a preset outlier number threshold. It can intuitively reflect the uniformity change through the number of outliers and determine the uniformity analysis monitoring results. Compared with the complex calculations of the prior art, it is simpler, thereby improving the detection efficiency.

[0108] Compared with the prior art, the present application first obtains the rolling pressure data and light transmittance data of the test sample through the pressure sensing array and the light intensity detection array, which can improve the data acquisition speed and reduce the detection delay, and then performs time series analysis and processing on the rolling pressure data and the light transmittance data to obtain the pressure change curve and the light transmittance change curve, and then obtains the uniformity curve of the test sample based on the interpolation algorithm, while considering the pressure exerted on the flash steamed nonwoven fabric during rolling and the light transmittance after rolling. Compared with the prior art that only considers one parameter, it is more comprehensive and can improve the detection confidence and accuracy, and then the outlier point set is marked based on the maximum likelihood estimation method, and the clustering analysis result is obtained based on the density clustering algorithm to obtain the uniformity detection result of the test sample. The detection delay can be further reduced by directly performing analysis and statistics through mathematical statistical methods, thereby improving the efficiency of uniformity detection while ensuring the uniformity detection accuracy of the flash steamed nonwoven fabric.

[0109] Corresponding to the previous method, see Figure 2 The embodiment of the present application provides a flash nonwoven fabric uniformity detection device, including a data acquisition module 210, a time series processing module 220, an interpolation processing module 230, a likelihood estimation module 240 and a cluster analysis module 250.

[0110] The data acquisition module 210 is used to acquire the rolling pressure data and light transmission intensity data of the test sample; wherein the rolling pressure data is the surface pressure collected when the test sample is rolled; and the light transmission intensity data is obtained by detecting the test sample through a light intensity detection array.

[0111] The time series processing module 220 is used to perform time series analysis and processing on the rolling pressure data and the light transmission intensity data to obtain a pressure change curve and a light transmission change curve.

[0112] The interpolation processing module 230 is used to obtain a uniformity curve of the detection sample based on an interpolation algorithm according to the pressure change curve and the light transmittance change curve.

[0113] The likelihood estimation module 240 is used to mark the outliers of the uniformity curve of the detected sample based on the maximum likelihood estimation method to obtain an outlier set; wherein the outliers are points whose uniformity outlier probability is greater than a preset threshold.

[0114] The cluster analysis module 250 is used to obtain cluster analysis results of outlier points according to a density-based clustering algorithm, and obtain uniformity test results of the test samples according to the cluster analysis results, so as to adjust the process parameters according to the uniformity test results.

[0115] In some embodiments of the present application, the timing processing module 220 includes a pressure processing unit, a light intensity processing unit, and a transformation processing unit.

[0116] The pressure processing unit is used to perform Fourier transform on the rolling pressure data to obtain first pressure frequency domain data, and filter the first pressure frequency domain data according to a preset vibration frequency threshold to obtain second pressure frequency domain data.

[0117] The light intensity processing unit is used to perform wavelet transform decomposition on the light transmission intensity data to obtain a plurality of first light transmission frequency domain data, and filter the plurality of first light transmission frequency domain data based on a median filtering algorithm to obtain a plurality of second light transmission frequency domain data.

[0118] The transformation processing unit is used to perform inverse Fourier transformation on the second pressure frequency domain data to obtain a pressure change curve, and perform inverse wavelet transformation synthesis on multiple second light transmission frequency domain data to obtain a light transmission change curve.

[0119] In some embodiments of the present application, the interpolation processing module 230 includes an alignment processing unit, an interpolation processing unit, and a splicing processing unit.

[0120] An alignment processing unit is used to align the pressure change curve and the transmittance change curve according to the timestamp, and divide and combine the pressure change curve and the transmittance change curve according to a plurality of preset partitions according to a first transmission speed corresponding to the pressure change curve and a second transmission speed corresponding to the transmittance change curve, so as to obtain a plurality of regional continuous point sets.

[0121] The interpolation processing unit is used to obtain a plurality of uniformity curves corresponding to a plurality of preset partitions based on the pressure and transmittance of each point in the continuous point set of each region based on an interpolation algorithm.

[0122] The splicing processing unit is used to splice multiple uniformity curves in the order of timestamps to obtain the uniformity curve of the test sample.

[0123] In some embodiments of the present application, the likelihood estimation module 240 includes a curve sampling unit and a likelihood estimation unit.

[0124] The curve sampling unit is used to sample the uniformity curve of the detection sample based on a preset sampling interval to obtain a sampling point set.

[0125] The likelihood estimation unit is used to calculate the outlier probability of each point in the sampling point set based on the maximum likelihood estimation method, and mark the outlier points according to the outlier probability threshold to obtain the outlier point set.

[0126] In some embodiments of the present application, the clustering analysis module 250 includes a density clustering unit, a clustering analysis unit, and a uniformity detection unit.

[0127] The density clustering unit is used to cluster all outlier points in the outlier point set according to a density-based clustering algorithm to obtain multiple cluster clusters.

