A method for improving the shelf-life stability of infant formula milk powder

Abstract

This invention discloses a method for improving the shelf-life stability of infant formula milk powder, relating to the field of milk powder production management technology. The method includes: setting key nodes and critical trigger values ​​for the activation of a synergistic deterioration positive feedback loop; selecting corresponding targeted blocking functional components and integrating all functional components to obtain a multi-target synergistic blocking milk powder product; dividing temperature control zones into graded ranges and setting corresponding control rules; synchronously collecting cumulative temperature and humidity exposure data across the entire supply chain; establishing sample detection rules adapted to the finished milk powder product; fitting the synergistic amplification coefficients of various deterioration reactions to construct a dynamic shelf-life prediction model; and establishing graded early warning systems based on sample detection results and model outputs to achieve full life-cycle management of the finished milk powder product. This invention provides a clear targeted design basis for improving the shelf-life stability of milk powder, solving the problems of existing technologies that often focus on a single deterioration reaction and lack targeted regulation.

Description

Technical Field

[0001] This invention relates to the field of milk powder production management technology, and in particular to a method for improving the shelf-life stability of infant formula milk powder. Background Technology

[0002] Infant formula is the only acceptable nutritional alternative for infants when adequate breastfeeding is not possible. Through a scientifically formulated blend of protein, fat, lactose, minerals, and trace elements, it provides essential energy, protective factors, and developmental support for infant growth and development. With the increasing prevalence of early breastfeeding, the market demand for infant formula continues to rise. Its product quality, nutritional stability, and safety have become core concerns for manufacturers, regulatory authorities, and consumers alike.

[0003] The shelf life of infant formula milk powder is divided into the regular shelf life in the unopened, sealed state, and a conservative 28-day secondary shelf life generally labeled by manufacturers after opening. Currently, most of the industry's technological research and quality control focuses on ensuring the shelf life in the unopened state, while existing research and applications are significantly insufficient regarding the quality deterioration mechanisms and stability improvement technologies within the secondary shelf life after opening.

[0004] In existing technologies, methods to improve the shelf-life stability of infant formula milk powder mainly fall into three categories: First, at the formulation optimization level, unsaturated fatty acids are processed through microencapsulation technology, or general antioxidant components such as vitamin E are added within the limits permitted by national standards to inhibit lipid oxidation. Second, at the process control level, the content of free fat on the surface of milk powder is reduced by optimizing the production process, or general guidelines for storage temperature and duration are provided to reduce the impact of the external environment on the quality of milk powder. Third, at the shelf-life prediction level, a predictive model for the shelf life of milk powder is established by combining a single physicochemical indicator or multiple indicators in parallel with the Arrhenius equation.

[0005] Current academic research techniques have several shortcomings. For example, there is a systematic gap in our understanding of the deterioration mechanism of milk powder during its second shelf life after opening. As early as 2012, a core Chinese industry review published in *Food and Fermentation Industries* clearly pointed out that "the factors causing quality deterioration in milk powder storage and processing are mainly three types of chemical reactions: lipid oxidation, Maillard reaction, and lactose crystallization. These reactions interact with each other, but the mechanism is not yet clear." Current technology has never revealed the closed-loop synergistic positive feedback loop between these three deterioration reactions, nor has it realized that this loop is the core driving force behind the rapid deterioration of milk powder quality after opening, resulting in all existing control measures being ineffective. Targeted, individual treatments for single reactions cannot fundamentally block the core drivers of degradation. Furthermore, existing regulatory methods are incomplete and cannot simultaneously inhibit the three major degradation reactions, resulting in a fundamental bottleneck in improving efficacy. Single antioxidant additions can only partially inhibit primary lipid oxidation, failing to block the transmission of lipid oxidation to the other two reactions. In fact, the compensatory acceleration of the Maillard reaction may exacerbate browning and flavor degradation in milk powder. Single protein modification methods can only reduce protein oxidation but cannot block the initiation of lipid oxidation, ultimately leading to the restart of protein oxidation and the Maillard reaction due to the continuous generation of free radicals.

[0006] Therefore, existing technologies and academic research have not revealed the positive feedback loop mechanism of synergistic deterioration during the secondary shelf life of infant formula milk powder, nor have they designed a systematic closed-loop blocking scheme for this loop, thus failing to fundamentally solve the industry pain point of insufficient shelf-life stability of milk powder. Summary of the Invention

[0007] Therefore, it is necessary to provide a method for improving the shelf-life stability of infant formula milk powder to address the aforementioned technical problems.

[0008] In a first aspect, the present invention provides a method for improving the shelf-life stability of infant formula milk powder, comprising:

[0009] S1. A synergistic deterioration positive feedback loop among various deterioration reactions during the storage of infant formula milk powder was established through accelerated storage experiments. Key nodes of the synergistic deterioration positive feedback loop and critical trigger values ​​for its activation were set. A graded quantitative threshold was established based on the critical trigger values.

[0010] S2. Based on the key nodes of the synergistic deterioration positive feedback loop, select the corresponding targeted blocking functional components, and combine and integrate all functional components to obtain a multi-target synergistic blocking milk powder product.

[0011] S3. Based on the critical trigger value of the collaborative deterioration positive feedback loop, combined with the deterioration kinetics of finished milk powder products, divide the temperature control range into graded ranges and set corresponding control rules; simultaneously collect the cumulative exposure data of temperature and humidity throughout the entire chain, and dynamically adjust the control rules;

[0012] S4. Based on graded and quantified thresholds, establish sample testing rules suitable for finished milk powder products; and combine the synergistic deterioration positive feedback loop, deterioration kinetic characteristics and exposure data to fit the synergistic amplification coefficients of various deterioration reactions and construct a dynamic prediction model for shelf life; establish graded early warning based on sample testing results and model output to achieve full life cycle management of finished milk powder products.

[0013] Furthermore, a synergistic deterioration positive feedback loop among various deterioration reactions during the storage of infant formula milk powder was established through accelerated storage experiments. Key nodes and critical trigger values ​​for the activation of this synergistic deterioration positive feedback loop were defined. Based on these critical trigger values, a graded quantitative threshold system was established, including:

[0014] S11. Using infant formula milk powder as a unified basic sample, with fixed relative humidity and non-temperature variable parameters, multiple sets of controllable parallel storage experiments were configured, and full-cycle change data of various degradation reactions were collected simultaneously. The interaction relationship between different degradation reactions was quantitatively verified through statistical analysis. Degradation reactions include lipid oxidation, protein oxidation and Maillard reaction.

[0015] S12. Based on the quantitative verification results, sort out the transmission path and mutual promotion logic between different degradation reactions, divide the complete process of synergistic degradation from the beginning to accelerated runaway, and select the key nodes of the synergistic degradation positive feedback loop, including the initiation trigger node, the transmission amplification node and the accelerated runaway node.

[0016] S13. Collect full-cycle kinetic data of the deterioration characteristic indicators of formula milk powder under different storage conditions as storage time changes, and set the critical trigger value for the activation of the synergistic deterioration positive feedback loop by analyzing the deterioration rate change law of the synergistic deterioration positive feedback loop.

[0017] S14. Based on the critical trigger value, and combined with the different startup states of the loop, the hierarchical quantization thresholds for the three states of the corresponding loop—not started, pre-started, and started—are obtained.

[0018] Furthermore, full-cycle kinetic data of the degradation characteristic indicators of formula milk powder under different storage conditions were collected, and the degradation rate variation law of the synergistic degradation positive feedback loop was analyzed. The critical trigger values ​​for the activation of the synergistic degradation positive feedback loop were set as follows:

[0019] S131. Obtain quantitative detection data of deterioration characteristic indicators of formula milk powder under different temperature gradients and different storage times in the full cycle change data, and form full cycle kinetic data of deterioration characteristic indicators changing with storage time and storage temperature.

[0020] S132. Nonlinear piecewise regression fitting is performed on the full-cycle dynamic data using nonlinear regression analysis software. Piecewise function equations of the deterioration characteristic index as a function of storage time are obtained under different temperature conditions, and the deterioration rate variation law of different stages is divided.

[0021] S133. Based on the piecewise function equation obtained by nonlinear piecewise regression fitting, select the abrupt change point in the change of degradation rate that causes an exponential leap, and determine the abrupt change point as the critical inflection point for the start of the collaborative degradation positive feedback loop; and extract the degradation characteristic index content value corresponding to the critical inflection point, and use the content value as the critical trigger value for the start of the collaborative degradation positive feedback loop.

[0022] Furthermore, based on the key nodes of the synergistic deterioration positive feedback loop, corresponding targeted blocking functional components are selected, and all functional components are combined and integrated to obtain multi-target synergistic blocking milk powder products, including:

[0023] S21. Based on the initiation triggering node of the synergistic deterioration positive feedback loop, with the goal of blocking the triggering source of lipid oxidation initiation and switching loop, a source-blocking functional component for lipid oxidation is selected as the first target functional component.

[0024] S22. Based on the conduction amplification node of the synergistic deterioration positive feedback loop, with the goal of cutting off the conduction pathway from lipid oxidation to protein oxidation and Maillard reaction, a blocking functional component targeting trans-reaction conduction is selected as the second target functional component.

[0025] S23. Based on the accelerated runaway node of the synergistic deterioration positive feedback loop, with the goal of inhibiting the reverse catalysis of the Maillard reaction on lipid oxidation and protein reaction and terminating the closed loop positive feedback, a blocking functional component targeting the loop closure is selected as the third target functional component.

[0026] S24. Select the first target functional component, the second target functional component, and the third target functional component, and combine and integrate them according to nutritional requirements and addition limits to obtain a multi-target synergistic blocking milk powder product.

[0027] Furthermore, based on the critical trigger value for the activation of the collaborative deterioration positive feedback loop, and combined with the deterioration kinetics of finished milk powder products, graded temperature control intervals are defined and corresponding control rules are set; cumulative temperature and humidity exposure data across the entire supply chain are collected simultaneously, and the control rules are dynamically adjusted, including:

[0028] S31. Based on the critical trigger value of the collaborative deterioration positive feedback loop, and combined with the kinetic analysis results of the measured deterioration rate of finished milk powder at different temperatures, graded temperature control intervals are divided, and corresponding temperature ranges and cumulative safe storage time rules are set for each temperature control interval; the graded temperature control intervals include safe storage intervals, warning storage intervals, and prohibited storage intervals.

[0029] S32. Integrate the entire life cycle of finished milk powder products from production to consumption, and formulate control rules that are compatible with temperature control ranges and can be implemented. The entire life cycle includes production and delivery, logistics and warehousing, terminal display and terminal consumption.

[0030] S33. Continuously collect temperature and humidity data throughout the entire life cycle of the finished milk powder product, and simultaneously record the cumulative storage time of each stage to obtain complete cumulative temperature and humidity exposure data for the entire chain.

