Production and preparation method of functional polypeptide formed by enzymolysis of soybean meal and peanut meal

Abstract

The invention discloses a production and preparation method of functional polypeptide formed by enzymolysis of soybean meal and peanut meal, and belongs to the technical field of bioengineering. According to the method, the problems that the enzymolysis process is unstable and the polypeptide yield and the activity index are greatly fluctuated due to the fact that an existing method depends on correlation analysis, adopts fixed parameters to control the enzymolysis process and lacks causal inference and a real-time dynamic adjustment mechanism are solved, and the contact area of enzyme and a substrate is increased through a two-stage crushing mechanism; an ultrasonic-assisted enzymolysis technology is combined, so that the contact between enzyme and a substrate is enhanced; fat and fat-soluble anti-nutritional factors can be effectively removed by utilizing a gradient ethanol-supercritical combined extraction technology, and the purity of the raw materials is improved; impurities can be effectively removed through a multi-stage filtering mechanism and a nano-filtration technology, and the purity of the polypeptide is improved; the activity of the enzyme can be quickly inactivated by microwave-assisted enzyme deactivation treatment, and the potential influence of high temperature on the structure and activity of the polypeptide is reduced.

Description

Technical Field

[0001] This invention relates to the field of bioengineering technology, specifically to a method for producing functional peptides by enzymatic hydrolysis of soybean meal and peanut meal. Background Technology

[0002] Soybean meal and peanut meal are major byproducts of grain and oil processing, rich in protein, and are excellent raw materials for obtaining plant-derived protein peptides. Current technologies mostly employ a single enzyme or a simple complex enzyme for a single enzymatic hydrolysis under fixed conditions, but this approach has the following drawbacks:

[0003] Traditional grinding methods cannot fully break down the cell walls of soybean meal and peanut meal, resulting in a limited contact area between enzymes and substrates and low enzymatic hydrolysis efficiency. Degreasing methods using single ethanol extraction are difficult to effectively remove bound fats and fat-soluble anti-nutritional factors, resulting in insufficient purity of raw materials. Furthermore, conventional heating to inactivate enzymes may damage peptide structure and activity, reducing product quality.

[0004] Traditional enzymatic hydrolysis processes typically employ fixed parameter control, lacking real-time monitoring and dynamic adjustment mechanisms. This inability to dynamically adjust enzymatic hydrolysis parameters based on real-time data leads to instability in the enzymatic hydrolysis process, resulting in significant fluctuations in peptide yield and activity indicators.

[0005] Therefore, to meet current needs, a method for producing functional peptides by enzymatic hydrolysis of soybean meal and peanut meal is proposed. Summary of the Invention

[0006] The purpose of this invention is to provide a method for producing functional peptides by enzymatic hydrolysis of soybean meal and peanut meal. This method utilizes a two-stage pulverization mechanism to increase the contact area between the enzyme and the substrate; combines ultrasound-assisted enzymatic hydrolysis technology to enhance the contact between the enzyme and the substrate; and employs a gradient ethanol-supercritical fluid hydrolysis technique. The combined extraction technology can effectively remove fats and fat-soluble anti-nutritional factors, improving the purity of raw materials; the multi-stage filtration mechanism and nanofiltration technology can effectively remove impurities, improving the purity of peptides; the microwave-assisted enzyme inactivation treatment can rapidly inactivate enzyme activity, reducing the potential impact of high temperature on peptide structure and activity, thus solving the problems mentioned in the background technology.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A method for producing functional peptides by enzymatic hydrolysis of soybean meal and peanut meal includes the following steps:

[0009] S1: Grind the soybean meal and peanut meal separately and pass them through an 80-mesh sieve;

[0010] The pulverized soybean meal and peanut meal were degreased separately using a gradient ethanol-supercritical fluid degreasing process. Combined extraction: first soak in 60% ethanol solution for 12 hours to remove free fats;

[0011] Then, supercritical conditions were carried out at 30 MPa and 35℃. Extract for 5 hours to selectively remove bound fats and fat-soluble anti-nutritional factors, then dry after filtration;

[0012] Mix defatted soybean meal powder and peanut meal powder at a mass ratio of 1:2, add an appropriate amount of water to make the moisture content of the mixture reach 55%;

[0013] S2: Add the compound enzyme preparation to the mixture, adjust the pH of the mixture to 8.0, control the temperature at 50℃, and stir for 3 hours for enzymatic hydrolysis; during the enzymatic hydrolysis process, introduce ultrasonic-assisted enzymatic hydrolysis technology to enhance the contact between the enzyme and the substrate;

[0014] S3: Heat the enzymatically hydrolyzed material to 92℃ and maintain for 12 minutes to inactivate the enzyme; centrifuge the enzyme-inactivated material using variable frequency centrifugation technology, automatically adjusting the centrifugation speed according to the density and viscosity of the material, with a rotation speed of 4500 rpm and a centrifugation time of 12 minutes;

[0015] The supernatant was collected and filtered using a functional membrane element with a molecular weight of 5 kDa. Nanofiltration technology was introduced during the separation and purification process to effectively remove impurities through the nanofiltration membrane.

