Cosmetic preservative component monitoring system

Through sliding window technology, local weighted scattered smoothing model, isolated forest algorithm and near-infrared spectroscopy technology, the cosmetic anti-corrosion ingredient monitoring system is optimized, and the missed and false detection problems in traditional systems when monitoring preservative changes are solved, improving real-time performance and accuracy to ensure the safe use of preservatives in cosmetics.

CN118918970AActive Publication Date: 2024-11-08GUANGZHOU FUAN TESTING TECHNOLOGY CO LTD +1

Patent Information

Application Number
CN202410902872.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-08
Publication Date
2024-11-08
Estimated Expiration
2044-07-08

AI Technical Summary

Technical Problem

When monitoring preservatives with rapid changes and sensitive to slight concentration changes, traditional cosmetic anticorrosion monitoring systems are difficult to capture key data in a timely manner, resulting in missed or mis-checked, and are cumbersome to operate and long analysis cycles, which affects production efficiency and market response capabilities.

Method used

Sliding window technology and local weighted scattered smoothing model, combined with isolated forest algorithm and near-infrared spectroscopy technology, Kalman filtering optimizes the concentration estimate of anticorrosion components, and builds a dynamic update mechanism to adjust the monitoring threshold to realize real-time abnormality detection and early warning.

Benefits of technology

It improves the real-time tracking ability of preservative concentration changes, enhances data accuracy and processing speed, ensures accurate control of the amount of preservatives in cosmetics, complies with safety standards, and reduces the risk of excessive use.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118918970A_ABST
    Figure CN118918970A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of component monitoring, in particular to a cosmetic preservative component monitoring system which comprises a component data acquisition module, a component anomaly detection module, a component pedigree analysis module and a monitoring and feedback adjustment module. According to the invention, by introducing the sliding window technology and the local weighted scatter smooth model, the real-time performance and the data accuracy of anticorrosion component monitoring are enhanced, the real-time tracking capability on the concentration change of the preservative is improved, the system is allowed to quickly respond to the tiny change, the abrupt change point in the time sequence data is analyzed by adopting the isolation forest algorithm, and the accuracy of the monitoring result is improved. An efficient anomaly detection method is provided for cosmetics, the algorithm is suitable for a big data environment, the processing speed and accuracy are remarkably improved, the accuracy of predicting the concentration of the preservative is improved by combining the near infrared spectrum technology and the Kalman filtering technology, a manufacturer can accurately control the dosage of the preservative in the cosmetics, the safety standard is met, and the method is suitable for popularization and application. And the risk of excessive use is reduced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of ingredient monitoring, and in particular to a cosmetic preservative ingredient monitoring system. Background Art

[0002] Ingredient monitoring technology involves the use of various sensors and analytical equipment to detect and determine the composition of substances. This technology is widely used in many industries such as food safety, environmental monitoring, medical diagnosis, and cosmetics manufacturing. In the cosmetics industry, ingredient monitoring technology ensures the quality and safety of products by accurately measuring the content of active ingredients in the products, including preservatives, antioxidants and other additives. This technology usually includes the use of spectroscopy, chromatography, mass spectrometry and electrochemical analysis to perform quantitative and qualitative analysis to ensure that all ingredients meet strict safety standards.

[0003] Among them, the cosmetic preservative ingredient monitoring system is a system specially designed to track and analyze the preservative content in cosmetics. Preservatives are an important component in cosmetics and are used to prevent the growth of microorganisms and extend the shelf life of products. Its main purpose is to ensure that the amount of preservatives used in cosmetics is within safety standards while avoiding exceeding the maximum limit allowed by regulations. By real-time monitoring of the chemical composition of products, it can help manufacturers maintain product quality, protect consumer safety, and comply with industry regulations.

[0004] When monitoring ingredients such as preservatives that change rapidly and are sensitive to small changes in concentration, traditional systems are unable to capture key data in a timely manner, which can easily lead to missed detections or false detections. In addition, traditional technologies are cumbersome to operate and have long analysis cycles, making them unsuitable for fast-paced production environments, affecting production efficiency and the ability to respond to market changes in a timely manner. For example, for products with new formulas or rapidly changing ingredients, traditional systems find it difficult to provide efficient quality assurance, which can easily affect the product's market response speed and consumer trust. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a cosmetic preservative ingredient monitoring system.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: a cosmetic preservative ingredient monitoring system, the system comprising:

[0007] The ingredient data collection module uses sliding window technology to calculate the statistical characteristics of the data in each window based on the continuously collected preservative ingredient concentration data, and establishes a local weighted scatter point smoothing model based on the reaction characteristics of the preservative types, including phenoxyethanol, ethylhexylglycerin, phenylpropanol and caprylyl glycol, to generate trend model data;

[0008] The component anomaly detection module uses the isolation forest algorithm based on the trend model data to analyze the mutation points in the time series, identify the abnormal changes in the concentration of the preservative components in the data, and compare the identified abnormal data points with the historical data pattern, refine the abnormal type, and generate a component abnormality feature set;

[0009] The ingredient spectrum analysis module collects spectral data of cosmetic samples based on near-infrared spectroscopy technology, performs multivariate analysis, compares spectral analysis results with machine vision data, and optimizes the estimated concentration of preservative ingredients through Kalman filtering to generate ingredient concentration optimization indicators;

[0010] The monitoring and feedback adjustment module constructs a dynamic update mechanism to adjust the monitoring threshold based on the abnormal component feature set and the component concentration optimization index, updates the monitoring parameter settings in real time, optimizes the monitoring and early warning process of the preservative type and component stability, and generates component monitoring feedback results.

