A method and system for adaptive temperature and humidity control in black tea fermentation

Abstract

This invention provides a method and system for adaptive temperature and humidity control in black tea fermentation, relating to the field of tea processing technology. The method includes: dividing the temperature and humidity control cycle of black tea fermentation into stages, with each stage having a target temperature and humidity value; acquiring basic environmental parameters, as well as adjustable ranges for temperature and humidity parameters; performing fuzzy configuration of temperature and humidity control to generate fuzzy control parameters for temperature and humidity at each stage; acquiring multiple black tea fermentation images and identifying fermentation effects to establish a fermentation effect evaluation function; and optimizing parameter configuration to obtain the optimal temperature and humidity control parameters. This invention solves the technical problem that existing technologies, which typically rely on simple temperature and humidity sensors for temperature and humidity regulation, often struggle to maintain precise control throughout the fermentation process, leading to localized fluctuations in temperature and humidity, and consequently, quality variations in black tea fermentation.

Description

Technical Field

[0001] This invention relates to the field of tea processing technology, specifically to a method and system for adaptive temperature and humidity control in black tea fermentation. Background Technology

[0002] Temperature and humidity are two crucial environmental factors in black tea fermentation, affecting enzyme activity, chemical reaction rates, and the aroma and taste of the tea. In actual production, the precision and stability of temperature and humidity control significantly impact the fermentation process. Current technologies typically rely on simple temperature and humidity sensors and control systems, often failing to achieve precise control. Even modern equipment may struggle to accurately control target temperature and humidity due to environmental factors or improper equipment layout, leading to localized fluctuations. This, in turn, causes variations in black tea quality during fermentation, resulting in unstable taste and aroma, ultimately reducing the consistency and market competitiveness of the final product. Summary of the Invention

[0003] This application provides a method and system for adaptive temperature and humidity control for black tea fermentation, aiming to solve the technical problem that existing technologies, which are usually based on simple temperature and humidity sensors for temperature and humidity regulation, often fail to maintain precise control throughout the fermentation process, leading to local fluctuations in temperature and humidity, and consequently causing quality differences in the black tea fermentation process.

[0004] The first aspect disclosed in this application provides a method for adaptive temperature and humidity control in black tea fermentation. The method includes: dividing the temperature and humidity control cycle of black tea fermentation into stages, determining the initial fermentation stage, the main reaction stage, and the fermentation completion stage, each stage having a target temperature value and a target humidity value; acquiring basic environmental parameters of the black tea fermentation environment, as well as adjustable ranges for temperature and humidity parameters, the basic environmental parameters including spatial dimensions, distribution locations of temperature and humidity control devices; based on the target temperature and humidity values ​​of each stage, and in conjunction with the basic environmental parameters, performing fuzzy configuration of temperature and humidity control at the starting point of each stage, generating fuzzy temperature control parameters and fuzzy humidity control parameters for each stage; during the temperature and humidity control process using the fuzzy temperature and humidity control parameters, acquiring multiple black tea fermentation images through an image acquisition device, and identifying the fermentation effect based on the multiple black tea fermentation images, establishing a fermentation effect evaluation function; and optimizing the parameter configuration of the fuzzy temperature and humidity control parameters within the adjustable ranges for temperature and humidity parameters according to the fermentation effect evaluation function, obtaining the optimal temperature control parameters and optimal humidity control parameters.

[0005] The second aspect of this application discloses a temperature and humidity adaptive control system for black tea fermentation. This system is used in the aforementioned temperature and humidity adaptive control method for black tea fermentation. The system includes: a stage division module for dividing the temperature and humidity control cycle of black tea fermentation into stages, determining the initial fermentation stage, the main reaction stage, and the fermentation completion stage, each stage having a target temperature value and a target humidity value; an environmental basic parameter acquisition module for acquiring the environmental basic parameters of the black tea fermentation environment, including adjustable temperature and humidity ranges, the environmental basic parameters including spatial dimensions, the distribution locations of temperature control equipment, and the distribution locations of humidity control equipment; and a fuzzy configuration module for configuring the system based on the target temperature and target humidity values ​​of each stage. The system combines the environmental parameters to perform fuzzy configuration of temperature and humidity control at the starting point of each stage, generating fuzzy temperature control parameters and fuzzy humidity control parameters for each stage. The fermentation effect recognition module acquires multiple black tea fermentation images using an image acquisition device during temperature and humidity control using the aforementioned fuzzy temperature and humidity control parameters, and identifies the fermentation effect based on these images, establishing a fermentation effect evaluation function. The parameter configuration optimization module optimizes the configuration of the fuzzy temperature and humidity control parameters within the adjustable ranges of the temperature and humidity parameters according to the fermentation effect evaluation function, obtaining the optimal temperature and humidity control parameters.

[0006] One or more technical solutions provided in this application have at least the following beneficial effects:

[0007] By dividing the temperature and humidity control cycle of the black tea fermentation process into stages, it is ensured that the temperature and humidity at each stage accurately correspond to specific fermentation requirements, enabling targeted control strategies at different stages. Accurate acquisition of the basic parameters of the black tea fermentation environment allows for a better understanding and prediction of the impact of different environmental factors on temperature and humidity control. This allows for more precise adaptation to environmental changes, avoiding instability caused by improper equipment layout or space constraints. Furthermore, the configuration of fuzzy control parameters for temperature and humidity allows for fine-tuning of the target temperature and humidity at different fermentation stages. The fuzzy control method ensures that the temperature and humidity control system does not fluctuate excessively but gradually and smoothly reaches the target values, effectively guaranteeing the temperature and humidity distribution throughout the entire fermentation process. Uniformity enhances the fermentation effect of black tea. Through image acquisition and fermentation effect recognition, the system can monitor the state changes of black tea during fermentation in real time, promptly capturing potential quality problems. The established fermentation effect evaluation function can comprehensively evaluate the fermentation progress and quality of black tea, providing data support for dynamic temperature and humidity adjustment, making the control process more precise and adaptive. By optimizing parameter configuration within the adjustable range of temperature and humidity parameters, the system can automatically explore the most suitable temperature and humidity control combination for the current fermentation conditions. This not only improves the adaptability of the control process but also ensures the maximization of fermentation effect. Guided by the fermentation effect evaluation function, it can avoid human adjustment errors, automatically optimize, and ensure that the fermentation quality of black tea always meets expectations.

