A rapid method for determining the stability of modified asphalt based on near-infrared spectral characteristic parameters

Abstract

This invention discloses a rapid stability determination method for modified asphalt based on near-infrared spectral characteristic parameters, belonging to the field of optical detection technology. The method includes processing the effective reflectance spectrum using a standard normal variable and performing Savitzky-Golay differentiation; integrating the effective reflectance spectrum within an oxidation-sensitive window, a segregation-sensitive window, and a reference window to obtain the three-window integral, and constructing an oxidation index and a segregation index; centering the segregation index and removing components linearly correlated with the oxidation index to obtain a segregation orthogonal index; using the oxidation index for applicable domain gate determination, and comparing the absolute value of the segregation orthogonal index with a stability threshold when the gate is passed, outputting a stable or unstable conclusion. This invention suppresses the aliasing effect of oxidation or aging signals on segregation determination through the segregation orthogonal index, improving the accuracy and consistency of determination; and reduces the risk of misjudgment caused by threshold drift by pre-intercepting and verifying samples exceeding the calibrated oxidation level through applicable domain gate determination.

Description

Technical Field

[0001] This invention relates to the field of optical detection technology, and in particular to a rapid method for determining the stability of modified asphalt based on near-infrared spectral characteristic parameters. Background Technology

[0002] Modified asphalt, by incorporating polymers, rubber powders, or composite modification systems, achieves superior high-temperature rutting resistance and low-temperature crack resistance. However, the stability of the material's phase state and component distribution still needs to be monitored during storage, transportation, and mixing. In engineering and research, segregation evaluation methods after thermal storage are commonly used (such as layered sampling after tubular thermal storage and comparison of differences in softening point, viscosity, or rheological parameters). This is combined with microscopic characterization and conventional physicochemical indicators to analyze the dispersion state, compatibility, and aging evolution of the modified system. Simultaneously, near-infrared spectroscopy, as a carrier of overtone and combination frequency information of molecular vibrations, can be used to obtain the spectral characteristics of materials and establish characteristic parameters related to performance indicators, enabling rapid characterization of the material's state.

[0003] Regarding "rapid discrimination," conventional thermal storage stratification evaluation has two limitations: First, it relies on processes such as thermal storage and stratified sampling, resulting in a long testing cycle and high requirements for sampling consistency and environmental control, making it difficult to meet the needs of rapid screening and batch release in the production process; Second, discrimination based on a single physical indicator or direct spectral features is easily affected by the simultaneous influence of thermo-oxidative aging and oxidation reactions, causing the signals related to segregation to couple with the signals related to oxidation, which in turn causes the feature threshold to drift with thermal history and batch changes, reducing the transferability and comparability of stability discrimination. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a rapid method for determining the stability of modified asphalt based on near-infrared spectral characteristic parameters, which solves the problems of overlapping determination of oxidation or aging effects and segregation effects, as well as the drift of determination threshold with batch and thermal history in the prior art.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: This invention provides a rapid method for determining the stability of modified asphalt based on near-infrared spectral characteristic parameters, which includes heating the modified asphalt to be tested to the detection temperature and stirring it to reconstitute it before placing it into a sample cup, and using a diffuse reflectance near-infrared spectrometer to collect the effective reflectance spectrum that passes through the mass gate. Read the preprocessing parameters and wavenumber window according to the method parameter table, convert the effective reflectance spectrum into absorbance spectrum and perform standard normal variable processing and Savitzky-Golay first-order derivative calculation, and integrate the absolute value of the derivative spectrum in the oxidation sensitive window, the separation sensitive window and the reference window respectively to obtain the three-window integral. The oxidation index and segregation index are constructed using the three-window integral, and the mean and orthogonalization coefficient determined by the calibration set in the method parameter table are called to center the segregation index and remove the components related to the oxidation index to obtain the segregation orthogonal index. The applicable domain boundary and stability threshold in the method parameter table are called. The applicable domain gate is determined by the oxidation index. When the applicable domain gate is passed, the absolute value of the segregation orthogonality index is compared with the stability threshold, and the judgment of stability or instability is output.

[0007] As a preferred embodiment of the rapid stability determination method for modified asphalt based on near-infrared spectral characteristic parameters described in this invention, the steps include: heating the modified asphalt to be tested to the detection temperature and stirring to reconcile it before placing it into a sample cup; and using a diffuse reflectance near-infrared spectrometer to collect the reflectance spectrum passing through the mass gate. The specific steps are as follows: Place the modified asphalt sample to be tested into a clean metal container, put the metal container into a constant temperature heating device to heat it to the test temperature, and keep it stable. After the constant temperature period is over, place the metal container under the same temperature conditions and use a mechanical stirrer to homogenize the sample. Diffuse reflectance was measured using a Fourier transform near-infrared spectrometer to obtain a single reflectance spectrum; The reflectance spectrum is subjected to quality gate judgment to obtain the effective reflectance spectrum.

[0008] As a preferred embodiment of the rapid stability determination method for modified asphalt based on near-infrared spectral characteristic parameters described in this invention, the quality access control determination includes: calculating the average reflectance of the reflectance spectrum within a reference window; if the average reflectance is within the upper and lower limits of the average reflectance in the reference window, then the access control indicator is determined to be passed. Calculate the mean absolute value of the reflectance difference between adjacent wavenumber points in the reflectance spectrum. If the mean absolute value of the reflectance difference between adjacent wavenumber points is not greater than the jitter threshold, then the access control indicator two is deemed to have passed. If both access control criteria one and access control criteria two are met, the current reflectance spectrum quality is determined to pass the access control, and the current reflectance spectrum is saved as a valid reflectance spectrum.

