Method and device for quantitative detection of irradiated milk powder, electronic equipment and storage medium

Abstract

The application provides a milk powder irradiation quantitative detection method and device, electronic equipment and storage medium, and belongs to the technical field of food irradiation detection. The method comprises the following steps: acquiring the signal intensity of target milk powder after irradiation; analyzing the physicochemical properties of the target milk powder after irradiation to obtain key component parameters, and extracting characteristic parameters in a resonance spectrum of the target milk powder after irradiation; determining the target matrix type of the target milk powder based on the key component parameters and the characteristic parameters, determining the dose-response slope curve and the free radical decay rate constant corresponding to the target matrix type, and calculating the preliminary irradiation dose of the target milk powder; and correcting the preliminary irradiation dose based on a correction coefficient to obtain the final irradiation dose of the target milk powder. The application can solve the problems of matrix interference, time-dependent error and low detection efficiency in the irradiation detection of traditional milk powder.

Description

Technical Field

[0001] This application belongs to the field of food irradiation detection technology, and more specifically, relates to a quantitative detection method and device for milk powder irradiation, electronic equipment, and storage medium. Background Technology

[0002] Irradiation technology is commonly used to extend the shelf life of products such as milk powder and ensure their microbiological safety. Accurate quantification of irradiation dose is crucial for ensuring product quality and regulatory compliance. Traditional electron paramagnetic resonance (EPR) methods are primarily used for qualitative determination of whether a sample has been irradiated, lacking precise quantitative capabilities and failing to meet the standardized regulatory requirements for food irradiation doses. Furthermore, different brands and formulations of milk powder have varying component contents, leading to significant differences in the free radical signal responses generated after irradiation. Using a single calibration model would introduce substantial errors. Therefore, it is essential to develop a quantitative detection method for milk powder irradiation to address the problems existing in current technologies. Summary of the Invention

[0003] The purpose of this application is to provide a method, device, electronic equipment, and storage medium for quantitative detection of milk powder irradiation, so as to solve the problems of matrix interference, time-dependent error and low detection efficiency in traditional milk powder irradiation detection, and to realize the quantitative detection of milk powder irradiation dose.

[0004] A first aspect of this application provides a method for quantitative detection of milk powder irradiation, comprising:

[0005] The signal intensity of the irradiated target milk powder was obtained by detecting the irradiated target milk powder using an electron paramagnetic resonance spectrometer.

[0006] The physicochemical properties of the irradiated target milk powder were analyzed to obtain key component parameters. Characteristic parameters were extracted from the resonance spectrum of the irradiated target milk powder, including peak shape, spectral splitting factor, and linewidth.

[0007] The target matrix type of the target milk powder is determined based on key component parameters and characteristic parameters. The dose-response slope curve and free radical decay rate constant corresponding to the target matrix type are queried in the target milk powder matrix database.

[0008] The initial irradiation dose of the target milk powder is calculated based on the signal intensity and dose-response slope curve. A correction coefficient is determined based on the storage time of the target milk powder from the irradiation date and the free radical decay rate constant. The initial irradiation dose is corrected based on the correction coefficient to obtain the final irradiation dose of the target milk powder.

[0009] In one possible implementation, the quantitative detection method for milk powder irradiation also includes:

[0010] The distribution parameters of free radicals in the target milk powder are obtained, including the distribution uniformity index and the fine spectral features.

[0011] The accuracy of the target matrix type is judged based on the distribution parameters of free radicals to obtain the judgment result;

[0012] If the judgment result is accurate, then the matrix type of the target milk powder is determined to be the target matrix type;

[0013] If the judgment result is inaccurate, the target matrix type is updated based on the distribution parameters of free radicals to obtain the updated target matrix type.

[0014] In one possible implementation, the accuracy of the target matrix type is determined based on the distribution parameters of the free radicals to obtain the judgment result, including:

[0015] The distribution parameters of free radicals of the target milk powder are compared with the benchmark range of the corresponding parameters of the target matrix type stored in the target milk powder matrix database;

[0016] Calculate the normalized offset of each distribution parameter value of the target milk powder relative to its corresponding benchmark range center value, and perform a weighted summation based on the preset weight of each distribution parameter value in the type discrimination to obtain the comprehensive deviation score;

[0017] If the overall deviation score is less than or equal to the first threshold, the judgment result is determined to be accurate.

[0018] If the overall deviation score is greater than the first threshold and less than or equal to the second threshold, the judgment result is determined to be inaccurate, and a prompt signal requesting review is generated; wherein the first threshold is less than the second threshold.

[0019] If the overall deviation score is greater than the second threshold, the judgment result is determined to be inaccurate.

[0020] In one possible implementation, the correction coefficient is determined based on the storage time of the target milk powder since the irradiation date and the free radical decay rate constant, including:

[0021] Based on the storage time of the target milk powder from the date of irradiation and the free radical decay rate constant, the correction coefficient is determined using the following correction formula:

[0022] , where A is the correction coefficient, λ is the free radical decay rate constant, and t is the storage time of the target milk powder from the irradiation date.

[0023] In one possible implementation, the milk powder matrix database includes multiple dose-response slope curves, each corresponding to a milk powder matrix; the generation process of each dose-response slope curve includes:

[0024] For each milk powder sample: determine multiple gradient dose values ​​for the milk powder sample, irradiate the milk powder sample based on the multiple gradient dose values, and obtain the signal intensity of the milk powder sample at each gradient dose value;

[0025] The mean value of the signal intensity of multiple milk powder samples under the same gradient dose value is calculated to obtain the dose-response slope curve corresponding to the milk powder matrix.

[0026] In one possible implementation, the physicochemical properties of the irradiated target milk powder are analyzed to obtain key component parameters, including:

[0027] The physicochemical properties of the irradiated target milk powder were analyzed to obtain the moisture content, ash content, fat content, and protein content.

[0028] In one possible implementation, the target matrix type of the target milk powder is determined based on key component parameters and characteristic parameters, including:

[0029] The target matrix type is selected from the parameter database based on key component parameters and characteristic parameters. The parameter database includes multiple milk powder matrix types, the values ​​of key component parameters and characteristic parameters corresponding to each milk powder matrix type.

[0030] A second aspect of this application provides a quantitative detection device for milk powder irradiation, comprising:

[0031] The signal acquisition module is used to acquire the signal intensity of the irradiated target milk powder. The signal intensity is obtained by detecting the irradiated target milk powder using an electron paramagnetic resonance spectrometer.

