Method for identifying heat extractant of preservative film by infrared spectrum
By repeatedly acquiring and processing infrared spectra, the problems of weak signal, background interference, and peak position drift in the detection of heat-induced precipitates from plastic wrap were solved, resulting in a more stable discrimination effect, forming a consistent evidence chain across temperatures, and improving the detection accuracy of heat-induced precipitates from plastic wrap.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- GUANGDONG OCEAN UNIVERSITY
- Filing Date
- 2026-04-21
- Publication Date
- 2026-07-21
AI Technical Summary
Existing technologies for detecting heat-extracted substances from plastic wrap suffer from problems such as weak signals and significant background interference, failure to effectively convert multiple temperature information, peak position drift leading to unstable peak selection, and lack of statistical evidence chains, making it difficult to make stable and reliable judgments.
By repeatedly acquiring infrared spectra and blank control spectra, the threshold and first spectrum are obtained through linear interpolation and normalization. The hit ratio is calculated to form a consistent evidence chain across temperatures, and the presence and risk level of heat-induced precipitates from the plastic wrap are determined.
It significantly enhances the detectability of newly precipitated signals due to heat, improves discrimination sensitivity and signal-to-noise ratio, reduces false negatives and false positives, forms a robust statistical evidence chain, and improves the stability and interpretability of discrimination.
Smart Images
Figure CN122432436A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of food contact material safety testing, specifically to a method for identifying heat-induced precipitates from food preservation film using infrared spectroscopy. Background Technology
[0002] With increasingly stringent regulations on food safety and food contact materials, plastic wrap may experience "heat-induced leaching" phenomena, such as additive migration and the formation of thermal oxidation products, when used in hot food, hot soup, microwave / steaming, and other scenarios. Infrared spectroscopy, due to its advantages of speed, non-destructiveness, and sensitivity to organic functional groups, is widely used for material identification and organic matter detection. However, in the scenario of "heat-induced leaching from plastic wrap," the leached substances often exhibit low concentration, weak absorption, and susceptibility to background and drift interference, making it difficult to obtain stable conclusions through conventional spectral interpretation or single-temperature detection. Therefore, there is an urgent need for an infrared spectral discrimination method that addresses the mechanism of heat-induced leaching and can form a chain of evidence.
[0003] Currently, conventional infrared spectroscopy methods are used to detect and identify substances released from plastic wrap when heated, but the following problems exist: Weak signals and significant background interference: Low precipitate concentration and weak absorption peaks, along with fluctuations in the medium / container / temperature background and baseline, can obscure key peaks, resulting in blurred signals and poor repeatability.
[0004] Multi-temperature information failed to be effectively converted into evidence of "thermally enhanced precipitation": directly averaging or simply comparing multiple temperature spectra easily introduces fixed background and non-temperature-related components into the results, thus diluting the "temperature-increasing" new precipitation signal.
[0005] Peak position drift leads to unstable peak extraction window: factors such as temperature, ATR contact, and instrument wavenumber axis cause peak position shift, which can lead to missed / false detections when extracting peaks from a fixed peak position window, especially in the case of weak peaks.
[0006] Lack of statistical evidence chain and consistency rules: Single spectrum discrimination is easily affected by random noise and local perturbation, and lacks interpretable discrimination basis of "multiple repetitions + cross-temperature consistency", making it difficult to use for stable screening and classification. Summary of the Invention
[0007] To overcome the problems of weak signal and significant background interference, failure to effectively convert multiple temperature information, unstable peak selection due to peak position drift, and lack of statistical consistency in the prior art, this invention provides a method for identifying heat-extracted substances from plastic wrap using infrared spectroscopy.
[0008] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: A method for identifying heat-extracted substances from plastic wrap using infrared spectroscopy includes the following steps: S1: Repeatedly collect infrared spectra of containers containing plastic wrap and blank control spectra containing only containers at different temperatures; S2: Preprocess the infrared spectrum and blank control spectrum to obtain the threshold and the first spectrum; S3: Obtain the global threshold and the second spectrum based on the threshold and the first spectrum; S4: Calculate the hit ratio based on the global threshold and the second spectrum, determine whether there are precipitates and the risk level based on the hit ratio, and complete the identification of heat-induced precipitates from the plastic wrap.
