Intelligent optimization control system for aroma-enhancing compound leavening agent

Abstract

The invention relates to the technical field of fermentation control, in particular to an intelligent optimization control system of an aroma-enhancing compound leavening agent, which comprises a strain culture selection module, a fermentation process monitoring module, an environment optimization adjustment module and a fermentation effect evaluation module. According to the method, accurate analysis and optimization of microbial reaction in the fermentation process are realized, the selection efficiency of the strain culture combination is improved, the fermentation process can be carried out under the optimal condition, the efficiency of biological reaction is remarkably improved, potential risks and behavior abnormalities can be recognized in time through analysis of time sequence data, and the safety of the fermentation process is improved. The fermentation control process is optimized by dynamically adjusting key parameters, environmental parameters are adjusted through a nonlinear regression model and monitored in real time, the environmental stability is enhanced, it is ensured that the fermentation process is continuous and efficient, data of the whole process are comprehensively analyzed, the fermentation effect is accurately evaluated, multiple performance indexes are quantified, and the quality of a leavening agent is improved.

Description

Technical Field

[0001] The invention relates to the technical field of fermentation control, and in particular to an intelligent optimization control system for a flavor-enhancing composite leavening agent. Background Art

[0002] The field of fermentation control technology involves the use of automated and precise control systems to monitor and adjust various parameters in the biological fermentation process, such as temperature, pH value, oxygen supply and nutrient concentration. Its purpose is to ensure the stability and efficiency of the fermentation process, optimize the microbial growth environment, and improve the quality and output of products. In many industries such as food industry, pharmaceuticals, biofuels and environmental engineering, fermentation control technology is a key component. The control system usually integrates sensors, controllers and actuators, and can collect data in real time and automatically adjust operating variables to achieve the predetermined fermentation goals.

[0003] Among them, the intelligent optimization control system of the flavor-enhancing composite fermentation agent refers to an automated system specially designed to control and optimize the preparation process of the flavor-enhancing composite fermentation agent. The system uses control technology to adjust and control the environmental conditions during the fermentation process, such as temperature, humidity and nutrient supply, to ensure that the fermentation agent is produced under optimal conditions. Its main purpose is to improve the fermentation efficiency and flavor-enhancing effect, thereby improving the flavor and quality of the product. In the food industry, such as brewing, dairy products and fermented beverage production, the application of such systems is particularly important.

[0004] Existing technologies lack real-time adaptation and accurate control of dynamic changes in microbial reactions, resulting in the inability of traditional systems to effectively adapt to new problems under rapidly changing production conditions, such as diverse strain requirements and changing fermentation conditions. The lack of data analysis tools limits the ability to process complex data and timely predict potential risks, hinders the maximization of fermentation process efficiency, and leads to inconsistencies in product quality and output. Especially in the preparation of fermentation products that require high-precision control, inaccurate control can cause fluctuations in product taste and texture, which not only affects consumer experience, but also has a negative impact on production efficiency. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose an intelligent optimization control system for a flavor-enhancing composite leavening agent.

[0006] In order to achieve the above object, the present invention adopts the following technical scheme: an intelligent optimization control system for a flavor-enhancing composite leavening agent, the system comprising: The strain culture selection module uses support vector machines to analyze microbial activity and evaluate microbial responses based on temperature, pH value, and oxygen concentration data, and selects the optimal strain culture combination based on the evaluation results to obtain the strain combination optimization result; The fermentation process monitoring module analyzes the time series data based on the strain combination optimization results, identifies potential risks and behavioral anomalies in the fermentation process, and dynamically adjusts key fermentation parameters to optimize the fermentation control process and obtain fermentation parameter optimization information; The environmental optimization and adjustment module uses a nonlinear regression model to analyze and calculate based on the fermentation parameter optimization information and the strain combination optimization results, adjusts the gas flow and temperature, and adjusts the fermentation environment stability according to real-time data to obtain an environmental steady-state adjustment index; The fermentation effect evaluation module formulates evaluation indicators based on the environmental steady-state regulation indicators and combines the whole process data of fermentation to analyze the fermentation effect in real time, and quantitatively evaluates multiple performance indicators of the preparation process to generate fermentation control evaluation results.