[0032] The feature extraction module is used to analyze the physicochemical properties of the irradiated target milk powder to obtain key component parameters and extract feature parameters from the resonance spectrum of the irradiated target milk powder. The feature parameters include peak shape, spectral splitting factor and linewidth.

[0033] The data calculation module is used to determine the target matrix type of the target milk powder based on key component parameters and characteristic parameters, and to query the dose-response slope curve and free radical decay rate constant corresponding to the target matrix type in the target milk powder matrix database;

[0034] The data correction module is used to calculate the initial irradiation dose of the target milk powder based on the signal intensity and dose-response slope curve, determine the correction coefficient based on the storage time of the target milk powder from the irradiation date and the free radical decay rate constant, and correct the initial irradiation dose based on the correction coefficient to obtain the final irradiation dose of the target milk powder.

[0035] In one possible implementation, the quantitative detection device for milk powder irradiation also includes a judgment module;

[0036] The judgment module is used to: obtain the distribution parameters of free radicals in the target milk powder, including the distribution uniformity index and the fine spectral features;

[0037] The accuracy of the target matrix type is judged based on the distribution parameters of free radicals to obtain the judgment result;

[0038] If the judgment result is accurate, then the matrix type of the target milk powder is determined to be the target matrix type;

[0039] If the judgment result is inaccurate, the target matrix type is updated based on the distribution parameters of free radicals to obtain the updated target matrix type.

[0040] In one possible implementation, the decision module is specifically used for:

[0041] The distribution parameters of free radicals of the target milk powder are compared with the benchmark range of the corresponding parameters of the target matrix type stored in the target milk powder matrix database;

[0042] Calculate the normalized offset of each distribution parameter value of the target milk powder relative to its corresponding benchmark range center value, and perform a weighted summation based on the preset weight of each distribution parameter value in the type discrimination to obtain the comprehensive deviation score;

[0043] If the overall deviation score is less than or equal to the first threshold, the judgment result is determined to be accurate.

[0044] If the overall deviation score is greater than the first threshold and less than or equal to the second threshold, the judgment result is determined to be inaccurate, and a prompt signal requesting review is generated; wherein the first threshold is less than the second threshold.

[0045] If the overall deviation score is greater than the second threshold, the judgment result is determined to be inaccurate.

[0046] In one possible implementation, the data correction module is specifically used for:

[0047] The correction coefficient is determined based on the storage time of the target milk powder from the date of irradiation, the free radical decay rate constant, and the following correction formula:

[0048] , where A is the correction coefficient, λ is the free radical decay rate constant, and t is the storage time of the target milk powder from the irradiation date.

[0049] In one possible implementation, the data computation module is specifically used to determine the generation process of each dose-response slope curve, including:

[0050] For each milk powder sample: determine multiple gradient dose values ​​for the milk powder sample, irradiate the milk powder sample based on the multiple gradient dose values, and obtain the signal intensity of the milk powder sample at each gradient dose value;

[0051] The mean value of the signal intensity of multiple milk powder samples under the same gradient dose value is calculated to obtain the dose-response slope curve corresponding to the milk powder matrix.

[0052] In one possible implementation, the feature extraction module is specifically used for:

[0053] The physicochemical properties of the irradiated target milk powder were analyzed to obtain the moisture content, ash content, fat content, and protein content.

[0054] In one possible implementation, the data computation module is specifically used for:

[0055] The target matrix type is selected from the parameter database based on key component parameters and characteristic parameters. The parameter database includes multiple milk powder matrix types, the values ​​of key component parameters and characteristic parameters corresponding to each milk powder matrix type.

[0056] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the above-described quantitative detection method for milk powder irradiation.

[0057] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described quantitative detection method for milk powder irradiation.

[0058] The beneficial effects of the quantitative detection method and apparatus for milk powder irradiation, electronic equipment, and storage medium provided in this application are as follows:

[0059] On the one hand, the embodiments of this application improve the accuracy of detection. These embodiments acquire the signal intensity of the target milk powder, analyze its physicochemical properties to obtain key component parameters, and extract resonance spectrum characteristic parameters to determine the target matrix type. Then, they query the corresponding dose-response slope curve and free radical decay rate constant from the database. The preliminary irradiation dose is then calculated using these parameters, and corrected by combining storage time and the free radical decay rate constant to obtain the final irradiation dose. This multi-parameter comprehensive analysis method fully considers the different free radical signal responses caused by variations in the composition of different brands and formula milk powders, effectively avoiding the huge errors caused by using a single calibration model, greatly improving the accuracy of irradiation dose detection, and meeting the standardized regulatory requirements for food irradiation doses.

[0060] On the other hand, the embodiments of this application enhance the practicality of the detection. The method provided in the embodiments of this application constructs a target milk powder matrix database, storing the dose-response slope curves and free radical decay rate constants corresponding to different matrix types. In actual detection, the matrix type only needs to be determined according to the parameters of the target milk powder to quickly obtain relevant parameters for calculation. The operation is simple and efficient, and it can adapt to the quantitative detection of irradiation of different brands and formula milk powders, providing a practical and reliable technical means for the accurate detection of milk powder irradiation dose. Attached Figure Description

[0061] To more clearly illustrate the technical solutions in the embodiments of this application, 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 this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0062] Figure 1 A schematic flowchart of a quantitative detection method for milk powder irradiation provided in an embodiment of this application;

[0063] Figure 2 This is a structural block diagram of a milk powder irradiation quantitative detection device provided in an embodiment of this application;

[0064] Figure 3 This is a schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0065] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0066] It should be noted that the terms "first," "second," etc., used in the specification, claims, and drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in sequences other than those illustrated or described herein.

[0067] In related technologies, long-lived free radicals generated by irradiation in milk powder gradually decay with storage time and environmental conditions (such as temperature and humidity). Traditional static detection models cannot correct for this signal loss, often leading to a significant underestimation of the irradiation dose. Furthermore, the analysis and quantitative calculation of EPR spectra are highly dependent on operator experience, lacking automation and intelligent support, which affects detection efficiency and result consistency. To address the above technical problems, embodiments of this application provide a method for quantitative detection of milk powder irradiation.

[0068] To make the objectives, technical solutions, and advantages of this application clearer, the following description will be provided in conjunction with the accompanying drawings and specific embodiments.

[0069] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a quantitative detection method for milk powder irradiation according to an embodiment of this application. The method may include:

[0070] S101: Obtain the signal intensity of the irradiated target milk powder. The signal intensity is obtained by detecting the irradiated target milk powder using an electron paramagnetic resonance spectrometer.