[0009] Furthermore, the preprocessing of the infrared spectrum and the blank control spectrum to obtain the threshold and the first spectrum includes: Linear interpolation with a unified wavenumber grid is applied to the infrared spectrum and the blank control spectrum to obtain a unified infrared spectrum and a unified blank control spectrum. The wavenumber grid is a set of discrete sampling points taken at fixed intervals on a continuous wavenumber axis of the infrared spectrum. The difference between the unified infrared spectrum and the unified blank control spectrum is calculated and normalized to obtain a normalized difference spectrum. The mean difference spectrum is obtained by taking the mean of multiple samples of the normalized difference spectrum. The mean and standard deviation of the normalized difference spectrum within the first range of the wavenumber grid are calculated, and the threshold is obtained based on the mean and standard deviation. Calculate the root mean square of the mean difference spectrum within the second range of the wavenumber grid, and normalize the mean difference spectrum based on the root mean square to obtain the normalized mean difference spectrum. The temperature difference spectrum is obtained by subtracting the normalized mean difference spectrum of adjacent temperatures, and the temperature difference spectrum is weighted and fused to obtain the first spectrum.
[0010] Furthermore, the wavenumber grid w It is the set of discrete sampling points taken at fixed intervals on the continuous wavenumber axis of the infrared spectrum, specifically:
[0011] In the formula, a Indicates the sampling point.
[0012] Furthermore, the step of obtaining the normalized difference spectrum by subtracting and normalizing the unified infrared spectrum and the unified blank control spectrum includes: The difference spectrum is obtained by subtracting the unified infrared spectrum and the unified blank control spectrum. The normalized difference spectrum is then obtained by using the difference spectrum and the normalization coefficient, as shown in the formula: , , ,
[0013] In the formula, Represented as temperature T The difference spectrum is collected a number of times, r. Indicates temperature as T A unified infrared spectrum collected a number of times, r. Indicates temperature as T Unified blank control spectrum, Expressed as contact strength factor, in units of cm 2 / ml , Represented as contact area, Expressed as the volume of the medium, Indicates temperature as T The normalization coefficient for the number of data collections, r. Indicates the standard contact strength factor. Indicates temperature as T The contact intensity factor is collected a number of times, r. Indicates the standard contact area. Indicates the standard medium volume. Indicates temperature as T The contact area is collected in r samples. Indicates temperature as T The volume of the medium is sampled a number of times, r. Indicates temperature as T The normalized difference spectrum is collected a number of times, r.
[0014] Furthermore, the mean difference spectrum is obtained by averaging multiple samples of the normalized difference spectrum, and the mean and standard deviation of the normalized difference spectrum within the first range of the wavenumber grid are calculated. A threshold is then obtained based on the mean and standard deviation, using the following formula: , ,
[0015] In the formula, Indicates temperature as T The mean difference spectrum, where R represents the total number of samples. Indicates the first range of the wavenumber grid. This represents the mean. Indicates standard deviation, This represents the threshold.
[0016] Furthermore, the root mean square of the mean difference spectrum within the second range of the wavenumber grid is calculated, and the normalized mean difference spectrum is obtained by normalizing the mean difference spectrum based on the root mean square, as shown in the formula: , ,
[0017] In the formula, Indicates the second range of the wavenumber grid. Represents the root mean square. This represents the number of wavenumber grid points in the mean difference spectrum that fall within the second range of the wavenumber grid. Represented as temperature T The normalized mean difference spectrum, It is a very small positive number.
[0018] Furthermore, the temperature difference spectrum is obtained by subtracting the normalized mean difference spectrum of adjacent temperatures, and the first spectrum is obtained by weighted fusion of the temperature difference spectrum, as shown in the formula:
[0019] ,
[0020] In the formula, Represents the temperature difference spectrum. , , , , , , Indicates the first i A key peak strength. Indicates the first i Each weight, This indicates the first spectrum.
[0021] Furthermore, obtaining the global threshold and the second spectrum based on the threshold and the first spectrum includes: The threshold that is the largest among all temperatures is taken as the global threshold. The formula is:
[0022] Then, the second spectrum is obtained based on the global threshold and the first spectrum, specifically: Define drift range , Pick , Indicates the step size; The first spectrum is interpolated and shifted according to any drift value within the drift range to obtain the shifted first spectrum. The key peak intensity is calculated based on the shifted first spectrum. The key peak intensity is compared with the global threshold, and the drift value corresponding to the key peak intensity being greater than the global threshold is selected. The first spectrum is then interpolated and shifted according to the drift value to obtain the second spectrum.
[0023] Furthermore, the hit ratio is calculated based on the global threshold and the second spectrum using the following formula: According to the second spectrum Calculate the intensity of all critical peaks, and then calculate the hit ratio based on the critical peak intensity and the global threshold. The formula is as follows:
[0024]
[0025] In the formula, Indicates temperature as T The second spectrum collected r times. Indicates temperature as T The number of collections is r, the th i The key peak strength. Indicates the first i The temperature is T The hit rate This is an indicator function.
