Cosmetic preservative detection method based on solid phase extraction

Abstract

The invention discloses a cosmetic preservative detection method based on solid phase extraction, and relates to the technical field of cosmetic detection.The method comprises the steps that M pieces of batch information of M production batches of cosmetics to be detected is obtained; obtaining M fusion quality detection information neighborhood sets; determining M sample extraction sub-schemes; analyzing the quality detection information sets in the M fusion sets and preset solid-phase extraction parameters by using an extraction parameter recognizer to obtain M adjusted solid-phase extraction parameters; m to-be-detected sample sets are obtained, and extraction, washing and elution are performed to obtain M sample extraction solution sets; m sample preservative type sets and M sample preservative concentration sets are obtained; and obtaining a preservative detection result of the to-be-detected cosmetic. The technical problem that the detection result is inaccurate due to low sample selection reliability during cosmetic preservative detection in the prior art is solved, and the technical effects of improving the preservative detection efficiency and the detection result accuracy are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of cosmetics detection, and in particular to a cosmetics preservative detection method based on solid phase extraction. Background Art

[0002] At present, traditional cosmetic preservative testing usually involves random sampling without fully considering the production volume, raw material ratio and quality fluctuations of each batch. If the samples in a batch fail to cover all possible quality differences due to random selection, potential problems in certain batches may be overlooked. For example, in a certain cosmetic batch, there is a significant fluctuation in the preservative content, but random sampling fails to select the problematic sample. The final test results may show that the preservative content meets the standard, but in fact there are quality risks. In this case, insufficient sample extraction leads to distortion of the test results, which cannot accurately reflect the preservative situation of the overall product.

[0003] The existing technology has the technical problem of low reliability of sample selection when testing cosmetic preservatives, resulting in inaccurate test results. Summary of the invention

[0004] The present application provides a cosmetic preservative detection method based on solid phase extraction, which is used to solve the technical problem of low reliability of sample selection in cosmetic preservative detection in the prior art, resulting in inaccurate detection results.

[0005] In view of the above problems, the present application provides a method for detecting cosmetic preservatives based on solid phase extraction, the method comprising: Obtaining M batch information of M production batches of the cosmetics to be tested, wherein the M batch information includes M product sequences, M quality inspection information sequences and M batch outputs; Performing neighbor quality division according to the degree of difference of adjacent quality detection information in the M quality detection information sequences, and performing neighborhood fusion on the division results to obtain M fused quality detection information neighborhood sets, wherein the M fused quality detection information neighborhood sets include M fused concentrated quality detection information sets; Interactively preset the preservative detection scheme to extract the sample detection quantity, combine the M batch outputs and the M fusion quality detection information neighborhood sets to identify the sample extraction scheme, and determine M sample extraction sub-schemes; Using an extraction parameter identifier to analyze the M fused centralized quality detection information sets and preset solid phase extraction parameters to obtain M adjusted solid phase extraction parameters; Based on the M sample extraction sub-schemes, the M product sequences are respectively subjected to sample extraction to obtain M sets of samples to be tested, and M solid phase extraction devices are configured according to the M adjusted solid phase extraction parameters, and the M sets of samples to be tested are respectively and slowly loaded onto the M activated solid phase extraction columns of the M solid phase extraction devices in sequence, and after the extraction time requirement is met, the M solid phase extraction columns are washed and eluted to obtain M sets of sample extraction solutions; The M sample extraction solution sets are sequentially sent to a preservative detection instrument to detect the preservative content, thereby obtaining M sample preservative type sets and M sample preservative concentration sets; Taking the preservative type as an index, the M sample preservative type sets and the M sample preservative concentration sets are subjected to similar aggregation mean processing to obtain the preservative detection result of the cosmetic to be tested, wherein the preservative detection result includes multiple preservative concentration means of multiple preservative types.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: The present application obtains M batch information of M production batches of the cosmetics to be tested, wherein the M batch information includes M product sequences, M quality inspection information sequences and M batch outputs, and then performs neighbor quality division according to the degree of difference of adjacent quality inspection information in the M quality inspection information sequences, and performs neighborhood fusion on the division results to obtain M fused quality inspection information neighborhood sets, wherein the M fused quality inspection information neighborhood sets include M fused centralized quality inspection information sets, and then interactively preset preservative detection schemes to extract the number of sample detections, combine the M batch outputs and the M fused quality inspection information neighborhood sets to identify the sample extraction scheme, determine M sample extraction sub-schemes, and then use an extraction parameter identifier to analyze the M fused centralized quality inspection information sets and preset solid phase extraction parameters to obtain M adjusted solid phase extraction parameters, based on the M samples. The product extraction sub-scheme extracts samples from M product sequences to obtain M sets of samples to be tested, and configures M solid phase extraction devices according to M adjusted solid phase extraction parameters, and slowly loads the M sets of samples to be tested onto the activated M solid phase extraction columns of the M solid phase extraction devices in sequence, and after the extraction time requirements are met, washes and elutes the M solid phase extraction columns to obtain M sets of sample extraction solutions; the M sets of sample extraction solutions are sequentially sent to the preservative detection instrument for preservative content detection to obtain M sets of sample preservative types and M sets of sample preservative concentrations, and the M sets of sample preservative types and M sets of sample preservative concentrations are processed by the same type of aggregation mean with the preservative type as the index to obtain the preservative detection results of the cosmetics to be tested, wherein the preservative detection results include multiple preservative concentration means of multiple preservative types. The technical effect of performing differentiated sample extraction on multiple batches of cosmetics and improving the efficiency of preservative detection and the reliability of detection results is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0008] Figure 1 A schematic diagram of a process for detecting cosmetic preservatives based on solid phase extraction provided in an embodiment of the present application; Figure 2 A schematic diagram of a process for obtaining M fused quality detection information neighborhood sets in a cosmetic preservative detection method based on solid phase extraction provided in an embodiment of the present application. DETAILED DESCRIPTION