[0016] S4: Spray dry the filtered peptide solution to obtain functional peptide powder; perform quality testing on the final product, including peptide content, purity, antioxidant activity, and immunomodulatory activity indicators.

[0017] Further, in S1, the soybean meal and peanut meal are separately pulverized and passed through an 80-mesh sieve, including the following steps:

[0018] A two-stage pulverization mechanism is constructed. The first stage adopts low-temperature shear pulverization that mimics molars, simulating the human chewing mechanism, and pulverizes the raw materials to 40 mesh at <15℃, so as to maintain the natural conformation of the protein to the greatest extent.

[0019] The second stage uses high-frequency vibration grinding with dental enamel to further refine the material to 80 mesh through a vibration field of a specific frequency; at the same time, it generates nanoscale cracks to increase the enzyme contact area.

[0020] Further, the feature is that, in step S3, the supernatant is taken and filtered using a functional membrane element with a molecular weight of 5 kDa, including the following steps:

[0021] A multi-stage filtration mechanism is adopted. First, a membrane element with a molecular weight of 10kDa is used for coarse filtration to remove large molecular impurities, and then a membrane element with a molecular weight of 5kDa is used for fine filtration.

[0022] Bioactive factors, including coenzyme Q10, were added to the filtered peptide solution; and nanoemulsification technology was used to uniformly disperse coenzyme Q10 in the peptide solution, so that it could better combine with the peptide solution.

[0023] Furthermore, the compound enzyme preparation is composed of alkaline protease, papain, and flavor protease in an enzyme activity ratio of 1:2:1.

[0024] Furthermore, it also includes the following steps:

[0025] S5: Establish a quality traceability system based on RIFD radio frequency technology to record production process data and quality inspection results for each batch of products, thereby achieving full traceability and quality monitoring of products.

[0026] S6: An adaptive control strategy is introduced, which uses online sensors to monitor the pH, temperature and enzyme activity of the enzymatic hydrolysate in real time, and dynamically adjusts the enzymatic hydrolysis parameters to ensure the stability and efficiency of the enzymatic hydrolysis process.

[0027] Furthermore, in S6, an adaptive control strategy is introduced to dynamically adjust the enzymatic hydrolysis parameters, including the following steps:

[0028] Each batch of soybean meal / peanut meal raw material will be rapidly tested to obtain basic data on different batches of soybean meal and peanut meal, including: protein content, fat content, ash content, and various parameters during the enzymatic hydrolysis process, including: temperature, pH value, enzyme dosage, and hydrolysis time; as well as various indicators after enzymatic hydrolysis, including: peptide yield, purity, antioxidant activity, and immunomodulatory activity.

[0029] Near-infrared spectroscopy was introduced to rapidly detect the protein structure of soybean meal and peanut meal, and to predict the enzymatic hydrolysis efficiency of proteins and the potential activity of peptides.

[0030] The collected data were analyzed based on a decision tree model, with basic data and enzymatic hydrolysis parameters as input variables and peptide yield, purity, and activity indicators as output variables.

[0031] By training a decision tree model, the correlation coefficients between each input variable and the output variable are calculated to determine the key parameters and their impact on peptide preparation.

[0032] Furthermore, through training the decision tree model, the correlation coefficients between each input variable and the output variable are calculated, including the following steps:

[0033] Causal inference analysis is introduced, and multiple causal trees are constructed to comprehensively analyze the causal effect of each input variable and determine the causal influence of the input variable on the output variable.

[0034] For each input variable, calculate its causal effect on each output variable;

[0035] The impact of input variables on peptide preparation efficiency was evaluated by combining correlation coefficients and causal effects.

[0036] Based on the correlation coefficient and the magnitude of the causal effect, the key parameters and their impact on the peptide preparation effect are determined;

[0037] For each key parameter, assess its impact on different output variables;

[0038] Based on the evaluation results, a list of feature importance is generated, showing the relative importance of each input variable to the output variable; and the importance values ​​are normalized to form correlation coefficients.

[0039] Furthermore, after generating a list of feature importance based on the evaluation results, the following steps are included:

[0040] Obtain the list of feature importance generated by the decision tree model to understand the degree of influence of each input variable on the output variable;

[0041] The current state of the enzymatic hydrolysis process is assessed based on real-time data and predictions from the decision tree model.

[0042] Based on the prediction results of the decision tree model, formulate parameter adjustment rules;

[0043] Determine the adjustment range and step size for each parameter, and determine the frequency of parameter adjustment based on the dynamic characteristics of the enzymatic hydrolysis process and production requirements.

[0044] Furthermore, after generating the feature importance list based on the evaluation results, the following steps are also included:

[0045] The evaluation cycle is set according to the actual production process;

[0046] During the evaluation period, various data from the enzymatic hydrolysis process are continuously collected and analyzed to assess whether the adaptive control strategy has achieved the expected goals; the actual production data is compared with the prediction results of the decision tree model to analyze the reasons for the deviation.

[0047] Set evaluation indicators to measure the effectiveness of the adaptive control strategy; the evaluation indicators include: peptide yield increase rate, antioxidant activity achievement rate, parameter adjustment stability, and production cost change.

