Tea leaf fermentation intelligent control system and method

Abstract

The application discloses a kind of tea fermentation intelligent control system and method, it is related to tea processing technical field, including the following steps: S001, establish synchronous acquisition process, continuous record wind speed rhythm, spray rhythm, air pressure, humidity and leaf surface temperature, and generate wet heat intersection diagram according to time sequence, for identifying the overlapping relationship of wind flow and humidification rhythm;S002, based on wet heat intersection diagram identification rhythm overlapping area, set up spray convergence window and stop spraying buffer window in rhythm overlapping area, by adjusting spray dwell time and diffusion range, generate phase pointer data for time sequence coordination.The application realizes rhythm overlapping identification by multi-parameter synchronous acquisition and wet heat intersection diagram, realizes wind mist dislocation control based on phase pointer, and releases evaporation stress by combining measures such as mist particle preposition, temperature rise gentle slope, wind direction guide, constructs closed-loop control based on respiratory rhythm index, realizes self-adaptive steady-state regulation of fermentation environment.

Description

Technical Field

[0001] This invention relates to the field of tea processing technology, specifically to an intelligent control system and method for tea fermentation. Background Technology

[0002] Intelligent regulation of tea fermentation refers to a comprehensive control method that utilizes a combination of sensing, data analysis, and intelligent algorithms to monitor, dynamically analyze, and adaptively adjust key environmental parameters affecting fermentation quality (such as temperature, humidity, oxygen content, carbon dioxide concentration, leaf water activity, and aroma substance release patterns) throughout the entire fermentation process. By establishing a digital twin model or data-driven model of the tea fermentation process, the system can automatically determine the fermentation stage and identify abnormal fluctuations based on real-time data. Through intelligent algorithms, it adjusts actions such as heating, ventilation, humidification, and turning frequency to achieve precise control and stable optimization of fermentation conditions. This intelligent regulation not only overcomes the lag and uncertainty inherent in traditional manual regulation but also significantly improves the consistency of tea aroma, taste, and color, ensuring that the fermentation process takes place under optimal conditions, thereby achieving standardized and high-end tea production.

[0003] The existing technology has the following shortcomings:

[0004] In existing technologies, high humidity control during tea fermentation typically relies on a fixed-rhythm humidification and constant-speed airflow strategy to maintain ambient humidity and airflow circulation, preventing uneven drying of the tea leaves. However, because existing technologies do not fully consider the dynamic coupling relationship between humidification rhythm and airflow adjustment, when these two factors overlap in the time domain, they can easily trigger a resonance response within the environmental field. At this point, the water vapor released by the humidifier has not yet completely dispersed, and the airflow switching causes synchronous oscillations in local air pressure and humidity, forming high-frequency evaporative stress pulses. This causes the cell membranes on the surface of the tea leaves to undergo rapid expansion and contraction changes in a short period. As the micro-periodic resonance continues to accumulate, the amplitude of the evaporative stress continuously amplifies, ultimately leading to an increased cell membrane rupture rate, leaf structure collapse, premature termination of enzymatic reactions, and disruption of the aroma generation chain, resulting in irreversible quality degradation.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent control system and method for tea fermentation to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for intelligent control of tea fermentation, comprising the following steps:

[0008] S001, Establish a synchronous acquisition process to continuously record wind speed beat, spray beat, air pressure, humidity and leaf surface temperature, and generate a heat-humidity convergence map in chronological order to identify the overlap between airflow and humidification beat;

[0009] S002, based on the wet and heat convergence map, identify the beat overlap area, set the spray convergence window and the stop spray buffer window in the beat overlap area, and generate phase pointer data for timing coordination by adjusting the spray residence time and diffusion range;

[0010] S003 adjusts airflow control based on phase pointer data, delays airflow initiation during peak spraying periods, and applies light winds earlier during low spraying periods, forming a rhythm misalignment mechanism and generating a list of safe windows for coordinated operation of airflow and water vapor.

[0011] S004, according to the safety window list, perform evaporative stress release operation, stabilize the force on the leaf water film and extract the respiratory beat index by releasing mist particles in advance, controlling the temperature rise slowly and guiding the wind direction to one side of rotation.

[0012] S005 dynamically adjusts the fermentation environment based on the respiratory beat index. Through beat misalignment guidance, phase diversion, and the construction of alternating static wind and water vapor circulation paths, the airflow and water vapor form an inverse balance in the time dimension, guiding the evaporation stress into the slow release range and completing dynamic steady-state control.

[0013] Preferably, step S001 includes:

[0014] By deploying wind speed detection devices, spray frequency detection devices, air pressure sensing devices, humidity detection devices, and contact surface temperature sensing devices in the fermentation space, wind speed beat, spray beat, air pressure, humidity, and leaf surface temperature are collected synchronously at a fixed frequency.

[0015] The collected data is labeled with sampling time, device number and measurement point coordinates, and aligned with a unified timestamp to form a synchronized time series basic dataset;

[0016] The time series of each parameter are arranged vertically and superimposed to construct a heat-humidity convergence diagram, which is used to identify the rhythmic overlap relationship between airflow changes and humidification rhythm.

[0017] Based on the high and low switching points of wind speed beats, a time window is set, and the changing trends of spray beat, air pressure, humidity and leaf surface temperature are combined to extract the beat overlap area and generate a beat overlap feature list.

[0018] Preferably, step S002 includes:

[0019] In the heat-humidity convergence map, the wind speed beat curve is used as the positioning reference to identify the boundary points of the wind speed change slope and combine the spray beat, humidity, air pressure and leaf surface temperature change trends to determine the beat overlap area.

[0020] Within the overlapping area of ​​the spray beats, spray convergence windows and spray stop buffer windows are set. By adjusting the spray dwell time and spatial diffusion range, the time period and range of spray release behavior are limited.

[0021] Based on the slope of wind speed change, time anchors are constructed, and spray release density, humidity fluctuation amplitude, air pressure oscillation frequency and leaf surface temperature peak-valley spacing are mapped to the time axis to form a time series reference structure and extract phase paths.