[0011] The present invention is improved in that the calculation steps of the statistical characteristics of the data in the window are specifically as follows:

[0012] Based on the continuously collected data on the concentration of anti-corrosion ingredients, the sliding window size is set, the anti-corrosion ingredient concentration data sequence is collected, and the data sequence is sliced ​​according to the window size, using the formula:

[0013]

[0014] Calculate the required number of windows n and generate a data subset, where S is the collected data sequence of the preservative component concentration and w is the sliding window size;

[0015] For each of the data subsets, the statistic is calculated using the formula:

[0016]

[0017] and

[0018]

[0019] Get the average value μ of the i-th window i and standard deviation σ i , where x is D i A data point in i is a data subset, p i is the weight of data point x, which is determined according to the time sequence of the data points, λ is the smoothing coefficient, which is used to adjust the sensitivity of the standard deviation, and w is the sliding window size;

[0020] The mean and standard deviation are combined into a statistical characteristic vector of the data subset, and the formula is:

[0021] Vi =(μ i ,σ i )

[0022] Get the data statistical characteristics V in the sliding window i , where μ i is the mean value, σ i is the standard deviation, and according to V i Perform iterative analysis including outlier detection or trend prediction.

[0023] The present invention is improved in that the step of establishing the local weighted scattered point smoothing model is specifically as follows:

[0024] Based on the statistical characteristics of the data, the smoothing parameter is selected and the bandwidth of the local weight is determined, and the weight function is adjusted. The formula is:

[0025] w(x,h,β)=exp(-β·(x 2 / h 2 )

[0026] Generate an adjusted weight function w(x,h,β), where h represents the bandwidth, which determines the local range of weight distribution, β represents the attenuation coefficient, which adjusts the sensitivity of the weight function, and x is the independent variable distance of the data point;

[0027] Based on the adjusted weight function, the LOWESS formula is applied:

[0028]

[0029] in, Calculate for each data point x i The local weighted regression value of is used to obtain the local weighted scattered point smoothing model. represents the predicted value calculated by LOWESS, w ij Represents point x j Point x i The local weight of n represents the total number of data points, x i and x j Represent the independent variable values ​​of the i-th and j-th data points, y j represents the dependent variable value of the jth data point, |x i -x j | is the distance between point i and point j on the independent variable axis, and w is the weight function.

[0030] The present invention is improved in that the step of identifying the change in the concentration of the antiseptic component is specifically:

[0031] Based on the trend model data, the isolation forest algorithm is used to analyze the data points in the time series, using the formula:

[0032]

[0033] Calculate the anomaly score of each data point and obtain a list of preliminarily identified anomalies, where S(x) is the anomaly score of data point x, d(x) is the distance from the data point to the adjacent decision boundary, m is the average distance from the data point to the decision boundary, κ is the density adjustment coefficient, p(x) is the prior probability of data point x, and σ is the standard deviation of the data point;

[0034] Based on the anomaly list, set a threshold to filter data points with high anomaly scores, using the formula:

[0035] E(x)={x|S(x)>θ×α}

[0036] Identify the abnormal data points that exceed the threshold and generate a list of data points with high anomaly scores, where θ is the threshold for identifying high anomaly scores, α is the threshold adjustment factor, E(x) is the set of abnormal data points, S(x) is the anomaly score of data point x, and x is the data point;

[0037] The list of high anomaly scoring data points is validated by checking the context and neighboring values ​​of the data points using the formula:

[0038]

[0039] Generate abnormal preservative ingredient concentration change characteristics, where V(x) is a verified set of abnormal data points, E is a set of abnormal data points, μ is the average value of the data points, which is used to determine the degree of abnormality, λ is an adjustment parameter used to affect the sensitivity of the abnormality, and x is a data point.

[0040] The present invention is improved in that the comparison step of the historical data mode is specifically as follows:

[0041] Collect the historical data of the concentration change of the preservative component, and for each currently identified abnormal data point, calculate the difference from the historical data using the formula:

[0042]

[0043] Generate a list D(x) of deviations between the current data point x and the historical pattern, where h i is the concentration value of the antiseptic component at the historical data point, N represents the total number of historical data points, and w i is the weight of the ith data point, γ represents the power exponent used to adjust the sensitivity of the deviation calculation, δ is the parameter used for smoothing, and x is the concentration value of the preservative ingredient at the data point;

[0044] According to the deviation list, the types of deviations are analyzed, including periodic changes, sudden increases or sudden decreases, and abnormality types are determined to obtain a component abnormality feature set.

[0045] The present invention is improved in that the steps of collecting the spectral data are specifically as follows:

[0046] Based on the near-infrared spectroscopy technology, the near-infrared spectrometer parameters are set, the light source intensity and the detector sensitivity are adjusted, and the characteristics of the cosmetic sample are matched, using the formula:

[0047]

[0048] Generate the spectral device configuration parameters P, where L represents the light source intensity, η represents the device efficiency, S represents the sensitivity of the detector, A represents the absorbance of the sample, and c represents the optical path length;

[0049] According to the spectral equipment configuration parameters, the spectral data of the cosmetic samples are collected, and the reflectance of each sample is recorded, using the formula:

[0050]

[0051] Calculate the reflectance R of the sample to obtain a spectral data set of the cosmetic sample, where I represents the reflected light intensity, I0 represents the incident light intensity, and β represents a correction factor used to adjust the error in the reflectance calculation;

[0052] The spectral data set is formatted into a data file using the data organization formula:

[0053]

[0054] Generate spectral data file D, where R n represents the reflectivity of the data point, γ is the normalization factor, and N represents the total number of data points.