[0007] The present invention is improved in that the step of analyzing the activity of microorganisms is specifically as follows: Based on the temperature, pH value and oxygen concentration data, the data were statistically analyzed using the formula: ; and ; Calculate the mean and variance and pass Get statistical characteristic data ,in, is a single data point, is the total number of data points, and Respectively represent the mean and volatility of the data; Based on the statistical characteristic data, a support vector machine is used, using the formula: ; Get activity prediction results ,in, represents the support vector machine model, It is the statistical characteristic data including mean and variance; Based on the activity prediction results and combined with the influence of environmental parameters on microbial activity, the formula is adopted: ; Get microbial activity score ,in, It corresponds to The activity prediction results of the parameters are Indicates The weight of each parameter is combined with the influence of environmental factors.

[0008] The present invention is improved in that the step of selecting the optimal bacterial culture combination is specifically: Based on the evaluation results, strains with scores above the threshold were selected for analysis using the formula: ; Get the list of strains after screening ,in, To score the activity of individual strains, is the activity score threshold; The screened strain list was subjected to a combination test to establish a differential culture combination, and the growth and activity performance of each combination was tested using the formula: ; Get combined effect data ,in, It is The proportion of strains in the combination, It is Activity index of strains, is the total number of data points; The combination effect data is evaluated, the strain combination with the best effect is selected, and the cultivation plan is determined by a decision algorithm, the formula is: ; Get the best strain combination ,in, It is the combined effect data.

[0009] The present invention is improved in that the steps of identifying potential risks and abnormal behaviors are specifically as follows: Based on the strain combination optimization results, key parameter data including temperature, pressure and pH value were collected, and the data were processed, including filtering and denoising, using the formula: ; Get the processed data set ,in, is the processing function, including filtering and denoising, is the original time series dataset; Based on the processed data set, an autoregressive moving average model is applied to identify abnormal patterns in the data, and the deviation calculation is performed by comparing the model prediction and the current data, using the formula: ; Get anomaly score ,in, are the observations in the processing dataset, is the value predicted by the ARMA model, is the total number of data points; Based on the anomaly score, risk points and behavioral anomalies are identified through the formula: ; Identify abnormal points that exceed the threshold, classify them as potential risks or behavioral anomalies, and obtain risk and abnormal points ,in, is the anomaly score, is the anomaly recognition threshold, yes A single anomaly score in .

[0010] The present invention is improved in that the optimization steps of the fermentation control process are specifically as follows: Based on the potential risks and behavioral anomalies, analyze the impact on current fermentation parameters and adjust key fermentation parameters using the formula: ; Get the adjusted parameters ,in, is the current fermentation control parameter, is the adjustment factor, which determines the sensitivity of the adjustment, Risks and outliers. is the hyperbolic tangent function; Based on the adjusted parameters, the fermentation control process is dynamically adjusted and the control settings are updated in real time, using the formula: ; Get control optimization feedback results ,in, Represents the update function of the fermentation process control flow, according to Make adjustments, is the adjusted parameter.

[0011] The present invention is improved in that the steps of adjusting the gas flow rate and temperature are specifically as follows: Based on the fermentation parameter optimization information and strain combination optimization results, a nonlinear regression model was used to analyze and calculate the target gas flow rate and temperature setting using the formula: ; and ; Get new traffic and temperature ,in, and is the current gas flow and temperature, and are data-driven parameters that affect flow and temperature adjustments, and is a model tuning parameter used to control tuning sensitivity; Based on the new flow and temperature, the fermentation control is updated using the formula: ; in, Is the update execution action, is the control function of the fermentation process, It’s new traffic. It's a new temperature. is the adjustment coefficient of feedback control; Monitor the effect of adjusted gas flow and temperature on the fermentation process and analyze stability and efficiency using the formula: ; Get the analysis results ,in, is an evaluation function used to analyze the effect of adjusted parameters. It’s new traffic. It's a new temperature. is an adjustment parameter during the analysis process that affects the sensitivity of the stability assessment. It is the associated data collected during the fermentation process.