[0022] The anchor points in the phase path are integrated and sorted, dynamically calibrated, and behavior labels are added to form a phase pointer data set used to drive the control logic.

[0023] Preferably, each anchor point in the phase pointer data set corresponds to a beat segment within a wind speed change cycle, and is attached with behavioral labels for controlling spray initiation, spray pause, delayed airflow initiation, and early airflow initiation, in order to achieve precise timing coordination between spray and airflow.

[0024] Preferably, step S003 includes:

[0025] Based on phase pointer data, the start time of the airflow during the high spray intensity phase is delayed, and the start time of the airflow during the low spray intensity phase is advanced, thus creating a temporal misalignment between wind and fog.

[0026] Based on the timing of the staggered execution, adjust the wind speed level and wind flow path structure, set a wind speed gradual increase curve in the delayed stage of wind flow, set a gentle push wind speed in the advanced stage of wind flow and guide the wind direction to bypass one side.

[0027] Based on the time misalignment and parameter response between airflow and spray, a list of high-stability time periods with stable water vapor diffusion, weak air pressure fluctuations, and stable leaf surface temperature is extracted to form a safety window list.

[0028] The safety window list is tagged with avoidance and advancement tags to construct a time-series control basis for subsequent dynamic adjustment.

[0029] Preferably, the safety window list is generated based on the time offset between wind speed and spray release, and simultaneously meets the comprehensive judgment conditions of stable humidity decrease, continuous air pressure change and mild leaf surface temperature fluctuation.

[0030] Preferably, step S004 includes:

[0031] Before the safety window begins, release the fine-particle droplets in advance and form a uniform wet film on the leaf pile surface through low-intensity spraying, so that water vapor and the leaf surface can complete the initial exchange before the wind starts.

[0032] After the wet film is formed, the ambient temperature is gradually increased. By adjusting the heat output rate of the heating device, the temperature rises slowly and enters the thermal stability zone before the target threshold, so as to reduce sudden changes in the leaf surface evaporation rate.

[0033] After the temperature rises slowly, the wind direction is rotated to one side to guide the wind. The wind guide structure forms a spiral streamline, allowing the airflow to slide along the blade surface to distribute wind energy evenly and stabilize evaporation stress.

[0034] After wind direction guidance is completed, real-time monitoring of leaf surface temperature, humidity and wind speed changes is performed. Data on the consistency of temperature fluctuation cycle, humidity response speed and wind speed frequency are extracted to generate a breathing rhythm index that reflects environmental stability.

[0035] Preferably, the respiratory beat index is extracted based on the periodic fluctuations of leaf surface temperature, ambient humidity and wind speed over a continuous period of time, and the phase consistency and fluctuation amplitude stability of the three are used as the judgment criteria.

[0036] Preferably, step S005 includes:

[0037] Based on the respiratory beat index, the fermentation cycle is divided into multiple micro-cycle beat units. By adjusting the time offset of airflow initiation and humidification behavior, beat misalignment guidance is performed to avoid wind and fog overlapping in the same time window.

[0038] Based on the phase trend between rhythm variables, a deflection channel is introduced into the airflow path and the humidification triggering time is delayed to perform phase diversion operation, so that the rhythm of airflow, temperature and humidity maintains a staggered and complementary relationship in the time dimension.

[0039] Based on rhythm coordination, alternating airflow static control is implemented, dividing the fermentation cycle into an operating phase and a static phase. During the static phase, airflow and humidification are simultaneously shut off to achieve a transition from evaporation stress to a slow-release range.

[0040] By constructing a water vapor circulation pathway network consisting of vertical ventilation units and grid diffusion zones, a closed-loop guidance of air heat conduction, water vapor diffusion, and leaf surface evaporation behavior is achieved, forming a dynamic steady-state control mechanism for evaporation stress.

[0041] A smart control system for tea fermentation includes a data acquisition module, a cycle analysis module, a collaborative control module, a stress relief module, and a steady-state control module.

[0042] The data acquisition module establishes a synchronous acquisition process, continuously records wind speed beat, spray beat, air pressure, humidity and leaf surface temperature, and generates a heat and humidity convergence map in chronological order to identify the overlap between airflow and humidification beat.

[0043] The beat analysis module identifies beat overlap areas based on the wet-heat convergence map, sets spray convergence windows and stop spray buffer windows in the beat overlap areas, and generates phase pointer data for timing coordination by adjusting the spray residence time and diffusion range.

[0044] The coordinated control module adjusts the airflow control based on the phase pointer data, delaying the start of airflow during the peak spraying period and applying light wind in advance during the low spraying period, forming a rhythm misalignment mechanism and generating a list of safe windows for coordinated operation of airflow and water vapor.

[0045] The stress relief module performs evaporative stress relief operations according to the safety window list. It stabilizes the stress on the leaf surface water film and extracts the respiratory rhythm index by releasing mist particles in advance, controlling the temperature rise slowly and guiding the wind direction to one side.

[0046] The steady-state control module dynamically adjusts the fermentation environment based on the respiratory beat index. Through beat misalignment guidance, phase diversion, and the construction of alternating static wind and water vapor circulation paths, the airflow and water vapor form an inverse balance in the time dimension, guiding the evaporation stress into the slow release range and completing dynamic steady-state control.

[0047] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0048] This invention achieves precise positioning of the rhythm overlap zone by constructing a multi-parameter synchronous acquisition mechanism and a humidity-heat convergence map. It also guides the wind and fog behavior rhythm misalignment through phase pointer data, ensuring dynamic avoidance between wind and fog over time, thus effectively preventing pressure resonance and sudden changes in leaf stress caused by localized humidity-heat superposition. Simultaneously, by implementing combined slow-release measures such as fog particle pre-positioning, gradual temperature rise, and wind direction guidance within a safe window, the evaporation behavior of the leaf water film tends to stabilize. Furthermore, a real-time closed-loop control strategy based on the respiratory rhythm index enables the fermentation environment to have adaptive adjustment and disturbance suppression capabilities, ultimately achieving steady-state operation of the tea fermentation process under dynamic changes. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0050] Figure 1 This is a flowchart of a method for intelligent regulation of tea fermentation according to the present invention.

