Temperature and humidity self-adaptive cooperative control system for medicinal material fermentation tank

Abstract

This invention belongs to the field of automatic control technology for fermentation processes, specifically relating to an adaptive and coordinated control system for temperature and humidity in a medicinal herb fermentation tank. The system includes: a parameter acquisition module, which acquires the core temperature and volumetric moisture content of the fermentation tank and obtains the actual output power of the heating equipment; a latent heat calculation module, which calculates the average rate of decrease in volumetric moisture content within a preset time window as the evaporation rate, and multiplies the evaporation rate by a preset vaporization constant to obtain the latent heat of vaporization; and a heat production identification module, which acquires the heat storage and convective heat dissipation terms corresponding to the core temperature of the fermentation tank, adds the heat storage and convective heat dissipation terms to the latent heat of vaporization, and subtracts the actual output power to obtain the equivalent heat production rate. The equivalent heat production rate reflects the heat power generated by microbial metabolism. This invention incorporates latent heat of vaporization into the heat balance, achieving dynamic complementarity between heating power and microbial heat production, stabilizing temperature and reducing energy consumption, and maintaining the activity of the microbial community.

Description

Technical Field

[0001] This invention belongs to the field of automatic control technology for fermentation processes, specifically relating to an adaptive and coordinated control system for temperature and humidity in medicinal herb fermentation tanks. Background Technology

[0002] The fermentation process of medicinal herbs requires the fermentation pile to be maintained within a specific temperature range for a relatively long period of time to ensure the continuous activity of functional microorganisms and the sufficient accumulation of metabolites. During the fermentation process, the heat generated by microbial metabolism changes dynamically with the growth stage of the microbial community. External heating equipment needs to adjust its output power in real time according to the self-heating level of the microorganisms in order to stably control the core temperature of the fermentation pile within the target range while avoiding temperature overshoot.

[0003] Currently, patent application CN121348888A discloses an intelligent control method and system for an integrated fermentation and steaming cabinet. This method uses the carbon dioxide concentration inside the cabinet as an indicator of the fermentation degree, dynamically adjusts the membership functions of each output in the fuzzy PID control model according to the fermentation degree, and uses the rate of change of carbon dioxide concentration as the fermentation activity index to determine the integral inhibition coefficient. Based on the mixed control output, the PID controller parameters are adjusted to achieve adaptive temperature control of the integrated fermentation and steaming cabinet.

[0004] However, the above method indirectly reflects microbial activity using carbon dioxide concentration, failing to include the latent heat carried away by moisture evaporation from the pile in the heat budget calculation. In the scenario of medicinal herb fermentation tanks, moisture evaporation intensifies significantly after the pile temperature exceeds 50 degrees Celsius, with latent heat of evaporation accounting for 30% to 50% of the total heat dissipation. If this heat loss is ignored, the controller's estimation of the actual heat production level of microorganisms will produce a systematic deviation, resulting in an inaccurate complementary relationship between the output power of the heating equipment and the heat production of microorganisms. Consequently, the pile temperature will fluctuate repeatedly or continuously deviate from the target value during the high-temperature stage. Summary of the Invention

[0005] This invention provides an adaptive and coordinated control system for temperature and humidity in medicinal herb fermentation tanks to solve the technical problem in the prior art that the latent heat of vaporization is not included in the heat balance calculation, resulting in the inability to form an accurate complementary relationship between the output power of heating equipment and the heat generated by microorganisms.

[0006] This invention provides an adaptive and coordinated control system for temperature and humidity in a medicinal herb fermentation tank, comprising: The parameter acquisition module collects the core temperature and volumetric water content of the reactor body, and obtains the actual output power of the heating equipment. The latent heat calculation module calculates the average rate of decrease in volumetric water content within a preset time window as the evaporation rate, and multiplies the evaporation rate by a preset vaporization constant to obtain the latent heat of evaporation. The heat generation identification module obtains the heat storage and convective heat dissipation items corresponding to the core temperature of the reactor body. After adding the heat storage, convective heat dissipation, and latent heat of vaporization items, the actual output power is subtracted to obtain the heat generation rate equivalent. The heat generation rate equivalent is used to reflect the heat power generated by microbial metabolism. The gain determination module performs linear fitting on the heat production rate equivalent of multiple historical sequences, uses the slope obtained from the fitting as the heat production rate trend value, and determines the current fermentation stage based on the heat production rate trend value; it obtains the quotient of the heat production rate equivalent and the preset full load equivalent, and subtracts the quotient from the constant 1 to obtain the adaptive gain coefficient. The component calculation module subtracts the heat generation rate equivalent from the convective heat dissipation term to obtain the feedforward component, and uses a proportional-integral controller to process the temperature deviation between the preset target temperature and the core temperature of the reactor to obtain the feedback component. The power output module sums the feedforward and feedback components, multiplies them by an adaptive gain coefficient to generate the target output power, and uses the target output power to control the heating equipment.