[0008] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0009] Figure 1 This is a schematic diagram of the temperature and humidity adaptive control method for black tea fermentation provided in an embodiment of this application.

[0010] Figure 2 This is a schematic diagram of the temperature and humidity adaptive control system for black tea fermentation provided in an embodiment of this application.

[0011] Figure labeling: Stage division module 10, environmental basic parameter acquisition module 20, fuzzy configuration module 30, fermentation effect recognition module 40, parameter configuration optimization module 50. Detailed Implementation

[0012] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0013] Example 1, as Figure 1 As shown in the embodiments of this application, an adaptive temperature and humidity control method for black tea fermentation is provided, the method comprising:

[0014] The temperature and humidity control cycle for black tea fermentation is divided into stages, defining the initial fermentation stage, the main reaction stage, and the fermentation completion stage. Each stage has target temperature and target humidity values.

[0015] The initial fermentation stage is the early stage of black tea fermentation, which usually requires a relatively high temperature and moderate humidity to promote the initial fermentation reaction. The main reaction stage is the core stage of fermentation, where the fermentation reaction is most active, and usually requires relatively stable temperature and humidity to ensure the smooth progress of the reaction. The completion stage is mainly to end the fermentation process, where the temperature and humidity gradually stabilize and are maintained under certain control until the fermentation process is complete. Each stage has specific target temperature and humidity values, which are determined according to the characteristics of different stages.

[0016] The basic environmental parameters of the black tea fermentation environment, as well as the adjustable ranges of temperature and humidity parameters, are obtained. The basic environmental parameters include the spatial dimensions, the distribution locations of temperature control equipment, and the distribution locations of humidity control equipment.

[0017] Among the basic environmental parameters, space size refers to the size of the space in the black tea fermentation environment. This affects factors such as air circulation and heat distribution. The larger the space, the more difficult it is to regulate temperature and humidity, requiring a more complex regulation system. The distribution of temperature regulation equipment affects the uniformity of temperature regulation. If the equipment is too close together, it will lead to uneven temperature and localized overheating. If the temperature regulation equipment is too far apart, the temperature may not be effectively regulated. Humidity regulation equipment includes humidifiers and dehumidifiers, used to ensure even humidity distribution. The adjustable ranges for temperature and humidity parameters are the temperature and humidity ranges set according to the equipment and environmental conditions. For example, in the initial stage of black tea fermentation, the temperature is set between 28℃ and 32℃, and the humidity is set between 80% and 85%. This range sets the maximum and minimum adjustable values.

[0018] Based on the target temperature and humidity values ​​for each stage, and in conjunction with the aforementioned environmental parameters, fuzzy configuration of temperature and humidity control is performed at the starting node of each stage, generating fuzzy temperature control parameters and fuzzy humidity control parameters for each stage.

[0019] Due to physical limitations of temperature and humidity control equipment, such as the distance between the equipment and the tea leaves, and the equipment's power, the temperature and humidity cannot be directly adjusted to the target values. Doing so could lead to localized overheating or excessive humidity, affecting the fermentation process. Therefore, a fuzzy control strategy is needed to gradually adjust the temperature and humidity. The adjustment step size and adjustment time determine the speed and magnitude of the temperature and humidity adjustment. The adjustment step size determines the magnitude of each adjustment, while the adjustment time determines the time required for each adjustment. Typically, temperature and humidity adjustment requires a relatively long time to stabilize, avoiding uneven temperature and humidity caused by rapid adjustments. Through fuzzy logic control rules, the temperature and humidity control parameters for each stage are refined into fuzzy values. For example, the target temperature and humidity values ​​are transformed into actual adjustment parameters, including the adjustment step size and adjustment time, according to the fuzzy control rules. During the fuzzy configuration process, the location and number of temperature and humidity control devices affect the efficiency of the control. Devices that are too close can lead to excessively high local temperatures and humidity, while devices that are too far away may not be able to adjust the environment in a timely manner. Therefore, it is necessary to gradually adjust the temperature and humidity to avoid excessive fluctuations, ensuring that the temperature and humidity gradually stabilize and ultimately reach the target values.

[0020] During the temperature and humidity control process using the aforementioned temperature fuzzy control parameters and humidity fuzzy control parameters, multiple black tea fermentation images are acquired using an image acquisition device, and the fermentation effect is identified based on the multiple black tea fermentation images to establish a fermentation effect evaluation function.

[0021] Image acquisition devices, such as high-definition cameras, are used to record changes in tea leaves during the fermentation process. Image acquisition should cover the entire fermentation environment to ensure that the state changes of the tea leaves are captured. By acquiring image data at multiple time points, changes in the tea leaves at different stages are captured. This image data can reflect the effectiveness of temperature and humidity regulation and details in the fermentation process. Relevant fermentation characteristics of black tea are extracted from the acquired images, including color, texture, and leaf shape. Color changes usually reflect the fermentation progress, and leaf color changes reveal the degree of oxidation. The color, texture, and shape features of multiple tea leaves obtained through image analysis are combined with fermentation progress and quality data to construct a fermentation effect evaluation function. This evaluation function correlates image features with actual fermentation quality to assess the current fermentation state.

[0022] Based on the fermentation effect evaluation function, the parameter configuration of the temperature fuzzy control parameter and the humidity fuzzy control parameter is optimized within the adjustable range of the temperature parameter and the adjustable range of the humidity parameter to obtain the optimal temperature control parameter and the optimal humidity control parameter.

[0023] The temperature and humidity control parameters are optimized using adjustable ranges for both temperature and humidity to ensure optimal fermentation of black tea. A fermentation effect evaluation function is used to analyze the impact of current temperature and humidity settings on fermentation progress and quality. If the deviation in fermentation effect caused by the current control parameters is small, the temperature and humidity control parameters are adjusted accordingly. During the optimization process, methods such as genetic algorithms, particle swarm optimization, and gradient ascent are employed to search for the optimal control parameters within the target temperature and humidity range at each stage, thus obtaining the optimal temperature and humidity control parameters to ensure the best quality of black tea fermentation.