[0055] The present invention is improved in that the step of obtaining the component concentration optimization index is specifically as follows:

[0056] The spectral data is analyzed to identify the spectral characteristics of multiple components in the sample using the formula:

[0057]

[0058] Generate multivariate analysis results M, where x i represents a variable observation in the sample, w i represents the weight assigned to the i-th variable, α represents the power exponent used to amplify the impact of the observed value, λ and δ i represents the attenuation coefficient, and N represents the total number of variables;

[0059] The multivariate analysis results were compared with the data acquired by machine vision using the formula:

[0060]

[0061] Generate a differential analysis log C, where M j represents the results from multivariate analysis, V j represents the data from machine vision, K represents the total number of data points compared, and β and γ represent the exponents used to adjust the sensitivity of the difference calculation;

[0062] The Kalman filter was applied to optimize the concentration of the preservative component predicted based on the spectral and visual data, using the formula:

[0063] F k =F k-1 +K k ·(Z k -F k-1 ) g

[0064] Generate component concentration optimization index, where F k Represents the estimated concentration value, K k represents the Kalman gain, Z k represents the current observation value, F k-1 Represents the previous estimate, and g represents the adjustment exponent of the update step size.

[0065] The present invention is improved in that the steps of obtaining the component monitoring feedback result are specifically as follows:

[0066] Based on the component anomaly feature set and component concentration optimization index, real-time monitoring data is collected from the monitoring process adjusted by the dynamic update mechanism, using the formula:

[0067]

[0068] Generate real-time monitoring data, where D represents the total amount of real-time monitoring data collected, c i represents the concentration of the anticorrosive component at the i-th monitoring point, λ i represents the weight coefficient assigned according to the characteristics of the i-th monitoring point, and N represents the total number of monitoring points;

[0069] To evaluate the deviation of the real-time monitoring data from the threshold, the formula is used:

[0070]

[0071] Generate monitoring data deviation evaluation information, where E represents the weighted average deviation between the monitoring data and the threshold, c i represents the real-time monitoring of the concentration of the preservative component, T set represents the concentration threshold of the preservative ingredient, w i represents the weight coefficient of the deviation, and N represents the total number of monitoring points;

[0072] Based on the deviation assessment information, using the formula:

[0073] F = "adjust" if E>θ else "hold"

[0074] Generate component monitoring feedback result F, where E is the weighted average deviation between the monitoring data and the threshold, and θ is the decision threshold.

[0075] Compared with the prior art, the advantages and positive effects of the present invention are:

[0076] In the present invention, by introducing sliding window technology and local weighted scatter point smoothing model, the real-time and data accuracy of preservative ingredient monitoring are enhanced, the real-time tracking capability of preservative concentration changes is improved, the system is allowed to respond quickly to small changes, and the isolation forest algorithm is used to analyze mutation points in time series data, providing a highly efficient anomaly detection method for cosmetics. The algorithm is suitable for big data environments and significantly improves processing speed and accuracy. Combined with near-infrared spectroscopy technology and Kalman filtering technology, the accuracy of predicting preservative concentration is improved, so that manufacturers can accurately control the amount of preservatives in cosmetics, meet safety standards, and reduce the risk of excessive use. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 The module diagram of the cosmetic preservative ingredient monitoring system proposed by the present invention;

[0078] Figure 2 The present invention proposes a flow chart for calculating the statistical characteristics of data in a window in a cosmetic preservative ingredient monitoring system;

[0079] Figure 3 The present invention proposes a flow chart for establishing a local weighted scatter point smoothing model in a cosmetic preservative ingredient monitoring system;

[0080] Figure 4 The present invention proposes a flow chart for identifying changes in the concentration of preservative ingredients in a cosmetic preservative ingredient monitoring system;

[0081] Figure 5 A comparative flow chart of historical data modes in a cosmetic preservative ingredient monitoring system is proposed for the present invention;

[0082] Figure 6 The present invention proposes a flow chart for collecting spectral data in a cosmetic preservative ingredient monitoring system;

[0083] Figure 7 The present invention proposes a flow chart for obtaining the ingredient concentration optimization index in the cosmetic preservative ingredient monitoring system;

[0084] Figure 8The present invention proposes a flow chart for obtaining ingredient monitoring feedback results in a cosmetic preservative ingredient monitoring system. DETAILED DESCRIPTION

[0085] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0086] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0087] Example

[0088] See also Figure 1 The present invention provides a technical solution: a cosmetic preservative ingredient monitoring system comprises:

[0089] The ingredient data acquisition module is based on the continuously collected preservative ingredient concentration data and uses sliding window technology to calculate the statistical characteristics of the data within each window. It also establishes a local weighted scatter point smoothing model based on the reaction characteristics of the preservative type, which includes phenoxyethanol, ethylhexylglycerin, phenylpropanol and caprylyl glycol, to generate trend model data. The local weighted scatter point smoothing model is used to capture the nonlinear effect of preservative type on concentration changes, so as to provide more accurate trend analysis and abnormal warning basis.

[0090] The component anomaly detection module uses the isolation forest algorithm based on the trend model data to analyze the mutation points in the time series, identify the abnormal changes in the concentration of preservative components in the data, and compare the historical data patterns with the identified abnormal data points, refine the anomaly types, and generate a component anomaly feature set;

[0091] The ingredient spectrum analysis module is based on near-infrared spectroscopy technology. It collects spectral data of cosmetic samples, performs multivariate analysis, compares spectral analysis results with machine vision data, and optimizes the estimated concentration of preservative ingredients through Kalman filtering to generate ingredient concentration optimization indicators. Machine vision data is collected by cameras on the production line and is used to monitor the color and texture of cosmetics, etc.

[0092] The monitoring and feedback adjustment module builds a dynamic update mechanism to adjust the monitoring threshold based on the abnormal component feature set and component concentration optimization index, updates the monitoring parameter settings in real time, optimizes the monitoring and early warning process of preservative types and component stability, and generates component monitoring feedback results.

[0093] Trend model data include concentration change trends, seasonal adjustment results, and abnormal warning indicators. The component abnormality feature set includes identified abnormal events, abnormality degree assessment results, and influencing factor analysis results. The component concentration optimization indicators include concentration stability score information, prediction accuracy information, and adjustment reaction sensitivity.

