Fragrance synthetic product component analysis method based on chromatographic data

Abstract

The invention relates to the technical field of chemical analysis data processing, in particular to a spice synthetic product component analysis method based on chromatographic data. The method comprises the following steps: acquiring chromatographic data of a plurality of historical qualified batches and batches to be detected, and calculating a gold mean value and a gold standard deviation of characteristics of all the historical qualified batches; constructing a perceptual stability mask, and decomposing the deviation between the features of each target batch and the corresponding golden mean value into perceptual deviation and amplitude deviation by using the perceptual stability mask; determining a dynamic process boundary; and judging whether the sensing deviation and the amplitude deviation of the batch to be detected are within the dynamic process boundary or not so as to realize analysis on the components of the perfume synthetic product. According to the scheme, the abnormal types in the components of the perfume synthetic product can be accurately diagnosed, the false alarm rate is reduced, and the detection sensitivity of key trace risks is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of chemical analysis data processing. More particularly, the present application relates to a method for analyzing components of a perfume synthesis product based on chromatographic data. BACKGROUND

[0002] Gas chromatography or liquid chromatography is a standard method for analyzing product purity, main component content, byproducts and trace impurity distribution. Furthermore, by automatically comparing and analyzing chromatographic fingerprints of successive production batches, the stability of the production process can be monitored in real time.

[0003] At present, the industry generally uses multi-dimensional statistical process control technology, such as principal component analysis (PCA) combined with a box plot method, to achieve product component analysis and thus achieve the purpose of production process monitoring. The above method first collects a large amount of historical qualified batch chromatographic data to establish a PCA model, which compresses high-dimensional chromatographic data into several key principal components, which represent the directions with the largest variance in the data. Then, the scores of these principal components are statistically analyzed using a box plot, and a boundary is set, which is usually dependent on the upper quartile Q3 and the lower quartile Q1 of the data, the interquartile range IQR is calculated, and a fixed statistical multiplier (usually 1.5) is used to define the boundary. When the principal component scores of a new batch fall outside the boundary, it is determined to be a process anomaly.

[0004] However, the prior art has defects in the specific scenario of perfume quality control. First, the mathematical principle of the PCA model is to find the direction with the largest variance in the data. In perfume production, most of the variance may come from slight fluctuations in several major perfume components. However, those trace impurities with extremely low content and extremely small variance, such as allergens or odor components, often have a decisive influence on the quality of the aroma. The standard PCA, in the dimensionality reduction process, will preferentially extract the principal component variance, discarding these key trace impurity signals as noise, resulting in a model that is not sensitive to this perception-critical anomaly, causing false negatives. Second, the statistical basis of the 1.5 times IQR rule of the box plot is to assume that the data follows a symmetric normal distribution. However, in actual chemical production, due to physical or chemical limitations of the process, the data distribution often presents an asymmetric skewness. At this time, if a symmetric and fixed boundary is forcibly used, it will result in a large number of false positives at the long tail, while the boundary is too wide at the short tail, causing the problem of false negatives. SUMMARY

[0005] The purpose of the present application is to propose a method for analyzing components of a perfume synthesis product based on chromatographic data, to solve the problem that the principal component analysis in the prior art is not sensitive to key trace impurities and the statistical rule of the standard box plot is not suitable for skewed data, thus causing false negatives and false positives. To this end, the present application provides a solution in the following aspect.

[0006] The application provides a perfume synthesis product component analysis method based on chromatographic data, comprising: obtaining chromatographic data of a plurality of historical qualified batches and a to-be-tested batch, performing binning processing on all the chromatographic data to obtain a plurality of features, and calculating golden mean values and golden standard deviations of each feature of all the historical qualified batches; constructing a perception stability mask, the perception stability mask being negatively correlated with a coefficient of variation corresponding to the feature, the coefficient of variation being obtained from the golden mean value and the golden standard deviation of each feature, and decomposing a deviation of the feature of each target batch from the corresponding golden mean value into a perception deviation and an amplitude deviation by using the perception stability mask, the target batch being any batch in all the historical qualified batches and the to-be-tested batch; obtaining sequences of all the historical qualified batches in the perception deviation and the amplitude deviation, calculating robust skewness of each sequence, and constructing an adjustment factor based on the robust skewness to determine a dynamic process boundary; judging whether the perception deviation and the amplitude deviation of the to-be-tested batch are within the dynamic process boundary, so as to realize analysis of components of a perfume synthesis product.