[0007] Its effect is as follows: the feedforward component, feedback component, and adaptive gain coefficient work together in the generation process of the target output power. The feedforward component predicts the heat gap based on the difference between the convective heat dissipation term and the equivalent heat production rate, so that the feedback controller does not have to bear the entire adjustment load, reducing the risk of overshoot; the adaptive gain coefficient introduces the microbial heat production level into the gain channel of the control loop, so that the output of the heating equipment and the biological heat production are proportionally complementary, avoiding temperature overshoot or insufficient compensation caused by the superposition of the two; the feedback component eliminates the residual deviation of the feedforward compensation and external disturbances through the proportional-integral controller, ensuring that the core temperature of the reactor body stably tracks the preset target temperature.

[0008] Furthermore, the core temperature and volumetric water content of the reactor were collected, including: Obtain the first detected value of the core temperature and the second detected value of the volumetric water content of the reactor body for the current cycle; In response to the first detection value being within a first preset range and the second detection value being within a second preset range, the first detection value is determined as the core temperature of the reactor body, and the second detection value is determined as the volumetric water content; In response to the first detection value exceeding the first preset range, or the second detection value exceeding the second preset range, the first and second detection values ​​of the current period are removed and an abnormal alarm is triggered.

[0009] Its effect is that: the first and second detection values ​​collected by the sensor are respectively checked within a preset range, and abnormal readings caused by sensor failure or interference are eliminated before the data enters the heat balance calculation. This avoids the deviation of the target output power after a single abnormal data is amplified step by step through the evaporation rate calculation and heat generation rate identification, and ensures that the control decision is always based on the effective physical quantity input.

[0010] Furthermore, before collecting the core temperature and volumetric water content of the reactor, the following steps are also included: With the core temperature of the reactor body in the initial temperature range before biological heat generation occurs, the heating equipment is driven by constant test power to run for a calibration time. The temperature rise curve within the calibration time is obtained. Based on the correspondence between the slope of the temperature rise curve and the constant test power, the equivalent heat capacity and heat dissipation coefficient corresponding to the core temperature of the reactor are calibrated.

[0011] Furthermore, calculating the average rate of decrease in volumetric water content within a preset time window as the evaporation rate includes: Obtain a sliding smooth window containing a specific number of acquisition cycles; Calculate the difference in volumetric water content between the first and last periods within the sliding smoothing window; Divide the difference by the time span of the sliding smoothing window to obtain the evaporation rate.

[0012] Furthermore, obtaining the heat storage and convective heat dissipation terms corresponding to the core temperature of the reactor body includes: Obtain the temperature change of the core temperature of the reactor body between two adjacent acquisition cycles; Divide the temperature change by the data acquisition period, and multiply the result of the division by the equivalent heat capacity to obtain the heat storage term; Obtain the actual temperature difference between the core temperature of the reactor and the current ambient temperature, and multiply the actual temperature difference by the heat dissipation coefficient to obtain the convective heat dissipation term.

[0013] Furthermore, by adding the heat storage term, the convective heat dissipation term, and the latent heat of vaporization term, and subtracting the actual output power, the equivalent heat production rate is obtained, which includes: Calculate the sum of the values ​​of the heat storage term, the convective heat dissipation term, and the latent heat of vaporization term; Subtract the actual output power from the sum of the numerical values ​​to obtain the instantaneous heat generation rate; The moving average of multiple instantaneous heat production rates continuously acquired within the historical data smoothing window is calculated, and the mean calculation result is determined as the heat production rate equivalent.

[0014] Furthermore, the current fermentation stage is determined based on the trend value of the heat production rate, including: When the heat production rate trend value is greater than the first slope threshold, the fermentation stage is determined to be the temperature rise activation period. If the absolute value of the heat production rate trend value is not greater than the first slope threshold, the fermentation stage is determined to be a high-temperature maintenance period. When the heat production rate trend value is less than the second slope threshold, the fermentation stage is determined to be the cooling decline period, where the second slope threshold is a negative number.

[0015] Its effect is as follows: based on the heat production rate trend value, the fermentation process is divided into three stages: the temperature rise activation period, the high temperature maintenance period, and the temperature drop decline period. The stage determination is made by using the heat production rate change trend rather than the absolute temperature value, which allows the control strategy to gain a time lead. The turning point of the heat production rate trend occurs before the temperature turning point, and the heating equipment can switch the adjustment direction in advance before the pile temperature changes significantly, reducing the temperature fluctuation amplitude during the stage transition period.