[0094] See also Figure 2 , the calculation steps of the statistical characteristics of the data in the window are as follows:

[0095] Based on the continuously collected data on the concentration of anti-corrosion ingredients, the sliding window size is set, the anti-corrosion ingredient concentration data sequence is collected, and the data sequence is sliced ​​according to the window size, using the formula:

[0096]

[0097] Calculate the required number of windows n and generate a data subset, where S is the collected data sequence of the preservative component concentration and w is the sliding window size;

[0098] For each data subset, calculate the statistic using the formula:

[0099]

[0100] and

[0101]

[0102] Get the average value μ of the i-th window i and standard deviation σ i , where x is D i A data point in i is a data subset, p i is the weight of data point x, which is determined according to the time sequence of the data points, λ is the smoothing coefficient, which is used to adjust the sensitivity of the standard deviation, and w is the sliding window size;

[0103] The mean and standard deviation are combined into a statistical feature vector of a data subset, and the formula is:

[0104] V i =(μ i ,σ i )

[0105] Get the data statistical characteristics V in the sliding window i , where μ iis the mean value, σ i is the standard deviation, and according to V i Perform iterative analysis including outlier detection or trend prediction.

[0106] In calculating the number of windows, assume the following data:

[0107] Assume that the window size w=10 (a window contains 10 data points), and there is a data sequence S with length |S|=150.

[0108] Calculation process:

[0109] Determine the window size w=10;

[0110] Measurement data sequence length |S|=150;

[0111] Apply the formula to calculate

[0112] This means that the data sequence is divided into 15 windows.

[0113] In calculating the weighted mean and adjusted standard deviation, the following data are assumed:

[0114] Assume that the weight p i Data points are assigned based on their position in the window (e.g., earlier data points have lower weights), and λ (smoothing coefficient) is set to 2. Example values ​​of data points x within the window are [5, 10, 15, 20, 25, 30, 35, 40, 45, 50].

[0115] Calculation process:

[0116] For each data point x in the window, set p i The value of, for example, p i =0.1i (where i is the index position of the data point in the window);

[0117] Calculate the weighted average:

[0118]

[0119] Calculate the adjusted standard deviation:

[0120] First, we calculate the cube of the difference between each data point and the mean:

[0121] For example |5-33.5| 3 ,|10-33.5| 3 , and so on.

[0122] Apply the formula to calculate the standard deviation:

[0123]

[0124] Weighted mean μ i =33.5 represents the average value of the data points in the window after taking the weight into account;

[0125] Adjusted standard deviation σ i =29.8 indicates the degree of dispersion of the data point distribution, which reflects a greater sensitivity after adjustment;

[0126] Apply Formula V i =(μ i ,σ i ), combining the mean and standard deviation into a statistical characteristic vector, which represents the statistical characteristics of the data in the window.

[0127] See also Figure 3 , the specific steps of establishing the local weighted scattered point smoothing model are:

[0128] Based on the statistical characteristics of the data, the smoothing parameter is selected and the bandwidth of the local weight is determined, and the weight function is adjusted. The formula is:

[0129] w(x,h,β)=exp(-β·(x 2 / h 2 )

[0130] Generate an adjusted weight function w(x,h,β), where h represents the bandwidth, which determines the local range of weight distribution, β represents the attenuation coefficient, which adjusts the sensitivity of the weight function, and x is the independent variable distance of the data point;

[0131] Based on the adjusted weight function, the LOWESS formula is applied:

[0132]

[0133] in, Calculate for each data point x i The local weighted regression value of is used to obtain the local weighted scattered point smoothing model. represents the predicted value calculated by LOWESS, w ij Represents point x j Point x i The local weight of n represents the total number of data points, x i and x j Represent the independent variable values ​​of the i-th and j-th data points, y j represents the dependent variable value of the jth data point, |x i -x j | is the distance between point i and point j on the independent variable axis, and w is the weight function.

[0134] Assume the following data:

[0135] Assuming that h=5 (bandwidth, affecting the local smoothing range) and β=0.5 (attenuation coefficient, adjusting the sensitivity of the weight function) are selected, the weight at distance x=3 will be calculated.

[0136] Calculation process:

[0137] Determine the bandwidth h = 5 and the attenuation coefficient β = 0.5;

[0138] Substitute x=3 into the weight function to calculate the weight:

[0139] w(3,5,0.5)=exp(-0.5·(3 2 / 5 2 ))

[0140] =exp(-0.18)=0.835

[0141] means point x j Distance prediction point x i The weight for the 3-unit distance is 0.835, indicating that closer points have a greater impact on the predicted value.

[0142] Assume the following data:

[0143] Assume n=3, the data point location x and the value of the dependent variable y are as follows:

[0144] x1=0, y1=2

[0145] x2=3,y2=3

[0146] x3=5,y3=5

[0147] Predict x i = Value at 3

[0148] Calculation process:

[0149] Calculate the weight w for each data point ij , for example, for j=1:

[0150]

[0151] Calculate using weights and corresponding y values

[0152]

[0153] Finally, the predicted point x i = Value at 3 It is approximately 5.183, which represents the value of the dependent variable y predicted by the LOWESS model at x=3.

