A tea processing equipment adaptive temperature control system based on fuzzy PID

Abstract

The application discloses a tea processing equipment adaptive temperature control system based on fuzzy PID, and relates to the technical field of tea processing equipment temperature control, comprising a data synchronization module, a thermal response detection module, a thermal response determination module, an adaptive control module and an execution safety module. The system is associated with process temperature targets, temperature, exhaust, equipment state and actuator feedback, and forms two thermal command segments in opposite directions around the same baseline when detection is allowed. The response is window-aligned and local baseline-eliminated to form a control state. The fuzzy PID parameters are adjusted according to the control state, and the write, keep and remove paths of conventional integral memory or evaporation compensation memory are selected. The candidate thermal command is limited in amplitude, speed and interlocked verification, and then the thermal actuator is driven. The system is used for temperature control in the processes of tea fixation, drying or aroma enhancement, and helps to reduce the risk of continued heating after the temporary exhaust load exits and the temperature exceeding the process allowed temperature band.

Description

Technical Field

[0001] This invention relates to the field of temperature control technology for tea processing equipment, and specifically to an adaptive temperature control system for tea processing equipment based on fuzzy PID. Background Technology

[0002] The processing of tea, including withering, drying, and aroma enhancement, typically involves regulating the temperature of the processing zone using thermal actuators such as electric heating power regulators, fuel supply regulators, or steam valves. Temperature control adjusts the heat input based on the deviation between the target process temperature and the controlled temperature. Fuzzy PID control can adjust the proportional, integral, and derivative actions according to the deviation and its changing trend. During processing, the continuous dehumidification of the tea leaves creates a transient endothermic load; simultaneously, changes in feeding, turning, fan operation, door status, and actuator status also alter the correlation between temperature response and dehumidification measurement.

[0003] If control parameters are adjusted solely based on temperature deviations or single temperature and humidity changes, it's difficult to distinguish between delayed temperature response caused by tea leaf dehumidification and changes in heat source capacity, abnormal heat transfer, or actuator malfunction. Transient deviations may also be continuously written into the conventional integral memory. Even after the dehumidification load decreases, the accumulated heating effect may continue to drive the thermal actuator, potentially causing the temperature to exceed the process's allowable temperature range and requiring repeated corrections. Furthermore, historical response relationships lack comparability when process conditions change, the response window is incomplete, or actuator feedback is abnormal. Therefore, a reliable control basis for the current thermal process needs to be established when temperature, humidity, equipment status, and actuator feedback are all effective and safety conditions permit. Summary of the Invention

[0004] The technical problem to be solved by this invention is that when the dehumidification evaporation load, slow temperature trend and actuator feedback conditions coexist in tea processing, conventional fuzzy PID has difficulty distinguishing between normal steady-state deviation, dehumidification transient heat absorption and the failure to execute heat command, which may cause the compensation formed during dehumidification to enter the conventional integral memory and continue to act after the load is removed.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0006] The data synchronization module is used to associate the process temperature target, temperature, humidity, equipment status and actuator feedback at the sampling time to form a synchronized measurement frame and output the data validity.

[0007] The thermal response detection module is used to determine whether detection is allowed based on the synchronous measurement frame and data validity. When detection is allowed, two thermal command segments with opposite directions are formed around the same baseline and confirmed by actuator feedback. Between the two segments, the actuator feedback and temperature trend meet the baseline recovery conditions. In case of an anomaly, the command segment is terminated and the baseline thermal command is restored through the execution safety module.

[0008] The thermal response determination module is used to perform window alignment and local baseline elimination on the temperature response and dehumidification response of the two command segments, separate the reversible thermal response quantity and the common drift quantity, and combine the dehumidification continuity and actuator feedback to form one of the following states: normal control state, evaporative load state, load transfer state or invalid control state.

[0009] The adaptive control module is used to adjust the fuzzy PID parameters according to the control state, select the write, hold and remove paths of conventional integral memory or evaporation compensation memory, and synthesize candidate thermal commands.

[0010] The execution safety module is used to limit the amplitude, speed, and interlock of candidate hot commands, drive the hot actuator, and provide feedback on the actual execution results.

[0011] Preferably, the process temperature target, temperature, dehumidification, equipment status, and actuator feedback are correlated at the sampling time to form a synchronous measurement frame and output the data validity, including:

[0012] The comparable operating condition range is defined as the continuous operating period during which the process temperature target, equipment operation mode, feeding status, air path status, and heat transfer connection are kept comparable.

[0013] Within the comparable operating condition range, continuous and valid samples excluding door opening, feeding start and stop, fan switching, or thermal actuator malfunction are used to form a local temperature baseline trend, a temperature trend natural fluctuation envelope, a dehumidification natural fluctuation envelope, a baseline thermal command, and an actuator feedback envelope.

[0014] Preferably, the comparable operating condition range is defined as a continuous operating period during which the process temperature target, equipment operating mode, feeding status, airflow status, and heat transfer connection remain comparable, including:

[0015] When the thermal response relationship becomes incomparable due to changes in process temperature target switching, door operation, sudden changes in feeding or turning, fan switching, or control mode switching, the current comparable operating condition range ends.

[0016] The reference data formed by the completed detection records and used for subsequent thermal response comparison, as well as the incomplete detection records and response results not yet used for control state determination, are invalidated within the comparable operating condition range. Evaporation compensation memory migration is prohibited, and a new comparable operating condition range is established after the equipment state is re-stabilized.

[0017] Preferably, when detection is permitted, two thermal command segments with opposite directions and confirmed by actuator feedback are formed around the same baseline, and the actuator feedback and temperature trend between the two segments satisfy the baseline recovery condition, including:

[0018] Detection is only permitted when the process temperature target, temperature trend, feeding, turning, fan, door and measurement signals are stable and effective, the thermal actuator has distinguishable upper and lower adjustment margins, the actuator feedback is normal and in an automatic control mode without alarms or hard interlock actions.

[0019] Detection is not performed when the detection amplitude exceeds the minimum resolvable change and feedback noise boundary of the thermal actuator, and is limited by temperature safety margin, actuator margin and command change rate. Detection is not performed when there is no detection amplitude that simultaneously satisfies safety and resolvability.

[0020] Preferably, when detection is permitted, two thermal command segments with opposite directions and confirmed by actuator feedback are formed around the same baseline, and the actuator feedback and temperature trend between the two segments satisfy the baseline recovery condition, including:

[0021] The actual feedback amplitudes of the actuators relative to the baseline in the two thermal command segments correspond to each other within the allowable deviation range determined by the minimum resolvable change of the thermal actuator and the feedback noise, and the durations of the two command segments are the same.

[0022] First, execute one command segment. After the actuator feedback recovers to the stable feedback range before detection, and the deviation of the temperature trend from the local temperature trend before detection recovers to the natural fluctuation range determined by the stable effective temperature sample and the temperature measurement uncertainty, then execute another command segment.

[0023] If any command segment touches the output boundary, the controlled temperature is predicted to reach or exceed the temperature safety boundary according to the current temperature change trend, the actuator feedback error or timeout occurs, the equipment status changes or interlocking action is performed, the current command segment is terminated and the baseline thermal command is restored, and incomplete detection records are discarded.

[0024] Preferably, window alignment and local baseline elimination are performed on the temperature response and dehumidification response of the two command segments to separate the reversible thermal response quantity and the common drift quantity, including:

[0025] Align the two temperature response windows with the moment when the actual actuator feedback begins to change, and after thermal hysteresis, use the same temperature change rate fitting rule to obtain the temperature response increment of each relative local baseline trend.