[0031] S34. Based on the collected exposure data, automatically identify abnormal events that exceed the safe storage range, trace the stage and duration of the abnormal event, and trigger the corresponding handling plan; and combine the early warning results output from step S4 to dynamically optimize and adjust the control rules in a closed loop.

[0032] Furthermore, based on the critical trigger value for the activation of the synergistic deterioration positive feedback loop, and combined with the kinetic analysis results of the measured deterioration rate changes of finished milk powder at different temperatures, graded temperature control intervals were determined, and corresponding temperature ranges and cumulative safe storage time rules were set for each temperature control interval, including:

[0033] S311. Conduct a multi-gradient temperature full-cycle storage experiment on the finished milk powder, obtain storage experiment data, and fit the degradation rate constant of the finished milk powder in different temperature ranges to quantify the influence of ambient temperature on the trigger probability and trigger time of the synergistic degradation positive feedback loop.

[0034] S312. Based on storage experiments and quantitative results, the temperature range in which the synergistic deterioration positive feedback loop has no risk of being triggered and can fully guarantee the shelf life of the preset number of days is defined as the safe storage range for finished milk powder products, and the longest storage time and recommended consumption period within the safe storage range are selected.

[0035] S313. Based on storage experiments and quantitative results, the temperature range in which the collaborative deterioration positive feedback loop has a pre-start risk and the deterioration rate is increased by more than a preset multiple compared with the safe storage range is defined as the warning storage range, and the upper limit of the cumulative safe storage time of the warning storage range is selected.

[0036] S314. Based on storage experiments and quantification results, the temperature range that can directly trigger the collaborative degradation loop within a preset time period is defined as the prohibited storage range, and the longest allowable cumulative storage time and abnormal handling signal within the prohibited storage range are selected.

[0037] Furthermore, based on the collected exposure data, abnormal events exceeding the safe storage range are automatically identified, and the stage and duration of the abnormal event are traced, triggering corresponding handling measures; and combined with the early warning results from the output feedback, the control rules are dynamically optimized and adjusted in a closed loop, including:

[0038] S341. Based on the exposure data collected across the entire chain, identify abnormal events that exceed the safe storage range, classify the abnormal events into different abnormal levels according to the abnormal temperature range and cumulative exposure duration, and select the trigger threshold for each level.

[0039] S342. Based on the identified abnormal events, trace the stage where the abnormal event occurred, the start and end time, the cumulative temperature and time exposure, and form a complete abnormal event tracing file;

[0040] S343. For abnormal events at different levels, implement emergency response plans and trigger corresponding sample testing rules to perform dynamic shelf-life prediction and graded early warning.

[0041] S344. Based on the risk classification and early warning results of milk powder quality deterioration, combined with the statistical analysis of abnormal events in the whole chain, the temperature control threshold, monitoring frequency and emergency response process of each link are optimized in reverse to realize the dynamic adjustment of process control rules.

[0042] Furthermore, based on graded and quantified thresholds, sample testing rules adapted to finished milk powder products are established; and by combining synergistic degradation positive feedback loops, degradation kinetics, and exposure data, synergistic amplification coefficients for various degradation reactions are fitted to construct a dynamic shelf-life prediction model; based on sample testing results and model outputs, graded early warning systems are established to achieve full life-cycle management of finished milk powder products, including:

[0043] S41. Based on graded quantification thresholds, establish sample detection rules for headspace sampling, gas chromatography, and ion mobility spectrometry that are adapted to the current matrix characteristics of milk powder products, and determine the content of characteristic compounds.

[0044] S42. Based on the synergistic degradation relationship, the basic synergistic amplification coefficient of different degradation reactions under different isothermal conditions is fitted, and the basic synergistic amplification coefficient is corrected according to the degradation rate contribution of different temperature ranges; combined with the actual storage temperature history and cumulative storage time of the finished milk powder, the real-time synergistic amplification coefficient at the corresponding time node is calculated.

[0045] S43. Based on the real-time collaborative amplification factor and the Arrhenius dynamic equation, a dynamic shelf-life prediction model adapted to the entire life cycle of milk powder products is constructed.

[0046] S44. Based on the measured results of the sample detection rules and the output results of the shelf life dynamic prediction model, establish graded early warning rules corresponding to different deterioration risk levels, and feed the early warning results back to the abnormal event identification stage in real time to complete the dynamic optimization of the control rules.

[0047] Furthermore, based on the real-time collaborative amplification factor and the Arrhenius kinetic equation, a dynamic shelf-life prediction model adapted to the entire life cycle of finished milk powder products is constructed, including:

[0048] S431. Based on the Arrhenius kinetic equation, the real-time cooperative amplification factor is introduced into the Arrhenius kinetic equation as a correction factor to establish a dynamic prediction model framework for the remaining shelf life with the total content of characteristic compounds as the deterioration index and storage time and storage temperature as independent variables.

[0049] S432. Through multi-gradient temperature accelerated storage experiments, collect full-cycle kinetic data of the content of characteristic compounds in finished milk powder at different temperatures with storage time, and use nonlinear regression method to fit and obtain fixed values ​​of pre-exponential factor and activation energy in the model.

[0050] S433, the shelf life dynamic prediction model includes the measured initial characteristic compound content of milk powder, exposure data of the whole chain, and real-time temperature of the storage environment. The core output parameters are the remaining safe shelf life of milk powder and the corresponding deterioration risk level.

[0051] S434. The model prediction results are corrected by periodically measuring the total content of characteristic compounds. Based on the deviation between the predicted value and the measured value, the model parameters are dynamically adjusted by a rolling optimization method so that the prediction matching degree is always kept above the preset threshold.

[0052] Furthermore, based on the measured results of the sample detection rules and the output results of the shelf-life dynamic prediction model, graded early warning rules corresponding to different deterioration risk levels are established, and the warning results are fed back to the abnormal event identification stage in real time, completing the dynamic optimization of the control rules, including:

[0053] S441. Based on the remaining safe shelf life output by the graded quantitative threshold and the dynamic prediction model of shelf life, the risk of quality deterioration of milk powder products is divided into different risk levels, and the remaining shelf life percentage threshold corresponding to each level is set respectively.

[0054] S442. Set corresponding early warning trigger conditions for different risk levels, and set the target audience and response time limit for each level of early warning;

[0055] S443. Feed back the graded early warning results, the corresponding deterioration risk level, remaining shelf life, and recommended handling measures to the abnormal event identification system as input parameters for dynamic adjustment of control rules.

[0056] Secondly, the present invention also provides a system for improving the shelf-life stability of infant formula milk powder, the system comprising:

[0057] The mechanism verification module is used to build a synergistic deterioration positive feedback loop among various deterioration reactions during the storage of infant formula milk powder through accelerated storage experiments, and to set the key nodes of the synergistic deterioration positive feedback loop and the critical trigger value for the activation of the synergistic deterioration positive feedback loop; and to establish a graded quantitative threshold based on the critical trigger value.

[0058] The component synergy module is used to select corresponding targeted blocking functional components based on the key nodes of the synergistic deterioration positive feedback loop, and to combine and integrate all functional components to obtain a multi-target synergistic blocking milk powder product.

[0059] The dynamic control module is used to divide the temperature control range into different levels and set corresponding control rules based on the critical trigger value of the collaborative deterioration positive feedback loop and the deterioration kinetic characteristics of the finished milk powder product; it also collects the cumulative exposure data of temperature and humidity throughout the entire chain and dynamically adjusts the control rules.

[0060] The shelf-life prediction module is used to establish sample testing rules adapted to finished milk powder products based on graded quantitative thresholds; and to combine the synergistic deterioration positive feedback loop, deterioration kinetic characteristics and exposure data to fit the synergistic amplification coefficients of various deterioration reactions and construct a dynamic shelf-life prediction model; and to establish graded early warning based on sample testing results and model output to achieve full life cycle management of finished milk powder products.

[0061] The beneficial effects of this invention are as follows:

[0062] 1. Through experimental methods including basic kinetic correlation analysis and single-node blocking causality verification, the synergistic positive feedback relationship among the three core deterioration reactions—lipid oxidation, protein oxidation, and Maillard reaction—during the storage of infant formula milk powder was clarified. The complete transmission path of the deterioration reaction from initiation to acceleration was elucidated, and the three corresponding key nodes were identified. Furthermore, by analyzing the intrinsic correlation of the synergistic deterioration of multiple reactions in milk powder, a clear targeted design basis was provided for improving the shelf-life stability of milk powder, solving the problem that existing technologies often focus on a single deterioration reaction and lack targeted regulation.

[0063] 2. By constructing a stability enhancement scheme that combines targeted blocking of the formula with full-chain process control, corresponding targeted blocking functional components were designed for three core key nodes to inhibit the occurrence and transmission of deterioration reactions at the product formulation level. At the same time, based on the critical trigger characteristics of deterioration, graded temperature control ranges were defined, and feasible control rules covering the entire process of production, logistics, terminal, and consumption were formulated, realizing the synergistic adaptation of formula design and process control, and effectively improving the storage stability of milk powder throughout its entire life cycle.

[0064] 3. The detection method, which employs headspace sampling, gas chromatography, and ion mobility spectrometry, eliminates the need for complex pretreatment and can quickly determine the deterioration state of milk powder, meeting the quality control needs of the entire industry chain. Furthermore, based on the synergistic effect of deterioration, a correction factor is introduced to construct a dynamic shelf-life prediction model suitable for the entire lifecycle of milk powder. This model can accurately predict shelf life by combining actual storage data and can also optimize control rules based on early warning results, forming a complete quality control closed loop and effectively ensuring the nutritional quality and safety of milk powder throughout its entire lifecycle. Attached Figure Description

[0065] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings:

[0066] Figure 1 This is a flowchart of a method for improving the shelf-life stability of infant formula milk powder according to an embodiment of the present invention;

[0067] Figure 2 This is a flowchart of step S3 in a method for improving the shelf-life stability of infant formula milk powder according to an embodiment of the present invention;

[0068] Figure 3 This is a flowchart of step S4 in a method for improving the shelf-life stability of infant formula milk powder according to an embodiment of the present invention;

[0069] Figure 4 These are scanning electron microscope (SEM) images of infant formula milk powder stored at different temperatures for 28 days according to an embodiment of the present invention.

[0070] Figure 5 This is a system principle block diagram of an infant formula milk powder shelf-life stability improvement system according to an embodiment of the present invention.

[0071] The reference numerals are: 1. Mechanism verification module; 2. Component synergy module; 3. Dynamic control module; 4. Shelf life prediction module. Detailed Implementation

[0072] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0073] Please see Figure 1 A method for improving the shelf-life stability of infant formula milk powder is provided, including:

[0074] S1. A synergistic deterioration positive feedback loop among various deterioration reactions during the storage of infant formula milk powder was established through accelerated storage experiments. Key nodes of the synergistic deterioration positive feedback loop and critical trigger values ​​for its activation were set. A graded quantitative threshold was established based on the critical trigger values.