[0024] Furthermore, it also includes:

[0025] A twin model of the black tea fermentation environment is constructed, covering the fermentation environment of the black tea. This twin model is synchronously associated with the distribution locations of the temperature and humidity control devices. Based on the fuzzy temperature and humidity control parameters, temperature and humidity control simulations are performed within the twin model to locate temperature and humidity fluctuation sensitive areas. The temperature and humidity fluctuation sensitive areas are identified by temperature extremes and temperature change rates, respectively, and the humidity and humidity fluctuation sensitive areas are identified by humidity extremes and humidity change rates. The overlapping areas of the temperature and humidity fluctuation sensitive areas are obtained, and the percentage of overlapping areas is calculated. Based on the temperature extremes, temperature and humidity change rates, humidity extremes, humidity change rates, and the percentage of overlapping areas, temperature and humidity fluctuation analysis is performed. Parameter configuration optimization constraints are generated based on the results of the temperature and humidity fluctuation analysis.

[0026] By digitizing the physical characteristics of the actual black tea fermentation environment, including spatial dimensions, the location of temperature and humidity control equipment, and the equipment's operational capabilities, a twin model of the black tea fermentation environment is created that changes synchronously with the actual environment. This model can provide real-time feedback on temperature and humidity changes and equipment operating status. The location of temperature and humidity control equipment significantly impacts the efficiency of environmental regulation. By accurately simulating the layout of these devices in the twin model, the real-time temperature and humidity distribution at each location can be obtained, thereby helping to optimize control strategies.

[0027] Temperature and humidity fuzzy control parameters are synchronized to a twin model of the black tea fermentation environment to simulate how temperature and humidity are distributed and change during fermentation. This simulation process incorporates factors such as the equipment's adjustability, the physical structure of the environment, and airflow within the space to gradually derive the environmental temperature and humidity distribution under different parameters. During fermentation, due to factors such as equipment location and heat transfer, the temperature in some areas may fluctuate significantly. Through simulation, areas with large temperature fluctuations can be identified, and the extreme temperatures and rates of change in these areas can be marked. Similarly, humidity in some areas may be affected by factors such as uneven equipment adjustment or airflow. Through simulation, areas with large humidity fluctuations can also be identified, and the extreme humidity values ​​and rates of change in humidity can be marked.

[0028] Overlapping areas refer to the regions where temperature-sensitive and humidity-sensitive areas meet. Within these overlapping areas, both temperature and humidity fluctuations are significant, potentially having a more pronounced impact on black tea fermentation. The percentage of overlapping areas represents the proportion of the overlapping portion of temperature and humidity-sensitive areas within the overall environment. A larger percentage indicates that this area requires special attention during regulation, necessitating optimization of temperature and humidity control strategies to avoid excessive fluctuations.

[0029] Based on the simulation results, the temperature and humidity fluctuations in each region are analyzed, especially in temperature and humidity sensitive areas. The analysis focuses on extreme temperature values, temperature change rates, extreme humidity values, and humidity change rates. These indicators reflect the severity of temperature and humidity changes, helping to optimize control strategies. Based on the analysis of temperature and humidity fluctuations, parameter configuration optimization constraints are generated. These constraints specify the tolerable range of temperature and humidity changes during the control process. For example, in sensitive areas, the rate of change of temperature and humidity cannot exceed a certain threshold to avoid adverse effects on the fermentation process. The parameter configuration optimization constraints provide the conditions that must be met when adjusting the control parameters, ensuring that temperature and humidity changes do not negatively impact the black tea fermentation process. Through these constraints, the temperature and humidity control system can be guided to automatically adjust parameter configurations to ensure optimal fermentation results.

[0030] Furthermore, the fermentation effect recognition based on the multiple black tea fermentation images includes:

[0031] Based on the black tea fermentation data records, a set of sample black tea fermentation images is obtained. The structural features of the black tea leaves in the sample black tea fermentation image set are analyzed to obtain a set of sample black tea leaf color features, a set of sample black tea leaf texture features, and a set of sample black tea leaf shape features. Fermentation effect is then marked to obtain a set of sample black tea fermentation progress information and a set of sample black tea fermentation quality information. Based on a convolutional neural network, using the sample black tea leaf color feature set, sample black tea leaf texture feature set, and sample black tea leaf shape feature set as input, and using the sample black tea fermentation progress information set and sample black tea fermentation quality information set as output, a fermentation progress recognition path and a fermentation quality recognition path are constructed to obtain a black tea leaf feature analysis channel. The multiple black tea fermentation images are input into the black tea leaf feature analysis channel, and multiple black tea fermentation progress information and multiple black tea fermentation quality information are output.

[0032] The black tea fermentation data record includes historical images of black tea leaves, which correspond to different time points in the black tea fermentation process. The sample black tea fermentation image set covers images of different fermentation progresses from the initial stage to the completion stage. It can be collected multiple times to form a time series dataset.

[0033] Color features reflect the changes in black tea leaves. For example, as fermentation progresses, the color of black tea leaves changes from green to different shades such as red and brown. These color changes can be quantified and the degree of fermentation can be analyzed by extracting the color distribution of the image, such as color features in RGB or HSV space. Texture features include the surface texture, wrinkles, and unevenness of the leaves. These change as fermentation progresses. Image processing methods, such as gray-level co-occurrence matrix and local binary mode, can be used to extract texture features and identify surface detail changes in black tea leaves. Shape features mainly focus on the shape changes of black tea leaves, such as whether the leaves are curled, damaged, or deformed. The shape features of the leaves can be quantified by edge detection, contour extraction, and other methods.

[0034] The extracted color, texture, and shape features are analyzed, and the fermentation progress and quality information corresponding to each image are labeled. The fermentation progress information includes "initial fermentation", "mid-term fermentation" or "complete fermentation", etc., and the fermentation quality information involves the structural quality characteristics of tea leaves.