[0026] The difference between the two temperature response increments is normalized according to the actual heat input change amplitude confirmed by the actuator feedback to form a reversible thermal response quantity, and the temperature trend retained by both is used to form a common drift quantity;

[0027] When the window is incomplete, the actual feedback amplitude does not correspond, the sample is insufficient, or the data does not belong to the same comparable operating condition range, the reversible thermal response quantity and the common drift quantity are not updated.

[0028] Preferably, the combination of continuous dehumidification and actuator feedback forms one of the following states: normal control state, evaporative load state, load transfer state, or ineffective control state:

[0029] An effective response envelope is formed and updated using complete detection records showing normal actuator feedback, no detected dehumidification, and stable operating conditions. This envelope is used to characterize the allowable range of change in reversible thermal response under normal control conditions. The effective response envelope is frozen under evaporative load and load transfer conditions.

[0030] When the current reversible thermal response is below the lower boundary of the effective response envelope, the two command segments of the dehumidification span continuously deviate in the same direction, and when the controlled temperature is below the process temperature target, the common temperature change trend of the two command windows deviates relative to the local temperature trend before detection in the direction of decreasing the heating rate or increasing the cooling rate. When this deviation exceeds the natural fluctuation range jointly determined by the natural fluctuation of the temperature trend of the stable effective temperature sample and the temperature measurement uncertainty, an evaporation load state is formed. When the necessary input is invalid or the actuator feedback is abnormal, an invalid control state is preferentially formed.

[0031] Preferably, the fuzzy PID parameters are adjusted according to the control state, and the write, hold, and delete paths for conventional integral memory or evaporation compensation memory are selected, including:

[0032] Under normal control conditions, update the conventional integral memory and keep the evaporation compensation memory at zero;

[0033] Under evaporative load conditions, the integral increment that is intended to be written into the conventional integral memory and will increase the heat input will be redirected to the evaporative compensation memory.

[0034] In load transfer mode, stop adding new evaporation compensation and gradually remove existing compensation. In ineffective control mode, freeze the increment of control parameters and the addition of two types of memory, and switch to restricted basic fuzzy PID control or safety handling.

[0035] Preferably, the write, hold, and erase paths for conventional integral memory or evaporation-compensated memory are selected, including:

[0036] Write only when the temperature deviation has not converged to the target, the thermal actuator has a safety margin, the controlled temperature predicted according to the current temperature change trend remains within the process-allowed temperature range, and the reversible thermal response is valid;

[0037] The upper limit of evaporation compensation memory is determined by the temperature safety margin, the actuator output margin, the minimum effective value of the reversible thermal response determined by the effective detection record under normal control conditions, and the anti-saturation condition.

[0038] When the dehumidification decreases, the reversible thermal response recovers, and the temperature rises, the compensation is removed according to the minimum resolvable reduction of the thermal actuator until the compensation returns to zero, the evaporative load state is re-established, or the system enters an ineffective control state.

[0039] Preferably, the process includes limiting, speeding, and interlocking candidate thermal commands, driving the thermal actuator, and providing feedback on the actual execution results, including:

[0040] When the output is saturated, the saturation direction is fed back to prevent both types of memory from increasing along that direction.

[0041] When the actuator reports an error, timeout, or interface interruption, the current detection record and corresponding control status become invalid, and a safety power command, a heat stop command, or a manual takeover request preset by the equipment safety configuration is sent according to the temperature and interlock status.

[0042] After the fault is eliminated and the equipment status is stable, a new comparable operating condition range is established, a new baseline is formed, and a set of effective opposite thermal command segments are completed before adaptive control is resumed without loading the evaporation compensation memory of the previous comparable operating condition range.

[0043] The above technical solution can achieve the following beneficial effects:

[0044] When process conditions and safety requirements permit, this invention obtains comparable responses using opposite thermal command segments confirmed by actuator feedback under the same baseline. Through window alignment and local baseline elimination, it distinguishes the reversible thermal response that changes with command reversal from the temperature drift retained by both windows. Dehumidification continuity, equipment status, and actuator feedback further participate in control state determination, enabling the system to identify evaporative load within the current comparable operating range and reducing the likelihood of mistaking actuator malfunctions, airflow changes, or isolated dehumidification fluctuations for evaporative load. This provides a more targeted control basis for adjusting fuzzy PID parameters.

[0045] After confirming the evaporation load, the system writes the temporary integral increment that increases heat input into a separate evaporation compensation memory, without mixing it into the regular integral memory. When dehumidification decreases, the reversible thermal response recovers, and the temperature rises, the system stops adding new compensation and gradually removes it. This ensures that the heating effect corresponding to the transient dehumidification load exits as the load is removed, reducing the risk of continued heating, temperatures exceeding the process's allowable temperature range, and repeated corrections. Candidate heat commands also undergo amplitude limiting, rate limiting, and interlock verification. In the event of measurement, execution, or interlock anomalies, new commands are frozen and transferred to restricted basic control or safety procedures, thereby preventing the continued accumulation of compensation under invalid conditions. Attached Figure Description

[0046] Figure 1 This is a schematic diagram of the system structure of the present invention.

[0047] Figure 2 This is a schematic diagram of the thermal response detection process of the present invention.

[0048] Figure 3 This is a schematic diagram illustrating the relationship between the control state and the memory path in this invention. Detailed Implementation

[0049] This embodiment provides an adaptive temperature control system for tea processing equipment based on fuzzy PID. This system is suitable for tea processing equipment that controls processing temperature through adjustable heat sources during processes such as fixation, drying, and aroma enhancement. It can be deployed on the equipment's industrial controller, edge controller, or other computing platforms with real-time data acquisition and control capabilities. The tea processing equipment is equipped with thermal actuators to provide heat to the processing area, temperature sensors to reflect the heating state of the processing area, a dehumidification measurement device to reflect the dehumidification process of the tea leaves, and over-temperature, electrical, combustion, and mechanical safety interlocks operating independently of the system. The thermal actuators can be electric heating power regulators, fuel supply regulators, steam valves, or other actuators capable of increasing or decreasing heat input around the current operating point.

[0050] The system includes a data synchronization module, a thermal response detection module, a thermal response determination module, an adaptive control module, and an execution safety module. The data synchronization module connects to the process recipe interface, temperature measurement interface, dehumidification measurement interface, equipment status interface, and thermal actuator feedback interface, respectively, to obtain the process temperature target, controlled temperature, dehumidification signal, equipment operating status, thermal commands, and actuator feedback for the current processing stage. The thermal response detection module connects to the data synchronization module and the execution safety module, and is used to generate two thermal command segments with opposite directions when process and safety conditions permit. The thermal response determination module receives the temperature response, dehumidification response, and actuator feedback corresponding to the two thermal command segments to form the control state of the current thermal process. The adaptive control module adjusts the fuzzy PID parameters according to the control state and selects conventional integral memory or evaporation compensation memory as the writing path for temperature deviation. The execution safety module performs amplitude limiting, speed limiting, and interlock verification on the synthesized candidate thermal commands, drives the actual thermal actuator, and feeds back the actual execution results to the next control cycle.

[0051] Please see Figure 1 The adaptive temperature control system for tea processing equipment in this embodiment includes a data synchronization module, a thermal response detection module, a thermal response determination module, an adaptive control module, and an execution safety module. Each module transmits measurement information, control status, thermal commands, and execution results according to the relationship shown in the diagram.