[0012] The present invention is improved in that the step of obtaining the environmental steady-state regulation index is specifically as follows: Based on the real-time data, analyze the key steady-state factors and evaluate the difference between the current environmental state and the ideal state, using the formula: ; Environmental homeostasis factor ,in, is the temperature data, is the humidity data, is the gas concentration data, is the weight coefficient, which is used to adjust the balance between factors. is the weighted average function; Based on the environmental steady-state factors, calculate the required adjustments and estimate the key adjustments using the formula: ; Get the required adjustment ,in, is the environmental homeostatic factor, and are the adjustment coefficients and response coefficients in the adjustment model, is the response function; Based on the required adjustment amount, the environmental parameter adjustment is performed, and the adjusted environmental data is monitored to verify the adjustment effect. The formula is: ; Get feedback on the adjusted results ,in, is the required adjustment, is the execution accuracy control parameter, To execute the function.

[0013] The present invention is improved in that the step of obtaining the fermentation control evaluation result is specifically as follows: Based on the environmental steady-state regulation index, the whole process fermentation data, including environmental conditions and biological reaction data, are collected, and the formula is used: ; Get fermentation process data ,in, It's time Environmental data, It's time The biological response data, is the weight coefficient, representing time The criticality of data is the total number of time steps; Based on the fermentation process data, logarithmic transformation was performed using the formula: ; Obtain standardized fermentation process data ,in, is the fermentation process data, is the maximum value of the fermentation process data; Based on the standardized fermentation process data, combined with the evaluation criteria, the formula is: ; Get fermentation control assessment results ,in, It is the standardized fermentation process data. , and It is a tuning parameter used to adjust the sensitivity and bias of the evaluation results.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, accurate analysis and optimization of microbial reactions in the fermentation process are achieved, the selection efficiency of bacterial culture combinations is improved, the fermentation process can be carried out under optimal conditions, and the efficiency of biological reactions is significantly improved. The analysis of time series data enables potential risks and behavioral abnormalities to be identified in a timely manner, and then the fermentation control process is optimized by dynamically adjusting key parameters. Environmental parameters are adjusted and monitored in real time through a nonlinear regression model, which enhances environmental stability and ensures that the fermentation process is continuous and efficient. The data of the entire process are comprehensively analyzed, the fermentation effect is accurately evaluated, multiple performance indicators are quantified, and the quality of the starter is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a system flow chart of the present invention; Figure 2 A flow chart for analyzing microbial activity in the present invention; Figure 3 This is a flow chart for selecting the optimal bacterial culture combination in the present invention; Figure 4 A flowchart for identifying potential risks and behavioral anomalies in the present invention; Figure 5 is an optimized flow chart of the fermentation control process in the present invention; Figure 6 The figure is a flow chart of regulating gas flow and temperature in the present invention; Figure 7 This is a flow chart for obtaining the environmental steady-state regulation index in the present invention; Figure 8 The figure is a flow chart for obtaining the fermentation control evaluation results in the present invention. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0017] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined. Example

[0018] See also Figure 1 The present invention provides a technical solution: an intelligent optimization control system for a flavor-enhancing composite fermentation agent comprises: The strain culture selection module uses support vector machines to analyze microbial activity and evaluate microbial responses based on temperature, pH value, and oxygen concentration data, and selects the optimal strain culture combination based on the evaluation results to obtain the strain combination optimization result; The fermentation process monitoring module analyzes time series data based on the strain combination optimization results, identifies potential risks and behavioral anomalies in the fermentation process, and dynamically adjusts key fermentation parameters to optimize the fermentation control process and obtain fermentation parameter optimization information; The environmental optimization and adjustment module uses a nonlinear regression model to analyze and calculate based on the fermentation parameter optimization information and the strain combination optimization results, adjusts the gas flow and temperature, and adjusts the fermentation environment stability according to real-time data to obtain the environmental steady-state adjustment index; The fermentation effect evaluation module is based on the environmental steady-state regulation index and combines the whole process data of fermentation to formulate evaluation indicators, analyze the fermentation effect in real time, and quantitatively evaluate multiple performance indicators of the preparation process to generate fermentation control evaluation results.

