Fuzhuan tea fermentation environment intelligent monitoring system based on AI optimization

Abstract

The invention discloses a Fuzhuan tea fermentation environment intelligent monitoring system based on AI optimization, and relates to the technical field of Dendrobium huoshanense processing, and the Fuzhuan tea fermentation environment intelligent monitoring system comprises the following steps: monitoring a fermentation environment in real time, and obtaining environment monitoring data from each sensor according to a preset time interval; by monitoring the fermentation environment, based on the AIAI analysis module, the environment fluctuation prediction model and the fermentation state evaluation model are constructed, the prediction of the fermentation environment change trend and the accurate evaluation of the fermentation state are realized, and according to the combination of the environment data prediction value, the stage change rate and the real-time environment monitoring data, the evaluation of the fermentation state is realized. The stage fermentation state of Fu tea fermentation is judged, overall analysis of the fermentation state is carried out, a fermentation stage state instruction of the environment is generated, basic operation parameters of equipment are automatically and dynamically adjusted according to the state instruction, intelligent and automatic regulation and control of the fermentation environment are achieved, and the Fu tea fermentation environment can be accurately monitored and regulated and controlled in real time. The fermentation is carried out in an optimal environment.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an intelligent monitoring system for a Fu'er tea fermentation environment based on AI optimization. Background Art

[0002] In recent years, artificial intelligence (AI) technology has achieved rapid development and has been widely used in various fields. In the field of environmental monitoring, AI technology has shown great advantages due to its powerful data processing, pattern recognition and intelligent decision-making capabilities. AI has the ability to self-learn and optimize. It is suitable for integrating environmental intelligent monitoring into the fermentation process of Fucha. As a special variety of dark tea, Fucha has unique processing technology and quality characteristics. Its fermentation process is an extremely complex and critical link, involving the growth and metabolic activities of various microorganisms, which plays a decisive role in the formation of tea quality. The fermentation process of Fucha is highly sensitive to environmental factors. Parameters such as temperature, humidity, and oxygen content will significantly affect the growth and reproduction of microorganisms and the activity of enzymes, thereby affecting the fermentation process and final quality of Fucha. During the traditional Fucha fermentation process, environmental parameter monitoring relies primarily on traditional sensors. While these sensors can provide basic environmental information, they have numerous shortcomings. The monitoring data is limited, relying on limited parameters such as temperature, humidity, and oxygen concentration, and cannot fully reflect the complex fermentation process. Traditional monitoring methods are also unable to respond promptly and effectively to environmental changes. Manual judgment and regulation based on limited monitoring data is not only inefficient, but also difficult to achieve precise environmental control due to human subjective factors and errors in judgment. In view of the above technical defects, a solution is now proposed. Summary of the Invention

[0003] The purpose of the present invention is to use artificial intelligence algorithms to analyze and process multi-source data, build an environmental fluctuation prediction model and a fermentation status assessment model, so as to predict environmental parameter fluctuations in advance and accurately judge the fermentation stage and status.

[0004] In order to achieve the above objectives, the present invention adopts the following technical solutions: an intelligent monitoring system for Fucha fermentation environment based on AI optimization, comprising a data monitoring module, an AI analysis module, a fluctuation analysis module, and an intelligent control module; The data monitoring module is used to monitor the fermentation environment in real time. It obtains environmental monitoring data from various sensors at preset time intervals and divides the external and internal data sets according to the environmental monitoring data. The AI analysis module includes a data analysis unit and a fusion analysis unit. The data analysis unit is used to obtain environmental monitoring data during the fermentation environment monitoring process, classify the data into four categories based on quarterly boundaries, and analyze the influence coefficient between the fermentation environment and the external environment in each category to generate an influence ratio value of the fermentation environment to obtain different influence ratio sets; The fusion analysis unit obtains environmental monitoring data and intercepts real-time image data at the time when the abnormality occurs in Fucha fermentation. It then divides the data into three stages based on the abnormal time point: normal stage state, pre-abnormal state, and abnormal state. It analyzes the fermentation process time change rate of each stage to obtain stage change rate data. The fluctuation analysis module is used to obtain environmental monitoring data, build an environmental fluctuation model based on the environmental monitoring data, and output the environmental data prediction value for the future time period. It also combines the environmental data prediction value, the stage change rate, and the real-time environmental monitoring data to determine the fermentation state of Fucha fermentation, conduct an overall analysis of the fermentation state, and generate the fermentation stage state instruction of the environment; The intelligent control module is used to control the equipment in the fermentation environment, obtain the fermentation stage status instructions, and control the equipment in real time to adjust the operating parameters of the fermentation environment according to the fermentation stage status instructions of the environment.