[0051] Figure 2 This is a schematic diagram of a module of an intelligent control system for tea fermentation according to the present invention. Detailed Implementation

[0052] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0053] This invention provides, for example Figure 1 The method for intelligent control of tea fermentation shown includes the following steps:

[0054] S001, Establish a synchronous acquisition process, continuously record wind speed beat, spray beat, air pressure, humidity and leaf surface temperature through a synchronous acquisition device, and arrange the recorded data in chronological order to generate a humidity and heat convergence diagram, which is used to identify the beat overlap relationship between airflow changes and humidification rhythm.

[0055] To achieve high-precision rhythm recognition and dynamic coordinated control during tea fermentation, a data structure based on multi-parameter time-series information must first be constructed to reveal the potential overlapping interference between airflow rhythm and humidification rhythm. This will provide a precise decision-making basis for subsequent phase misalignment control and evaporation stress mitigation. The specific steps are as follows:

[0056] Within the fermentation space, data acquisition devices are deployed to measure environmental changes. These include a wind speed detector to measure changes in airflow velocity, a spray frequency detector to sense the release pattern of humidified water mist, a pressure sensor to monitor air pressure fluctuations in the enclosed environment, a humidity detector to record changes in regional relative humidity, and a contact surface temperature sensor to capture microscopic temperature fluctuations on the tea leaf surface. These devices are positioned at representative locations within the tea fermentation space and collect data in parallel at a fixed frequency. Each sampling session covers a complete airflow-stop-re-airflow cycle and its corresponding humidification process. To ensure the temporal integrity and spatial representativeness of the data, each data record is labeled with the specific sampling time, device number, and measurement point coordinates, and undergoes unified timestamp alignment to form a precisely synchronized time-series dataset, ensuring temporal consistency and response traceability in subsequent processing steps.

[0057] After obtaining parameter data with a unified time reference, each physical quantity is arranged longitudinally according to the time dimension to construct a cross-parameter time series set. By superimposing and visualizing the wind speed beat variation curve, the intermittent release rhythm of the spray, the compression and expansion process caused by the air pressure switching with the wind flow, the rise and fall trajectory of humidity per unit time, and the slight fluctuation curve of leaf surface temperature response on the coordinate axis, these physical quantities form a spatial-temporal coupled expression under a unified time window. This processing method can reveal the correlation between multiple parameters in the time domain. For example, in certain specific time periods, frequent spray releases coinciding with wind speed switching nodes may form a nonlinear interaction behavior where water vapor is not fully diffused before being pushed away by the wind flow, thus providing explicit evidence for the preliminary identification of beat overlap. In particular, superimposing the real-time response trend of leaf surface temperature on the humidity change path can intuitively reflect the dynamic thermo-humidity response characteristics of the tea surface microenvironment under the overlapping effects of spray and wind flow.

[0058] After constructing the multi-dimensional time series overlay relationship, key time segments are dynamically marked. Using wind speed beats as a benchmark, symmetrical time windows are set before and after each high-low switching point to observe the frequency changes of the spray beats and whether pulsed spraying behavior occurs within the corresponding time period. Further analysis is conducted to determine whether air pressure and humidity fluctuate synchronously within the same time period. If multiple parameters show highly consistent amplitude trends within the same time window, this time period can be preliminarily identified as a potential beat overlap zone. Subsequently, the rise and fall of leaf surface temperature within this time period is combined to identify whether the temperature curve shows drastic fluctuations, thereby verifying whether airflow and water vapor interference form an actual evaporative stress response on the leaf surface. All time segments that satisfy the above patterns are extracted, and the corresponding wind speed and spray release start time points are marked at their boundary points to establish a baseline index for subsequent rhythm reconstruction. During this process, all extracted key data nodes retain their time index and spatial location parameters, forming a beat overlap feature list with a complete spatiotemporal mapping relationship.

[0059] Based on the feature list, a correspondence map between airflow rhythm and humidification rhythm is further constructed, generating a data structure called a heat-humidity cross-plot. This plot encodes parameters such as time interval, response amplitude, duration, and intervention level between each pair of wind speed changes and spray release events, mapping them to specific colors, line types, or graphic symbols, and arranging them sequentially in chronological order to form a visual chart that reflects the dynamic coupling of multiple parameters. By observing this plot, the beat synchronization zone, phase overlap segment, and key nodes that may induce disturbances in the fermentation process can be clearly identified, thus providing a quantitative basis for rhythm regulation. This heat-humidity cross-plot not only has the characteristics of temporal integrity and high variable resolution, but also enables data comparison under different batches and spatial distribution conditions, exhibiting good versatility and scalability. It can serve as a decision-making reference for subsequent beat analysis, phase coordination, and wind-mist misalignment execution.

[0060] S002, based on the rhythm changes of the heat-humidity convergence map, identifies the beat overlap area corresponding to the wind speed change threshold, sets the spray convergence window and the stop spray buffer window in the beat overlap area, and outputs phase pointer data for timing coordination by adjusting the spray residence time and spatial diffusion range.

[0061] To further utilize the multi-parameter coupling variation patterns revealed in the heat-humidity convergence map, a rhythm coordination mechanism needs to be established within the identified key airflow variation zones. This involves finely segmenting and windowing overlapping rhythm areas to generate phase pointer data for timing coordination, providing a highly timely rhythm reference benchmark for subsequent wind-fog misalignment control strategies. The specific steps are as follows:

[0062] In the generated humidity and heat convergence map, using the wind speed beat curve as the positioning benchmark, the slope of its change within a unit time window is retrieved, and boundary points with significant increases or decreases in slope are selected as wind flow change threshold nodes. Equally spaced time segments are constructed between each pair of consecutive threshold nodes, and analyzed against the spray beat curve to identify whether there is a trend of periodic increased release, sudden increase in spray frequency, or increase in spray density within the wind speed change segment. Simultaneously, it is observed whether the humidity curve exhibits high-frequency fluctuations, whether the air pressure curve shows periodic fluctuations, and whether the leaf surface temperature exhibits rapid rises and falls within this segment. If multiple physical quantities show synchronous amplitude changes within the same wind speed change range, this segment can be identified as a typical beat overlap area and highlighted in the map using a color-coded stripe. Through this method, typical overlapping periods at the wind speed change threshold are accurately located, providing accurate temporal and spatial positioning for intervention spray behavior.

