Intelligent regulation and control method for tea fermentation process and regulation and control system based on microbial activity monitoring

Abstract

The invention discloses an intelligent regulation and control method for a tea fermentation process and a regulation and control system based on microbial activity monitoring, and relates to the technical field of tea processing. According to the method, microbial activity and environmental parameters are monitored in real time, and a microbial metabolism model and a neural network algorithm are combined, so that dynamic optimization and accurate regulation and control of the tea leaf fermentation process are realized, firstly, microbial activity data and environmental parameters in a fermentation environment are collected by utilizing a sensor array; the method comprises the following steps: firstly, calculating a metabolite generation rate and a concentration change trend through a microbial metabolism model, accurately identifying abnormal fluctuation in a fermentation process, secondly, adjusting fermentation conditions in real time based on the metabolite concentration change trend and a dynamic regulation and control algorithm to ensure the stability of the fermentation process, and finally, carrying out real-time fermentation. Tea quality parameters after fermentation condition adjustment are predicted through a neural network algorithm, a regulation and control strategy is optimized, and stable improvement of tea quality is achieved.

Description

Technical Field

[0001] This invention relates to the field of tea processing technology, specifically to an intelligent control method for the tea fermentation process and a control system based on microbial activity monitoring. Background Technology

[0002] Tea fermentation is a crucial process determining tea quality, involving complex interactions between microbial metabolism and environmental factors. It directly impacts the flavor, aroma, and nutritional value of tea. With increasing consumer demand for high-quality tea, precisely controlling the fermentation process to improve quality has become a vital industry challenge. Traditional fermentation methods often rely on manual experience or simple timed control, making it difficult to adapt to the dynamic changes in microbial activity and environmental conditions during fermentation. This often leads to uneven fermentation, unstable quality, and even resource waste. For example, fixed temperature and humidity control cannot address fluctuations in microbial metabolic rates, affecting the optimization of fermentation results. The core challenge lies in how to monitor the dynamic changes in microbial activity in real time and precisely correlate them with environmental parameters such as temperature, humidity, and oxygen content to achieve dynamic optimization of fermentation conditions. Existing technologies are insufficient in terms of real-time monitoring of microbial activity and the intelligence of control strategies, making it difficult to ensure the timeliness and accuracy of parameter adjustments in complex and ever-changing environments. In addition, the lack of intelligent algorithms that can efficiently integrate microbial metabolic data with environmental control strategies makes it difficult for the fermentation process to reach an ideal state. Therefore, how to develop an intelligent control method by real-time monitoring of microbial activity and its metabolites, combined with changes in environmental parameters, to dynamically optimize fermentation conditions and ensure the stable improvement of tea quality has become a key issue that urgently needs to be addressed. Summary of the Invention

[0003] The purpose of this invention is to provide an intelligent control method for the tea fermentation process and a control system based on microbial activity monitoring.

[0004] The objective of this invention can be achieved through the following technical solutions:

[0005] This application provides a method for intelligent control of the tea fermentation process, comprising the following steps:

[0006] Collect microbial activity data and environmental parameter data in the fermentation environment to obtain real-time values ​​of microbial activity and environmental parameters;

[0007] Based on the collected data, calculate the generation rate and concentration trend of metabolites, and output the concentration trend of metabolites.

[0008] When the trend of metabolite concentration changes exceeds a preset threshold, environmental parameter optimization suggestions are generated.

[0009] Based on the optimization suggestions for the generation environment parameters, a dynamic control algorithm is used to adjust the fermentation conditions and output temperature control values, humidity adjustment values ​​and oxygen content set values.

[0010] Collect data on the adjusted fermentation conditions, determine whether the trend of metabolite concentration changes is stabilizing, and then use a neural network algorithm to predict the tea quality parameters after the fermentation conditions are adjusted, and output the tea quality score.

[0011] Based on the predicted results of tea quality parameters, the control strategy is optimized and applied to the next fermentation condition adjustment cycle.

[0012] Furthermore, real-time values ​​of microbial activity and environmental parameters are obtained, specifically including:

[0013] Microbial activity data and environmental parameter data in the fermentation environment are collected by a sensor array to obtain real-time values. The real-time values ​​are then divided into time series according to the collection frequency to obtain segmented data.

[0014] The dynamic correlation between microbial activity and environmental parameters was calculated using segmented data. Based on this dynamic correlation, when the temperature sensor data exceeded a preset threshold, the humidity sensor and oxygen concentration data were weighted and adjusted to obtain corrected values.

[0015] The correlation coefficient between the corrected values ​​and microbial activity is obtained, the influence weight of the parameters is determined, and then the support vector machine algorithm is used to predict microbial activity to obtain the prediction results.

[0016] By comparing the predicted results with real-time values, abnormal fluctuations in the fermentation environment can be identified, providing a basis for adjustments.

[0017] Furthermore, the trend of changes in metabolite concentrations is output, specifically including:

[0018] By acquiring data on microbial activity, temperature, humidity, and oxygen content through sensors, a complete dataset of environmental parameters is obtained. Then, a pre-established microbial metabolic model is used to process the complete dataset of environmental parameters and output the rate of metabolite generation.