[0009] The present application provides a cosmetic preservative detection method based on solid phase extraction, which is used to solve the technical problem of low reliability of sample selection in cosmetic preservative detection in the prior art, resulting in inaccurate detection results.

[0010] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0011] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0012] Embodiment 1, as Figure 1 As shown, the present application provides a method for detecting cosmetic preservatives based on solid phase extraction, wherein the method comprises: S100: Obtaining M batch information of M production batches of the cosmetics to be tested, wherein the M batch information includes M product sequences, M quality inspection information sequences and M batch outputs; In a possible embodiment, the cosmetics to be tested are any cosmetics that need to be tested for preservatives, such as liquid foundation, lotion, or blush. When the cosmetics to be tested are produced in large quantities, there are multiple production batches. In the production process, due to factors such as the state of production equipment, quality of raw materials, mixing time, and temperature control, the quality of products produced on the same production line at different production times also varies greatly. Therefore, the production information of each batch is extracted to obtain the M batch information. Among them, the M batch information reflects the cosmetic production situation of the cosmetics to be tested in the M production batches.

[0013] Optionally, the M batch information includes M product sequences, M quality inspection information sequences and M batch outputs. Among them, the product sequence is a sequence of cosmetics produced in the batch determined in chronological order according to the production time, and each product has a unique identification code. The M quality inspection information sequences are historical quality inspection information of M production batches obtained from the quality control department or the testing center. Preferably, each quality inspection information contains physical properties (such as pH value, viscosity), chemical properties (such as preservative content, active ingredient concentration), and microbial test results. The M batch outputs reflect the processing volume of M production batches. Exemplarily, the batch output can be specific data such as "batch 1: 5000 bottles, batch 2: 7000 bottles".

[0014] By obtaining the M batch information and describing the batch sources of the cosmetics to be tested, the product conditions of each production batch are systematically and completely collected, thus achieving the technical effect of paving the way for subsequent reliable preservative testing.

[0015] S200: performing neighbor quality division according to the degree of difference of adjacent quality detection information in the M quality detection information sequences, and performing neighborhood fusion on the division results to obtain M fused quality detection information neighborhood sets, wherein the M fused quality detection information neighborhood sets include M fused concentrated quality detection information sets; Further, such as Figure 2 As shown, neighbor quality division is performed according to the degree of difference of adjacent quality detection information in the M quality detection information sequences, and neighborhood fusion is performed on the division results to obtain M fused quality detection information neighborhood sets. Step S200 of the embodiment of the present application also includes: Taking the M median quality detection information of the M quality detection information sequences as M division starting points respectively, and constructing initial division neighborhoods to the left and right sides respectively according to a preset step length, to obtain M left initial division neighborhoods and M right initial division neighborhoods; Traversing the M left initial division neighborhoods and the M right initial division neighborhoods to perform neighbor quality difference authentication, and determining M left differences and M right differences; Determine whether the difference between the M left difference degrees and the M right difference degrees is less than or equal to a preset difference threshold value, if not, generate M division ridge lines at the M division starting points, and use the quality detection information located at the left end of the M left initial division neighborhoods as the M left division starting points, and use the quality detection information located at the right end of the M right initial division neighborhoods as the M right division starting points; Based on the M left division starting points and the M right division starting points, respectively, neighbor quality division is performed toward the left and right sides according to the preset step length until the first and last positions of the M quality detection information sequences are reached, and the division is stopped to obtain M division ridge line sets, and the M quality detection information sequences are divided based on the M division ridge line sets to obtain M initial division neighborhood sets; The number of information in the initial partitioned neighborhoods of the M initial partitioned neighborhood sets is traversed and counted, and the M initial partitioned neighborhood sets are subjected to neighborhood fusion according to the statistical results to obtain the M fused quality detection information neighborhood sets, and the mean of the M fused quality detection information neighborhood sets is calculated to determine the M fused concentrated quality detection information sets.