[0071] In this embodiment, the target milk powder is a specific milk powder sample to be irradiated and detected. The signal intensity is the response amplitude corresponding to the irradiation-induced products (free radicals or trapped electrons) in the milk powder, measured by a spectrometer, such as an EPR spectrometer, after the target milk powder has been irradiated and stored for a certain period of time.

[0072] In this embodiment, the irradiation dosage is unknown, and the purpose of this application is to accurately detect the irradiation dose of the milk powder. In this embodiment, the target milk powder, after being irradiated and stored for a certain period of time, is placed into a paramagnetic test tube, and the EPR spectrometer parameters are set to obtain the signal intensity of the target milk powder.

[0073] S102: Analyze the physicochemical properties of the irradiated target milk powder to obtain key component parameters, and extract characteristic parameters from the resonance spectrum of the irradiated target milk powder, including peak shape, spectral splitting factor and linewidth.

[0074] In this embodiment, the physicochemical properties of the target milk powder refer to the collective physical and chemical properties possessed by the irradiated target milk powder. These are the inherent core attributes of the milk powder itself, which can be analyzed using conventional food testing methods. The physical properties of the target milk powder include moisture content, ash content, particle size, color, and powder looseness; the chemical properties include protein content, fat content, and reducing sugar content, etc.

[0075] In one embodiment, the physicochemical properties of the irradiated target milk powder are analyzed to obtain key component parameters, including: the analysis of the physicochemical properties of the irradiated target milk powder to obtain moisture content, ash content, fat content and protein content.

[0076] In this embodiment, key component parameters refer to core quantitative indicators that are directly related to the irradiation effect and can characterize the milk powder matrix, selected from the physicochemical property analysis results of the milk powder. For example, high-fat whole milk powder and skim milk powder have different fat contents. High-fat whole milk powder has a fat content ≥26% (national standard for whole milk powder), while skim milk powder has a fat content ≤1.5% (national standard for skim milk powder).

[0077] For irradiated target milk powder, it is also necessary to extract characteristic parameters from its resonance spectrum. The resonance spectrum specifically refers to the characteristic spectrum obtained by an EPR spectrometer from irradiated target milk powder. It is a spectrum formed by the resonance signal generated by unpaired electrons (free radicals) induced by irradiation in milk powder under the action of a magnetic field. It is the core spectral carrier reflecting the characteristics of free radicals in milk powder after irradiation and is used to extract characteristic parameters such as peak shape, spectral splitting factor, and linewidth.

[0078] Peak shape represents the overall morphology and contour characteristics of the signal peaks in the resonance spectrum, including peak symmetry, number of peaks, peak height distribution, peak sharpness / bluntness, and the presence of shoulder peaks / extra peaks. Peak shape directly reflects the types, distribution, and interactions of irradiated free radicals in the milk powder, and is the most intuitive characteristic parameter of the resonance spectrum. The spectral splitting factor, also known as the g-factor, is a core characteristic constant of the resonance spectrum, characterizing the magnetic moment characteristics of free radicals resonating in a magnetic field. Its value is determined by the chemical environment and electron spin state of the free radicals in the milk powder; different types of irradiated free radicals correspond to specific g-factor values. The spectral splitting factor g value can be calculated using the formula g = 71.448ν / B, where ν is the microwave frequency of the device and B is the magnetic induction intensity. Linewidth, also known as resonance linewidth, refers to the magnetic field width corresponding to the half-height position of the signal peak in the resonance spectrum (commonly measured in Gaussian G or millitalas mT). It reflects the spin-lattice relaxation and spin-spin relaxation processes of free radicals in milk powder, and is directly related to the stability and lifetime of free radicals and the degree of binding of free radicals by the milk powder matrix. It is an important parameter for evaluating the free radical activity of milk powder after irradiation.

[0079] It is worth noting that the signal intensity of the irradiated target milk powder obtained in S101 and the characteristic parameters of the resonance spectrum of the irradiated target milk powder extracted in S102 are extracted simultaneously and are both obtained by an EPR detector.

[0080] S103: Determine the target matrix type of the target milk powder based on key component parameters and characteristic parameters, and query the dose-response slope curve and free radical decay rate constant corresponding to the target matrix type in the target milk powder matrix database.

[0081] In this embodiment, the target matrix type is high-fat whole milk powder, skim milk powder, low-glycemic milk powder, or whole milk powder with added sugar. The matrix type is determined by the fat content, protein / lactose ratio, and EPR characteristic parameters after irradiation of the milk powder.

[0082] Determining the target matrix type of a target milk powder based on key component parameters and characteristic parameters includes at least two implementation methods. One method involves selecting the corresponding target matrix type from a parameter database based on key component parameters and characteristic parameters. The parameter database contains pre-set threshold ranges for key component parameters and characteristic parameters for high-fat whole milk powder, skim milk powder, low-glycemic milk powder, and whole milk powder with added sugar. The measured parameters of the milk powder to be tested (target milk powder) are compared one by one; if all parameters fall within a certain matrix range, the milk powder is determined to be of the corresponding matrix type.

[0083] Secondly, the key component parameters and feature parameters are input into a classification model to obtain the target matrix type of the target milk powder. This classification model is obtained after training and testing on multiple historical milk powder samples. The classification model can be a neural network model such as a support vector machine or a random forest. Since the above model is a known model, and no improvements have been made to its structure in this embodiment, the details of the above model will not be elaborated further.

[0084] In this embodiment, the target milk powder matrix database refers to a pre-constructed dedicated data storage system for milk powder matrices, which stores complete associated data for four matrix types: high-fat whole milk powder, skim milk powder, low-glycemic milk powder, and whole milk powder with added sugar. Specifically, it includes: standard key component parameter thresholds for each matrix type, standard EPR characteristic parameter ranges, and dose-response slope curves and free radical decay rate constants unique to each matrix type. The database supports retrieving and matching core data by using "target matrix type" as the search condition.

[0085] In this embodiment, the dose-response slope curve represents the linear relationship between the irradiation dose value and the EPR signal intensity of the milk powder. The horizontal axis is the irradiation dose (unit kGy), and the vertical axis is the EPR signal intensity (peak height / peak area, unit au). The "slope" of the curve represents the rate of change of the EPR signal intensity of the base milk powder with the increase of the irradiation dose. The slope value and linear applicable range of each type of base have significant specific differences.

[0086] The dose-response slope curve model can be constructed through the following steps:

[0087] Representative milk powder samples were selected and subjected to gradient dose irradiation. To ensure the accuracy of the model, the irradiation dose points should cover the regulatory standard range; for example, values ​​such as 0 kGy, 1 kGy, 2 kGy, 4 kGy, 5 kGy, 6 kGy, or 8 kGy can be set to establish a more accurate curve. The signal intensity (S) of the samples at each dose point was detected using an electron paramagnetic resonance (EPR) spectrometer to establish an initial dose-response slope curve. The test report shows that the EPR signal intensity of the milk powder samples gradually increases with increasing irradiation dose.