[0026] Furthermore, the determination of the presence and risk level of precipitates based on the hit ratio includes: When the hit rate is at temperature T At or above 100℃, and At least at one of the temperatures, it is greater than or equal to 0.6 and When the value is greater than or equal to 0.6, it is determined that there are precipitates and a high risk; when and At least at one of the temperatures, it is greater than or equal to 0.6, but Less than 0.6, or when and If the temperature is equal to 0.6 at only one of the three temperatures, it is considered that there is a suspected presence of precipitates and a retest is recommended; otherwise, it is considered that there are no precipitates and the risk is low.
[0027] Compared with the prior art, the beneficial effects of the technical solution of the present invention are: It significantly enhances the signal of newly precipitated substances when heated and improves the detectability of weak peaks. It suppresses fixed background and non-temperature-related components from a mechanistic perspective, making the infrared spectral peaks of newly precipitated substances in the plastic wrap more prominent, thereby improving the discrimination sensitivity and signal-to-noise ratio. It also reduces false negatives and false positives, and improves the stability of characteristic peak window values. It forms a statistical evidence chain that is consistent across temperatures, making the discrimination more robust and interpretable, and reducing misjudgments caused by random noise. Attached Figure Description
[0028] Figure 1 This is a flowchart of a method for identifying substances released from plastic wrap when heated using infrared spectroscopy.
[0029] Figure 2 This is a comparison of the infrared spectrum and blank control spectrum of a method for identifying heat-induced precipitates from plastic wrap using infrared spectroscopy.
[0030] Figure 3 This is a mean difference spectrum of a method for identifying heat-induced precipitates from plastic wrap using infrared spectroscopy.
[0031] Figure 4 This is a temperature difference spectrum comparison diagram of a method for identifying heat-induced precipitates from plastic wrap using infrared spectroscopy.
[0032] Figure 5 This is the first spectral (fusion spectrum) of a method for identifying heat-extracted substances from plastic wrap using infrared spectroscopy. Detailed Implementation
[0033] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.
[0034] The terms "first," "second," "third," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.
[0035] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.
[0036] In a first embodiment of the present invention, a method for identifying heat-extracted substances from plastic wrap using infrared spectroscopy is provided, such as... Figure 1 and Figure 2 As shown: Includes the following steps: S1: Repeatedly collect infrared spectra of containers containing plastic wrap and blank control spectra containing only containers at different temperatures; S2: Preprocess the infrared spectrum and blank control spectrum to obtain the threshold and the first spectrum; S3: Obtain the global threshold and the second spectrum based on the threshold and the first spectrum; S4: Calculate the hit ratio based on the global threshold and the second spectrum, determine whether there are precipitates and the risk level based on the hit ratio, and complete the identification of heat-induced precipitates from the plastic wrap.
[0037] In a further embodiment, step S1 simulates the thermal precipitation of the same batch of cling film under a preset temperature sequence, and performs multiple repeated samplings at each temperature point (e.g., n=20), simultaneously acquiring blank control spectra at the same temperature; at the same time, key experimental conditions such as contact area, medium volume, and heat preservation time are recorded, establishing a data structure of "multiple temperature gradients + multiple repeated statistics," providing a traceable and quantifiable chain of evidence for subsequent construction of temperature difference spectra and cross-temperature consistency discrimination, specifically: Because the amount of substances released from plastic wrap usually increases with increasing temperature (migration rate, dissolution / evaporation, and the formation of thermal oxidation products are all temperature-dependent), the temperature range should cover the gradient from "light heating → moderate heating → near cooking / hot food contact" and the temperature difference between adjacent temperatures should be sufficient to create an increment of ΔT, as shown in Table 1.
[0038] Table 1
[0039] The infrared spectrum of the sample measurement object is defined as follows:
[0040] In the formula, The infrared spectrum of the r-th precipitation sample at temperature T; The blank sample (i.e., the blank control spectrum) is defined as:
[0041] In the formula, To and Infrared spectra obtained under the same temperature, medium, container, and heating process, but without the use of plastic wrap; In addition, to ensure comparability across temperatures, step S1 also records the observable / recordable experimental conditions as shown in Table 2.
[0042] Table 2
[0043] In a specific embodiment, step S1 collected infrared spectra under the experimental conditions shown in Table 3, as follows: Figure 2 As shown.
[0044] Table 3
[0045] Simultaneously, the collected infrared spectra were standardized, specifically as follows: Building the dataset:
[0046] Blank set:
[0047] This step ensures that each You can find the corresponding one. ; Define the repeated measures index: The test was performed 20 times at each temperature to prevent "random noise peaks" from being misidentified as precipitates. The results were compared at each temperature T and for each characteristic peak j, and the hit rate was calculated at the end.