[0063] Within each marked overlap zone of the spray cycle, two key control windows are defined based on the coupling pattern between the spray release time series and the airflow change path. The first window is the spray convergence window, with its starting point set one time unit before the maximum upward slope of the wind speed change and its ending point set at the edge of the stable zone before the wind speed reaches steady state. This is used to control the spray release intensity by contracting it before the airflow fully forms the dominant direction, limiting the spatial diffusion range of water vapor to a local area above the blade stack and preventing high-speed movement. The second window is the spray stop buffer window, with its starting point set at the end of the wind speed stabilization period and its ending point one time unit before the wind speed begins to decrease. This is used to briefly pause spray release before the airflow switch begins, allowing residual water vapor in the space sufficient time to diffuse and preventing high-concentration water vapor from being re-concentrated and transferred after the wind direction changes, thus preventing the overlapping effect of pressure fluctuations and humidity peaks. These two windows are not only closely related to the wind speed cycle boundary in terms of time location, but also directly related to the spray diffusion radius and water film formation area in terms of spatial control, ensuring that each spraying action is executed within stable and controllable boundaries.

[0064] After setting the spray convergence window and the spray stop buffer window, key characteristic variables for each overlapping area of ​​each cycle are extracted segment by segment. These include: wind speed change slope, spray release density, humidity fluctuation amplitude, air pressure oscillation frequency, and leaf surface temperature peak-to-valley spacing. These variables are matched one by one, and a time anchor point is established with wind speed as the dominant variable. The changes in other parameters are mapped to the time axis formed by this anchor point, ultimately constructing a time-series reference structure centered on the abrupt change point of the wind speed slope. In this structure, the changing trends of each physical quantity are encoded as specific time nodes and change amplitudes, arranged linearly to form a phase path. This path is the framework for the coordinated performance of wind and fog in the time domain, reflecting the relative phase relationship between airflow and humidification, response delay time, and peak superposition patterns. Furthermore, based on this path, the most representative time positions are extracted as reference anchor points, forming a set of phase pointer data to guide regulatory behavior.

[0065] The generated phase pointer data set is integrated and sorted, and dynamically calibrated based on the dynamic changes in the actual operating sequence during fermentation. The calibration process uses the wind speed change cycle as the main periodic framework, inserting multiple subdivided beat segments within each complete cycle. Each anchor point in the phase pointer data set is mapped to the precise time position within the corresponding beat segment, constructing a multi-resolution temporal alignment relationship. Based on this, corresponding behavioral labels are added to the phase pointer data, such as suggestions for appropriate airflow activation, humidification pause, delayed airflow activation, or pre-spray release. This gives the phase pointers not only time reference attributes but also regulatory action command attributes. Finally, all phase pointer data is embedded into the subsequent control logic's drive flow, enabling the system to automatically identify beat overlap trends during each cycle's execution and dynamically adjust the timing and intensity of spray and airflow activation accordingly. This ensures precise coordination between airflow changes and humidification behavior in the time dimension, mitigating adverse factors such as high humidity disturbance superposition, pressure resonance, and wet film rupture risks from the source, significantly improving the environmental stability and quality consistency of the tea fermentation process.

[0066] S003 adjusts the airflow control logic according to the phase pointer data, delays the airflow start time during the high spray intensity stage, and starts the light wind circulation in advance during the low spray intensity stage, forming a rhythm misalignment mechanism where the spray trough corresponds to the light wind push and the spray peak corresponds to the airflow delay, and generates a safety window list for the coordinated operation of airflow and water vapor.

[0067] After obtaining the phase pointer data for rhythm coordination, the timing of airflow initiation and termination needs to be adjusted based on these timing anchors to maintain a time misalignment between wind speed changes and spray release. This breaks the superposition of disturbance peaks caused by the original rhythm overlap, achieving dynamic coordination between wind and fog, and further constructing a list of stable operating windows under the synergistic effect of airflow and water vapor. The specific steps are as follows:

[0068] Based on the generated phase pointer data, time adjustments are made to all wind speed control behaviors within the current fermentation cycle. Specifically, the inflection point of the wind speed change slope is used as the basic anchor point, and the corresponding spray release intensity information is extracted to calculate the peak spray frequency and duration within the time period of that anchor point. When the phase pointer indicates that the current time period is in a stage of significantly increased spray release intensity, i.e., a high-intensity spray period, the wind flow initiation time is delayed within this stage, postponing the trigger point for wind speed increase to a time unit after the spray intensity reaches its peak. This avoids the wind flow forcibly cutting in at the initial stage of spray diffusion, causing large-scale disturbance of unevenly diffused water vapor. At the same time, when the phase pointer indicates a stage of relatively low spray intensity, i.e., a spray trough period, the wind flow channel is opened in advance, guiding low-speed wind flow slowly through the blade pile surface before the spray significantly intensifies, allowing residual water vapor to be smoothly diluted and guided for dispersion, establishing a flexible flow field. This adjustment of the time difference between delay and advance constitutes a rhythm misalignment mechanism between wind and fog, which is a key link in the subsequent formation of a coordinated control rhythm.

[0069] Based on the timing adjustment of airflow regulation, further fine-tuning of wind speed levels and airflow channel structure is performed to adapt air movement patterns to the water vapor diffusion state at different spray stages. During the peak spray intensity phase, i.e., the period when airflow initiation is delayed, the wind speed is controlled below the minimum initiation threshold, creating a short period of silence to prevent high-energy airflow from disturbing fine mist particles that have not yet settled. When airflow restarts, its wind speed is set in a gradual increase mode, transitioning slowly from low wind speed to the target operating value, making the pressure rise process more gentle and reducing sudden changes in water film stress caused by local pressure differences. During the trough spray intensity phase, i.e., the period when airflow intervenes early, a gentle push airflow method is used to initiate the airflow path, maintaining the wind speed within 30% of the average operating wind speed. Airflow buffer structures are installed on both sides of the blade surface, allowing the airflow to advance along a single-sided axial direction, guiding the air to circulate along the blade surface arc, forming a low-speed vortex exhaust method that does not interfere with the water film structure on the blade surface. This adjustment method enables the intensity and direction of airflow to adapt to changes in the state of water vapor, ensuring that the dynamic behavior between water vapor and airflow is decoupled in physical space.