[0088] The free radical decay rate constant represents the characteristic constant of the natural decay of unpaired electrons (free radicals) generated by irradiation under a specific type of milk powder matrix, characterizing the stability and decay law of free radicals. A larger value indicates that the free radicals in the milk powder matrix decay faster, and the signal intensity decreases more significantly over time. In this embodiment, each type of milk powder matrix corresponds to a free radical decay rate constant.

[0089] S104: Calculate the initial irradiation dose of the target milk powder based on the signal intensity and dose-response slope curve, determine the correction coefficient based on the storage time of the target milk powder from the irradiation date and the free radical decay rate constant, and correct the initial irradiation dose based on the correction coefficient to obtain the final irradiation dose of the target milk powder.

[0090] In this embodiment, the initial irradiation dose of the target milk powder is retrieved from the dose-response slope curve corresponding to the milk powder matrix type based on the signal intensity. The retrieval method is to determine the position of the signal intensity on the vertical axis of the curve, and then determine the value of the horizontal axis based on the position of the vertical axis. The value of the horizontal axis corresponds to the initial irradiation dose of the target milk powder.

[0091] In this embodiment, calculating the preliminary irradiation dose of the target milk powder based on signal intensity and dose-response slope curve further includes: constructing a fitting equation based on the dose-response slope curve corresponding to the target matrix type, inputting the signal intensity of the target milk powder into the fitting equation, and solving for the irradiation dose through inverse operation. In one embodiment, if the response is linear, a linear regression equation y=kx+b is constructed, where k represents the response slope value, x represents the irradiation dose, b represents a constant, and y represents the signal intensity. The signal intensity y0 of the target milk powder is input into the above formula to obtain the value of x0, which is the preliminary irradiation dose of the target milk powder.

[0092] In this embodiment, the storage time of the target milk powder from the irradiation date represents the actual storage time (unit: days) of the target milk powder from the time the irradiation is completed to the time the EPR signal is detected. This storage time, combined with the free radical decay rate constant, can help to understand the decay amount of free radicals in the target milk powder and realize the correction of the initial irradiation dose.

[0093] As can be seen from the above, the embodiments of this application improve the accuracy of detection. Traditional EPR methods can only qualitatively determine whether milk powder has been irradiated, lacking precise quantitative capabilities. The method provided in this application, however, obtains the signal intensity of the target milk powder, analyzes its physicochemical properties to obtain key component parameters, extracts resonance spectrum characteristic parameters to determine the target matrix type, and then queries the database for the corresponding dose-response slope curve and free radical decay rate constant. Using these parameters, a preliminary irradiation dose is first calculated, and then a correction coefficient is determined based on storage time and the free radical decay rate constant to obtain the final irradiation dose. This multi-parameter comprehensive analysis fully considers the different free radical signal responses caused by differences in the composition of different brands and formula milk powders, effectively avoiding the huge errors caused by using a single calibration model, greatly improving the accuracy of irradiation dose detection, and meeting the standardized regulatory requirements for food irradiation doses.

[0094] On the other hand, the embodiments of this application enhance the practicality of the detection. The method provided in the embodiments of this application constructs a target milk powder matrix database, storing the dose-response slope curves and free radical decay rate constants corresponding to different matrix types. In actual detection, the matrix type only needs to be determined according to the parameters of the target milk powder to quickly obtain relevant parameters for calculation. The operation is simple and efficient, and it can adapt to the quantitative detection of irradiation of different brands and formula milk powders, providing a practical and reliable technical means for the accurate detection of milk powder irradiation dose.

[0095] In one embodiment of this application, before obtaining the signal intensity of the target milk powder, the quantitative detection method for milk powder irradiation further includes:

[0096] Preprocessing and signal enhancement steps: The target milk powder sample was placed in an environment with constant temperature and humidity for at least 24 hours to achieve temperature and humidity equilibration; then, it was pre-irradiated with microwaves of a specific frequency for a short time and with low power to stabilize the background free radical signal present in the sample; finally, the sample signal and the reference blank signal were acquired simultaneously using dual-channel differential detection technology, and the background and environmental noise were directly subtracted from the original spectrum to obtain a pure electron spin resonance signal induced only by the target irradiation.

[0097] By adding a control sample, this embodiment can eliminate some environmental interference, making the signal intensity of the target milk powder more accurate.

[0098] In one embodiment of this application, the quantitative detection method for milk powder irradiation further includes:

[0099] The distribution parameters of free radicals in the target milk powder are obtained, including the distribution uniformity index and the fine spectral features.

[0100] The accuracy of the target matrix type is judged based on the distribution parameters of free radicals to obtain the judgment result;

[0101] If the judgment result is accurate, then the matrix type of the target milk powder is determined to be the target matrix type;

[0102] If the judgment result is inaccurate, the target matrix type is updated based on the distribution parameters of free radicals to obtain the updated target matrix type.

[0103] The microscopic distribution characteristics of free radicals in irradiated milk powder are matrix-specific. In this embodiment, by detecting the free radical distribution parameters, the determination results of the milk powder matrix type are verified / corrected, avoiding the deviation in quantitative detection of irradiation dose due to misjudgment of matrix type, and finally achieving accurate quantitative detection of milk powder irradiation dose.

[0104] In this embodiment, the distribution parameters include the distribution uniformity index and the fine spectral characteristics. The distribution uniformity index is a numerical indicator that quantifies the degree of distribution and dispersion of irradiated free radicals in the microscopic region of milk powder, and is the basis for judging whether free radicals are evenly distributed. After irradiation, the matrix components (fat, protein, and lactose) of milk powder have different abilities to capture free radicals. For example, whole milk powder, due to its high fat content, allows free radicals to easily aggregate around fat particles, resulting in a low uniformity index; skim milk powder has a more uniform distribution of free radicals, resulting in a high uniformity index.

[0105] Fine spectral features include spectral asymmetry factor, hyperfine splitting constant, and relative intensity of satellite peaks.