[0075] In the description of this invention, a synergistic deterioration positive feedback loop among various deterioration reactions during the storage of infant formula milk powder is established through accelerated storage experiments. Key nodes of the synergistic deterioration positive feedback loop and critical trigger values ​​for its activation are set. A graded quantification threshold is established based on the critical trigger values, including:

[0076] S11. Using infant formula milk powder as a unified basic sample, with fixed relative humidity and non-temperature variable parameters, multiple sets of controllable parallel storage experiments were configured, and full-cycle change data of various degradation reactions were collected simultaneously. The interaction relationship between different degradation reactions was quantitatively verified through statistical analysis. Degradation reactions include lipid oxidation, protein oxidation and Maillard reaction.

[0077] Specifically, this invention uses commercially available stage 1 infant formula milk powder that meets the requirements of GB10765-2021 "Infant Formula Food" as a uniform base sample. Four temperature gradient experimental groups were set up at 25℃, 30℃, 37℃ and 42℃. Six parallel samples were set up in each experimental group. The relative humidity of the environment in all experimental groups was kept constant at 55%, and all other non-temperature experimental conditions were completely consistent. Samples were taken simultaneously on days 0, 3, 6, 9, 12, 15, 18, 21, 24 and 27 of the experiment. After sampling, the samples were immediately sealed and frozen in the dark to avoid secondary deterioration.

[0078] The core indicators of three major deterioration reactions were detected simultaneously: lipid oxidation index was determined by titration method I (GB5009.227-2016) for peroxide value and liquid chromatography method II (GB5009.181-2016) for malondialdehyde content; protein oxidation index was determined by Ellman method for free sulfhydryl content; Maillard reaction index was determined by FAST index method for protein denaturation degree; 1-hexanal-M, 1-hexanal-D, 4,5-dihydro-3(2H)-thiophenone, and 2-methylhexanal were also detected. The contents of six characteristic volatile compounds, namely propionic acid, cyclopentanone-M, and 1-pentanol-D, were measured. Pearson correlation analysis was performed on the full-cycle detection data using SPSS 26.0 software to calculate the correlation coefficients and significance levels between the three types of deterioration reactions, thus completing the quantitative verification of the interaction relationship. All experimental data were obtained from at least three independent replicate experiments, and the results are expressed as mean ± standard deviation. The differences between groups were analyzed using t-tests, with p < 0.05 indicating significant differences and p < 0.01 indicating highly significant differences.

[0079] S12. Based on the quantitative verification results, sort out the transmission paths and mutual promotion logic between different degradation reactions, divide the complete process of coordinated degradation from the start to accelerated runaway, and select the key nodes of the coordinated degradation positive feedback loop, including the initiation trigger node, the transmission amplification node and the accelerated runaway node.

[0080] Specifically, based on the correlation analysis results (the pairwise positive correlation coefficients of the three types of deterioration reactions are all >0.92, p<0.01), combined with the results of the single-node blocking causal verification experiment, the complete transmission pathway and closed-loop promotion logic of the three major deterioration reactions were clarified: the oxidation of unsaturated fatty acids generates free radicals and aldehyde products, which then attack whey proteins to induce protein oxidation. At the same time, the lipid oxidation products and the active sites of oxidized proteins accelerate the initiation of the Maillard reaction. The dicarbonyl compounds generated by the Maillard reaction catalyze lipid oxidation and protein cross-linking in the reverse direction, forming a closed-loop positive feedback.

[0081] Based on the above transmission logic, the complete process of synergistic degradation is divided into three stages: initiation triggering, transmission amplification, and accelerated runaway. Correspondingly, three core key nodes of the synergistic degradation positive feedback loop are selected: 1. Initiation triggering node: Lipid oxidation is initiated, and free radicals begin to be generated, which is the core source of loop initiation; 2. Transmission amplification node: Protein oxidation and Maillard reaction are initiated simultaneously, realizing the trans-reaction transmission and amplification of degradation, which is the core bridge for the continuous operation of the loop; 3. Accelerated runaway node: Maillard reaction reverses and catalyzes the other two types of reactions, the closed-loop positive feedback is formally formed, and the degradation rate jumps exponentially, which is the core switch for the collapse of milk powder quality.

[0082] S13. Collect full-cycle kinetic data of the deterioration characteristic indicators of formula milk powder under different storage conditions as storage time changes, and set the critical trigger value for the activation of the synergistic deterioration positive feedback loop by analyzing the deterioration rate change law of the synergistic deterioration positive feedback loop.

[0083] In the description of this invention, full-cycle kinetic data of the degradation characteristic indicators of formula milk powder under different storage conditions are collected as storage time changes. By analyzing the degradation rate variation law of the synergistic degradation positive feedback loop, the critical trigger value for the activation of the synergistic degradation positive feedback loop is set as follows:

[0084] S131. Obtain quantitative detection data of deterioration characteristic indicators of formula milk powder under different temperature gradients and different storage times in the full cycle change data, and form full cycle kinetic data of deterioration characteristic indicators changing with storage time and storage temperature.

[0085] Specifically, using the six characteristic volatile compounds identified in S11 as the core characteristic indicators of deterioration, quantitative detection data of characteristic compounds at all sampling time points within the four temperature gradients and the entire storage cycle were compiled. With storage time as the x-axis, total content of characteristic compounds as the y-axis, and storage temperature as the grouping variable, all detection data were standardized and compiled to form a full-cycle dynamic dataset of deterioration characteristic indicators changing with storage time and storage temperature.

[0086] It should be noted that the microscopic morphology changes of milk powder particles under different storage conditions were observed using scanning electron microscopy (SEM), and the results are as follows: Figure 4 As shown. Before storage, the milk powder sample (D0) had intact particles with good sphericity, smooth and dense surface, and no obvious agglomeration or pores. After 28 days of storage, as the storage temperature increased, the microstructure damage of the milk powder particles gradually intensified: the milk powder particles stored at 25℃ still maintained good integrity, with only slight agglomeration; the milk powder particles stored at 30℃ showed obvious pores on the surface, and the degree of agglomeration intensified; the milk powder particles stored at 37℃ and 42℃ showed severe wall material cracking, structural collapse and agglomeration, and a large number of porous structures formed on the particle surface.

[0087] The aforementioned microstructural changes perfectly match the synergistic degradation loop mechanism confirmed by this invention: the cracking of the milk powder particle wall material and the increase in surface porosity lead to the exposure of embedded unsaturated fatty acids to oxygen, accelerating the initiation of lipid oxidation, and thus triggering a synergistic degradation positive feedback loop. This confirms the key influence of temperature on the initiation of milk powder degradation at the microstructural level.

[0088] S132. Nonlinear piecewise regression fitting is performed on the full-cycle dynamic data using nonlinear regression analysis software. Piecewise function equations of the deterioration characteristic index as a function of storage time are obtained under different temperature conditions, and the deterioration rate variation law of different stages is divided.

[0089] Specifically, nonlinear regression analysis software, such as Origin 2023, was used to perform piecewise regression fitting on the full-cycle kinetic dataset. First-order kinetic equations were used to fit piecewise function equations of the total content of characteristic compounds changing with storage time under different temperature conditions. The optimal segmentation point was determined by the R² maximum goodness-of-fit method. Based on the slope of the piecewise function equations, the degradation rate constants of different stages were calculated to clarify the low-speed stage of slow growth of characteristic compound content, the medium-speed stage of linear growth, and the high-speed stage of exponential growth, thus obtaining the degradation rate change law corresponding to the three stages.

[0090] S133. Based on the piecewise function equation obtained by nonlinear piecewise regression fitting, the abrupt change point where an exponential jump occurs in the degradation rate change is selected, and the abrupt change point is determined as the critical inflection point for the initiation of the synergistic degradation positive feedback loop. The degradation characteristic index content value corresponding to the critical inflection point is extracted, and the content value is used as the critical trigger value for the initiation of the synergistic degradation positive feedback loop.

[0091] Specifically, by comparing the degradation rate constants at different stages, the point where the degradation rate constant jumps by two orders of magnitude compared to the medium-speed stage is identified as the critical inflection point for the initiation of the synergistic degradation positive feedback loop. The total content of the six characteristic compounds corresponding to all critical inflection points under four temperature gradients is extracted, and the average value is taken as the final critical trigger value. The critical trigger value for the initiation of the synergistic degradation positive feedback loop is determined to be 0.82 mg / kg of the total content of the six characteristic volatile compounds.

[0092] S14. Based on the critical trigger value, and combined with the different startup states of the loop, the hierarchical quantization thresholds for the three states of the corresponding loop—not started, pre-started, and started—are obtained.

[0093] Specifically, using a critical trigger value of 0.82 mg / kg as the core benchmark, and combining the activation process of the synergistic degradation positive feedback loop, a three-level quantitative threshold system is obtained: 1. Safety threshold: The total content of the six characteristic compounds is <0.25 mg / kg, that is, below 30% of the critical trigger threshold, corresponding to the synergistic degradation loop not being activated, and the milk powder quality fully meets the national standards; 2. Warning threshold: The total content of the six characteristic compounds is 0.25 mg / kg to 0.57 mg / kg, that is, 30% to 70% of the critical trigger threshold, corresponding to the pre-activation state of the synergistic degradation loop, with an extremely high risk of loop activation; 3. Runaway threshold: The total content of the six characteristic compounds is >0.57 mg / kg, that is, exceeding 70% of the critical trigger threshold, corresponding to the activated state of the synergistic degradation loop, and the milk powder has an irreversible degradation risk.

[0094] S2. Based on the key nodes of the synergistic deterioration positive feedback loop, select the corresponding targeted blocking functional components, and combine and integrate all functional components to obtain a multi-target synergistic blocking milk powder product.

[0095] In the description of this invention, based on the key nodes of the synergistic deterioration positive feedback loop, corresponding targeted blocking functional components are selected respectively, and all functional components are compounded and integrated to obtain a multi-target synergistic blocking milk powder product, including:

[0096] S21. Based on the initiation triggering node of the synergistic deterioration positive feedback loop, with the goal of blocking the triggering source of lipid oxidation initiation and switching loop, a source-blocking functional component for lipid oxidation is selected as the first target functional component.

[0097] Specifically, the first target functional component consists of two parts: unsaturated fatty acids encapsulated in microcapsules with double-walled material and a complex antioxidant system. All raw materials meet the relevant national standards for infant formula.

[0098] The inner wall material of the microcapsule is sodium octenyl succinate starch, with an addition amount of 0.8 g / kg (which meets the limit requirement of 1.0 g / kg for the maximum use in infant formula in GB2760-2014). The outer wall material is a complex system of whey protein and maltodextrin in a mass ratio of 2:1, which encapsulates DHA, ARA, linoleic acid, and linolenic acid unsaturated functional oils in the formula.