[0019] The concentration change value is calculated based on the rate of metabolite formation to obtain the preliminary result of the concentration change of metabolites. When the concentration change value exceeds the preset threshold, the change trend is classified by the support vector machine algorithm to determine the abnormal fluctuation range.

[0020] Temperature, humidity, and oxygen content data within the abnormal fluctuation range are obtained to identify key environmental factors affecting microbial activity. The parameters of the microbial metabolic model are adjusted based on these key environmental factors to obtain the optimized trend of metabolite concentration changes.

[0021] Furthermore, environmental parameter optimization suggestions are generated, specifically including:

[0022] The concentration data of metabolites is acquired in real time by sensors, the concentration change trend is calculated, and the change trend value is obtained. When the change trend exceeds the preset threshold, the current environmental parameters are extracted from the database to obtain the environmental parameter set.

[0023] By comparing the set of environmental parameters with the historical best parameter dataset, the direction of parameter deviation is determined, the deviation value is determined, and the direction of temperature control adjustment is calculated based on the deviation value and preset rules to obtain the temperature optimization parameters.

[0024] A temperature optimization parameter and humidity regulation correlation model is used to calculate the humidity regulation range and obtain the humidity optimization parameter. Then, the direction of oxygen content adjustment is determined by the correlation rule between the humidity optimization parameter and oxygen content to obtain the oxygen optimization parameter.

[0025] Based on the optimized parameters for temperature, humidity, and oxygen, suggestions for optimizing environmental parameters are generated.

[0026] Furthermore, the output temperature control value, humidity adjustment value, and oxygen content setpoint specifically include:

[0027] By collecting real-time values ​​of environmental parameters through sensors, the changing trend of metabolite concentration is determined. Based on the changing trend, a dynamic control algorithm is used to calculate the temperature control value and output the adjustment command.

[0028] The deviation between the real-time value and the preset threshold is obtained to determine the humidity adjustment value. When the deviation exceeds the preset range, the fermentation conditions are adjusted according to the oxygen content setting value.

[0029] Based on the adjusted fermentation conditions, the concentration changes of metabolites are analyzed to obtain updated trend data. The output value of the dynamic control algorithm is then optimized. The stability of environmental parameters is judged by the optimized output value, and the final control parameters are generated.

[0030] Furthermore, after outputting the temperature control value, humidity adjustment value, and oxygen content setpoint, it also includes:

[0031] The algorithm generation module is activated by a preset dynamic control strategy to output the initial set value. Real-time status data is obtained by acquiring the current operating data of the heating device, humidification device and ventilation device from the sensors of the fermentation equipment.

[0032] The random forest algorithm is used to compare real-time status data with set values ​​to determine the deviation range. When the deviation exceeds the preset threshold, the set value is updated through dynamic adjustment logic to obtain the optimized set value.

[0033] By optimizing the set values ​​to drive the operation of the heating, humidifying, and ventilation devices inside the fermentation equipment, the adjusted status data is obtained.

[0034] The fermentation condition change trend is calculated based on the adjusted state data to obtain condition optimization data. By comparing the condition optimization data with the target fermentation conditions, it is determined whether iterative adjustment of the set value is necessary.

[0035] Furthermore, determining whether the trend of metabolite concentration changes tends to stabilize includes:

[0036] Data preprocessing techniques were used to clean and standardize the adjusted fermentation condition data to obtain a processed dataset. Microbial activity data and metabolite concentration data were extracted from the processed dataset to determine time series characteristics.

[0037] By analyzing time series data, the trend of metabolite concentration is determined and a trend curve is obtained. When the absolute value of the slope of the trend curve is less than a preset threshold, the trend is determined to be stable, and a stability judgment result is obtained.

[0038] Based on the stability assessment results and combined with microbial activity data, the support vector machine algorithm was used to model the relationship between fermentation conditions and metabolite concentration to obtain an optimized model.

[0039] By optimizing the model, we can analyze the impact of the adjusted fermentation conditions on the stability trend of metabolite concentration data and determine the degree of condition optimization.

[0040] Furthermore, a neural network algorithm is used to predict tea quality parameters after adjusting fermentation conditions, specifically including:

[0041] The data integration module fuses the collected microbial activity data, metabolite concentration data, and environmental parameter data to generate a multidimensional feature dataset. The dataset is then processed using multidimensional feature extraction technology to obtain a structured feature set.

[0042] A neural network algorithm is used to train a structured feature set to determine a quality score prediction model. Initial fermentation conditions are input, and condition change data is generated after adjustment. When the condition change data exceeds a preset threshold, the adjusted quality score value is predicted by the neural network.

[0043] Based on the predicted quality score, the trend of tea quality change is determined. Through trend analysis, optimized fermentation condition parameters are obtained. The dataset is updated using the optimized fermentation condition parameters to determine the final quality score.

[0044] Furthermore, optimize the regulatory strategy, specifically including:

[0045] The quality score of tea is obtained by using a neural network algorithm. The initial quality data is determined by the score. Key feature values ​​are then extracted from the initial quality data. The parameter weights are adjusted by a dynamic control algorithm to obtain the weight update result.

[0046] An optimized control strategy is generated based on the weight update results. The trend of fermentation conditions is judged by the optimized control strategy. When the trend exceeds the preset threshold, the fermentation conditions are adjusted through the condition update module to obtain the updated fermentation parameters.