[0005] Furthermore, when monitoring the fermentation environment, the data monitoring module constructs a sensor implementation diagram according to the fermentation environment layout, which specifically includes the following steps: Step 1: Obtain the overall shape, size, door and window locations, and equipment distribution points of the fermentation workshop, and generate the initial implementation drawing based on the workshop construction drawing; Step 2: Select sensors based on the requirements of the Fucha fermentation process and perform high-density coverage in the monitoring area. Set the specific distance between each sensor. The distance setting varies depending on the type of sensor. Set the installation distance at different heights based on the workshop space layout and the nature of the sensor. Step 3: Mark the installation location of each sensor in the initial implementation diagram. Different sensors use different types of marking symbols. With the center point of the workshop as the origin, generate a sensor link diagram and set it as a dynamic monitoring node to obtain the constructed sensor implementation diagram.

[0006] Furthermore, the AI analysis module also includes a data processing step: Based on the different sensor types, a database for collecting data in different environments is established, and each database is automatically coded and sorted. When data is transmitted, it is automatically classified based on AI recognition of data features; The database will judge and identify the data passed in, automatically identify the data, and remove noise, erroneous data and outlier processing steps, and automatically adjust the cleaning method steps according to the passed data; According to the preset time interval, data is obtained from each sensor and stored, and the sensors are set to the same gateway communication state when connecting to the network end.

[0007] Furthermore, different influence ratio sets are obtained, specifically including the following: S100, obtaining environmental monitoring data during the fermentation environment monitoring process, dividing the data into four categories based on quarterly boundaries, generating each type of external environment and each environmental data set based on the environmental monitoring data of each category, and performing classification and labeling; S101. Obtain the fermentation environment and external environment data sets of each category, analyze the influence coefficient between the fermentation environment and external environment data sets in each category, and generate an influence ratio value of the fermentation environment. The influence ratio value is calculated as follows: Internal environmental impact coefficient: , External environment impact coefficient: , Internal environmental impact ratio: , the proportion of external environmental impact: , Where n is the number of internal environmental factors, M is the number of external environmental factors, is the influence weight set, The data of the Jth internal environment factor in the internal environment data set, is the influence weight set, is the data of the Kth external environmental factor in the external environment data set; S102: Substitute the environmental monitoring data of each category into the above formula to obtain different impact ratio values under the four categories, sort them out to obtain different impact ratio sets, and set the data mean of each category as the warning threshold.

[0008] Furthermore, the phase change rate data is obtained, including the following: S200, obtaining image information of Fu tea during fermentation through monitoring equipment, intercepting real-time image data before and after the abnormality occurs according to the time point when the Fu tea fermentation is abnormal, and dividing the process into three stages based on the abnormality time point: a normal stage state, a pre-abnormal state, and an abnormal state, wherein the normal stage state is a stage when no abnormality occurs, the pre-abnormal state is a state close to the time when the abnormality occurs, and the abnormal state is a state in which the abnormality is revealed; S201. Obtain the time of each stage of the normal stage state, the state before abnormality, and the abnormal state, analyze the change time between each stage, and perform time change rate analysis of the image fermentation process based on the change time to obtain stage change rate data. The calculation process is as follows: Normal stage state: The image taken in the normal stage is recorded as , the shooting time is recorded as ; State before abnormality: The image before the abnormality occurs, recorded as , the shooting time is recorded as ; Generate abnormal state: The image when the abnormality occurs is recorded as , the shooting time is recorded as ; Similarity objective function M: M= , The rate of change from the normal stage state to the state before abnormality is R1: R1= , The rate of change from the state before the abnormality to the abnormal state is: R2= ; S202 , performing a preset stage change rate R according to the change rates R1 and R2 .