[0009] As a preferred embodiment of the rapid stability determination method for modified asphalt based on near-infrared spectral characteristic parameters described in this invention, the method parameter table includes wavenumber window parameters, preprocessing parameters, calibration parameters required for orthogonalization, applicable domain gate parameters, and stability thresholds. The wavenumber window parameters include the left and right endpoints of the oxidation-sensitive window, the left and right endpoints of the separation-sensitive window, and the left and right endpoints of the reference window; The preprocessing parameters include the number of Savitzky-Golay first-order derivative window points and the order of the Savitzky-Golay fitting polynomial. The calibration parameters required for orthogonalization include the mean oxidation index of the calibration set, the mean segregation index of the calibration set, and the orthogonalization coefficient; The applicable domain access control parameters include the lower boundary of the applicable domain and the upper boundary of the applicable domain.

[0010] As a preferred embodiment of the rapid stability determination method for modified asphalt based on near-infrared spectral characteristic parameters described in this invention, the specific steps of converting the effective reflectance spectrum into an absorbance spectrum and performing standard normal variable processing and Savitzky-Golay first-order differentiation are as follows: For each effective reflectance spectrum, the same conversion is performed at each wavenumber point to obtain the absorbance spectrum; For each absorbance spectrum, the mean and standard deviation are calculated over the entire wavenumber range, and the absorbance value at each wavenumber point is scaled to obtain the normalized absorbance spectrum. Based on the normalized absorbance spectrum, using the preprocessing parameters given in the method parameter table, a local window is taken near each wavenumber point. Polynomial fitting is performed on the data within the local window and the derivative is directly calculated to obtain the first derivative value at the wavenumber point.

[0011] As a preferred embodiment of the rapid stability determination method for modified asphalt based on near-infrared spectral characteristic parameters described in this invention, the method involves integrating the absolute value of the derivative spectrum within the oxidation-sensitive window, the segregation-sensitive window, and the reference window to obtain the three-window integral. The specific steps are as follows: Take the absolute value of the first derivative at each wavenumber point to obtain the absolute derivative spectrum; Based on the absolute value derivative spectrum, integrals are accumulated in the wavenumber window respectively, and the arithmetic mean of the integrals accumulated in the wavenumber window is obtained to obtain the three-window integral. The wavenumber window includes an oxidation-sensitive window, a separation-sensitive window, and a reference window; The three-window integral includes the oxidation-sensitive window integral, the separation-sensitive window integral, and the reference window integral; The integral accumulation follows a trapezoidal accumulation method.

[0012] As a preferred embodiment of the rapid determination method for the stability of modified asphalt based on near-infrared spectral characteristic parameters described in this invention, wherein: the construction of oxidation index and segregation index using three window integral quantities respectively means that the reference window integral quantity is used as a normalization benchmark, and the oxidation sensitive window integral quantity and the segregation sensitive integral quantity are normalized respectively to construct oxidation index and segregation index.

[0013] As a preferred embodiment of the rapid stability determination method for modified asphalt based on near-infrared spectral characteristic parameters described in this invention, the specific steps for centering the segregation index and removing components related to the oxidation index to obtain the segregation orthogonal index are as follows: The centralized oxidation index and centralized separation index are obtained by subtracting the mean values ​​of the calibration sets for the oxidation index and the separation index, respectively. The product of the central oxidation index and the orthogonalization coefficient is used as the relevant component deducted from the segregation side; The orthogonal exponent of segregation is obtained by subtracting the central segregation index from the relevant components deducted from the segregation side.

[0014] As a preferred embodiment of the rapid stability determination method for modified asphalt based on near-infrared spectral characteristic parameters described in this invention, the method involves: using the oxidation index to determine the applicable domain threshold; when the applicable domain threshold is passed, comparing the absolute value of the segregation orthogonal index with the stability threshold, and outputting a judgment of stability or instability. The specific steps are as follows: The oxidation index is used as the value for access control in the applicable domain. Access control rules for the applicable domain are defined, and the stability determination is determined based on the access control rules for the applicable domain. Using the absolute value of the segregation orthogonality index as the discriminant, a stability determination rule is defined. Based on the stability determination rule, the sample to be tested is determined to be stable or unstable.

[0015] As a preferred embodiment of the rapid stability determination method for modified asphalt based on near-infrared spectral characteristic parameters described in this invention, the applicable domain access control rule includes: if the oxidation index of the sample to be tested is within the upper and lower bounds of the applicable domain, then the sample to be tested is determined to meet the applicable domain access control and can be used for stability determination. If the oxidation index of the sample to be tested is not within the upper or lower bound of the applicable range, the sample to be tested is determined to be outside the applicable range and no stability determination is performed. The stability determination rule includes determining that if the discrimination value is not greater than the stability threshold, the sample to be tested is stable. If the discriminant is greater than the stability threshold, the sample to be tested is determined to be unstable.