[0070] After completing the entire process of staggered airflow execution, the performance of each rhythmic coordination segment throughout the fermentation period is evaluated. The time periods with the most stable water vapor response, the weakest air pressure fluctuations, and the slowest leaf surface temperature changes are extracted and recorded as high-stability periods. Each high-stability segment must possess the following characteristics: First, there is a clear time offset between the airflow intervention time and the spray release time; second, the residual humidity decreases synchronously with the change in airflow direction after spraying stops; third, there are no abrupt fluctuations in leaf surface temperature; fourth, there are no periodic fluctuations in air pressure within the local space. These high-stability segments are summarized to generate a list of safe operating time intervals, identifying the time periods when a relative balance is achieved between water vapor diffusion and air circulation under specific airflow behavior. These time periods constitute the safety window, representing the optimal timing for implementing key regulatory behaviors during fermentation.

[0071] Based on the obtained list of safety windows, each safety window is tagged, giving each time period not only static descriptive attributes but also structured features for dynamic control. For windows where airflow is delayed after the spray peak, an avoidance tag is added to indicate that any strong interference that would damage the leaf surface water film is not suitable during this period. For windows where airflow intervenes early during the spray trough, a promotion tag is added to indicate that low-speed ventilation or slight temperature rise operations are appropriate to assist in water vapor dissipation. For multiple consecutive window segments, the time intervals, response trends, and continuity of environmental variables between them are analyzed to construct a time-domain coherence model, so that subsequent control strategies can be dynamically followed according to the window sequence during execution. In this way, the safety window list is no longer a single set of time segments but becomes a structured control basis, enabling all subsequent control actions to be anchored to time nodes with stable rhythm and water vapor balance, improving environmental adaptability and response accuracy during fermentation, and providing a highly reliable operating platform for subsequent evaporation stress slow-release operations and respiratory rhythm index extraction.

[0072] S004, according to the safety window list, perform evaporation stress slow release operation. By implementing early release of mist particles, slow rise control of ambient temperature and unilateral rotation guidance of wind direction, the stress distribution of water film on the leaf surface tends to be more balanced, thereby stabilizing the evaporation rate of the tea surface and extracting the breathing rhythm index that reflects the stability of environmental changes.

[0073] To minimize stress accumulation caused by water vapor disturbance within the stable period specified in the safety window list, proactive stress mitigation operations are required for the leaf evaporation process. This involves precisely controlling mist release, heat conduction, and airflow guidance to achieve dynamic stability of evaporation behavior. Based on this, key time-series parameters reflecting resilience to environmental changes can be extracted and used as feedback for subsequent cycle control. The specific steps are as follows:

[0074] Before entering the calibrated safe operating time window, a pre-treatment operation is performed on the spray release behavior, namely, the early release of microparticle-sized droplets within a time unit before the start or switching of airflow. This droplet release is not intended to provide a large amount of moisture, but rather to form a uniformly distributed, ultra-thin wet film on the leaf surface through low-intensity spraying, allowing for preliminary exchange of moisture between the leaf surface and the airflow before the airflow begins. The key to this operation is controlling the droplet size and spray velocity, ensuring that the moisture diffuses immediately upon contact with the leaf surface to form a monolayer water film, avoiding intercellular spaces and acting only on the outer surface of the cells, serving as a buffer against subsequent airflow impact. During this early release operation, the spray nozzle must maintain a stationary angle, with the spray angle controlled within 30 degrees, and cover the entire leaf surface in a fan-shaped distribution to avoid localized accumulation and ensure that all leaves receive an equal water film coating. This pre-wetting not only buffers the impact of subsequent airflow on the dry surface but also keeps the evaporation rate within a low-amplitude range in the initial stage, preventing violent evaporation caused by a temperature surge.

[0075] After the water film is formed by the spray release, the ambient temperature gradually increases. During this stage, the heat output rate of the heating device within the fermentation space is adjusted to ensure a linear and slow temperature increase per unit time, with the rate of increase remaining no higher than 0.2 degrees Celsius per minute, until the set target fermentation temperature threshold is reached. The core of this gradual temperature increase control lies in reducing the temperature gradient and the temperature difference pressure between the leaf surface and the airflow, thereby suppressing sudden changes in evaporation acceleration caused by a sudden temperature rise in the water film. In practice, the heating element of the temperature control device must maintain a constant power, using an intermittent on / off power-on method to maintain a slow temperature rise curve. A heat diffusion hood is added to the top of the fermentation space to allow the rising heat flow to form a sinking air mass, preventing the temperature rise from directly affecting the top of the leaf pile and creating strong convective disturbances. Simultaneously, within the first two time units when the temperature approaches the target value, further heating is stopped, allowing the system to enter a thermally stable zone. This ensures that the surface moisture of the leaves evaporates naturally without additional thermal shock, thereby reducing the concentration of local compressive stress in the cell membrane caused by sudden changes in surface tension.

[0076] After the temperature gradually rises, a unilateral rotational guidance system is implemented, combined with the airflow path, to create a stable spiral propulsion pattern as the airflow passes through the blade stack. The core operation in this stage is to adjust the originally multi-directional cross-flow into a swirling flow with air entering in one direction and exiting in another, causing the gas to run along a spiral path on the blade stack surface, avoiding local disturbances caused by direct airflow. During implementation, rotatable air guide structures are set on both sides of the blade stack, with guide vanes deflected towards the center at an angle between 20 and 40 degrees, so that the main airflow direction and the blade stack alignment direction form a non-perpendicular angle. When the airflow enters the guiding area, it is affected by the microstructural resistance of the blade stack surface, gradually forming an arc-shaped streamline, rotating and sliding along the blade surface, achieving uniform distribution of wind energy on the contact surface. Since the airflow no longer forms a direct impact, the shear force on the blade surface is significantly reduced, and the evaporation stress also tends to stabilize. Throughout the entire rotational guidance process, it is necessary to maintain wind speed stability, ensuring that it does not fluctuate abruptly throughout the entire safety window period, and to maintain a constant total airflow in conjunction with the top exhaust channel, forming a dynamic equilibrium state.