[0048] Based on the evaluation results, update the decision tree model, re-evaluate the influence of each input variable on the output variable, and update the feature importance list.

[0049] Based on the model update results and actual production data, optimize the parameter adjustment rules, including: redetermining the adjustment range and step size of each parameter based on the new model prediction results;

[0050] Based on the dynamic characteristics of the enzymatic hydrolysis process and production requirements, the frequency of parameter adjustment is redefined;

[0051] Adjust the adaptive control strategy based on the evaluation results.

[0052] Furthermore, based on the prediction results of the decision tree model, parameter adjustment rules are formulated, including the following steps:

[0053] The real-time monitoring data is compared with the prediction results of the decision tree model to assess the current status of the enzymatic hydrolysis process;

[0054] Based on the deviation between the prediction results of the decision tree model and real-time data, determine the direction and magnitude of parameter adjustments, including:

[0055] Based on the prediction results of the decision tree model, the temperature of the enzymatic hydrolysate is maintained within the predicted range to optimize peptide yield.

[0056] Based on the prediction results of the decision tree model, the pH value of the enzymatic hydrolysate was maintained within the predicted range to optimize the antioxidant activity of the peptides.

[0057] The amount of enzyme is dynamically adjusted according to changes in enzyme activity to ensure the efficient execution of the enzymatic hydrolysis reaction.

[0058] Based on changes in peptide yield and antioxidant activity, the enzymatic hydrolysis time is dynamically adjusted to optimize the quality of the final product.

[0059] Compared with the prior art, the beneficial effects of the present invention are:

[0060] In this invention, a two-stage pulverization mechanism increases the contact area between the enzyme and the substrate, thereby improving enzymatic hydrolysis efficiency. Combining this with ultrasound-assisted enzymatic hydrolysis technology further enhances the contact between the enzyme and the substrate, further improving hydrolysis efficiency. The invention also utilizes a gradient ethanol-supercritical fluid dynamics process. The combined extraction technology can effectively remove fats and fat-soluble anti-nutritional factors, improving the purity of raw materials; the multi-stage filtration mechanism and nanofiltration technology can effectively remove impurities, improving the purity of peptides; the nanoemulsification technology can uniformly disperse coenzyme Q10 in the peptide solution, enhancing the bioactivity of peptides; the microwave-assisted enzyme inactivation treatment can rapidly inactivate enzyme activity, reduce the potential impact of high temperature on peptide structure and activity, and further improve product stability. Attached Figure Description

[0061] Figure 1 This is a flowchart of the method for producing functional peptides by enzymatic hydrolysis of soybean meal and peanut meal according to the present invention. Detailed Implementation

[0062] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0063] To address the technical issues arising from existing methods' reliance on correlation analysis and fixed parameter control of the enzymatic digestion process, which lack causal inference and real-time dynamic adjustment mechanisms, resulting in unscientific optimization strategies, unstable enzymatic digestion processes, and large fluctuations in peptide yield and activity indicators, please refer to [link to relevant documentation]. Figure 1 This embodiment provides the following technical solution:

[0064] A method for producing functional peptides by enzymatic hydrolysis of soybean meal and peanut meal includes the following steps:

[0065] S1: Soybean meal and peanut meal are pulverized separately and passed through an 80-mesh sieve; including: constructing a two-stage pulverization mechanism. The first stage uses low-temperature shear pulverization that mimics molars to simulate the human chewing mechanism, pulverizing the raw materials to 40 mesh at <15℃ to maintain the natural conformation of the protein to the greatest extent. The second stage uses high-frequency vibration grinding that mimics tooth enamel, further refining the material to 80 mesh through a vibration field of a specific frequency; at the same time, it generates nanoscale cracks to increase the enzyme contact area.

[0066] The pulverized soybean meal and peanut meal were degreased separately using a gradient ethanol-supercritical fluid degreasing process. Combined extraction: first, soaking in 60% ethanol solution for 12 hours to remove free fats; then supercritical extraction at 30 MPa and 35℃. Extract for 5 hours to selectively remove bound fats and fat-soluble anti-nutritional factors, and then dry after filtration; mix defatted soybean meal powder and peanut meal powder at a mass ratio of 1:2, add an appropriate amount of water to make the moisture content of the mixture reach 55%.

[0067] S2: Add a compound enzyme preparation to the mixture. The compound enzyme preparation consists of alkaline protease, papain and flavor protease in an enzyme activity ratio of 1:2:1. Adjust the pH of the mixture to 8.0, control the temperature at 50℃, and stir for 3 hours for enzymatic hydrolysis. During the enzymatic hydrolysis process, introduce ultrasonic-assisted enzymatic hydrolysis technology to enhance the contact between the enzyme and the substrate.