[0099] The microcapsule preparation process is as follows: the wall material is mixed with deionized water to prepare an aqueous phase with a solid content of 30%, and stirred in a water bath at 55°C for 2 hours until completely dissolved. After adding unsaturated oil, the microcapsule is emulsified by high-speed shearing for 10 minutes to obtain a primary emulsion. The microcapsule is homogenized twice under high pressure at 40 MPa and then prepared by spray drying at an inlet air temperature of 180°C and an outlet air temperature of 80°C. The final fat encapsulation rate is ≥98.5%.

[0100] The composite antioxidant system uses vitamin E and ascorbate palmitate in a mass ratio of 3:1. The amount of vitamin E added is 100 mg / kg (meeting the limit requirements of GB10765-2021), and the amount of ascorbate palmitate added is 33 mg / kg (meeting the relevant requirements of GB2760-2014). The microcapsule powder is premixed with the composite antioxidant system to obtain the first target functional component targeting the initiation trigger node.

[0101] S22. Based on the conduction amplification node of the synergistic deterioration positive feedback loop, with the goal of cutting off the conduction pathway from lipid oxidation to protein oxidation and Maillard reaction, a blocking functional component targeting trans-reaction conduction is selected as the second target functional component.

[0102] Specifically, the second target functional component consists of two parts: a pre-modified protein base and a complex blocking system. All raw materials meet the relevant national standards for infant formula.

[0103] The pre-modified protein base material uses wet-process glycosylated pre-modified whey protein and casein. The specific process is as follows: whey protein and lactose are mixed at a mass ratio of 4:1 to prepare a protein solution with a solid content of 15%. The pH is adjusted to 7.0, dissolved in a water bath at 55°C, and then freeze-dried. The solution is then placed in an environment of 60°C and 65% relative humidity for 4 hours. The endpoint is controlled as follows: degree of glycosylation 15%±2%, browning index ΔE<5, and protein active site blocking rate ≥85%.

[0104] Premodified whey protein and casein were optimized at a mass ratio of 6.5:3.5 to form a premodified protein base. The composite blocking system used L-cysteine ​​and tea polyphenols at a mass ratio of 3:100, with L-cysteine ​​added at 6 mg / kg (compliant with relevant requirements of GB2760-2014) and tea polyphenols added at 200 mg / kg (compliant with relevant requirements of GB2760-2014). The premodified protein base and the composite blocking system were premixed evenly to obtain a second target functional component targeting the conduction amplification node.

[0105] S23. Based on the accelerated runaway node of the synergistic deterioration positive feedback loop, with the goal of inhibiting the reverse catalysis of the Maillard reaction on lipid oxidation and protein reaction and terminating the closed loop positive feedback, a blocking functional component targeting the loop closure is selected as the third target functional component.

[0106] Specifically, the third target functional component consists of two parts: acidity regulator and anti-caking agent. All raw materials meet the relevant national standards for infant formula.

[0107] The acidity regulator is sodium citrate, added at a rate of 0.5 g / kg (compliant with relevant requirements of GB2760-2014). It can stabilize the pH value of the rehydrated milk powder system at 6.6-6.7, inhibit the formation of late-stage active intermediates in the Maillard reaction, and eliminate its reverse catalytic effect on the other two types of deterioration reactions. The anti-caking agent is silica, added at a rate of 5 g / kg (compliant with the limit requirement of 15 g / kg for maximum addition of milk powder in GB2760-2014). It can inhibit the aggregation of milk powder particles and lactose crystallization, maintain the structural stability of milk powder particles, and avoid exposure of oxidized substrates due to damage to the encapsulation matrix. Sodium citrate and silica are premixed evenly to obtain the third target functional component targeting the accelerated runaway node.

[0108] S24. Select the first target functional component, the second target functional component, and the third target functional component, and combine and integrate them according to nutritional requirements and addition limits to obtain a multi-target synergistic blocking milk powder product.

[0109] Specifically, based on GB10765-2021 "Infant Formula Foods", a basic nutritional framework for infant formula milk powder is established. The functional components of the first, second, and third targets are added in the amount designed in the formula, and then dry-mixed with basic raw materials such as lactose, compound vitamins, and compound minerals. The amount of all raw materials added meets the national standard nutritional requirements and additive limits.

[0110] After compounding, aseptic filling with nitrogen is used, with residual oxygen content ≤2% in each can, resulting in a targeted modified milk powder product with synergistic blocking of three targets. Before leaving the factory, the total content of six characteristic compounds must be tested to ensure that the initial total content is <0.25mg / kg, a safe threshold. After 28 days of storage at 37℃ and 55% relative humidity, the total amount of the six characteristic compounds in the three-target synergistic formula is reduced by 82.3%±1.1%, far exceeding the expected effect of simple superposition of single targets. At the same time, it achieves simultaneous inhibition of the three major deterioration reactions, extending the shelf life of the milk powder from the industry-standard 28 days to more than 45 days.

[0111] S3. Based on the critical trigger value of the collaborative deterioration positive feedback loop, combined with the deterioration kinetics of finished milk powder products, divide the graded temperature control range and set corresponding control rules; simultaneously collect the cumulative exposure data of temperature and humidity throughout the entire chain, and dynamically adjust the control rules.

[0112] In the description of this invention, as Figure 2 Based on the critical trigger value for the activation of the collaborative deterioration positive feedback loop, and combined with the deterioration kinetics of finished milk powder products, graded temperature control intervals are defined and corresponding control rules are set. Simultaneously, cumulative temperature and humidity exposure data across the entire supply chain are collected, and the control rules are dynamically adjusted, including:

[0113] S31. Based on the critical trigger value of the collaborative deterioration positive feedback loop, and combined with the kinetic analysis results of the measured deterioration rate of finished milk powder at different temperatures, a graded temperature control interval is obtained, and a corresponding temperature range and cumulative safe storage time rule are set for each temperature control interval; the graded temperature control interval includes a safe storage interval, a warning storage interval, and a prohibited storage interval.

[0114] In the description of this invention, based on the critical trigger value for the activation of the synergistic deterioration positive feedback loop, and combined with the kinetic analysis results of the measured deterioration rate changes of finished milk powder at different temperatures, graded temperature control intervals are defined, and corresponding temperature ranges and cumulative safe storage time rules are set for each temperature control interval, including:

[0115] S311. Conduct multi-gradient temperature full-cycle storage experiments on finished milk powder products, obtain storage experiment data, and fit the degradation rate constant of finished milk powder products in different temperature ranges to quantify the influence of ambient temperature on the trigger probability and trigger time of the synergistic degradation positive feedback loop.

[0116] Specifically, the prepared targeted modified milk powder was placed in a constant temperature and humidity environment with five temperature gradients (25℃, 30℃, 37℃, 45℃, and 55℃) and a constant relative humidity of 55% for a total storage period of 90 days. Samples were taken every 7 days to detect the total content of six characteristic compounds. Based on the detection data, the degradation rate constant of the milk powder was obtained by fitting the Arrhenius kinetic equation in different temperature ranges. A quantitative functional relationship was established between ambient temperature and degradation rate constant, loop triggering time, and triggering probability. It was found that for every 5℃ increase in temperature, the degradation rate constant increases by 1.8 to 2.2 times, and the loop triggering time is shortened by more than 50%, thus quantifying the influence of ambient temperature on the synergistic degradation loop.

[0117] S312. Based on storage experiments and quantitative results, the temperature range within which the synergistic deterioration positive feedback loop has no risk of being triggered and can fully guarantee the shelf life of the preset number of days is defined as the safe storage range for finished milk powder products, and the longest storage time and recommended consumption period within the safe storage range are selected.

[0118] Specifically, based on the results of quantitative analysis, an ambient temperature of ≤25℃ is defined as the safe storage range. Within this range, there is no risk of triggering the positive feedback loop of synergistic deterioration, which can fully guarantee the manufacturer's stated 28-day shelf life after opening. It is also clearly stated that within this range, it is recommended to consume within 28 days after opening, with a maximum storage time of 45 days. In the unopened and sealed state, it can fully guarantee the 18-month regular shelf life.

[0119] S313. Based on storage experiments and quantification results, the temperature range in which the collaborative deterioration positive feedback loop has a pre-start risk and the deterioration rate is increased by more than a preset multiple compared with the safe storage range is defined as the warning storage range, and the upper limit of the cumulative safe storage time of the warning storage range is selected.

[0120] Specifically, based on the results of quantitative analysis, an ambient temperature of 25℃~30℃ was designated as the warning storage range. Within this range, the rate of deterioration of the finished milk powder product is more than twice that of the safe storage range, and there is a risk of pre-activation of the positive feedback loop of synergistic deterioration. It was also determined that the maximum cumulative safe storage time within this range is 10 days. If the cumulative storage time exceeds 10 days, a special re-test of the content of characteristic compounds in the finished milk powder product is required.

[0121] S314. Based on storage experiments and quantification results, the temperature range that can directly trigger the collaborative degradation loop within a preset time period is defined as the prohibited storage range, and the longest allowable cumulative storage time and abnormal handling signal within the prohibited storage range are selected.

[0122] Specifically, based on the results of quantitative analysis, an ambient temperature >30℃ is defined as a prohibited storage range. Within this range, a positive feedback loop of synergistic deterioration can be directly triggered within 72 hours, and the deterioration rate of the finished milk powder product increases exponentially. It is also clarified that the maximum allowable cumulative storage time within this range is 48 hours. If the cumulative storage time exceeds 48 hours, an abnormal handling signal for product sealing and comprehensive testing will be immediately triggered.

[0123] S32. Integrate the entire life cycle of finished milk powder products from production to consumption, and formulate control rules that are compatible with temperature control ranges and can be implemented. The entire life cycle includes production and delivery, logistics and warehousing, terminal display and terminal consumption.

[0124] Specifically, standardized control rules have been developed for each of the four core stages, fully adapted to the three-level temperature control range. All rules comply with relevant national standards and are compatible with existing industry-standard technologies. The entire lifecycle includes: 1. Production and delivery stage: The safety thresholds determined by S1 are incorporated into the mandatory quality control standards for product delivery. Each batch of product must be accompanied by an initial quality file, recording background kinetic parameters, initial characteristic compound content, and production batch information. The product label must indicate the storage guidelines and dynamic consumption period corresponding to the three-level temperature control range, in accordance with GB7718-2011 requirements; 2. Logistics and warehousing stage: Trunk line transportation 1. **Regional Warehousing:** The temperature in the regional storage environment is stably controlled at ≤25℃, and the relative humidity is 50%~55%. Temperature and humidity data are recorded every 15 minutes using an IoT temperature and humidity recorder and uploaded to the cloud every 2 hours. 2. **Terminal Display:** Products are displayed away from heaters, freezer vents, and direct sunlight. Temperature and humidity in the display area are monitored and recorded three times a day (morning, noon, and evening). Online warehousing enables continuous 24-hour temperature and humidity monitoring. 3. **Household Consumption:** Once opened, products should be stored in an environment ≤25℃. No more than 5 times of opening and taking powder per day. The can should be sealed immediately after taking powder, and refrigeration is strictly prohibited. Discontinue consumption immediately if any sensory abnormalities occur.