[0019] The optimization results of strain combinations include metabolic rate, acid resistance, and high temperature resistance. The fermentation parameter optimization information includes temperature setting value, pH setting value, and oxygen flow rate. The environmental steady-state regulation indicators include environmental pressure, control accuracy, and response time. The fermentation control evaluation results include total fermentation time, total output, and by-product content.

[0020] See also Figure 2 , the specific steps for analyzing microbial activity are: Based on the temperature, pH value and oxygen concentration data, the data were statistically analyzed using the formula: ; and ; Calculate the mean and variance and pass Get statistical characteristic data ,in, is a single data point, is the total number of data points, and Respectively represent the mean and volatility of the data; Based on the statistical characteristic data, a support vector machine is used, using the formula: ; Get activity prediction results ,in, represents the support vector machine model, It is the statistical characteristic data including mean and variance; Based on the activity prediction results and combined with the influence of environmental parameters on microbial activity, the formula is adopted: ; Get microbial activity score ,in, It corresponds to The activity prediction results of the parameters are Indicates The weight of each parameter is combined with the influence of environmental factors.

[0021] Suppose the following three data points of oxygen concentration are collected:

[0022] Calculate the mean:

[0023] Calculate the variance:

[0024]

[0025]

[0026] The SVM algorithm requires training data to build a classification model. is the input feature, including the mean and variance. Assuming that the decision function of SVM is based on the linear kernel, its classification decision form is:

[0027] in, and are the model parameters obtained through training, assuming a model where .

[0028] Using the result of step 1:

[0029] Since the result is positive, , indicating that the data point is classified as positive.

[0030] Assume that there are three environmental parameters with weights: And the SVM prediction results of each parameter are :

[0031] result It means that the comprehensive score is positive, indicating that under the given parameter weights and prediction results, the microbial activity tends to respond positively.

[0032] See also Figure 3 , the specific steps for selecting the optimal bacterial culture combination are: Based on the evaluation results, strains with scores above the threshold were selected for analysis using the formula: ; Get the list of strains after screening ,in, To score the activity of individual strains, is the activity score threshold; The screened strain list was subjected to a combination test to establish a differential culture combination, and the growth and activity performance of each combination was tested using the formula: ; Get combined effect data ,in, It is The proportion of strains in the combination, It is Activity index of strains, is the total number of data points; The combination effect data is evaluated, the strain combination with the best effect is selected, and the cultivation plan is determined by a decision algorithm, the formula is: ; Get the best strain combination ,in, It is the combined effect data.

[0033] Assume that there is the following activity score data:

[0034] Setting Thresholds .

[0035] Check if each score is greater than 20; Filter out strains with scores greater than 20:

[0036] Filtered list All strains with activity scores exceeding the threshold are included, and the strains will enter the next step of combination test.

[0037] Assume that there are two strains after screening, and the ratio and activity index are and .

[0038] The calculation process is as follows: Calculate the contribution for each strain:

[0039] and

[0040] Calculate the overall combined effect:

[0041] It represents the comprehensive activity index of the strain combination at a specific ratio and is used to compare the effects of different combinations.

[0042] Assume that in multiple different combination tests, the combined effect data obtained are:

[0043] The calculation process is as follows: Select the maximum value:

[0044] This indicates that among all the tested combinations, the combination with an activity index of 25.5 is the best and will be determined as the culture scheme.