[0016] Furthermore, step S200 of the embodiment of the present application further includes: If so, the M left initial division neighborhoods and the M right initial division neighborhoods are merged into one neighborhood to obtain M initial division neighborhoods, and the quality detection information located at both ends of the M initial division neighborhoods is updated to M left division starting points and M right division starting points respectively.

[0017] Furthermore, step S200 of the embodiment of the present application further includes: Extracting M sets of initial partitioned neighborhoods to be merged whose number is less than a preset number threshold from the statistical results; The M initial partition neighborhood sets to be fused are integrated into the initial partition areas corresponding to the adjacent minimum difference values, and the M initial partition neighborhood sets are updated according to the fusion results to obtain the M fusion quality detection information neighborhood sets.

[0018] In one possible embodiment, in order to improve the efficiency and accuracy of preservative detection, it is necessary to master the distribution of product quality in M ​​production batches. Therefore, according to the degree of difference between adjacent quality detection information in the M quality detection information sequences, it is determined whether adjacent quality detection information can be divided into the same neighborhood. In other words, when the degree of difference is large, two adjacent quality detection information may represent two production states of the production batch and need to be divided into different neighborhoods. When the degree of difference is small, it indicates that the production state of the production batches corresponding to the adjacent quality detection information at these two production nodes does not change much. When sampling is performed, the two adjacent quality detection information can be divided into the same neighborhood for random sampling, which has little impact on the accuracy of the test results.

[0019] By performing neighborhood fusion on the division results, the M fusion quality detection information neighborhood sets are obtained. Then, the mean value of each fusion quality detection information neighborhood is calculated to determine the average quality level of each neighborhood, thereby avoiding excessive dispersion of data and increasing sampling difficulty, and achieving the technical effect of improving subsequent sampling efficiency.

[0020] In one embodiment, the median values ​​of the M quality detection information sequences, that is, the M median quality detection information, are taken as M division starting points, and initial division neighborhood construction is performed on the left and right sides according to a preset step length (the number of positions of a single neighborhood movement preset by a person skilled in the art), to obtain M left initial division neighborhoods and M right initial division neighborhoods. Among them, the M left initial division neighborhoods contain a number of quality detection information corresponding to the preset step length. Similarly, the M right initial division neighborhoods contain a number of quality detection information corresponding to the preset step length.

[0021] Traverse the M left initial division neighborhoods and the M right initial division neighborhoods to perform neighborhood quality difference authentication. Optionally, respectively calculate the variance of the quality detection information in the M left initial division neighborhoods and the variance of the quality detection information in the M right initial division neighborhoods to obtain the M left differences and the M right differences. The M left differences reflect the degree of fluctuation of the quality detection information in the M left initial division neighborhoods. The M right differences reflect the degree of fluctuation of the quality detection information in the M right initial division neighborhoods.

[0022] Determine whether the difference between the M left-side differences and the M right-side differences is less than or equal to a preset difference threshold; if not, it indicates that the neighborhoods on both sides of the M division starting points cannot be fused and the two neighborhoods need to be separated; therefore, M division ridge lines are generated at the M division starting points, and the quality detection information at the left end of the M left initial division neighborhoods is used as the M left-side division starting points, and the quality detection information at the right end of the M right-side initial division neighborhoods is used as the M right-side division starting points.

[0023] If so, it indicates that the degree of fluctuation of the quality detection information on the left and right sides of the M division starting points is not much different. At this time, the M left initial division neighborhoods and the M right initial division neighborhoods can be merged into one neighborhood to obtain M initial division neighborhoods, and the quality detection information located at both ends of the M initial division neighborhoods are updated to M left division starting points and M right division starting points respectively.

[0024] Then, according to the preset step size, the M left-side division starting points and the M right-side division starting points are moved to the left and right sides respectively, and neighbor quality division is performed according to the same division principle as above, until the first and last positions of the M quality detection information sequences are reached, and the division is stopped to obtain M division ridge line sets, and the M quality detection information sequences are divided based on the M division ridge line sets to obtain M initial division neighborhood sets.

