A rapeseed oil quality detection method and device and a storage medium

Abstract

The present application relates to the technical field of physical property analysis, and particularly relates to a rapeseed oil quality detection method and device and a storage medium, which comprises the following steps: obtaining the concentration of oleic acid and the concentration of erucic acid of a rapeseed oil sample; determining the Raman spectrum data of the rapeseed oil sample based on the concentration of oleic acid and the concentration of erucic acid; determining a peak intensity contribution coefficient based on the Raman spectrum data; constructing a rapeseed oil quality detection model based on the concentration of oleic acid, the concentration of erucic acid and the peak intensity contribution coefficient; and performing quality detection on the Raman peak characteristics of a rapeseed oil to be predicted based on the rapeseed oil quality detection model to obtain a detection result of the rapeseed oil to be predicted. The method improves the prediction accuracy of the quality of rapeseed oil.

Description

Technical Field

[0001] This invention relates to the field of physical property analysis technology, specifically to a method, apparatus, and storage medium for detecting the quality of rapeseed oil. Background Technology

[0002] Currently, most research focuses on feature engineering, exploring and innovating to obtain new features for modeling based on the extraction, screening, fusion, and optimization of raw Raman spectra, thereby achieving better detection accuracy. The industry primarily uses partial least squares regression (PLS) modeling algorithms from spectrum to concentration. This data-driven approach suffers from poor interpretability and fails to fully utilize the clear chemical mechanisms and unique fingerprint properties of Raman spectroscopy. When the model is used to detect the content of target fatty acids, the interference of unknown fatty acids is difficult to handle, hindering model transfer to the quantitative detection of specific fatty acid components and resulting in poor model scalability. Due to limitations such as data-driven approaches and insufficient model fitting capabilities, machine learning models like PLS perform poorly in detecting components with large differences in concentration dimensions, such as high oleic acid (50%~70%) and low erucic acid (0%~3%). Modeling methods using deep learning models such as Artificial Neural Networks (ANN) show significant effectiveness in predicting low-concentration fatty acids, but they rely on large-scale, high-quality data, increasing the cost of sample collection and model training. Summary of the Invention

[0003] The purpose of this invention is to provide a method, apparatus, and storage medium for detecting the quality of rapeseed oil. The specific technical solution adopted is as follows:

[0004] In a first aspect, embodiments of the present invention provide a method for detecting the quality of rapeseed oil, the method comprising:

[0005] Obtain the oleic acid and erucic acid concentrations of rapeseed oil samples;

[0006] Based on the oleic acid and erucic acid concentrations, the Raman spectral data of the rapeseed oil sample were determined;

[0007] Based on the Raman spectral data, the peak intensity contribution coefficient is determined;

[0008] Based on the oleic acid concentration, erucic acid concentration and peak intensity contribution coefficient, a rapeseed oil quality detection model is constructed.

[0009] The quality of the rapeseed oil to be predicted is detected based on the Raman peak characteristics of the rapeseed oil to be detected by the rapeseed oil quality detection model, and the detection result of the rapeseed oil to be predicted is obtained.

[0010] Secondly, embodiments of the present invention provide a rapeseed oil quality testing device, the rapeseed oil quality testing device comprising:

[0011] The acquisition module is used to acquire the oleic acid and erucic acid concentrations of rapeseed oil samples;

[0012] The first determining module is used to determine the Raman spectral data of the rapeseed oil sample based on the oleic acid concentration and erucic acid concentration.

[0013] The second determining module is used to determine the peak intensity contribution coefficient based on the Raman spectral data;

[0014] A construction module is used to construct a rapeseed oil quality detection model based on the oleic acid concentration, erucic acid concentration, and peak intensity contribution coefficient.

[0015] The detection module is used to perform quality detection on the Raman peak characteristics of the rapeseed oil to be predicted based on the rapeseed oil quality detection model, and obtain the detection result of the rapeseed oil to be predicted.

[0016] Thirdly, a computer program product is provided, comprising: computer program code, which, when run on a computer, causes the computer to perform the method described in the first aspect.

[0017] Fourthly, a computer-readable storage medium is provided that stores computer program code, which, when executed on a computer, causes the computer to perform the method described in the first aspect.

[0018] This invention offers the following advantages: After obtaining the oleic acid and erucic acid concentrations of a rapeseed oil sample, the Raman spectral data of the sample is determined based on these concentrations, and a peak intensity contribution coefficient is calculated based on this data. This fully considers the chemical mechanism of Raman spectroscopy and calculates the peak intensity contribution coefficient based on the reference normalized Raman spectra of oleic acid and erucic acid molecules, thus distinguishing the different degrees of contribution of different components to the peak intensity and improving the prediction accuracy of each component's concentration. Subsequently, a rapeseed oil quality detection model is constructed based on the oleic acid concentration, erucic acid concentration, and the peak intensity contribution coefficient, improving the accuracy of the constructed model. Finally, for the rapeseed oil to be predicted, the Raman peak characteristics are extracted, and quality detection is performed using this model, yielding more accurate results and improving the prediction accuracy of rapeseed oil quality. Attached Figure Description

[0019] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. 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.