[0047] A new round of quality score values ​​are obtained based on the updated fermentation parameters. The adaptability of the control strategy is judged by the score values. At the same time, the parameter weights of the dynamic control algorithm are adjusted according to the adaptability results to obtain a new round of optimized control strategy.

[0048] By updating the fermentation conditions and adjusting the cycle with a new round of optimized control strategies, we can obtain cycle control data.

[0049] This embodiment provides a tea fermentation regulation system based on microbial activity monitoring, used to realize an intelligent regulation method for the tea fermentation process, including:

[0050] The sensor array module is used to collect microbial activity data and environmental parameter data in the fermentation environment. The sensors include a temperature sensor, a humidity sensor, and an oxygen concentration sensor, and the acquisition frequency is set to once per minute.

[0051] The data processing module is used to calculate the rate of metabolite production based on the collected microbial activity data and environmental parameter data using a pre-established microbial metabolic model. The model inputs include microbial activity, temperature, humidity and oxygen content, and the output is the trend of metabolite concentration changes.

[0052] The optimization judgment module is used to determine the direction of temperature control, humidity regulation and oxygen content adjustment when the trend of metabolite concentration changes exceeds the preset threshold by comparing the current environmental parameters with the historical best parameter dataset, and to obtain environmental parameter optimization suggestions.

[0053] The dynamic control module is used to adjust the fermentation conditions according to the optimization suggestions of environmental parameters using a dynamic control algorithm. The algorithm outputs temperature control values, humidity adjustment values ​​and oxygen content set values ​​based on the trend of metabolite concentration changes and real-time values ​​of environmental parameters.

[0054] The execution module is used to adjust the heating device, humidification device and ventilation device in the fermentation equipment according to the set value output by the dynamic control module, so as to obtain the adjusted fermentation condition data.

[0055] The feedback module is used to collect microbial activity data and metabolite concentration data again after obtaining the adjusted fermentation condition data, and to determine whether the adjusted fermentation conditions have made the trend of metabolite concentration change more stable.

[0056] The data integration module is used to fuse collected microbial activity data, metabolite concentration data, and environmental parameter data to generate a multidimensional feature dataset when the trend of metabolite concentration changes is unstable.

[0057] The quality prediction module is used to predict the tea quality parameters after the fermentation conditions are adjusted based on the generated multidimensional feature dataset using a neural network algorithm. The neural network input is the multidimensional feature dataset and the output is the tea quality score.

[0058] The optimization and adjustment module is used to adjust the parameter weights of the dynamic control algorithm based on the tea quality score output by the neural network algorithm, so as to obtain the optimized control strategy, which is then applied to the next fermentation condition adjustment cycle.

[0059] The beneficial effects of this invention are as follows:

[0060] This invention achieves dynamic optimization and precise control of the tea fermentation process by real-time monitoring of microbial activity and environmental parameters, combined with a microbial metabolic model and neural network algorithm. It utilizes a sensor array to collect microbial activity data and environmental parameters in the fermentation environment, and calculates the generation rate and concentration trends of metabolites using a microbial metabolic model to accurately identify abnormal fluctuations during fermentation. Secondly, based on the trend of metabolite concentration changes, a dynamic control algorithm adjusts fermentation conditions in real time to ensure the stability of the fermentation process. Finally, a neural network algorithm predicts the tea quality parameters after the fermentation conditions are adjusted, optimizing the control strategy and achieving a stable improvement in tea quality. This effectively solves the problems of uneven fermentation and unstable quality caused by fluctuations in environmental parameters in traditional fermentation methods, significantly improving the controllability and efficiency of tea production. Attached Figure Description

[0061] To better understand and implement this application, the technical solution is described in detail below with reference to the accompanying drawings.

[0062] Figure 1 This is a flowchart illustrating the intelligent control method for tea fermentation provided in Embodiment 1 of this application.

[0063] Figure 2 A schematic diagram illustrating the process of obtaining real-time values ​​of microbial activity and environmental parameters using the intelligent control method for tea fermentation provided in Embodiment 1 of this application;

[0064] Figure 3 This is a schematic diagram illustrating the output of the trend of metabolite concentration changes in the intelligent control method for tea fermentation provided in Embodiment 1 of this application.

[0065] Figure 4 This is a schematic diagram of a tea fermentation regulation system based on microbial activity monitoring, provided in Embodiment 2 of this application. Detailed Implementation

[0066] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, exemplary embodiments will be described in detail below, examples of which are illustrated in the accompanying drawings. In the following description relating to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of methods and systems consistent with some aspects of this application as detailed in the appended claims.

[0067] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0068] The following detailed description of the specific implementation methods, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided in detail.

[0069] Example 1

[0070] Please see Figures 1-3 This embodiment provides an intelligent control method for the tea fermentation process, including the following steps:

[0071] S1. Collect microbial activity data and environmental parameter data in the fermentation environment through a sensor array. The sensors include a temperature sensor, a humidity sensor, and an oxygen concentration sensor. The acquisition frequency is set to once per minute to obtain real-time values ​​of microbial activity and environmental parameters.