[0035] In convolutional neural networks (CNNs), the input is the extracted features of an image, including a set of features such as color, texture, and shape. The output is fermentation progress and fermentation quality information. CNNs extract deep features from images through multiple convolutional layers, pooling layers, and fully connected layers. During training, the backpropagation algorithm automatically adjusts the weights to optimize feature extraction and classification capabilities. Through a large amount of sample data, CNNs can learn to extract the most relevant features from images, thereby accurately identifying fermentation progress and quality. The fermentation progress identification path identifies the fermentation stage in the image; the model infers whether the tea is in the early, middle, or late fermentation stage based on changes in leaf color and texture. The fermentation quality identification path assesses fermentation quality, including the color, shape, texture, and other quality indicators of the tea leaves.

[0036] Multiple new images of black tea fermentation are input into a trained convolutional neural network. Each image is processed through different layers of the network, progressively extracting features such as color, texture, and shape. The output of each image includes information on the fermentation progress and quality of the black tea. This process eliminates the need for manual observation in monitoring the black tea fermentation process, allowing for automatic machine identification and analysis, thereby improving efficiency and accuracy.

[0037] Furthermore, a fermentation effect evaluation function is established, including:

[0038] Based on the current time point, obtain the expected fermentation progress and expected fermentation quality of black tea; based on the multiple black tea fermentation progress information and multiple black tea fermentation quality information, as well as the expected black tea fermentation progress and expected fermentation quality, construct the fermentation effect evaluation function.

[0039] The expected fermentation progress of black tea is based on the planned time and objectives of black tea fermentation. At the current time point, it is predicted that the black tea should reach the desired fermentation progress. This fermentation progress target is derived from actual experience or experimental data. The expected fermentation quality of black tea is based on the standards and requirements for black tea production, and the quality targets are set, including the standard color change and standard texture change of the leaves.

[0040] The actual fermentation progress and quality are compared with the expected targets to calculate the deviations. These deviations reflect the degree to which the targets are achieved during fermentation. Deviation values ​​can be positive or negative, indicating that the fermentation progress is ahead of schedule or behind schedule, and that the quality exceeds or falls below the standard. These deviations are then weighted for evaluation. Different fermentation stages and quality indicators have different impacts on the final quality of black tea. For example, fermentation quality is more important than fermentation progress; therefore, fermentation quality is given a higher weight in the evaluation. By using a weighted approach and integrating various information, a comprehensive fermentation effect score is obtained. This score reflects the overall effect of the current fermentation state. If the score is lower than the expected standard, it indicates that there are problems in the fermentation process, requiring adjustments through temperature and humidity control. If the score is higher than the expected standard, it indicates that the fermentation process is good, and the current control parameters may be close to optimal.

[0041] Furthermore, this includes:

[0042] Within the adjustable temperature and humidity ranges, starting with the fuzzy temperature and humidity control parameters, a parameter search is performed based on parameter configuration optimization constraints to generate multiple first temperature and humidity control parameter combinations. The fermentation effect evaluation function is used to evaluate these multiple first temperature and humidity control parameter combinations, generating multiple first fitness values. Gradient ascent calculations are performed on these multiple first fitness values, and multiple second temperature and humidity control parameter combinations with the maximum gradient ascent direction are obtained based on the gradient ascent results. Using these multiple second temperature and humidity control parameter combinations as update starting points, an iterative parameter search is performed based on parameter configuration optimization constraints and the fermentation effect evaluation function until a predetermined convergence condition is met, at which point the optimal temperature and humidity control parameters are output.

[0043] The temperature and humidity fuzzy control parameters are determined by the temperature and humidity control model from the previous stage. These parameters are used to gradually approach the target temperature and humidity at each stage. The goal of the optimization process is to find the most suitable temperature and humidity combination through an optimization algorithm. Parameter configuration optimization constraints specify the tolerance range for temperature and humidity changes during the control process. For example, in sensitive areas, the rate of change of temperature and humidity cannot exceed a certain threshold to avoid adverse effects on the fermentation process. Based on these parameters, a search algorithm is used to generate multiple candidate first temperature and humidity control parameter combinations. These parameter combinations represent attempts to explore different control methods for the current environment and stage.

[0044] By collecting images of black tea fermentation and using a fermentation effect analysis model, the effect of each combination of first temperature and humidity control parameters is evaluated using a fermentation effect evaluation function. The fitness value of each parameter combination is obtained. The fitness value indicates the quality of the effect of the parameter combination during fermentation. It is usually expressed as a digital score. The higher the fitness, the better the temperature and humidity control of the combination can meet the fermentation effect requirements.

[0045] Gradient ascent is an optimization method used to find the maximum value of a function. In this step, the gradient of the fitness value is calculated to find the optimal direction for parameter adjustment. Specifically, the gradient of the current fitness is calculated, i.e., the impact of changes in various temperature and humidity control parameters on fitness. Then, based on this gradient, the parameter combinations are adjusted to increase the fitness value. The direction of maximum gradient ascent is obtained through calculation, i.e., which combinations of temperature and humidity parameters can most quickly improve fermentation. During this process, the step size and time of temperature and humidity control are adjusted according to the gradient direction to more quickly approach the optimal fermentation effect. The results of gradient ascent calculation will provide new and better combinations of temperature and humidity parameters, i.e., the second combination of temperature and humidity control parameters. These parameter combinations are more likely to achieve the best fermentation effect than the first combination.

[0046] Using the second combination of temperature and humidity control parameters as a new starting point, iterative optimization continues. By applying parameter configuration optimization constraints and fermentation effect evaluation functions again, further searching and adjusting is performed until predetermined convergence conditions, such as the maximum number of iterations or the error range, are met. After multiple iterations, the optimal temperature control parameters and optimal humidity control parameters are finally output. These parameters can achieve the best temperature and humidity control effect during the fermentation process of black tea.

[0047] Furthermore, this includes:

[0048] By assigning membership degrees between 0 and 1 to the target temperature value, target humidity value, spatial dimensions, distribution location of temperature control equipment, and distribution location of humidity control equipment respectively through membership functions, fuzzy control rules are established; based on the fuzzy control rules, temperature and humidity control configuration reasoning is performed based on fuzzy logic to generate temperature fuzzy control parameters and humidity fuzzy control parameters for each stage.