[0020] Figure 1 This is a schematic diagram illustrating the implementation process of a rapeseed oil quality testing method provided in an embodiment of the present invention;

[0021] Figure 2 This is a schematic diagram of another implementation process of a rapeseed oil quality detection method provided in an embodiment of the present invention;

[0022] Figure 3 This is a schematic diagram of another implementation process of a rapeseed oil quality testing method provided in an embodiment of the present invention;

[0023] Figure 4 This is a schematic diagram of the composition structure of a rapeseed oil quality testing device provided in an embodiment of the present invention;

[0024] Figure 5 This is a schematic diagram of the structure of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0025] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a rapeseed oil quality testing method proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined from any suitable form.

[0026] In the description of the embodiments of the present invention, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of the present invention, "multiple" means two or more.

[0027] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0028] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0029] The specific scheme of the rapeseed oil quality testing method provided by the present invention will be described in detail below with reference to the accompanying drawings. Please refer to the accompanying drawings. Figure 1 The illustration shows a schematic flowchart of a rapeseed oil quality testing method provided by an embodiment of the present invention, the method comprising:

[0030] 101. Obtain the oleic acid and erucic acid concentrations of rapeseed oil samples.

[0031] Here, a certain number of rapeseed oil samples were collected. Using chemical methods such as gas chromatography, the concentrations of oleic acid and erucic acid in the collected samples were determined, forming a concentration input (independent variable) matrix. .in, 2 indicates the number of samples collected; 2 indicates that there are two dimensions of input features: oleic acid concentration (%) and erucic acid concentration (%).

[0032] Raman spectra of rapeseed oil samples were collected using a Raman spectrometer, and a light intensity output (dependent variable) matrix was constructed. .in, Indicates the number of samples collected; This indicates the number of Raman shifts (wavenumbers) in the sample spectrum.

[0033] In a specific example, 35 rapeseed oil samples were collected. Using chemical methods such as gas chromatography, the concentrations of oleic acid and erucic acid in the collected samples were determined, forming a concentration input (independent variable) matrix. .

[0034] 102. Based on the oleic acid concentration and erucic acid concentration, determine the Raman spectral data of the rapeseed oil sample.

[0035] Here, data preprocessing is performed using oleic acid and erucic acid concentrations to obtain more accurate Raman spectral data.

[0036] In some possible implementations, step 102 above can be achieved through the following steps 121 to 123 (not shown in the figures):

[0037] 121. Measure the number of Raman shifts of the rapeseed oil sample.

[0038] Here, the number of Raman shifts is the wavenumber.

[0039] 122. Based on the oleic acid concentration, erucic acid concentration, and the number of Raman shifts of the rapeseed oil sample, determine the light intensity matrix.

[0040] Here, the light intensity matrix is ​​calculated using the least squares method by combining the concentrations of oleic acid, erucic acid, and the number of Raman shifts of the rapeseed oil sample.

[0041] 123. Baseline correction is performed on the light intensity matrix to obtain the Raman spectral data.

[0042] Here, Raman spectral data of rapeseed oil samples are collected using a Raman spectrometer to form a light intensity matrix. .in, Indicates the number of samples collected; This represents the number of Raman shifts (wavenumbers) in the sample spectrum. The input is a two-dimensional concentration, and the output is a high-dimensional Raman spectrum. To reduce the output dimensionality and simultaneously reduce noise interference, the output matrix is ​​modified... The Raman characteristic peaks corresponding to specific molecular groups are numerically integrated, and the numerical integration is shown in formula (1):

[0043] (1);

[0044] in, Represents the vibrational modes of molecular groups The numerical integral value of the corresponding Raman characteristic peak, Represents the vibrational modes of molecular groups From Raman displacement To Raman shift The Raman characteristic peak matrix. This corresponds to the five molecular group vibration modes common to oleic acid and erucic acid. Based on this, the output results are derived from... The dimension is reduced to 5, at which point the output matrix... .

[0045] In a specific example, Raman spectra of these 35 rapeseed oil samples were collected using a Raman spectrometer, with each spectrum having a Raman shift (wavenumber) range of 191. up to 2649 A total of 512 Raman shifts (wavenumbers) constitute the light intensity output (dependent variable) matrix. For the output matrix Numerical integration is performed on the Raman characteristic peaks corresponding to specific molecular groups in the Raman spectrum. Based on this, the output dimension is reduced from 512 dimensions to 5 dimensions, at which point the output matrix... Thus, by using a Raman spectrometer to perform baseline correction on the acquired Raman spectra, the accuracy of the obtained Raman spectral data can be improved.

[0046] 103. Based on the Raman spectral data, determine the peak intensity contribution coefficient.