[0009] Furthermore, an environmental fluctuation model is constructed based on environmental monitoring data, specifically including the following: S300, obtaining environmental monitoring data, dividing the environmental monitoring data into a training set, a test set, and a validation set, preprocessing the environmental monitoring data to obtain relevant feature data, using the environmental feature data as sample data as input to the input layer, analyzing the feature data in different time periods as the processing layer, and outputting predicted environmental data values for future time periods; S301, when the processing layer performs feature analysis on the input: , Where, is the data value of N at the current moment, is the average rate of change of data over a period of time, To predict the data value at a future time, is the volatility factor, and T is the predicted time from the current moment.

[0010] Furthermore, the determination of the fermentation stage of Fucha tea includes the following: S400, obtaining a real-time environmental data prediction value, identifying a data value whose environmental data prediction value exceeds a standard environmental data prediction value, identifying the real-time image data according to a preset stage change rate, obtaining a change rate of the real-time image data, and combining it with the real-time environmental monitoring data to calculate the real-time fermentation evaluation status of Fu'er tea. The calculation process is as follows: S= , Where S is the evaluation value, are all weight coefficients, E is the actual predicted value, D is the real-time environmental data value, M is the similarity value between the previous image and the standard image, m is the similarity value between the current image and the standard image, S>F, then it is the third category, F>S>f, then it is the second category, S<f, then it is the third category; S401. Obtain an evaluation value to determine the fermentation stage of the Fu tea. When the evaluation value is Class I, the Fu tea is in the normal fermentation stage. When the evaluation value is Class II, the Fu tea is in the abnormal fermentation stage. When the evaluation value is Class III, the Fu tea is in the abnormal fermentation stage. Different fermentation stage status instructions are generated according to different stages.

[0011] Furthermore, it also includes an equipment control unit for dynamically adjusting the internal environment of the fermentation site, and dynamically adjusting the basic environmental parameters of the fermentation site according to the fermentation stage combined with the warning threshold. When the fermentation status assessment determines that the fermentation is in a certain stage and the environmental parameters need to be adjusted, the equipment can work together, adjust the operating status at the same time, and judge the basic environmental parameters according to the warning threshold.

[0012] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: This AI-optimized intelligent monitoring system for Fu'er tea fermentation environment monitors the fermentation environment and, based on the AI analysis module, builds an environmental fluctuation prediction model and a fermentation status assessment model to predict the fermentation environment change trend and accurately assess the fermentation status. It also determines the fermentation status of Fu'er tea fermentation based on a combination of environmental data prediction values, stage change rates, and real-time environmental monitoring data, conducts an overall analysis of the fermentation status, generates fermentation stage status instructions for the environment, and automatically and dynamically adjusts the basic operating parameters of the equipment according to the status instructions, realizing intelligent and automated fermentation environment regulation. It can monitor and regulate the Fu'er tea fermentation environment in real time and accurately, ensuring that fermentation is carried out in the optimal environment, thereby improving the stability of Fu'er tea quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 The figure shows the overall structure diagram of the system of the present invention. DETAILED DESCRIPTION

[0014] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0015] Example 1: like Figure 1 As shown in the figure, the AI-optimized intelligent monitoring system for Fucha fermentation environment includes a data monitoring module, an AI analysis module, a fluctuation analysis module, and an intelligent control module. The data monitoring module is used to monitor the fermentation environment in real time. It obtains environmental monitoring data from various sensors at preset time intervals and divides the external and internal data sets according to the environmental monitoring data. The AI analysis module includes a data analysis unit and a fusion analysis unit. The data analysis unit is used to obtain environmental monitoring data during the fermentation environment monitoring process, classify the data into four categories based on quarterly boundaries, and analyze the influence coefficient between the fermentation environment and the external environment in each category to generate an influence ratio value of the fermentation environment to obtain different influence ratio sets; The fusion analysis unit obtains environmental monitoring data and intercepts real-time image data at the time when the abnormality occurs in Fucha fermentation. It then divides the data into three stages based on the abnormal time point: normal stage state, pre-abnormal state, and abnormal state. It analyzes the fermentation process time change rate of each stage to obtain stage change rate data. The fluctuation analysis module is used to obtain environmental monitoring data, build an environmental fluctuation model based on the environmental monitoring data, and output the environmental data prediction value for the future time period. It also combines the environmental data prediction value, the stage change rate, and the real-time environmental monitoring data to determine the fermentation state of Fucha fermentation, conduct an overall analysis of the fermentation state, and generate the fermentation stage state instruction of the environment; The intelligent control module is used to control the equipment in the fermentation environment, obtain the fermentation stage status instructions, and control the equipment in real time to adjust the operating parameters of the fermentation environment according to the fermentation stage status instructions of the environment.