[0007] The above scheme replaces the variance-driven logic of traditional PCA by constructing a two-dimensional evaluation space of the perception deviation and the amplitude deviation, and replaces the fixed symmetric threshold of the standard box chart by constructing a dynamic process boundary, thereby solving the defects that PCA is difficult to distinguish trace impurities and the standard box chart misjudges skewed data, and achieving accurate diagnosis of abnormal types, while reducing the false positive rate and improving the detection sensitivity of key trace risks.

[0008] Optionally, a calculation formula of the perception stability mask is: ; In the formula, perception stability mask of the i-th feature is coefficient of variation of the i-th feature is median of the coefficients of variation of all the features is hyperbolic tangent function is The above construction of the perception stability mask can nonlinearly lower the weight of the feature corresponding to the coefficient of variation (i.e., a key region with low signal and small fluctuation), and suppress the weight of the feature with high coefficient of variation (such as a fluctuating main peak), thereby providing a basis for subsequent separation of the perception deviation and the amplitude deviation. Optionally, the perception deviation is:

[0009] ;

[0010] In the formula, perception deviation of the i-th feature is coefficient of variation of the i-th feature is median of the coefficients of variation of all the features is hyperbolic tangent function is For target batch Perceptual bias, For target batch The Middle The value of each feature, For the first The golden mean of each characteristic, For the first Perceptual stability mask for each feature The total number of features.

[0011] By calculating the sensing bias, the degree of deviation of the corresponding target batch in the critical trace area can be prioritized for evaluation, making the fluctuations of trace impurities that would be submerged in PCA significant, thus ensuring the sensitivity of sensing anomalies.

[0012] Optionally, the amplitude deviation is: ; In the formula, For target batch amplitude deviation, For target batch The Middle The value of each feature, For the first The golden mean of each characteristic, For the first Perceptual stability mask for each feature The total number of features.

[0013] Optionally, the adjustment factor includes an upper adjustment factor at the upper boundary and a lower adjustment factor at the lower boundary, wherein the upper adjustment factor and the lower adjustment factor are respectively: ; ; In the formula, and These are the up-regulation factor and down-regulation factor of the target sequence, respectively. For the robustness skewness of the target sequence, Based on the multipliers, For the maximum adjustment range, For skewness sensitivity, To find the maximum value function, The function is a hyperbolic tangent, and the target sequence is either a sequence with perceptual bias or a subsequence with amplitude bias.

[0014] By constructing adjustment factors, statistical boundaries can be automatically adjusted according to the true skewed distribution of the data, such as widening the boundaries on the long tail side of the data, thereby greatly reducing false alarms caused by data skewness.

[0015] Optionally, the robust skewness is obtained by using the median of medians.

[0016] Optionally, the dynamic process boundary comprises an upper boundary and a lower boundary, and the upper boundary and the lower boundary are respectively: In the formula, , are respectively an upper boundary and a lower boundary of the target sequence, and are respectively a lower quartile and an upper quartile of the target sequence, is a quartile range of the target sequence, , are respectively an upper adjustment factor and a lower adjustment factor of the target sequence.

[0017] The dynamic process boundary is automatically expanded on the long tail side of the data skewness, the boundary is significantly relaxed, the normal drift of the process is correctly determined as normal, and the false alarm rate is greatly reduced.

[0018] Optionally, the judgment of whether the perceptual deviation and the amplitude deviation of the to-be-tested batch are within the dynamic process boundary comprises: If the perceptual deviation and the amplitude deviation of the to-be-tested batch are both outside the corresponding dynamic process boundary, it is determined that a composite anomaly occurs; otherwise, it is determined that the to-be-tested batch is a process stable batch, and the perfume ingredient is qualified; If the perceptual deviation of the to-be-tested batch is greater than the upper boundary of the sequence of the perceptual deviation, and the amplitude deviation of the to-be-tested batch is normal, it is determined that a perceptual anomaly occurs, and a new impurity in trace amount or the loss of a key trace component affecting the aroma occurs; if the amplitude deviation of the to-be-tested batch is greater than the upper boundary of the sequence of the amplitude deviation, and the perceptual deviation of the to-be-tested batch is normal, it is determined that an amplitude anomaly occurs, and the proportion or total yield of the perfume ingredient fluctuates.