[0047] Here, after obtaining the Raman spectral data, it is possible to calculate the peak intensity contribution coefficient of each molecule to the vibrational mode of the molecular group.

[0048] In some possible implementations, firstly, based on the Raman spectral data, the structural influence of any molecule in the rapeseed oil sample on the vibrational mode of the molecular group is determined; then, the number of molecular groups corresponding to the vibrational mode of the molecular group in the any molecule is determined; finally, based on the structural influence and the number, the peak intensity contribution coefficient of the any molecule to the vibrational mode of the molecular group is determined.

[0049] Here, while the number of different molecular groups in different molecules is readily available, the structural influence of different molecules on these groups is uncertain. Therefore, this invention combines both factors into a single influencing factor: Among them, this influence factor , called molecules Vibrational modes of molecular groups The peak intensity contribution coefficient, j represents the vibrational mode of the j-th molecule group, and m represents the m-th molecule; This represents the vibrational mode of the m-th molecule against its molecular group. The resulting structural impact, Represents the vibrational modes of molecular groups in the m-th molecule. The number of corresponding molecular groups. Thus, by analyzing the structural influence of a molecule on the vibrational modes of molecular groups, the peak intensity contribution coefficient of that molecule to the vibrational modes of molecular groups can be accurately calculated.

[0050] 104. Based on the oleic acid concentration, erucic acid concentration and peak intensity contribution coefficient, a rapeseed oil quality detection model is constructed.

[0051] Here, a preliminary detection model is constructed using oleic acid concentration, erucic acid concentration, and the peak intensity contribution coefficient. Then, logarithmic processing is performed on the preliminary detection model to construct a rapeseed oil quality detection model.

[0052] In some possible implementations, step 104 above can be achieved through... Figure 2 The steps shown are to be implemented as follows:

[0053] 201. Determine the concentration of any molecule in the vibrational mode of a molecular group in a rapeseed oil sample.

[0054] Here, the concentration of any individual molecule is calculated, not the concentration of molecular groups. The concentration of the m-th molecule in the rapeseed oil sample, representing the vibrational mode of its molecular group, is calculated. .

[0055] 202. The concentration of any molecule is adjusted based on the peak intensity contribution coefficient to obtain the adjustment result.

[0056] Here, by multiplying the peak intensity contribution coefficient by the concentration of any molecule, the peak intensity contribution coefficient is adjusted to the concentration of any molecule, thus obtaining the adjusted result.

[0057] In some possible implementations, the concentration of any molecule is adjusted by fusing the peak intensity contribution coefficient with the number of molecular groups corresponding to the vibrational modes of any molecular group in any molecule, and then adjusting the concentration of any molecule based on the fusion result.

[0058] In related technologies, a partial least squares (PLS) model is established based on Raman spectroscopy to detect oleic acid and erucic acid concentrations, from Raman spectrum I to concentration C. Then, collected samples are used for parameter training and fitting. This algorithm is purely data-driven, lacks sufficient model interpretability, and fails to fully utilize the clear chemical mechanisms and unique fingerprint properties of Raman spectroscopy.

[0059] Based on this, the starting point for the model algorithm design in this invention is the clear chemical mechanism and fingerprint characteristics of Raman spectroscopy. Specifically, Raman spectroscopy involves a laser of a specific frequency striking a sample. Due to Raman scattering, vibrational modes of certain molecular groups in the sample absorb the laser light at specific frequencies. This specific absorption frequency corresponds to the wavenumber or Raman shift. A higher Raman peak intensity corresponding to a specific wavenumber or Raman shift indicates a greater number of photons absorbed at that frequency by the vibrational mode of that specific molecular group, thus implying a stronger signal from that specific molecular group vibrational mode. Each Raman peak and its intensity correspond to a specific molecular group vibrational mode and its signal intensity; this is the unique fingerprint characteristic of Raman spectroscopy.

[0060] Based on this, the higher the concentration of a specific molecular group in the sample, the stronger the corresponding Raman peak, and a proportional relationship can be established as shown in formula (2):

[0061] (2);

[0062] in, Indicates the vibrational modes of molecular groups in the sample The corresponding Raman characteristic peak intensity, Indicates the vibrational modes of molecular groups in the sample The signal strength.

[0063] Based on the chemical mechanism of molecular group vibration modes, the signal intensity is affected by both the concentration of the molecular group and the molecular structure that influences the vibration mode. Therefore, it is reasonable to assume the existence of a mathematical relationship as shown in formula (3):

[0064] (3);

[0065] in, Represents the vibrational modes of molecular groups The concentration of the corresponding molecular groups, Indicates the vibrational mode of the molecular group. The molecular structure influences it.