[0016] Furthermore, subtracting the quotient from the preset value of 1 yields the adaptive gain coefficient, which includes: The initial gain is determined by subtracting the quotient from the preset value of 1. The initial gain is subjected to a limiting process based on a set lower limit and a set upper limit to obtain a limited gain; During the transition period when the fermentation stage is switched, a linear interpolation operation is performed on the historical limiting gain of the previous fermentation stage and the current limiting gain, and the interpolation result is determined as the adaptive gain coefficient.

[0017] Its effects are as follows: the initial gain is limited and linear interpolation is used during stage switching. The lower limit of the limit keeps the heating tube in a hot standby state, thereby eliminating the response lag of the actuator when starting from cold. The linear interpolation makes the gain coefficient transition smoothly during stage switching instead of a step change, preventing instantaneous power output jumps from causing oscillations in the reactor temperature, and ensuring the continuity and stability of the heating equipment output.

[0018] Furthermore, the feedback components obtained by processing the temperature deviation between the preset target temperature and the core temperature of the reactor using a proportional-integral controller include: The temperature deviation is obtained by subtracting the core temperature of the reactor from the preset target temperature. Multiply the temperature deviation by the set proportional coefficient to obtain the proportional control term; The temperature deviation is integrally accumulated, and the integral accumulation result is multiplied by a set integral coefficient to obtain the integral control term; When the target output power reaches the rated upper limit and the temperature deviation is positive, or when the target output power reaches the power output lower limit and the temperature deviation is negative, stop accumulating the temperature deviation to freeze the integral control term; The feedback component is obtained by summing the proportional control term and the integral control term.

[0019] Furthermore, after controlling the heating equipment using the target output power, it also includes: In response to the reactor core temperature not being lower than the set safe upper limit temperature, the pre-calculation is ignored, and the control command sent to the heating equipment is forcibly reset to zero until the reactor core temperature drops back to the safe recovery range. In response to multiple consecutive invalid sampling cycles of volumetric water content, the latent heat of vaporization is calculated using the most recent valid evaporation rate. In response to the duration of invalid volumetric moisture content reaching the degradation limit, the latent heat of vaporization is forcibly set to 0, and a sensor maintenance prompt is triggered.

[0020] Its effects are as follows: setting up over-temperature forced zeroing protection and a graded degradation strategy for moisture content sensors, the over-temperature protection serves as the final safety layer independent of the control algorithm, which constrains the pile temperature within the range where the functional microbial community can survive; the graded degradation strategy maintains normal control accuracy by using the most recent effective evaporation rate when the sensor fails for a short time, and degrades to a simplified mode without evaporation correction when the sensor fails for a long time, but can still maintain basic closed-loop function, thus avoiding the shutdown of the entire control process due to a single sensor failure.

[0021] The beneficial effects are as follows: This invention collects the core temperature and volumetric moisture content of the fermentation pile in real time during the fermentation process. Based on the heat balance equation, it explicitly incorporates the latent heat of vaporization into the energy budget calculation, identifies the equivalent heat production rate generated by microbial metabolism, and dynamically adjusts the output power of the heating equipment accordingly. This ensures that electric heating compensation and microbial self-heating complement each other throughout the fermentation cycle. Compared to a constant gain feedback control method based solely on temperature deviation, this scheme enables the heating equipment to automatically match the output ratio according to the microbial activity level. It reduces the heating power in advance during the temperature rise and activation period to avoid temperature overshoot, and promptly increases the heating power during the temperature decline period to maintain the target temperature. This reduces temperature oscillation amplitude, lowers heating energy consumption, and simultaneously ensures the continuous activity of functional microorganisms within the optimal temperature range. Attached Figure Description

[0022] Figure 1 This is a block diagram of the adaptive and coordinated control system for temperature and humidity in a medicinal herb fermentation tank.

[0023] Figure 2 This is a comparison chart showing the trend of the change in the heat production rate equivalent of the entire fermentation cycle and the target output power of the present invention. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] An embodiment of the temperature and humidity adaptive and coordinated control system for medicinal material fermentation tanks provided by this invention: This embodiment is applied to a medicinal material fermentation tank with electric heating pipes laid at the bottom and natural ventilation openings at the top. Medicinal materials are piled up in the tank for microbial fermentation. The electric heating pipes serve as heating equipment to assist in raising the temperature of the pile in the early stage of fermentation or at low temperatures, so that the core temperature of the pile is maintained in the optimal activity range of microorganisms.

[0026] like Figure 1 As shown, the temperature and humidity adaptive and coordinated control system for the medicinal herb fermentation tank includes: The parameter acquisition module collects the core temperature and volumetric water content of the reactor body, and obtains the actual output power of the heating equipment.