[0019] Optionally, the binning processing of all the chromatographic data comprises: dividing each chromatographic data into a plurality of equal-width time intervals; taking the peak area or average signal strength in each time interval as a corresponding feature.

[0020] Optionally, the method further comprises a step of pre-processing each chromatographic data, and the pre-processing comprises performing baseline correction of each chromatographic data by using an adaptive iteratively reweighted penalized least squares method.

[0021] The beneficial effects of the present application are: ​​First, this invention decouples batch deviation into perceived deviation and amplitude deviation by constructing a perceived stability mask. This enables the monitoring system not only to determine whether an anomaly exists, but also to diagnose the type of anomaly, providing valuable decision-making guidance for quality control personnel (for example, an anomaly in perceived deviation may indicate the need to investigate raw material contamination, while an anomaly in amplitude deviation may indicate the need to adjust reaction time). In other words, this invention overcomes the shortcomings of existing technologies that mix all variances in the principal components of the PCA model, making it impossible to distinguish between trace critical anomalies and regular fluctuations in principal components, thus achieving accurate diagnosis of anomaly types.

[0022] Secondly, by constructing adjustment factors and then constructing dynamic process boundaries, this invention solves the problem of misjudgment in standard box plots during quality control and improves the sensitivity to key risks. Attached Figure Description

[0023] Figure 1 This illustration schematically shows a flowchart of a method for analyzing the components of a fragrance synthesis product based on chromatographic data in this embodiment. Figure 2 The illustration shows a comparison between the traditional boundary and the adaptive boundary obtained by this invention in the batch quality monitoring of spices. Detailed Implementation

[0024] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0025] like Figure 1 As shown in this embodiment, a method for analyzing the components of a fragrance synthesis product based on chromatographic data includes the following steps: S101: Collect and preprocess chromatographic data from multiple historical batches and the batch to be tested, obtain chromatographic data from multiple historical qualified batches, and calculate the gold mean vector and gold standard deviation vector.

[0026] The process of obtaining the golden mean vector and the golden standard deviation vector in this embodiment is as follows: First, chromatographic data from multiple historical qualified batches were obtained.

[0027] In this embodiment, data collection Chromatographic data, such as gas chromatography (GC) data, of historical batches of flavoring products, and filtering from them. A batch that is fully qualified by manual verification or laboratory physicochemical indicators (such as aroma, purity, refractive index, etc.) is designated as a historical qualified batch (i.e., a gold batch), and chromatographic data of the gold batch is obtained.

[0028] For example, in this embodiment, data was collected. Chromatographic data from a number of historical batches can be filtered to select These historical batches are considered the "golden batches".

[0029] In this embodiment, it is also necessary to... All chromatographic data from the gold batch and the batch to be tested underwent uniform preprocessing. This preprocessing included: (1) Baseline correction: The drifted baseline is subtracted by adaptive iterative reweighted penalized least squares (airPLS) or morphological methods.

[0030] (2) Noise removal: Savitzky-Golay smoothing filter is used to remove high-frequency noise while retaining peak characteristics.

[0031] (3) Retention time alignment: Use correlation optimization algorithm (COW) or dynamic time warping (DTW) to eliminate retention time drift caused by column aging or flow rate fluctuations, and ensure that the characteristic peaks of all spectra are aligned.

[0032] Secondly, construct the golden feature matrix.

[0033] In this embodiment, the preprocessed chromatographic data are binned at fixed time intervals, that is, each preprocessed chromatographic data is divided into bins. The system calculates the peak area or average signal strength within each time interval as a feature, using equal-width time intervals.

[0034] It should be noted that although the chromatographic data are numerical sequences, they are generally presented in the form of chromatograms. Therefore, by integrating the signal intensity of the chromatograms, the peak area within each time interval can be obtained.

[0035] For example, in this embodiment, N bin-splitting operations can be performed during bin-splitting, thus obtaining one... The golden characteristic matrix. Among them, the features in the golden characteristic matrix... Representing the The first batch of gold in the Peak area over each time interval.

[0036] For example, if If there are bins, then the size of the golden characteristic matrix is .

[0037] The sequence number of the time interval mentioned above is the same as the sequence number of the feature.

[0038] Then, calculate the golden mean vector and the golden standard deviation vector.

[0039] Based on the golden characteristic matrix, two core benchmark vectors are calculated as the golden mean vector and the golden standard deviation vector. Both benchmark vectors are... A vector of dimension, specifically as follows: (1) Golden Mean Vector: The mean of the golden feature matrix is ​​calculated column by column (feature). , Let be the golden mean of the j-th feature in the golden mean vector.