[0049] The membership function represents the degree to which a variable belongs to a certain fuzzy set. Its value ranges from 0 to 1, representing a continuous degree from "not belonging" to "completely belonging." In this step, the following important parameters are fuzzified: target temperature value, target humidity value, spatial dimensions, distribution location of temperature control equipment, and distribution location of humidity control equipment. The membership degree of each variable is set within a specific range. For example, for the target temperature value, such as 30℃, several fuzzy sets are defined: low temperature, suitable temperature, and high temperature. Each set corresponds to a membership function, representing the degree to which a temperature value belongs to that set. For instance, when the temperature is close to the low temperature value, the membership degree is close to 1, and the further away it is, the smaller the membership degree becomes. When the temperature is close to the high temperature value, the membership degree is close to 1, and the membership degree gradually decreases as it moves away from the high temperature value.

[0050] Fuzzy control rules are used to derive output results from input variables. Each rule is expressed in the form of "if, then" and inference depends on the membership degree of the input conditions. For example, the following fuzzy control rules can be defined: if the temperature is low and the humidity is high, then increase the temperature and decrease the humidity; if the temperature is moderate and the humidity is moderate, then maintain the current temperature and humidity; if the temperature is high and the humidity is low, then decrease the temperature and increase the humidity.

[0051] Employing a fuzzy inference mechanism based on membership degrees and fuzzy rules, specific control actions are derived through the relationships between fuzzy sets. The inference process utilizes fuzzy set operations, such as fuzzy synthesis, minimization, or weighted averaging, to derive appropriate control quantities, including changes in temperature and humidity regulation, based on the input membership values. Ultimately, a fuzzy output of the temperature and humidity control configuration is generated, representing the magnitude and direction of the regulation. For example, increasing the temperature can be expressed as a 0.5°C increase, and decreasing the humidity can be expressed as a 5% decrease. Based on the results of the fuzzy inference, fuzzy temperature and humidity control parameters are generated for each stage.

[0052] Furthermore, multiple images of black tea fermentation are acquired using an image acquisition device, including:

[0053] Multiple initial black tea fermentation images are acquired by moving the image acquisition device; after preprocessing the multiple initial black tea fermentation images, black tea leaf target tracking and identification are performed to determine whether the number of black tea leaf targets meets the predetermined number of black tea leaf targets; if it does not meet the target, the parameters of the image acquisition device are adjusted according to the quantity deviation until it meets the target, and then the image acquisition is performed again to obtain the multiple black tea fermentation images.

[0054] A mobile image acquisition device means that the image acquisition equipment can move freely within the fermentation environment to acquire images from multiple angles and in all directions. This helps to capture the state of the black tea leaves more comprehensively. Multiple initial images of black tea fermentation are acquired through the image acquisition device for subsequent image processing and analysis.

[0055] The preprocessing steps include: using filtering techniques to remove background noise; enhancing contrast and brightness to make the tea leaves stand out more in the image; and using algorithms such as Canny to detect the edges of the tea leaves for easier subsequent tracking. The purpose of preprocessing is to improve image quality, reduce noise, and enhance the recognizability of the tea leaves.

[0056] After image processing, target tracking of the red tea leaves in the image is performed through image segmentation and shape recognition. For example, leaves are identified by features such as shape, color, and texture. The number of detected red tea leaves in the current image is counted, and it is determined whether it meets the predetermined target number, such as how many leaves should be monitored at a certain fermentation stage. If the target number meets the predetermined requirements, subsequent analysis can continue. If the target number is insufficient, it indicates a problem with the image acquisition quality, or there may be image occlusion, acquisition angle issues, etc.

[0057] The system determines the discrepancy between the currently detected number of leaves and the predetermined target number. For example, if the goal is to capture at least 50 leaves in each image, but only 40 are detected in the current image, there is a discrepancy of 10 leaves. In this case, parameter feedback adjustments are made, including adjusting the camera's position, angle, focal length, or exposure, to ensure that more tea leaves are captured. After the feedback adjustments, image acquisition restarts, and the images are processed until the number of tea leaves reaches the predetermined requirement.

[0058] Furthermore, it also includes:

[0059] During the temperature and humidity control process using the aforementioned fuzzy control parameters for temperature and humidity, the temperature and humidity control duration window and the temperature and humidity change curve are recorded. When the temperature and humidity change curve does not meet the parameter configuration optimization constraints, the temperature and humidity control duration window is adjusted accordingly.

[0060] In actual fermentation, pre-set fuzzy control parameters for temperature and humidity are used to adjust the equipment's operating status, thereby controlling the temperature and humidity within the fermentation chamber. The temperature and humidity control duration window refers to a specific time period for temperature and humidity control operations. For example, in a certain stage, the temperature and humidity are adjusted to a target value within a certain time to ensure the stability of environmental conditions. The duration window is usually related to the adjustment step size and adjustment time of temperature and humidity. The temperature and humidity change curve is a graph showing the changes in temperature and humidity over time. This curve illustrates the trend of temperature and humidity changes throughout the entire control process.

[0061] The parameter configuration optimization constraints define the tolerable range of temperature and humidity changes during the control process. For example, in sensitive areas, the rate of change of temperature and humidity cannot exceed a certain threshold to avoid adverse effects on the fermentation process. If the actual recorded temperature and humidity change curves deviate from the predetermined conditions, such as excessively rapid, large, or slow temperature and humidity changes, it means that the current control process has not achieved the expected effect and may affect the fermentation effect of black tea. In this case, feedback adjustments are made to the temperature and humidity control time window, including: if the temperature and humidity changes are insufficient or too rapid within the current time period, the adjustment time can be extended or shortened to ensure that the temperature and humidity changes are stable and gradually approach the target value; if the temperature and humidity change curves are too drastic, the adjustment step size can be reduced and the adjustment time can be extended; if the temperature and humidity changes are too slow, the adjustment step size can be increased and the adjustment time can be shortened. Through this feedback mechanism, the control process can be continuously optimized to achieve the optimal temperature and humidity state, thereby improving the quality and effect of black tea fermentation.