[0128] The cluster analysis unit is used to determine the cluster analysis results of multiple clusters according to the number of outliers in each of the multiple clusters and a preset outlier number threshold, and obtain the cluster analysis results of the outlier set according to the cluster analysis results of the multiple clusters.

[0129] The uniformity detection unit is used to determine the uniformity detection result of the detection sample according to the cluster analysis result.

[0130] The present application first obtains the rolling pressure data and light transmittance data of the test sample through the pressure sensing array and the light intensity detection array, which can improve the data acquisition speed and reduce the detection delay, and then performs time series analysis and processing on the rolling pressure data and the light transmittance data to obtain the pressure change curve and the light transmittance change curve, and then obtains the uniformity curve of the test sample based on the interpolation algorithm, while considering the pressure exerted on the flash steamed nonwoven fabric during rolling and the light transmittance after rolling. Compared with the prior art that only considers one parameter, it is more comprehensive and can improve the detection confidence and accuracy, and then the outlier point set is marked based on the maximum likelihood estimation method, and the clustering analysis result is obtained based on the density clustering algorithm to obtain the uniformity detection result of the test sample. The direct analysis and statistics are performed by mathematical statistical methods, which can further reduce the detection delay, thereby improving the efficiency of uniformity detection while ensuring the uniformity detection accuracy of the flash steamed nonwoven fabric.

[0131] It should be understood that the device provided in the embodiment of the present application corresponds to the aforementioned method, and the flash steamed nonwoven fabric uniformity detection device provided in the embodiment of the present application can implement the flash steamed nonwoven fabric uniformity detection method provided in any embodiment of the present application.

[0132] Adaptively, the embodiments of the present application also provide a computer device and a computer-readable storage medium.

[0133] The computer device comprises: a processor, a memory and a computer program stored in the memory and configured to be executed by the processor; wherein, when the processor executes the computer program, a flash steamed nonwoven fabric uniformity detection method of the present application is implemented.

[0134] The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute a flash-steamed nonwoven fabric uniformity detection method of the present application.

[0135] The above are some implementation methods of the present application, and the purpose, technical scheme and beneficial effects of the present application are further described in detail. It should be clear that the above-mentioned partial implementation methods of the present application cannot be understood as limiting the present application. It is particularly pointed out that for those skilled in the art, any changes, modifications, equivalent substitutions and variations made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A flash nonwoven fabric uniformity detection method, characterized in that: include: Acquire rolling pressure data and light transmission intensity data of the test sample; wherein the rolling pressure data is the surface pressure collected when the test sample is rolled; Performing time series analysis on the rolling pressure data and the light transmission intensity data to obtain a pressure change curve and a light transmission change curve; According to the pressure change curve and the light transmittance change curve, based on an interpolation algorithm, a uniformity curve of the test sample is obtained; Based on the maximum likelihood estimation method, outliers of the uniformity curve of the test sample are marked to obtain an outlier set; wherein the outliers are points whose uniformity outlier probability is greater than a preset threshold; According to a density-based clustering algorithm, a clustering analysis result of the outlier set is obtained, and according to the clustering analysis result, a uniformity detection result of the detection sample is obtained, so as to adjust the process parameters according to the uniformity detection result.

2. The method for detecting uniformity of flash-steamed nonwoven fabrics according to claim 1, characterized in that: The performing time series analysis on the rolling pressure data and the light transmission intensity data to obtain a pressure change curve and a light transmission change curve specifically includes: Performing Fourier transformation on the rolling pressure data to obtain first pressure frequency domain data, and filtering the first pressure frequency domain data according to a preset vibration frequency threshold to obtain second pressure frequency domain data; Performing wavelet transform decomposition on the light transmission intensity data to obtain a plurality of first light transmission frequency domain data, and filtering the plurality of first light transmission frequency domain data based on a median filter algorithm to obtain a plurality of second light transmission frequency domain data; The second pressure frequency domain data is subjected to inverse Fourier transformation to obtain a pressure variation curve, and the plurality of second light transmission frequency domain data are subjected to inverse wavelet transformation synthesis to obtain a light transmission variation curve.

3. The method for detecting uniformity of flash-steamed nonwoven fabrics according to claim 1, characterized in that: The step of obtaining the uniformity curve of the detected sample based on the pressure change curve and the light transmittance change curve and an interpolation algorithm specifically includes: The pressure change curve and the light transmittance change curve are aligned according to the timestamps, and the pressure change curve and the light transmittance change curve are divided and combined according to a plurality of preset partitions according to a first transmission speed corresponding to the pressure change curve and a second transmission speed corresponding to the light transmittance change curve, so as to obtain a plurality of regional continuous point sets; Based on the interpolation algorithm, a plurality of uniformity curves corresponding to the plurality of preset partitions are obtained according to the pressure and transmittance of each point in the continuous point set of each region; The multiple uniformity curves are spliced ​​in the order of timestamps to obtain the uniformity curve of the detected sample.