[0077] After implementing three stress mitigation measures—early release of fog particles, gradual temperature rise, and wind direction guidance—the process enters a stable observation phase of the evaporation process to extract key time-series parameters from the environmental resilience index, namely the breathing rhythm index. This index is constructed by real-time monitoring of subtle fluctuations in leaf surface temperature, the response speed of spatial humidity, and the amplitude of wind speed feedback changes. In the time dimension, it comprehensively reflects the consistency of the frequency, amplitude, and rhythm of changes in various variables within a stable period. Specifically, based on a leaf surface temperature sensor, the temperature fluctuation curve within a unit of time is recorded. This is combined with cross-analysis of the humidity change rate and wind speed fluctuation frequency within the same time period to construct a spectrum graph with time as the horizontal axis and the variable change rate as the vertical axis. In the spectrum graph, if multiple variables exhibit periodic fluctuations within a continuous time period and maintain relative phase consistency, and their fluctuation amplitude does not exceed a set stability threshold, then that time period is considered to possess high resilience characteristics. The fluctuation period, peak-to-valley difference, and phase consistency values ​​are recorded and combined to constitute the breathing rhythm index. This index not only reflects the stable adaptability of the current fermentation environment to disturbances, but can also be used as a feedback criterion for the next stage of dynamic control strategy, ensuring the operational foundation for maintaining stable environmental rhythms, intact leaf structure, and continuous optimization of quality development throughout the entire fermentation process.

[0078] S005, based on the dynamic adjustment of the tea fermentation environment by the respiratory beat index, through micro-periodic beat misalignment guidance, phase beat symmetrical flow diversion, alternating airflow static control and water vapor circulation conduction path construction, so that the airflow and water vapor form an inverse balance in the time dimension, induce the evaporation stress to enter the slow release range, and realize the dynamic steady-state control of the tea fermentation process.

[0079] To achieve dynamic steady-state control of the fermentation environment in the time domain, a control strategy based on time misalignment, phase diversion, and alternating airflow and water vapor needs to be established, building upon the extracted respiratory rhythm index. This strategy drives the airflow and water vapor to form a predictable reverse coupling state in terms of rhythm, thereby actively guiding the evaporation stress into the slow-release range. The specific steps are as follows:

[0080] Based on the extracted respiratory beat index, the entire fermentation cycle is divided into time axes, further refined into multiple micro-cycle beat units. Within each micro-cycle, it is identified whether the current evaporation state exhibits an excessively rapid release trend or a deviation in temperature and humidity fluctuations. When the respiratory beat index shows tightening characteristics—that is, an increase in the frequency of temperature fluctuations, a lag in humidity response, and a decrease in the synchronicity of wind speed response—it can be determined that the evaporation behavior is developing excessively and positively, requiring immediate beat-shifting guidance. In this operation, a fixed time offset is applied to the originally synchronized airflow initiation and humidification behavior, delaying the airflow initiation time after the humidification initiation time and advancing the airflow termination time before the humidification termination time, thereby breaking the linear overlap between wind and mist in the time dimension. Through this guidance method, the spatial distribution of water vapor no longer overlaps with the high-kinetic-energy period of airflow, mitigating its accumulation effect in local areas and avoiding pressure disturbances on the leaf surface. Simultaneously, in the delayed airflow stage, a stepped wind speed gradual increase curve is set to extend the airflow transition period, forming a buffer zone in time and establishing a wider stress release window. This type of micro-cycle timing misalignment operation dynamically adjusts according to the changing frequency of the respiratory timing index, achieving proactive peak-shifting control of the evaporation process.

[0081] Based on the control of wind and fog rhythm misalignment, phase diversion operation is performed according to the phase trend shown by the breathing rhythm index. In the rhythm sequence of different parameters, if the humidity peak is found to be delayed after the wind speed trough, while the temperature peak is found to be earlier than the humidity response, it indicates that the rhythm of the three is unbalanced, and phase diversion is required to achieve synchronous adjustment. The core of phase diversion operation is to re-organize the interaction path between the three main control lines of airflow, humidity, and temperature, and redistribute the response points of the three to a coordinated time frame. To this end, multiple airflow deflection points are introduced in the airflow path, so that the main airflow no longer advances linearly along a fixed channel, but forms multiple diversions when passing through the blade stack, with some airflow detouring to the bottom of the blade stack, forming a slow upward return channel; at the same time, a slight delay triggering mechanism is introduced at the humidity source, so that the humidification behavior is slightly delayed after the initial temperature rise, so that the evaporation peak and the airflow peak remain parallel in time, rather than overlapping. By using this type of diversion method, multiple rhythm variables no longer compete for the same time window, but instead form a staggered and complementary pattern, reducing the intensity of the combined effects of temperature difference, wind force and water pressure on the leaf surface at the same time.

[0082] After phase coordination is completed, an alternating airflow static control phase is introduced to simulate a natural intermittent breathing pattern, further buffering the impact of rapid environmental variable switching on the tea structure. In this operation, the fermentation cycle is divided into an alternating operation phase and a static transition phase. During the operation phase, the coordinated behavior of normal airflow and continuous moisture release is maintained. During the static phase, both the airflow and humidification devices are simultaneously shut off, causing the leaf pile space to be in a short-term closed state, entering a buffer recovery state. The duration of this phase is determined based on the lower limit of the stable period in the respiratory rhythm index, typically set within two time units. In the static state, since the airflow no longer interferes with the water film structure, residual moisture can diffuse naturally along the leaf surface, and the temperature tends to moderate due to the lack of external heat input. At this time, the evaporative stress of the leaf cells shifts from a high-fluctuation state to a slow-release range, which is beneficial for the self-regulation of cell membrane tension. In each round of alternating operation, the start and end points of airflow and moisture must be precisely controlled by a time control device to ensure that the switching process is non-overlapping and non-delayed, ensuring that a complete simulated breathing state is formed within each cycle.