[0016] The beneficial effects of this invention are as follows: by centering the segregation index and removing components linearly related to the oxidation index to obtain the segregation orthogonal index, the aliasing effect of oxidation or aging signals on stability discrimination is reduced, thereby improving the accuracy and consistency of stability discrimination conclusions; by introducing an applicable domain gate with the oxidation index as input before threshold discrimination and using upper and lower boundaries for interval judgment, samples exceeding the calibrated oxidation level range are pre-intercepted and verified, reducing the risk of misjudgment caused by threshold drift. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a rapid method for determining the stability of modified asphalt based on near-infrared spectral characteristic parameters.

[0019] Figure 2 A flowchart for obtaining the effective reflectance spectrum.

[0020] Figure 3 This is a flowchart for quality access control judgment.

[0021] Figure 4 This is a flowchart for determining the stability of the sample to be tested.

[0022] Figure 5 This is a statistical characteristic graph of the segregation index under different segregation level parameters.

[0023] Figure 6 This represents the statistical characteristics of the orthogonality index under different segregation level parameters. Detailed Implementation

[0024] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0025] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0026] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0027] Reference Figures 1-6 This is one embodiment of the present invention, which provides a rapid method for determining the stability of modified asphalt based on near-infrared spectral characteristic parameters, including the following steps: S1. The modified asphalt to be tested is heated to the test temperature and stirred until homogeneous, then placed into a sample cup. The effective reflectance spectrum passing through the quality gate is collected using a diffuse reflectance near-infrared spectrometer.

[0028] Furthermore, the modified asphalt sample to be tested is placed in a clean metal container, and the metal container is placed in a constant temperature heating device to heat the sample to the test temperature, such as 160°C, and kept stable. The constant temperature holding time is usually set to 30 minutes to ensure that the internal temperature of the sample is uniform and reaches a stable flow state.

[0029] After the constant temperature is achieved, the metal container is placed under the same temperature conditions, and the sample is homogenized using a mechanical stirrer. After stirring, the sample is allowed to stand to release the air bubbles introduced by stirring and stabilize the surface fluctuations.

[0030] The hot modified asphalt was poured into a heat-resistant sample cup and then placed on a constant temperature stage to maintain the sample at a constant temperature throughout the entire spectral acquisition process.

[0031] Diffuse reflectance was measured using a Fourier transform near-infrared spectrometer to obtain a single reflectance spectrum.

[0032] It should be noted that before sample measurement, a PTFE standard diffuse reflectance plate was used to collect the background spectrum, and the background was updated at a fixed frequency; for example, the background was re-collected after every 5 consecutive samples to reduce the impact of light source drift on the consistency of reflectance data.

[0033] Furthermore, the same sample was repeatedly sampled until at least three effective reflectance spectra that passed the quality gate were obtained.

[0034] Specifically, the calculation and determination of the quality access control involves reading the wavenumber interval of the reference window from the method parameter table and defining the wavenumber point index set corresponding to the reference window; calculating the mean reflectance of the reflectance spectrum within the reference window. ;like If the access control indicator is met, the access control is deemed to have passed; otherwise, it is deemed to have failed. The mean of the absolute values ​​of the reflectance differences between adjacent wavenumber points is calculated for the reflectance spectrum. ;like If both access control indicators are met, the access control system is considered to have passed; otherwise, it is considered to have failed. If both access control indicators are met, the access control system is considered to have passed and the value is 1; otherwise, it is considered to have failed and the value is 0.

[0035] in, This represents the lower limit threshold of the mean reflectance of the reference window. This represents the upper limit threshold of the average reflectance of the reference window. The lower limit threshold and the upper limit threshold of the average reflectance of the reference window are determined by statistical analysis of the reflectance data of standard background samples and stable samples in long-term operation, respectively, to determine the normal reflectance range, and are set according to the energy response stability of the instrument.

[0036] The jitter threshold is represented by the upper bound of the normal fluctuation of the jitter mean at adjacent points, which is obtained by repeatedly collecting data from standard background samples and stable retained samples under fixed geometric and constant temperature conditions, and then fixing the upper bound of the normal fluctuation.

[0037] It should be noted that the fixed lower limit threshold of the mean reflectance of the reference window, the upper limit threshold of the mean reflectance of the reference window, and the jitter threshold should be written into the method parameter table for use in calling.

[0038] When both access control criteria 1 and 2 are met, the quality access control passes, the quality access control is set to 1, and the reflectance spectrum is saved as an effective reflectance spectrum. When either access control criterion is not met, the quality access control fails, the quality access control is set to 0, the current reflectance spectrum is discarded, and a new sample is collected until at least three effective reflectance spectra are obtained, at which point the process ends.

[0039] S2. Read the preprocessing parameters and wavenumber window according to the method parameter table, convert the effective reflectance spectrum into absorbance spectrum and perform standard normal variable processing and Savitzky-Golay first-order differentiation. Integrate the absolute value of the derivative spectrum in the oxidation sensitive window, the separation sensitive window and the reference window respectively to obtain the three-window integral.

[0040] Furthermore, the endpoint wavenumbers of the curing oxidation-sensitive window, the endpoint wavenumbers of the segregation-sensitive window, the endpoint wavenumbers of the reference window, the number of window points required for Savitzky-Golay differentiation, and the order of the fitted polynomial are called from the method parameter table.