[0040] The aforementioned golden mean vector represents the standard fingerprint of fragrance chromatography.

[0041] (2) Golden standard deviation vector: The standard deviation of the golden feature matrix is ​​calculated column by column (feature). , Let be the golden standard deviation of the j-th feature in the golden standard deviation vector.

[0042] The above-mentioned golden standard deviation vector reflects the allowable fluctuation range of each feature in normal production.

[0043] For example, suppose a gold batch Number of features The golden characteristic matrix at this time for: ; where columns 1, 2, and 4 represent principal components (large signal values), and column 3 represents trace impurities (small signal values).

[0044] Based on the above gold feature matrix, the gold mean vector is obtained as follows: The golden standard deviation vector is .

[0045] The above-mentioned standard models for subsequent comparison and evaluation were obtained using chromatographic data from all gold batches, namely the gold mean vector and the gold standard deviation vector.

[0046] S102: Based on the golden mean vector and the golden standard deviation vector, calculate the coefficient of variation of each data feature and construct a perceptual stability mask. Use the perceptual stability mask to decompose the deviation between the features of the target batch and the corresponding golden mean into perceptual deviation and amplitude deviation.

[0047] The process of obtaining the perceived stability mask in this embodiment is as follows: First, calculate the coefficient of variation for each feature to reflect its relative volatility.

[0048] In flavor analysis, the most critical features are those with low (trace) signals and high stability (small fluctuations). A feature with a mean and standard deviation close to 0 (i.e., a stable baseline region or trace impurities) is considered a major anomaly once it undergoes a significant change; therefore, it is necessary to obtain the relative volatility of the feature.

[0049] Specifically, the coefficients of variation for each feature are: ; In the formula, For the first The coefficient of variation of each feature For the first The gold standard deviation of each feature For the first The golden mean of each characteristic, A small offset constant is used to prevent [the following from happening]: The baseline region is subject to division by zero errors, ensuring the numerical stability of the calculation.

[0050] In one embodiment, it can be set .

[0051] For example, when the golden mean vector is The golden standard deviation vector is , At this point, we can obtain four coefficients of variation, which are: , , , .

[0052] As can be seen from the above example, the coefficient of variation of the third feature, which belongs to the trace features, is... The relative fluctuations are relatively large, while the coefficient of variation of the fourth characteristic, which belongs to the principal characteristics, is relatively large. The relative fluctuations are relatively small; therefore, the aforementioned trace characteristics are those that require attention.

[0053] Secondly, a perceptual stability mask is constructed to amplify "low CV" features (critical regions) and suppress "high CV" features (non-critical regions).

[0054] Specifically, the formula for calculating the perceived stability mask is: ; In the formula, For the first Perceptual stability mask for each feature For the first The coefficient of variation of each feature The median of the coefficients of variation for all features. It is the hyperbolic tangent function.

[0055] When the number of coefficients of variation is even, the mean of the two middle coefficients of variation can be used as the median.

[0056] The value range of the above-mentioned perceptual stability mask is [0,1]; This value serves as a normalization mechanism, exhibiting robustness and remaining unaffected by fluctuations in a few extreme cases of the coefficient of variation; when the coefficient of variation... Perceptual stability mask when much smaller than the median Approaching 1; when Perceptual stability mask when much larger than the median Close to 0.

[0057] In this embodiment, the perceptual stability masks of all features constitute the perceptual stability mask vector.

[0058] After obtaining the perceptual stability mask, the perceptual stability mask is used to obtain the following: Perception and amplitude deviation of the gold batch and the batch to be tested.

[0059] Specifically, the perception bias is used to assess the corresponding batch in the perception critical area ( The degree of deviation (of the region).

[0060] Among them, taking gold batch l as the target batch, the perceptual bias is calculated. The calculation formula is as follows: ; In the formula, Gold batch The Middle Values ​​on each feature Let j be the golden mean value corresponding to the j-th feature. For the first The perceptual stability mask for each feature, where N is the total number of features. The calculation of the perceptual bias described above is actually a weighted absolute bias.

[0061] For example, when a batch of gold The characteristic data of the chromatographic data after binning are as follows: And the constructed perceptual stability mask vector is So, gold batches The perceptual bias is approximately .