[0062] Furthermore, it also includes:

[0063] At the end of each stage, a random sample of black tea leaves is extracted. The fermentation effect of the sampled black tea leaves is then evaluated using predetermined fermentation effect assessment indicators to obtain fermentation effect evaluation data and a fermentation effect evaluation coefficient. These predetermined fermentation effect evaluation indicators include fermentation aroma, moisture content, leaf weight, and tea liquor color. When the fermentation effect evaluation coefficient does not meet the current fermentation effect evaluation coefficient threshold, the optimal temperature control parameters and optimal humidity control parameters are adjusted based on the fermentation effect evaluation data.

[0064] Random sampling means selecting a certain number of black tea leaves for inspection each time in different locations or under different conditions in the fermentation environment. This method helps to avoid bias in the test results and ensure quality control of the fermentation process.

[0065] Among the predetermined fermentation effect evaluation indicators, fermentation aroma is a key characteristic produced during the fermentation process of black tea. The intensity, type, and changes of aroma are usually used to assess the degree of fermentation. Good fermentation should produce a distinct and standard aroma. During fermentation, the tea leaves gradually lose moisture, and the moisture content has a direct impact on the fermentation effect. Too much or too little moisture may lead to uneven fermentation and affect the quality of the tea. As the fermentation process progresses, the weight of the tea leaves will change. Ideal fermentation will result in an appropriate change in the weight of the tea leaves. If the leaves are too heavy or too light, it indicates that there is a problem with the fermentation process. The color change of the tea soup can indirectly reflect the fermentation status of black tea. A color that is too dark or too light indicates that the fermentation process has not proceeded as expected.

[0066] These evaluation indicators are used to obtain fermentation effect data for each piece of black tea leaves, and a set of corresponding evaluation coefficients are calculated. These evaluation coefficients quantify the effect of each indicator, forming a comprehensive evaluation of the fermentation effect.

[0067] The current evaluation thresholds for black tea fermentation are standards determined based on experience or research. For example, the threshold for aroma intensity is a specific range, and the range for moisture content also has standard values. If the actual evaluation results are lower than these thresholds, it indicates that fermentation has not reached the ideal state. When the fermentation effect of black tea at a certain stage does not meet the standards, a feedback mechanism is used to adjust the control parameters. For example, if the aroma, color, or other indicators do not meet expectations, the temperature needs to be increased or decreased to promote the fermentation process or slow down the excessively rapid reaction rate; if the moisture content does not meet the standards, the humidity needs to be adjusted to ensure that the tea leaves maintain suitable moisture and avoid being too dry or too wet, which would affect fermentation. The goal of these adjustments is to optimize the temperature and humidity control parameters so that the tea leaves can achieve better fermentation results in the next stage.

[0068] Example 2, based on the same inventive concept as the temperature and humidity adaptive control method for black tea fermentation in the foregoing examples, such as... Figure 2 As shown in the embodiment of this application, a temperature and humidity adaptive control system for black tea fermentation is provided, the system comprising:

[0069] The stage division module 10 is used to divide the temperature and humidity control cycle of black tea fermentation into stages, determining the initial fermentation stage, the main reaction stage, and the fermentation completion stage. Each stage has a target temperature value and a target humidity value. The environmental basic parameter acquisition module 20 is used to acquire the environmental basic parameters of the black tea fermentation environment, as well as the adjustable range of the temperature parameter and the adjustable range of the humidity parameter. The environmental basic parameters include spatial dimensions, the distribution location of temperature control equipment, and the distribution location of humidity control equipment. The fuzzy configuration module 30 is used to perform fuzzy configuration of temperature and humidity control at the starting node of each stage based on the target temperature value and target humidity value of each stage, combined with the environmental basic parameters. The system generates fuzzy temperature control parameters and fuzzy humidity control parameters for each stage. A fermentation effect recognition module 40 is used to acquire multiple black tea fermentation images via an image acquisition device during the temperature and humidity control process using the aforementioned fuzzy temperature control parameters and fuzzy humidity control parameters, and to identify the fermentation effect based on these images, establishing a fermentation effect evaluation function. A parameter configuration optimization module 50 is used to optimize the parameter configuration of the fuzzy temperature control parameters and fuzzy humidity control parameters within the adjustable range of the temperature and humidity parameters according to the fermentation effect evaluation function, obtaining the optimal temperature control parameters and optimal humidity control parameters.

[0070] Furthermore, the parameter configuration optimization module 50 is used to perform the following operation steps:

[0071] A twin model of the black tea fermentation environment is constructed, covering the fermentation environment of the black tea. This twin model is synchronously associated with the distribution locations of the temperature and humidity control devices. Based on the fuzzy temperature and humidity control parameters, temperature and humidity control simulations are performed within the twin model to locate temperature and humidity fluctuation sensitive areas. The temperature and humidity fluctuation sensitive areas are identified by temperature extremes and temperature change rates, respectively, and the humidity and humidity fluctuation sensitive areas are identified by humidity extremes and humidity change rates. The overlapping areas of the temperature and humidity fluctuation sensitive areas are obtained, and the percentage of overlapping areas is calculated. Based on the temperature extremes, temperature and humidity change rates, humidity extremes, humidity change rates, and the percentage of overlapping areas, temperature and humidity fluctuation analysis is performed. Parameter configuration optimization constraints are generated based on the results of the temperature and humidity fluctuation analysis.

[0072] Furthermore, the fermentation effect recognition module 40 is used to perform the following operation steps:

[0073] Based on the black tea fermentation data records, a set of sample black tea fermentation images is obtained. The structural features of the black tea leaves in the sample black tea fermentation image set are analyzed to obtain a set of sample black tea leaf color features, a set of sample black tea leaf texture features, and a set of sample black tea leaf shape features. Fermentation effect is then marked to obtain a set of sample black tea fermentation progress information and a set of sample black tea fermentation quality information. Based on a convolutional neural network, using the sample black tea leaf color feature set, sample black tea leaf texture feature set, and sample black tea leaf shape feature set as input, and using the sample black tea fermentation progress information set and sample black tea fermentation quality information set as output, a fermentation progress recognition path and a fermentation quality recognition path are constructed to obtain a black tea leaf feature analysis channel. The multiple black tea fermentation images are input into the black tea leaf feature analysis channel, and multiple black tea fermentation progress information and multiple black tea fermentation quality information are output.