[0016] When monitoring the fermentation environment, the data monitoring module constructs a sensor implementation diagram according to the fermentation environment layout, specifically including the following steps: Step 1: Obtain the overall shape, size, door and window locations, and equipment distribution points of the fermentation workshop, and generate the initial implementation drawing based on the workshop construction drawing; Step 2: Select sensors based on the requirements of the Fucha fermentation process and perform high-density coverage in the monitoring area. Set the specific distance between each sensor. The distance setting varies depending on the type of sensor. Set the installation distance at different heights based on the workshop space layout and the nature of the sensor. Step 3: Mark the installation location of each sensor in the initial implementation diagram. Different sensors use different types of marking symbols. With the center point of the workshop as the origin, generate a sensor link diagram and set it as a dynamic monitoring node to obtain the constructed sensor implementation diagram.

[0017] The AI analysis module also includes data processing steps: Based on the different sensor types, a database for collecting data in different environments is established, and each database is automatically coded and sorted. When data is transmitted, it is automatically classified based on AI recognition of data features; The database will judge and identify the data passed in, automatically identify the data, and remove noise, erroneous data and outlier processing steps, and automatically adjust the cleaning method steps according to the passed data; According to the preset time interval, data is obtained from each sensor and stored, and the sensors are set to the same gateway communication state when connecting to the network end.

[0018] To obtain different influence ratio sets, specifically include the following: S100, obtaining environmental monitoring data during the fermentation environment monitoring process, dividing the data into four categories based on quarterly boundaries, generating each type of external environment and each environmental data set based on the environmental monitoring data of each category, and performing classification and labeling; S101. Obtain the fermentation environment and external environment data sets of each category, analyze the influence coefficient between the fermentation environment and external environment data sets in each category, and generate an influence ratio value of the fermentation environment. The influence ratio value is calculated as follows: Internal environmental impact coefficient: , External environment impact coefficient: , Internal environmental impact ratio: , the proportion of external environmental impact: , Where n is the number of internal environmental factors, M is the number of external environmental factors, is the influence weight set, The data of the Jth internal environment factor in the internal environment data set, is the influence weight set, is the data of the Kth external environmental factor in the external environment data set; S102: Substitute the environmental monitoring data of each category into the above formula to obtain different impact ratio values under the four categories, sort them out to obtain different impact ratio sets, and set the data mean of each category as the warning threshold.

[0019] The stage change rate data is obtained, including the following: S200, obtaining image information of Fu tea during fermentation through monitoring equipment, intercepting real-time image data before and after the abnormality occurs according to the time point when the Fu tea fermentation is abnormal, and dividing the process into three stages based on the abnormality time point: a normal stage state, a pre-abnormal state, and an abnormal state, wherein the normal stage state is a stage when no abnormality occurs, the pre-abnormal state is a state close to the time when the abnormality occurs, and the abnormal state is a state in which the abnormality is revealed; S201. Obtain the time of each stage of the normal stage state, the state before abnormality, and the abnormal state, analyze the change time between each stage, and perform time change rate analysis of the image fermentation process based on the change time to obtain stage change rate data. The calculation process is as follows: Normal stage state: The image taken in the normal stage is recorded as , the shooting time is recorded as ; State before abnormality: The image before the abnormality occurs, recorded as , the shooting time is recorded as ; Generate abnormal state: The image when the abnormality occurs is recorded as , the shooting time is recorded as ; Similarity objective function M: M= , The rate of change from the normal stage state to the state before abnormality is R1: R1= , The rate of change from the state before the abnormality to the abnormal state is: R2= ; S202 , performing a preset stage change rate R according to the change rates R1 and R2 .