[0072] Furthermore, in step S1, real-time values ​​of microbial activity and environmental parameters are obtained, specifically including:

[0073] S11. Collect microbial activity data and environmental parameter data in the fermentation environment through a sensor array to obtain real-time values. Divide the real-time values ​​into time series according to the acquisition frequency to obtain segmented data.

[0074] S12. Calculate the changing trends of microbial activity and environmental parameters using segmented data to obtain a dynamic correlation. Based on this dynamic correlation, if the temperature sensor data exceeds a preset threshold, then weighted adjustments are made to the humidity sensor and oxygen concentration data to obtain corrected values.

[0075] S13. Obtain the correlation coefficient between the corrected value and microbial activity, determine the influence weight of the parameters, and then use the support vector machine algorithm to predict microbial activity and obtain the prediction results.

[0076] S14. By comparing the predicted results with real-time values, abnormal fluctuations in the fermentation environment are identified, and a basis for adjustment is obtained.

[0077] Specifically, by collecting high-frequency data on microbial activity and environmental parameters in the fermentation environment, combined with time series analysis and dynamic correlation calculation, abnormal fluctuations in the fermentation process can be accurately identified, and microbial activity can be predicted based on support vector machine, providing a basis for optimizing and adjusting fermentation conditions, thereby effectively improving the stability and quality consistency of the tea fermentation process.

[0078] S2. Based on the collected microbial activity data and environmental parameter data, the pre-established microbial metabolic model is used to calculate the metabolic product generation rate. The model input includes microbial activity, temperature, humidity and oxygen content, and the output is the trend of metabolic product concentration change.

[0079] Among them, when the trend of metabolite concentration changes exceeds the preset threshold, the direction of temperature control, humidity regulation and oxygen content adjustment is determined by comparing the current environmental parameters with the historical best parameter dataset, and environmental parameter optimization suggestions are generated.

[0080] Furthermore, in step S2, the trend of metabolite concentration changes is output, specifically including:

[0081] S21. Obtain microbial activity, temperature data, humidity data and oxygen content through sensors to obtain a complete dataset of environmental parameters. Then, process the complete dataset of environmental parameters using a pre-established microbial metabolic model to output the metabolic product generation rate.

[0082] S22. Calculate the concentration change value based on the metabolic product generation rate to obtain the preliminary result of the metabolic product concentration change. When the concentration change value exceeds the preset threshold, the change trend is classified by the support vector machine algorithm to determine the abnormal fluctuation range.

[0083] S23. Obtain temperature, humidity and oxygen content within the abnormal fluctuation range, identify key environmental factors affecting microbial activity, adjust the parameters of the microbial metabolic model through key environmental factors, and obtain the optimized trend of metabolite concentration changes.

[0084] Among them, the microbial metabolic model is based on the metabolic network and biochemical reaction pathways of microorganisms. It is constructed through mathematical modeling and computational methods, combined with steps such as genome annotation, metabolic network reconstruction, and experimental data calibration. Its principle is to use constraints such as mass conservation and energy balance to simulate the metabolic behavior of microorganisms under specific environmental conditions, thereby predicting the generation rate and concentration trend of metabolites, and providing theoretical support for the optimization of fermentation processes and the improvement of metabolite yield.

[0085] Specifically, the generation rate and concentration trend of metabolites are calculated using a microbial metabolic model. When the trend exceeds a threshold, the abnormal fluctuations are classified using a support vector machine algorithm to identify key environmental factors and adjust model parameters, thereby generating environmental parameter optimization suggestions to optimize fermentation conditions and ensure the stability and quality consistency of the fermentation process.

[0086] Furthermore, in step S2, environmental parameter optimization suggestions are generated, specifically including:

[0087] The concentration data of metabolites is acquired in real time by sensors, the concentration change trend is calculated, and the change trend value is obtained. When the change trend exceeds the preset threshold, the current environmental parameters are extracted from the database to obtain the environmental parameter set.

[0088] By comparing the set of environmental parameters with the historical best parameter dataset, the direction of parameter deviation is determined, the deviation value is determined, and the direction of temperature control adjustment is calculated based on the deviation value and preset rules to obtain the temperature optimization parameters.

[0089] A temperature optimization parameter and humidity regulation correlation model is used to calculate the humidity regulation range and obtain the humidity optimization parameter. Then, the direction of oxygen content adjustment is determined by the correlation rule between the humidity optimization parameter and oxygen content to obtain the oxygen optimization parameter.

[0090] Based on the optimized parameters for temperature, humidity, and oxygen, suggestions for optimizing environmental parameters are generated.

[0091] Among them, the temperature optimization parameter and humidity regulation correlation model is established by collecting temperature and humidity data in the environment and using mathematical modeling or machine learning methods (such as PID algorithm, neural network, etc.) to establish the coupling relationship between temperature and humidity. Through data-driven training and optimization process, the model parameters are adjusted to improve prediction accuracy, thereby realizing the strategy of dynamically adjusting humidity based on temperature changes, achieving the goal of coordinated temperature and humidity control, and improving the stability and control accuracy of the system.