[0027] Platinum resistance temperature sensors are deployed in the core layer of the reactor to collect the core temperature, and dielectric moisture content sensors are deployed simultaneously to collect the volumetric moisture content. The acquisition cycle is set to 5 minutes. The thermal inertia time constant of the fermented medicinal materials is usually over 30 minutes, and the 5-minute acquisition cycle is approximately one-sixth of this time constant, meeting the requirements for tracking the dynamic temperature process. The actual output power of the heating equipment is directly read by the heating equipment controller in each acquisition cycle.

[0028] To ensure the reliability of the input data, the validity of the sensor readings is verified in each acquisition cycle. The system acquires the first detected value of the core temperature and the second detected value of the volumetric moisture content for the current cycle. If both the first and second detected values ​​fall within a first preset range and a second preset range, the first detected value is determined as the core temperature and the second detected value as the volumetric moisture content, for use in subsequent calculations. The first preset range is set to 10°C to 85°C, with the lower limit corresponding to the lowest reasonable temperature when the pool is not activated under extreme low-temperature conditions in winter, and the upper limit corresponding to the highest temperature that the pool may experience under extreme abnormal operating conditions. The second preset range is set to 20% to 80%, with the lower limit corresponding to the drying limit of the medicinal material's moisture content and the upper limit corresponding to the moisture content boundary when the material's pores are completely saturated. If either the first or second detected value exceeds the first preset range, the current cycle's first and second detected values ​​are discarded, and an anomaly alarm is triggered, notifying maintenance personnel to check the status of the corresponding sensors. Data from this cycle is not used in subsequent heat balance calculations.

[0029] Before entering the control closed loop, the thermophysical parameters of the reactor core are calibrated online. With the core temperature of the reactor core in the initial temperature range before biothermal generation, the heating equipment is driven by a constant test power for calibration time. The initial temperature range corresponds to the temperature range when the material is first introduced into the pool and the microorganisms have not yet entered the active metabolic stage; in this embodiment, it is taken as ambient temperature to 5 degrees Celsius above ambient temperature. The constant test power is set to 50% of the rated power of the heating equipment, and the calibration time is set to 30 minutes. Selecting 50% of the rated power ensures sufficient temperature rise to guarantee calibration accuracy while avoiding excessively high local temperatures caused by full-power heating, which could trigger premature microbial activation. During the 30-minute calibration time, the temperature rise of the reactor core is approximately 3 to 5 degrees Celsius, sufficient to form a stable linear temperature rise range.

[0030] The temperature rise curve within the calibration period is obtained. Based on the correspondence between the slope of the temperature rise curve and the constant test power, the equivalent heat capacity and heat dissipation coefficient corresponding to the core temperature of the reactor are calibrated. The linear segment of the temperature rise curve is fitted using the standard least squares method to obtain the temperature rise slope. The equivalent heat capacity is equal to the constant test power divided by the temperature rise slope, representing the amount of heat input required to raise the core temperature of the reactor by 1 degree Celsius. After calibration, the heating equipment is turned off, and the temperature drop curve during the natural cooling phase is recorded. The heat dissipation coefficient is equal to the equivalent heat capacity multiplied by the absolute value of the cooling slope and then divided by the difference between the core temperature of the reactor and the ambient temperature, representing the heat power lost by the reactor to the environment per unit temperature difference. Both parameters are considered constants under the condition of the same fermentation batch and unchanged material ratio.

[0031] The synchronous acquisition of core temperature and volumetric moisture content of the reactor body ensures the time alignment of temperature and humidity field information, and the online calibration of equivalent heat capacity and heat dissipation coefficient eliminates the deviation of model parameters caused by the difference in material ratio of different batches.

[0032] The latent heat calculation module calculates the average rate of decrease in volumetric water content within a preset time window as the evaporation rate, and multiplies the evaporation rate by a preset vaporization constant to obtain the latent heat of evaporation term.

[0033] The evaporation rate was calculated using a sliding smoothing window method, which obtained a sliding smoothing window containing a specific number of acquisition cycles. In this embodiment, the specific number was 24 acquisition cycles, corresponding to a time span of 2 hours. The evaporation process is driven by both the surface temperature of the pile and the ventilation status. The change in moisture content within a single acquisition cycle is extremely small and fluctuates significantly due to limitations in sensor measurement accuracy. The 2-hour window span can effectively smooth out measurement noise while still capturing the trend of increased evaporation during the fermentation heating period.

[0034] Calculate the difference in volumetric moisture content between the first and last periods within the sliding smoothing window. The first period is the earliest sampling period within the window, and the last period is the latest sampling period. The difference equals the volumetric moisture content of the first period minus the volumetric moisture content of the last period. Divide this difference by the time span of the sliding smoothing window to obtain the evaporation rate. During fermentation, the moisture in the pile continuously evaporates, and the volumetric moisture content decreases monotonically with time. The moisture content of the first period is greater than that of the last period, resulting in a positive evaporation rate.