[0025] After obtaining the M initial partition neighborhood sets, in order to improve analysis efficiency and reduce the amount of analysis data, neighborhood fusion is performed on the M initial partition neighborhood sets according to the amount of information of each initial partition neighborhood in the M initial partition neighborhood sets.

[0026] Preferably, the number of information in the initial partition neighborhood of the M initial partition neighborhood sets is traversed and counted, and the M initial partition neighborhood sets are subjected to neighborhood fusion according to the statistical results to obtain the M fusion quality detection information neighborhood sets. The M initial partition neighborhood sets to be fused whose number is less than a preset number threshold in the statistical results are extracted, and then the M initial partition neighborhood sets to be fused are integrated into the initial partition domain corresponding to the minimum difference value of the adjacent difference degree, and the M initial partition neighborhood sets are updated according to the fusion results to obtain the M fusion quality detection information neighborhood sets.

[0027] Then, the mean of the M fusion quality detection information neighborhood sets is calculated to determine the average quality of each fusion quality detection information neighborhood, and M fusion quality detection information sets are obtained. The technical effect of performing neighborhood division and fusion on the M quality detection information sequences and improving the reliability and efficiency of subsequent detection sampling is achieved.

[0028] S300: interactively preset a preservative detection scheme to extract the number of sample tests, identify a sample extraction scheme in combination with the M batch outputs and the M fusion quality detection information neighborhood sets, and determine M sample extraction sub-schemes; Further, the interactive preset preservative detection scheme is used to extract the sample detection quantity, the sample extraction scheme is identified by combining the M batch outputs and the M fusion quality detection information neighborhood sets, and M sample extraction sub-schemes are determined. Step S300 of the embodiment of the present application also includes: Dividing the output of the M batches by the sum of the output of the M batches respectively to obtain M sampling coefficients, and multiplying the M sampling coefficients by the number of sample tests to obtain the number of sample tests of M batches; Extracting a first fusion quality detection information neighborhood set from the M fusion quality detection information neighborhood sets; Traversing and counting a plurality of information quantities of a plurality of first fusion quality detection information neighborhoods of the first fusion quality detection information neighborhood set; Determine a first sample sampling sub-scheme based on the plurality of information quantities, the first fusion quality detection information neighborhood set and the number of M batches of sample detections; According to the number of sample detections in the M batches, a sample extraction scheme is identified for the M fusion quality detection information neighborhood sets to determine the M sample extraction sub-schemes.

[0029] Furthermore, step S300 of the embodiment of the present application further includes: Match the M batches of sample detection quantities according to the first fusion quality detection information neighborhood set to obtain the first batch of sample detection quantities; Based on the multiple information quantities, multiple sampling probabilities are calculated, and the first batch of sample detection quantity is constrained, and the first fusion quality detection information neighborhood set is randomly sampled according to the multiple sampling probabilities as a first sample extraction sub-scheme.

[0030] In a possible embodiment, the preset preservative detection scheme is a scheme for preservative detection of the cosmetics to be tested set by a person skilled in the art, including information such as the number of sample detections, detection methods, and detection devices. The preset preservative detection scheme is extracted with the number of sample detections as an index, and the number of batch sample detections that can be allocated to the M production batches is analyzed in combination with the M batch outputs. Furthermore, the number of detections that should be extracted in each fused quality detection information neighborhood is determined in combination with the M fused quality detection information neighborhood sets corresponding to the M production batches, thereby determining the M sample extraction sub-schemes.

[0031] Preferably, the proportion of the output of each M batch in the total output is calculated respectively. The output of each batch is divided by the sum of the output of the M batches to obtain the sampling coefficient of each batch. The sampling coefficient is used to represent the proportion of each batch in the whole, so as to guide the subsequent allocation of the number of sample tests. According to the number of sample tests previously extracted (this is the preset value in the preset preservative test plan), it is multiplied by the sampling coefficient of each batch to obtain the number of sample tests of M batches. The number of sample tests of each batch reflects the number of samples that need to be extracted from the batch for preservative testing.

[0032] Select a fusion quality detection information neighborhood set from the M fusion quality detection information neighborhood sets for analysis. Traverse and count multiple quality detection information neighborhoods in the first fusion quality detection information neighborhood set, calculate the number of detection information contained in each neighborhood, and obtain multiple information quantities. The purpose of this step is to analyze the data volume of each neighborhood to ensure the rationality and accuracy of subsequent sampling.

[0033] Based on the statistical number of multiple information, the coverage of the first fusion quality detection information neighborhood set, and the number of sample tests in M ​​batches, the first sample extraction sub-scheme of the batch is determined. Through this scheme, it is ensured that the number of samples extracted from each batch meets the preservative detection requirements. For example, for neighborhoods with a large number of information, more sample extraction weights can be assigned to ensure that the covered samples are representative.