[0062] In the above embodiments, the weak signals of the sensing critical area are extracted separately and amplified onto the sensing deviation axis using a sensing stability mask, making the minute impurity fluctuations that would be submerged in PCA extremely significant, thus ensuring the detection sensitivity of high-risk sensing anomalies.

[0063] Among them, the amplitude deviation is used to assess the batch within the normal fluctuation range ( The degree of deviation (of the region). Specifically, the magnitude deviation. for: ; In the formula, Gold batch The Middle Values ​​on each feature Let j be the golden mean value corresponding to the j-th feature. For the first A perceptual stability mask for each feature, where N is the total number of features.

[0064] The above calculation of amplitude deviation is actually a weighted root mean square error. It uses... As a weight, it complements the perceptual bias.

[0065] For example, when a batch of gold The characteristic data of the chromatographic data after binning are as follows: So, gold batches The amplitude deviation is approximately .

[0066] In the above embodiments, The standard dataset of gold batches from The 1D space is mapped to a more information-dense 2D space composed of perception bias and amplitude bias, resulting in a 2D distribution cloud, which is the normal process model; where a 2D vector in the 2D distribution cloud is... , Gold batch amplitude deviation, Gold batch Perceptual bias.

[0067] It should be noted that the above embodiments are used to replace standard PCA. Their purpose is not to find the direction of maximum variance, but to separate the biases based on the importance of the chemical components.

[0068] Thus, by constructing a two-dimensional space, the deviations of all gold batches are decomposed into perceptual deviations reflecting changes in key trace areas and amplitude deviations reflecting changes in conventional major areas, solving the problem that PCA cannot clearly distinguish key impurities.

[0069] S103: Obtain the sequences of perceived deviations and amplitude deviations for all gold batches, calculate the robust skewness of each sequence, and construct the adjustment factor for the corresponding deviation based on the robust skewness, thereby determining the asymmetric dynamic process boundary.

[0070] It should be noted that this embodiment does not use a fixed 1.5 times. The boundary is not the point of reference, but rather the point where the amplitude deviation occurs. Where the axis and perception deviation are located Dynamic process boundaries are constructed for each axis.

[0071] Using the sequence of perceived deviation or the sequence of amplitude deviation as the target sequence, a dynamic process boundary for the target sequence is constructed. The specific process is as follows: First, robust statistics of the target sequence are calculated, including the upper quartile, lower quartile, interquartile range, and robust skewness.

[0072] Get The target sequence of gold batches .

[0073] Specifically, the upper quartiles of the above sequence are calculated. Lower quartiles Interquartile range ( and robust skewness .

[0074] The robustness skewness mentioned above was calculated using the Medcouple median. The value range is [-1, 1], which can accurately describe the data skewness and is not affected by outliers themselves.

[0075] Since the Medcouple method is an existing technology, its specific process will not be described in detail here.

[0076] For example, in obtaining After analyzing the target sequences of each gold batch, we can obtain the following statistical results: , , , .in, This indicates that the data is strongly right-skewed.

[0077] Since each deviation (amplitude deviation or perception deviation) in the target sequence is theoretically always Therefore, its distribution is very likely to be strongly right-skewed. ).

[0078] Second, calculate the adjustment factor.

[0079] against The right-skewed distribution requires an upward adjustment factor on the upper boundary of the axis containing the target sequence. The lower bound adjustment factor Then should be maintained (No need to tighten the short tail side). It should be noted that relaxing the upper boundary can prevent false alarms.

[0080] Specifically, the formulas for calculating the adjustment factor are as follows: ; ; In the formula, and These are the upper adjustment factor for the upper boundary and the lower adjustment factor for the lower boundary of the dynamic process boundary corresponding to the target sequence, respectively. For the robustness skewness of the target sequence, Let the basic multipliers be denoted as... This represents the baseline of the standard box plot. To set the maximum adjustment range, set it to... , For skewness sensitivity, set to .

[0081] The above formula uses the hyperbolic tangent function to construct a unidirectional, non-linear adjustment relationship, which allows the upper boundary to be dynamically amplified while the lower boundary remains unchanged. constant.

[0082] Then, based on the adjustment factor, the adaptive boundary is determined.

[0083] The axis where the target sequence is located in this embodiment ( shaft or The upper and lower boundaries of the final dynamic process boundary of the axis are: ; ; in, , The target sequence corresponds to the upper and lower boundaries of the dynamic process boundary, respectively.