[0068] S3: Heat the enzymatically hydrolyzed material to 92℃ and maintain for 12 minutes to inactivate the enzyme; centrifuge the enzyme-inactivated material using variable frequency centrifugation technology, automatically adjusting the centrifugation speed according to the material's density and viscosity, at 4500 rpm for 12 minutes; collect the supernatant and filter it using a 5kDa functional membrane element, including: employing a multi-stage filtration mechanism, first using a 10kDa membrane element for coarse filtration to remove large molecular impurities; then using a 5kDa membrane element for fine filtration; adding bioactive factors, including coenzyme Q10, to the filtered peptide solution; and using nanoemulsification technology to uniformly disperse coenzyme Q10 in the peptide solution, allowing it to better combine with the peptide solution; during the separation and purification process, nanofiltration technology is introduced to effectively remove impurities through the nanofiltration membrane.

[0069] S4: Spray dry the filtered peptide solution to obtain functional peptide powder; perform quality testing on the final product, including peptide content, purity, antioxidant activity, and immunomodulatory activity indicators.

[0070] The beneficial effects achieved by the above are as follows: the two-stage pulverization mechanism increases the contact area between the enzyme and the substrate, thus improving the enzymatic hydrolysis efficiency; the combination of ultrasound-assisted enzymatic hydrolysis technology further enhances the contact between the enzyme and the substrate, improving the enzymatic hydrolysis efficiency; and the utilization of gradient ethanol-supercritical fluid dynamics further enhances the enzymatic hydrolysis efficiency. The combined extraction technology can effectively remove fats and fat-soluble anti-nutritional factors, improving the purity of raw materials; the multi-stage filtration mechanism and nanofiltration technology can effectively remove impurities, improving the purity of peptides; the nanoemulsification technology can uniformly disperse coenzyme Q10 in the peptide solution, enhancing the bioactivity of peptides; the microwave-assisted enzyme inactivation treatment can rapidly inactivate enzyme activity, reduce the potential impact of high temperature on peptide structure and activity, and further improve product stability.

[0071] S5: Establish a quality traceability system based on RIFD radio frequency technology to record production process data and quality inspection results for each batch of products, thereby achieving full traceability and quality monitoring of products.

[0072] S6: An adaptive control strategy is introduced, which uses online sensors to monitor the pH, temperature, and enzyme activity of the hydrolysate in real time, and dynamically adjusts the hydrolysis parameters to ensure the stability and efficiency of the hydrolysis process; including the following steps:

[0073] Rapid testing was performed on each batch of soybean meal / peanut meal raw materials to obtain basic data on different batches of soybean meal and peanut meal, including: protein content, fat content, and ash content. Various parameters during the enzymatic hydrolysis process were recorded, including: temperature, pH value, enzyme dosage, and hydrolysis time. Various indicators after enzymatic hydrolysis were measured, including: peptide yield, purity, antioxidant activity, and immunomodulatory activity.

[0074] Near-infrared spectroscopy was introduced to rapidly detect the protein structure of soybean meal and peanut meal, predicting the enzymatic hydrolysis efficiency of proteins and the potential activity of peptides; the steps included are as follows:

[0075] Spectral data of soybean meal and peanut meal powders were acquired using a near-infrared spectrometer, covering the wavelength range of 900-2500 nm. The acquired spectral data underwent preprocessing, including scattering correction, differential processing, and wavelength selection, to eliminate physical interference and enhance chemical information, obtaining a set of spectral variables reflecting the protein structural characteristics of the samples. This set of spectral variables was then input into an established partial least squares regression prediction model. By analyzing the mapping relationship between spectral characteristics and enzymatic hydrolysis kinetic parameters in the calibration database, a quantitative prediction value of the enzymatic hydrolysis efficiency of the samples was output. The predicted values ​​include: the predicted time required to reach the target degree of hydrolysis under standard enzymatic hydrolysis conditions and the predicted maximum theoretical degree of hydrolysis. The prediction model includes at least one of the following: instantaneous degree of hydrolysis prediction curves at specific time points; inputting the same set of spectral variables into the established prediction model, and analyzing the correlation between spectral characteristics and the functional activities of peptide products in the calibration database, outputting the potential bioactivity prediction value of the peptide products obtained after the test sample undergoes standard enzymatic hydrolysis; the prediction value includes at least one of the following: in vitro antioxidant activity index of peptide products, half-maximal inhibitory concentration prediction value of angiotensin-converting enzyme inhibitory activity, and relative abundance prediction value of characteristic active peptides; integrating the prediction results to generate a comprehensive raw material quality prediction report with enzymatic hydrolysis efficiency rating and peptide activity profile, which is used to guide the selection of process route or value assessment of this batch of raw materials.

[0076] The collected data are analyzed based on a decision tree model, using basic data and enzymatic hydrolysis parameters as input variables, and peptide yield, purity, and activity indicators as output variables; the process includes the following steps:

[0077] The collected data is standardized to ensure fair comparison of data of different dimensions and orders of magnitude; the completeness and consistency of the data are checked, and missing and outlier values ​​are handled; the data is divided into training and test sets, typically in a 7:3 or 8:2 ratio; the decision tree model is trained using the training set data, progressively building the decision tree based on the relationship between input and output variables; by selecting the optimal split point, the data is continuously split into smaller subsets until a stopping condition is met, such as reaching the maximum tree depth or a node purity threshold; the model's performance is evaluated using the test set data to ensure its accuracy and generalization ability; within the decision tree model, the effect of each input variable on the output is evaluated through the model's split points and paths. The system assesses the impact of input variables and calculates the contribution of each input variable to the output variable. For example, if an input variable, such as temperature, is frequently used at multiple split points in the decision tree, and that split point has a significant impact on the prediction of the output variable, such as peptide yield, then the input variable is considered to have a high correlation with the output variable. Based on the assessment results, a feature importance list is generated, showing the relative importance of each input variable to the output variable. The importance values ​​are then normalized to form a correlation coefficient. The correlation coefficient ranges from -1 to 1; the larger the absolute value, the more significant the impact of the input variable on the output variable. For example, if the correlation coefficient for temperature is 0.8, it indicates that temperature has a significant positive correlation with peptide yield.