[0092] Specifically, by monitoring the trend of metabolite concentration changes in real time and combining it with historical best parameter datasets, this method can accurately determine the direction of deviation of environmental parameters and calculate optimization parameters, thereby generating optimization suggestions for temperature, humidity and oxygen content. This process uses microbial metabolic models and machine learning algorithms (such as support vector machines) to dynamically optimize the fermentation process, ensuring that the fermentation conditions are always in the optimal state, thereby improving the stability and quality consistency of tea fermentation.

[0093] S3. Based on the environmental parameter optimization suggestions, the fermentation conditions are adjusted using a dynamic control algorithm. The algorithm outputs temperature control values, humidity adjustment values, and oxygen content set values ​​based on the trend of metabolite concentration changes and real-time environmental parameter values. Then, by executing the set values ​​output by the dynamic control algorithm, the heating device, humidification device, and ventilation device in the fermentation equipment are adjusted to obtain the adjusted fermentation condition data.

[0094] Furthermore, in step S3, the output of temperature control values, humidity adjustment values, and oxygen content setpoints specifically includes:

[0095] By collecting real-time values ​​of environmental parameters through sensors, the changing trend of metabolite concentration is determined. Based on the changing trend, a dynamic control algorithm is used to calculate the temperature control value and output the adjustment command.

[0096] The deviation between the real-time value and the preset threshold is obtained to determine the humidity adjustment value. When the deviation exceeds the preset range, the fermentation conditions are adjusted according to the oxygen content setting value.

[0097] Based on the adjusted fermentation conditions, the concentration changes of metabolites are analyzed to obtain updated trend data. The output value of the dynamic control algorithm is then optimized. The stability of environmental parameters is judged by the optimized output value, and the final control parameters are generated.

[0098] Furthermore, in step S3, after outputting the temperature control value, humidity adjustment value, and oxygen content setpoint, the following is also included:

[0099] The algorithm generation module is activated by a preset dynamic control strategy to output the initial set value. Real-time status data is obtained by acquiring the current operating data of the heating device, humidification device and ventilation device from the sensors of the fermentation equipment.

[0100] The random forest algorithm is used to compare real-time status data with set values ​​to determine the deviation range. When the deviation exceeds the preset threshold, the set value is updated through dynamic adjustment logic to obtain the optimized set value.

[0101] By optimizing the set values ​​to drive the operation of the heating, humidifying, and ventilation devices inside the fermentation equipment, the adjusted status data is obtained.

[0102] The fermentation condition change trend is calculated based on the adjusted state data to obtain condition optimization data. By comparing the condition optimization data with the target fermentation conditions, it is determined whether iterative adjustment of the set value is necessary.

[0103] Specifically, by combining dynamic control algorithms with real-time monitoring data, this method can accurately calculate and output optimized setpoints for temperature, humidity, and oxygen content, adjust fermentation conditions in real time, and further utilize random forest algorithms to compare real-time state data with preset thresholds, dynamically update control parameters, ensure the stability and consistency of the fermentation process, and effectively improve the quality and efficiency of tea fermentation.

[0104] S4. After obtaining the adjusted fermentation condition data, the microbial activity data and metabolite concentration data are collected again through the sensor array to determine whether the adjusted fermentation conditions have made the metabolite concentration change trend more stable. Then, the neural network algorithm is used to predict the tea quality parameters after the fermentation conditions are adjusted, and the output is the tea quality score.

[0105] Furthermore, in step S4, it is determined whether adjusting the fermentation conditions stabilizes the trend of metabolite concentration changes, specifically including:

[0106] Data preprocessing techniques were used to clean and standardize the adjusted fermentation condition data to obtain a processed dataset. Microbial activity data and metabolite concentration data were extracted from the processed dataset to determine time series characteristics.

[0107] By analyzing time series data, the trend of metabolite concentration is determined and a trend curve is obtained. When the absolute value of the slope of the trend curve is less than a preset threshold, the trend is determined to be stable, and a stability judgment result is obtained.

[0108] Based on the stability assessment results and combined with microbial activity data, the support vector machine algorithm was used to model the relationship between fermentation conditions and metabolite concentration to obtain an optimized model.

[0109] By optimizing the model, we can analyze the impact of the adjusted fermentation conditions on the stability trend of metabolite concentration data and determine the degree of condition optimization.

[0110] Specifically, the adjusted fermentation conditions are evaluated through data preprocessing, time series analysis, and support vector machine modeling. First, the fermentation condition data is cleaned and standardized, and time series characteristics of microbial activity and metabolite concentration are extracted. Then, time series analysis is used to determine the trend of metabolite concentration changes; when the absolute value of the slope of the trend curve is less than a preset threshold, the trend is considered to be stable. Finally, combined with microbial activity data, an optimization model is established using the support vector machine algorithm to analyze the impact of fermentation conditions on the stable trend of metabolite concentration, thereby determining the degree of condition optimization. This process effectively verifies the stability of the adjusted fermentation conditions, ensures the optimization effect of the fermentation process, and provides a scientific basis for improving tea quality.

[0111] Furthermore, in step S4, a neural network algorithm is used to predict the tea quality parameters after the fermentation conditions have been adjusted, specifically including:

[0112] The data integration module fuses the collected microbial activity data, metabolite concentration data, and environmental parameter data to generate a multidimensional feature dataset. The dataset is then processed using multidimensional feature extraction technology to obtain a structured feature set.