[0125] S33. Continuously collect temperature and humidity data throughout the entire life cycle of the finished milk powder product, and simultaneously record the cumulative storage time of each stage to obtain complete cumulative temperature and humidity exposure data for the entire chain.

[0126] Specifically, temperature and humidity monitoring points are deployed at five core nodes: production and delivery, trunk transportation and transshipment, regional warehousing and storage, terminal display and shelf placement, and consumer can opening, to achieve uninterrupted coverage throughout the entire lifecycle. All monitoring points use a unified timestamp, a temperature accuracy of ±0.2℃, a humidity accuracy of ±2%RH, and a collection frequency of 15 minutes per collection point. All data is encrypted and uploaded to the cloud traceability system, making it tamper-proof and fully traceable. Based on the collected continuous temperature and humidity data and cumulative storage time, the cumulative temperature and time exposure of a single batch of products is calculated according to the degradation rate weights of different temperature ranges (safe range weight 1, warning range weight 3, and prohibited range weight 10), forming a standardized dataset, which is updated in real time as the product flows.

[0127] S34. Based on the collected exposure data, automatically identify abnormal events that exceed the safe storage range, trace the stage and duration of the abnormal event, and trigger the corresponding handling plan. Combined with the early warning results output from step S4, dynamically optimize and adjust the control rules in a closed-loop manner.

[0128] In the description of this invention, based on the collected exposure data, abnormal events exceeding the safe storage range are automatically identified, and the stage and duration of the abnormal event are traced, triggering corresponding handling measures; and combined with the output feedback early warning results, the control rules are dynamically optimized and adjusted in a closed loop, including:

[0129] S341. Based on the exposure data collected across the entire chain, identify abnormal events that exceed the safe storage range, classify the abnormal events into different abnormal levels according to the abnormal temperature range and cumulative exposure duration, and select the trigger threshold for each level.

[0130] Specifically, based on the cumulative exposure data of temperature and humidity collected throughout the entire chain, the system automatically identifies temperature anomalies where the ambient temperature exceeds the safe storage range. According to the abnormal temperature range and cumulative exposure duration, the anomalies are divided into three levels: 1. General anomaly: Cumulative exposure duration in the warning storage range is 4-10 days; 2. Moderate anomaly: Cumulative exposure duration in the warning storage range is >10 days, or cumulative exposure duration in the prohibited storage range is 24-48 hours; 3. Severe anomaly: Cumulative exposure duration in the prohibited storage range is >48 hours. The system defines the trigger threshold for each level and automatically triggers the corresponding level of warning.

[0131] S342. Based on the identified abnormal events, trace the stage where the abnormal event occurred, the start and end time, the cumulative temperature and time exposure, and form a complete abnormal event traceability file.

[0132] Specifically, for identified temperature anomalies, the entire chain of data from the cloud-based traceability system is used to pinpoint the specific stage where the anomaly occurred, its start and end times, cumulative temperature, exposure time, and responsible party. This creates a complete anomaly traceability file containing the anomaly batch, anomaly level, anomaly details, traceability data, and handling recommendations, which is then simultaneously pushed to the corresponding responsible party.

[0133] S343. For abnormal events at different levels, implement emergency response plans and trigger corresponding sample testing rules to perform dynamic shelf-life prediction and graded early warning.

[0134] Specifically, a tiered emergency response plan is implemented for different levels of anomalies: 1. General anomaly: Sampling and testing of the corresponding batch of products are conducted, and the total content of characteristic compounds is determined using the S41 rapid detection method. Dynamic prediction of remaining shelf life is initiated simultaneously. 2. Moderate anomaly: The corresponding batch of products is sealed and subjected to specialized testing. Based on the test results, the product shelf life is dynamically shortened, and an important early warning is issued. 3. Severe anomaly: The corresponding batch of products is immediately and completely sealed. An emergency early warning and comprehensive testing are initiated. The product's compliance is determined based on the test results. Non-compliant products are subject to a recall process.

[0135] S344. Based on the risk classification and early warning results of milk powder quality deterioration, combined with the statistical analysis of abnormal events in the whole chain, the temperature control threshold, monitoring frequency and emergency response process of each link are optimized in reverse to realize the dynamic adjustment of process control rules.

[0136] Specifically, based on the risk classification and early warning results of milk powder quality deterioration output by S4, combined with the statistical analysis of monthly abnormal events across the entire chain, high-risk links and control loopholes are identified, and control rules are optimized in reverse: for high-risk links, the frequency of temperature and humidity monitoring is increased from 15 minutes / time to 5 minutes / time; for high-temperature seasons, the temperature control threshold for terminal displays is dynamically tightened and the re-inspection cycle is shortened; for the effective handling process verified by the early warning, it is updated into the standardized control rules to achieve dynamic iteration of process control.

[0137] S4. Based on graded and quantified thresholds, establish sample testing rules suitable for finished milk powder products; and combine the synergistic deterioration positive feedback loop, deterioration kinetic characteristics and exposure data to fit the synergistic amplification coefficients of various deterioration reactions and construct a dynamic prediction model for shelf life; establish graded early warning based on sample testing results and model output to achieve full life cycle management of finished milk powder products.

[0138] In the description of this invention, as Figure 3As shown, based on graded quantification thresholds, sample testing rules adapted to finished milk powder products are established; and by combining synergistic degradation positive feedback loops, degradation kinetics, and exposure data, synergistic amplification coefficients for various degradation reactions are fitted, and a dynamic shelf-life prediction model is constructed. Based on sample testing results and model outputs, a graded early warning system is established to achieve full life-cycle management of finished milk powder products, including:

[0139] S41. Based on graded quantification thresholds, establish sample detection rules for headspace sampling, gas chromatography, and ion mobility spectrometry that are adapted to the matrix characteristics of current finished milk powder products, and determine the content of characteristic compounds.

[0140] Specifically, this invention establishes a rapid detection method using headspace sampling, gas chromatography, and ion mobility spectrometry adapted to target and improve the matrix characteristics of finished milk powder products. The specific parameters are as follows: 1. Pretreatment: Accurately weigh 1.00 g of milk powder sample into a 20 mL headspace injection vial, seal, and incubate at 80 °C for 20 min, shaking at 500 rpm during incubation. After incubation, inject 500 μL of headspace gas; no organic reagent extraction is required. 2. Instrument conditions: Use a weakly polar capillary column, maintain the column temperature at 60 °C for 30 min, use high-purity nitrogen as the carrier gas, and program the carrier gas flow rate as 2 mL / min for 0-2 min, 2-1 The flow rate was linearly increased to 10 mL / min at 0 min, linearly increased to 100 mL / min at 10-20 min, and maintained at 100 mL / min at 20-30 min; the drift tube temperature was 45℃, and the drift gas flow rate was 150 mL / min; 3. Qualitative and quantitative analysis: Standards of six target characteristic compounds were used for dual qualitative analysis by retention time and migration time, and a standard curve was established using the external standard method for quantitative detection. The method detection limit was 0.01 mg / kg, and the quantitation limit was 0.03 mg / kg; Based on the detection results and combined with the grading and quantification threshold, the quality status of the milk powder was rapidly determined within 2 hours.

[0141] S42. Based on the synergistic degradation relationship, the basic synergistic amplification coefficients of different degradation reactions under different isothermal conditions are fitted, and the basic synergistic amplification coefficients are corrected according to the degradation rate contribution of different temperature ranges; combined with the actual storage temperature history and cumulative storage time of the finished milk powder, the real-time synergistic amplification coefficients at the corresponding time nodes are calculated.

[0142] Specifically, based on single-node blocking experimental data, the basic synergistic amplification factor k of the three major degradation reactions was obtained under different isothermal conditions of 25℃, 30℃, 37℃, and 42℃. The basic synergistic amplification factor is defined as the factor by which the degradation rate increases due to the mutual promotion of the three major degradation reactions, and the calculation formula is k = actual degradation rate / (independent rate of lipid oxidation + independent rate of protein oxidation + independent rate of Maillard reaction). The functional relationship between the basic synergistic amplification factor k and the absolute temperature T is established as k = 0.0028 × T - 0.5421, and the goodness of fit R² = 0.992.

[0143] Based on the cumulative temperature exposure dataset across the entire supply chain, weighting coefficients are assigned according to the contribution of degradation rate in different temperature ranges, and the basic synergistic amplification coefficient is dynamically corrected. Combining the actual storage temperature history and cumulative storage time of the finished milk powder, the real-time synergistic amplification coefficient at the corresponding time node is calculated and synchronously updated to the shelf-life dynamic prediction model.

[0144] Furthermore, the weighting coefficient ω corresponding to the contribution of the degradation rate in different temperature ranges is calculated using the following standardized formula:

[0145] ;

[0146] In the formula, ω i t represents the weighting coefficient corresponding to the i-th temperature range; i t represents the cumulative exposure time of milk powder in the i-th temperature range; total k represents the total cumulative storage time of the milk powder. i k is the milk powder deterioration rate constant corresponding to the i-th temperature range; 25℃ The rate constant of milk powder degradation corresponding to a safe storage range of 25°C. Real-time co-amplification factor k. 实时 The final calculation formula is:

[0147] ;

[0148] In the formula, k 基础 The basic synergistic amplification factor is obtained by fitting the temperature function relationship; n is the total number of temperature ranges experienced by the milk powder; through this standardized formula, those skilled in the art can directly calculate a unique and repeatable real-time synergistic amplification factor based on the temperature exposure data of the entire chain, without any fuzzy adjustment space.

[0149] S43. Based on the real-time collaborative amplification factor and the Arrhenius dynamic equation, a dynamic shelf-life prediction model adapted to the entire life cycle of finished milk powder products is constructed.

[0150] In the description of this invention, the shelf-life dynamic prediction model adapted to the entire life cycle scenario of finished milk powder products, based on the real-time collaborative amplification factor and the Arrhenius kinetic equation, includes:

[0151] S431. Based on the Arrhenius kinetic equation, the real-time cooperative amplification factor is introduced into the Arrhenius kinetic equation as a correction factor to establish a dynamic prediction model framework for the remaining shelf life with the total content of characteristic compounds as the deterioration index and storage time and storage temperature as independent variables.

[0152] Specifically, based on the classical Arrhenius kinetic equation, the real-time synergistic amplification factor k is introduced into the equation as a correction factor for the reaction rate. A dynamic prediction model framework for the remaining shelf life is established with the total content of six characteristic compounds as the core deterioration index and storage time and storage temperature as the core independent variables. The core logic is: by using the corrected reaction rate, the remaining time when the total content of characteristic compounds reaches the critical trigger threshold is predicted, which is the remaining safe shelf life of the milk powder.