[0020] An environmental fluctuation model is constructed based on environmental monitoring data, specifically including the following: S300, obtaining environmental monitoring data, dividing the environmental monitoring data into a training set, a test set, and a validation set, preprocessing the environmental monitoring data to obtain relevant feature data, using the environmental feature data as sample data as input to the input layer, analyzing the feature data in different time periods as the processing layer, and outputting predicted environmental data values for future time periods; S301, when the processing layer performs feature analysis on the input: , Where, is the data value of N at the current moment, is the average rate of change of data over a period of time, To predict the data value at a future time, is the volatility factor, and T is the predicted time from the current moment.

[0021] The determination of the fermentation stage of Fucha tea includes the following: S400, obtaining a real-time environmental data prediction value, identifying a data value whose environmental data prediction value exceeds a standard environmental data prediction value, identifying the real-time image data according to a preset stage change rate, obtaining a change rate of the real-time image data, and combining it with the real-time environmental monitoring data to calculate the real-time fermentation evaluation status of Fu'er tea. The calculation process is as follows: S= , Where S is the evaluation value, are all weight coefficients, E is the actual predicted value, D is the real-time environmental data value, M is the similarity value between the previous image and the standard image, m is the similarity value between the current image and the standard image, S>F, then it is the third category, F>S>f, then it is the second category, S<f, then it is the third category; S401. Obtain an evaluation value to determine the fermentation stage of the Fu tea. When the evaluation value is Class I, the Fu tea is in the normal fermentation stage. When the evaluation value is Class II, the Fu tea is in the abnormal fermentation stage. When the evaluation value is Class III, the Fu tea is in the abnormal fermentation stage. Different fermentation stage status instructions are generated according to different stages.

[0022] It also includes an equipment control unit for dynamically adjusting the internal environment of the fermentation site, and dynamically adjusting the basic environmental parameters of the fermentation site according to the fermentation stage and the early warning threshold. When the fermentation status assessment determines that the fermentation is at a certain stage and the environmental parameters need to be adjusted, the equipment can work together, adjust the operating status at the same time, and judge the basic environmental parameters according to the early warning threshold.

[0023] In this solution, the data monitoring module can monitor the fermentation environment in real time and obtain data from various sensors at preset time intervals to ensure timely capture of changes in the fermentation environment and provide accurate and up-to-date data support for subsequent analysis. The division of external and internal data sets based on environmental monitoring data helps to analyze the different effects of external and internal environmental factors on Fucha fermentation, making subsequent analysis more targeted. Impact coefficient and ratio analysis: The AI analysis module's data analysis unit categorizes data by quarter, analyzes the impact coefficients of the fermentation environment and the external environment, and generates impact ratio values, helping to understand the extent to which environmental factors affect fermentation in different seasons. When abnormalities occur in the fermentation of Fucha tea, the fusion analysis unit intercepts real-time image data and divides the fermentation process into different stages to analyze the time change rate; Environmental fluctuation modeling and prediction: The fluctuation analysis module builds an environmental fluctuation model based on environmental monitoring data and outputs predicted values of future environmental data. This can predict environmental change trends in advance and prepare for adjusting the fermentation environment. Combining the predicted value of environmental data, the stage change rate and real-time environmental monitoring data, the stage and status of Fucha fermentation can be comprehensively judged, which can more accurately and reliably evaluate the fermentation status and avoid the one-sidedness of single data judgment. The intelligent control module controls the equipment in the fermentation environment in real time according to the status instructions of the fermentation stage, automatically adjusts the operating parameters, and realizes intelligent and automatic control of the fermentation environment. It does not require frequent human intervention, improves production efficiency and product quality stability, and reduces fermentation problems caused by untimely human operation. Through the collaborative work of the above modules, the system can gain an in-depth understanding of the impact of environmental factors on the fermentation process of Fu'er tea, detect abnormalities in a timely manner and adjust the fermentation environment, thereby optimizing the entire fermentation process, improving the quality and production efficiency of Fu'er tea, and reducing production costs.