[0052] The process formulation interface provides the data synchronization module with the current processing stage's process temperature target, permissible temperature range, temperature safety boundary, stage identifier, and configuration version. The process temperature target is provided by the existing process configuration of the tea processing equipment; this system does not automatically generate processing formulas based on tea varieties, nor does it presuppose a fixed temperature for a specific tea. The data synchronization module verifies the unit, version, and value range of the process temperature target and confirms that it does not exceed the equipment's rated temperature control capability. If the target version changes or the current processing stage ends, the system stops using previously obtained thermal response relationships to avoid treating data under different temperature targets as the same thermal process.

[0053] Temperature sensors are positioned to reflect the heating environment near the processing area or tea materials. The temperature measurement interface provides at least the measured value, sampling time, valid status, and calibration status. When the equipment uses multiple temperature measurement points, the data synchronization module can generate a controlled temperature based on the physical correspondence between the measurement points and the controlled area, while retaining the differences between the measurement points to determine if the temperature distribution is abnormal. When the differences between multiple measurement points exceed the allowable envelope determined by the equipment based on the measurement point accuracy, spatial location, and safety requirements, the system stops active thermal response detection to avoid masking localized overheating or uneven heating of the material with a simple average value.

[0054] The dehumidification measurement device is preferably a humidity sensor installed in the exhaust duct. Its measurement location should reflect changes in the humidity of the exhaust airflow and avoid direct adhesion of condensate water causing continuous distortion. For equipment using condensation dehumidification, changes in condensate drainage within the control window can also be used as the equivalent dehumidification amount. Regardless of the dehumidification measurement method used, the data synchronization module simultaneously reads the fan or airflow path status. When changes occur in fan speed, damper position, or dehumidification path that alter the measurement correspondence, the original dehumidification baseline and any incomplete response windows become invalid, preventing misinterpretation of airflow path changes as changes in tea evaporation load.

[0055] The equipment status interface provides the data synchronization module with information on feeding, turning, fan, door, automatic or manual control modes, equipment alarms, and safety interlock status. In each control cycle, the data synchronization module correlates the process temperature target, controlled temperature, dehumidification signal, current thermal command, actuator feedback, and equipment status according to the sampling time, forming a synchronized measurement frame. The communication bus or data encapsulation method used by the interface is not limited, as long as the source, timing, validity, and corresponding control cycle of each signal can be uniquely identified. Duplicate data, out-of-order data, or data re-transmitted after an interruption that cannot be uniquely correlated are not appended to the thermal response window.

[0056] In this embodiment, a comparable operating period is defined as a continuous operating time during which the process temperature target, equipment operating mode, feeding status, airflow status, and key heat transfer connections remain comparable. A new comparable operating period is established upon system startup or a change in the process temperature target, and the effective response envelope of this comparable operating period is marked as not established. Door operation, sudden changes in feeding method, changes in tumbling mode, fan status switching, automatic / manual mode switching, and other events sufficient to alter the thermal response relationship all terminate the current comparable operating period. When a comparable operating period ends, the effective response envelope of that period becomes invalid, incomplete thermal response detection records, and response results not yet used to form a control state or update the effective response envelope are discarded. Evaporation compensation memory is not migrated to the new comparable operating period; conventional integral memory is clamped or reinitialized according to the equipment's existing non-disruptive switching rules.

[0057] Within the same comparable operating range, the data synchronization module checks the validity of the temperature and humidity signals. For isolated spikes caused by sensor noise, amplitude limiting, median processing, or equivalent conventional denoising methods can be used based on the sensor calibration uncertainty. However, the processing used must not change the true direction of temperature or humidity rise or fall, nor should it produce trends not present in the original signal. The phase delay caused by filtering should be included in the response window alignment process; if the filtering delay cannot be determined, the signal should not be used to form active detection evidence.

[0058] When the process temperature target and equipment status are stable and no active detection is implemented, the data synchronization module forms a local temperature baseline trend, a natural temperature trend fluctuation envelope, a natural humidity fluctuation envelope, a current baseline thermal command envelope, and an actuator feedback envelope based on continuous effective samples. The baseline window must not cross different comparable operating condition intervals, nor include door opening, feeding start / stop, fan switching, or thermal actuator malfunction. The natural temperature trend fluctuation envelope is formed from continuous effective temperature samples within the current comparable operating condition interval during which no active detection is implemented. The data synchronization module divides the continuous effective temperature samples into multiple continuous baseline sub-windows. Each baseline sub-window uses the same temperature change rate fitting rule as the thermal response window to obtain its corresponding temperature change rate. The average temperature change rate of multiple baseline sub-windows is used as the local baseline reference change rate. The deviation of the temperature change rate of each baseline sub-window relative to the local baseline reference change rate is calculated to obtain the temperature trend deviation. Based on the normal maximum and minimum values ​​of the temperature trend deviation, the temperature change rate margin obtained by converting the temperature sensor measurement uncertainty is extended to both sides to form the natural temperature trend fluctuation envelope. The natural humidity fluctuation envelope is determined by the stable humidity fluctuation in the current comparable operating range and the calibration uncertainty of the humidity measurement device. Therefore, there is no need to pre-write a fixed humidity threshold for a specific tea variety.

[0059] The adaptive control module first runs basic fuzzy PID control using the initial safety parameters obtained from equipment commissioning. The basic fuzzy PID takes the deviation between the target process temperature and the current effective temperature, as well as the trend of this deviation, as input. Based on pre-configured fuzzy rules, it determines the adjustment direction and magnitude of the proportional, integral, and derivative parameters. The fuzzy input range is determined based on the process's permissible temperature band, temperature measurement uncertainty, and the equipment's permissible temperature change trend. The parameter increments obtained through fuzzy inference must not exceed the parameter boundaries given by equipment commissioning or safety configuration.

[0060] In one specific implementation, the basic fuzzy PID is based on temperature deviation. and the rate of change of temperature deviation As a fuzzy control input. Let the first... The process temperature target for each control cycle is The controlled temperature is The control cycle length is ,but:

[0061]

[0062]

[0063] Based on the process-allowed temperature range and the equipment-allowed temperature change rate, respectively... and Normalization is performed to ensure that the set falls within a predefined fuzzy universe. The fuzzy universe is divided into seven overlapping fuzzy sets: negative large, negative medium, negative small, zero, positive small, positive medium, and positive large. The middle fuzzy set uses a triangular membership function, while the two fuzzy sets use a trapezoidal membership function.

[0064] Fuzzy inference employs a Mamdani-type max-min inference method, determining the increment of the proportional parameter based on the temperature deviation and its changing trend. Increment of integral parameters and differential parameter increment The fuzzy output is obtained, and the centroid method is used for defuzzification.

[0065] The fuzzy rules are established according to the following control principles: when the temperature deviation is large and continues to move away from the target, the proportional correction action is increased and the integral action is limited; when the temperature deviation is large but is approaching the target, the proportional correction action is maintained and the incremental integral action is reduced; when the temperature deviation is small but begins to move away from the target, the proportional and integral actions are increased; when the temperature deviation is small and continues to exist in the same direction, the integral action used to eliminate steady-state error is increased under normal control conditions; when the temperature has crossed the target and continues to deviate, the proportional and integral actions are reduced and the derivative action to suppress temperature changes is enhanced; when the deviation of the temperature change trend from the local baseline trend falls within the natural fluctuation envelope of the temperature trend, the derivative action is not increased due to noise.