[0041] Among them, the endpoint wavenumber of the oxidation sensitive window is used to determine the endpoint wavenumber of the oxidation sensitive window; the endpoint wavenumber of the separation sensitive window is used to determine the endpoint wavenumber of the separation sensitive window; the endpoint wavenumber of the reference window is used to determine the reference window; the number of window points required for Savitzky-Golay differentiation and the order of the fitting polynomial are used to perform first-order differentiation on the effective reflectance spectrum.

[0042] It should be noted that, to obtain the endpoints of the oxidation-sensitive window, specifically, using the oxidation control set as a sample, the correlation intensity between the spectral value and the oxidation reference amount at each wavenumber point is calculated through correlation operations to obtain the correlation intensity of each wavenumber point with oxidation; the wavenumber point with the largest absolute value of the oxidation correlation intensity is taken as the oxidation response center, and an oxidation attenuation ratio coefficient is set, and the oxidation-sensitive window is obtained according to the following A1-A3 method: A1. Move from the oxidation response center towards lower wavenumbers to find the condition that satisfies... The first wavenumber point is taken as the left endpoint of the oxidation sensitivity window.

[0043] in, Indicates at the wavenumber point The correlation intensity of the spectrum at that location with oxidation This represents the oxidation attenuation ratio coefficient. Indicates the central wavenumber of the oxidation response. Oxidation-related strength at the location.

[0044] A2. Moving from the center of the oxidation response towards higher wavenumbers, find the first wavenumber point that satisfies the same attenuation condition as the right endpoint of the oxidation-sensitive window, and finally obtain the oxidation-sensitive window. .

[0045] in, Indicates the left endpoint of the oxidation-sensitive window. This represents the right endpoint of the oxidation-sensitive window.

[0046] A3. Perform correlation calculations on the separation control set as well, obtain the correlation strength of each wavenumber point with separation, and require that the separation correlation strength within the oxidation sensitivity window does not exceed the upper limit of oxidation anti-aliasing. , represented as: ; in, Indicates at the wavenumber point The correlation intensity of the spectrum at the point of separation.

[0047] If the conditions are not met, the oxidation-sensitive window will be narrowed according to the narrowing window rule until the upper limit position of anti-aliasing is met.

[0048] The narrowing window rule involves moving from both ends of the oxidation-sensitive window towards the center simultaneously until the anti-aliasing limit is met.

[0049] The upper limit of anti-aliasing oxidation is obtained by calculating the maximum distribution of the segregation correlation intensity at each wavenumber point within the oxidation sensitivity window based on the segregation control set, and then fixing the statistical upper bound of the maximum distribution, such as the 95th percentile of the above quantile.

[0050] The oxidation attenuation ratio coefficient is used to calculate the oxidation correlation intensity at each wavenumber point within the oxidation sensitive window. The peak shape of the oxidation correlation intensity is obtained, and the minimum attenuation ratio value is obtained so that the oxidation sensitive window just covers the effective width of the main peak and satisfies that the discrete correlation intensity within the window does not exceed the upper limit of anti-aliasing.

[0051] It should be noted that, to obtain the endpoints of the segregation sensitivity window, specifically, using the segregation control set as a sample, the correlation intensity between the spectral value and the segregation reference at each wavenumber point is calculated through correlation calculation to obtain the correlation intensity of each wavenumber point with segregation; the wavenumber point with the largest absolute value of the segregation correlation intensity is taken as the segregation response center, and a segregation attenuation ratio coefficient is set. The segregation sensitivity window is then obtained following the same steps as A1-A3. ;in, This represents the left endpoint of the separation-sensitive window. This represents the right endpoint of the separation sensitive window.

[0052] The method for obtaining the segregation attenuation ratio coefficient and the segregation anti-aliasing upper limit in the segregation sensitivity window is the same as that for obtaining the oxidation attenuation ratio coefficient and the oxidation anti-aliasing upper limit.

[0053] It should be noted that the acquisition of the reference window endpoints specifically involves setting the candidate window width across the entire spectrum. For each candidate window Calculate the sensitivity to oxidation and segregation, and calculate the in-window stability; wherein, the candidate window width satisfies the following conditions: .

[0054] The oxidation and separation sensitivities were determined by taking the maximum oxidation correlation intensity and the maximum separation correlation intensity within the window as the oxidation sensitivity and separation sensitivity, respectively.

[0055] In-window stability is determined by averaging the variances of the in-window spectral values ​​over the entire sample.

[0056] The oxidation sensitivity, separation sensitivity, and in-window stability are weighted and summed according to preset weights to form the objective function value of the candidate window. The candidate window with the smallest objective function value is selected as the reference window.

[0057] It should be noted that the preset weights are determined by cross-validating a set of candidate weight combinations. The optimization criteria are to minimize the sensitivity of the reference window to oxidation and segregation and minimize the variance of the signal within the window. The combination of weights that minimizes the objective function and has the best stability is selected as the preset weights. Generally, the value range of each preset weight is [0,1], and the sum of each preset weight is 1.

[0058] It should be noted that the number of window points required for the Savitzky-Golay derivative and the order of the fitted polynomial are obtained by specifically calculating the oxidation difference spectrum and the separation difference spectrum respectively using the absolute value of the difference between the mean absorbance spectra of the two extreme samples of oxidation and separation, and obtaining the difference intensity spectrum. , represented as: ; in, This represents the average absorbance spectrum of the high-variance group. This represents the average absorbance spectrum of the low-variance group.