[0066] Furthermore, since the objective of this invention is to determine the concentration of molecules, rather than the concentration of molecular groups, when several molecules possess the same molecular group, theoretically the concentration of that molecular group should be the weighted sum of the concentrations of all molecules. Here, the weight refers to the contribution of each molecule to the concentration of that group, i.e., the ratio of the number of molecules with that group in each molecule. Therefore, in formula (3)... The following mathematical relationship is satisfied, as shown in formula (4):

[0067] (4);

[0068] in, Represents the vibrational modes of molecular groups in the m-th molecule. The number of corresponding molecular groups, This represents the concentration of the m-th molecule.

[0069] Similarly, molecular structure influences It also satisfies formula (5):

[0070] (5);

[0071] in, This represents the vibrational mode of the m-th molecule against its molecular group. The resulting structural impact, This represents the concentration of the m-th molecule.

[0072] Furthermore, integrating the above formulas, we can obtain formula (6):

[0073] (6);

[0074] Based on this, the relationship shown in formula (7) is obtained:

[0075] (7);

[0076] Therefore, for the detection of oleic acid and erucic acid concentrations in rapeseed oil, formula (7) can be used to obtain formula (8):

[0077] (8);

[0078] in, , These represent the concentrations of oleic acid and erucic acid, respectively. , These represent the vibrational modes of molecular groups in oleic acid and erucic acid molecules, respectively. The number of corresponding molecular groups, , These represent the vibrational modes of oleic acid and erucic acid molecules relative to their molecular groups, respectively. The resulting structural impact.

[0079] Since the number of different molecular groups of different molecules is easy to obtain, but the structural influence of different molecules on different molecular groups is uncertain, the two are combined into a single influencing factor, as shown in formula (9):

[0080] (9);

[0081] This impact factor , called molecules Vibrational modes of molecular groups The peak intensity contribution coefficient is given by formula (10):

[0082] (10);

[0083] Where N represents the number of molecules. This is the result of adjusting the peak intensity contribution coefficient for the concentration of any molecule.

[0084] 203. Based on the adjustment results of the oleic acid concentration, erucic acid concentration and multiple molecules, the rapeseed oil quality detection model is constructed.

[0085] Here, the Raman characteristic peak intensity corresponding to the vibrational mode of molecular groups in the rapeseed oil sample is calculated by adjusting the results of multiple molecules; the oleic acid concentration, erucic acid concentration and Raman characteristic peak intensity in the rapeseed oil sample are combined to construct the rapeseed oil quality detection model.

[0086] In some possible implementations, it can be achieved through Figure 3 The steps shown are to be implemented as follows:

[0087] 301, obtain the adjustment results of the peak intensity contribution coefficient on the oleic acid concentration and the adjustment results of the peak intensity contribution coefficient on the erucic acid concentration.

[0088] 302. The adjustment results of the peak intensity contribution coefficient on the oleic acid concentration and the adjustment results of the peak intensity contribution coefficient on the erucic acid concentration are summed to obtain the concentration summation result.

[0089] 303. The proportional coefficients of the vibrational modes of the molecular groups and the summation results of the concentrations are fused to construct a preliminary detection model.

[0090] 304. Logarithmic processing is performed on the preliminary detection model to obtain the processing result.

[0091] 305. Based on the processing results and the concentrations of oleic acid and erucic acid in the rapeseed oil sample, a rapeseed oil quality detection model is constructed.

[0092] Here, multiple weighting coefficients are determined for the oleic acid concentration and erucic acid concentration in the rapeseed oil sample, respectively (e.g., ) and bias coefficients (e.g., Then, the multiple product results and processing results are weighted and summed using multiple weighting coefficients to obtain a weighted result; finally, the weighted result and the bias coefficient are combined to construct a rapeseed oil quality detection model.

[0093] In steps 301 to 305 above, based on formula (10), the oleic acid and erucic acid molecules in rapeseed oil are as shown in formula (11):

[0094] (11);

[0095] This invention calculates the peak intensity contribution coefficient based on the reference normalized Raman spectra of oleic acid and erucic acid molecules. The possible values ​​are as follows:

[0096]

[0097]

[0098] Therefore, a preliminary detection model for rapeseed oil quality testing can be established, as shown in formula (12):

[0099] (12);

[0100] in, Represents molecular vibrational modes The proportionality coefficient, These represent the vibrational modes of oleic acid and erucic acid molecules relative to their molecular groups, respectively. The peak intensity contribution coefficient. This refers to the adjustment results of the peak intensity contribution coefficient to the oleic acid concentration and the adjustment results of the peak intensity contribution coefficient to the erucic acid concentration. This is the result of summing the concentrations.

[0101] Specifically, considering the saturation effect of concentration, a logarithmic function was introduced to obtain the logarithmic processing result of the preliminary detection model; considering the intermolecular forces, an interaction term was introduced; considering the nonlinear effect, a higher-order term was introduced; considering the influence of unknown molecules, a residual (bias) was introduced, and finally the following rapeseed oil detection model was established, as shown in formula (13):

[0102]

[0103] ;

[0104] Therefore, it can be seen that, compared to light intensity To concentration The PLS model, a rapeseed oil detection model based on chemical mechanisms, establishes a system for detecting oleic acid and erucic acid concentrations. To light intensity The multiple regression model.