[0066] Obtained through defuzzification , and These parameters act on the current proportional, integral, and derivative parameters respectively, and are limited by the upper and lower boundaries of the parameters provided by equipment debugging or safety configuration. The aforementioned basic fuzzy PID is only used to form the proportional, integral increment, and derivative actions required for normal temperature control. The control state determination for evaporation load and the selection of the write path between conventional integral memory and evaporation compensation memory are performed independently as described later.

[0067] After the basic control brings the controlled temperature into the permissible detection range, the thermal response detection module determines whether it is appropriate to perform thermal response detection. This determination considers whether the process temperature target is stable, whether the controlled temperature predicted according to the current temperature change trend is still within the safe range defined by the temperature safety boundary, whether the thermal actuator has distinguishable upward and downward adjustment margins, whether the feeding and turning status is stable, whether the fan and door status is stable, whether the actuator feedback is normal, whether there are any alarms or hard interlock actions in the equipment, and whether the current control mode is automatic. The detection process only begins when all permissible conditions are met; if any condition is not met, active detection is not performed in this control cycle.

[0068] Rapid heating, switching of process temperature target, equipment shutdown, thermal actuator approaching unidirectional output limit, door opening or imminent activation, sudden change in feeding or turning status, fan status change, invalid temperature and humidity measurement, manual intervention, equipment alarm, and hard interlock activation all constitute conditions prohibiting detection. Prohibition of detection does not necessarily indicate a malfunction in the tea material or equipment. For situations where there is simply a lack of safe detection space, the system continues to use constrained basic fuzzy PID control; for situations where there are simultaneous sensor, actuator, or interlock anomalies, the execution safety module will initiate the corresponding safety handling path.

[0069] The thermal response detection module uses the currently stable baseline candidate thermal command as the baseline thermal command. thermal command around the baseline This creates a positive command segment that increases heat input and a negative command segment that decreases heat input. Both command segments are relative to the baseline heat command. The commands issued correspond to each other in magnitude and have the same duration, ensuring that the incremental heat input of the entire detection group relative to the baseline remains as balanced as possible. The reverse command segment indicates a reduction in heat input relative to the baseline, and does not require the tea processing equipment to have active cooling capabilities.

[0070] The detection command amplitude uses power, valve position, or normalized control units consistent with the thermal actuator interface. The lower bound of the detection command amplitude is higher than the minimum resolvable change and feedback noise boundary of the thermal actuator, ensuring that the issued command and its response can be identified. Its upper bound is jointly limited by the margin between the current temperature and the temperature safety boundary, the available actuator margin, the allowable command change rate of the equipment, and the thermal response range determined based on existing reliable commissioning data or verified valid detection records within the current comparable operating range. When no detection command amplitude simultaneously satisfies both safety and identifiability requirements, the thermal response detection module will not perform the current detection.

[0071] The duration of the two command segments is determined based on the thermal actuator response time, temperature sensor response time, and current device thermal hysteresis. It only needs to reach a level sufficient to identify the direction of the temperature response; a steady-state temperature is not required. When reliable commissioning data is available, the initial thermal hysteresis range can be determined from this data. After obtaining valid detection records, the thermal hysteresis can be updated based on the temporal relationship between the actuator feedback starting to change and the temperature trend deviating from the local baseline trend until it consistently exceeds the natural fluctuation envelope of the temperature trend. When there are neither valid historical records nor available commissioning ranges, the system only runs basic fuzzy PID control and does not forcibly implement active detection.

[0072] When a valid response envelope has been established, the system prioritizes using the upper boundary of the valid response envelope. This serves as the upper boundary of the response capability for the current thermal process. Let the current controlled temperature be... The current local baseline temperature change rate is Based on the verified command-feedback relationship of the thermal actuator, the expected change in the actual actuator feedback generated by the corresponding detection command is as follows: The duration of the command segment is The predicted temperature of the positive command segment. Determined according to the following formula:

[0073]

[0074] The reverse command segment uses the same response capability boundary for corresponding predictions in the direction of decreasing heat input relative to the baseline. Only the predicted values ​​are available. The corresponding detection command segment is only allowed to be executed when the temperature remains within the safe range defined by the temperature safety boundary. Determined based on reliable commissioning data of the thermal actuator, the current actuator feedback envelope, or verified command-feedback records; otherwise, it cannot be reliably determined. When no active detection is performed, and an effective response envelope has not yet been established, the thermal response range is determined using existing reliable debugging data of the equipment; when there is neither a valid historical record nor a usable debugging range, the system only runs the basic fuzzy PID and does not forcibly perform active detection.

[0075] Please see Figure 2 When both detection and safety conditions are met, the thermal response detection module sequentially generates two thermal command segments in opposite directions, which are confirmed by feedback from the actuator. When an abnormality occurs, the detection is terminated and the baseline is restored. A complete detection record is only formed when the feedback amplitude corresponds and the response window is complete.

[0076] When implementing a set of thermal response probes, the thermal response probe module first freezes the baseline thermal command. The probe command amplitude is superimposed on top of this to form a positive command segment, and the actuator feedback confirms that the thermal input has changed in the increasing direction. After the positive command segment ends, the system restores the baseline thermal command. The system waits for feedback from the actual actuator to return to the baseline feedback envelope, and waits for the difference between the temperature trend and the local baseline trend to return to the range defined by the natural fluctuation envelope of the temperature trend. The return of actuator feedback to the baseline feedback envelope and the recovery of the temperature trend to within the natural fluctuation envelope of the temperature trend are considered together as the baseline recovery condition. After the baseline recovery condition is met, the baseline thermal command is then executed. Subtract the magnitude of the detection command from the above to form the reverse command segment.

[0077] The order of the two command segments can be interchanged. To reduce the time direction offset caused by the device's natural heating or cooling, adjacent effective probes can alternately use the order of increasing then decreasing and decreasing then increasing. Regardless of the order, the baseline recovery condition must be met between the two command segments, and the response calculation should use the actual actuator feedback rather than just the issued command. If the thermal actuator is limited, causing the actual feedback amplitudes of the two command segments to not correspond, the command value cannot be used to replace the actual feedback amplitude, and the entire set of probe records is invalidated.

[0078] After the two command segments are completed, the system determines the actual change magnitude of the forward and reverse command segments relative to the baseline feedback based on the actual actuator feedback. The absolute value of the actual feedback change magnitude of the forward command segment is denoted as... The absolute value of the actual feedback change in the reverse command segment is denoted as... .

[0079] The resolution margin is determined by the minimum resolvable feedback change of the thermal actuator. The feedback noise margin is determined by the actuator feedback fluctuation under steady-state conditions. The sum of the two serves as the allowable deviation of the actual feedback amplitude between the two command segments. :

[0080]

[0081] When the following formula is satisfied, the actual feedback amplitudes of the two command segments are considered to correspond to each other; otherwise, the entire set of probe records is invalid:

[0082]

[0083] when and When within the allowable deviation range, the actual change in heat input used to respond to normalization. Determined as:

[0084]

[0085] The The control unit consistent with the thermal actuator feedback is adopted and used for the normalization calculation of subsequent reversible thermal response, and the amplitude of the detection command issued by the controller is not used to replace the actual actuator feedback change amplitude.