[0035] The calculation process is illustrated with a set of actual collected data: the volumetric water content at the beginning of the window is 58.4%, and at the end of the window it is 57.6%. The window time span is 120 minutes, and the evaporation rate is (58.4%-57.6%) / 120=0.0067% / min, indicating that the volumetric water content of the pile is decreasing at a rate of 0.0067 percentage points per minute within the current 2 hours.

[0036] The latent heat of vaporization is obtained by multiplying the evaporation rate by a preset vaporization constant. The heat removed by evaporation per minute is equal to the mass of water evaporated in that minute multiplied by the latent heat of vaporization of water. The mass of water evaporated per unit time is obtained by converting the evaporation rate to the dry basis mass of the pile: the evaporation rate is expressed as a percentage per minute, multiplied by the dry basis mass of the pile and divided by 100 to convert it to kilograms per minute. The latent heat of vaporization of water, the dry basis mass of the pile, and the percentage conversion factor are combined into a preset vaporization constant, which is a constant within a single fermentation batch. The latent heat of vaporization of water is taken as the physical property constant at standard atmospheric pressure, 2260 kJ / kg. The dry basis mass of the pile is obtained from the loading record, which is 200 kg in this embodiment. The preset vaporization constant is: 2260 × 200 / 100 = 4520 kJ / %. The latent heat of vaporization is equal to the evaporation rate multiplied by the preset vaporization constant: 0.0067 × 4520 = 30.3 kJ / min, which is approximately 0.505 kW.

[0037] The physical meaning of the latent heat of vaporization is the amount of heat carried away from the reactor core per unit time when water in the reactor changes from a liquid to a gaseous state. When the core temperature of the reactor core exceeds 50 degrees Celsius, the evaporation rate increases significantly, and the latent heat of vaporization can account for 30% to 50% of the total heat dissipation.

[0038] At the start of each data collection cycle, the sliding window advances one cycle, incorporating the latest volumetric moisture content reading and discarding the earliest reading, and recalculating the evaporation rate and latent heat of vaporization. If an invalid cycle is discarded due to validity verification within the window, this invalid cycle is not included in the selection of the beginning or end of the cycle, and the window automatically extends forward to the cycle containing the most recent valid reading, ensuring that the difference calculation is always based on valid data.

[0039] The explicit quantification of the latent heat of vaporization enables the subsequent heat balance equation to accurately separate the heat consumed by water evaporation from the heat generated by microbial metabolism, avoiding confusion between the two and causing systematic bias in the identification of heat production rate.

[0040] The heat generation identification module obtains the heat storage and convective heat dissipation items corresponding to the core temperature of the reactor. After adding the heat storage, convective heat dissipation, and latent heat of vaporization items, the actual output power is subtracted to obtain the heat generation rate equivalent. The heat generation rate equivalent is used to reflect the heat power generated by microbial metabolism.

[0041] The heat balance of the reactor core during each data collection cycle follows the law of energy conservation: the sum of heat generated by microbial metabolism and the heat input from the heating equipment equals the sum of the heat accumulated in the reactor core, the heat lost to the environment through convection, and the latent heat carried away by water evaporation. The only unmeasurable term in this conservation relationship, microbial heat generation, is designated as the variable to be identified.

[0042] The calculation of the heat storage term depends on the core temperature of the reactor core between two adjacent acquisition cycles. The temperature change of the core temperature between two adjacent acquisition cycles is obtained. This temperature change is divided by the acquisition cycle span, and the result is multiplied by the equivalent heat capacity to obtain the heat storage term. The temperature change is the core temperature of the current cycle minus the core temperature of the previous cycle; the acquisition cycle span is 5 minutes. The physical meaning of the heat storage term is the heat power accumulated or released by the reactor core per unit time due to temperature changes. The heat storage term is positive when the temperature rises and negative when the temperature falls.

[0043] The calculation of the convective heat dissipation term depends on the temperature difference between the core temperature of the reactor and the ambient temperature. The actual temperature difference between the core temperature and the current ambient temperature is obtained, and multiplied by a heat dissipation coefficient to arrive at the convective heat dissipation term. The ambient temperature is measured by an ambient temperature sensor within the fermentation tank. The convective heat dissipation term characterizes the continuous heat power lost from the reactor to the environment due to natural convection driven by the temperature difference and heat conduction through the tank walls; the greater the temperature difference, the stronger the heat dissipation.

[0044] In this embodiment, the equivalent heat capacity is calibrated to 180 kJ / °C, and the heat dissipation coefficient is calibrated to 0.12 kW / °C. Assume the current cycle core temperature is 52.3°C, the previous cycle was 52°C, the ambient temperature is 22°C, the actual output power of the heating equipment is 1.2 kW, and the latent heat of vaporization, calculated in the previous step, is 0.505 kW.