[0048] In specific embodiments, the repeatability metrics are shown in Table 4: Table 4
[0049] Furthermore, the preprocessing of the infrared spectrum and the blank control spectrum to obtain the threshold and the first spectrum includes: Linear interpolation with a unified wavenumber grid is applied to the infrared spectrum and the blank control spectrum to obtain a unified infrared spectrum and a unified blank control spectrum. The wavenumber grid is a set of discrete sampling points taken at fixed intervals on a continuous wavenumber axis of the infrared spectrum. The difference between the unified infrared spectrum and the unified blank control spectrum is calculated and normalized to obtain a normalized difference spectrum. The mean difference spectrum is obtained by taking the mean of multiple samples of the normalized difference spectrum. The mean and standard deviation of the normalized difference spectrum within the first range of the wavenumber grid are calculated, and the threshold is obtained based on the mean and standard deviation. Calculate the root mean square of the mean difference spectrum within the second range of the wavenumber grid, and normalize the mean difference spectrum based on the root mean square to obtain the normalized mean difference spectrum. The temperature difference spectrum is obtained by subtracting the normalized mean difference spectrum of adjacent temperatures, and the temperature difference spectrum is weighted and fused to obtain the first spectrum.
[0050] In this embodiment, the sample spectrum and blank spectrum at each temperature are unified to the same wavenumber grid, and then the blank spectrum at the same temperature is differentially analyzed to obtain the difference spectrum signal. The difference spectrum signal is then normalized according to the contact intensity factor (such as A / V) to obtain the normalized difference spectrum set and its mean difference spectrum for each temperature. At the same time, the noise mean and standard deviation are statistically analyzed in the preset noise range to form the basis for threshold setting, thereby ensuring that the cross-temperature difference, peak selection and discrimination have consistent dimensions and reliable threshold sources.
[0051] Furthermore, the wavenumber grid w It is the set of discrete sampling points taken at fixed intervals on the continuous wavenumber axis of the infrared spectrum, specifically:
[0052] That is, 4000 → 400cm -1 , every 2cm -1 a little, a Indicates the sampling point; For any original spectrum X(w), resample to a uniform grid using linear interpolation:
[0053] In a specific embodiment, as shown in Table 5, assuming that a certain sample spectrum has two points in the original derivation: w a =1736cm -1 Time X(w) a )=0.205 w b =1734cm -1 Time X(w) b )=0.213 A unified grid requires w k =1735cm -1 Linear interpolation:
[0054] Table 5
[0055] Furthermore, the step of obtaining the normalized difference spectrum by subtracting and normalizing the unified infrared spectrum and the unified blank control spectrum includes: The difference spectrum is obtained by subtracting the unified infrared spectrum and the unified blank control spectrum. The normalized difference spectrum is then obtained by using the difference spectrum and the normalization coefficient, as shown in the formula: , , ,
[0056] In the formula, Represented as temperature T The difference spectrum is collected a number of times, r. Indicates temperature as T A unified infrared spectrum collected a number of times, r. Indicates temperature as T Unified blank control spectrum, Expressed as contact strength factor, in units of cm 2 / ml , Represented as contact area, Expressed as the volume of the medium, Indicates temperature as T The normalization coefficient for the number of data collections, r. Indicates the standard contact strength factor. Indicates temperature as T The contact intensity factor is collected a number of times, r. Indicates the standard contact area. Indicates the standard medium volume. Indicates temperature as T The contact area is collected in r samples. Indicates temperature as T The volume of the medium is sampled a number of times, r. Indicates temperature as T The normalized difference spectrum collected a number of times r; The difference spectrum is obtained by subtracting the uniform infrared spectrum and the uniform blank control spectrum because temperature will change the medium baseline and the system response; at the same temperature, subtracting the blank control can cancel out the variable factors of "medium + container + temperature background" to the greatest extent, so that the remaining changes in the difference spectrum are closer to those caused by precipitates. The intensity of plastic wrap precipitation is directly related to the contact area and the volume of the medium. If the A / V ratio is inconsistent at different temperatures, the "difference in contact conditions" will be misjudged as the "increase in precipitation due to temperature difference". Therefore, the purpose of calculating the normalized difference spectrum is to ensure that the temperature difference only reflects the difference in precipitates and is not affected by other factors.