[0106] The spectral asymmetry factor is a quantitative indicator of the degree of asymmetry between the left and right sides of an EPR spectral peak. It is typically obtained by calculating the ratio of signal intensity to the left and right sides of the spectral line, the half-width at half-maximum (WHM) ratio, or the symmetry coefficient of the fitted curve. The components of milk powder matrix, such as fat, protein, and lactose, have varying abilities to bind free radicals. For example, in whole milk powder, the hydrophobic environment of fat molecules causes uneven electron cloud distribution around free radicals, resulting in significant asymmetry in the EPR spectrum and a larger asymmetry factor. In contrast, in skim milk powder, free radicals are mainly distributed in the hydrophilic environment of lactose / protein, resulting in a relatively uniform electron cloud distribution, high spectral symmetry, and a smaller asymmetry factor.

[0107] The hyperfine splitting constant is a parameter representing the spacing between spectral lines in an EPR spectrum. Essentially, it represents the distance between unpaired electrons of a free radical and adjacent magnetic nuclei (such as ¹H, ¹⁰). 4 The strength of spin-spin coupling between N and N. Differences in the composition of different milk powder matrices can alter the type and number of atomic nuclei adjacent to free radicals. For example, nitrogenous nutrients added to formula milk powder can cause a large number of ¹ atoms to exist around free radicals. 4 The N-core has strong spin coupling, a large hyperfine splitting constant, and its spectral lines split into multiple characteristic peaks; while the free radicals in pure milk powder are mostly surrounded by ¹H nuclei, which have relatively weak coupling, a small splitting constant, and fewer spectral splitting peaks.

[0108] The relative intensity of satellite peaks refers to the ratio of the signal intensity of secondary characteristic peaks (satellite peaks) other than the main peak in an EPR spectrum to the signal intensity of the main peak. Satellite peaks are generated due to the isotopic effect of free radicals or their special spatial conformations. The purity of the milk powder matrix and the types of additives affect the appearance and intensity of satellite peaks. For example, in compound milk powders with added probiotics and prebiotics, the special conformation of free radicals induces characteristic satellite peaks with relatively high intensity; while in pure milk powder, the free radical conformation is uniform, resulting in weak or even absent satellite peak signals with relative intensities approaching zero.

[0109] This embodiment, by acquiring the free radical distribution uniformity index and fine spectral features, can verify and dynamically update the target matrix type of milk powder, avoid quantitative irradiation deviation caused by matrix type misjudgment, and improve the accuracy and reliability of quantitative detection of milk powder irradiation.

[0110] In one embodiment of this application, the accuracy of determining the target matrix type based on the distribution parameters of free radicals is used to obtain a determination result, including:

[0111] The distribution parameters of free radicals of the target milk powder are compared with the benchmark range of the corresponding parameters of the target matrix type stored in the target milk powder matrix database;

[0112] Calculate the normalized offset of each distribution parameter value of the target milk powder relative to its corresponding benchmark range center value, and perform a weighted summation based on the preset weight of each distribution parameter value in the type discrimination to obtain the comprehensive deviation score;

[0113] If the overall deviation score is less than or equal to the first threshold, the judgment result is determined to be accurate.

[0114] If the overall deviation score is greater than the first threshold and less than or equal to the second threshold, the judgment result is determined to be inaccurate, and a prompt signal requesting review is generated; wherein the first threshold is less than the second threshold.

[0115] If the overall deviation score is greater than the second threshold, the judgment result is determined to be inaccurate.

[0116] This embodiment first retrieves a preset target milk powder matrix database and compares the measured values ​​of the target milk powder free radical distribution uniformity index, fractal dimension, and radial distribution function with the reference ranges of each parameter corresponding to the target matrix type in the database. Next, it calculates the normalized offset of each distribution parameter value relative to the center value of its reference range to eliminate dimensional differences between parameters. Then, it performs a weighted summation according to the preset weights of each parameter in matrix type discrimination to obtain a comprehensive deviation score that comprehensively reflects the overall deviation of the parameters. Finally, it completes the judgment based on a preset grading threshold. The reference ranges of each parameter in this embodiment can be obtained through experimental determination.

[0117] This embodiment obtains a comprehensive deviation score through parameter benchmark comparison, normalized offset calculation, and weighted summation. Then, it combines this with a dual-threshold grading system to determine the accuracy of the target matrix type. This achieves a quantitative assessment of matrix type matching while avoiding the limitations and dimensional interference of single-parameter comparison, thus improving the accuracy of the judgment score. Furthermore, this embodiment sets a request-for-review threshold range, balancing detection efficiency with the rigor of judgment in special scenarios, effectively avoiding misjudgments of matrix type and providing assurance for subsequent quantitative detection of milk powder irradiation, further improving the reliability and stability of the detection results.

[0118] In one embodiment of this application, a correction coefficient is determined based on the storage time of the target milk powder from the irradiation date and the free radical decay rate constant, including:

[0119] Based on the storage time of the target milk powder from the date of irradiation and the free radical decay rate constant, the correction coefficient is determined using the following correction formula:

[0120] , where A is the correction coefficient, λ is the free radical decay rate constant, and t is the storage time of the target milk powder from the irradiation date.

[0121] In this embodiment, λ is the free radical decay rate constant, and λ is different for each matrix type. Correcting the initial irradiation dose based on a correction factor can compensate for signal intensity loss due to time decay. The corrected irradiation dose... for:

[0122]

[0123] in, This is the initial irradiation dose.

[0124] This embodiment determines the correction coefficient based on the storage time of the target milk powder from the irradiation date, the free radical decay rate constant, and the following correction formula. The initial irradiation dose is then corrected using the correction coefficient. This can specifically compensate for the detection loss caused by the decay of free radical signal intensity with storage time, effectively avoid dose measurement deviations caused by storage time and matrix differences, and improve the timeliness and accuracy of quantitative detection of milk powder irradiation dose.

[0125] In one embodiment of this application, the milk powder matrix database includes multiple dose-response slope curves, each dose-response slope curve corresponding to a milk powder matrix; the generation process of each dose-response slope curve includes:

[0126] For each milk powder sample: determine multiple gradient dose values ​​for the milk powder sample, irradiate the milk powder sample based on the multiple gradient dose values, and obtain the signal intensity of the milk powder sample at each gradient dose value;

[0127] The mean value of the signal intensity of multiple milk powder samples under the same gradient dose value is calculated to obtain the dose-response slope curve corresponding to the milk powder matrix.

[0128] In this embodiment, for each milk powder sample: multiple gradient dose values ​​are determined for the milk powder sample, for example, the irradiation dose value can be set to values ​​such as 0 kGy, 1 kGy, 2 kGy, 4 kGy, 5 kGy, 6 kGy, or 8 kGy. The milk powder sample is irradiated based on multiple gradient dose values ​​to obtain the signal intensity of the milk powder sample at each gradient dose value. Then, the signal intensity of the milk powder samples at the same gradient dose value is averaged to obtain the dose-response slope curve corresponding to the milk powder matrix.