[0059] It should be noted that, and The mean absorbance spectra of the two groups of samples with the greatest difference are calculated separately for oxidation and separation, and the narrower one is taken.

[0060] Find the main peak in the difference intensity spectrum and calculate the full width at half maximum (FWHM) of the main peak. Take the smallest FWHM as the narrowest effective characteristic bandwidth.

[0061] Based on the narrowest effective feature bandwidth, the window number is obtained by taking the smallest odd number of window points covering the current bandwidth, and is expressed as: ; in, This represents the number of window points used in the Savitzky-Golay derivative. Indicates the narrowest effective feature bandwidth. The representative value representing the wavenumber sampling interval.

[0062] It should be noted that the representative value of the wavenumber sampling interval refers to the interval between adjacent sampling wavenumber points in the instrument's output spectrum, and is usually expressed as the median of the difference between adjacent wavenumbers.

[0063] Determining the order of a polynomial based on the number of window points in Savitzky-Golay differentiation. Specifically, take the condition that satisfies and The smallest integer is used as the order of the polynomial.

[0064] Write the endpoints of the oxidation sensitive window, the segregation sensitive window, the wavenumbers of the reference window, the number of window points required for Savitzky-Golay differentiation, and the order of the fitted polynomial into the method parameter table. Also write the oxidation attenuation ratio coefficient, the segregation attenuation ratio coefficient, the upper limit of oxidation anti-aliasing, the upper limit of segregation anti-aliasing, the differential intensity spectrum, and the preset weights of the objective function into the method parameter table.

[0065] Furthermore, near-infrared diffuse reflectance measurement directly outputs a reflectance spectrum to improve the stability of subsequent processing.

[0066] The reflectance spectrum is converted into an absorbance spectrum. Specifically, for each effective reflectance spectrum, the same conversion is performed at each wavenumber point, that is, the reflectance is taken as the base 10 logarithm and the sign is negative to obtain the absorbance value.

[0067] Furthermore, modified asphalt is a medium with strong absorption and strong scattering. Minor disturbances on the sample surface and slight changes in probe distance can cause differences in the overall amplitude of the entire spectrum. In order to reduce the impact of the overall scale difference on the subsequent integration,

[0068] For each absorbance spectrum, standard normal variable processing is performed. Specifically, the mean and standard deviation of each absorbance spectrum over the entire wavenumber range are calculated. Then, the absorbance value at each wavenumber point is subtracted from the mean of the entire spectrum and divided by the standard deviation of the entire spectrum to obtain an absorbance spectrum with a mean of 0 and scale normalized.

[0069] Furthermore, even after standard normal variable processing, the absorbance spectrum may still exhibit slow baseline drift or low-frequency distortion. In order to further suppress low-frequency background and highlight local spectral changes, further measures should be taken.

[0070] Based on the standard normal variable processing, the Savitzky-Golay first derivative is performed. Specifically, with the number of window points and the order of the polynomial given in the method parameter table as fixed settings, a local window is taken near each wavenumber point, and the data within the local window is polynomial fitted and directly differentiated to obtain the first derivative value at the wavenumber point.

[0071] Furthermore, after completing the differentiation, the absolute value of the first derivative at each wavenumber point is taken to obtain the absolute value derivative spectrum; based on the absolute value derivative spectrum, the integrals are then accumulated in the three wavenumber windows respectively, and the arithmetic mean of the accumulated integrals in the three wavenumber windows is obtained to obtain the three-window integral; where the three wavenumber windows refer to the oxidation sensitive window, the separation sensitive window, and the reference window.

[0072] The integral accumulation follows a trapezoidal rule, expressed as: ; in, Indicates the first Absorbance spectra in wavenumber window The integral within, This indicates the number of wavenumber points within the window. and Wavenumber window The two adjacent wavenumber points of the intrinsic component in the trapezoidal integral, Indicates the first The absolute derivative spectrum of the absorbance spectrum at the wavenumber point The value at that location, Indicates the first The absolute derivative spectrum of the absorbance spectrum at the wavenumber point The value at that location.

[0073] Among them, the three-window integral refers to the integral of the oxidation-sensitive window, the integral of the separation-sensitive window, and the integral of the reference window.

[0074] S3. Construct the oxidation index and segregation index using the three-window integral, and call the mean and orthogonalization coefficient determined by the calibration set in the method parameter table to center the segregation index and remove the components related to the oxidation index to obtain the segregation orthogonal index.

[0075] Furthermore, using the integral of the reference window as a normalization benchmark, the integrals of the oxidation-sensitive window and the separation-sensitive window are normalized respectively to construct the oxidation index and the separation index.

[0076] Furthermore, since the segregation index and oxidation index will vary with the batch of asphalt raw materials and differences in instruments, in order to ensure comparability between different times and different batches,

[0077] Before orthogonalization, the oxidation index and the segregation index are centered. Specifically, the centered oxidation index and the centered segregation index are obtained by subtracting the mean of the calibration set from the oxidation index and the segregation index, respectively.