[0086] During the detection process, the safety module continuously checks the temperature trend, output margin, equipment status, and interlock status. If any command segment touches the output boundary, the controlled temperature is predicted to reach or exceed the temperature safety boundary based on the current temperature change trend, the actuator feedback direction is incorrect, feedback timeout occurs, equipment status changes, or a safety interlock is activated, the thermal response detection module immediately terminates the current command segment and restores the baseline thermal command. The resulting incomplete detection records must not be concatenated with subsequent command segments, nor used to update thermal response determination results.

[0087] A complete detection record should include at least the comparable operating condition range and the baseline thermal command. The system includes two opposing command segments, actual actuator feedback, temperature response window and dehumidification response window for each command segment, local baseline trend, and record validity. The thermal response determination module only processes complete records within the same comparable operating condition range where both command segments are confirmed by actuator feedback and the baseline recovery conditions between the two segments are met. The inability to continuously form complete records only indicates that active identification conditions are not currently available, and cannot be used to directly determine the presence of evaporative load.

[0088] The thermal response determination module uses the moment when the actual actuator feedback changes as the response alignment starting point and determines the temperature response window based on the current thermal hysteresis. For the forward command segment, temperature time-series data is obtained from the effective sample segment after the actuator feedback confirms that the thermal command actually changes in the increasing direction and passes through the current thermal hysteresis, according to the sampling time of the response window. For the reverse command segment, temperature time-series data is obtained from the effective sample segment after the actuator feedback confirms that the thermal command actually changes in the decreasing direction and passes through the current thermal hysteresis, according to the same sampling time scale. For each temperature response window, the effective temperature samples arranged according to the sampling time within the window are subjected to first-order least-squares linear fitting to obtain the temperature change rate.

[0089] Assume there are a total of The first valid sample, the... The sampling time for each sample is Temperature value The corresponding window's temperature change rate Determined according to the following formula:

[0090]

[0091]

[0092]

[0093] The forward command window, the reverse command window, and the temperature samples used to determine the local baseline trend all employ the same first-order least squares fitting rule as described above. Let the temperature change rate obtained by fitting the forward command window be... The rate of temperature change corresponding to its local baseline trend The temperature change rate obtained by reverse command window fitting is The rate of temperature change corresponding to its local baseline trend Then the positive response increment and reverse response increment They are respectively:

[0094]

[0095]

[0096] Positive response increment and reverse response increment Both represent the change in the rate of temperature change caused by the command segment relative to the local baseline trend, rather than a simple difference between the two temperature values ​​at the beginning and end of the command segment. When the number of valid samples is insufficient to complete a first-order linear fit, or the fitting residual exceeds the allowable range determined by the temperature measurement uncertainty and the fluctuation of the stable temperature samples, or the sample temperature change trend is inconsistent, the entire set of detection records is invalid and does not form the two response increments mentioned above.

[0097] The thermal response determination module determines the thermal response based on the positive response increment. Incremental Reverse Response and actual heat input change Formation of reversible thermal response quantity and common drift amount Their relationships are as follows:

[0098]

[0099]

[0100] Among them, positive response increment The positive command window represents the rate of temperature change increment relative to its local baseline, while the negative response increment represents the positive response increment. The temperature change rate increment of the reverse command window relative to its local baseline, both using the same unit of temperature change rate; actual heat input change. The absolute value of the actual feedback change amplitude confirmed by the actuator for both forward and reverse command segments. and The arithmetic mean of the reversible thermal response is expressed in the same control unit as the thermal actuator feedback; This indicates the strength of the temperature response as it reverses with the opposite thermal command, used to determine the available control capability of the current thermal execution chain; common drift. This indicates the temperature trend that is retained in both opposite command windows. It is used to determine the heat and humidity load together with the change in dehumidification, and does not represent the moisture content of tea leaves separately.

[0101] Both the forward command segment and the reverse command segment are hot commands around the current baseline. The resulting localized heat input variation, where the reverse command segment only represents the heat command relative to the baseline. Reducing heat input does not require the controlled temperature to actually decrease, nor does it require the tea processing equipment to have active cooling capabilities.

[0102] Positive response increment and reverse response increment Each response window uses its corresponding local baseline trend as a reference, therefore the reversible thermal response quantity It is used to characterize the bidirectional local response capability between actual heat input changes and temperature response near the current operating point, without requiring strict symmetry between the absolute temperature changes of the forward and reverse processes.

[0103] For inherent response deviations caused by thermal actuator hysteresis, heat transfer nonlinearity, or other normal directional differences near the operating point, the effective response envelope is formed by valid detection records within the same comparable operating condition range, showing normal actuator feedback, no detected dehumidification, and stable operating conditions. Only the current An evaporative load state is formed only when the normal effective response envelope is weakened and the conditions of continuous dehumidification and common drift direction are met simultaneously.

[0104] Reversible thermal response Only the actual heat input variation Update the reversible thermal response when the response is above the resolvable boundary, both response windows are intact, actuator feedback is valid, the actual feedback amplitudes of both command segments meet the allowable deviation requirements, baseline recovery conditions are met between the two segments, and the data belong to the same comparable operating condition range. If any of these conditions are not met, the reversible thermal response... and common drift None are updated. Reversible thermal response. When the direction of the temperature response is inconsistent with the direction that should be caused by the increase in heat input, the system checks the sensor polarity, thermal actuator direction mapping, window alignment, and measurement point validity, and stops the current adaptive update.

[0105] In the new comparable operating condition range, the thermal response determination module forms a normal and valid detection record based on complete complementary detection with normal actuator feedback, undetected dehumidification, and stable process temperature target and equipment status. Let the reversible thermal response quantities corresponding to the normal and valid detection records already obtained in the current comparable operating condition range be sequentially... The number of valid records is This forms a reference value for the reversible thermal response. :

[0106]

[0107] Based on the temperature measurement uncertainty, the thermal actuator feedback resolution, and the allowable thermal response changes under normal equipment conditions, each factor is converted to the dimensions of the reversible thermal response quantity, and an allowable response deviation is formed. .by An effective response envelope is formed around the center, and its lower boundary and upper boundary They are respectively:

[0108]

[0109]

[0110] The effective response envelope is used to characterize the reversible thermal response under normal control conditions within the current comparable operating range. The permissible range of variation. Subsequent updates will only use valid detection records showing normal actuator feedback, undetected dehumidification, and stable operating conditions. The corresponding effective response envelope is defined; during updates, single changes to the envelope boundary are restricted, ensuring that new records can only correct the existing envelope within the range determined by temperature measurement uncertainty, command resolution capability, and allowable response changes of the equipment. The effective response envelope is frozen under evaporative load and load transfer states. If the effective response envelope has not yet been established, the thermal response determination module outputs an invalid control state, and the adaptive control module only runs the restricted basic fuzzy PID, without adding evaporative compensation.

[0111] For the current comparable operating condition range, the current reversible thermal response quantity Compare with the effective response envelope. Current reversible thermal response quantity. satisfy At that time, it is determined that the reversible thermal response of the current thermal execution chain is within the normal and effective range; the current reversible thermal response quantity At that time, it is determined that the reversible thermal response quantity is weakened relative to the normal effective response; the current reversible thermal response quantity In such cases, it is not used as evidence of evaporative load, but rather as a condition for re-probing or entering an ineffective control state due to an abnormal change in the thermal response relationship. A decrease in the reversible thermal response only indicates a reduction in the reversible temperature effect produced by the same actual heat input change; it cannot determine whether the decrease originates from tea leaf dehumidification, changes in heat source capacity, or other heat transfer anomalies.

