CN-121983158-A - Intelligent control optimization method, system and storage medium for collaborative fermentation of Polygonatum sibiricum enzyme

CN121983158ACN 121983158 ACN121983158 ACN 121983158ACN-121983158-A

Abstract

The application relates to the field of microbial fermentation engineering, and discloses an intelligent control optimization method, system and storage medium for collaborative fermentation of sealwort bacterial enzymes. The method comprises the steps of collecting real-time data of fermentation parameters, analyzing flora imbalance and enzyme system imbalance indexes to generate system state description, preprocessing and extracting feature vectors, analyzing correlations to determine core influence factors, calculating parameter coupling interaction strength through a support vector machine, simulating a multi-parameter coordination scene to generate potential adjustment schemes and enzyme system collaborative prediction data when the strength exceeds a standard, screening candidate schemes, evaluating correction effects of the candidate schemes on the flora imbalance by using random forests, determining optimal parameter combinations, adjusting parameters according to the optimal combinations, calculating real-time feedback deviation, and iteratively optimizing pH values and stirring rates if the deviation exceeds the limit to obtain a final scheme. The application solves the problem that the multi-parameter coupling interaction in the collaborative fermentation of the Polygonatum sibiricum enzyme is difficult to quantitatively identify and coordinate and regulate, and improves the stability of a fermentation system and the quality of products.

Inventors

WANG XUEFANG
WANG XUEBING
LI XIAO
SUN HEYANG
LI XINYUE
SHI KAI
WANG XUELING
ZHANG HUAQIANG
SU SHUCHENG
HAO ZHONGHUA

Assignees

河南农业大学

Dates

Publication Date: 20260505
Application Date: 20260123

Claims (10)