[0105] In some possible implementations, the weighted result and the bias coefficient are fused to construct a training detection model for rapeseed oil quality; then, in the training detection model, a concentration feature matrix characterizing oleic acid concentration and erucic acid concentration is determined; finally, based on the least squares method and the concentration feature matrix, the bias coefficient and weight coefficient are adjusted to train the training detection model, thereby obtaining the rapeseed oil quality detection model and realizing the training process of the training detection model.

[0106] Here, it appears to be a nonlinear model, but during training, four new features—the logarithm of the weighted sum, the interaction term, and the higher-order term—are directly created to establish the following linear model, as shown in formula (14):

[0107]

[0108] Right now: (14);

[0109] in, It is a proportional coefficient matrix, which needs to be trained and optimized using samples. It is a concentration feature matrix constructed based on the mechanism. This represents the residual (bias coefficient) indicating the influence of the unknown component on the Raman bee intensity. Based on the collected rapeseed oil samples, the least squares method was used to perform parameter analysis. , Optimization.

[0110] In a specific example, taking 35 rapeseed oil samples as an example... , representing the proportional coefficient matrix, requires training and optimization using samples. , representing the concentration feature matrix constructed based on the mechanism. , representing the residual indicating the influence of the unknown component on the Raman bee strength. , representing the output matrix of Raman spectral characteristic peak intensities. Based on 35 collected rapeseed oil samples, parameters were determined using the least squares method. , Optimization.

[0111] 105. Based on the rapeseed oil quality detection model, the Raman peak characteristics of the rapeseed oil to be predicted are used for quality detection to obtain the detection results of the rapeseed oil to be predicted.

[0112] Here are the Raman spectral data of newly collected rapeseed oil samples to be predicted. Data preprocessing and feature engineering were performed to obtain the processed Raman peak features. Then, The input is fed into the trained model, and the nonlinear equations are solved using a constrained quasi-Newton method to reconstruct the oleic acid concentration. and erucic acid concentration .

[0113] In a specific example, the Raman spectral data of the collected rapeseed oil sample Leave-one-out cross-validation was used, and data preprocessing and feature engineering were performed to obtain the processed Raman peak features. (i.e., Raman spectral data). Then, The input is fed into the trained model, and the nonlinear equations are solved using a constrained quasi-Newton method to reconstruct the oleic acid concentration. and erucic acid concentration Cross-validation results show that the model has the following average coefficient of determination for detecting oleic acid. The mean root mean square error reached 0.9645. The coefficient of determination for oleic acid obtained by the traditional PLS model is 1.9546. The mean root mean square error reached 0.9485. The average coefficient of determination for erucic acid is 2.9610. The mean root mean square error reached 0.9423. The coefficient of determination for oleic acid obtained by the traditional PLS model is 0.9675. The mean root mean square error reached 0.9278. It is 1.1984.

[0114] In this embodiment of the invention, after obtaining the oleic acid and erucic acid concentrations of a rapeseed oil sample, the Raman spectral data of the rapeseed oil sample is determined based on these concentrations, and a peak intensity contribution coefficient is determined based on the Raman spectral data. This fully considers the chemical mechanism of Raman spectroscopy and calculates the peak intensity contribution coefficient based on the reference normalized Raman spectra of oleic acid and erucic acid molecules to distinguish the different degrees of contribution of different components to the peak intensity, thereby improving the prediction accuracy of each component concentration. Subsequently, based on the oleic acid concentration, erucic acid concentration, and the peak intensity contribution coefficient, a rapeseed oil quality detection model is constructed, which can improve the accuracy of the constructed rapeseed oil quality detection model. Finally, for the rapeseed oil to be predicted, the Raman peak characteristics of the rapeseed oil to be predicted are extracted, and quality detection is performed using this rapeseed oil detection model, which can obtain more accurate detection results.

[0115] In this embodiment of the invention, the rapeseed oil detection model fully considers the chemical mechanism of Raman spectroscopy and designs a peak intensity contribution coefficient based on the reference normalized Raman spectra of oleic acid and erucic acid molecules to distinguish the different degrees of contribution of different components to the peak intensity, thereby improving the prediction accuracy of the concentration of each component. This rapeseed oil detection model can be used to model the concentration prediction of any specific fatty acid component (such as oleic acid and erucic acid), and the proportionality coefficient between concentration and peak intensity, as well as the influence of unknown fatty acid components on specific Raman peaks, can be solved in a data-driven manner. If the modeling method is based on light intensity to concentration, the influence of unknown fatty acids is difficult to handle. Therefore, it can be extended to the detection of more material components, thus being more suitable for the application expansion of material component content detection. Oleic acid concentration is 50%~70%, while erucic acid concentration is only 1%-3%, a large difference in dimensional range. If deep learning modeling methods relying on large-scale data are not used, and modeling is purely based on machine learning, the prediction effect for outputs with such significant dimensional differences is not good. The mechanistic model established in this embodiment of the invention has a better predictive effect on components such as low-concentration erucic acid, and thus has a better detection effect on high oleic acid and low erucic acid content with large dimensional differences.