[0078] By training a decision tree model, the correlation coefficients between each input and output variable are calculated to determine key parameters and their impact on peptide preparation; this includes the following steps:

[0079] Causal inference analysis is introduced, and multiple causal trees are constructed to comprehensively analyze the causal effect of each input variable and determine the causal influence of input variables on output variables. For each input variable, its causal effect on each output variable is calculated. For example, through causal forest analysis, the causal effect of temperature on peptide yield is determined to be 0.75, indicating that increased temperature significantly improves peptide yield at the causal level. The impact of input variables on peptide preparation is evaluated by combining correlation coefficients and causal effects. For example, although the correlation coefficient of a certain input variable may be low, its causal effect may be high, indicating that this variable has a significant causal impact on the output variable. The magnitude of correlation coefficients and causal effects determines the key parameters and their impact on peptide preparation. For example, if the correlation coefficient for temperature is 0.8 and the causal effect is 0.75, it indicates that temperature is a key parameter affecting peptide yield. Similarly, if the correlation coefficient for pH is 0.6 and the causal effect is 0.65, it indicates that pH is also a key parameter affecting peptide purity. For each key parameter, its impact on different output variables is assessed. For example, temperature not only significantly affects peptide yield but may also have some impact on the antioxidant activity of peptides. Through comprehensive analysis, the overall impact of each key parameter in the peptide preparation process can be determined.

[0080] An adaptive control strategy is introduced, using online sensors to monitor the pH, temperature, and enzyme activity of the enzymatic hydrolysate in real time. Combined with the prediction results of a decision tree model, the enzymatic hydrolysis parameters are dynamically adjusted to ensure the stability and efficiency of the hydrolysis process. For example, if model analysis shows that temperature has a significant impact on peptide yield, the temperature can be fine-tuned within a certain range to improve peptide yield. The process includes the following steps:

[0081] Obtain the feature importance list generated by the decision tree model to understand the degree of influence of each input variable on the output variable; for example, if the model analysis finds that temperature has a significant impact on peptide yield, then temperature will be a key focus in subsequent dynamic adjustments; assess the current state of the enzymatic hydrolysis process based on real-time data and the prediction results of the decision tree model; for example, if real-time monitoring finds that the temperature deviates from the optimal range predicted by the decision tree model, then immediately activate the adjustment mechanism; formulate parameter adjustment rules based on the prediction results of the decision tree model; for example, if temperature has a significant positive correlation with peptide yield, and when the real-time monitored temperature is lower than the optimal value predicted by the decision tree model, then automatically adjust the heating system to increase the temperature; conversely, if the temperature is too high, then activate the cooling system to decrease the temperature; determine the adjustment range and step size for each parameter; for example, the temperature adjustment range can be set to ±2℃, with a step size of 0.5℃; determine the frequency of parameter adjustments based on the dynamic characteristics of the enzymatic hydrolysis process and production needs; for example, in the early stages of the enzymatic hydrolysis reaction, due to the rapid reaction rate, the adjustment frequency can be set to once every 5 minutes; in the later stages of the reaction, as the reaction rate slows down, the adjustment frequency can be reduced to once every 15 minutes.

[0082] Based on the prediction results of the decision tree model, parameter adjustment rules are formulated, including the following steps:

[0083] The real-time monitoring data is compared with the prediction results of the decision tree model to assess the current state of the enzymatic hydrolysis process. For example, if the real-time monitored temperature is 48℃, while the model predicts an optimal temperature of 50℃, there is a deviation of 2℃; if the real-time monitored pH value is 7.2, while the model predicts an optimal pH value of 7.5, there is a deviation of 0.3. Based on the deviation between the prediction results of the decision tree model and the real-time data, the direction and magnitude of parameter adjustments are determined, including:

[0084] Based on the prediction results of the decision tree model, the temperature of the enzymatic hydrolysate is maintained within the predicted range to optimize peptide yield. If the real-time temperature is lower than the optimal temperature predicted by the decision tree model, the heating system is activated to gradually increase the temperature until the optimal temperature range is reached. If the real-time temperature is higher than the optimal temperature predicted by the decision tree model, the cooling system is activated to gradually decrease the temperature until the optimal temperature range is reached. Each temperature adjustment is made in increments of 0.5℃ to avoid rapid temperature changes impacting the enzymatic hydrolysis process. In the initial stage of the enzymatic hydrolysis reaction (first 2 hours), the temperature is monitored every 10 minutes and adjusted every 30 minutes. In the later stage of the enzymatic hydrolysis reaction (after 2 hours), the temperature is monitored every 15 minutes and adjusted every 45 minutes.