[0045] Calculation of heat storage term: The temperature change is 52.3-52=0.3 degrees Celsius, the temperature change rate is 0.3 / 5=0.06 degrees Celsius / min, and the heat storage term is (0.06×180) / 60=0.18kW.

[0046] Convection heat dissipation calculation: The actual temperature difference is 52.3 - 22 = 30.3 degrees Celsius, and the convection heat dissipation is 30.3 × 0.12 = 3.636 kW.

[0047] The sum of the values ​​for heat storage, convective heat dissipation, and latent heat of vaporization is calculated as: 0.18 + 3.636 + 0.505 = 4.321 kW. Subtracting the actual output power from this sum, 4.321 - 1.2 = 3.121 kW, yields the instantaneous heat production rate.

[0048] Because the temperature change in a single cycle is relatively small, sensor quantization noise significantly affects the calculation of the temperature change rate, resulting in periodic fluctuations in the instantaneous heat production rate. A moving average is calculated for multiple instantaneous heat production rates continuously acquired within a historical data smoothing window, and the average result is determined as the heat production rate equivalent. The historical data smoothing window consists of six consecutive acquisition cycles, corresponding to 30 minutes. This duration is on the same order of magnitude as the thermal inertia time constant of the reactor, effectively suppressing periodic fluctuations introduced by sensor noise, while avoiding delays in responding to changes in microbial heat production trends due to excessive smoothing. The window slides forward one unit each cycle, incorporating the latest value and discarding the earliest value, and the arithmetic mean is recalculated.

[0049] Let the instantaneous heat production rates of the most recent 6 cycles be 3.05, 3.12, 3.08, 3.15, 3.1, and 3.12 kW respectively. The arithmetic mean is (3.05+3.12+3.08+3.15+3.1+3.12) / 6=3.103kW, which is the heat production rate equivalent at the current moment.

[0050] The heat production rate equivalent explicitly identifies the heat generated by microbial metabolism from the pile heat balance equation, providing a quantitative basis for determining subsequent fermentation stages and adjusting gain. If the latent heat of vaporization term is omitted, the heat production rate equivalent in the above example will be underestimated by about 15.7%, and the deviation can reach more than 30% in the high-temperature stage of intense evaporation, directly affecting the accuracy of subsequent gain coefficient calculations.

[0051] The gain determination module performs linear fitting on the heat production rate equivalent of multiple historical sequences, uses the slope obtained from the fitting as the heat production rate trend value, and determines the current fermentation stage based on the heat production rate trend value; it obtains the quotient of the heat production rate equivalent and the preset full load equivalent, and subtracts the quotient from the constant 1 to obtain the adaptive gain coefficient.

[0052] The heat production rate equivalent exhibits an evolutionary pattern of first increasing, then stabilizing, and finally decreasing during fermentation, with its trend preceding changes in the core temperature of the fermentation pile. A historical sequence was constructed using the heat production rate equivalent output values ​​from the most recent 12 collection periods. Using the period number as the independent variable and the corresponding heat production rate equivalent as the dependent variable, a univariate linear fit was performed using the standard least squares method. The slope obtained from the fit was taken as the heat production rate trend value. Twelve periods correspond to one hour. Since the timescale of microbial population turnover is typically on the order of several hours, a one-hour window is sufficient to identify the inflection point of the heat production rate trend, while avoiding a window that is too long and would cause a lag in the response to stage transitions. A positive heat production rate trend value indicates that the microbial metabolic intensity is increasing, a negative value indicates that it is declining, and a value close to 0 indicates a stable state.

[0053] Based on the trend value of heat production rate, the current fermentation stage is determined, and the fermentation process is divided into three stages: the temperature rise activation period, the high temperature maintenance period, and the temperature drop decline period.

[0054] When the heat production rate trend value exceeds the first slope threshold, the fermentation stage is determined to be the temperature rise activation period. The microbial community enters the logarithmic growth phase, and metabolic heat production continues to rise. The heating equipment needs to gradually reduce its output to avoid temperature overshoot caused by superposition with biological heat production. The first slope threshold is set as follows: divide the rated power of the heating equipment by the equivalent heat capacity to obtain the full-power temperature rise rate, and then take 5% of that rate. In this embodiment, the rated power of the heating equipment is 3kW, the equivalent heat capacity is 180kJ / degree Celsius, the full-power temperature rise rate is 3 / (180 / 60) = 1 degree Celsius / min, and the first slope threshold is 1×5% = 0.05kW / cycle.