[0034] Based on the same principle as the first sample sampling sub-scheme, the M fusion quality inspection information neighborhood sets of all M production batches are analyzed, and combined with the number of sample inspections of each batch, M sample extraction sub-schemes are finally identified. Each sub-scheme should be targeted at the quality characteristics and output of each batch to ensure the reasonable allocation of inspection samples.

[0035] Preferably, according to the first fusion quality detection information neighborhood set determined previously, the number of sample tests in M ​​batches is matched to obtain the number of sample tests in the first batch corresponding to the batch. For example, if the number of tests in a batch is 100, and the batch has multiple fusion quality detection information neighborhoods, the probability calculation is performed according to the number of information in the neighborhood.

[0036] Based on the amount of information, the sampling probability of multiple neighborhoods is calculated. Optionally, the multiple information quantities are divided by the sum of the multiple information quantities to obtain multiple sampling probabilities. Taking the number of sample detections in the first batch as a constraint, random sampling is performed in the first fusion quality detection information neighborhood set according to multiple sampling probabilities to ensure that the samples are random and representative. For example, if a neighborhood accounts for 30% of the total amount of information, there is a 30% probability that the samples in the neighborhood will be extracted for detection. Therefore, the number of sample detections in the first batch is constrained, and the first fusion quality detection information neighborhood set is randomly sampled according to the multiple sampling probabilities as the first sample extraction sub-scheme. Through the above steps, the technical effect of improving the rationality and accuracy of sample extraction is achieved.

[0037] Since the number of samples tested in each batch is determined based on batch output and neighborhood quality information, the sample extraction sub-scheme can effectively optimize resource allocation and avoid the problem of over-testing or under-testing.

[0038] S400: Analyzing the M fused centralized quality detection information sets and the preset solid phase extraction parameters using an extraction parameter identifier to obtain M adjusted solid phase extraction parameters; Further, the M fusion concentrated quality detection information sets and the preset solid phase extraction parameters are analyzed by using an extraction parameter identifier to obtain M adjusted solid phase extraction parameters. Step S400 of the embodiment of the present application further includes: Acquire a set of quality detection information of multiple sample fusions, preset solid phase extraction parameters of multiple samples, and adjusted solid phase extraction parameters of multiple samples as training data, perform supervised training on a framework based on a feedforward neural network, and update the network parameters according to the output during training until the training converges, thereby obtaining the extraction parameter identifier that has been trained; The trained extraction parameter identifier is used to adjust the M fused centralized quality detection information sets and preset solid phase extraction parameters to obtain the M adjusted solid phase extraction parameters.

[0039] In one embodiment of the present application, the extraction parameter identifier is used to intelligently adjust the preset solid phase extraction parameters according to the actual product quality of M production batches. The M adjusted solid phase extraction parameters are obtained by inputting M fusion centralized quality detection information sets and preset solid phase extraction parameters into the extraction parameter identifier. Among them, the M adjusted solid phase extraction parameters are solid phase extraction parameters that meet the actual conditions of M production batches. The preset solid phase extraction parameters are solid phase extraction parameters pre-set by a person skilled in the art in the preset preservative detection scheme.

[0040] Preferably, a feedforward neural network framework is constructed based on the principle of supervised learning. The network includes an input layer, a hidden layer and an output layer. The input layer receives the fused centralized mass detection information of the sample and the preset solid phase extraction parameters; the hidden layer processes the input data through a multi-layer nonlinear function to capture the complex relationship therein; the output layer is used to predict the optimized solid phase extraction parameters (adjusted parameters). The framework constructed based on the feedforward neural network is supervised and trained using training data. During the training, the weights and bias parameters in the network are gradually adjusted so that the neural network can accurately predict the optimized solid phase extraction parameters according to the input sample data until the training converges to obtain the trained extraction parameter identifier. At this time, the trained extraction parameter identifier is used to perform parameter adjustment analysis on the M fused centralized mass detection information sets and the preset solid phase extraction parameters to obtain the M adjusted solid phase extraction parameters. The technical effect of greatly improving the adjustment efficiency of the solid phase extraction parameters during the solid phase extraction process is achieved.