[0084] It should be noted that when When the value is less than 0, the lower boundary value is 0, which is due to the amplitude deviation score. Its lower boundary is usually truncated by 0, so the actual lower boundary is 0.

[0085] In this embodiment, a set of asymmetric two-dimensional rectangular boundaries was obtained, namely The asymmetric two-dimensional rectangular boundaries of this group are dynamic boundaries.

[0086] The above method constructs an asymmetric boundary that automatically scales with data skewness by calculating robust skewness and adjustment factor, thus solving the problem of false alarms easily generated by traditional fixed boundaries on skewed data.

[0087] S104: Compare whether the perception deviation and amplitude deviation of the batch to be tested are within the dynamic process boundary, so as to judge the batch to be tested and realize the analysis of the components of the fragrance synthesis product.

[0088] In this embodiment, the batch to be tested is obtained. The amplitude deviation and perception deviation, i.e. It is then compared with the dynamic process boundary to determine and diagnose anomalies.

[0089] Specifically, the process of anomaly identification and diagnosis is as follows: (1) Case 1: If ( If the result is normal, it is judged as a sensory abnormality (high risk), and a high-priority alarm is immediately triggered. The diagnosis is: the batch being tested has deviated from the critical sensory area, possibly indicating the presence of trace amounts of new impurities that affect the aroma or the absence of key trace components.

[0090] (2) Case 2: If ( The result was normal, but it was determined to be an abnormal amplitude (medium risk), triggering a medium-priority alarm. The diagnosis was: the tested batch deviated from the normal fluctuation range, indicating that the proportion or total yield of the main flavoring components fluctuated significantly, and the process stability decreased.

[0091] (3) Scenario 3: If and All exceeded the boundary, which was determined to be a composite anomaly, triggering the highest priority alarm.

[0092] (4) Case 4: If the scores of both axes are within their respective adaptive boundaries, the batch to be tested is determined to be a process-stable batch.

[0093] For example, such as Figure 2 As shown, it is a two-dimensional scatter plot, with the X-axis representing the amplitude deviation ( The Y-axis represents perceptual bias. ).

[0094] If the final adaptive boundary is Boundary is , The boundary is: So: For Figure 2 For batch A to be tested, ,at this time and If so, then the batch A to be tested is a batch with stable process.

[0095] for Figure 2 For batch B to be tested, ,at this time If so, then the batch B to be tested is of abnormal amplitude.

[0096] for Figure 2 For batch C to be tested, ,at this time If the batch C to be tested is found to be perceptually abnormal (high risk), then the batch C to be tested is considered to be abnormal (high risk).

[0097] in, Figure 2 The blue dots in the middle represent The distribution of gold batches in two-dimensional space is right-skewed on both axes. The red dashed box represents a fixed 1.5x. The rule, with its overly strict boundary, incorrectly classifies many gold batches (blue dots) and long-tail normal batches (such as the blue square test batch D) outside the bounding box, resulting in a large number of false alarms.

[0098] The green solid line box represents the dynamic boundary of the invention. Its upper and right boundaries extend significantly outward. The green box completely encompasses all gold batches and normal batch D, eliminating false alarms, while still correctly detecting batch B with X-axis anomalies (amplitude anomalies) and batch C with Y-axis anomalies (sensory anomalies). This demonstrates that the invention maintains high detection sensitivity for true anomalies while eliminating false alarms.

[0099] Therefore, from Figure 2 As can be seen, by comparing the two-dimensional spatial deviation of the batch under test with the dynamically constructed boundary, it is possible not only to accurately determine whether the batch is abnormal, but also to diagnose the specific type of abnormality, providing a precise basis for production control decisions.

[0100] The solution of this invention solves the shortcomings of PCA in distinguishing trace impurities and the shortcomings of standard box plots in misjudging skewed data, and can achieve accurate diagnosis of anomaly types.

[0101] In the description of this specification, "multiple" means at least two, such as two, three or more, etc., unless otherwise expressly and specifically defined.

[0102] While various embodiments of the invention have been shown and described in this specification, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention.