[0024] The size of the interval and threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by technical personnel in this field for each set of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value.

[0025] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by those skilled in the art according to actual conditions. In the two embodiments provided in this application, it should be understood that the disclosed system can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the module division is merely a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not implemented. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or module, which may be electrical, mechanical or other forms. The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. An AI-optimized intelligent monitoring system for Fucha fermentation environment, characterized in that: Including data monitoring module, AI analysis module, fluctuation analysis module, and intelligent control module; The data monitoring module is used to monitor the fermentation environment in real time. It obtains environmental monitoring data from various sensors at preset time intervals and divides the external and internal data sets according to the environmental monitoring data. The AI analysis module includes a data analysis unit and a fusion analysis unit. The data analysis unit is used to obtain environmental monitoring data during the fermentation environment monitoring process, classify the data into four categories based on quarterly boundaries, and analyze the influence coefficient between the fermentation environment and the external environment in each category to generate an influence ratio value of the fermentation environment to obtain different influence ratio sets; The fusion analysis unit obtains environmental monitoring data and intercepts real-time image data at the time when the abnormality occurs in Fucha fermentation. It then divides the data into three stages based on the abnormal time point: normal stage state, pre-abnormal state, and abnormal state. It analyzes the fermentation process time change rate of each stage to obtain stage change rate data. The fluctuation analysis module is used to obtain environmental monitoring data, build an environmental fluctuation model based on the environmental monitoring data, and output the environmental data prediction value for the future time period. It also combines the environmental data prediction value, the stage change rate, and the real-time environmental monitoring data to determine the fermentation state of Fucha fermentation, conduct an overall analysis of the fermentation state, and generate the fermentation stage state instruction of the environment; The intelligent control module is used to control the equipment in the fermentation environment, obtain the fermentation stage status instructions, and control the equipment in real time to adjust the operating parameters of the fermentation environment according to the fermentation stage status instructions of the environment.

2. The AI-optimized intelligent monitoring system for Fucha fermentation environment according to claim 1 is characterized in that: When monitoring the fermentation environment, the data monitoring module constructs a sensor implementation diagram according to the fermentation environment layout, specifically including the following steps: Step 1: Obtain the overall shape, size, door and window locations, and equipment distribution points of the fermentation workshop, and generate the initial implementation drawing based on the workshop construction drawing; Step 2: Select sensors based on the requirements of the Fucha fermentation process and perform high-density coverage in the monitoring area. Set the specific distance between each sensor. The distance setting varies depending on the type of sensor. Set the installation distance at different heights based on the workshop space layout and the nature of the sensor. Step 3: Mark the installation location of each sensor in the initial implementation diagram. Different sensors use different types of marking symbols. With the center point of the workshop as the origin, generate a sensor link diagram and set it as a dynamic monitoring node to obtain the constructed sensor implementation diagram.

3. The AI-optimized intelligent monitoring system for Fucha fermentation environment according to claim 1 is characterized in that: The AI analysis module also includes data processing steps: Based on the different sensor types, a database for collecting data in different environments is established, and each database is automatically coded and sorted. When data is transmitted, it is automatically classified based on AI recognition of data features; The database will judge and identify the data passed in, automatically identify the data, and remove noise, erroneous data and outlier processing steps, and automatically adjust the cleaning method steps according to the passed data; According to the preset time interval, data is obtained from each sensor and stored, and the sensors are set to the same gateway communication state when connecting to the network end.

4. The AI-optimized intelligent monitoring system for Fucha fermentation environment according to claim 1 is characterized in that: To obtain different influence ratio sets, specifically include the following: S100, obtaining environmental monitoring data during the fermentation environment monitoring process, dividing the data into four categories based on quarterly boundaries, generating each type of external environment and each environmental data set based on the environmental monitoring data of each category, and performing classification and labeling; S101. Obtain the fermentation environment and external environment data sets of each category, analyze the influence coefficient between the fermentation environment and external environment data sets in each category, and generate an influence ratio value of the fermentation environment. The influence ratio value is calculated as follows: Internal environmental impact coefficient: , External environment impact coefficient: , Internal environmental impact ratio: , the proportion of external environmental impact: , Where n is the number of internal environmental factors, M is the number of external environmental factors, is the influence weight set, The data of the Jth internal environment factor in the internal environment data set, is the influence weight set, is the data of the Kth external environmental factor in the external environment data set; S102: Substitute the environmental monitoring data of each category into the above formula to obtain different impact ratio values under the four categories, sort them out to obtain different impact ratio sets, and set the data mean of each category as the warning threshold.