[0045] See also Figure 4 , the specific steps for identifying potential risks and behavioral anomalies are: Based on the results of strain combination optimization, key parameter data including temperature, pressure and pH value were collected and processed, including filtering and denoising, using the formula: ; Get the processed data set ,in, is the processing function, including filtering and denoising, is the original time series dataset; Based on the processed data set, an autoregressive moving average model is applied to identify abnormal patterns in the data, and the deviation calculation is performed by comparing the model prediction and the current data, using the formula: ; Get anomaly score ,in, are the observations in the processing dataset, is the value predicted by the ARMA model, is the total number of data points; Based on the anomaly score, risk points and behavioral anomalies are identified through the formula: ; Identify abnormal points that exceed the threshold, classify them as potential risks or behavioral anomalies, and obtain risk and abnormal points ,in, is the anomaly score, is the anomaly recognition threshold, yes A single anomaly score in .

[0046] Assume that the collected data points such as temperature and pH are:

[0047] and

[0048] Calculation process: For the sliding average filter (assuming a window size of 2), the temperature and pH values ​​are processed as follows:

[0049]

[0050] Processed dataset:

[0051]

[0052] This step smoothes the data, reduces the impact of noise, and facilitates subsequent analysis.

[0053] Assume that the model predicts:

[0054] and

[0055] Calculation process: For processed and model-predicted data:

[0056]

[0057] Total Anomaly Score:

[0058] Anomaly score It indicates the deviation between the model prediction and the actual observation value. A lower value indicates that the deviation is small and the model prediction is more accurate.

[0059] Assume it is 0.1.

[0060] Calculation process: because ,Compare and

[0061]

[0062] Exception Collection Indicates that below the threshold, potential risks or behavioral anomalies are identified, requiring further investigation or action to optimize the process.

[0063] See also Figure 5 , the optimization steps of the fermentation control process are as follows: Based on potential risks and behavioral anomalies, analyze the impact on current fermentation parameters and adjust key fermentation parameters using the formula: ; Get the adjusted parameters ,in, is the current fermentation control parameter, is the adjustment factor, which determines the sensitivity of the adjustment, Risks and outliers. is the hyperbolic tangent function; Based on the adjusted parameters, the fermentation control process is dynamically adjusted and the control settings are updated in real time, using the formula: ; Get control optimization feedback results ,in, Represents the update function of the fermentation process control flow, according to Make adjustments, is the adjusted parameter.

[0064] Assume the data is as follows: Assume that the temperature is 37°C for the temperature control parameter; Assume +3; Assuming it is 0.5, Calculate the risk adjustment factor:

[0065] Calculate the adjusted temperature:

[0066] The adjusted temperature is , this adjustment is based on the identified risks and aims to optimize the growth conditions of bacteria and increase fermentation efficiency.

[0067] for , the adjusted parameters Input to the control system, assuming that the control system updates the temperature setting in real time and provides feedback on the success of the operation; The system was updated successfully, and the new temperature setting is , indicating that the control system can respond to parameter adjustments in a timely manner and maintain the stability and efficiency of the fermentation process.

[0068] See also Figure 6 , the specific steps for adjusting gas flow and temperature are: Based on the fermentation parameter optimization information and strain combination optimization results, a nonlinear regression model was used to analyze and calculate the target gas flow rate and temperature setting using the formula: ; and ; Get new traffic and temperature ,in, and is the current gas flow and temperature, and is a data-driven parameter that affects flow and temperature adjustments, and is a model tuning parameter used to control tuning sensitivity; Based on the new flow and temperature, the fermentation control is updated using the formula: ; in, Is the update execution action, is the control function of the fermentation process, It’s new traffic. It's a new temperature. is the adjustment coefficient of feedback control; Monitor the effect of adjusted gas flow and temperature on the fermentation process and analyze stability and efficiency using the formula: ; Get the analysis results ,in, is an evaluation function used to analyze the effect of adjusted parameters. It’s new traffic. It's a new temperature. is an adjustment parameter during the analysis process that affects the sensitivity of the stability assessment. It is the associated data collected during the fermentation process.

[0069] Assume the data is as follows: L / min; ; L / min; ; ; , Calculate the new gas flow rate:

[0070] Calculate the new temperature:

[0071] The new gas flow rate is 100.5L / min and the new temperature is , indicating that subtle adjustments after model analysis are aimed at improving fermentation efficiency and strain growth environment.

[0072] Assume the following: L / min; ℃; .