[0153] The core applicable boundaries and limitations of the shelf-life dynamic prediction model are as follows: 1. Applicable objects: It is primarily applicable to stage 1 infant formula milk powder that meets the requirements of GB10765-2021 "Infant Formula Food", and has equal compatibility with stage 2 and 3 follow-up infant and toddler formula milk powder that meets the requirements of GB10767-2021; 2. Applicable temperature range: 0℃~55℃, covering all conventional and extreme storage scenarios in the entire milk powder industry chain; 3. Maximum prediction duration: The maximum prediction duration is 18 months in the unopened sealed state and 90 days in the opened state; 4. Core limitations: The characteristic compound indicators input to the model are detected using the headspace sampling, gas chromatography, and ion mobility spectrometry methods disclosed in this invention to ensure the consistency and accuracy of the input data.

[0154] S432. Through multi-gradient temperature accelerated storage experiments, the full-cycle kinetic data of the content of characteristic compounds in the finished milk powder at different temperatures with storage time were collected, and the fixed values ​​of the pre-exponential factor and activation energy in the model were obtained by fitting the model using nonlinear regression method.

[0155] Specifically, accelerated storage experiments were conducted at multiple temperature gradients (25℃, 30℃, 37℃, 45℃, and 55℃) to collect full-cycle kinetic data on the total content of six characteristic compounds in the finished milk powder at different temperatures over storage time. Nonlinear regression fitting was performed using Origin 2023 software to obtain the model's fixed parameters: the pre-exponential factor A had a fitted value of 2.37 × 10⁻⁶. 7 day -1 Activation energy of the reaction E a The fitted value was 42.3 kJ / mol, the gas constant R was 8.314 J / (mol·K), and all parameters passed the significance test, p < 0.01.

[0156] Furthermore, the fitting validation criteria for the model with fixed parameters are as follows: The fitting is performed using the nonlinear least squares method, and the fitting results must meet two mandatory requirements: 1. Goodness of fit R0 2 1. A p-value ≥ 0.99 ensures a linear fit of the parameters; 2. A p-value < 0.01 in the significance test of the fitted parameters ensures that the parameters are statistically significant; The pre-exponential factor A obtained by fitting the parameters in this invention is 2.37 × 10⁻⁶. 7 day -1 Activation energy of reaction E a =42.3kJ / mol, all of which meet the above verification criteria, and have been repeatedly verified by independent experiments in 3 different batches and 5 temperature gradients, with the relative standard deviation of the parameters being ≤2.1%.

[0157] S433, the shelf-life dynamic prediction model includes the measured initial characteristic compound content of milk powder, exposure data of the whole chain, and real-time temperature of the storage environment. The core output parameters are the remaining safe shelf life of milk powder and the corresponding deterioration risk level.

[0158] Specifically, the core input parameters of the model include: the total content of initial characteristic compounds C0 measured at the factory of milk powder, the real-time updated cumulative exposure data of temperature and time throughout the entire chain, the real-time absolute temperature T of the storage environment, and the real-time collaborative amplification factor k calculated by S42.

[0159] The core output parameters of the model are: the total content C of the characteristic compounds corresponding to the prediction time t, the remaining safe shelf life of the milk powder when it reaches the critical trigger threshold, and the corresponding degradation risk level of the synergistic degradation loop; and ensure that the collection specifications and data accuracy requirements of all input parameters are met to ensure the accuracy of the model input.

[0160] S434. The model prediction results are corrected by periodically measuring the total content of characteristic compounds. Based on the deviation between the predicted value and the measured value, the model parameters are dynamically adjusted by a rolling optimization method so that the prediction matching degree is always kept above the preset threshold.

[0161] Specifically, a rapid sample testing rule is adopted, and milk powder samples are tested every 7 days. The total content of characteristic compounds measured in the test is used to correct the error of the model prediction results. When the relative deviation between the predicted value and the measured value is >5%, a rolling optimization method is adopted to dynamically adjust the model's co-amplification coefficient and pre-exponential factor parameters based on the latest measured data. Through continuous parameter correction, the model's prediction matching degree for the remaining shelf life is ensured to remain above 98%.

[0162] S44. Based on the measured results of the sample detection rules and the output results of the shelf life dynamic prediction model, establish graded early warning rules corresponding to different deterioration risk levels, and feed the early warning results back to the abnormal event identification stage in real time to complete the dynamic optimization of the control rules.

[0163] In the description of this invention, based on the measured results of sample detection rules and the output results of the shelf-life dynamic prediction model, graded early warning rules corresponding to different deterioration risk levels are established, and the warning results are fed back to the abnormal event identification stage in real time, completing the dynamic optimization of control rules, including:

[0164] S441. Based on the remaining safe shelf life output by the graded quantitative threshold and the dynamic shelf life prediction model, the quality deterioration risk of milk powder products is divided into different risk levels, and a percentage threshold of the remaining shelf life corresponding to each level is set.

[0165] Specifically, based on the grading and quantification thresholds and the remaining safe shelf life output by the model, the risk of deterioration in milk powder quality is divided into three levels: 1. Low risk: total content of characteristic compounds < 0.25 mg / kg, remaining safe shelf life ≥ 28 days, corresponding to general warning; 2. Medium risk: total content of characteristic compounds 0.25 mg / kg~0.57 mg / kg, remaining safe shelf life 14~28 days, corresponding to important warning; 3. High risk: total content of characteristic compounds > 0.57 mg / kg, remaining safe shelf life < 14 days, corresponding to emergency warning; the thresholds corresponding to each risk level are clearly defined to achieve accurate grading of deterioration risk.

[0166] S442. Set corresponding early warning trigger conditions for different risk levels, and set the target audience and response time limit for each level of early warning.

[0167] Specifically, tiered early warning triggering rules are established for different risk levels: 1. General warning: Triggered when the cumulative temperature exposure reaches 80% of the upper limit of the safe range under low-risk conditions, and pushed to warehouse / terminal management personnel, with a response time limit of 24 hours; 2. Important warning: Triggered immediately under medium-risk conditions, and pushed to quality control department and supply chain management personnel, with a response time limit of 12 hours; 3. Emergency warning: Triggered immediately under high-risk conditions, and pushed to enterprise management, quality control, legal, and supply chain departments, with a response time limit of 4 hours.

[0168] S443. Feed back the graded early warning results, the corresponding deterioration risk level, remaining shelf life, and recommended handling measures to the abnormal event identification system as input parameters for dynamic adjustment of control rules.

[0169] Specifically, the results of graded early warnings, deterioration risk assessment reports, remaining shelf life data, and standardized handling recommendations will be synchronized in real time to the S3 full-chain management system as the core reference for identifying abnormal temperature events, determining risk levels, and handling emergencies. At the same time, the early warning data will be included in the monthly management effectiveness evaluation as the core input parameter for dynamic optimization of management rules, adjustment of monitoring frequency, and updating of handling procedures, forming a complete quality control closed loop.

[0170] Please see Figure 5 The present invention also provides a system for improving the shelf-life stability of infant formula milk powder, the system comprising:

[0171] Mechanism verification module 1 is used to establish a synergistic deterioration positive feedback loop among various deterioration reactions during the storage of infant formula milk powder through accelerated storage experiments, and to set the key nodes of the synergistic deterioration positive feedback loop and the critical trigger value for its activation. A graded quantitative threshold is established based on the critical trigger value.

[0172] Component synergy module 2 is used to select corresponding targeted blocking functional components based on key nodes of the synergistic deterioration positive feedback loop, and to combine and integrate all functional components to obtain a multi-target synergistic blocking milk powder product.

[0173] The dynamic control module 3 is used to divide the temperature control range into graded ranges and set corresponding control rules based on the critical trigger value of the collaborative deterioration positive feedback loop and the deterioration kinetics of the finished milk powder product. It synchronously collects the cumulative exposure data of temperature and humidity throughout the entire chain and dynamically adjusts the control rules.

[0174] Module 4, the shelf-life prediction module, is used to establish sample testing rules adapted to finished milk powder products based on graded quantification thresholds. It combines synergistic degradation positive feedback loops, degradation kinetics, and exposure data to fit synergistic amplification coefficients for various degradation reactions, constructing a dynamic shelf-life prediction model. Based on sample testing results and model outputs, a graded early warning system is established to achieve full lifecycle management of finished milk powder products.

[0175] To further clarify the technical solution, complete implementation process, and repeatable technical effects of the present invention, the following specific embodiments and parallel comparative examples are provided.

[0176] This embodiment fully implements all technical steps S1-S4 of the present invention, and completely reproduces the entire process of improving the shelf-life stability of infant formula milk powder. The specific operation is as follows:

[0177] Step S1: Cooperative deterioration loop verification and threshold system establishment:

[0178] 1. Commercially available Stage 1 infant formula milk powder conforming to GB10765-2021 was used as a uniform basic sample. Four temperature gradient experimental groups were set up at 25℃, 30℃, 37℃ and 42℃, with 6 parallel samples in each group. The relative humidity of the environment in all groups was kept constant at 55%, and all other non-temperature conditions were completely consistent. Samples were taken simultaneously on days 0, 3, 6, 9, 12, 15, 18, 21, 24 and 27 of the experiment. After sampling, the samples were immediately sealed and protected from light, and frozen at 20℃.

[0179] 2. The contents of three core indicators—lipid oxidation, protein oxidation, and Maillard reaction—and six characteristic volatile compounds were simultaneously detected. Pearson correlation analysis was performed using SPSS 26.0 software, which confirmed that the pairwise positive correlation coefficients of the three types of deterioration reactions were all >0.92 (p<0.01). Through single-node blocking causal verification experiments, it was confirmed that after blocking any reaction, the rates of the other two types of reactions decreased significantly (p<0.01), clarifying the closed-loop causal promoting relationship of the three major deterioration reactions.

[0180] 3. Clarify the three core key nodes of the synergistic deterioration loop: initiation triggering, conduction amplification, and accelerated runaway; use Origin2023 software to perform nonlinear piecewise regression fitting on the full-cycle kinetic data to determine that the total content of 6 characteristic compounds of 0.82 mg / kg is the critical triggering threshold for loop initiation, and establish a three-level quantitative threshold system: safety threshold < 0.25 mg / kg, warning threshold 0.25~0.57 mg / kg, and runaway threshold > 0.57 mg / kg.

[0181] Step S2, Preparation of the finished milk powder product with synergistic blocking of three targets:

[0182] 1. Using sodium octenyl succinate starch (addition amount 0.8g / kg) as the inner wall material and whey protein and maltodextrin in a mass ratio of 2:1 as the outer wall material, DHA, ARA, linoleic acid and linolenic acid were microencapsulated. The microencapsulation was prepared by high pressure homogenization at 40MPa and spray drying at 180℃ / 80℃, with a fat encapsulation rate ≥98.5%. A compound antioxidant system of vitamin E (100mg / kg) and ascorbate palmitate (33mg / kg) in a mass ratio of 3:1 was premixed to obtain the first target functional component.