[0129] The sample size of this application embodiment is large enough that the dose-response slope curve determined based on this embodiment is accurate and can be used for quantitative detection of the irradiation of the milk powder to be tested.

[0130] In one embodiment of this application, to address the problem that existing detection models are fixed and cannot adapt to batch variations in milk powder and minor environmental changes, and to further improve detection accuracy, a long short-term memory network is introduced to perform iterative parameter updates. The specific steps are as follows:

[0131] The final irradiation dose is obtained based on the target milk powder matrix database. The final irradiation dose, the actual irradiation dose feedback value, the EPR spectrum features of the target milk powder, environmental data, and storage time are input into a pre-trained long short-term memory network to obtain the output parameter increment. , Let be the slope increment of the response of the i-th matrix. This represents the increment of the free radical decay rate constant for the i-th matrix. The output parameter increment is used to update the target milk powder matrix database.

[0132] In this embodiment, the multi-dimensional inputs, including "final irradiation dose, actual irradiation dose feedback value, EPR spectrum features, environmental data, and storage time," are first converted into a time-series feature vector that can be processed by LSTM.

[0133] Dimensionality reduction encoding is performed on the EPR spectral features (usually high-dimensional sequences); scalar / low-dimensional data such as "dose, environmental data, and storage time" are concatenated with the EPR-encoded features to form a single-time-step feature vector (if the input is a time series, it will be expanded into a multi-step vector along the time dimension).

[0134] Secondly, LSTM uses its own "memory cells" and "input gate / forget gate / output gate" to complete feature extraction and state transfer based on the input temporal feature vector.

[0135] Among them, the forget gate is used to determine whether to retain the historical state of milk powder irradiation parameters previously stored by LSTM;

[0136] The input gate is used to encode the currently input information such as "dose, EPR characteristics, environmental data and storage time" into a new state and store it in the memory unit;

[0137] The output gate is used to combine the state of the current memory cell to output the hidden state of the current time step. This hidden state integrates "historical information + current input features" for subsequent calculations.

[0138] Finally, the hidden state output by the LSTM is passed to the subsequent fully connected layer. The fully connected layer maps the hidden state to the output parameter increment and outputs the final result through the activation function.

[0139] In this embodiment, the quantitative detection method for milk powder irradiation further includes:

[0140] Based on historical EPR spectral features, environmental data, time-dimensional data, and historical detection bias sequences, a time-series feature vector is constructed. This time-series feature vector is then input into a long short-term memory network to train the model, resulting in a trained long short-term memory network.

[0141] Furthermore, in response to receiving the actual irradiation dose feedback value, the EPR spectrum features of the milk powder corresponding to the actual irradiation dose feedback value, environmental data, time dimension data, and predicted dose are input into the trained long short-term memory network to obtain the output parameter increment. The output parameter increment includes the response slope increment and the free radical decay rate constant increment. The loss function of the long short-term memory network is constructed based on the predicted dose and the actual irradiation dose feedback value.

[0142] A new response slope is obtained based on the response slope increment, and a new free radical decay rate constant is obtained based on the free radical decay rate constant increment.

[0143] When the loss function value is less than the preset error threshold, the model training is stopped and the model parameters of the new long short-term memory network are obtained; the feedback value of the actual radiation dose received in the future is calculated based on the model parameters of the new long short-term memory network.

[0144] In one embodiment, the quantitative detection method for milk powder irradiation further includes:

[0145] The parameters in the target milk powder matrix database are updated based on the incremental output parameters of the new long short-term memory network, resulting in a new response slope and a new free radical decay rate constant.

[0146] In this embodiment, the Long Short-Term Memory network first needs to be trained. The training data consists of historical detection bias sequences and the corresponding EPR spectral features, environmental data, time dimension data, and other parameters.

[0147] EPR spectral features include the peak shape eigenvector, spectral splitting factor g, and linewidth W of the target milk powder;

[0148] Environmental data includes the temperature (T) and humidity (H) of the storage and monitoring environment;

[0149] Time-dimensional data includes storage time t;

[0150] The historical detection deviation sequence includes the calculated value (final irradiation dose) and the actual value (actual irradiation dose feedback value) from N historical detections. In this embodiment, the parameter update process in the target milk powder matrix database is only triggered when the actual irradiation dose feedback value is received. This is because receiving the actual irradiation dose feedback value indicates that the calculated data may have a deviation, and the calculated value needs to be corrected in a timely manner.

[0151] In this embodiment, the model parameters of the Long Short-Term Memory (LSTM) network and the parameters in the target milk powder matrix database can be updated online. The construction method of the LTM network loss function in this embodiment differs from existing technologies; the loss function in this embodiment is Loss = (Dreal - Dpred). 2 Where Dreal is the actual irradiation dose feedback value and Dpred is the predicted dose. The model parameters are updated by minimizing the loss function value.

[0152] In this embodiment, the output parameter increments include the response slope increment and the free radical decay rate constant increment. The response slope increment is used to compensate for the response sensitivity drift caused by minor fluctuations in the composition of milk powder batches, and the free radical decay rate constant increment is used to compensate for changes in the free radical decay rate caused by seasonal temperature and humidity variations. The parameters in the target milk powder matrix database are updated based on the output parameter increments, using the following update formula:

[0153] .

[0154] The updated response slope, The slope of the response before the update. In response to the slope increment, This is the updated free radical decay rate constant. The free radical decay rate constant before the update. This represents the increment of the free radical decay rate constant. , To prevent drastic parameter fluctuations when updating the step size or learning rate, you can set it based on experience.

[0155] In one embodiment of this application, the overall process for quantitative detection of milk powder irradiation is as follows:

[0156] Step 1, Sample Preparation and Testing:

[0157] Weigh approximately 65 mg of the milk powder sample to be tested and place it into a 4 mm EPR paramagnetic test tube.

[0158] Set the temperature to room temperature and set the EPR spectrometer parameters, such as: microwave frequency 9.83379 GHz, microwave power 1.00 mW, modulation amplitude 2 Gauss, and receiver gain 303.

[0159] The EPR signal intensity (S) of the sample was obtained by testing on the machine.

[0160] Step 2, Matrix Classification and Parameter Calling:

[0161] Based on the method described in the above embodiments, the fat content of the milk powder sample was determined to be 28%, and the ash content was 2.5%.

[0162] Analysis of its EPR spectrum yielded a g value of 2.001 and a linewidth of 0.6 mT.