[0073] Feedback control applications: The system adjusts the control logic to ensure the accurate implementation of the new parameters. Control, adjust the equipment according to the new parameters to ensure the fermentation process is carried out at 100.5L / min and 38℃, and the feedback control coefficient helps maintain the stability of this new setting.

[0074] Assumptions is 0.1; Assume that the product concentration is recorded every 10 minutes.

[0075] Data Collection: Assume that before and after adjusting the parameters, the concentration of the collected fermentation products is as follows: Before adjustment (unit: g / L):

[0076] After adjustment (unit: g / L):

[0077] Effect evaluation calculation, using linear regression model to evaluate the slope change of product concentration growth trend and the effect of parameter adjustment The slope calculation formula is: is the number of data points, and is the data mean:

[0078] Applied to data before and after adjustment:

[0079]

[0080] Compare whether the slope change exceeds Set thresholds to evaluate stability and efficiency improvements; if Significantly higher than And the difference exceeds , it indicates that the parameter adjustment effectively increases the product concentration growth rate, indicating that the stability and efficiency of the fermentation environment have been substantially improved.

[0081] See also Figure 7 , the specific steps for obtaining the environmental steady-state regulation index are: Based on real-time data, analyze key steady-state factors and evaluate the difference between the current environmental state and the ideal state, using the formula: ; Environmental homeostasis factor ,in, is the temperature data, is the humidity data, is the gas concentration data, is the weight coefficient, which is used to adjust the balance between factors. is the weighted average function; Based on the environmental steady-state factors, calculate the required adjustments and estimate the key adjustments using the formula: ; Get the required adjustment ,in, is the environmental homeostatic factor, and are the adjustment coefficients and response coefficients in the adjustment model, is the response function; Based on the required adjustment amount, the environmental parameter adjustment is performed, and the adjusted environmental data is monitored to verify the adjustment effect. The formula is: ; Get feedback on the adjusted results ,in, is the required adjustment, is the execution accuracy control parameter, To execute the function.

[0082] Assume the following data: =25℃; =60%; =0.05% (assuming CO2); =0.3.

[0083] Calculation process: Assuming steady-state factor Calculated by the weighted average formula:

[0084] Temperature and gas concentrations were normalized by simple scaling conversions to bring all parameters to the same magnitude; Insert specific values ​​to calculate:

[0085]

[0086]

[0087] Calculated steady-state factor Indicates the comprehensive deviation between the current environmental state and the ideal state. The higher the value, the greater the deviation.

[0088] Assume the following data: ; ; .

[0089] Calculation process: Assuming the required adjustment is a direct multiplication of the steady-state factor, taking into account the adjustment of sensitivity and response coefficient:

[0090] Insert specific values ​​to calculate:

[0091]

[0092]

[0093]

[0094] Calculated Adjustment =0.96255, indicating that in order to achieve the ideal environmental stability, the environmental control parameters need to be adjusted accordingly.

[0095] Assume the following data: is 0.96255; is 0.1.

[0096] Calculation process: Assuming that the actual adjustment is performed by directly applying the adjustment amount, consider the execution accuracy:

[0097] Insert specific values ​​to calculate:

[0098] Results of the adjustments implemented =0.866295 indicates the actual effect after adjustment. Taking into account the impact of execution accuracy, the calculated adjustment amount cannot be fully achieved, indicating the actual loss during the adjustment execution process.

[0099] See also Figure 8 , the specific steps for obtaining the fermentation control evaluation results are as follows: Based on the environmental steady-state regulation index, the whole process fermentation data, including environmental conditions and biological reaction data, are collected using the formula: ; Get fermentation process data ,in, It's time Environmental data, It's time The biological response data, is the weight coefficient, representing time The criticality of data is the total number of time steps; Based on the fermentation process data, logarithmic transformation was performed using the formula: ; Obtain standardized fermentation process data ,in, is the fermentation process data, is the maximum value of the fermentation process data; Based on the standardized fermentation process data, combined with the evaluation criteria, the formula is: ; Get fermentation control assessment results ,in, It is the standardized fermentation process data. , and It is a tuning parameter used to adjust the sensitivity and bias of the evaluation results.