[0041] S500: extracting test samples from the M product sequences based on the M sample extraction sub-schemes to obtain M sets of samples to be tested, configuring M solid phase extraction devices according to the M adjusted solid phase extraction parameters, and sequentially and slowly loading the M sets of samples to be tested onto the M activated solid phase extraction columns of the M solid phase extraction devices, and after the extraction time requirement is met, washing and eluting the M solid phase extraction columns to obtain M sets of sample extraction solutions; In one embodiment, based on the M sample extraction sub-schemes determined previously, samples are extracted from the M product neighborhood sets corresponding to the M fusion quality detection information neighborhood sets of the M product batches according to multiple sampling probabilities to obtain M sets of samples to be tested. Each sample extraction sub-scheme is designed based on factors such as batch output and quality detection information neighborhood to ensure that the extracted samples are representative and reasonable. For example, if the output of a batch is large, the sample extraction sub-scheme may require more samples to be extracted from the batch to ensure the accuracy of the test results. These sample sets will represent the main quality characteristics of each batch and will be used for subsequent solid phase extraction tests.

[0042] Preferably, the M adjusted solid phase extraction parameters obtained above are used to perform personalized configuration on each batch of solid phase extraction devices to obtain M solid phase extraction devices. Optionally, each adjusted solid phase extraction parameter includes adjusting the type and size of the solid phase extraction column, adjusting the selection of the pretreatment solvent, adjusting the sample loading speed (flow rate), adjusting the extraction time and the washing solvent, adjusting the elution conditions, etc.

[0043] In each set of samples to be tested, the samples to be tested are slowly loaded one by one into the activated solid phase extraction column. Preferably, the solid phase extraction column needs to be activated in advance to ensure that the adsorption material is in the best condition. The sample should be loaded at a uniform and stable speed to ensure that the target substance (preservative, etc.) can fully contact the packing of the extraction column to achieve efficient separation and adsorption.

[0044] After each sample to be tested is loaded, an appropriate extraction time is maintained to ensure that target substances such as preservatives are effectively adsorbed onto the packing of the solid phase extraction column. Different samples and preservatives may require different times, so the operation should be carried out according to the time set in the adjusted solid phase extraction parameters. During this period, it may be necessary to maintain the equilibrium time of the sample in the column to ensure the best adsorption effect. After the extraction is completed, the solid phase extraction column is washed with an appropriate washing solvent to remove impurities or interfering substances in the sample. This step ensures that target substances such as preservatives remain in the column, while other non-target components are effectively removed. The choice of washing solvent should be based on the parameters adjusted in the early stage, such as the use of solvents such as methanol, water or acetonitrile.

[0045] After washing, a specific elution solvent is used to elute the target substance from the solid phase extraction column to form the final sample extraction solution. The elution solution contains the concentrated preservative substance to be detected for subsequent preservative content analysis. The flow rate and volume of the elution solvent should be controlled in this process to ensure that the preservative is completely eluted.

[0046] The sample extraction solutions of each batch are collected into corresponding containers to form M sets of sample extraction solutions. These solutions will be used for subsequent preservative detection and analysis, such as quantitative analysis by means of liquid chromatography or gas chromatography. This achieves the technical effect of providing high-quality sample extraction solutions for subsequent preservative detection, thereby improving detection accuracy and reliability.

[0047] S600: sending the M sample extraction solution sets to a preservative detection instrument in sequence to detect the preservative content, thereby obtaining M sample preservative type sets and M sample preservative concentration sets; In a possible embodiment, before starting the test, ensure that the preservative detection instrument (such as high performance liquid chromatograph HPLC, gas chromatograph GC or mass spectrometer MS, etc.) is in good working condition. According to the operating manual of the instrument, perform baseline calibration, mobile phase adjustment and detector sensitivity calibration to ensure the accuracy and repeatability of the test.

[0048] Optionally, the filtered M sample extraction solution sets are respectively divided into vials of an automatic sampler. Each sample should be clearly marked to ensure that each batch of samples can be accurately identified in subsequent steps. Using the instrument's automatic injection system, the M sample extraction solution sets are sequentially sent into the preservative detection instrument in accordance with a preset detection sequence. This step should ensure that each sample can correctly enter the chromatography system or mass spectrometry system in the set order, and after analysis, the M sample preservative type sets and M sample preservative concentration sets are obtained. The technical effect of detecting the content and type of preservatives in the sample extraction solution is achieved.

[0049] S700: Using the preservative type as an index, perform the same type of aggregation mean processing on the M sample preservative type sets and the M sample preservative concentration sets to obtain the preservative detection result of the cosmetic to be tested, wherein the preservative detection result includes multiple preservative concentration means of multiple preservative types.