[0078] It should be noted that the calibration set mean in the centralized processing is accessed through the method parameter table. Specifically, the calibration set mean is obtained by selecting modified asphalt samples within the same application range to form a calibration set sample sequence, obtaining the three-window integral of the calibration samples, and constructing the oxidation index and segregation index of the calibration samples based on the three-window integral. Subsequently, the oxidation index and segregation index of all calibration samples are arithmetically averaged to obtain the calibration set mean. The calibration set mean includes the oxidation index calibration set mean and the segregation index calibration set mean. The selected modified asphalt samples should cover different batches of base asphalt and modifier sources, sample states with different thermal histories or oxidation levels, and duplicate samples under the same process conditions. Finally, the calibration set mean is written into the method parameter table.

[0079] Furthermore, in the actual thermal process of modified asphalt, aging and oxidation may increase the oxidation index, and the segregation index may also show a corresponding change. If the segregation index is used directly for stability judgment, the common changes caused by oxidation may be misjudged as segregation enhancement. In order to reduce the overlapping phenomenon of segregation enhancement.

[0080] The centralized segregation index undergoes decorrelation processing, specifically, the component linearly correlated with the centralized oxidation index is subtracted from the centralized segregation index. This is achieved by multiplying the centralized oxidation index and the orthogonality coefficient to obtain the correlated component to be subtracted from the segregation side. This component is then subtracted from the centralized segregation index to obtain the segregation orthogonality index, expressed as: ; in, Indicates the orthogonality index of the separation. Indicates the centralization segregation index. Denotes the orthogonalization coefficients. This represents the central oxidation index.

[0081] It should be noted that the orthogonalization coefficients are accessed through the method parameter table. Specifically, the orthogonalization coefficients are obtained by first calculating the oxidation index and segregation index of each calibration sample in the calibration set according to the same calibrator as the detection operation, then calculating the mean of the calibration set, and centering the oxidation index and segregation index. Subsequently, the slope of "univariate linear fitting of the centered oxidation index to the centered segregation index" in the least squares sense is used as the orthogonalization coefficient, and the orthogonalization coefficient is written into the method parameter table.

[0082] In this embodiment, to verify the effect of the segregation orthogonality index on aliasing, the detection environment was abstracted according to common field conditions. Modified asphalt samples were heated and stirred at the detection temperature and then placed into sample cups. The effective reflectance spectrum was collected using diffuse reflectance near-infrared spectroscopy, and the conversion and preprocessing of reflectance to absorbance were completed according to the agreed method. Then, the absolute value of the derivative spectrum was integrated in the oxidation sensitive window, segregation sensitive window, and reference window to obtain the three-window integral quantity, and the oxidation index and segregation index were constructed. On this basis, the segregation index was centered using the calibration set statistics and the components linearly related to the oxidation index were removed to obtain the segregation orthogonality index. In terms of materials, the test object was abstracted as "modified asphalt samples with different segregation level parameters S", where S=-1, 0, and 1 represent three sample sets with segregation degree from low to high, respectively. At the same time, the variation of oxidation or aging level was introduced to reflect the differences and uncertainties of field samples.

[0083] Figure 5 The figure shows the statistical characteristics of the segregation index changing with the oxidation index under different segregation level parameters within the operating conditions. The horizontal axis represents the center of the bins after the oxidation index is binned with a fixed width (step size 0.5), and the vertical axis represents the segregation index. Each broken line corresponds to three sample sets with segregation level parameters S=-1, S=0, and S=1, respectively. The broken line points in the figure represent the statistical center of the segregation index D in each bin (using robust statistics), and the vertical error bars reflect the dispersion of D within each bin. The background scatter plot represents single sample data sampled from each bin, used to visually represent the field fluctuations. It can be seen that the three curves show a significant upward trend with the oxidation index, that is, the segregation index D is not only affected by the segregation level, but also undergoes systematic drift with changes in the oxidation level, which is manifested as the segregation discriminant and the oxidation signal overlapping. This phenomenon means that under the same segregation level, the segregation index may shift due to differences in oxidation or aging levels, thereby reducing the consistency of the stability discrimination conclusion and increasing the risk of misjudgment near the threshold.

[0084] Figure 6 In order to be in Figure 5Under the same operating conditions within the domain, the statistical characteristics of the segregation orthogonality index changing with the oxidation index under different segregation level parameters S; the horizontal axis represents the center of the bins after the oxidation index is divided into bins with a fixed width, and the vertical axis represents the segregation orthogonality index; the three broken lines correspond to the three sample sets with S=-1, S=0, and S=1, respectively. Figure 5 In comparison, the three curves are generally distributed horizontally within the oxidation index binning range, and the drift with changes in oxidation index is significantly reduced. Moreover, the error bars converge relatively at most binning points, indicating that the sensitivity of the segregation orthogonal index to changes in oxidation level is significantly reduced. This result corresponds to the technical feature of this invention, which is to "center the segregation index and remove the linearly correlated component with the oxidation index to obtain the segregation orthogonal index". By eliminating the linear component in the segregation index that changes in the same direction as the oxidation index, the segregation discrimination value is more focused on the segregation itself rather than oxidation or aging perturbations, thereby reducing the aliasing effect, improving the consistency of stability discrimination conclusions among samples with different oxidation levels, and providing a more stable input index for subsequent threshold discrimination.

[0085] S4. Call the applicable domain boundary and stability threshold in the method parameter table, and use the oxidation index to determine the applicable domain gate. When the applicable domain gate is passed, compare the absolute value of the segregation orthogonality index with the stability threshold, and output the judgment of stability or instability.