[0163] By analyzing the above parameters, the milk powder sample was classified as a "high-fat milk powder" matrix, and the dose-response slope curve and free radical decay rate constant corresponding to this matrix were retrieved from the database.

[0164] Step 3, Dosage Calculation and Dynamic Correction:

[0165] The initial irradiation dose D0 is calculated based on the measured signal intensity S and the slope of the call.

[0166] According to the inquiry, the sample had been stored at 25°C for 60 days (t=60).

[0167] Applying dynamic correction formula The final corrected irradiation dose is then calculated.

[0168] Step 4, Optimization:

[0169] If subsequent tracing reveals that the actual irradiation dose of the sample was 4.5 kGy, while the method's detection result is 4.7 kGy, the operator can report the true value back to the system. Upon receiving this feedback, the system optimization engine automatically learns the error and fine-tunes the slope and free radical decay rate constant under the "high-fat milk powder" matrix category, making the system more accurate when detecting similar samples in the future.

[0170] Based on the same inventive concept, this application also provides a milk powder irradiation quantitative detection device for implementing the above-mentioned milk powder irradiation quantitative detection method. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations of one or more milk powder irradiation quantitative detection device embodiments provided below can be found in the limitations of the milk powder irradiation quantitative detection method described above, and will not be repeated here.

[0171] This application provides a device for quantitative detection of milk powder irradiation, such as... Figure 2 As shown, the milk powder irradiation quantitative detection device 20 includes: a signal acquisition module 21, a feature extraction module 22, a data calculation module 23, and a data correction module 24.

[0172] Among them, the signal acquisition module 21 is used to acquire the signal intensity of the irradiated target milk powder. The signal intensity is obtained by detecting the irradiated target milk powder using an electron paramagnetic resonance spectrometer.

[0173] The feature extraction module 22 is used to analyze the physicochemical properties of the irradiated target milk powder to obtain key component parameters and extract feature parameters from the resonance spectrum of the irradiated target milk powder. The feature parameters include peak shape, spectral splitting factor and linewidth.

[0174] Data calculation module 23 is used to determine the target matrix type of the target milk powder based on key component parameters and characteristic parameters, and to query the dose-response slope curve and free radical decay rate constant corresponding to the target matrix type in the target milk powder matrix database;

[0175] The data correction module 24 is used to calculate the initial irradiation dose of the target milk powder based on the signal intensity and dose-response slope curve, determine the correction coefficient based on the storage time of the target milk powder from the irradiation date and the free radical decay rate constant, and correct the initial irradiation dose based on the correction coefficient to obtain the final irradiation dose of the target milk powder.

[0176] The milk powder irradiation quantitative detection device 20 also includes a judgment module;

[0177] The judgment module is used to: obtain the distribution parameters of free radicals in the target milk powder, including the distribution uniformity index and the fine spectral features;

[0178] The accuracy of the target matrix type is judged based on the distribution parameters of free radicals to obtain the judgment result;

[0179] If the judgment result is accurate, then the matrix type of the target milk powder is determined to be the target matrix type;

[0180] If the judgment result is inaccurate, the target matrix type is updated based on the distribution parameters of free radicals to obtain the updated target matrix type.

[0181] In one embodiment of this application, the determination module is specifically used for:

[0182] The distribution parameters of free radicals of the target milk powder are compared with the benchmark range of the corresponding parameters of the target matrix type stored in the target milk powder matrix database;

[0183] Calculate the normalized offset of each distribution parameter value of the target milk powder relative to its corresponding benchmark range center value, and perform a weighted summation based on the preset weight of each distribution parameter value in the type discrimination to obtain the comprehensive deviation score;

[0184] If the overall deviation score is less than or equal to the first threshold, the judgment result is determined to be accurate.

[0185] If the overall deviation score is greater than the first threshold and less than or equal to the second threshold, the judgment result is determined to be inaccurate, and a prompt signal requesting review is generated; wherein the first threshold is less than the second threshold.

[0186] If the overall deviation score is greater than the second threshold, the judgment result is determined to be inaccurate.

[0187] In one embodiment of this application, the data correction module 24 is specifically used for:

[0188] Based on the storage time of the target milk powder from the date of irradiation and the free radical decay rate constant, the correction coefficient is determined using the following correction formula:

[0189] , where A is the correction coefficient, λ is the free radical decay rate constant, and t is the storage time of the target milk powder from the irradiation date.

[0190] In one embodiment of this application, the data calculation module 23, when determining the generation of each dose-response slope curve, is specifically used for:

[0191] For each milk powder sample: determine multiple gradient dose values ​​for the milk powder sample, irradiate the milk powder sample based on the multiple gradient dose values, and obtain the signal intensity of the milk powder sample at each gradient dose value;

[0192] The mean value of the signal intensity of multiple milk powder samples under the same gradient dose value is calculated to obtain the dose-response slope curve corresponding to the milk powder matrix.

[0193] In one embodiment of this application, the feature extraction module 22 is specifically used for:

[0194] The physicochemical properties of the irradiated target milk powder were analyzed to obtain the moisture content, ash content, fat content, and protein content.

[0195] In one embodiment of this application, the data calculation module 23 is specifically used for:

[0196] The target matrix type is selected from the parameter database based on key component parameters and characteristic parameters.

[0197] See Figure 3 , Figure 3This is a schematic block diagram of an electronic device provided according to an embodiment of this application. Figure 3 The electronic device 300 in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 store computer programs, including program instructions. The processors 301 execute the program instructions stored in the memories 304. Specifically, the processors 301 are configured to invoke the program instructions to perform the functions of each module / unit in the above-described device embodiments, for example... Figure 2 The functions of the signal acquisition module 21, feature extraction module 22, data calculation module 23, and data correction module 24 are shown.

[0198] It should be understood that, in the embodiments of this application, the processor 301 may be a central processing unit (CPU), but it may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0199] Input device 302 may include a touchpad, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint orientation information), a microphone, etc., and output device 303 may include a display (LCD, etc.), a speaker, etc.

[0200] The memory 304 may include read-only memory and random access memory, and provides instructions and data to the processor 301.

[0201] In specific implementations, the processor 301, input device 302, and output device 303 described in the embodiments of this application can execute the implementation method described in the quantitative detection method of milk powder irradiation provided in the embodiments of this application, or they can execute the implementation method of the electronic device described in the embodiments of this application, which will not be repeated here.