[0112] The thermal response determination module also compares the dehumidification signals in the two command segments with the natural dehumidification fluctuation envelope before detection. If the dehumidification offset exceeds the sensor calibration uncertainty and the natural fluctuation range, and remains in the same direction across two opposite command segments, the dehumidification is determined to be in a continuous state; if the dehumidification offset gradually returns from a continuous state to the natural dehumidification fluctuation envelope, the dehumidification is determined to be in a decaying state; if the dehumidification signal is within the natural dehumidification fluctuation envelope, the dehumidification is determined to be in an undetected state. A valid dehumidification determination is not formed if the airflow path changes, the dehumidification sample is missing, the dehumidification measurement is invalid, or the offset directions in the two windows contradict each other.

[0113] Please see Figure 3 The thermal response determination module forms a unique control state according to the abnormality priority rule, and the adaptive control module then selects the update, maintenance, removal or freezing path of conventional integral memory or evaporation compensation memory based on the control state.

[0114] The thermal response determination module prioritizes abnormalities, identifying a unique control state from among normal control, evaporative load, load transfer, and invalid control states. An invalid control state is prioritized if any necessary input is invalid, actuator feedback fails to follow, detection records are incomplete, the response direction does not conform to the physical relationship of thermal actuation, or safety conditions are violated. Even if the dehumidification signal shows a continuous offset, it cannot cover up actuator feedback abnormalities.

[0115] The thermal response determination module binds the formed control state to the corresponding comparable operating condition interval identifier, the version of the most recent valid detection record, and the time of control state formation. Between two valid detections, the control state is maintained for use in each control cycle only if the controlled temperature, dehumidification signal, equipment status, actuator feedback, process temperature target, and valid response envelope remain valid, and the effective period determined by equipment thermal hysteresis, current state change rate, and safety refresh requirements has not exceeded from the time of control state formation. If the refresh conditions required by equipment thermal hysteresis or state change are met, the effective period expires, or any binding condition changes, the thermal response determination module immediately invalidates the old control state. If no new valid detection record has been obtained, the thermal response determination module outputs an invalid control state, the adaptive control module freezes the online PID parameter increments and the additions to the conventional integral memory and evaporation compensation memory, and switches to restricted basic control or safety handling based on the current temperature, actuator, and safety status.

[0116] With actuator feedback normal and current reversible thermal response... Below the lower boundary of the current valid response envelope Furthermore, when dehumidification is continuous, it only occurs when the controlled temperature is below the process temperature target and the common drift amount... A common drift is considered to occur only when the common temperature change trend of the two command windows deviates from the local baseline trend in the direction of decreasing heating rate or increasing cooling rate, and this deviation exceeds the natural fluctuation envelope of the temperature trend. The condition is compatible with a slow temperature response and forms an evaporative load state; otherwise, this condition does not form. This state indicates that the current low temperature deviation has at least transient endothermic evidence consistent with the continuous dehumidification of the tea leaves, and the thermal command has been actually implemented by the actuator. This includes an isolated humidity peak, an occasional temperature drop, or only a weak reversible thermal response. None of these are sufficient to form an evaporative load state.

[0117] The system was previously under evaporative load, and subsequently dehumidification changed from a continuous state to a decreasing state. New effective detection shows reversible thermal response. When the temperature recovers and begins to rise again, a load transfer state is formed. This state indicates that the original transient heat-absorbing load is withdrawing, and the effect of the same heat input on the controlled temperature is increasing. It is necessary to stop adding temporary compensation and check whether the existing compensation can be removed. The load transfer state cannot be formed solely due to the passage of a predetermined time, nor can it be formed solely based on a single temperature rise.

[0118] When reversible thermal response quantity satisfy A normal control state is formed when the actuator feedback is normal, dehumidification is in an undetected state or has returned to the natural dehumidification fluctuation envelope, and there are no other abnormalities. Reversible thermal response quantity. In the absence of evidence of continuous dehumidification, the slow temperature is not interpreted as an evaporation load of the tea leaves. Instead, it is considered an ineffective control state or a state of limited basic control, in order to check the heat source capacity, heat transfer contact, material blockage, air path, and temperature measurement points.

[0119] In each control cycle, the adaptive control module first performs fuzzy inference based on the temperature deviation and its trend, and then determines the integral write path based on the control state output by the thermal response determination module. This sequence ensures that the control state is determined before the error enters the memory, avoiding the situation where the deviation caused by transient dehumidification is written into the regular integral memory first, and then an attempt is made to separate the temporary load from the ordinary integral memory afterward.

[0120] Fuzzy PID rules must at least distinguish the direction and magnitude of the deviation, as well as whether the temperature is approaching, moving away from, crossing the target, or falling within the noise range. When the deviation is large and continues to move away from the target, the proportional gain and variation suppression effects are enhanced, while limiting the continued accumulation of integral data. When the deviation is large but approaching the target, the necessary proportional gain is maintained, and the increase in integral gain is reduced. When the deviation is small but begins to move away from the target, the proportional and integral gains are moderately increased. When the deviation is small and exists stably in the same direction, steady-state error is eliminated through conventional integral memory under normal control conditions. When the temperature continues to deviate after crossing the target, the proportional and integral gains are reduced, and the variation suppression effect is enhanced. When the deviation of the temperature change trend from the local baseline trend falls within the natural fluctuation envelope of the temperature trend, noise is not interpreted as an effective trend, and the derivative gain is not frequently increased accordingly.

[0121] Under normal control conditions, the conventional integral memory is updated according to fuzzy rules, while the evaporation compensation memory remains at zero. The adjustment of the proportional and derivative parameters is influenced by the current reversible thermal response. The corresponding control capability boundaries and equipment safety parameter boundaries are constrained. Conventional integral memory only receives integral increments used to eliminate sustained temperature deviations under normal controllable thermal response conditions, and is limited by output margin and anti-integral saturation conditions.

[0122] When under evaporative load conditions, the adaptive control module does not integrate all temperature deviations separately, nor does it establish an additional controller unrelated to the fuzzy PID controller. Instead of writing the integral increment that the fuzzy PID controller would have originally intended to write into the regular integral memory for the current control cycle, which would have increased the heat input, it redirects the same increment to the evaporation compensation memory. Reverse integral adjustments that could cause the system to exit output saturation or reduce heat input can still be performed according to safety rules.

[0123] Evaporation-compensated memory has independent source states, write boundaries, and lifecycles. Each time a write operation is prepared, the adaptive control module checks whether the temperature deviation has not yet converged to the target, whether the thermal actuator has a safe output margin, whether the controlled temperature predicted based on the current temperature change trend remains within the process-allowed temperature range, and the reversible thermal response. Is it still valid? The compensation increment for this cycle will only be accepted if all the above conditions are met; if any condition is not met, the existing compensation will be maintained or reduced. The upper limit of the evaporation compensation memory is determined by the current temperature safety margin, the actuator output margin, and the lower boundary of the current effective response envelope. The minimum effective value of the reversible thermal response quantity and the anti-saturation condition are jointly determined, and fixed values ​​that are universal across devices are not used.

[0124] When in load transfer mode, the adaptive control module stops writing new values ​​to the evaporation compensation memory and uses the minimum resolvable decrement of the thermal actuator as the trial removal step size. The system determines the lower boundary of the current effective response envelope. The lowest effective value of the reversible thermal response is used to determine whether the controlled temperature, predicted according to the current temperature change trend after removing this step size, is still within the process's allowable temperature range and will not deviate far from the process temperature target. If the condition is met, the removal is accepted; if not, the existing compensation is maintained, and a new judgment is made in the next effective control cycle or after obtaining new effective detection results.