[0183] 2. Whey protein was pre-modified using a wet glycosylation process. The reaction endpoint was controlled at a degree of glycosylation of 15% ± 2% and a browning index ΔE < 5. The pre-modified whey protein and casein were optimized to obtain a protein base by mass ratio of 6.5:3.5. A compound blocking system of L-cysteine ​​(6 mg / kg) and tea polyphenols (200 mg / kg) with a mass ratio of 3:100 was prepared and premixed to obtain the second target functional component.

[0184] 3. Combine sodium citrate (0.5g / kg) acidity regulator and silica (5g / kg) anticaking agent, and premix evenly to obtain the third target functional component.

[0185] 4. Based on GB10765-2021, a basic nutritional framework was established. The three groups of targeted functional components were dry-mixed and compounded with basic raw materials such as lactose, compound vitamins, and compound minerals. Aseptic nitrogen filling was used, and the residual oxygen content of each can was ≤2%. This resulted in a finished milk powder product with synergistic blocking of three targets. The initial total content of the six characteristic compounds was 0.12 mg / kg when the product was tested before leaving the factory, which met the safety threshold requirements.

[0186] Step S3: Execution of dynamic hierarchical control across the entire chain:

[0187] 1. Based on the critical trigger threshold and the deterioration kinetics of finished milk powder products, three control zones are defined: safe storage zone ≤25℃, warning storage zone 25~30℃, and prohibited storage zone >30℃. The cumulative safe storage time rules for each zone are clearly defined.

[0188] 2. During the production and manufacturing process, an initial quality file is established for each batch of products, and the label indicates the graded storage guidelines; during the logistics and warehousing process, an IoT temperature and humidity recorder is used to collect data every 15 minutes, and the ambient temperature is stably controlled at ≤25℃; during the terminal display process, the temperature and humidity are monitored 3 times a day, and the product is kept away from heat sources; during the household consumption process, clear specifications are provided for storage and powder dispensing after opening the can.

[0189] 3. Set up monitoring points at five core nodes: production, transfer, storage, display, and can opening, continuously collect temperature and humidity data, calculate the cumulative temperature exposure over time throughout the entire chain to form a standardized dataset and update it in real time.

[0190] 4. Establish a three-tiered mechanism for handling abnormal events, and optimize control rules in reverse based on early warning results.

[0191] Step S4: Implementation of the rapid detection and dynamic shelf-life early warning system:

[0192] 1. Using the headspace sampling-gas chromatography-ion mobility spectrometry detection method disclosed in this invention, the quantitative detection of 6 characteristic compounds and the determination of the deterioration state of milk powder can be completed within 2 hours.

[0193] 2. Based on single-node blocking experimental data, a functional relationship between the synergistic amplification factor and absolute temperature was obtained by fitting the data. Combined with the cumulative exposure data of the entire chain, the real-time synergistic amplification factor was calculated.

[0194] 3. Based on the Arrhenius dynamic equation, a real-time collaborative amplification factor is introduced to construct a dynamic prediction model for the remaining shelf life. The model input and output parameters, fixed fitting values, and correction rules are defined to ensure that the prediction matching degree is ≥98%.

[0195] 4. Establish a three-level risk early warning system, and feed the early warning results back to the S3 control system in real time to achieve dynamic optimization of the control rules and form a closed loop of quality control throughout the entire life cycle.

[0196] To verify the synergistic effect of the technical solution of the present invention, four parallel comparative examples were set up. Except for the explicitly defined variables, the basic formula, experimental environment, and storage conditions were completely consistent with those of Example 1:

[0197] Comparative Example 1: Commercially available ordinary Stage 1 infant formula milk powder (compliant with GB10765-2021), without the three-target component of this invention, without graded dynamic control, and using the industry's conventional storage method.

[0198] Comparative Example 2: Only the first target functional component of the present invention (lipid oxidation source blocking) was used, and the rest of the basic formulation was the same as that of Example 1. There was no graded dynamic control, and conventional storage methods were used.

[0199] Comparative Example 3: Only the second and third target functional components of the present invention are used, and the rest of the basic formulation is the same as that of Example 1. There is no graded dynamic control, and conventional storage method is used.

[0200] Comparative Example 4: Using existing conventional antioxidant formulations, only vitamin E 100mg / kg was added, with no other targeted components. The rest of the basic formulation was the same as in Example 1, without graded dynamic control, and conventional storage methods were used.

[0201] All validation experiments used constant temperature and humidity accelerated storage conditions of 37℃ and 55% relative humidity to simulate the home storage scenario after opening the milk powder can. Six parallel samples were set up in each group. Experimental data are expressed as mean ± standard deviation. Differences between groups were analyzed using t-tests. p<0.05 was considered significant and p<0.01 was considered highly significant.

[0202] 1. Comparison of core deterioration indicators (stored for 28 days), the results are shown in Table 1.

[0203] Table 1 Comparison of core indicators of deterioration

[0204]

[0205] Experimental results confirm that the total amount of the six characteristic compounds generated in Example 1 of this invention is reduced by 82.3% compared with Comparative Example 1, far exceeding the expected effect of simple superposition of single target components. At the same time, it achieves the simultaneous inhibition of the three major deterioration reactions, completely solving the pain point of the prior art that is incomplete in addressing one aspect while neglecting another.

[0206] 2. Comparison of shelf-life stability: The longest storage time during which the total content of the six characteristic compounds in the milk powder reaches the critical trigger threshold of 0.82 mg / kg and all physicochemical indicators meet the national standards is defined as the shelf-life stability time. The experimental results are as follows:

[0207] Example 1: Stability duration ≥ 45 days under accelerated conditions at 37℃, stability duration ≥ 45 days after opening at room temperature (25℃), and stability duration ≥ 18 months in sealed condition without opening;

[0208] Comparative Example 1: The stabilization time under accelerated conditions at 37℃ was only 28 days, and the stabilization time after opening at room temperature was only 28 days, which is consistent with the industry standard.

[0209] Comparative Examples 2-4: The stabilization time under accelerated conditions at 37℃ was 30-35 days, which was significantly shorter than that of Example 1 of the present invention (p<0.01).

[0210] 3. Validation of the accuracy of the shelf life prediction model: The dynamic prediction model of this invention was used to predict the remaining shelf life of milk powder in Example 1 at 25℃, 30℃, and 37℃. The results were compared with the actual storage days. The results showed that the average prediction matching degree of the model of this invention was ≥98.2%, and the prediction error was ≤±1 day. In contrast, the average prediction matching degree of the conventional Arrhenius single-index model of the prior art was only 39.5%, and the prediction error was more than ±7 days. This confirms that the model of this invention has extremely high prediction accuracy and practicality.

[0211] In summary, by utilizing the technical solutions described above in this invention, and through experimental methods involving basic kinetic correlation analysis and single-node blocking causality verification, the synergistic positive feedback relationship among the three core deterioration reactions—lipid oxidation, protein oxidation, and Maillard reaction—during the storage of infant formula milk powder was clarified. The complete transmission pathway of the deterioration reaction from initiation to acceleration was elucidated, and three corresponding key nodes were identified. Furthermore, by analyzing the intrinsic correlation of the synergistic deterioration of multiple reactions in milk powder, a clear targeted design basis was provided for improving the shelf-life stability of milk powder, solving the problems of existing technologies focusing on single deterioration reactions and insufficient targeted regulation. By constructing a stability improvement scheme combining targeted blocking of the formula with full-chain process control, corresponding targeted blocking functional components were designed for the three key nodes to inhibit the occurrence and transmission of deterioration reactions at the product formulation level. Simultaneously, based on the critical triggering characteristics of deterioration, graded temperature control ranges were defined, and implementable control rules covering the entire process from production, logistics, terminal, to consumption were formulated, achieving synergistic adaptation between formulation design and process control, effectively improving the storage stability of milk powder throughout its entire life cycle. The detection method, built using headspace sampling, gas chromatography, and ion mobility spectrometry, eliminates the need for complex pretreatment and can quickly determine the deterioration state of milk powder, meeting the quality control needs of the entire industry chain. Furthermore, by introducing a correction factor based on the synergistic effect of deterioration, a dynamic shelf-life prediction model adapted to the entire lifecycle of milk powder has been constructed. This model can accurately predict shelf life by combining actual storage data and can also optimize control rules based on early warning results, forming a complete quality control closed loop that effectively ensures the nutritional quality and safety of milk powder throughout its entire lifecycle.

[0212] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

Claims

1. A method for improving the shelf-life stability of infant formula milk powder, characterized in that, include: S1. A synergistic deterioration positive feedback loop among various deterioration reactions during the storage of infant formula milk powder was established through accelerated storage experiments, and the key nodes of the synergistic deterioration positive feedback loop and the critical trigger value for the activation of the synergistic deterioration positive feedback loop were set. Establish a graded quantization threshold based on the critical trigger value; S2. Based on the key nodes of the synergistic deterioration positive feedback loop, select the corresponding targeted blocking functional components, and combine and integrate all functional components to obtain a multi-target synergistic blocking milk powder product. S3. Based on the critical trigger value of the collaborative deterioration positive feedback loop, combined with the deterioration kinetics of finished milk powder products, divide the temperature control range into graded ranges and set corresponding control rules; simultaneously collect the cumulative exposure data of temperature and humidity throughout the entire chain, and dynamically adjust the control rules; S4. Based on graded and quantified thresholds, establish sample testing rules suitable for finished milk powder products; and combine the synergistic deterioration positive feedback loop, deterioration kinetic characteristics and exposure data to fit the synergistic amplification coefficients of various deterioration reactions and construct a dynamic prediction model for shelf life; establish graded early warning based on sample testing results and model output to achieve full life cycle management of finished milk powder products.

2. The method for improving the shelf-life stability of infant formula milk powder according to claim 1, characterized in that, The experiment establishes a synergistic deterioration positive feedback loop among various deterioration reactions during the storage of infant formula milk powder through accelerated storage experiments, and sets the key nodes of the synergistic deterioration positive feedback loop and the critical trigger value for the activation of the synergistic deterioration positive feedback loop. Establishing hierarchical quantization thresholds based on critical trigger values ​​includes: S11. Using infant formula milk powder as a unified basic sample, with fixed relative humidity and non-temperature variable parameters, multiple sets of controllable parallel storage experiments were configured, and full-cycle change data of various degradation reactions were collected simultaneously. The interaction relationship between different degradation reactions was quantitatively verified through statistical analysis. The degradation reactions include lipid oxidation, protein oxidation and Maillard reaction. S12. Based on the quantitative verification results, sort out the transmission path and mutual promotion logic between different degradation reactions, divide the complete process of synergistic degradation from the beginning to accelerated runaway, and select the key nodes of the synergistic degradation positive feedback loop, including the initiation trigger node, the transmission amplification node and the accelerated runaway node. S13. Collect full-cycle kinetic data of the deterioration characteristic indicators of formula milk powder under different storage conditions as they change with storage time, and set the critical trigger value for the activation of the synergistic deterioration positive feedback loop by analyzing the deterioration rate change law of the synergistic deterioration positive feedback loop. S14. Based on the critical trigger value, and combined with the different startup states of the loop, the hierarchical quantization thresholds for the three states of the corresponding loop—not started, pre-started, and started—are obtained.