[0057] In a specific embodiment, following the example of step S1, 80 o C is the 7th iteration, k=1.25, and we assume a wavenumber point w=1735cm on a uniform grid. -1 : Sample spectrum: (After interpolation) Blank spectrum:
[0058] Difference spectrum:
[0059] Normalized difference spectrum:
[0060] Furthermore, such as Figure 3 As shown, the mean difference spectrum is obtained by averaging multiple samples of the normalized difference spectrum. The mean and standard deviation of the normalized difference spectrum within the first range of the wavenumber grid are calculated, and the threshold is obtained based on the mean and standard deviation. The formula is as follows: , ,
[0061] In the formula, Indicates temperature as T The mean difference spectrum, where R represents the total number of samples. Indicates the first range of the wavenumber grid. This represents the mean. Indicates standard deviation, Indicates the threshold; The purpose of obtaining the mean difference spectrum is to ensure comparability across temperatures.
[0062] In this embodiment, a band that is typically relatively flat and has few characteristic peaks is selected as the noise estimation region, such as... ; Calculate noise statistics (e.g., 80) o C): Take the point set of the difference spectrum in the noise region each time:
[0063] Calculate the noise mean and standard deviation: ,
[0064] Calculate threshold For each peak j in the peak group library, the threshold can be uniformly set as follows:
[0065] Commonly used This indicates that the peak intensity exceeds a significant level of random noise fluctuations, as shown in Table 6.
[0066] Table 6
[0067] Furthermore, the root mean square of the mean difference spectrum within the second range of the wavenumber grid is calculated, and the normalized mean difference spectrum is obtained by normalizing the mean difference spectrum based on the root mean square, as shown in the formula: , ,
[0068] In the formula, Indicates the second range of the wavenumber grid. Represents the root mean square. This represents the number of wavenumber grid points in the mean difference spectrum that fall within the second range of the wavenumber grid. Represented as temperature T The normalized mean difference spectrum, It is a very small positive number.
[0069] In a specific embodiment, step S2 has already yielded... Because the precipitation intensity varies at different temperatures, if you directly perform... The difference is affected by the "overall scale," causing the high-temperature segment to dominate everything, which is not conducive to the fair integration of the "new information" of each ΔT segment. Therefore, the main discriminant interval is introduced: This range often includes carbonyl groups, C–O and other absorptions related to additive migration / oxidation; it is directly related to the purpose of "precipitate discrimination"; then the root mean square is calculated to obtain the normalized mean difference spectrum, as shown in Table 7.
[0070] Table 7
[0071] The specific calculation process is as follows: Assume that at a certain wavenumber point w=1735cm -1 (Near the carbonyl group)
[0072]
[0073] First, return to one:
[0074]
[0075] Temperature difference spectrum (60→80):
[0076] Furthermore, such as Figure 4 and Figure 5 As shown, the temperature difference spectrum is obtained by subtracting the normalized mean difference spectrum of adjacent temperatures, and the first spectrum is obtained by weighted fusion of the temperature difference spectrum, as shown in the formula:
[0077] ,
[0078] In the formula, Represents the temperature difference spectrum. , , , , , , Indicates the first i A key peak strength. Indicates the first i Each weight, Indicates the first spectrum; The purpose of obtaining the temperature difference spectrum by subtracting the normalized mean difference spectrum of adjacent temperatures is to significantly offset the influence of factors other than temperature on the peak intensity of the precipitates.
[0079] In a specific embodiment, in order to strongly bind the fusion with the "discrimination of exudates from heated plastic wrap", the weights do not use generalized SNR, but instead use the "intensity of newly added exudate key peaks".
[0080] In this embodiment, the sources of the key peak set are: 1. Common migration peaks of acetic acid / additives: 1735 cm⁻¹ (C=O), 1240 cm⁻¹ (C=O) (commonly used as indicator peaks in infrared discrimination of migration / additives / oxidation products in food contact materials) 2. Given peak window tolerance (derived from instrument resolution / residual drift, which S4 will also compensate for). 3. 1735±8cm -1 1240±10cm -1 ; A new key peak intensity A is added. The definition and calculation of A are as follows: For each △T segment (e.g., 60->80)
[0081] The specific calculation process is as follows: Taking a window of 1735±8 from 60 to 80 as an example: the window is [1727, 1743]; With a uniform grid step size of 2cm⁻¹, the wavenumber points within the window are: w={1743,1741,1739,1737,1735,1733,1731,1729,1727} Take the absolute value point by point:
[0082] The maximum value is the peak intensity contribution of that window.
[0083] The same applies to 1240±10.
[0084] The weight β calculation and fusion spectrum definition (from A△ to β to D_fus) are shown in Table 8.