[0050] Further, taking the preservative type as an index, the M sample preservative type sets and the M sample preservative concentration sets are subjected to the same type of aggregate mean processing to obtain the preservative detection result of the cosmetic to be tested. Step S700 of the embodiment of the present application further includes: Taking the preservative type as an index, performing cross-batch aggregation of the M sample preservative type sets and the M sample preservative concentration sets to obtain a plurality of aggregated sample preservative concentration sets, wherein each aggregated sample preservative concentration set corresponds to one preservative type; Performing mean processing on the plurality of aggregate sample preservative concentration sets to obtain the plurality of preservative concentration means; The multiple preservative types and the average values ​​of the multiple preservative concentrations are used as the preservative detection result.

[0051] In a possible embodiment, after obtaining the M sample preservative type sets and the M sample preservative concentration sets, it is necessary to perform similar aggregation using the preservative type as an index, and perform mean processing on the aggregation results, so as to integrate the sample preservative test results of M production batches to determine the preservative test results of the cosmetics to be tested. The preservative test results include multiple preservative concentration means of multiple preservative types.

[0052] Using each preservative type as an index, data aggregation is performed on all M sample preservative type sets and M sample preservative concentration sets. The aggregation process aggregates the preservative data of the same type in each batch to form a new aggregated sample preservative concentration set. For example, if benzoic acid is detected in all M samples, the preservative concentration data related to benzoic acid in these samples will be merged into the same aggregated sample preservative concentration set of benzoic acid.

[0053] Then, the plurality of aggregate sample preservative concentration sets are respectively processed by averaging to obtain the plurality of preservative concentration means. The goal of determining the average preservative concentration of each preservative type is achieved, and then the plurality of preservative types and the plurality of preservative concentration means are used as the preservative detection result. The technical effect of improving the accuracy and representativeness of the detection result is achieved.

[0054] In summary, the embodiments of the present application have at least the following technical effects: This application achieves the technical effect of optimizing the allocation of testing resources and avoiding over- or under-testing of samples from different production batches by improving the rationality and accuracy of sample extraction, thereby achieving the goal of ensuring the coverage and randomness of samples, making the final test results more representative, and reducing detection errors caused by unreasonable sample extraction.

[0055] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0056] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0057] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A method for detecting cosmetic preservatives based on solid phase extraction, characterized in that: The method comprises: Obtaining M batch information of M production batches of the cosmetics to be tested, wherein the M batch information includes M product sequences, M quality inspection information sequences and M batch outputs; Performing neighbor quality division according to the degree of difference of adjacent quality detection information in the M quality detection information sequences, and performing neighborhood fusion on the division results to obtain M fused quality detection information neighborhood sets, wherein the M fused quality detection information neighborhood sets include M fused concentrated quality detection information sets; Interactively preset the preservative detection scheme to extract the sample detection quantity, combine the M batch outputs and the M fusion quality detection information neighborhood sets to identify the sample extraction scheme, and determine M sample extraction sub-schemes; Using an extraction parameter identifier to analyze the M fused centralized quality detection information sets and preset solid phase extraction parameters to obtain M adjusted solid phase extraction parameters; Based on the M sample extraction sub-schemes, the M product sequences are respectively subjected to sample extraction to obtain M sets of samples to be tested, and M solid phase extraction devices are configured according to the M adjusted solid phase extraction parameters, and the M sets of samples to be tested are respectively and slowly loaded onto the M activated solid phase extraction columns of the M solid phase extraction devices in sequence, and after the extraction time requirement is met, the M solid phase extraction columns are washed and eluted to obtain M sets of sample extraction solutions; The M sample extraction solution sets are sequentially sent to a preservative detection instrument to detect the preservative content, thereby obtaining M sample preservative type sets and M sample preservative concentration sets; Taking the preservative type as an index, the M sample preservative type sets and the M sample preservative concentration sets are subjected to similar aggregation mean processing to obtain the preservative detection result of the cosmetic to be tested, wherein the preservative detection result includes multiple preservative concentration means of multiple preservative types.

2. A method for detecting cosmetic preservatives based on solid phase extraction as claimed in claim 1, characterized in that: According to the difference degree of adjacent quality detection information in the M quality detection information sequences, neighbor quality division is performed, and neighborhood fusion is performed on the division results to obtain M fused quality detection information neighborhood sets, including: Taking the M median quality detection information of the M quality detection information sequences as M division starting points respectively, and constructing initial division neighborhoods to the left and right sides respectively according to a preset step length, to obtain M left initial division neighborhoods and M right initial division neighborhoods; Traversing the M left initial division neighborhoods and the M right initial division neighborhoods to perform neighbor quality difference authentication, and determining M left differences and M right differences; Determine whether the difference between the M left difference degrees and the M right difference degrees is less than or equal to a preset difference threshold value, if not, generate M division ridge lines at the M division starting points, and use the quality detection information located at the left end of the M left initial division neighborhoods as the M left division starting points, and use the quality detection information located at the right end of the M right initial division neighborhoods as the M right division starting points; Based on the M left division starting points and the M right division starting points, respectively, neighbor quality division is performed toward the left and right sides according to the preset step length until the first and last positions of the M quality detection information sequences are reached, and the division is stopped to obtain M division ridge line sets, and the M quality detection information sequences are divided based on the M division ridge line sets to obtain M initial division neighborhood sets; The number of information in the initial partitioned neighborhoods of the M initial partitioned neighborhood sets is traversed and counted, and the M initial partitioned neighborhood sets are subjected to neighborhood fusion according to the statistical results to obtain the M fused quality detection information neighborhood sets, and the mean of the M fused quality detection information neighborhood sets is calculated to determine the M fused concentrated quality detection information sets.