[0086] Furthermore, the access control rules for the applicable domain are defined. Specifically, when the oxidation index of the sample to be tested is greater than or equal to the lower boundary of the applicable domain and less than or equal to the upper boundary of the applicable domain, the access control flag of the applicable domain is set to 1; otherwise, the access control flag is set to 0.

[0087] If the applicable domain access control flag is 0, the access control result of "out of applicable range" will be output, and the oxidation index of the sample to be tested and the upper and lower boundaries of the applicable domain will be recorded as traceability information.

[0088] If the applicable domain access control flag is 1, then a stability check is performed.

[0089] Among them, stability determination uses the amplitude of the orthogonality index as the discriminant. The amplitude refers to the absolute value of the orthogonality index, which is used to characterize the intensity of the segregation correlation change without distinguishing the direction.

[0090] That is, the discriminant is defined as the absolute value of the orthogonality index.

[0091] The discriminant is compared with the stability threshold. The stability determination rule is as follows: when the discriminant is less than or equal to the stability threshold, the stability determination flag is set to 1; otherwise, it is set to 0.

[0092] If the stability determination flag is set to 1, the output of the test sample is "stable".

[0093] If the stability determination flag is 0, the output of the test sample is "unstable".

[0094] The final output includes the oxidation index, the upper and lower boundaries of the applicable domain, the segregation orthogonality index, the discriminant, and the stability threshold.

[0095] It should be noted that the upper and lower boundaries of the applicable domain and the stability threshold are accessed through the method parameter table. Specifically, to obtain the upper and lower boundaries of the applicable domain, a calibration sample set consistent with the application system of the sample to be tested is selected. The oxidation index of each calibration sample is calculated using the same spectral acquisition and preprocessing procedures, and the mean and dispersion of the oxidation index of the calibration set are obtained. Then, using the mean of the calibration set as the center, the boundary is expanded downwards and upwards by a preset multiple, such as 3 times the standard deviation, to form the lower and upper boundaries of the applicable domain. These boundaries are then written into the method parameter table as the applicable domain boundaries. The domain range is as follows: Specifically, the stability threshold is obtained by selecting several batches of modified asphalt samples that have been confirmed to be stable by existing standard tests as a stable reference set. Following the same spectral acquisition, preprocessing, and orthogonalization calculation process as the test sample, the segregation orthogonal index is obtained batch by batch, and the absolute value of the segregation orthogonal index is taken as the discriminant. Then, the mean and dispersion of the discriminant in the stable reference set are calculated. Subsequently, the stability threshold is formed by expanding upward by a preset multiple, such as 3 times the standard deviation, with the mean of the stable reference set as the center. The stability threshold is then written into the method parameter table.

[0096] In summary, this invention achieves a reduction in the aliasing effect of oxidation or aging signals on stability discrimination by centering the segregation index and removing components linearly correlated with the oxidation index to obtain the segregation orthogonal index, thereby improving the accuracy and consistency of stability discrimination conclusions. By introducing an applicable domain gate with the oxidation index as input before threshold discrimination and using upper and lower boundaries for interval judgment, this invention enables pre-interception and verification prompts for samples exceeding the calibrated oxidation level range, reducing the risk of misjudgment caused by threshold drift.

[0097] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A rapid method for determining the stability of modified asphalt based on near-infrared spectral characteristic parameters, characterized in that: include, The modified asphalt to be tested was heated to the test temperature and stirred until homogeneous before being placed into a sample cup. The effective reflectance spectrum passing through the mass gate was collected using a diffuse reflectance near-infrared spectrometer. Read the preprocessing parameters and wavenumber window according to the method parameter table, convert the effective reflectance spectrum into absorbance spectrum and perform standard normal variable processing and Savitzky-Golay first-order derivative calculation, and integrate the absolute value of the derivative spectrum in the oxidation sensitive window, the separation sensitive window and the reference window respectively to obtain the three-window integral. The oxidation index and segregation index are constructed using the three-window integral, and the mean and orthogonalization coefficient determined by the calibration set in the method parameter table are called to center the segregation index and remove the components related to the oxidation index to obtain the segregation orthogonal index. The applicable domain boundary and stability threshold in the method parameter table are called. The applicable domain gate is determined by the oxidation index. When the applicable domain gate is passed, the absolute value of the segregation orthogonality index is compared with the stability threshold, and the judgment of stability or instability is output.

2. The rapid stability determination method for modified asphalt based on near-infrared spectral characteristic parameters as described in claim 1, characterized in that: The modified asphalt to be tested is heated to the detection temperature and stirred until homogeneous, then placed into a sample cup. The reflectance spectrum passing through the mass gate is collected using a diffuse reflectance near-infrared spectrometer. The specific steps are as follows: Place the modified asphalt sample to be tested into a clean metal container, put the metal container into a constant temperature heating device to heat it to the test temperature, and keep it stable. After the constant temperature period is over, place the metal container under the same temperature conditions and use a mechanical stirrer to homogenize the sample. Diffuse reflectance was measured using a Fourier transform near-infrared spectrometer to obtain a single reflectance spectrum; The reflectance spectrum is subjected to quality gate judgment to obtain the effective reflectance spectrum.