[0083] Through the combined effects of staggered timing, phase guidance, and alternating static operation, a comprehensive water vapor circulation pathway network is constructed, enabling balanced guidance of air and water vapor in both the temporal and spatial domains throughout the fermentation environment. In the specific construction process, the fermentation space is divided into multiple vertically layered ventilation units. Each unit has independent air inlets and outlets, and a permeable grid structure is designed at the intersection of the water vapor release area and the airflow channel, allowing water vapor to enter the transfer zone for natural diffusion before encountering the wind. As the airflow passes through each layer, its direction changes from horizontal advancement to vertical penetration, decreasing layer by layer from top to bottom, ultimately completing a closed loop at the bottom. Because the water vapor is evenly distributed in the grid diffusion area, it no longer accumulates during its guidance into each layer of the leaf stack with the airflow, but instead continuously permeates the entire fermentation space at a linear velocity, avoiding the formation of regional high-pressure or high-humidity retention zones. In this way, the heat conduction of air, the mass transfer of water vapor, and the evaporation of water film on the leaf surface form a closed-loop flow network in the entire spatial structure, truly realizing the time misalignment, path guidance, and intensity balance between wind and fog. This keeps the evaporation stress within the slow-release range, preventing sudden changes or disturbances, and ensuring that the tea fermentation process maintains a steady-state structure amidst dynamic changes, providing the optimal environmental foundation for quality development.

[0084] This invention achieves precise positioning of the rhythm overlap zone by constructing a multi-parameter synchronous acquisition mechanism and a humidity-heat convergence map. It also guides the wind and fog behavior rhythm misalignment through phase pointer data, ensuring dynamic avoidance between wind and fog over time, thus effectively preventing pressure resonance and sudden changes in leaf stress caused by localized humidity-heat superposition. Simultaneously, by implementing combined slow-release measures such as fog particle pre-positioning, gradual temperature rise, and wind direction guidance within a safe window, the evaporation behavior of the leaf water film tends to stabilize. Furthermore, a real-time closed-loop control strategy based on the respiratory rhythm index enables the fermentation environment to have adaptive adjustment and disturbance suppression capabilities, ultimately achieving steady-state operation of the tea fermentation process under dynamic changes.

[0085] This invention provides, for example Figure 2 The tea fermentation intelligent control system shown includes a data acquisition module, a cycle analysis module, a collaborative control module, a stress relief module, and a steady-state control module.

[0086] The data acquisition module establishes a synchronous acquisition process, continuously records wind speed beat, spray beat, air pressure, humidity and leaf surface temperature, and generates a heat and humidity convergence map in chronological order to identify the overlap between airflow and humidification beat.

[0087] The beat analysis module identifies beat overlap areas based on the wet-heat convergence map, sets spray convergence windows and stop spray buffer windows in the beat overlap areas, and generates phase pointer data for timing coordination by adjusting the spray residence time and diffusion range.

[0088] The coordinated control module adjusts the airflow control based on the phase pointer data, delaying the start of airflow during the peak spraying period and applying light wind in advance during the low spraying period, forming a rhythm misalignment mechanism and generating a list of safe windows for coordinated operation of airflow and water vapor.

[0089] The stress relief module performs evaporative stress relief operations according to the safety window list. It stabilizes the stress on the leaf surface water film and extracts the respiratory rhythm index by releasing mist particles in advance, controlling the temperature rise slowly and guiding the wind direction to one side.

[0090] The steady-state control module dynamically adjusts the fermentation environment based on the respiratory beat index. Through beat misalignment guidance, phase diversion, and the construction of alternating static wind and water vapor circulation paths, the airflow and water vapor form an inverse balance in the time dimension, guiding the evaporation stress into the slow release range and completing dynamic steady-state control.

[0091] The present invention provides a method for intelligent control of tea fermentation, which is implemented by the above-mentioned intelligent control system for tea fermentation. For details of the specific method and process of the intelligent control system for tea fermentation, please refer to the embodiment of the above-mentioned method for intelligent control of tea fermentation, which will not be repeated here.