Claims

1. A method for component analysis of fragrance synthesis products based on chromatographic data, characterized in that, include: Acquire chromatographic data from multiple historical qualified batches and batches to be tested, divide all chromatographic data into bins to obtain multiple features, and calculate the gold mean and gold standard deviation of each feature for all historical qualified batches. A perceptual stability mask is constructed, which is negatively correlated with the coefficient of variation of the corresponding feature. The coefficient of variation is obtained from the golden mean and golden standard deviation of each feature. The perceptual stability mask is used to decompose the deviation between the feature of each target batch and the corresponding golden mean into perceptual deviation and amplitude deviation. The target batch is any batch among all historical qualified batches and the batch to be tested. Obtain the sequences of all historical qualified batches in terms of perception deviation and amplitude deviation, calculate the robust skewness of each sequence, and construct an adjustment factor based on the robust skewness to determine the dynamic process boundary. Determine whether the perception deviation and amplitude deviation of the batch to be tested are within the dynamic process boundary in order to analyze the composition of the fragrance synthesis product.

2. The method for analyzing the components of fragrance synthesis products based on chromatographic data according to claim 1, characterized in that, The formula for calculating the perceived stability mask is: ； In the formula, For the first Perceptual stability mask for each feature For the first The coefficient of variation of each feature The median of the coefficients of variation for all features. It is the hyperbolic tangent function.

3. The method for analyzing the components of fragrance synthesis products based on chromatographic data according to claim 1 or 2, characterized in that, The perception bias is: ； In the formula, For target batch Perceptual bias, For target batch The Middle The value of each feature, For the first The golden mean of each characteristic, For the first Perceptual stability mask for each feature The total number of features.

4. The method for analyzing the components of fragrance synthesis products based on chromatographic data according to claim 3, characterized in that, The amplitude deviation is: ； In the formula, For target batch amplitude deviation, For target batch The Middle The value of each feature, For the first The golden mean of each characteristic, For the first Perceptual stability mask for each feature The total number of features.

5. The method for analyzing the components of fragrance synthesis products based on chromatographic data according to claim 1, characterized in that, The adjustment factor includes an upper boundary adjustment factor and a lower boundary adjustment factor, wherein the upper and lower adjustment factors are respectively: ； ； In the formula, and These are the up-regulation factor and down-regulation factor of the target sequence, respectively. For the robustness skewness of the target sequence, Based on the multipliers, For the maximum adjustment range, For skewness sensitivity, To find the maximum value function, The function is a hyperbolic tangent, and the target sequence is either a sequence with perceptual bias or a subsequence with amplitude bias.

6. The method for analyzing the components of fragrance synthesis products based on chromatographic data according to claim 1, characterized in that, The robustness skewness is obtained using the Meadow median.

7. The method for analyzing the components of fragrance synthesis products based on chromatographic data according to claim 5, characterized in that, The dynamic process boundary includes an upper boundary and a lower boundary, wherein the upper boundary and the lower boundary are respectively: ； ； In the formula, , These are the upper and lower boundaries of the target sequence, respectively. and These are the lower quartile and upper quartile of the target sequence, respectively. The interquartile range of the target sequence. , These are the up-regulation factor and down-regulation factor of the target sequence, respectively.

8. The method for analyzing the components of fragrance synthesis products based on chromatographic data according to claim 1, characterized in that, The step of determining whether the perception deviation and amplitude deviation of the batch to be tested are within the dynamic process boundary, in order to achieve the analysis of the components of the fragrance synthesis product, includes: If both the perception deviation and amplitude deviation of the batch to be tested are outside the corresponding dynamic process boundary, it is determined to be a composite anomaly; otherwise, the batch to be tested is determined to be a process-stable batch and the fragrance components are qualified. If the perceptual deviation of the batch to be tested is greater than the upper boundary of the perceptual deviation sequence, and the amplitude deviation of the batch to be tested is normal, it is judged as perceptual abnormality, and there is a trace amount of new impurities or missing key trace components that affect the aroma; if the amplitude deviation of the batch to be tested is greater than the upper boundary of the amplitude deviation sequence, and the perceptual deviation of the batch to be tested is normal, it is judged as amplitude abnormality, and there is fluctuation in the proportion or total yield of flavor components.

9. The method for analyzing the components of fragrance synthesis products based on chromatographic data according to claim 1, characterized in that, The binning process for all chromatographic data yields several features, including: Each chromatographic data is divided into multiple time intervals of equal width; The peak area or average signal intensity within each time interval is used as the corresponding feature.

10. The method for analyzing the components of fragrance synthesis products based on chromatographic data according to claim 1, characterized in that, Also includes: The preprocessing steps for each chromatographic data include: baseline correction of each chromatographic data using an adaptive iterative reweighted penalized least squares method.

Images