4. The method for detecting uniformity of flash-steamed nonwoven fabrics according to claim 1, characterized in that: The method of marking the outliers of the uniformity curve of the detected sample based on the maximum likelihood estimation method to obtain an outlier set specifically includes: Based on a preset sampling interval, sampling the uniformity curve of the detection sample to obtain a sampling point set; Based on the maximum likelihood estimation method, the outlier probability of each point in the sampling point set is calculated, and the outlier points are marked according to the outlier probability threshold to obtain an outlier point set.

5. The method for detecting uniformity of flash-steamed nonwoven fabrics according to claim 1, characterized in that: The step of obtaining the cluster analysis result of the outlier set according to the density-based clustering algorithm, and obtaining the uniformity test result of the test sample according to the cluster analysis result, specifically includes: Clustering all outlier points in the outlier point set according to a density-based clustering algorithm to obtain a plurality of clusters; Determining cluster analysis results of the multiple clusters according to the number of outliers of each of the multiple clusters and a preset outlier number threshold, and obtaining a cluster analysis result of the outlier set according to the cluster analysis results of the multiple clusters; According to the cluster analysis result, the uniformity test result of the test sample is determined.

6. A flash nonwoven fabric uniformity detection device, characterized in that: It includes a data acquisition module, a time series processing module, an interpolation processing module, a likelihood estimation module and a cluster analysis module; The data acquisition module is used to acquire rolling pressure data and light transmission intensity data of the test sample; wherein the rolling pressure data is the surface pressure collected when the test sample is rolled; The time series processing module is used to perform time series analysis on the rolling pressure data and the light transmission intensity data to obtain a pressure change curve and a light transmission change curve; The interpolation processing module is used to obtain the uniformity curve of the detection sample based on the pressure change curve and the light transmittance change curve and the interpolation algorithm; The likelihood estimation module is used to mark the outliers of the uniformity curve of the test sample based on the maximum likelihood estimation method to obtain an outlier set; wherein the outliers are points whose uniformity outlier probability is greater than a preset threshold; The cluster analysis module is used to obtain the cluster analysis result of the outlier set according to the density-based clustering algorithm, and obtain the uniformity detection result of the detection sample according to the cluster analysis result, so as to adjust the process parameters according to the uniformity detection result.

7. A flash-steamed nonwoven fabric uniformity detection device according to claim 6, characterized in that: The timing processing module includes a pressure processing unit, a light intensity processing unit and a transformation processing unit; The pressure processing unit is used to perform Fourier transform on the rolling pressure data to obtain first pressure frequency domain data, and filter the first pressure frequency domain data according to a preset vibration frequency threshold to obtain second pressure frequency domain data; The light intensity processing unit is used to perform wavelet transform decomposition on the light transmission intensity data to obtain a plurality of first light transmission frequency domain data, and filter the plurality of first light transmission frequency domain data based on a median filter algorithm to obtain a plurality of second light transmission frequency domain data; The transformation processing unit is used to perform inverse Fourier transformation on the second pressure frequency domain data to obtain a pressure change curve, and perform inverse wavelet transformation synthesis on the multiple second light transmission frequency domain data to obtain a light transmission change curve.

8. The flash-steamed nonwoven fabric uniformity detection device according to claim 6, characterized in that: The interpolation processing module includes an alignment processing unit, an interpolation processing unit and a splicing processing unit; The alignment processing unit is used to align the pressure change curve and the light transmittance change curve according to the timestamp, and divide and combine the pressure change curve and the light transmittance change curve according to a plurality of preset partitions according to a first transmission speed corresponding to the pressure change curve and a second transmission speed corresponding to the light transmittance change curve, so as to obtain a plurality of regional continuous point sets; The interpolation processing unit is used to obtain a plurality of uniformity curves corresponding to the plurality of preset partitions based on the pressure and transmittance of each point in the continuous point set of each region based on an interpolation algorithm; The splicing processing unit is used to splice the multiple uniformity curves in the order of timestamps to obtain the uniformity curve of the detection sample.

9. The flash-steamed nonwoven fabric uniformity detection device according to claim 6, characterized in that: The likelihood estimation module includes a curve sampling unit and a likelihood estimation unit; The curve sampling unit is used to sample the uniformity curve of the detection sample based on a preset sampling interval to obtain a sampling point set; The likelihood estimation unit is used to calculate the outlier probability of each point in the sampling point set based on the maximum likelihood estimation method, and mark the outlier points according to the outlier probability threshold to obtain the outlier point set.

10. The flash-steamed nonwoven fabric uniformity detection device according to claim 6, characterized in that: The cluster analysis module includes a density clustering unit, a cluster analysis unit and a uniformity detection unit; The density clustering unit is used to cluster all outliers in the outlier point set according to a density-based clustering algorithm to obtain a plurality of clusters; The cluster analysis unit is used to determine the cluster analysis results of the multiple clusters according to the number of outliers in each of the multiple clusters and a preset outlier number threshold, and obtain the cluster analysis result of the outlier set according to the cluster analysis results of the multiple clusters; The uniformity detection unit is used to determine the uniformity detection result of the detection sample according to the cluster analysis result.

Images