[0055] If the absolute value of the heat production rate trend value is not greater than the first slope threshold, the fermentation stage is determined to be a high-temperature maintenance period. Microbial heat production tends to stabilize, and the pile temperature fluctuates slightly within the target range. The heating equipment only needs to cover the difference between heat loss and microbial heat production.

[0056] When the heat production rate trend value is less than the second slope threshold, the fermentation stage is determined to be in a cooling decline phase. The second slope threshold is negative; in this embodiment, it is taken as -0.05 kW / cycle, and its absolute value is equal to the first slope threshold. As microbial activity decreases and metabolic heat production declines, the heating equipment needs to gradually increase its output power to maintain the target temperature.

[0057] The adaptive gain coefficient complements the output power of the heating equipment with the self-heating of the microorganisms. The quotient of the heat production rate equivalent and the preset full-load equivalent is obtained. The preset full-load equivalent is the sum of the rated power of the heating equipment and the steady-state heat dissipation power of the reactor core, representing the maximum total heat power required by the reactor core at the target temperature. In this embodiment, the rated power of the heating equipment is 3kW, the difference between the target temperature of 55 degrees Celsius and the ambient temperature of 22 degrees Celsius is 33 degrees Celsius, the steady-state heat dissipation power is 0.12 × 33 = 3.96kW, and the preset full-load equivalent is 3 + 3.96 = 6.96kW, rounded down to 7kW for engineering purposes. The quotient is equal to the current heat production rate equivalent divided by the preset full-load equivalent. Subtracting this quotient from the preset value of 1 yields the initial gain.

[0058] When the heat production rate equivalent is 3.103 kW, the quotient is 3.103 / 7 = 0.443, and the initial gain is 1 - 0.443 = 0.557. The microorganisms currently handle 44.3% of the total heat load, and the heating equipment only needs to compensate for the remaining 55.7%. The initial gain is 1 when the microbial heat production is 0; the initial gain approaches 0 when the microbial heat production is close to the preset full-load equivalent.

[0059] The initial gain is limited based on a set lower and upper limit to obtain a limited gain. The lower limit is set to 0.05, and the upper limit is set to 1. The lower limit of 0.05 ensures that the heating element is always in a hot standby state, and can respond immediately once the heat production of microorganisms declines, without response lag due to the actuator starting from a completely cold state. The upper limit of 1 corresponds to full-scale heating when there is no heat production from microorganisms. The initial gain is set to 0.05 when it is below 0.05, 1 when it is above 1, and remains unchanged when it is in between. The result of this operation is the limited gain.

[0060] Within the transition period when a fermentation stage switch is detected, a linear interpolation operation is performed between the historical limiting gain of the previous fermentation stage and the current limiting gain. The interpolation result is then used as the adaptive gain coefficient. The stage switch determination condition is that the fermentation stage identifier of the current cycle is different from that of the previous cycle. The transition period is set to 6 acquisition cycles, corresponding to 30 minutes, matching the thermal inertia time constant of the reactor core. The limiting gain value of the cycle preceding the stage switch is recorded as the interpolation starting point, and the limiting gain value calculated according to normal logic after the switch is used as the interpolation ending point. The transition is performed linearly at equal intervals within the 6 transition cycles. Assuming the limiting gain is 0.82 one week before the switching time, and the calculated limiting gain for the first cycle after the switching is 0.55, the adaptive gain coefficient for the first cycle during the transition period is 0.82 + (0.55 - 0.82) × (1 / 6) = 0.775, for the second cycle it is 0.82 + (0.55 - 0.82) × (2 / 6) = 0.73, and it reaches 0.55 in the sixth cycle. After the transition ends, the adaptive gain coefficient is directly equal to the limiting gain, and it is updated synchronously with the heat generation rate equivalent update in each acquisition cycle.

[0061] The stage determination is based on the trend of heat generation rate change rather than the absolute temperature value, allowing the control strategy to gain a time lead and adjust the heating direction in advance before the temperature changes significantly. The limiting and smooth transition mechanism ensures the continuity of actuator action and prevents temperature oscillations caused by sudden power changes during stage switching.

[0062] The component calculation module subtracts the heat generation rate equivalent from the convective heat dissipation term to obtain the feedforward component, and uses a proportional-integral controller to process the temperature deviation between the preset target temperature and the core temperature of the reactor to obtain the feedback component.

[0063] The feedforward component predicts heating demand based on the current heat balance. The convective heat loss term represents the heat power lost by the reactor to the environment at the current moment, and the heat production rate equivalent represents the metabolic heat production power of microorganisms at the current moment. The feedforward component is obtained by subtracting the heat production rate equivalent from the convective heat loss term. When the feedforward component is positive, heat loss exceeds heat production, and the heating equipment needs to make up the heat gap; when it is negative, the heat production of microorganisms has exceeded the heat loss, and the reactor has a tendency to heat up autonomously. At the current moment, the convective heat loss term is 3.636 kW, the heat production rate equivalent is 3.103 kW, and the feedforward component is 3.636 - 3.103 = 0.533 kW.