1. An intelligent control optimization method for collaborative fermentation of Polygonatum sibiricum enzymes is characterized by comprising the following steps: s101, collecting real-time data of temperature, dissolved oxygen, pH value and stirring rate in a sealwort bacterial enzyme collaborative fermentation system, and analyzing a bacterial flora imbalance index and an enzyme system imbalance index according to the real-time data to generate a system overall state description; Step S102, preprocessing the system overall state description, extracting feature vectors, analyzing the correlation between flora imbalance and enzyme system imbalance, and determining core influence factors; Step S103, inputting the core influence factors into a pre-trained support vector machine model, and calculating parameter coupling interaction strength values; Step S104, if the parameter coupling interaction intensity value exceeds a preset intensity threshold value, simulating a multi-parameter coordination change scene to generate a potential adjustment scheme, analyzing the influence of the potential adjustment scheme on enzyme system coordination and generating corresponding enzyme system coordination maintenance prediction data, and screening to obtain candidate schemes; step 105, inputting the enzyme system collaborative maintenance prediction data into a pre-trained random forest model, evaluating the influence degree of the candidate scheme on the flora imbalance correction, and determining an optimal parameter combination; step S106, adjusting the operation parameters of the fermentation system according to the optimal parameter combination, collecting real-time feedback data, and calculating a deviation value between enzyme system cooperative maintenance prediction data corresponding to the optimal parameter combination; And step S107, if the deviation value exceeds a preset deviation range, updating the pH value and the stirring rate through an iterative optimization algorithm to obtain a final coordination change scheme.
2. The method according to claim 1, wherein the step S101 includes: collecting temperature, dissolved oxygen, pH value and stirring rate data in a Polygonatum sibiricum enzyme synergistic fermentation system in real time, and integrating to form a dynamic data set; Calculating a temperature change rate and a dissolved oxygen fluctuation range based on the dynamic data set, and taking the ratio of the temperature change rate to the dissolved oxygen fluctuation range as a flora imbalance index; Calculating a pH value deviation and a stirring rate adjustment frequency based on the dynamic data set, and taking the product of the pH value deviation and the stirring rate adjustment frequency as an enzyme system imbalance index; And generating a system overall state description representing the running condition of the fermentation system based on the flora imbalance index and the enzyme system imbalance index.
3. The method according to claim 2, wherein the step S102 includes: Preprocessing data in the system overall state description to obtain standardized state data; Based on the standardized state data, extracting a flora imbalance characteristic vector, an enzyme system imbalance characteristic vector and an environmental parameter characteristic vector by adopting a principal component analysis method, and constructing a characteristic data set; aiming at the characteristic data set, calculating the correlation coefficient of the flora imbalance characteristic vector and the enzyme system imbalance characteristic vector, and screening out a high coefficient index with the absolute value of the correlation coefficient larger than a preset correlation threshold; And constructing a decision tree model based on the high coefficient index, pruning the decision tree model, and determining core influence factors.
4. The method according to claim 1, wherein the step S103 includes: converting the core influence factors into characteristic factor vectors, and inputting a pre-trained support vector machine model; Classifying dominant bacteria inhibition requirements and weak bacteria activation potentials of the Polygonatum sibiricum fermentation system through the support vector machine model, and outputting classification probability vectors; based on the classification probability vector, calculating parameter interaction coefficients among the temperature, dissolved oxygen, pH value and stirring rate corresponding to each core influence factor; And according to the weight duty ratio of the classification probability vector and the parameter interaction coefficient, obtaining a parameter coupling interaction strength value through weighted summation.
5. The method according to claim 1, wherein the step S104 includes: If the parameter coupling interaction intensity value exceeds a preset intensity threshold, taking the temperature, the dissolved oxygen amount, the pH value and the stirring speed in the fermentation system as regulating and controlling objects, simulating a plurality of groups of multi-parameter coordination change scenes, and generating a plurality of potential regulating schemes; Analyzing the influence of each potential adjustment scheme on enzyme system coordination in a Polygonatum sibiricum enzyme coordination fermentation system, and quantifying the enzyme activity balance degree and enzyme system imbalance correction potential; Generating enzyme system cooperative maintenance prediction data corresponding to each potential adjustment scheme according to the influence analysis result; and screening potential adjustment schemes meeting preset enzyme system cooperation standards in the enzyme system cooperation maintenance prediction data to serve as candidate schemes.
6. The method according to claim 1, wherein the step S105 includes: The enzyme system cooperative maintenance prediction data corresponding to the candidate scheme is associated and integrated with the flora imbalance index to form a model input data set; inputting the model input data set into a pre-trained random forest model, and outputting flora imbalance correction influence coefficients corresponding to each candidate scheme; based on the flora unbalance correction influence coefficient, combining enzyme system cooperative maintenance prediction data corresponding to the candidate schemes, and quantifying the comprehensive optimization value of each candidate scheme; And screening out parameter combinations of temperature, dissolved oxygen, pH value and stirring rate corresponding to the candidate scheme with the highest comprehensive optimization value, and determining the parameter combinations as optimal parameter combinations.
7. The method according to claim 1, wherein the step S106 includes: Based on the optimal parameter combination, adjusting the temperature, dissolved oxygen, pH value and stirring rate in the Polygonatum sibiricum enzyme synergistic fermentation system; Continuously collecting real-time feedback data of temperature, dissolved oxygen, pH value and stirring rate in the fermentation system after parameter adjustment, and integrating the real-time feedback data according to time sequence to form a feedback data set; matching the feedback data set with enzyme system cooperative maintenance prediction data corresponding to the optimal parameter combination according to corresponding parameter dimensions; and calculating a deviation value between the real-time feedback data and enzyme system cooperative maintenance prediction data corresponding to the optimal parameter combination based on the matching result.
8. The method according to claim 1, wherein the step S107 includes: if the deviation value exceeds a preset deviation range, analyzing the association reason of the flora imbalance and the enzyme system imbalance based on the deviation value, and determining the fine adjustment direction and interval of the pH value and the stirring rate; Adopting a particle swarm optimization algorithm, taking the coordination of flora balance and an enzyme system as an optimization target, and updating the pH value and the stirring rate according to the determined fine adjustment direction and interval; Collecting temperature, dissolved oxygen, pH value and stirring rate in a Polygonatum sibiricum enzyme collaborative fermentation system after parameter updating, and calculating a new deviation value; Judging whether the new deviation value is in a preset deviation range, if not, repeating the steps of updating and data acquisition and calculation until the new deviation value accords with the preset deviation range; recording final pH value and stirring speed parameters, and combining the temperature and dissolved oxygen parameters in the optimal parameter combination to form a final coordination change scheme.
9. An intelligent control optimizing system for collaborative fermentation of a sealwort bacterial enzyme, which is used for realizing the intelligent control optimizing method for collaborative fermentation of a sealwort bacterial enzyme according to any one of claims 1 to 8, and is characterized in that the intelligent control optimizing system for collaborative fermentation of a sealwort bacterial enzyme comprises: The data acquisition and analysis module acquires real-time data of temperature, dissolved oxygen, pH value and stirring rate in the Polygonatum sibiricum enzyme collaborative fermentation system, and analyzes the flora imbalance index and the enzyme system imbalance index according to the real-time data to generate system overall state description; the influence factor determining module is used for preprocessing the system overall state description, extracting characteristic vectors, analyzing the correlation between flora imbalance and enzyme system imbalance and determining core influence factors; the coupling interaction calculation module is used for inputting the core influence factors into a pre-trained support vector machine model and calculating parameter coupling interaction strength values; The candidate scheme screening module is used for simulating a multi-parameter coordination change scene to generate a potential adjustment scheme when the parameter coupling interaction intensity value exceeds a preset intensity threshold value, analyzing the influence of the potential adjustment scheme on enzyme system coordination and generating corresponding enzyme system coordination maintenance prediction data, and screening to obtain a candidate scheme; The optimal parameter determining module is used for inputting the enzyme system collaborative maintenance prediction data into a pre-trained random forest model, evaluating the influence degree of the candidate scheme on the flora imbalance correction and determining an optimal parameter combination; The feedback deviation calculation module is used for adjusting the operation parameters of the fermentation system according to the optimal parameter combination, collecting real-time feedback data and calculating a deviation value between enzyme system cooperative maintenance prediction data corresponding to the optimal parameter combination; And the parameter optimization adjustment module is used for updating the pH value and the stirring rate through an iterative optimization algorithm when the deviation value exceeds a preset deviation range, so as to obtain a final coordination change scheme.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being run by a processor, causes the processor to perform the intelligent control optimization method of co-fermentation of a sealwort bacterial enzyme according to any one of claims 1 to 8.