[0116] This invention provides a rapeseed oil quality testing device. Please refer to [link / reference]. Figure 4 The diagram illustrates the structural composition of a rapeseed oil quality testing device according to an embodiment of the present invention. The device 400 includes:

[0117] The acquisition module 401 is used to acquire the oleic acid concentration and erucic acid concentration of rapeseed oil samples;

[0118] The first determining module 402 is used to determine the Raman spectral data of the rapeseed oil sample based on the oleic acid concentration and erucic acid concentration.

[0119] The second determining module 403 is used to determine the peak intensity contribution coefficient based on the Raman spectral data;

[0120] Module 404 is used to construct a rapeseed oil quality detection model based on the oleic acid concentration, erucic acid concentration and peak intensity contribution coefficient.

[0121] The detection module 405 is used to perform quality detection on the Raman peak characteristics of the rapeseed oil to be predicted based on the rapeseed oil quality detection model, and obtain the detection result of the rapeseed oil to be predicted.

[0122] In some possible implementations, the detection module 405 is further configured to determine the concentration of any molecule in the vibrational mode of molecular groups in the rapeseed oil sample; adjust the concentration of any molecule based on the peak intensity contribution coefficient to obtain the adjustment result; and construct the rapeseed oil quality detection model based on the oleic acid concentration, erucic acid concentration and the adjustment results of multiple molecules.

[0123] In some possible implementations, the detection module 405 is further configured to determine the Raman characteristic peak intensity corresponding to the vibrational mode of molecular groups in the rapeseed oil sample based on the adjustment results corresponding to multiple molecules; and to construct the rapeseed oil quality detection model based on the oleic acid concentration, erucic acid concentration and the Raman characteristic peak intensity in the rapeseed oil sample.

[0124] In some possible implementations, the detection module 405 is further configured to: acquire the adjustment results of the peak intensity contribution coefficient on the oleic acid concentration and the adjustment results of the peak intensity contribution coefficient on the erucic acid concentration; sum the adjustment results of the peak intensity contribution coefficient on the oleic acid concentration and the adjustment results of the peak intensity contribution coefficient on the erucic acid concentration to obtain a concentration summation result; fuse the proportionality coefficient of the molecular group vibration mode with the concentration summation result to construct a preliminary detection model; perform logarithmic processing on the preliminary detection model to obtain a processing result; and construct the rapeseed oil quality detection model based on the processing result, the oleic acid concentration, and the erucic acid concentration in the rapeseed oil sample.

[0125] In some possible implementations, the detection module 405 is further configured to determine multiple weighting coefficients and bias coefficients for the oleic acid concentration and erucic acid concentration in the rapeseed oil sample, respectively; multiply the oleic acid concentration and erucic acid concentration pairwise to obtain multiple product results; weight and sum the multiple product results and the processing result based on the multiple weighting coefficients to obtain a weighted result; and construct the rapeseed oil quality detection model based on the weighted result and the bias coefficients.

[0126] In some possible implementations, the detection module 405 is further configured to fuse the weighted result and the bias coefficient to construct a detection model to be trained for rapeseed oil quality; in the detection model to be trained, a concentration feature matrix characterizing oleic acid concentration and erucic acid concentration is determined; based on the least squares method and the concentration feature matrix, the bias coefficient and weight coefficient are adjusted to train the detection model to be trained, thereby obtaining the rapeseed oil quality detection model.

[0127] In some possible implementations, the first determining module 402 is further configured to measure the number of Raman shifts of the rapeseed oil sample; determine the light intensity matrix based on the oleic acid concentration, erucic acid concentration and the number of Raman shifts of the rapeseed oil sample; and perform baseline correction on the light intensity matrix to obtain the Raman spectral data.

[0128] In some possible implementations, the second determining module 403 is further configured to, based on the Raman spectral data, determine the structural influence of any molecule in the rapeseed oil sample on the vibrational mode of the molecular group; determine the number of molecular groups corresponding to the vibrational mode of the molecular group in any molecule; and, based on the structural influence and the number, determine the peak intensity contribution coefficient of any molecule to the vibrational mode of the molecular group.

[0129] Optionally, the transmission medium can be a wired link (e.g., but not limited to, coaxial cable, optical fiber, and Digital Subscriber Line (DSL)) or a wireless link (e.g., but not limited to, Wireless Fidelity (WIFI), Bluetooth, and mobile device networks). It should be noted that the apparatus provided in the above embodiments is only illustrative of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. Furthermore, the method embodiments provided in the above embodiments belong to the same concept, and their specific implementation processes are detailed in the method embodiments, and will not be repeated here.