[0085] Based on the prediction results of the decision tree model, the pH value of the enzymatic hydrolysate was maintained within the predicted range to optimize the antioxidant activity of the peptides. If the real-time pH value was lower than the optimal pH value predicted by the decision tree model, an alkaline regulator, such as sodium hydroxide solution, was added to gradually increase the pH value until the optimal pH value range was reached. If the real-time pH value was higher than the optimal pH value predicted by the decision tree model, an acidic regulator, such as hydrochloric acid solution, was added to gradually decrease the pH value until the optimal pH value range was reached. The pH value was adjusted by 0.1 each time to ensure stable pH changes. In the early stage of the enzymatic hydrolysis reaction (first 2 hours), the pH value was monitored every 10 minutes and adjusted every 30 minutes. In the later stage of the enzymatic hydrolysis reaction (after 2 hours), the pH value was monitored every 15 minutes and adjusted every 45 minutes.

[0086] The enzyme dosage is dynamically adjusted based on changes in enzyme activity to ensure efficient enzymatic hydrolysis. If the real-time monitored enzyme activity is lower than a set threshold, such as 80% of the initial enzyme activity, the enzyme dosage is automatically increased, for example, by 10% of the initial enzyme dosage each time. If the real-time monitored enzyme activity is higher than a set threshold, such as 120% of the initial enzyme activity, the enzyme dosage is automatically reduced, for example, by 5% of the initial enzyme dosage each time. Enzyme activity is monitored every 30 minutes and adjusted every 60 minutes.

[0087] Based on changes in peptide yield and antioxidant activity, the enzymatic hydrolysis time is dynamically adjusted to optimize the quality of the final product. If the real-time monitored peptide yield is lower than 80% of the optimal yield predicted by the decision tree model, and the antioxidant activity is lower than 70% of the optimal activity predicted by the decision tree model, the enzymatic hydrolysis time is extended by 15 minutes each time. If the real-time monitored peptide yield is higher than 120% of the optimal yield predicted by the decision tree model, and the antioxidant activity is higher than 110% of the optimal activity predicted by the decision tree model, the enzymatic hydrolysis time is shortened by 10 minutes each time. Peptide yield and antioxidant activity are monitored every 60 minutes, and adjustments are made every 120 minutes.

[0088] In one embodiment, assuming the decision tree model predicts that the peptide yield is highest at a temperature of 50°C; real-time monitoring reveals that the temperature of the enzymatic hydrolysate is 48°C, lower than the optimal value predicted by the decision tree model; then, the heating system is automatically activated at the control center according to preset adjustment rules to raise the temperature to 50°C; after adjustment, the peptide yield is significantly improved, consistent with the prediction result of the decision tree model; through real-time monitoring and dynamic adjustment, it is ensured that the enzymatic hydrolysis process is always carried out under optimal conditions, thereby improving peptide yield and quality.

[0089] Based on the actual production process, set an evaluation cycle; for example, for continuous enzymatic hydrolysis, a comprehensive evaluation can be conducted weekly; for intermittent enzymatic hydrolysis, an evaluation can be conducted immediately after each batch. During the evaluation cycle, continuously collect and analyze various data from the enzymatic hydrolysis process to assess whether the adaptive control strategy has achieved the expected goals. Compare the actual production data with the prediction results of the decision tree model to analyze the reasons for the deviation; for example, if the actual peptide yield is found to be lower than the model prediction, it is necessary to analyze whether it is due to inaccurate temperature control, decreased enzyme activity, or other factors. Set evaluation indicators to measure the effectiveness of the adaptive control strategy; evaluation indicators include: peptide yield increase rate, antioxidant activity achievement rate, parameter adjustment stability, and production cost changes. Based on the evaluation results, update the decision tree model and re-evaluate the influence of each input variable on the output variable, updating the features accordingly. Importance list; Based on model update results and actual production data, optimize parameter adjustment rules, including: redetermining the adjustment range and step size of each parameter based on the new model prediction results; for example: if the temperature adjustment range is found to be too wide, leading to reaction instability, then narrow the adjustment range; if the pH adjustment step size is found to be too small, resulting in low adjustment efficiency, then increase the step size; redetermining the frequency of parameter adjustment based on the dynamic characteristics of the enzymatic hydrolysis process and production needs; for example: if enzyme activity changes rapidly, the frequency of enzyme activity monitoring needs to be increased, and the frequency of enzyme dosage adjustment needs to be increased accordingly; adjusting the adaptive control strategy based on the evaluation results; for example: if the reaction is delayed or overshooted after adjusting certain parameters, then adding a more complex feedback mechanism, such as PID control; applying the updated model and optimized control strategy to actual production, continuously monitoring the implementation effect, and ensuring that the adjusted strategy can effectively improve production efficiency and product quality.