Description

Intelligent control optimization method, system and storage medium for collaborative fermentation of Polygonatum sibiricum enzyme Technical Field The application relates to the field of microbial fermentation engineering, in particular to an intelligent control optimization method, system and storage medium for collaborative fermentation of Polygonatum sibiricum enzymes. Background Rhizoma Polygonati is used as an important medicinal and edible raw material, and the active ingredients are usually transformed and synergized through the synergized fermentation of microorganisms and enzymes to form a rhizoma polygonati bacterial enzyme synergistic fermentation system. The process involves dynamic regulation and control of a plurality of key parameters such as temperature, dissolved oxygen, pH, stirring rate and the like, and complex coupling interaction relation exists among the parameters, so that the balance of a flora structure in a system and the synergy of enzyme activity are directly influenced. The precise and intelligent control of the system is realized, and the method is a core technical challenge for improving the quality, yield and process stability of the rhizoma polygonati fermentation product. The method for regulating and controlling the fermentation process in the prior art has obvious limitation, and is difficult to effectively solve the complex dynamic problem of multi-parameter strong coupling in the collaborative fermentation of the Polygonatum sibiricum enzymes. First, common methods rely on empirical rules based on a single variable or fixed threshold, such as cooling when the temperature exceeds a preset upper limit, or increasing the stirring rate when dissolved oxygen is below a lower limit. The method ignores that temperature change can affect dissolved oxygen solubility, microorganism metabolic rate and pH value drift at the same time, and the adjustment of stirring rate can adversely affect shearing force and dissolved oxygen distribution, and chain reaction among parameters can easily cause 'failure to account for' to cause flora imbalance or enzyme system imbalance. Second, some approaches attempt to introduce simple feedback control, such as a PID controller to adjust pH, but their design is typically only directed to a single controlled variable, and cannot handle the nonlinear interactions and synergy requirements between multiple variables such as temperature, dissolved oxygen, pH, agitation rate, etc. When a plurality of parameters deviate from an ideal interval at the same time, the independent control strategy can not identify and quantify the coupling interaction strength between the parameters, so that the adjustment actions are uncoordinated, even mutually conflict, and the system instability is aggravated. Furthermore, although there are methods for predicting by using a data model, such as predicting a certain parameter trend by using a single regression model, there is a lack of joint analysis and feature extraction of two core biological status indexes, namely "bacterial flora imbalance" and "enzyme system imbalance", and it is not possible to identify core influencing factors from the overall status of the system. Therefore, the defects that the identification of the multiparameter coupling relation is delayed, the adjustment scheme lacks cooperativity, the coordination scheme cannot be generated in advance when the interaction strength exceeds the standard and the like commonly exist in the prior art, so that the stability of the collaborative fermentation process of the polygonatum fungi enzyme is poor and the efficiency is low. Aiming at the defects, the application solves the difficult problems that the multi-parameter coupling interaction strength is difficult to quantitatively identify and coordinate and regulate in the collaborative fermentation of the Polygonatum sibiricum enzymes by combining the real-time multi-parameter data acquisition, the support vector machine with the random forest model intelligent analysis and the feedback-based iterative optimization, realizes the advanced prediction and collaborative correction of the flora imbalance and the enzyme system imbalance, and improves the stability, the product quality and the overall control intelligent level of a fermentation system. Disclosure of Invention The application provides an intelligent control optimization method, system and storage medium for collaborative fermentation of Polygonatum sibiricum enzymes, which solve the problem that the multi-parameter coupling interaction strength is difficult to quantitatively identify and coordinate and regulate in collaborative fermentation of Polygonatum sibiricum enzymes, realize the advanced prediction and collaborative correction of flora imbalance and enzyme system imbalance, and improve the stability, product quality and overall control intelligent level of a fermentation system. In a first aspect, the application provides an intelligent control