[0130] Figure 5 This is a schematic diagram of the structure of a computer device provided in an embodiment of the present invention. For example, as shown... Figure 5 As shown, the computer device 500 includes: a memory 501, a processor 502, and a computer program 503 stored in the memory 501 and running on the processor 502, wherein when the processor 502 executes the computer program 503, the computer device can execute any of the rapeseed oil quality detection methods described above.

[0131] Furthermore, this embodiment of the invention also protects a system that may include a memory and a processor. The memory stores executable program code, and the processor is used to call and execute the executable program code to perform a rapeseed oil quality detection method provided by this embodiment of the invention. This embodiment can divide the system into functional modules based on the above method example. For example, each module can correspond to a specific function, or two or more functions can be integrated into a single processing module. The integrated module can be implemented in hardware. It should be noted that the module division in this embodiment is illustrative and only represents a logical functional division; other division methods may exist in actual implementation. It should also be noted that all relevant content of each step involved in the above method embodiment can be referenced to the functional description of the corresponding functional module, and will not be repeated here.

[0132] It should be understood that the apparatus provided in this embodiment is used to execute the above-described rapeseed oil quality detection method, and therefore can achieve the same effect as the above-described implementation method. When using integrated units, the apparatus may include a processing module and a storage module. When the apparatus is applied to a device, the processing module can be used to control and manage the device's operations. The storage module can be used to support the device in executing mutual program code, etc. The processing module may be a processor or a controller, which can implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this invention. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of Digital Signal Processing (DSP) and a microprocessor, etc., and the storage module may be a memory.

[0133] Furthermore, the apparatus provided in the embodiments of the present invention may specifically be a chip, component, or module. The chip may include a connected processor and a memory; wherein, the memory is used to store instructions, and when the processor calls and executes the instructions, the chip can execute the rapeseed oil quality detection method provided in the above embodiments. This embodiment also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer, the computer executes the above-mentioned method steps to implement the rapeseed oil quality detection method provided in the above embodiments.

[0134] This embodiment also provides a computer program product. When the computer program product is run on a computer, it causes the computer to perform the aforementioned related steps to implement the rapeseed oil quality detection method provided in the above embodiment. The device, computer-readable storage medium, computer program product, or chip provided in this embodiment are all used to execute the corresponding methods provided above. Therefore, the beneficial effects they achieve can be referred to the beneficial effects in the corresponding methods provided above, and will not be repeated here. Through the description of the above embodiments, those skilled in the art can understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In the embodiments provided by this invention, it should be understood that the disclosed device and method can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of modules or units is merely a logical functional division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, and can be electrical, mechanical or other forms.

[0135] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multiple task processing and parallel processing are possible or may be advantageous. The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. The above content is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for detecting the quality of rapeseed oil, characterized in that, The rapeseed oil quality testing method includes: Obtain the oleic acid and erucic acid concentrations of rapeseed oil samples; Based on the oleic acid and erucic acid concentrations, the Raman spectral data of the rapeseed oil sample were determined; Based on the Raman spectroscopy data, the structural influence of any molecule in the rapeseed oil sample on the vibrational modes of molecular groups is determined; wherein, any molecule in the rapeseed oil sample includes: any oleic acid molecule and any erucic acid molecule; Determine the number of molecular groups corresponding to the vibrational modes of molecular groups in any molecule; wherein the number of molecular groups includes: the number of molecular groups corresponding to the vibrational modes of molecular groups in oleic acid molecules, and the number of molecular groups corresponding to the vibrational modes of molecular groups in erucic acid molecules. Based on the structural influence and the quantity, the peak intensity contribution coefficient of any molecule to the vibrational mode of the molecular group is determined; wherein, the peak intensity contribution coefficient includes: the peak intensity contribution coefficient of oleic acid molecule to the vibrational mode of the molecular group, and the peak intensity contribution coefficient of erucic acid molecule to the vibrational mode of the molecular group; the expression for the peak intensity contribution coefficient is: , For molecules Vibrational modes of molecular groups The peak intensity contribution coefficient, Indicates the first Vibrational modes of individual molecules and molecular groups The resulting structural impact, Indicates the first Vibrational modes of molecular groups in each molecule The number of corresponding molecular groups; Determine the concentration of any molecule in the rapeseed oil sample that represents a specific vibrational mode of a molecular group; The peak intensity contribution coefficient of oleic acid molecules to the vibrational mode of molecular groups is multiplied by the oleic acid concentration of the group vibrational mode to obtain the adjustment result of oleic acid concentration. Similarly, the peak intensity contribution coefficient of erucic acid molecules to the vibrational mode of molecular groups is multiplied by the erucic acid concentration of the group vibrational mode to obtain the adjustment result of erucic acid concentration. The results of adjusting the oleic acid concentration and the erucic acid concentration are summed to obtain the concentration summation result; The proportionality coefficients of the vibrational modes of the molecular groups and the summation results of the concentrations are fused to construct a preliminary detection model; The preliminary detection model is subjected to logarithmic processing to obtain the processing result; Based on the processing results and the oleic acid and erucic acid concentrations in the rapeseed oil samples, a rapeseed oil quality detection model was constructed. Based on the rapeseed oil quality detection model, the Raman peak characteristics of the rapeseed oil to be predicted are used for quality detection, and the detection results of the rapeseed oil to be predicted are obtained.