[0074] Furthermore, the fermentation effect recognition module 40 is used to perform the following operation steps:

[0075] Based on the current time point, obtain the expected fermentation progress and expected fermentation quality of black tea; based on the multiple black tea fermentation progress information and multiple black tea fermentation quality information, as well as the expected black tea fermentation progress and expected fermentation quality, construct the fermentation effect evaluation function.

[0076] Furthermore, the parameter configuration optimization module 50 is used to perform the following operation steps:

[0077] Within the adjustable temperature and humidity ranges, starting with the fuzzy temperature and humidity control parameters, a parameter search is performed based on parameter configuration optimization constraints to generate multiple first temperature and humidity control parameter combinations. The fermentation effect evaluation function is used to evaluate these multiple first temperature and humidity control parameter combinations, generating multiple first fitness values. Gradient ascent calculations are performed on these multiple first fitness values, and multiple second temperature and humidity control parameter combinations with the maximum gradient ascent direction are obtained based on the gradient ascent results. Using these multiple second temperature and humidity control parameter combinations as update starting points, an iterative parameter search is performed based on parameter configuration optimization constraints and the fermentation effect evaluation function until a predetermined convergence condition is met, at which point the optimal temperature and humidity control parameters are output.

[0078] Furthermore, the fuzzy configuration module 30 is used to perform the following operation steps:

[0079] By assigning membership degrees between 0 and 1 to the target temperature value, target humidity value, spatial dimensions, distribution location of temperature control equipment, and distribution location of humidity control equipment respectively through membership functions, fuzzy control rules are established; based on the fuzzy control rules, temperature and humidity control configuration reasoning is performed based on fuzzy logic to generate temperature fuzzy control parameters and humidity fuzzy control parameters for each stage.

[0080] Furthermore, the fermentation effect recognition module 40 is used to perform the following operation steps:

[0081] Multiple initial black tea fermentation images are acquired by moving the image acquisition device; after preprocessing the multiple initial black tea fermentation images, black tea leaf target tracking and identification are performed to determine whether the number of black tea leaf targets meets the predetermined number of black tea leaf targets; if it does not meet the target, the parameters of the image acquisition device are adjusted according to the quantity deviation until it meets the target, and then the image acquisition is performed again to obtain the multiple black tea fermentation images.

[0082] Furthermore, the fermentation effect recognition module 40 is used to perform the following operation steps:

[0083] During the temperature and humidity control process using the aforementioned fuzzy control parameters for temperature and humidity, the temperature and humidity control duration window and the temperature and humidity change curve are recorded. When the temperature and humidity change curve does not meet the parameter configuration optimization constraints, the temperature and humidity control duration window is adjusted accordingly.

[0084] Furthermore, the parameter configuration optimization module 50 is used to perform the following operation steps:

[0085] At the end of each stage, a random sample of black tea leaves is extracted. The fermentation effect of the sampled black tea leaves is then evaluated using predetermined fermentation effect assessment indicators to obtain fermentation effect evaluation data and a fermentation effect evaluation coefficient. These predetermined fermentation effect evaluation indicators include fermentation aroma, moisture content, leaf weight, and tea liquor color. When the fermentation effect evaluation coefficient does not meet the current fermentation effect evaluation coefficient threshold, the optimal temperature control parameters and optimal humidity control parameters are adjusted based on the fermentation effect evaluation data.

[0086] 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 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 temperature and humidity self-adaptive control method for black tea fermentation, characterized in that, The method comprises: The temperature and humidity control cycle of black tea fermentation is divided into stages to determine the starting fermentation stage, the main reaction stage, and the fermentation completion stage, each stage having a target temperature value and a target humidity value; Obtain the environmental basic parameters of the black tea fermentation environment, as well as the temperature parameter adjustable interval and the humidity parameter adjustable interval, the environmental basic parameters including the spatial size, the temperature adjustment device distribution position, and the humidity adjustment device distribution position; According to the target temperature value and the target humidity value of each stage, combined with the environmental basic parameters, the temperature and humidity control fuzzy configuration of the starting node of each stage is performed to generate the temperature fuzzy control parameter and the humidity fuzzy control parameter of each stage; During the temperature and humidity control process using the temperature fuzzy control parameter and the humidity fuzzy control parameter, a plurality of black tea fermentation images are collected by an image acquisition device, and a fermentation effect recognition is performed based on the plurality of black tea fermentation images to establish a fermentation effect evaluation function; According to the fermentation effect evaluation function, the parameter configuration optimization of the temperature fuzzy control parameter and the humidity fuzzy control parameter is performed within the temperature parameter adjustable interval and the humidity parameter adjustable interval to obtain the optimal temperature control parameter and the optimal humidity control parameter.

2. The temperature and humidity self-adaptive control method for black tea fermentation according to claim 1, characterized in that, Further comprising: A black tea fermentation environment twin model covering the black tea fermentation environment is constructed, and the black tea fermentation environment twin model synchronously associates the temperature adjustment device distribution position and the humidity adjustment device distribution position; Based on the temperature fuzzy control parameter and the humidity fuzzy control parameter, temperature adjustment simulation and humidity adjustment simulation are performed in the black tea fermentation environment twin model to locate temperature fluctuation sensitive areas and humidity fluctuation sensitive areas, the temperature fluctuation sensitive areas being marked with temperature extreme values and temperature change rates, and the humidity fluctuation sensitive areas being marked with humidity extreme values and humidity change rates; An overlapping area of the temperature fluctuation sensitive areas and the humidity fluctuation sensitive areas is obtained, and an overlapping area proportion is calculated; Based on the temperature extreme values, the temperature change rates, the humidity extreme values, the humidity change rates, and the overlapping area proportion, a temperature and humidity fluctuation analysis is performed, and a parameter configuration optimization constraint is generated according to the temperature and humidity fluctuation analysis result.