[0100] Suppose there are data at three time points :

[0101]

[0102]

[0103] Calculation process:

[0104]

[0105] Represents the weighted comprehensive data of the fermentation process calculated according to the set weights.

[0106] Assumptions =100:

[0107]

[0108] Represents the normalized ratio of the result to the theoretical maximum value.

[0109] Assumptions is 0.777; ; ; .

[0110] Calculation process: calculate :

[0111] Calculate the denominator : 1+0.604=1.604 Compute the overall fraction:

[0112] application and Perform the calculation:

[0113] It indicates the fermentation control evaluation result, and the value reflects the effect of the current fermentation process control.

[0114] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. An intelligent optimization control system for a flavoring composite fermentation agent, characterized in that: The system comprises: The strain culture selection module uses support vector machines to analyze microbial activity and evaluate microbial responses based on temperature, pH value, and oxygen concentration data, and selects the optimal strain culture combination based on the evaluation results to obtain the strain combination optimization result; The fermentation process monitoring module analyzes the time series data based on the strain combination optimization result, identifies the potential risks and behavioral abnormalities of the fermentation process, and dynamically adjusts the key fermentation parameters to optimize the fermentation control process and obtain the fermentation parameter optimization information; The environmental optimization and adjustment module uses a nonlinear regression model to analyze and calculate based on the fermentation parameter optimization information and the strain combination optimization results, adjusts the gas flow and temperature, and adjusts the fermentation environment stability according to real-time data to obtain an environmental steady-state adjustment index; The fermentation effect evaluation module formulates evaluation indicators based on the environmental steady-state regulation indicators and combines the whole process data of fermentation to analyze the fermentation effect in real time, and quantitatively evaluates multiple performance indicators of the preparation process to generate fermentation control evaluation results.

2. The intelligent optimization control system for the flavoring composite leavening agent according to claim 1, characterized in that: The steps of analyzing the microbial activity are specifically as follows: Based on the temperature, pH value and oxygen concentration data, the data were statistically analyzed using the formula: ； and ； Calculate the mean and variance and pass Get statistical characteristic data ,in, is a single data point, is the total number of data points, and Respectively represent the mean and volatility of the data; Based on the statistical characteristic data, a support vector machine is used, using the formula: ； Get activity prediction results ,in, represents the support vector machine model, It is the statistical characteristic data including mean and variance; Based on the activity prediction results and combined with the influence of environmental parameters on microbial activity, the formula is adopted: ； Get microbial activity score ,in, It corresponds to The activity prediction results of the parameters are Indicates The weight of each parameter is combined with the influence of environmental factors.

3. The intelligent optimization control system for the flavoring composite leavening agent according to claim 1, characterized in that: The steps for selecting the optimal bacterial culture combination are specifically as follows: Based on the evaluation results, strains with scores above the threshold were selected for analysis using the formula: ； Get the list of strains after screening ,in, To score the activity of individual strains, is the activity score threshold; The screened strain list was subjected to a combination test to establish a differential culture combination, and the growth and activity performance of each combination was tested using the formula: ； Get combined effect data ,in, It is The proportion of strains in the combination, It is Activity index of strains, is the total number of data points; The combination effect data is evaluated, the strain combination with the best effect is selected, and the cultivation plan is determined by a decision algorithm, the formula is: ； Get the best strain combination ,in, It is the combined effect data.

4. The intelligent optimization control system for the flavoring composite leavening agent according to claim 1, characterized in that: The steps for identifying potential risks and behavioral anomalies are specifically as follows: Based on the strain combination optimization results, key parameter data including temperature, pressure and pH value were collected, and the data were processed, including filtering and denoising, using the formula: ； Get the processed data set ,in, is the processing function, including filtering and denoising, is the original time series dataset; Based on the processed data set, an autoregressive moving average model is applied to identify abnormal patterns in the data, and the deviation calculation is performed by comparing the model prediction and the current data, using the formula: ； Get anomaly score ,in, are the observations in the processing dataset, is the value predicted by the ARMA model, is the total number of data points; Based on the anomaly score, risk points and behavioral anomalies are identified through the formula: ； Identify abnormal points that exceed the threshold, classify them as potential risks or behavioral anomalies, and obtain risk and abnormal points ,in, is the anomaly score, is the anomaly recognition threshold, yes A single anomaly score in .