[0085] The newly added key peak intensities of each ΔT segment are normalized to a weight:
[0086] Final fusion spectrum:
[0087] Table 8
[0088] Taking a wavenumber point w=1735 as an example, assume the values of the three ΔT difference spectra at this point are as follows:
[0089]
[0090]
[0091]
[0092] Term-by-term multiplication: 0.3908 * 0.1237 = 0.0484 0.2846 · 0.1010 = 0.0287 0.3246 * 0.1155 = 0.0375 Add:
[0093] Similarly, by calculating each point of w, the entire line D can be obtained. fus (w), as shown in Table 9.
[0094] Table 9
[0095] Furthermore, obtaining the global threshold and the second spectrum based on the threshold and the first spectrum includes: The threshold that is the largest among all temperatures is taken as the global threshold. The formula is:
[0096] Then, the second spectrum is obtained based on the global threshold and the first spectrum, specifically: Define drift range , Pick , Indicates the step size; The first spectrum is interpolated and shifted according to any drift value within the drift range to obtain the shifted first spectrum. The key peak intensity is calculated based on the shifted first spectrum. The key peak intensity is compared with the global threshold, and the drift value corresponding to the key peak intensity being greater than the global threshold is selected. The first spectrum is then interpolated and shifted according to the drift value to obtain the second spectrum. Specifically, the drift value corresponding to the condition where all key peak intensities are greater than the global threshold is selected. If there are multiple corresponding drift values, the drift value corresponding to the minimum key peak intensity is selected.
[0097] In this embodiment, the drift search range and step size are set as follows:
[0098] △max source: The offset caused by instrument wavenumber axis drift and temperature / contact is usually within a small range (it is common in engineering to use ±10cm-1 as a conservative search window); The 1cm-1 step size is derived from the fact that it is on the same order of magnitude as the resolution / sampling interval of common FTIR systems and is simple to implement.
[0099] Example:
[0100] Peak group library is defined as: Establish key peak groups: ①Group A (Additives / Ester Migration Indicator): 1735cm -1 (C=0), 1240cm -1 (C-0) ②Group C (Indicator of Thermal Oxidation Products): 1710cm -1 (Oxidized carbonyl shoulder) Each peak is assigned a tolerance δ: ①1735±8,1240±10,1710±10 Step S2 has defined the noise region Ω noise And statistically analyze μ noise ,δ noise .
[0101] The threshold uses a simple 3δ significance test:
[0102] To be conservative, the largest θ among all temperatures can be used as the global threshold (more stable, fewer false positives):
[0103] Set the objective function: Peak group consistency (maximum hit rate). For any candidate drift Δ, first translate (interpolate and align) the entire fusion spectrum:
[0104] In the formula, Align is the interpolation translation (the interpolation method has been explained in step S2); Then, the peak intensity is calculated for each peak j:
[0105] Hit:
[0106] Define peak group consistency score:
[0107] in: #Hit A Count A is the number of hits in the two peaks of group A (0 / 1 / 2); #Hit C This is the number of hits in group C (0 / 1); choose:
[0108] If there are ties for the maximum value, take the one with the smallest |A| (which is more in line with the physical intuition of "minimizing drift");
[0109] Final compensation spectrum:
[0110] In a specific embodiment, it is assumed that the global threshold θ is obtained through S2. globa =0.0067, calculated for A∈[-6,0] (excerpt), as shown in Table 10.
[0111] Table 10
[0112] The maximum score of 3 occurs when Δ = -5, therefore: = 5 cm 1 And obtain Dcor(w).
[0113] Furthermore, the hit ratio is calculated based on the global threshold and the second spectrum using the following formula: According to the second spectrum Calculate the intensity of all critical peaks, and then calculate the hit ratio based on the critical peak intensity and the global threshold. The formula is as follows:
[0114]
[0115] In the formula, Indicates temperature as T The second spectrum collected r times. Indicates temperature as T The number of collections is r, the th i The key peak strength. Indicates the first i The temperature is T The hit rate This is an indicator function.
[0116] In this embodiment, to ensure consistency between the discrimination and drift compensation, the same Δ is used for the mean spectrum at each temperature or for each spectrum. * Alignment (overall translation):
[0117] Temperature-level peak hit ratio p T,j The specific calculation method is as follows: For each temperature T, each peak j:
[0118]
[0119] Hit ratio:
[0120] Furthermore, the determination of the presence and risk level of precipitates based on the hit ratio includes: When the hit rate is at temperature T At or above 100℃, and At least at one of the temperatures, it is greater than or equal to 0.6 and When the value is greater than or equal to 0.6, it is determined that there are precipitates and a high risk; when and At least at one of the temperatures, it is greater than or equal to 0.6, but Less than 0.6, or when and If the temperature is equal to 0.6 at only one of the three temperatures, it is considered that there is a suspected presence of precipitates and a retest is recommended; otherwise, it is considered that there are no precipitates and the risk is low.