[0125] The removal of evaporation compensation memory continues until it reaches zero, the control state reverts to evaporation load state, or an ineffective control state is entered due to an anomaly. Once compensation reaches zero, dehumidification returns to the natural fluctuation envelope of dehumidification, and the risk of temperature rise exceeding limits is eliminated, the system enters normal control state, and the conventional integral memory resumes its role in eliminating continuous static error. Therefore, the temporary heating generated during dehumidification will not be retained beyond its corresponding load lifecycle.

[0126] When in an invalid control state, the adaptive control module freezes the online PID parameter increments and new additions in both types of memory, discards probe results that have not yet been used to form a control state or update the effective response envelope, and notifies the execution safety module to select a rollback action. When there is a short-term temperature or communication data anomaly but the temperature is still within the safe range and the actuator can still respond normally, the restricted basic fuzzy PID can be run using the most recent safety parameters; when the dehumidification signal is invalid but the temperature and actuator are valid, the restricted control can also be maintained, but an evaporative load state or new evaporation compensation must not be formed.

[0127] The adaptive control module synthesizes proportional action, conventional integral memory, derivative action, and evaporative compensation memory into candidate thermal commands. The execution safety module first limits the amplitude according to the physical range of the thermal actuator, then limits the speed according to the allowable rate of change of the command, and verifies the automatic control mode, temperature safety boundary, and hard interlock status. Candidate thermal commands are only sent to the thermal actuator after passing the above constraints. When output saturation occurs, the execution safety module returns the saturation direction to the adaptive control module, preventing the conventional integral memory and evaporative compensation memory from continuing to increase along the saturation direction.

[0128] After a hot command is issued, the safety module compares the direction, timing, and amplitude of the actual actuator feedback to ensure they fall within the device's allowable envelope. If the feedback changes in the correct direction within the allowed time, the command is considered executed. If the feedback reaches the physical boundary, further increases in the corresponding direction cease. If the feedback direction is incorrect, the amplitude is incompatible, or the interface is interrupted, the current detection record and the load judgment based on that record become invalid. The system must not attribute slow temperature response to tea leaf dehumidification when the actuator fails to follow, nor may it attempt to mask execution failures by repeatedly issuing hot commands.

[0129] When an actuator malfunctions, the controlled temperature is predicted to reach or exceed the temperature safety boundary based on the current temperature change trend, an effective temperature cannot be formed, or a hard interlock is activated, the execution safety module sends a preset safety power command, a heat shutdown command, or a manual takeover request according to the equipment's existing safety configuration, and outputs corresponding alarms. After manual takeover, the automatic control stops active detection and memory writing, retaining only necessary status monitoring. If there is still no execution feedback for the safety command, the system maintains the heat shutdown request and relies on the equipment's independent safety interlock and predetermined manual handling, without automatically disengaging or overriding the safety device.

[0130] During fault recovery, it is first confirmed that the hard interlock and actuator faults have been eliminated according to the equipment procedures, and that the process temperature target, feeding, turning, fan, and door status have stabilized again. The data synchronization module then establishes a new comparable operating condition range, re-establishing the temperature baseline, dehumidification baseline, and actuator feedback envelope. After the actuator is checked through normal command feedback, the thermal response detection module re-completes a set of valid opposite thermal command segments, and the thermal response determination module confirms the reversible thermal response quantity. Only after confirming the physical direction and input validity can the system resume adaptive control. The evaporation compensation memory from the previous comparable operating condition interval is not reloaded after recovery.

[0131] In a complete operation, the system first loads the process temperature target and equipment safety configuration for the current processing stage, runs the basic fuzzy PID with safety initial parameters, and drives the thermal actuator through the execution safety module. The data synchronization module continuously acquires the controlled temperature, dehumidification signal, equipment status, and actuator feedback, establishing a comparable operating range and baseline when the state is stable. Once the temperature enters the allowable detection range and the thermal actuator has sufficient adjustment margin, the thermal response detection module generates two opposite command segments with mutually canceling incremental thermal inputs, and satisfies the baseline recovery condition between the two segments.

[0132] The thermal response determination module aligns the response windows corresponding to the two command segments and eliminates local baselines to form a reversible thermal response quantity. Common drift amount The system judges changes in humidity and combines this with actuator feedback to form a unique control state. The adaptive control module selects the write, hold, and remove paths for conventional integral memory or evaporation compensation memory based on this control state. The execution safety module constrains the candidate thermal commands and applies them to the actual thermal actuator, returning the actual feedback to the next control cycle, thus forming a closed-loop control relationship connected to the temperature, humidity removal process, and heat input of the tea processing area.

[0133] When tea leaves continuously release moisture, reversible thermal response quantity When the temperature decreases and the actuator feedback is normal, the system writes the temporary integral increment required to maintain the temperature into the evaporation compensation memory, preventing it from becoming a regular integral. This is due to the decrease in dehumidification and the reversible thermal response. Upon recovery, the system prioritizes removing evaporation compensation, preventing the transient endothermic load from continuously driving the thermal actuator. This function relies on effective temperature and humidity measurements, a safe detection window, comparable equipment status, and the thermal actuator's ability to adjust around the baseline with normal execution feedback. It helps reduce the risks of continued heating after the dehumidification load is removed, temperatures exceeding the process's allowable temperature range, and repeated corrections.

[0134] When the equipment cannot increase or decrease heat input around the current baseline, active detection is not allowed throughout the entire processing stage, there is no stable physical relationship between the humidity measurement and the actual humidity process, the temperature measurement point cannot reflect the controlled area, or the independent safety interlock is unavailable, the active thermal response detection and evaporation load adaptive path will not be activated. In this case, the system can only use the equipment-permitted basic temperature control or safety shutdown, and will not use historical tea variety, fixed humidity threshold, or unverified response data to replace the actual temperature, humidity, and actuator feedback evidence in the current comparable operating range.

[0135] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. An adaptive temperature control system for tea processing equipment based on fuzzy PID, characterized in that, include: The data synchronization module is used to associate the process temperature target, temperature, humidity, equipment status and actuator feedback at the sampling time to form a synchronized measurement frame and output the data validity. The thermal response detection module is used to determine whether detection is allowed based on the synchronous measurement frame and data validity. When detection is allowed, two thermal command segments with opposite directions are formed around the same baseline and confirmed by actuator feedback. Between the two segments, the actuator feedback and temperature trend meet the baseline recovery conditions. In case of an anomaly, the command segment is terminated and the baseline thermal command is restored through the execution safety module. The thermal response determination module is used to perform window alignment and local baseline elimination on the temperature response and dehumidification response of the two command segments, separate the reversible thermal response quantity and the common drift quantity, and combine the dehumidification continuity and actuator feedback to form one of the following states: normal control state, evaporative load state, load transfer state or invalid control state. The adaptive control module is used to adjust the fuzzy PID parameters according to the control state, select the write, hold and remove paths of conventional integral memory or evaporation compensation memory, and synthesize candidate thermal commands. The execution safety module is used to limit the amplitude, speed, and interlock of candidate hot commands, drive the hot actuator, and provide feedback on the actual execution results.