[0090] In one embodiment, it is assumed that an evaluation found that an excessively wide temperature adjustment range led to instability in the enzymatic hydrolysis reaction and large fluctuations in peptide yield. Based on the new prediction results of the decision tree model, the temperature adjustment range was narrowed from ±2℃ to ±1.5℃, and the step size was adjusted from 0.5℃ to 0.3℃. After optimization, the stability of the enzymatic hydrolysis reaction was significantly improved, the fluctuation of peptide yield was reduced, and the average yield increased by 8%.

[0091] The beneficial effects achieved by the above are as follows: By introducing decision tree models and causal inference analysis, it is possible not only to systematically analyze the relationship between input and output variables in the enzymatic hydrolysis of soybean meal and peanut meal, but also to further determine the causal influence of input variables on output variables, and to more accurately assess the impact of each input variable on peptide preparation, thereby determining key parameters and their degree of influence on different output variables; this not only improves the efficiency and quality of peptide preparation, but also ensures the stable improvement of peptide yield and activity indicators, thus achieving a more efficient and stable peptide preparation process in actual production.

[0092] Working principle: Near-infrared spectroscopy is used to rapidly detect the protein structure of raw materials and predict enzymatic hydrolysis efficiency and peptide activity. Based on decision tree model and causal inference analysis, the influence of each input variable on peptide yield, purity and activity is comprehensively evaluated to determine key parameters and their degree of influence. An adaptive control strategy is introduced to dynamically adjust the enzymatic hydrolysis parameters by real-time monitoring of enzymatic hydrolysis parameters and combining the prediction results, so as to ensure the stability and efficiency of the enzymatic hydrolysis process and significantly improve the efficiency and quality of peptide preparation.

[0093] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or high-voltage switchgear that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or high-voltage switchgear.

[0094] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for producing functional polypeptides by enzymatic hydrolysis of soybean meal and peanut meal, characterized in that, Includes the following steps: S1: Grind the soybean meal and peanut meal separately and pass them through an 80-mesh sieve; The pulverized soybean meal and peanut meal were degreased separately using a gradient ethanol-supercritical fluid degreasing process. Combined extraction: first soak in 60% ethanol solution for 12 hours to remove free fats; Then, supercritical conditions were carried out at 30 MPa and 35℃. Extract for 5 hours to selectively remove bound fats and fat-soluble anti-nutritional factors, then dry after filtration; Mix defatted soybean meal powder and peanut meal powder at a mass ratio of 1:2, add an appropriate amount of water to make the moisture content of the mixture reach 55%; S2: Add the compound enzyme preparation to the mixture, adjust the pH of the mixture to 8.0, control the temperature at 50℃, and stir for 3 hours for enzymatic hydrolysis; during the enzymatic hydrolysis process, introduce ultrasonic-assisted enzymatic hydrolysis technology to enhance the contact between the enzyme and the substrate; S3: Heat the enzymatically hydrolyzed material to 92℃ and maintain for 12 minutes to inactivate the enzyme; centrifuge the enzyme-inactivated material using variable frequency centrifugation technology, automatically adjusting the centrifugation speed according to the density and viscosity of the material, with a rotation speed of 4500 rpm and a centrifugation time of 12 minutes; The supernatant was collected and filtered using a functional membrane element with a molecular weight of 5 kDa. Nanofiltration technology was introduced during the separation and purification process to effectively remove impurities through the nanofiltration membrane. S4: Spray dry the filtered polypeptide solution to obtain functional polypeptide powder; The final product undergoes quality testing, including the content, purity, antioxidant activity, and immunomodulatory activity indicators of the peptides.

2. The method for producing functional polypeptides by enzymatic hydrolysis of soybean meal and peanut meal according to claim 1, characterized in that, In step S1, soybean meal and peanut meal are separately pulverized and passed through an 80-mesh sieve, including the following steps: A two-stage pulverization mechanism is constructed. The first stage adopts low-temperature shear pulverization that mimics molars, simulating the human chewing mechanism, and pulverizes the raw materials to 40 mesh at <15℃. The second stage uses high-frequency vibration grinding with dental enamel to further refine the material to 80 mesh through a vibration field of a specific frequency; at the same time, it generates nanoscale cracks to increase the enzyme contact area.

3. The method for producing functional polypeptides by enzymatic hydrolysis of soybean meal and peanut meal according to claim 1, characterized in that, In S3, the supernatant is collected and filtered using a functional membrane element with a molecular weight of 5 kDa, including the following steps: A multi-stage filtration mechanism is adopted. First, a membrane element with a molecular weight of 10kDa is used for coarse filtration to remove large molecular impurities, and then a membrane element with a molecular weight of 5kDa is used for fine filtration. Bioactive factors, including coenzyme Q10, were added to the filtered peptide solution; and nanoemulsification technology was used to uniformly disperse coenzyme Q10 in the peptide solution, so that it could better combine with the peptide solution.

4. The method for producing functional polypeptides by enzymatic hydrolysis of soybean meal and peanut meal according to claim 1, characterized in that, The compound enzyme preparation is composed of alkaline protease, papain, and flavor protease in an enzyme activity ratio of 1:2:

1.