[0121] In this embodiment, a "high-temperature segment cluster" is defined (source: thermal precipitation is more stable at higher temperatures):
[0122] The specific judgment rules are as follows: ① Existence (high risk): In Thigh, both peaks in group A satisfy the condition. (Meeting at least one high-temperature point is sufficient, or meeting both is even better) And at least one high-temperature point in group C satisfies:
[0123] ② Suspected: Group A meets the requirements but Group C does not, or Group A only meets the requirements at one temperature point (retesting is recommended). ③ Not detected / Low risk: Group A does not meet the criteria The cross-temperature hit ratio is shown in Table 11; Table 11
[0124] According to the rules: ① At 100℃: Both peaks in Group A are ≥0.6 and Group C is ≥0.6 → This satisfies the condition of "existence (high risk)". ②120℃ also meets the requirement → Stronger evidence Therefore, the output is: "Precipitates present (high risk: additive / ester migration + co-occurrence of oxidation products)". The final output table is shown in Table 12.
[0125] Table 12
[0126] A second embodiment of the present invention provides a system for identifying heat-extracted substances from plastic wrap using infrared spectroscopy, comprising a data acquisition module, a first module, a second module, and a discrimination module, wherein: The acquisition module is designed to repeatedly acquire infrared spectra of containers containing plastic wrap and blank control spectra containing only containers at different temperatures. The function of the first module is to preprocess the infrared spectrum and the blank control spectrum to obtain the threshold and the first spectrum; The function of the second module is to obtain the global threshold and the second spectrum based on the threshold and the first spectrum; The function of the discrimination module is to calculate the hit ratio based on the global threshold and the second spectrum, determine whether there are precipitates and the risk level based on the hit ratio, and complete the discrimination of heat-induced precipitates from the plastic wrap.
[0127] A third embodiment of the present invention provides a computer-readable storage medium, which is a non-volatile storage medium or a non-transient storage medium, and stores a computer program thereon. When the computer program is run by a processor, it executes the steps of any of the above-described methods for determining heat-induced precipitates from plastic wrap using infrared spectroscopy.
[0128] The terms used to describe positional relationships in the accompanying drawings are for illustrative purposes only and should not be construed as limiting this patent. Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art can make other variations or modifications based on the above description. It is neither necessary nor possible to exhaustively describe all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.
Claims
1. A method for identifying heat-extracted substances from plastic wrap using infrared spectroscopy, characterized in that, Includes the following steps: S1: Repeatedly collect infrared spectra of containers containing plastic wrap and blank control spectra containing only containers at different temperatures; S2: Preprocess the infrared spectrum and blank control spectrum to obtain the threshold and the first spectrum; S3: Obtain the global threshold and the second spectrum based on the threshold and the first spectrum; S4: Calculate the hit ratio based on the global threshold and the second spectrum, determine whether there are precipitates and the risk level based on the hit ratio, and complete the identification of heat-induced precipitates from the plastic wrap.
2. The method for determining the heat-extracted substances from plastic wrap using infrared spectroscopy according to claim 1, characterized in that, The preprocessing of the infrared spectrum and the blank control spectrum to obtain the threshold and the first spectrum includes: Linear interpolation with a unified wavenumber grid is applied to the infrared spectrum and the blank control spectrum to obtain a unified infrared spectrum and a unified blank control spectrum. The wavenumber grid is a set of discrete sampling points taken at fixed intervals on a continuous wavenumber axis of the infrared spectrum. The difference between the unified infrared spectrum and the unified blank control spectrum is calculated and normalized to obtain a normalized difference spectrum. The mean difference spectrum is obtained by taking the mean of multiple samples of the normalized difference spectrum. The mean and standard deviation of the normalized difference spectrum within the first range of the wavenumber grid are calculated, and the threshold is obtained based on the mean and standard deviation. Calculate the root mean square of the mean difference spectrum within the second range of the wavenumber grid, and normalize the mean difference spectrum based on the root mean square to obtain the normalized mean difference spectrum. The temperature difference spectrum is obtained by subtracting the normalized mean difference spectrum of adjacent temperatures, and the temperature difference spectrum is weighted and fused to obtain the first spectrum.
3. The method for determining the heat-extracted substances from plastic wrap using infrared spectroscopy according to claim 2, characterized in that, The wavenumber grid w It is the set of discrete sampling points taken at fixed intervals on the continuous wavenumber axis of the infrared spectrum, specifically: In the formula, a Indicates the sampling point.