3. A method for detecting cosmetic preservatives based on solid phase extraction as claimed in claim 2, characterized in that: include: If so, the M left initial division neighborhoods and the M right initial division neighborhoods are merged into one neighborhood to obtain M initial division neighborhoods, and the quality detection information located at both ends of the M initial division neighborhoods is updated to M left division starting points and M right division starting points respectively.

4. A method for detecting cosmetic preservatives based on solid phase extraction as claimed in claim 3, characterized in that: include: Extracting M sets of initial partitioned neighborhoods to be merged whose number is less than a preset number threshold from the statistical results; The M initial partition neighborhood sets to be fused are integrated into the initial partition areas corresponding to the adjacent minimum difference values, and the M initial partition neighborhood sets are updated according to the fusion results to obtain the M fusion quality detection information neighborhood sets.

5. A method for detecting cosmetic preservatives based on solid phase extraction as claimed in claim 1, characterized in that: The preservative detection scheme is interactively preset to extract the sample detection quantity, and the sample extraction scheme is identified by combining the M batch outputs and the M fusion quality detection information neighborhood sets, and M sample extraction sub-schemes are determined, including: Dividing the output of the M batches by the sum of the output of the M batches respectively to obtain M sampling coefficients, and multiplying the M sampling coefficients by the number of sample tests to obtain the number of sample tests of M batches; Extracting a first fusion quality detection information neighborhood set from the M fusion quality detection information neighborhood sets; Traversing and counting a plurality of information quantities of a plurality of first fusion quality detection information neighborhoods of the first fusion quality detection information neighborhood set; Determine a first sample sampling sub-scheme based on the plurality of information quantities, the first fusion quality detection information neighborhood set and the number of M batches of sample detections; According to the number of sample detections in the M batches, a sample extraction scheme is identified for the M fusion quality detection information neighborhood sets to determine the M sample extraction sub-schemes.

6. A method for detecting cosmetic preservatives based on solid phase extraction as claimed in claim 5, characterized in that: include: Match the M batches of sample detection quantities according to the first fusion quality detection information neighborhood set to obtain the first batch of sample detection quantities; Based on the multiple information quantities, multiple sampling probabilities are calculated, and the first batch of sample detection quantity is constrained, and the first fusion quality detection information neighborhood set is randomly sampled according to the multiple sampling probabilities as a first sample extraction sub-scheme.

7. A method for detecting cosmetic preservatives based on solid phase extraction as claimed in claim 1, characterized in that: The M fusion centralized quality detection information sets and the preset solid phase extraction parameters are analyzed by using an extraction parameter identifier to obtain M adjusted solid phase extraction parameters, including: Acquire a set of quality detection information of multiple sample fusions, preset solid phase extraction parameters of multiple samples, and adjusted solid phase extraction parameters of multiple samples as training data, perform supervised training on a framework based on a feedforward neural network, and update the network parameters according to the output during training until the training converges, thereby obtaining the extraction parameter identifier that has been trained; The trained extraction parameter identifier is used to adjust the M fused centralized quality detection information sets and preset solid phase extraction parameters to obtain the M adjusted solid phase extraction parameters.

8. A method for detecting cosmetic preservatives based on solid phase extraction as claimed in claim 1, characterized in that: Taking the preservative type as an index, the M sample preservative type sets and the M sample preservative concentration sets are subjected to the same type of aggregate mean processing to obtain the preservative test results of the cosmetics to be tested, including: Taking the preservative type as an index, performing cross-batch aggregation of the M sample preservative type sets and the M sample preservative concentration sets to obtain a plurality of aggregated sample preservative concentration sets, wherein each aggregated sample preservative concentration set corresponds to one preservative type; Performing mean processing on the plurality of aggregate sample preservative concentration sets to obtain the plurality of preservative concentration means; The multiple preservative types and the average values ​​of the multiple preservative concentrations are used as the preservative detection result.

Images