3. The rapid stability determination method for modified asphalt based on near-infrared spectral characteristic parameters as described in claim 2, characterized in that: The quality access control judgment includes calculating the average reflectance of the reflectance spectrum within a reference window. If the average reflectance is within the upper and lower limits of the average reflectance in the reference window, the access control indicator is determined to be passed. Calculate the mean absolute value of the reflectance difference between adjacent wavenumber points in the reflectance spectrum. If the mean absolute value of the reflectance difference between adjacent wavenumber points is not greater than the jitter threshold, then the access control indicator two is deemed to have passed. If both access control criteria one and access control criteria two are met, the current reflectance spectrum quality is determined to pass the access control, and the current reflectance spectrum is saved as a valid reflectance spectrum.

4. The rapid stability determination method for modified asphalt based on near-infrared spectral characteristic parameters as described in claim 3, characterized in that: The method parameter table includes wavenumber window parameters, preprocessing parameters, calibration parameters required for orthogonalization, applicable domain access control parameters, and stability threshold; The wavenumber window parameters include the left and right endpoints of the oxidation-sensitive window, the left and right endpoints of the separation-sensitive window, and the left and right endpoints of the reference window; The preprocessing parameters include the number of Savitzky-Golay first-order derivative window points and the order of the Savitzky-Golay fitting polynomial. The calibration parameters required for orthogonalization include the mean oxidation index of the calibration set, the mean segregation index of the calibration set, and the orthogonalization coefficient; The applicable domain access control parameters include the lower boundary of the applicable domain and the upper boundary of the applicable domain.

5. The rapid stability determination method for modified asphalt based on near-infrared spectral characteristic parameters as described in claim 4, characterized in that: The specific steps for converting the effective reflectance spectrum into an absorbance spectrum and performing standard normal variable processing and Savitzky-Golay first-order differentiation are as follows: For each effective reflectance spectrum, the same conversion is performed at each wavenumber point to obtain the absorbance spectrum; For each absorbance spectrum, the mean and standard deviation are calculated over the entire wavenumber range, and the absorbance value at each wavenumber point is scaled to obtain the normalized absorbance spectrum. Based on the normalized absorbance spectrum, using the preprocessing parameters given in the method parameter table, a local window is taken near each wavenumber point. Polynomial fitting is performed on the data within the local window and the derivative is directly calculated to obtain the first derivative value at the wavenumber point.

6. The rapid stability determination method for modified asphalt based on near-infrared spectral characteristic parameters as described in claim 5, characterized in that: The steps involve integrating the absolute value of the derivative spectrum within the oxidation-sensitive window, the separation-sensitive window, and the reference window respectively to obtain the three-window integral. Take the absolute value of the first derivative at each wavenumber point to obtain the absolute derivative spectrum; Based on the absolute value derivative spectrum, integrals are accumulated in the wavenumber window respectively, and the arithmetic mean of the integrals accumulated in the wavenumber window is obtained to obtain the three-window integral. The wavenumber window includes an oxidation-sensitive window, a separation-sensitive window, and a reference window; The three-window integral includes the oxidation-sensitive window integral, the separation-sensitive window integral, and the reference window integral; The integral accumulation follows a trapezoidal accumulation method.

7. The rapid stability determination method for modified asphalt based on near-infrared spectral characteristic parameters as described in claim 6, characterized in that: The construction of oxidation index and separation index using the three-window integral means that the reference window integral is used as a normalization benchmark, and the oxidation sensitive window integral and the separation sensitive integral are normalized respectively to construct oxidation index and separation index.

8. The rapid stability determination method for modified asphalt based on near-infrared spectral characteristic parameters as described in claim 7, characterized in that: The specific steps for centering the segregation index and removing components related to the oxidation index to obtain the segregation orthogonal index are as follows: The centralized oxidation index and centralized separation index are obtained by subtracting the mean values ​​of the calibration sets for the oxidation index and the separation index, respectively. The product of the central oxidation index and the orthogonalization coefficient is used as the relevant component deducted from the segregation side; The orthogonal exponent of segregation is obtained by subtracting the central segregation index from the relevant components deducted from the segregation side.

9. The rapid stability determination method for modified asphalt based on near-infrared spectral characteristic parameters as described in claim 8, characterized in that: The method of determining the applicable domain access control based on the oxidation index involves comparing the absolute value of the orthogonal index with a stability threshold when the access control passes, and outputting a judgment of stability or instability. The specific steps are as follows: The oxidation index is used as the value for access control in the applicable domain. Access control rules for the applicable domain are defined, and the stability determination is determined based on the access control rules for the applicable domain. Using the absolute value of the segregation orthogonality index as the discriminant, a stability determination rule is defined. Based on the stability determination rule, the sample to be tested is determined to be stable or unstable.

10. The rapid stability determination method for modified asphalt based on near-infrared spectral characteristic parameters as described in claim 9, characterized in that: The applicable domain access control determination rule includes that if the oxidation index of the sample to be tested is within the upper and lower bounds of the applicable domain, the sample to be tested is determined to meet the applicable domain access control and can be used for stability determination. If the oxidation index of the sample to be tested is not within the upper or lower bound of the applicable range, the sample to be tested is determined to be outside the applicable range and no stability determination is performed. The stability determination rule includes determining that if the discrimination value is not greater than the stability threshold, the sample to be tested is stable. If the discriminant is greater than the stability threshold, the sample to be tested is determined to be unstable.

Images