3. The method for improving the shelf-life stability of infant formula milk powder according to claim 2, characterized in that, The process involves collecting full-cycle kinetic data on the changes in the degradation characteristic indicators of formula milk powder under different storage conditions over storage time, and analyzing the degradation rate variation law of the synergistic degradation positive feedback loop. The critical trigger values ​​for initiating the synergistic degradation positive feedback loop are then defined as follows: S131. Obtain quantitative detection data of deterioration characteristic indicators of formula milk powder under different temperature gradients and different storage times in the full cycle change data, and form full cycle kinetic data of deterioration characteristic indicators changing with storage time and storage temperature. S132. Nonlinear piecewise regression fitting is performed on the full-cycle dynamic data using nonlinear regression analysis software. Piecewise function equations of the deterioration characteristic index as a function of storage time are obtained under different temperature conditions, and the deterioration rate variation law of different stages is divided. S133. Based on the piecewise function equation obtained by nonlinear piecewise regression fitting, select the abrupt change point in the change of degradation rate that causes an exponential leap, and determine the abrupt change point as the critical inflection point for the start of the collaborative degradation positive feedback loop; and extract the degradation characteristic index content value corresponding to the critical inflection point, and use the content value as the critical trigger value for the start of the collaborative degradation positive feedback loop.

4. The method for improving the shelf-life stability of infant formula milk powder according to claim 1, characterized in that, The key nodes of the synergistic deterioration positive feedback loop are selected with corresponding targeted blocking functional components, and all functional components are combined and integrated to obtain a multi-target synergistic blocking milk powder product, including: S21. Based on the initiation triggering node of the synergistic deterioration positive feedback loop, with the goal of blocking the triggering source of lipid oxidation initiation and switching loop, a source-blocking functional component for lipid oxidation is selected as the first target functional component. S22. Based on the conduction amplification node of the synergistic deterioration positive feedback loop, with the goal of cutting off the conduction pathway from lipid oxidation to protein oxidation and Maillard reaction, a blocking functional component targeting trans-reaction conduction is selected as the second target functional component. S23. Based on the accelerated runaway node of the synergistic deterioration positive feedback loop, with the goal of inhibiting the reverse catalysis of the Maillard reaction on lipid oxidation and protein reaction and terminating the closed loop positive feedback, a blocking functional component targeting the loop closure is selected as the third target functional component. S24. Select the first target functional component, the second target functional component, and the third target functional component, and combine and integrate them according to nutritional requirements and addition limits to obtain a multi-target synergistic blocking milk powder product.

5. The method for improving the shelf-life stability of infant formula milk powder according to claim 1, characterized in that, The critical trigger value based on the synergistic deterioration positive feedback loop, combined with the deterioration kinetics of finished milk powder, is used to divide the temperature control range into graded zones and set corresponding control rules; synchronously collecting cumulative temperature and humidity exposure data across the entire chain, and dynamically adjusting the control rules, includes: S31. Based on the critical trigger value of the collaborative deterioration positive feedback loop, and combined with the kinetic analysis results of the measured deterioration rate of finished milk powder at different temperatures, a graded temperature control interval is obtained, and a corresponding temperature range and cumulative safe storage time rule are set for each temperature control interval; the graded temperature control interval includes a safe storage interval, a warning storage interval, and a prohibited storage interval. S32. Integrate the entire life cycle of finished milk powder products from production to consumption, and formulate control rules that are compatible with temperature control ranges and can be implemented. The entire life cycle includes production and delivery, logistics and warehousing, terminal display and terminal consumption. S33. Continuously collect temperature and humidity data throughout the entire life cycle of the finished milk powder product, and simultaneously record the cumulative storage time of each stage to obtain complete cumulative temperature and humidity exposure data for the entire chain. S34. Based on the collected exposure data, automatically identify abnormal events that exceed the safe storage range, trace the stage and duration of the abnormal event, and trigger the corresponding handling plan; and combine the early warning results output from step S4 to dynamically optimize and adjust the control rules in a closed loop.

6. The method for improving the shelf-life stability of infant formula milk powder according to claim 5, characterized in that, The critical trigger value based on the synergistic deterioration positive feedback loop, combined with the kinetic analysis results of the measured deterioration rate changes of finished milk powder at different temperatures, divides the temperature control intervals into graded ranges. For each temperature control interval, corresponding temperature ranges and cumulative safe storage time rules are set, including: S311. Conduct a multi-gradient temperature full-cycle storage experiment on the finished milk powder, obtain storage experiment data, and fit the degradation rate constant of the finished milk powder in different temperature ranges to quantify the influence of ambient temperature on the trigger probability and trigger time of the synergistic degradation positive feedback loop. S312. Based on storage experiments and quantitative results, the temperature range in which the synergistic deterioration positive feedback loop has no risk of being triggered and can fully guarantee the shelf life of the preset number of days is defined as the safe storage range for finished milk powder products, and the longest storage time and recommended consumption period within the safe storage range are selected. S313. Based on storage experiments and quantification results, the temperature range in which the collaborative deterioration positive feedback loop has a pre-start risk and the deterioration rate is increased by more than a preset multiple compared to the safe storage range is defined as the warning storage range, and the upper limit of the cumulative safe storage time of the warning storage range is selected. S314. Based on storage experiments and quantification results, the temperature range that can directly trigger the collaborative degradation loop within a preset time period is defined as the prohibited storage range, and the longest allowable cumulative storage time and abnormal handling signal within the prohibited storage range are selected.

7. The method for improving the shelf-life stability of infant formula milk powder according to claim 5, characterized in that, Based on the collected exposure data, the system automatically identifies abnormal events that exceed the safe storage range, traces the stage and duration of the abnormal event, and triggers corresponding handling measures. In conjunction with the early warning results from the output feedback, the control rules are dynamically optimized and adjusted in a closed loop, including: S341. Based on the exposure data collected across the entire chain, identify abnormal events that exceed the safe storage range, classify the abnormal events into different abnormal levels according to the abnormal temperature range and cumulative exposure duration, and select the trigger threshold for each level. S342. Based on the identified abnormal events, trace the stage where the abnormal event occurred, the start and end time, the cumulative temperature and time exposure, and form a complete abnormal event tracing file; S343. For abnormal events at different levels, implement emergency response plans and trigger corresponding sample testing rules to perform dynamic shelf-life prediction and graded early warning. S344. Based on the risk classification and early warning results of milk powder quality deterioration, combined with the statistical analysis of abnormal events in the whole chain, the temperature control threshold, monitoring frequency and emergency response process of each link are optimized in reverse to realize the dynamic adjustment of process control rules.

8. The method for improving the shelf-life stability of infant formula milk powder according to claim 1, characterized in that, The method establishes sample detection rules adapted to finished milk powder products based on graded quantification thresholds; and combines the synergistic degradation positive feedback loop, degradation kinetic characteristics and exposure data to fit the synergistic amplification coefficients of various degradation reactions and construct a dynamic shelf life prediction model. A tiered early warning system is established based on sample testing results and model output to achieve full lifecycle management of finished milk powder products, including: S41. Based on graded quantification thresholds, establish sample detection rules for headspace sampling, gas chromatography, and ion mobility spectrometry that are adapted to the current matrix characteristics of milk powder products, and determine the content of characteristic compounds. S42. Based on the synergistic degradation relationship, the basic synergistic amplification coefficient of different degradation reactions under different isothermal conditions is fitted, and the basic synergistic amplification coefficient is corrected according to the degradation rate contribution of different temperature ranges; combined with the actual storage temperature history and cumulative storage time of the finished milk powder, the real-time synergistic amplification coefficient at the corresponding time node is calculated. S43. Based on the real-time collaborative amplification factor and the Arrhenius dynamic equation, a dynamic shelf-life prediction model adapted to the entire life cycle of milk powder products is constructed. S44. Based on the measured results of the sample detection rules and the output results of the shelf life dynamic prediction model, establish graded early warning rules corresponding to different deterioration risk levels, and feed the early warning results back to the abnormal event identification stage in real time to complete the dynamic optimization of the control rules.

9. The method for improving the shelf-life stability of infant formula milk powder according to claim 4, characterized in that, The shelf-life dynamic prediction model adapted to the entire life cycle of finished milk powder products, based on the real-time collaborative amplification factor and the Arrhenius kinetic equation, includes: S431. Based on the Arrhenius kinetic equation, the real-time cooperative amplification factor is introduced into the Arrhenius kinetic equation as a correction factor to establish a dynamic prediction model framework for the remaining shelf life with the total content of characteristic compounds as the deterioration index and storage time and storage temperature as independent variables. S432. Through multi-gradient temperature accelerated storage experiments, collect full-cycle kinetic data of the content of characteristic compounds in finished milk powder at different temperatures with storage time, and use nonlinear regression method to fit and obtain fixed values ​​of pre-exponential factor and activation energy in the model. S433, the shelf life dynamic prediction model includes the measured initial characteristic compound content of milk powder, exposure data of the whole chain, and real-time temperature of the storage environment. The core output parameters are the remaining safe shelf life of milk powder and the corresponding deterioration risk level. S434. The model prediction results are corrected by periodically measuring the total content of characteristic compounds. Based on the deviation between the predicted value and the measured value, the model parameters are dynamically adjusted by a rolling optimization method so that the prediction matching degree is always kept above the preset threshold.

10. The method for improving the shelf-life stability of infant formula milk powder according to claim 4, characterized in that, The measured results based on the sample detection rules and the output results of the shelf-life dynamic prediction model are used to establish graded early warning rules corresponding to different deterioration risk levels. The early warning results are fed back to the abnormal event identification stage in real time, and the dynamic optimization of the control rules is completed, including: S441. Based on the remaining safe shelf life output by the graded quantitative threshold and the dynamic prediction model of shelf life, the risk of quality deterioration of milk powder products is divided into different risk levels, and the remaining shelf life percentage threshold corresponding to each level is set respectively. S442. Set corresponding early warning trigger conditions for different risk levels, and set the target audience and response time limit for each level of early warning; S443. Feed back the graded early warning results, the corresponding deterioration risk level, remaining shelf life, and recommended handling measures to the abnormal event identification system as input parameters for dynamic adjustment of control rules.

Images