2. The method for detecting rapeseed oil quality according to claim 1, characterized in that, The process involves constructing a rapeseed oil quality detection model based on the processing results, the oleic acid concentration, and the erucic acid concentration in the rapeseed oil sample, including: Multiple weighting coefficients and bias coefficients were determined for the oleic acid concentration and erucic acid concentration in the rapeseed oil sample, respectively. The oleic acid concentration and erucic acid concentration are multiplied pairwise to obtain multiple product results; The weighted product results and the processing results are weighted and summed based on the multiple weighting coefficients to obtain a weighted result; Based on the weighted results and the bias coefficients, the rapeseed oil quality detection model is constructed.

3. The method for detecting rapeseed oil quality according to claim 2, characterized in that, The process of constructing the rapeseed oil quality detection model based on the weighted results and the bias coefficient includes: The weighted results and the bias coefficients are fused together to construct a training detection model for rapeseed oil quality. In the detection model to be trained, a concentration feature matrix characterizing oleic acid concentration and erucic acid concentration is determined; Based on the least squares method and the concentration feature matrix, the bias coefficient and weight coefficient are adjusted to train the detection model to be trained, thereby obtaining the rapeseed oil quality detection model.

4. The method for detecting rapeseed oil quality according to claim 1, characterized in that, The determination of the Raman spectral data of the rapeseed oil sample based on the oleic acid concentration and erucic acid concentration includes: The number of Raman shifts in the rapeseed oil sample was determined; The light intensity matrix is ​​determined based on the oleic acid concentration, erucic acid concentration, and the number of Raman shifts of the rapeseed oil sample. The light intensity matrix is ​​baseline corrected to obtain the Raman spectral data.

5. A rapeseed oil quality testing device, characterized in that, The rapeseed oil quality testing device includes: The acquisition module is used to acquire the oleic acid and erucic acid concentrations of rapeseed oil samples; The first determining module is used to determine the Raman spectral data of the rapeseed oil sample based on the oleic acid concentration and erucic acid concentration. The second determining module is used to determine, based on the Raman spectral data, the structural influence of any molecule in the rapeseed oil sample on the vibrational mode of molecular groups; wherein, any molecule in the rapeseed oil sample includes: any oleic acid molecule and any erucic acid molecule; determine the number of molecular groups corresponding to the vibrational mode of molecular groups in any molecule; wherein, the number of molecular groups includes: the number of molecular groups corresponding to the vibrational mode of molecular groups in oleic acid molecules and the number of molecular groups corresponding to the vibrational mode of molecular groups in erucic acid molecules; based on the structural influence and the number, determine the peak intensity contribution coefficient of any molecule on the vibrational mode of molecular groups; wherein, the peak intensity contribution coefficient includes: the peak intensity contribution coefficient of oleic acid molecules on the vibrational mode of molecular groups and the peak intensity contribution coefficient of erucic acid molecules on the vibrational mode of molecular groups; the expression for the peak intensity contribution coefficient is: , For molecules Vibrational modes of molecular groups The peak intensity contribution coefficient, Indicates the first Vibrational modes of individual molecules and molecular groups The resulting structural impact, Indicates the first Vibrational modes of molecular groups in each molecule The number of corresponding molecular groups; A construction module is used to determine the concentration of any molecule in the vibrational mode of a molecular group in the rapeseed oil sample; to adjust the oleic acid concentration by multiplying the peak intensity contribution coefficient of the oleic acid molecule to the vibrational mode of the molecular group and the oleic acid concentration in the vibrational mode of the group; to adjust the erucic acid concentration by multiplying the peak intensity contribution coefficient of the erucic acid molecule to the vibrational mode of the molecular group and the erucic acid concentration in the vibrational mode of the group; to sum the adjusted oleic acid concentration and the adjusted erucic acid concentration to obtain a concentration summation result; to fuse the proportionality coefficient of the vibrational mode of the molecular group and the concentration summation result to construct a preliminary detection model; to perform logarithmic processing on the preliminary detection model to obtain a processing result; and to construct a rapeseed oil quality detection model based on the processing result, the oleic acid concentration, and the erucic acid concentration in the rapeseed oil sample. The detection module is used to perform quality detection on the Raman peak characteristics of the rapeseed oil to be predicted based on the rapeseed oil quality detection model, and obtain the detection result of the rapeseed oil to be predicted.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed on an electronic device, cause the electronic device to perform the method as described in any one of claims 1 to 4.

Images