[0113] A neural network algorithm is used to train a structured feature set to determine a quality score prediction model. Initial fermentation conditions are input, and condition change data is generated after adjustment. When the condition change data exceeds a preset threshold, the adjusted quality score value is predicted by the neural network.

[0114] Based on the predicted quality score, the trend of tea quality change is determined. Through trend analysis, optimized fermentation condition parameters are obtained. The dataset is updated using the optimized fermentation condition parameters to determine the final quality score.

[0115] The tea quality rating prediction model is based on a neural network algorithm. Its principle is to use a neural network to learn and fit a multidimensional feature dataset, thereby establishing a non-linear mapping relationship between fermentation conditions and tea quality. Specifically, the data integration module fuses collected microbial activity data, metabolite concentration data, and environmental parameter data to generate a multidimensional feature dataset. Then, the dataset undergoes preprocessing and feature extraction to obtain a structured feature set. Next, the neural network algorithm is used to train this structured feature set. By inputting initial fermentation conditions and adjusted condition change data, the model can dynamically predict tea quality ratings. During training, by continuously optimizing network parameters and adjusting the structure, the model can accurately predict tea quality ratings, thus achieving precise control and quality prediction of the tea fermentation process.

[0116] Specifically, by integrating data on microbial activity, metabolite concentration, and environmental parameters to generate a multidimensional feature dataset, and training it using a neural network algorithm, this method can establish a tea quality scoring prediction model. This model can dynamically predict tea quality parameters based on changes in fermentation conditions, and optimize fermentation conditions by analyzing the prediction results, thereby achieving precise control of the tea fermentation process and effectively improving the stability of tea quality.

[0117] S5. Based on the tea quality score output by the neural network algorithm, adjust the parameter weights of the dynamic control algorithm to obtain the optimized control strategy, which is then applied to the next fermentation condition adjustment cycle.

[0118] Furthermore, in step S5, the optimized control strategy is obtained, specifically including:

[0119] The quality score of tea is obtained by using a neural network algorithm. The initial quality data is determined by the score. Key feature values ​​are then extracted from the initial quality data. The parameter weights are adjusted by a dynamic control algorithm to obtain the weight update result.

[0120] An optimized control strategy is generated based on the weight update results. The trend of fermentation conditions is judged by the optimized control strategy. When the trend exceeds the preset threshold, the fermentation conditions are adjusted through the condition update module to obtain the updated fermentation parameters.

[0121] A new round of quality score values ​​are obtained based on the updated fermentation parameters. The adaptability of the control strategy is judged by the score values. At the same time, the parameter weights of the dynamic control algorithm are adjusted according to the adaptability results to obtain a new round of optimized control strategy.

[0122] By updating the fermentation conditions and adjusting the cycle with a new round of optimized control strategies, we can obtain cycle control data.

[0123] Specifically, by dynamically adjusting the parameter weights of the control algorithm and combining them with the tea quality score predicted by the neural network, this method can optimize fermentation conditions in real time, ensuring that the fermentation process is always in the best state, thereby significantly improving the stability and consistency of tea quality, while also improving the controllability and efficiency of the fermentation process.

[0124] Example 2

[0125] Please see Figure 4 This embodiment provides a tea fermentation regulation system based on microbial activity monitoring, used to realize an intelligent regulation method for the tea fermentation process, including:

[0126] The sensor array module is used to collect microbial activity data and environmental parameter data in the fermentation environment. The sensors include a temperature sensor, a humidity sensor, and an oxygen concentration sensor, and the acquisition frequency is set to once per minute.

[0127] The data processing module is used to calculate the rate of metabolite production based on the collected microbial activity data and environmental parameter data using a pre-established microbial metabolic model. The model inputs include microbial activity, temperature, humidity and oxygen content, and the output is the trend of metabolite concentration changes.

[0128] The optimization judgment module is used to determine the direction of temperature control, humidity regulation and oxygen content adjustment when the trend of metabolite concentration changes exceeds the preset threshold by comparing the current environmental parameters with the historical best parameter dataset, and to obtain environmental parameter optimization suggestions.

[0129] The dynamic control module is used to adjust the fermentation conditions according to the optimization suggestions of environmental parameters using a dynamic control algorithm. The algorithm outputs temperature control values, humidity adjustment values ​​and oxygen content set values ​​based on the trend of metabolite concentration changes and real-time values ​​of environmental parameters.

[0130] The execution module is used to adjust the heating device, humidification device and ventilation device in the fermentation equipment according to the set value output by the dynamic control module, so as to obtain the adjusted fermentation condition data.

[0131] The feedback module is used to collect microbial activity data and metabolite concentration data again after obtaining the adjusted fermentation condition data, and to determine whether the adjusted fermentation conditions have made the trend of metabolite concentration change more stable.

[0132] The data integration module is used to fuse collected microbial activity data, metabolite concentration data, and environmental parameter data to generate a multidimensional feature dataset when the trend of metabolite concentration changes is unstable.

[0133] The quality prediction module is used to predict the tea quality parameters after the fermentation conditions are adjusted based on the generated multidimensional feature dataset using a neural network algorithm. The neural network input is the multidimensional feature dataset and the output is the tea quality score.