[0154] See also Figure 4 , the specific steps for identifying changes in the concentration of preservative ingredients are:

[0155] Based on the trend model data, the isolation forest algorithm is used to analyze the data points in the time series, using the formula:

[0156]

[0157] Calculate the anomaly score of each data point and obtain a list of preliminarily identified anomalies, where S(x) is the anomaly score of data point x, d(x) is the distance from the data point to the adjacent decision boundary, m is the average distance from all data points to the decision boundary, κ is the density adjustment coefficient, considering the crowding around the data point, p(x) is the prior probability of data point x, and σ is the standard deviation of the data point;

[0158] Based on the list of outliers, set a threshold to filter out data points with high anomaly scores, using the formula:

[0159] E(x)={x|S(x)>…×α}

[0160] Identify the abnormal data points that exceed the threshold and generate a list of data points with high anomaly scores, where θ is the threshold for identifying high anomaly scores, α is the threshold adjustment factor to adapt to different data characteristics, E(x) is the set of abnormal data points, S(x) is the anomaly score of data point x, and x is the data point;

[0161] Validate the list of high anomaly score data points, check the context and neighboring values ​​of the data points, and use the formula:

[0162]

[0163] Generate abnormal preservative ingredient concentration change characteristics, where V(x) is a verified set of abnormal data points, E is a set of abnormal data points, μ is the average value of the data points, which is used to determine the degree of abnormality, λ is an adjustment parameter used to affect the sensitivity of the abnormality, and x is a data point.

[0164] Assume the following data:

[0165] The position of data point x is x=10;

[0166] The distance to the neighboring decision boundary is d(x) = 2;

[0167] Average distance m = 5;

[0168] The density coefficient of data points is κ = 0.5;

[0169] The prior probability of the data point p(x) = 0.1;

[0170] Standard deviation σ=1.5.

[0171] Calculation process:

[0172] Calculate d(x)+κ·p(x)=2+0.5×0.1=2.05;

[0173] Calculate m+σ=5+1.5=6.5;

[0174] Substitute the values ​​into the anomaly score formula:

[0175]

[0176] This means that the anomaly score of the data point x=10 is approximately 0.729, which can be used in subsequent steps to compare with the anomaly threshold to determine whether it is an anomaly.

[0177] Assume the following data:

[0178] Anomaly score threshold θ = 0.6;

[0179] Adjustment factor α=1.2.

[0180] Calculation process:

[0181] Calculate the adjusted threshold θ×α=0.6×1.2=0.72;

[0182] Compare the above anomaly score S(x)=0.729 with the adjusted threshold value of 0.72;

[0183] Since 0.729>0.72, data point x=10 is confirmed as an anomaly;

[0184] It shows that after adjusting the threshold, data point x=10 is confirmed as abnormal.

[0185] Assume the following data:

[0186] The mean value of data point x is μ=8;

[0187] Adjust parameter λ=2.

[0188] Calculation process:

[0189] For the data point x=10, calculate |x-μ|=|10-8|=2;

[0190] Calculate λ·μ=2×8=16;

[0191] Substitute into the verification formula:

[0192]

[0193] The calculation shows that the verified anomaly score is 0.875, indicating that the data point x=10 is highly abnormal.

[0194] See also Figure 5 , the specific steps for comparing historical data patterns are:

[0195] Collect historical data on changes in the concentration of preservative ingredients, and for each currently identified abnormal data point, calculate the difference from the historical data using the formula:

[0196]

[0197] Generate a list D(x) of deviations between the current data point x and the historical pattern, where h i is the concentration value of the antiseptic component at the historical data point, N represents the total number of historical data points, and w i is the weight of the ith data point, based on time proximity, the closer the distance is to the present, the greater the weight, γ represents the power exponent used to adjust the sensitivity of the deviation calculation, δ is a parameter used for smoothing to ensure calculation stability, and x is the concentration value of the preservative ingredient at the data point;

[0198] According to the deviation list, the types of deviations are analyzed, including periodic changes, sudden increases or decreases, and the abnormality types are determined to obtain the component abnormality feature set.

[0199] Assume the following data:

[0200] Suppose there is a current data point x=50 and a small historical data set containing three points h=

[0201] 45, 55, 60];

[0202] The point weights are w = [0.5, 1, 0.5], γ = 2 and smoothing parameter δ = 1 are selected.

[0203] Calculation process:

[0204] Setting of weights and history points:

[0205] h1=45,w1=0.5

[0206] h2=55, w2=1

[0207] h3=60,w3=0.5

[0208] Calculate the weighted deviation for each historical point:

[0209] For h1:

[0210] w1·|x-h1| γ =0.5·|50-45| 2 =0.5·25=12.5

[0211] For h2:

[0212] w2·|x-h2| γ =1·|50-55| 2 =1·25=25

[0213] For h3:

[0214] w3·|x-h3| γ =0.5·|50-60| 2 =0.5·100=50

[0215] Calculation of the sum and mean deviation:

[0216] Total Deviation:

[0217] 12.5+25+50=87.5

[0218] Denominator:

[0219] N+δ=3+1=4

[0220] Calculate D(x):

[0221]

[0222] The calculated result D(x)=21.875 represents the average weighted deviation of the current data point x=50 relative to the given historical data pattern. A high value indicates that there is a significant difference between the current data point and the historical data, indicating abnormal activity or change.

[0223] See also Figure 6 , the specific steps for collecting spectral data are:

[0224] Based on near-infrared spectroscopy technology, set the near-infrared spectrometer parameters, adjust the light source intensity and detector sensitivity, match the characteristics of the cosmetic sample, and use the formula:

[0225]

[0226] Generate the spectroscopic device configuration parameters P, where L represents the intensity of the light source, η represents the device efficiency, which measures how the device performs in converting the light source energy, S represents the sensitivity of the detector, A represents the absorbance of the sample, and c represents the optical path length, which affects the distance that light travels in the device;

[0227] According to the spectral equipment configuration parameters, the spectral data of the cosmetic samples are collected, and the reflectance of each sample is recorded, using the formula:

[0228]

[0229] Calculate the reflectance R of the sample to obtain a spectral data set of the cosmetic sample, where I represents the reflected light intensity, I0 represents the incident light intensity, and β represents a correction factor used to adjust the error in the reflectance calculation;

[0230] Format the spectral dataset into a data file using the data wrangling formula:

[0231]

[0232] Generate spectral data file D, where R n represents the reflectance of the data point, γ is the normalization factor to ensure the consistency and comparability of the data, and N represents the total number of data points.