5. The method for producing functional polypeptides by enzymatic hydrolysis of soybean meal and peanut meal according to claim 1, characterized in that, It also includes the following steps: S5: Establish a quality traceability system based on RIFD radio frequency technology to record production process data and quality inspection results for each batch of products; S6: An adaptive control strategy is introduced, which uses online sensors to monitor the pH, temperature and enzyme activity of the enzymatic hydrolysate in real time and dynamically adjust the enzymatic hydrolysis parameters.

6. The method for producing functional polypeptides by enzymatic hydrolysis of soybean meal and peanut meal according to claim 5, characterized in that, In S6, an adaptive control strategy is introduced to dynamically adjust the enzymatic hydrolysis parameters, including the following steps: Each batch of soybean meal / peanut meal raw material will be rapidly tested to obtain basic data on different batches of soybean meal and peanut meal, including: protein content, fat content, ash content, and various parameters during the enzymatic hydrolysis process, including: temperature, pH value, enzyme dosage, and hydrolysis time; as well as various indicators after enzymatic hydrolysis, including: peptide yield, purity, antioxidant activity, and immunomodulatory activity. Near-infrared spectroscopy was introduced to rapidly detect the protein structure of soybean meal and peanut meal, and to predict the enzymatic hydrolysis efficiency of proteins and the potential activity of peptides. The collected data were analyzed based on a decision tree model, with basic data and enzymatic hydrolysis parameters as input variables and peptide yield, purity, and activity indicators as output variables. By training a decision tree model, the correlation coefficients between each input variable and the output variable are calculated to determine the key parameters and their impact on peptide preparation.

7. The method for producing functional polypeptides by enzymatic hydrolysis of soybean meal and peanut meal according to claim 5, characterized in that, The process of training a decision tree model involves calculating the correlation coefficients between each input variable and the output variable, including the following steps: Causal inference analysis is introduced, and multiple causal trees are constructed to comprehensively analyze the causal effect of each input variable and determine the causal influence of the input variable on the output variable. For each input variable, calculate its causal effect on each output variable; The impact of input variables on peptide preparation efficiency was evaluated by combining correlation coefficients and causal effects. Based on the correlation coefficient and the magnitude of the causal effect, the key parameters and their impact on the peptide preparation effect are determined; For each key parameter, assess its impact on different output variables; Based on the evaluation results, a list of feature importance is generated, showing the relative importance of each input variable to the output variable; and the importance values ​​are normalized to form correlation coefficients.

8. The method for producing functional polypeptides by enzymatic hydrolysis of soybean meal and peanut meal according to claim 7, characterized in that, After generating a list of feature importance based on the evaluation results, the following steps are included: Obtain the list of feature importance generated by the decision tree model to understand the degree of influence of each input variable on the output variable; The current state of the enzymatic hydrolysis process is assessed based on real-time data and predictions from the decision tree model. Based on the prediction results of the decision tree model, formulate parameter adjustment rules; Determine the adjustment range and step size for each parameter, and determine the frequency of parameter adjustment based on the dynamic characteristics of the enzymatic hydrolysis process and production requirements.

9. The method for producing functional polypeptides by enzymatic hydrolysis of soybean meal and peanut meal according to claim 8, characterized in that, After generating the feature importance list based on the evaluation results, the following steps are also included: The evaluation cycle is set according to the actual production process; During the evaluation period, various data from the enzymatic hydrolysis process are continuously collected and analyzed to assess whether the adaptive control strategy has achieved the expected goals; the actual production data is compared with the prediction results of the decision tree model to analyze the reasons for the deviation. Set evaluation indicators to measure the effectiveness of the adaptive control strategy; the evaluation indicators include: peptide yield increase rate, antioxidant activity achievement rate, parameter adjustment stability, and production cost change. Based on the evaluation results, update the decision tree model, re-evaluate the influence of each input variable on the output variable, and update the feature importance list. Based on the model update results and actual production data, optimize the parameter adjustment rules, including: redetermining the adjustment range and step size of each parameter based on the new model prediction results; Based on the dynamic characteristics of the enzymatic hydrolysis process and production requirements, the frequency of parameter adjustment is redefined; Adjust the adaptive control strategy based on the evaluation results.

10. The method for producing functional polypeptides by enzymatic hydrolysis of soybean meal and peanut meal according to claim 8, characterized in that, Based on the prediction results of the decision tree model, parameter adjustment rules are formulated, including the following steps: The real-time monitoring data is compared with the prediction results of the decision tree model to assess the current status of the enzymatic hydrolysis process; Based on the deviation between the prediction results of the decision tree model and real-time data, determine the direction and magnitude of parameter adjustments, including: Based on the prediction results of the decision tree model, the temperature of the enzymatic hydrolysate is maintained within the predicted range to optimize peptide yield. Based on the prediction results of the decision tree model, the pH value of the enzymatic hydrolysate was maintained within the predicted range to optimize the antioxidant activity of the peptides. The amount of enzyme is dynamically adjusted according to changes in enzyme activity to ensure the efficient execution of the enzymatic hydrolysis reaction. Based on changes in peptide yield and antioxidant activity, the enzymatic hydrolysis time is dynamically adjusted to optimize the quality of the final product.

Images