[0134] The optimization and adjustment module is used to adjust the parameter weights of the dynamic control algorithm based on the tea quality score output by the neural network algorithm, so as to obtain the optimized control strategy, which is then applied to the next fermentation condition adjustment cycle.

[0135] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for intelligent control of tea fermentation process, characterized in that: Includes the following steps: Collect microbial activity data and environmental parameter data in the fermentation environment to obtain real-time values ​​of microbial activity and environmental parameters; Based on the collected data, calculate the generation rate and concentration trend of metabolites, and output the concentration trend of metabolites. When the trend of metabolite concentration changes exceeds a preset threshold, environmental parameter optimization suggestions are generated. Based on the optimization suggestions for the generation environment parameters, a dynamic control algorithm is used to adjust the fermentation conditions and output temperature control values, humidity adjustment values ​​and oxygen content set values. Collect data on the adjusted fermentation conditions, determine whether the trend of metabolite concentration changes is stabilizing, and then use a neural network algorithm to predict the tea quality parameters after the fermentation conditions are adjusted, and output the tea quality score. Based on the predicted results of tea quality parameters, the control strategy is optimized and applied to the next fermentation condition adjustment cycle.

2. The intelligent control method for tea fermentation process according to claim 1, characterized in that: Obtain real-time values ​​of microbial activity and environmental parameters, specifically including: Microbial activity data and environmental parameter data in the fermentation environment are collected by a sensor array to obtain real-time values. The real-time values ​​are then divided into time series according to the collection frequency to obtain segmented data. The dynamic correlation between microbial activity and environmental parameters was calculated using segmented data. Based on this dynamic correlation, when the temperature sensor data exceeded a preset threshold, the humidity sensor and oxygen concentration data were weighted and adjusted to obtain corrected values. The correlation coefficient between the corrected values ​​and microbial activity is obtained, the influence weight of the parameters is determined, and then the support vector machine algorithm is used to predict microbial activity to obtain the prediction results. By comparing the predicted results with real-time values, abnormal fluctuations in the fermentation environment can be identified, providing a basis for adjustments.

3. The intelligent control method for tea fermentation process according to claim 1, characterized in that: The trends in the concentration changes of output metabolites include: By acquiring data on microbial activity, temperature, humidity, and oxygen content through sensors, a complete dataset of environmental parameters is obtained. Then, a pre-established microbial metabolic model is used to process the complete dataset of environmental parameters and output the rate of metabolite generation. The concentration change value is calculated based on the rate of metabolite formation to obtain the preliminary result of the concentration change of metabolites. When the concentration change value exceeds the preset threshold, the change trend is classified by the support vector machine algorithm to determine the abnormal fluctuation range. Temperature, humidity, and oxygen content data within the abnormal fluctuation range are obtained to identify key environmental factors affecting microbial activity. The parameters of the microbial metabolic model are adjusted based on these key environmental factors to obtain the optimized trend of metabolite concentration changes.

4. The intelligent control method for tea fermentation process according to claim 1, characterized in that: The system generates environmental parameter optimization suggestions, specifically including: The concentration data of metabolites is acquired in real time by sensors, the concentration change trend is calculated, and the change trend value is obtained. When the change trend exceeds the preset threshold, the current environmental parameters are extracted from the database to obtain the set of environmental parameters. By comparing the set of environmental parameters with the historical best parameter dataset, the direction of parameter deviation is determined, the deviation value is determined, and the direction of temperature control adjustment is calculated based on the deviation value and preset rules to obtain the temperature optimization parameters. A correlation model between temperature optimization parameters and humidity regulation is used to calculate the humidity regulation range and obtain humidity optimization parameters. Then, the direction of oxygen content adjustment is determined by the correlation rule between humidity optimization parameters and oxygen content to obtain oxygen optimization parameters. Based on the optimized parameters for temperature, humidity, and oxygen, suggestions for optimizing environmental parameters are generated.

5. The intelligent control method for tea fermentation process according to claim 1, characterized in that: Output temperature control values, humidity adjustment values, and oxygen content setpoints, specifically including: By collecting real-time values ​​of environmental parameters through sensors, the changing trend of metabolite concentration is determined. Based on the changing trend, a dynamic control algorithm is used to calculate the temperature control value and output the adjustment command. The deviation between the real-time value and the preset threshold is obtained to determine the humidity adjustment value. When the deviation exceeds the preset range, the fermentation conditions are adjusted according to the oxygen content setting value. Based on the adjusted fermentation conditions, the concentration changes of metabolites are analyzed to obtain updated trend data. The output value of the dynamic control algorithm is then optimized. The stability of environmental parameters is judged by the optimized output value, and the final control parameters are generated.

6. The intelligent control method for tea fermentation process according to claim 5, characterized in that: After outputting the temperature control value, humidity adjustment value, and oxygen content setpoint, it also includes: The algorithm generation module is activated by a preset dynamic control strategy to output the initial set value. Real-time status data is obtained by acquiring the current operating data of the heating device, humidification device and ventilation device from the sensors of the fermentation equipment. The random forest algorithm is used to compare real-time status data with set values ​​to determine the deviation range. When the deviation exceeds the preset threshold, the set value is updated through dynamic adjustment logic to obtain the optimized set value. By optimizing the set values ​​to drive the operation of the heating, humidifying, and ventilation devices inside the fermentation equipment, the adjusted status data is obtained. The fermentation condition change trend is calculated based on the adjusted state data to obtain condition optimization data. By comparing the condition optimization data with the target fermentation conditions, it is determined whether iterative adjustment of the set value is necessary.