[0233] Assume the following data:

[0234] Assume that the light source intensity L = 100 units;

[0235] Equipment efficiency η = 0.8 (80% efficiency);

[0236] Detector sensitivity S = 0.9;

[0237] The absorbance of the sample is A = 1.2;

[0238] Optical path length c = 2 meters.

[0239] Calculation process

[0240] Equipment parameter calculation:

[0241] Calculate L·η·S=100×0.8×0.9=72;

[0242] Calculate A·c = 1.2 × 2 = 2.4;

[0243] Substitute the values ​​into the device configuration parameter formula:

[0244]

[0245] It indicates that the configuration parameter P of the spectral device is 30, which is a synthetic parameter reflecting the overall sensitivity and efficiency of the device.

[0246] Assume the following data:

[0247] Reflected light intensity I = 50 units;

[0248] Incident light intensity I0 = 100 units;

[0249] Correction factor β = 1.1.

[0250] Reflectivity calculation:

[0251] Calculate the corrected incident light intensity:

[0252] I0·β=100×1.1=110

[0253] Calculate reflectivity:

[0254]

[0255] The reflectivity R of the sample is 0.455, indicating that a certain proportion of light is reflected by the sample.

[0256] Assume the following data:

[0257] Reflectivity data point R = [0.45, 0.50, 0.48];

[0258] Normalization factor γ = 0.5;

[0259] The total number of data points N=3.

[0260] Data collation:

[0261] Normalize each reflectivity and sum it:

[0262]

[0263] This means that the sum of the sorted spectral data file D is 2.86, which can be used for further analysis or storage.

[0264] See also Figure 7 , the steps for obtaining the component concentration optimization index are as follows:

[0265] Analyze the spectral data and identify the spectral characteristics of multiple components in the sample using the formula:

[0266]

[0267] Generate multivariate analysis results M, where x i represents a variable observation in the sample, w i represents the weight assigned to the i-th variable, α represents the power exponent used to amplify the impact of the observed value, λ and δ i represents the attenuation coefficient, which is used to simulate the reduction of the impact of environmental factors or time on variables, and N represents the total number of variables;

[0268] Compare the multivariate analysis results with the data acquired by machine vision using the formula:

[0269]

[0270] Generate a differential analysis log C, where M j represents the results from multivariate analysis, V j represents the data from machine vision, K represents the total number of data points compared, and β and γ represent the exponents used to adjust the sensitivity of the difference calculation;

[0271] Apply Kalman filtering to optimize the concentration of preservative ingredients predicted based on spectral and visual data using the formula:

[0272] F k =F k-1 +K k ·(Z k -F k-1 ) g

[0273] Generate component concentration optimization index, where F k Represents the estimated concentration value, K k Represents the Kalman gain, which is used to adjust the degree to which the estimated value is close to the actual observed value, Z k represents the current observation value, F k-1 Represents the previous estimate, and g represents the adjustment exponent of the update step size.

[0274] Assume the following data:

[0275] x i Assuming it is spectral intensity, take sample values ​​[100, 200, 150];

[0276] Assume the weight w i is [0.5, 1.0, 0.75];

[0277] α is set to 2;

[0278] λ is set to 0.05;

[0279] δ i Set to [1, 2, 3];

[0280] N3.

[0281] Calculation process:

[0282] For each i:

[0283] calculate

[0284] [100 2 , 200 2 , 150 2 ] = [10000, 40000, 22500]

[0285] calculate

[0286] [e -0.05·1 , e -0.05·2 , e -0.05·3 ] = [0.951, 0.905, 0.860]

[0287] Calculate each

[0288] [0.5 10000 0.951, 1.0 40000 0.905, 0.75 22500 0.860]

[0289] =[4755,36200,14512.5]

[0290] Calculate the sum M:

[0291] 4755+36200+14512.5=55467.5

[0292] The multivariate analysis result M=55449.5 represents the comprehensive analysis value obtained according to the weighted and adjusted spectral intensity, and a higher value indicates a high concentration or a strong response of a specific component.

[0293] Assume the following data:

[0294] M j Assume it is [55450, 55460, 55455];

[0295] V j Assume it is [55430, 55470, 55440];

[0296] K is 3;

[0297] β is set to 1.5;

[0298] γ is set to 0.5.

[0299] Calculation process:

[0300] For each j:

[0301] Compute the absolute value of the difference and weight it:

[0302] |M j -V j | β =|20| 1.5 ,|10| 1.5 ,|15| 1.5 ≈[89, 32, 58]

[0303] Compute the sum and normalize:

[0304]

[0305] The difference analysis report C = 103.35 represents the average adjusted difference between the multivariate analysis results and the machine vision data, which is used to evaluate the consistency or difference between the two technologies.

[0306] See also Figure 8, the specific steps for obtaining the component monitoring feedback results are:

[0307] Based on the component anomaly feature set and component concentration optimization index, real-time monitoring data is collected from the monitoring process adjusted by the dynamic update mechanism, using the formula:

[0308]

[0309] Generate real-time monitoring data, where D represents the total amount of real-time monitoring data collected, c i represents the concentration of the anticorrosive component at the i-th monitoring point, λ i represents the weight coefficient assigned according to the characteristics of the i-th monitoring point, and N represents the total number of monitoring points;

[0310] To evaluate the deviation of real-time monitoring data from the threshold, the formula is used:

[0311]

[0312] Generate monitoring data deviation evaluation information, where E represents the weighted average deviation between the monitoring data and the threshold, c i represents the real-time monitoring of the concentration of the preservative component, T set represents the concentration threshold of the preservative ingredient, w i Represents the weight coefficient of the deviation, which is used to adjust the influence of data from different monitoring points. N represents the total number of monitoring points;

[0313] Based on the deviation assessment information, use the formula:

[0314] F = "adjust" if E>θ else "maintain"

[0315] Generate component monitoring feedback results F, where E is the weighted average deviation between the monitoring data and the threshold, and θ is the decision threshold used to determine whether the monitoring strategy needs to be adjusted.