3. The method for temperature and humidity adaptive control for black tea fermentation as claimed in claim 1 wherein, The fermentation effect recognition based on the plurality of black tea fermentation images comprises: According to black tea fermentation data records, a sample black tea fermentation image set is obtained; Black tea leaf structure feature analysis is performed on the sample black tea fermentation image set to obtain a sample black tea leaf color feature set, a sample black tea leaf texture feature set, and a sample black tea leaf shape feature set, fermentation effect labels are marked, and a sample black tea fermentation progress information set and a sample black tea fermentation quality information set are obtained; Based on a convolutional neural network, the sample black tea leaf color feature set, the sample black tea leaf texture feature set, and the sample black tea leaf shape feature set are used as inputs, and the sample black tea fermentation progress information set and the sample black tea fermentation quality information set are used as outputs to construct a fermentation progress recognition path and a fermentation quality recognition path, and a black tea leaf feature analysis channel is obtained; The plurality of black tea fermentation images are input into the black tea leaf feature analysis channel, and a plurality of black tea fermentation progress information and a plurality of black tea fermentation quality information are output.

4. The method for temperature and humidity adaptive control for black tea fermentation as claimed in claim 3 wherein, Establish a fermentation effect evaluation function, including: According to the current time node, obtain the expected black tea fermentation progress and the expected black tea fermentation quality; According to the plurality of black tea fermentation progress information and the plurality of black tea fermentation quality information, and the expected black tea fermentation progress and the expected black tea fermentation quality, the fermentation effect evaluation function is constructed.

5. The method for temperature and humidity self-adaptive control of black tea fermentation according to claim 1, characterized in that, Including: In the temperature parameter adjustable interval, humidity parameter adjustable interval, with the temperature fuzzy control parameter, humidity fuzzy control parameter as the starting point, according to the parameter configuration optimization constraint parameter search, generate a plurality of first temperature and humidity control parameter combination; Using the fermentation effect evaluation function to evaluate the plurality of first temperature and humidity control parameter combination, generate a plurality of first fitness; The plurality of first fitness is calculated by gradient ascent, and the plurality of second temperature and humidity control parameter combination with the maximum gradient ascent direction is matched and obtained based on the gradient ascent result; With the plurality of second temperature and humidity control parameter combination as the update starting point, according to the parameter configuration optimization constraint and the fermentation effect evaluation function, parameter iterative search is carried out, and when the predetermined convergence condition is reached, the optimal temperature control parameter and the optimal humidity control parameter are output.

6. The method for temperature and humidity adaptive control for black tea fermentation as claimed in claim 1 wherein, Including: Through membership function, the target temperature value, target humidity value, space size, temperature regulation equipment distribution position and humidity regulation equipment distribution position are respectively given membership degree between 0-1, and fuzzy control rule is established; Based on the fuzzy control rule, the temperature and humidity control configuration reasoning is carried out based on fuzzy logic, and the temperature fuzzy control parameter and the humidity fuzzy control parameter of each stage are generated.

7. The method for temperature and humidity self-adaptive control of black tea fermentation according to claim 1, characterized in that, A plurality of black tea fermentation images are collected by an image acquisition device, including: A plurality of initial black tea fermentation images are collected by moving the image acquisition device; After preprocessing the plurality of initial black tea fermentation images, black tea leaf target tracking identification is carried out, and whether the number of black tea leaf targets meets the predetermined number of black tea leaf targets is judged; When it is not met, the parameter feedback adjustment of the image acquisition device is carried out according to the number deviation until it is met, and then image acquisition is carried out again to obtain the plurality of black tea fermentation images.

8. The method for temperature and humidity self-adaptive regulation and control of black tea fermentation as claimed in claim 1, wherein, Also including: In the process of temperature and humidity control using the temperature fuzzy control parameter and the humidity fuzzy control parameter, the temperature and humidity control time window and the temperature and humidity change curve are recorded; When the temperature and humidity change curve does not meet the parameter configuration optimization constraint, the temperature and humidity control time window is adjusted.

9. The method for temperature and humidity self-adaptive regulation and control of black tea fermentation as claimed in claim 1, wherein, Also including: At each stage end node, a random extraction of a black tea leaf set is carried out; The fermentation effect of the extraction of the black tea leaf set is carried out by using the predetermined fermentation effect evaluation index to obtain the black tea fermentation effect extraction data and the black tea fermentation effect evaluation coefficient, wherein the predetermined fermentation effect evaluation index includes fermentation aroma, moisture content, leaf weight and tea color; When the black tea fermentation effect evaluation coefficient does not meet the current black tea fermentation effect evaluation coefficient threshold, the feedback adjustment of the optimal temperature control parameter and the optimal humidity control parameter is carried out based on the black tea fermentation effect extraction data.

10. A temperature and humidity self-adaptive control system for black tea fermentation, characterized in that, The system is used to implement the temperature and humidity adaptive control method for black tea fermentation of any one of claims 1-9, and the system comprises: The stage division module is configured to divide the temperature and humidity regulation period of black tea fermentation into stages, determine a starting fermentation stage, a main reaction stage, and a fermentation completion stage, and each stage has a target temperature value and a target humidity value. The environmental basic parameter acquisition module is configured to acquire environmental basic parameters of a black tea fermentation environment, and a temperature parameter adjustable interval and a humidity parameter adjustable interval. The environmental basic parameters include spatial dimensions, temperature adjustment device distribution positions, and humidity adjustment device distribution positions. The fuzzy configuration module is configured to perform fuzzy configuration of temperature and humidity regulation at a starting node of each stage according to the target temperature value and the target humidity value of each stage and in combination with the environmental basic parameters, and generate temperature fuzzy regulation parameters and humidity fuzzy regulation parameters for each stage. The fermentation effect recognition module is configured to acquire a plurality of black tea fermentation images through an image acquisition device during temperature and humidity regulation using the temperature fuzzy regulation parameters and the humidity fuzzy regulation parameters, recognize fermentation effects based on the plurality of black tea fermentation images, and establish a fermentation effect evaluation function. The parameter configuration optimization module is configured to perform parameter configuration optimization of the temperature fuzzy regulation parameters and the humidity fuzzy regulation parameters in the temperature parameter adjustable interval and the humidity parameter adjustable interval according to the fermentation effect evaluation function, and obtain optimal temperature regulation parameters and optimal humidity regulation parameters.

Images