[0092] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A tea leaf fermentation intelligent regulation method, characterized in that, Includes the following steps: S001, Establish a synchronous acquisition process to continuously record wind speed beat, spray beat, air pressure, humidity and leaf surface temperature, and generate a heat-humidity convergence map in chronological order to identify the overlap between airflow and humidification beat; S002, based on the wet and heat convergence map, identify the beat overlap area, set the spray convergence window and the stop spray buffer window in the beat overlap area, and generate phase pointer data for timing coordination by adjusting the spray residence time and diffusion range; Step S002 includes: In the heat-humidity convergence map, the wind speed beat curve is used as the positioning reference to identify the boundary points of the wind speed change slope and combine the spray beat, humidity, air pressure and leaf surface temperature change trends to determine the beat overlap area. Within the overlapping area of ​​the spray beats, spray convergence windows and spray stop buffer windows are set. By adjusting the spray dwell time and spatial diffusion range, the time period and range of spray release behavior are limited. Based on the slope of wind speed change, time anchors are constructed, and spray release density, humidity fluctuation amplitude, air pressure oscillation frequency and leaf surface temperature peak-valley spacing are mapped to the time axis to form a time series reference structure and extract phase paths. The anchor points in the phase path are integrated and sorted, dynamically calibrated, and behavior labels are added to form a phase pointer data set for driving the control logic. S003 adjusts airflow control based on phase pointer data, delays airflow initiation during peak spraying periods, and applies light winds earlier during low spraying periods, forming a rhythm misalignment mechanism and generating a list of safe windows for coordinated operation of airflow and water vapor. S004, according to the safety window list, perform evaporative stress release operation, stabilize the force on the leaf water film and extract the respiratory beat index by releasing mist particles in advance, controlling the temperature rise slowly and guiding the wind direction to one side of rotation. Step S004 includes: Before the safety window begins, release microparticle-sized droplets in advance and form a uniform wet film on the leaf pile surface through low-intensity spraying, so that water vapor and the leaf surface can complete the initial exchange before the wind starts. After the wet film is formed, the ambient temperature is gradually increased. By adjusting the heat output rate of the heating device, the temperature rises slowly and enters the thermal stability zone before the target threshold, so as to reduce sudden changes in the leaf surface evaporation rate. After the temperature rises slowly, the wind direction is rotated to one side to guide the wind. The wind guide structure forms a spiral streamline, allowing the airflow to slide along the blade surface to distribute wind energy evenly and stabilize evaporation stress. After wind direction guidance is completed, real-time monitoring of leaf surface temperature, humidity and wind speed changes is performed. Data on the consistency of temperature fluctuation cycle, humidity response speed and wind speed frequency are extracted to generate a breathing rhythm index that reflects environmental stability. S005 dynamically adjusts the fermentation environment based on the respiratory beat index. Through beat misalignment guidance, phase diversion, and alternating static wind and water vapor circulation paths, the airflow and water vapor form an inverse balance in the time dimension, guiding the evaporation stress into the slow release range and completing dynamic steady-state control. Step S005 includes: Based on the respiratory beat index, the fermentation cycle is divided into multiple micro-cycle beat units. By adjusting the time offset of airflow initiation and humidification behavior, beat misalignment guidance is performed to avoid wind and fog overlapping in the same time window. Based on the phase trend between rhythm variables, a deflection channel is introduced into the airflow path and the humidification triggering time is delayed to perform phase diversion operation, so that the rhythm of airflow, temperature and humidity maintains a staggered and complementary relationship in the time dimension. Based on rhythm coordination, alternating airflow static control is implemented, dividing the fermentation cycle into an operating phase and a static phase. During the static phase, airflow and humidification are simultaneously shut off to achieve a transition from evaporation stress to a slow-release range. By constructing a water vapor circulation pathway network consisting of vertical ventilation units and grid diffusion zones, a closed-loop guidance of air heat conduction, water vapor diffusion, and leaf surface evaporation behavior is achieved, forming a dynamic steady-state control mechanism for evaporation stress.

2. The intelligent control method for tea fermentation according to claim 1, characterized in that, Step S001 includes: By deploying wind speed detection devices, spray frequency detection devices, air pressure sensing devices, humidity detection devices, and contact surface temperature sensing devices in the fermentation space, wind speed beat, spray beat, air pressure, humidity, and leaf surface temperature are collected synchronously at a fixed frequency. The collected data is labeled with sampling time, device number and measurement point coordinates, and aligned with a unified timestamp to form a synchronized time series basic dataset; The time series of each parameter are arranged vertically and superimposed to construct a heat-humidity convergence diagram, which is used to identify the rhythmic overlap relationship between airflow changes and humidification rhythm. Based on the high and low switching points of wind speed beats, a time window is set, and the changing trends of spray beat, air pressure, humidity and leaf surface temperature are combined to extract the beat overlap area and generate a beat overlap feature list.

3. The intelligent control method for tea fermentation according to claim 1, characterized in that, Each anchor point in the phase pointer data set corresponds to a beat segment within a wind speed change cycle, and is attached with behavioral labels for controlling spray initiation, spray pause, delayed airflow initiation, and early airflow initiation, in order to achieve precise timing coordination between spray and airflow.

4. The intelligent control method for tea fermentation according to claim 1, characterized in that, Step S003 includes: Based on phase pointer data, the start time of the airflow during the high spray intensity phase is delayed, and the start time of the airflow during the low spray intensity phase is advanced, thus creating a temporal misalignment between wind and fog. Based on the timing of the staggered execution, adjust the wind speed level and wind flow path structure, set a wind speed gradual increase curve in the delayed stage of wind flow, set a gentle push wind speed in the advanced stage of wind flow and guide the wind direction to bypass one side. Based on the time misalignment and parameter response between airflow and spray, a list of high-stability time periods with stable water vapor diffusion, weak air pressure fluctuations, and stable leaf surface temperature is extracted to form a safety window list. The safety window list is tagged with avoidance and advancement tags to construct a time-series control basis for subsequent dynamic adjustment.

5. The intelligent control method for tea fermentation according to claim 4, characterized in that, The safety window list is generated based on the time offset between wind speed and spray release, and simultaneously meets the comprehensive judgment conditions of a steady decrease in humidity, continuous changes in air pressure, and mild fluctuations in leaf temperature.

6. The intelligent control method for tea fermentation according to claim 1, characterized in that, The respiratory beat index is extracted based on the periodic fluctuations of leaf surface temperature, ambient humidity and wind speed over a continuous period of time, and the phase consistency and fluctuation amplitude stability of the three are used as the judgment criteria.

7. A tea fermentation intelligent control system, used to implement the tea fermentation intelligent control method according to any one of claims 1-6, characterized in that, It includes a data acquisition module, a cycle time analysis module, a coordinated control module, a stress relief module, and a steady-state control module. The data acquisition module establishes a synchronous acquisition process, continuously records wind speed beat, spray beat, air pressure, humidity and leaf surface temperature, and generates a heat and humidity convergence map in chronological order to identify the overlap between airflow and humidification beat. The beat analysis module identifies beat overlap areas based on the wet-heat convergence map, sets spray convergence windows and stop spray buffer windows in the beat overlap areas, and generates phase pointer data for timing coordination by adjusting the spray residence time and diffusion range. The coordinated control module adjusts the airflow control based on the phase pointer data, delaying the start of airflow during the peak spraying period and applying light wind in advance during the low spraying period, forming a rhythm misalignment mechanism and generating a list of safe windows for coordinated operation of airflow and water vapor. The stress relief module performs evaporative stress relief operations according to the safety window list. It stabilizes the stress on the leaf surface water film and extracts the respiratory rhythm index by releasing mist particles in advance, controlling the temperature rise slowly and guiding the wind direction to one side. The steady-state control module dynamically adjusts the fermentation environment based on the respiratory beat index. Through beat misalignment guidance, phase diversion, and the construction of alternating static wind and water vapor circulation paths, the airflow and water vapor form an inverse balance in the time dimension, guiding the evaporation stress into the slow release range and completing dynamic steady-state control.

Images