5. The AI-optimized intelligent monitoring system for Fucha fermentation environment according to claim 1 is characterized in that: The stage change rate data is obtained, including the following: S200, obtaining image information of Fu tea during fermentation through monitoring equipment, intercepting real-time image data before and after the abnormality occurs according to the time point when the Fu tea fermentation is abnormal, and dividing the process into three stages based on the abnormality time point: a normal stage state, a pre-abnormal state, and an abnormal state, wherein the normal stage state is a stage when no abnormality occurs, the pre-abnormal state is a state close to the time when the abnormality occurs, and the abnormal state is a state in which the abnormality is revealed; S201. Obtain the time of each stage of the normal stage state, the state before abnormality, and the abnormal state, analyze the change time between each stage, and perform time change rate analysis of the image fermentation process based on the change time to obtain stage change rate data. The calculation process is as follows: Normal stage state: The image taken in the normal stage is recorded as , the shooting time is recorded as ; State before abnormality: The image before the abnormality occurs, recorded as , the shooting time is recorded as ; Generate abnormal state: The image when the abnormality occurs is recorded as , the shooting time is recorded as ; Similarity objective function M: M= ， The rate of change from the normal stage state to the state before abnormality is R1: R1= , The rate of change from the state before the abnormality to the abnormal state is: R2= ; S202 , performing a preset stage change rate R according to the change rates R1 and R2 .

6. The AI-optimized intelligent monitoring system for Fucha fermentation environment according to claim 1 is characterized in that: An environmental fluctuation model is constructed based on environmental monitoring data, specifically including the following: S300, obtaining environmental monitoring data, dividing the environmental monitoring data into a training set, a test set, and a validation set, preprocessing the environmental monitoring data to obtain relevant feature data, using the environmental feature data as sample data as input to the input layer, analyzing the feature data in different time periods as the processing layer, and outputting predicted environmental data values for future time periods; S301, when the processing layer performs feature analysis on the input: ， Where, is the data value of N at the current moment, is the average rate of change of data over a period of time, To predict the data value at a future time, is the volatility factor, and T is the predicted time from the current moment.

7. The AI-optimized intelligent monitoring system for Fucha fermentation environment according to claim 1 is characterized in that: The determination of the fermentation stage of Fucha tea includes the following: S400, obtaining a real-time environmental data prediction value, identifying a data value whose environmental data prediction value exceeds a standard environmental data prediction value, identifying the real-time image data according to a preset stage change rate, obtaining a change rate of the real-time image data, and combining it with the real-time environmental monitoring data to calculate the real-time fermentation evaluation status of Fu'er tea. The calculation process is as follows: S= ， Where S is the evaluation value, are all weight coefficients, E is the actual predicted value, D is the real-time environmental data value, M is the similarity value between the previous image and the standard image, m is the similarity value between the current image and the standard image, S>F, then it is the third category, F>S>f, then it is the second category, S<f, then it is the third category; S401. Obtain an evaluation value to determine the fermentation stage of the Fu tea. When the evaluation value is Class I, the Fu tea is in the normal fermentation stage. When the evaluation value is Class II, the Fu tea is in the abnormal fermentation stage. When the evaluation value is Class III, the Fu tea is in the abnormal fermentation stage. Different fermentation stage status instructions are generated according to different stages.

8. The AI-optimized intelligent monitoring system for Fucha fermentation environment according to claim 1 is characterized in that: It also includes an equipment control unit for dynamically adjusting the internal environment of the fermentation site, and dynamically adjusting the basic environmental parameters of the fermentation site according to the fermentation stage and the early warning threshold. When the fermentation status assessment determines that the fermentation is at a certain stage and the environmental parameters need to be adjusted, the equipment can work together, adjust the operating status at the same time, and judge the basic environmental parameters according to the early warning threshold.

Images