[0316] Assume the following data:

[0317] c i is the concentration of the preservative component at the i-th monitoring point. Assume that the concentrations at the three monitoring points are 15ppm, 20ppm, and 25ppm respectively;

[0318] λ i The weights are set to 0.5, 1.0, and 1.5 to reflect the importance of different monitoring points;

[0319] N is set to 3.

[0320] For each monitoring point:

[0321] c1·λ1=15×0.5=7.5

[0322] c2·λ2=20×1.0=20

[0323] c3·λ3=25×1.5=37.5

[0324] Calculate the sum:

[0325] D = 7.5 + 20 + 37.5 = 65

[0326] The sum D=65 represents the sum of weighted real-time monitoring data, which is the total concentration value weighted based on the importance of the monitoring points.

[0327] Assume the following data:

[0328] c i 15ppm, 20ppm and 25ppm respectively;

[0329] T set Assume 18ppm;

[0330] w i 0.5, 1.0 and 1.5;

[0331] N is 3.

[0332] |c1-T set |·w1=|15-18|×0.5=1.5

[0333] |c2-T set |·w2=|20-18|×1.0=2.0

[0334] |c3-T set |·w3=|25-18|×1.5=10.5

[0335] Calculate the weighted mean deviation:

[0336]

[0337] The weighted mean deviation E = 4.67 represents the average deviation of the monitoring data relative to the preset threshold, taking into account the importance of different monitoring points.

[0338] The weighted mean deviation obtained for E is 4.67, and θ is assumed to be 3.

[0339] Check if E is greater than θ:

[0340] Compare E=4.67 with θ=3, because 4.67>3, F is adjusted.

[0341] Indicates that the monitoring strategy needs to be adjusted because the deviations shown by the real-time monitoring data exceed the set decision threshold.

[0342] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. The cosmetic preservative ingredient monitoring system is characterized by: The system comprises: The ingredient data collection module uses sliding window technology to calculate the statistical characteristics of the data in each window based on the continuously collected preservative ingredient concentration data, and establishes a local weighted scatter point smoothing model based on the reaction characteristics of the preservative types, including phenoxyethanol, ethylhexylglycerin, phenylpropanol and caprylyl glycol, to generate trend model data; The component anomaly detection module uses the isolation forest algorithm based on the trend model data to analyze the mutation points in the time series, identify the abnormal changes in the concentration of the preservative components in the data, and compare the identified abnormal data points with the historical data pattern, refine the abnormal type, and generate a component abnormality feature set; The ingredient spectrum analysis module collects spectral data of cosmetic samples based on near-infrared spectroscopy technology, performs multivariate analysis, compares spectral analysis results with machine vision data, and optimizes the estimated concentration of preservative ingredients through Kalman filtering to generate ingredient concentration optimization indicators; The monitoring and feedback adjustment module constructs a dynamic update mechanism to adjust the monitoring threshold based on the abnormal component feature set and the component concentration optimization index, updates the monitoring parameter settings in real time, optimizes the monitoring and early warning process of the preservative type and component stability, and generates component monitoring feedback results.

2. The cosmetic preservative ingredient monitoring system according to claim 1, characterized in that: The calculation steps of the data statistical characteristics in the window are specifically as follows: Based on the continuously collected data on the concentration of anti-corrosion ingredients, the sliding window size is set, the anti-corrosion ingredient concentration data sequence is collected, and the data sequence is sliced ​​according to the window size, using the formula: Calculate the required number of windows n and generate a data subset, where S is the collected data sequence of the preservative component concentration and w is the sliding window size; For each of the data subsets, the statistic is calculated using the formula: and Get the average value μ of the i-th window i and standard deviation σ i , where x is D i A data point in i is a data subset, p i is the weight of data point x, which is determined according to the time sequence of the data points, λ is the smoothing coefficient, which is used to adjust the sensitivity of the standard deviation, and w is the sliding window size; The mean and standard deviation are combined into a statistical characteristic vector of the data subset, and the formula is: V i =(μ i ,s i ) Get the data statistical characteristics V in the sliding window i , where μ i is the mean value, σ i is the standard deviation, and according to V i Perform iterative analysis including outlier detection or trend prediction.

3. The cosmetic preservative ingredient monitoring system according to claim 1, characterized in that: The steps of establishing the local weighted scattered point smoothing model are specifically as follows: Based on the statistical characteristics of the data, the smoothing parameter is selected and the bandwidth of the local weight is determined, and the weight function is adjusted. The formula is: w(x,h,β)=exp(-β·(x 2 / h 2 ) Generate an adjusted weight function w(x,h,β), where h represents the bandwidth, which determines the local range of weight distribution, β represents the attenuation coefficient, which adjusts the sensitivity of the weight function, and x is the independent variable distance of the data point; Based on the adjusted weight function, the LOWESS formula is applied: in, Calculate for each data point x i The local weighted regression value of is used to obtain the local weighted scattered point smoothing model. represents the predicted value calculated by LOWESS, w ij Represents point x j Point x i The local weight of n represents the total number of data points, x i and x j Represent the independent variable values ​​of the i-th and j-th data points, y j represents the dependent variable value of the jth data point, |x i -x j | is the distance between point i and point j on the independent variable axis, and w is the weight function.