5. The intelligent optimization control system for the flavoring composite leavening agent according to claim 1, characterized in that: The optimization steps of the fermentation control process are specifically as follows: Based on the potential risks and behavioral anomalies, analyze the impact on current fermentation parameters and adjust key fermentation parameters using the formula: ； Get the adjusted parameters ,in, is the current fermentation control parameter, is the adjustment factor, which determines the sensitivity of the adjustment, Risks and outliers. is the hyperbolic tangent function; Based on the adjusted parameters, the fermentation control process is dynamically adjusted and the control settings are updated in real time, using the formula: ； Get control optimization feedback results ,in, Represents the update function of the fermentation process control flow, according to Make adjustments, is the adjusted parameter.

6. The intelligent optimization control system for the flavoring composite leavening agent according to claim 1, characterized in that: The steps for adjusting the gas flow rate and temperature are specifically as follows: Based on the fermentation parameter optimization information and strain combination optimization results, a nonlinear regression model was used to analyze and calculate the target gas flow rate and temperature setting using the formula: ； and ； Get new traffic and temperature ,in, and is the current gas flow and temperature, and are data-driven parameters that affect flow and temperature adjustments, and is a model tuning parameter used to control tuning sensitivity; Based on the new flow and temperature, the fermentation control is updated using the formula: ； in, Is the update execution action, is the control function of the fermentation process, It’s new traffic. It's a new temperature. is the adjustment coefficient of feedback control; Monitor the effect of adjusted gas flow and temperature on the fermentation process and analyze stability and efficiency using the formula: ； Get the analysis results ,in, is an evaluation function used to analyze the effect of adjusted parameters. It’s new traffic. It's a new temperature. is an adjustment parameter during the analysis process that affects the sensitivity of the stability assessment. It is the associated data collected during the fermentation process.

7. The intelligent optimization control system for the flavoring composite leavening agent according to claim 1, characterized in that: The steps for obtaining the environmental steady-state regulation index are specifically as follows: Based on the real-time data, analyze the key steady-state factors and evaluate the difference between the current environmental state and the ideal state, using the formula: ； Environmental homeostasis factor ,in, is the temperature data, is the humidity data, is the gas concentration data, is the weight coefficient, which is used to adjust the balance between factors. is the weighted average function; Based on the environmental steady-state factors, calculate the required adjustments and estimate the key adjustments using the formula: ； Get the required adjustment ,in, is the environmental homeostatic factor, and are the adjustment coefficients and response coefficients in the adjustment model, is the response function; Based on the required adjustment amount, the environmental parameter adjustment is performed, and the adjusted environmental data is monitored to verify the adjustment effect. The formula is: ； Get feedback on the adjusted results ,in, is the required adjustment, is the execution accuracy control parameter, To execute the function.

8. The intelligent optimization control system for the flavoring composite leavening agent according to claim 1, characterized in that: The steps for obtaining the fermentation control evaluation results are specifically as follows: Based on the environmental steady-state regulation index, the whole process fermentation data, including environmental conditions and biological reaction data, are collected, and the formula is used: ； Get fermentation process data ,in, It's time Environmental data, It's time The biological response data, is the weight coefficient, representing time The criticality of data is the total number of time steps; Based on the fermentation process data, logarithmic transformation was performed using the formula: ； Obtain standardized fermentation process data ,in, is the fermentation process data, is the maximum value of the fermentation process data; Based on the standardized fermentation process data, combined with the evaluation criteria, the formula is: ； Get fermentation control assessment results ,in, It is the standardized fermentation process data. , and It is a tuning parameter used to adjust the sensitivity and bias of the evaluation results.

Images