4. The method for determining the heat-extracted substances from plastic wrap using infrared spectroscopy according to claim 3, characterized in that, The step of obtaining the normalized difference spectrum by subtracting and normalizing the unified infrared spectrum and the unified blank control spectrum includes: The difference spectrum is obtained by subtracting the unified infrared spectrum and the unified blank control spectrum. The normalized difference spectrum is then obtained by using the difference spectrum and the normalization coefficient, as shown in the formula: ， ， ， In the formula, Represented as temperature T The difference spectrum is collected a number of times, r. Indicates temperature as T A unified infrared spectrum collected a number of times, r. Indicates temperature as T Unified blank control spectrum, Expressed as contact strength factor, in units of cm 2 / ml , Represented as contact area, Expressed as the volume of the medium, Indicates temperature as T The normalization coefficient for the number of data collections, r. Indicates the standard contact strength factor. Indicates temperature as T The contact intensity factor is collected a number of times, r. Indicates the standard contact area. Indicates the standard medium volume. Indicates temperature as T The contact area is collected in r samples. Indicates temperature as T The volume of the medium is sampled a number of times, r. Indicates temperature as T The normalized difference spectrum is collected a number of times, r.
5. The method for determining the heat-extracted substances from plastic wrap using infrared spectroscopy according to claim 4, characterized in that, The mean difference spectrum is obtained by averaging multiple samples of the normalized difference spectrum. The mean and standard deviation of the normalized difference spectrum within the first range of the wavenumber grid are calculated, and the threshold is obtained based on the mean and standard deviation, using the following formula: ， ， In the formula, Indicates temperature as T The mean difference spectrum, where R represents the total number of samples. Indicates the first range of the wavenumber grid. This represents the mean. Indicates standard deviation, This represents the threshold.
6. The method for determining the heat-extracted substances from plastic wrap using infrared spectroscopy according to claim 5, characterized in that, The root mean square of the mean difference spectrum within the second range of the wavenumber grid is calculated, and the normalized mean difference spectrum is obtained by normalizing the mean difference spectrum based on the root mean square. The formula is as follows: ， ， In the formula, Indicates the second range of the wavenumber grid. Represents the root mean square. This represents the number of wavenumber grid points in the mean difference spectrum that fall within the second range of the wavenumber grid. Represented as temperature T The normalized mean difference spectrum, It is a very small positive number.
7. The method for determining the heat-extracted substances from plastic wrap using infrared spectroscopy according to claim 6, characterized in that, The temperature difference spectrum is obtained by subtracting the normalized mean difference spectrum of adjacent temperatures, and the first spectrum is obtained by weighted fusion of the temperature difference spectrum, as shown in the formula: ， In the formula, Represents the temperature difference spectrum. , , , , , , Indicates the first i A key peak strength. Indicates the first i Each weight, This indicates the first spectrum.
8. The method for determining the heat-extracted substances from plastic wrap using infrared spectroscopy according to claim 7, characterized in that, The step of obtaining the global threshold and the second spectrum based on the threshold and the first spectrum includes: The threshold that is the largest among all temperatures is taken as the global threshold. The formula is: Then, the second spectrum is obtained based on the global threshold and the first spectrum, specifically: Define drift range , Pick , Indicates the step size; The first spectrum is interpolated and shifted according to any drift value within the drift range to obtain the shifted first spectrum. The key peak intensity is calculated based on the shifted first spectrum. The key peak intensity is compared with the global threshold, and the drift value corresponding to the key peak intensity being greater than the global threshold is selected. The first spectrum is then interpolated and shifted according to the drift value to obtain the second spectrum.
9. The method for determining the heat-extracted substances from plastic wrap using infrared spectroscopy according to claim 8, characterized in that, The formula for calculating the hit ratio based on the global threshold and the second spectrum is as follows: According to the second spectrum Calculate the intensity of all critical peaks, and then calculate the hit ratio based on the critical peak intensity and the global threshold. The formula is as follows: In the formula, Indicates temperature as T The second spectrum collected r times. Indicates temperature as T The number of collections is r, the th i The key peak strength. Indicates the first i The temperature is T The hit rate This is an indicator function.
10. The method for determining the heat-extracted substances from plastic wrap using infrared spectroscopy according to claim 9, characterized in that, The determination of the presence of precipitates and risk level based on the hit ratio includes: When the hit rate is at temperature T At or above 100℃, and At least at one of the temperatures, it is greater than or equal to 0.6 and When the value is greater than or equal to 0.6, it is determined that there are precipitates and a high risk; when and At least at one of the temperatures, it is greater than or equal to 0.6, but Less than 0.6, or when and If the temperature is equal to 0.6 at only one of the three temperatures, it is considered that there is a suspected presence of precipitates and a retest is recommended; otherwise, it is considered that there are no precipitates and the risk is low.