7. The intelligent control method for tea fermentation process according to claim 1, characterized in that: To determine whether the trend of metabolite concentration changes is stabilizing, the following steps are taken: Data preprocessing techniques were used to clean and standardize the adjusted fermentation condition data to obtain a processed dataset. Microbial activity data and metabolite concentration data were extracted from the processed dataset to determine time series characteristics. By analyzing time series data, the trend of metabolite concentration is determined and a trend curve is obtained. When the absolute value of the slope of the trend curve is less than a preset threshold, the trend is determined to be stable, and a stability judgment result is obtained. Based on the stability assessment results and combined with microbial activity data, the support vector machine algorithm was used to model the relationship between fermentation conditions and metabolite concentration to obtain an optimized model. By optimizing the model, we can analyze the impact of the adjusted fermentation conditions on the stability trend of metabolite concentration data and determine the degree of condition optimization.

8. The intelligent control method for tea fermentation process according to claim 1, characterized in that: The neural network algorithm is used to predict tea quality parameters after fermentation conditions are adjusted, specifically including: The data integration module fuses the collected microbial activity data, metabolite concentration data, and environmental parameter data to generate a multidimensional feature dataset. The dataset is then processed using multidimensional feature extraction technology to obtain a structured feature set. A neural network algorithm is used to train a structured feature set to determine a quality score prediction model. Initial fermentation conditions are input, and condition change data is generated after adjustment. When the condition change data exceeds a preset threshold, the adjusted quality score value is predicted by the neural network. Based on the predicted quality score, the trend of tea quality change is determined. Through trend analysis, optimized fermentation condition parameters are obtained. The dataset is updated using the optimized fermentation condition parameters to determine the final quality score.

9. The intelligent control method for tea fermentation process according to claim 1, characterized in that: Optimize regulatory strategies, specifically including: The quality score of tea is obtained by using a neural network algorithm. The initial quality data is determined by the score. Key feature values ​​are then extracted from the initial quality data. The parameter weights are adjusted by a dynamic control algorithm to obtain the weight update result. An optimized control strategy is generated based on the weight update results. The trend of fermentation conditions is judged by the optimized control strategy. When the trend exceeds the preset threshold, the fermentation conditions are adjusted through the condition update module to obtain the updated fermentation parameters. A new round of quality score values ​​are obtained based on the updated fermentation parameters. The adaptability of the control strategy is judged by the score values. At the same time, the parameter weights of the dynamic control algorithm are adjusted according to the adaptability results to obtain a new round of optimized control strategy. By updating the fermentation conditions and adjusting the cycle with a new round of optimized control strategies, we can obtain cycle control data.

10. A tea fermentation control system based on microbial activity monitoring, used to implement the intelligent control method for the tea fermentation process as described in any one of claims 1-9, characterized in that: include: The sensor array module is used to collect microbial activity data and environmental parameter data in the fermentation environment. The sensors include a temperature sensor, a humidity sensor, and an oxygen concentration sensor, and the acquisition frequency is set to once per minute. The data processing module is used to calculate the generation rate of metabolites based on the collected microbial activity data and environmental parameter data using a pre-established microbial metabolic model. The microbial metabolic model inputs include microbial activity, temperature, humidity and oxygen content, and outputs the trend of metabolite concentration changes. The optimization judgment module is used to determine the direction of temperature control, humidity regulation and oxygen content adjustment when the trend of metabolite concentration changes exceeds the preset threshold by comparing the current environmental parameters with the historical best parameter dataset, and to obtain environmental parameter optimization suggestions. The dynamic control module is used to adjust the fermentation conditions according to the optimization suggestions of environmental parameters using a dynamic control algorithm. The algorithm outputs temperature control values, humidity adjustment values ​​and oxygen content set values ​​based on the trend of metabolite concentration changes and real-time values ​​of environmental parameters. The execution module is used to adjust the heating device, humidification device and ventilation device in the fermentation equipment according to the set value output by the dynamic control module, so as to obtain the adjusted fermentation condition data. The feedback module is used to collect microbial activity data and metabolite concentration data again after obtaining the adjusted fermentation condition data, and to determine whether the adjusted fermentation conditions have made the trend of metabolite concentration change more stable. The data integration module is used to fuse collected microbial activity data, metabolite concentration data, and environmental parameter data to generate a multidimensional feature dataset when the trend of metabolite concentration changes is unstable. The quality prediction module is used to predict the tea quality parameters after the fermentation conditions are adjusted based on the generated multidimensional feature dataset using a neural network algorithm. The neural network input is the multidimensional feature dataset and the output is the tea quality score. The optimization and adjustment module is used to adjust the parameter weights of the dynamic control algorithm based on the tea quality score output by the neural network algorithm, so as to obtain the optimized control strategy, which is used in the next fermentation condition adjustment cycle.

Images