4. The cosmetic preservative ingredient monitoring system according to claim 1, characterized in that: The steps for identifying the change in the concentration of the antiseptic component are specifically as follows: Based on the trend model data, the isolation forest algorithm is used to analyze the data points in the time series, using the formula: Calculate the anomaly score of each data point and obtain a list of preliminarily identified anomalies, where S(x) is the anomaly score of data point x, d(x) is the distance from the data point to the adjacent decision boundary, m is the average distance from the data point to the decision boundary, κ is the density adjustment coefficient, p(x) is the prior probability of data point x, and σ is the standard deviation of the data point; Based on the anomaly list, set a threshold to filter data points with high anomaly scores, using the formula: E(x)={x|S(x)>θ×α} Identify the abnormal data points that exceed the threshold and generate a list of data points with high anomaly scores, where θ is the threshold for identifying high anomaly scores, α is the threshold adjustment factor, E(x) is the set of abnormal data points, S(x) is the anomaly score of data point x, and x is the data point; The list of high anomaly scoring data points is validated by checking the context and neighboring values ​​of the data points using the formula: Generate abnormal preservative ingredient concentration change characteristics, where V(x) is a verified set of abnormal data points, E is a set of abnormal data points, μ is the average value of the data points, which is used to determine the degree of abnormality, λ is an adjustment parameter used to affect the sensitivity of the abnormality, and x is a data point.

5. The cosmetic preservative ingredient monitoring system according to claim 1, characterized in that: The comparison steps of the historical data mode are specifically as follows: Collect the historical data of the concentration change of the preservative component, and for each currently identified abnormal data point, calculate the difference from the historical data using the formula: Generate a list D(x) of deviations between the current data point x and the historical pattern, where h i is the concentration value of the antiseptic component at the historical data point, N represents the total number of historical data points, and w i is the weight of the ith data point, γ represents the power exponent used to adjust the sensitivity of the deviation calculation, δ is the parameter used for smoothing, and x is the concentration value of the preservative ingredient at the data point; According to the deviation list, the types of deviations are analyzed, including periodic changes, sudden increases or sudden decreases, and abnormality types are determined to obtain a component abnormality feature set.

6. The cosmetic preservative ingredient monitoring system according to claim 1, characterized in that: The steps for collecting the spectral data are specifically as follows: Based on the near-infrared spectroscopy technology, the near-infrared spectrometer parameters are set, the light source intensity and the detector sensitivity are adjusted, and the characteristics of the cosmetic sample are matched, using the formula: Generate the spectral device configuration parameters P, where L represents the light source intensity, η represents the device efficiency, S represents the sensitivity of the detector, A represents the absorbance of the sample, and c represents the optical path length; According to the spectral equipment configuration parameters, the spectral data of the cosmetic samples are collected, and the reflectance of each sample is recorded, using the formula: Calculate the reflectance R of the sample to obtain a spectral data set of the cosmetic sample, where I represents the reflected light intensity, I0 represents the incident light intensity, and β represents a correction factor used to adjust the error in the reflectance calculation; The spectral data set is formatted into a data file using the data organization formula: Generate spectral data file D, where R n represents the reflectivity of the data point, γ is the normalization factor, and N represents the total number of data points.

7. The cosmetic preservative ingredient monitoring system according to claim 1, characterized in that: The steps for obtaining the component concentration optimization index are specifically as follows: The spectral data is analyzed to identify the spectral characteristics of multiple components in the sample using the formula: Generate multivariate analysis results M, where x i represents a variable observation in the sample, w i represents the weight assigned to the i-th variable, α represents the power exponent used to amplify the impact of the observed value, λ and δ i represents the attenuation coefficient, and N represents the total number of variables; The multivariate analysis results were compared with the data acquired by machine vision using the formula: Generate a differential analysis log C, where M j represents the results from multivariate analysis, V j represents the data from machine vision, K represents the total number of data points compared, and β and γ represent the exponents used to adjust the sensitivity of the difference calculation; The Kalman filter was applied to optimize the concentration of the preservative component predicted based on the spectral and visual data, using the formula: F k =F k-1 +K k ·(Z k -F k-1 ) g Generate component concentration optimization index, where F k represents the estimated concentration value, K k represents the Kalman gain, Z k represents the current observation value, F k-1 Represents the previous estimate, and g represents the adjustment exponent of the update step size.

8. The cosmetic preservative ingredient monitoring system according to claim 1, characterized in that: The steps for obtaining the component monitoring feedback result are specifically as follows: Based on the component anomaly feature set and component concentration optimization index, real-time monitoring data is collected from the monitoring process adjusted by the dynamic update mechanism, using the formula: Generate real-time monitoring data, where D represents the total amount of real-time monitoring data collected, c i represents the concentration of the anticorrosive component at the i-th monitoring point, λ i represents the weight coefficient assigned according to the characteristics of the i-th monitoring point, and N represents the total number of monitoring points; To evaluate the deviation of the real-time monitoring data from the threshold, the formula is used: Generate monitoring data deviation evaluation information, where E represents the weighted average deviation between the monitoring data and the threshold, c i represents the real-time monitoring of the concentration of the preservative component, T set represents the concentration threshold of the preservative ingredient, w i represents the weight coefficient of the deviation, and N represents the total number of monitoring points; Based on the deviation assessment information, using the formula: F = "adjust" if E>θ else "keep" Generate component monitoring feedback result F, where E is the weighted average deviation between the monitoring data and the threshold, and θ is the decision threshold.

Citation Information

Patent Citations

  • Unattended urban and rural water supply operation and maintenance analysis system

    CN117670051A

  • Intelligent Internet of Things air quality monitoring system

    CN117953995A

  • Cable channel structure settlement monitoring system

    CN118260723A

  • Device state monitoring method based on multi-index cluster analysis

    WO2022252505A1

Cited By

  • Monitoring data optimization method for saline-alkali soil improvement equipment

    CN119807987A