2. The adaptive temperature control system for tea processing equipment based on fuzzy PID according to claim 1, characterized in that, By associating the sampling time with the process temperature target, temperature, dehumidification, equipment status, and actuator feedback, a synchronous measurement frame is formed, and the data validity is output, including: The comparable operating condition range is defined as the continuous operating period during which the process temperature target, equipment operation mode, feeding status, air path status, and heat transfer connection are kept comparable. Within the comparable operating condition range, continuous and valid samples excluding door opening, feeding start and stop, fan switching, or thermal actuator malfunction are used to form a local temperature baseline trend, a temperature trend natural fluctuation envelope, a dehumidification natural fluctuation envelope, a baseline thermal command, and an actuator feedback envelope.

3. The adaptive temperature control system for tea processing equipment based on fuzzy PID according to claim 2, characterized in that, The comparable operating condition range is defined as the continuous operating period during which the process temperature target, equipment operating mode, feeding status, airflow status, and heat transfer connections remain comparable. When the thermal response relationship becomes incomparable due to changes in process temperature target switching, door operation, sudden changes in feeding or turning, fan switching, or control mode switching, the current comparable operating condition range ends. The reference data formed by the completed detection records and used for subsequent thermal response comparison, as well as the incomplete detection records and response results not yet used for control state determination, are invalidated within the comparable operating condition range. Evaporation compensation memory migration is prohibited, and a new comparable operating condition range is established after the equipment state is re-stabilized.

4. The adaptive temperature control system for tea processing equipment based on fuzzy PID according to claim 1, characterized in that, During permitted detection, two thermal command segments are formed around the same baseline in opposite directions and confirmed by actuator feedback. Between the two segments, the actuator feedback and temperature trend satisfy the baseline recovery condition, including: Detection is only permitted when the process temperature target, temperature trend, feeding, turning, fan, door and measurement signals are stable and effective, the thermal actuator has distinguishable upper and lower adjustment margins, the actuator feedback is normal and in an automatic control mode without alarms or hard interlock actions. Detection is not performed when the detection amplitude exceeds the minimum resolvable change and feedback noise boundary of the thermal actuator, and is limited by temperature safety margin, actuator margin and command change rate. Detection is not performed when there is no detection amplitude that simultaneously satisfies safety and resolvability.

5. The adaptive temperature control system for tea processing equipment based on fuzzy PID according to claim 4, characterized in that, During permitted detection, two thermal command segments are formed around the same baseline in opposite directions and confirmed by actuator feedback. Between the two segments, the actuator feedback and temperature trend satisfy the baseline recovery condition, including: The actual feedback amplitudes of the actuators relative to the baseline in the two thermal command segments correspond to each other within the allowable deviation range determined by the minimum resolvable change of the thermal actuator and the feedback noise, and the durations of the two command segments are the same. First, execute one command segment. After the actuator feedback recovers to the stable feedback range before detection, and the deviation of the temperature trend from the local temperature trend before detection recovers to the natural fluctuation range determined by the stable effective temperature sample and the temperature measurement uncertainty, execute another command segment. If any command segment touches the output boundary, the controlled temperature is predicted to reach or exceed the temperature safety boundary according to the current temperature change trend, the actuator feedback error or timeout occurs, the equipment status changes or interlocking action is performed, the current command segment is terminated and the baseline thermal command is restored, and incomplete detection records are discarded.

6. The adaptive temperature control system for tea processing equipment based on fuzzy PID according to claim 1, characterized in that, Window alignment and local baseline elimination are performed on the temperature and dehumidification responses of the two command segments to separate reversible thermal response quantities from common drift quantities, including: Align the two temperature response windows with the moment when the actual actuator feedback begins to change, and after thermal hysteresis, use the same temperature change rate fitting rule to obtain the temperature response increment of each relative local baseline trend. The difference between the two temperature response increments is normalized according to the actual heat input change amplitude confirmed by the actuator feedback to form a reversible thermal response quantity, and the temperature trend retained by both is used to form a common drift quantity; When the window is incomplete, the actual feedback amplitude does not correspond, the sample is insufficient, or the data does not belong to the same comparable operating condition range, the reversible thermal response quantity and the common drift quantity are not updated.

7. The adaptive temperature control system for tea processing equipment based on fuzzy PID according to claim 1, characterized in that, The combination of continuous dehumidification and actuator feedback results in one of the following states: normal control state, evaporative load state, load transfer state, or ineffective control state, including: An effective response envelope is formed and updated using complete detection records showing normal actuator feedback, no detected dehumidification, and stable operating conditions. This envelope is used to characterize the allowable range of change in reversible thermal response under normal control conditions. The effective response envelope is frozen under evaporative load and load transfer conditions. When the current reversible thermal response is below the lower boundary of the effective response envelope, the two command segments of the dehumidification span continuously deviate in the same direction, and when the controlled temperature is below the process temperature target, the common temperature change trend of the two command windows deviates relative to the local temperature trend before detection in the direction of decreasing the heating rate or increasing the cooling rate. When this deviation exceeds the natural fluctuation range jointly determined by the natural fluctuation of the temperature trend of the stable effective temperature sample and the temperature measurement uncertainty, an evaporation load state is formed. When the necessary input is invalid or the actuator feedback is abnormal, an invalid control state is preferentially formed.

8. The adaptive temperature control system for tea processing equipment based on fuzzy PID according to claim 1, characterized in that, Adjust the fuzzy PID parameters according to the control state, and select the write, hold, and delete paths for conventional integral memory or evaporation compensation memory, including: Under normal control conditions, update the conventional integral memory and keep the evaporation compensation memory at zero; Under evaporative load conditions, the integral increment that is intended to be written into the conventional integral memory and will increase the heat input will be redirected to the evaporative compensation memory. In load transfer mode, stop adding new evaporation compensation and gradually remove existing compensation. In ineffective control mode, freeze the increment of control parameters and the addition of two types of memory, and switch to restricted basic fuzzy PID control or safety handling.

9. The adaptive temperature control system for tea processing equipment based on fuzzy PID according to claim 8, characterized in that, Choose the write, hold, and erase paths for conventional integral memory or evaporation-compensated memory, including: Write only when the temperature deviation has not converged to the target, the thermal actuator has a safety margin, the controlled temperature predicted according to the current temperature change trend remains within the process-allowed temperature range, and the reversible thermal response is valid; The upper limit of evaporation compensation memory is determined by the temperature safety margin, the actuator output margin, the minimum effective value of the reversible thermal response determined by the effective detection record under normal control conditions, and the anti-saturation condition. When the dehumidification decreases, the reversible thermal response recovers, and the temperature rises, the compensation is removed according to the minimum resolvable reduction of the thermal actuator until the compensation returns to zero, the evaporative load state is re-established, or the system enters an ineffective control state.

10. The adaptive temperature control system for tea processing equipment based on fuzzy PID according to claim 1, characterized in that, Verify the amplitude, speed, and interlock of candidate hot commands, drive the hot actuator, and provide feedback on the actual execution results, including: When the output is saturated, the saturation direction is fed back to prevent both types of memory from increasing along that direction. When the actuator reports an error, timeout, or interface interruption, the current detection record and corresponding control status become invalid, and a safety power command, a heat stop command, or a manual takeover request preset by the equipment safety configuration is sent according to the temperature and interlock status. After the fault is eliminated and the equipment status is stable, a new comparable operating condition range is established, a new baseline is formed, and a set of effective opposite thermal command segments are completed before adaptive control is resumed without loading the evaporation compensation memory of the previous comparable operating condition range.

Images