[0202] In another embodiment of this application, a computer-readable storage medium is provided. This computer-readable storage medium stores a computer program, which includes program instructions. When executed by a processor, the program instructions implement all or part of the processes in the methods described above. Alternatively, the computer program can instruct related hardware to complete the process. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0203] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the foregoing embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., provided on the electronic device. Furthermore, the computer-readable storage medium can include both internal and external storage units of the electronic device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0204] Those skilled in the art will recognize that the modules / units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0205] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the electronic devices and units described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0206] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of modules / units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules, units, or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces or modules / units, or it may be an electrical, mechanical, or other form of connection.

[0207] The modules / units described as separate components may or may not be physically separate. Similarly, the components shown as modules / units may or may not be physical modules / units; they may be located in one place or distributed across multiple network modules / units. Some or all of the modules / units can be selected to achieve the purpose of the embodiments of this application, depending on actual needs.

[0208] Furthermore, the functional modules / units in the various embodiments of this application can be integrated into one processing module / unit, or each module / unit can exist physically separately, or two or more modules / units can be integrated into one module / unit. The integrated modules / units described above can be implemented in hardware or in the form of software functional modules / units.

[0209] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for quantitative detection of milk powder irradiation, characterized in that, include: The signal intensity of the irradiated target milk powder is obtained by detecting the irradiated target milk powder using an electron paramagnetic resonance spectrometer. The physicochemical properties of the irradiated target milk powder were analyzed to obtain key component parameters, and characteristic parameters were extracted from the resonance spectrum of the irradiated target milk powder, including peak shape, spectral splitting factor, and linewidth. Based on the key component parameters and the characteristic parameters, the target matrix type of the target milk powder is determined. The dose-response slope curve and free radical decay rate constant corresponding to the target matrix type are queried in the target milk powder matrix database. The milk powder matrix database includes multiple dose-response slope curves, and each dose-response slope curve corresponds to a milk powder matrix. The generation process of each dose-response slope curve includes: For each milk powder sample: determine multiple gradient dose values ​​for the milk powder sample, irradiate the milk powder sample based on the multiple gradient dose values, and obtain the signal intensity of the milk powder sample at each gradient dose value; The mean value of the signal intensity of multiple milk powder samples under the same gradient dose value is calculated to obtain the dose-response slope curve corresponding to the milk powder matrix; The initial irradiation dose of the target milk powder is calculated based on the signal intensity and the dose-response slope curve, and a correction coefficient is determined based on the storage time of the target milk powder from the irradiation date and the free radical decay rate constant. The initial irradiation dose is corrected based on the correction factor to obtain the final irradiation dose of the target milk powder; The quantitative detection method for milk powder irradiation also includes: The distribution parameters of free radicals in the target milk powder are obtained, including the distribution uniformity index and the fine spectral features; The accuracy of the target matrix type is judged based on the distribution parameters of the free radicals to obtain the judgment result; If the judgment result is accurate, then the matrix type of the target milk powder is determined to be the target matrix type; If the judgment result is inaccurate, the target matrix type is updated based on the distribution parameters of the free radicals to obtain the updated target matrix type.

2. The method for quantitative detection of milk powder irradiation as described in claim 1, characterized in that, The determination of the accuracy of the target matrix type based on the distribution parameters of the free radicals to obtain the determination result includes: The distribution parameters of free radicals of the target milk powder are compared with the benchmark range of the corresponding parameters of the target matrix type stored in the target milk powder matrix database; Calculate the normalized offset of each distribution parameter value of the target milk powder relative to its corresponding reference range center value, and perform a weighted summation according to the preset weight of each distribution parameter value in the type discrimination to obtain the comprehensive deviation score; If the overall deviation score is less than or equal to the first threshold, the judgment result is determined to be accurate. If the overall deviation score is greater than the first threshold and less than or equal to the second threshold, the judgment result is determined to be inaccurate, and a prompt signal requesting review is generated; wherein the first threshold is less than the second threshold; If the overall deviation score is greater than the second threshold, the judgment result is determined to be inaccurate.

3. The quantitative detection method for milk powder irradiation as described in claim 1, characterized in that, The determination of the correction coefficient based on the storage time of the target milk powder from the irradiation date and the free radical decay rate constant includes: Based on the storage time of the target milk powder from the date of irradiation and the free radical decay rate constant, the correction coefficient is determined using the following correction formula: , where A is the correction coefficient, λ is the free radical decay rate constant, and t is the storage time of the target milk powder from the irradiation date.

4. The method for quantitative detection of milk powder irradiation as described in claim 1, characterized in that, The analysis of the physicochemical properties of the irradiated target milk powder yielded key component parameters, including: The physicochemical properties of the irradiated target milk powder were analyzed to obtain its moisture content, ash content, fat content, and protein content.

5. The quantitative detection method for milk powder irradiation as described in claim 1, characterized in that, The determination of the target matrix type of the target milk powder based on the key component parameters and the characteristic parameters includes: Based on the key component parameters and the characteristic parameters, the corresponding target matrix type is selected from the parameter database, which includes multiple milk powder matrix types, the values ​​of key component parameters and characteristic parameters corresponding to each milk powder matrix type.

6. A quantitative detection device for milk powder irradiation based on the quantitative detection method for milk powder irradiation as described in claim 1, characterized in that, include: The signal acquisition module is used to acquire the signal intensity of the irradiated target milk powder, wherein the signal intensity is obtained by detecting the irradiated target milk powder using an electron paramagnetic resonance spectrometer. The feature extraction module is used to analyze the physicochemical properties of the irradiated target milk powder to obtain key component parameters, and to extract feature parameters from the resonance spectrum of the irradiated target milk powder, including peak shape, spectral splitting factor and linewidth. The data calculation module is used to determine the target matrix type of the target milk powder based on the key component parameters and the characteristic parameters, and to query the dose-response slope curve and free radical decay rate constant corresponding to the target matrix type in the target milk powder matrix database; the milk powder matrix database includes multiple dose-response slope curves, and each dose-response slope curve corresponds to a milk powder matrix; The data calculation module is specifically used to determine the generation of each dose-response slope curve when: For each milk powder sample: determine multiple gradient dose values ​​for the milk powder sample, irradiate the milk powder sample based on the multiple gradient dose values, and obtain the signal intensity of the milk powder sample at each gradient dose value; The mean value of the signal intensity of multiple milk powder samples under the same gradient dose value is calculated to obtain the dose-response slope curve corresponding to the milk powder matrix; The data correction module is used to calculate the initial irradiation dose of the target milk powder based on the signal intensity and the dose-response slope curve, and to determine the correction coefficient based on the storage time of the target milk powder from the irradiation date and the free radical decay rate constant. The initial irradiation dose is corrected based on the correction factor to obtain the final irradiation dose of the target milk powder.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 5.

Images