[0064] The feedback component is used to eliminate residual biases in feedforward compensation and temperature drift caused by external disturbances. The temperature deviation is obtained by subtracting the core temperature of the reactor from the preset target temperature. The preset target temperature is set according to the fermentation process requirements; in this embodiment, it is 55 degrees Celsius, corresponding to the center value of the optimal activity temperature range for the main functional microbial communities in the herbal fermentation. A positive temperature deviation indicates that the current temperature is lower than the target value and heating is required, while a negative deviation indicates that the temperature is higher than the target value.

[0065] The proportional control term is obtained by multiplying the temperature deviation by a set proportionality coefficient. The set proportionality coefficient is the rated power of the heating equipment divided by the maximum allowable temperature deviation: Rated power is 3kW, maximum allowable temperature deviation is 5 degrees Celsius, and the set proportionality coefficient is 3 / 5 = 0.6kW / degree Celsius. This value ensures that the proportional control term outputs full power when the temperature deviation reaches 5 degrees Celsius. When the core temperature of the reactor is 52.3 degrees Celsius, the temperature deviation is 55 - 52.3 = 2.7 degrees Celsius, and the proportional control term is 2.7 × 0.6 = 1.62kW.

[0066] The temperature deviation is integrally accumulated, and the result is multiplied by a set integral coefficient to obtain the integral control term. The integral accumulation is the sum of the temperature deviations over each historical period. The set integral coefficient is one-tenth of the set proportional coefficient divided by the reactor thermal inertia time constant: (0.6 × 0.1) / 30 = 0.002 kW / (°C·min). The cumulative contribution of the integral term within one thermal inertia time constant does not exceed one-tenth of the proportional term to prevent integral saturation and overshoot. The integral control term equals the integral accumulation result multiplied by the set integral coefficient and then multiplied by the acquisition period span.

[0067] The proportional-integral controller is equipped with an integral anti-saturation mechanism. When the target output power reaches the rated upper limit and the temperature deviation is positive, or when the target output power reaches the power output lower limit and the temperature deviation is negative, the accumulation of the temperature deviation is stopped to freeze the integral control term. The rated upper limit is 3kW, and the power output lower limit is 0. The integral term accumulates normally only when the target output power is between the upper and lower limits.

[0068] The feedback component is obtained by summing the proportional control term and the integral control term. Assuming the current cumulative integral result is 40 degrees Celsius per minute after several cycles, the integral control term is 40 × 0.002 = 0.08 kW, and the feedback component is 1.62 + 0.08 = 1.7 kW.

[0069] The feedforward component predicts the heat gap based on the heat production rate equivalent, allowing the feedback loop to handle only the residual deviation caused by identification errors and external disturbances, thus reducing the regulation burden and overshoot risk. The integral freeze mechanism prevents the integral value from accumulating indefinitely during actuator saturation, ensuring that the control output quickly returns to normal regulation state after the limit is released.

[0070] The power output module sums the feedforward and feedback components, multiplies them by an adaptive gain coefficient to generate the target output power, and uses the target output power to control the heating equipment.

[0071] Summing the feedforward and feedback components: 0.533 + 1.7 = 2.233 kW. Multiplying the sum by the adaptive gain coefficient of 0.557, the target output power is generated: 2.233 × 0.557 = 1.244 kW. The adaptive gain coefficient acts on both components simultaneously. During the vigorous stage of microbial heat production, the gain coefficient is small, and the heating output automatically decreases; during the stage of declining microbial heat production, the gain coefficient is large, and the heating equipment quickly fills the heat gap.

[0072] The target output power is sent to the heating equipment after being limited, with a lower limit of 0 and an upper limit of 3kW (rated power).

[0073] After receiving control commands, the heating equipment performs a safety boundary constraint check. In response to the core temperature not falling below the set safety upper limit temperature, prior calculations are ignored, and the control commands sent to the heating equipment are forcibly reset to zero until the core temperature falls back to the safe recovery range. The safety upper limit temperature is set at 65 degrees Celsius, which is the critical temperature at which the survival rate of the main functional microbial communities in the fermentation of medicinal materials drops sharply. The safe recovery range is 3 degrees Celsius below the safety upper limit temperature, meaning that normal control output resumes when the core temperature falls below 62 degrees Celsius. This 3-degree Celsius hysteresis bandwidth prevents frequent switching of the heating equipment when the temperature repeatedly crosses the safety upper limit.

[0074] Moisture content sensors may experience intermittent failures in high-temperature and high-humidity environments. To address this, a tiered degradation strategy is implemented. In response to multiple consecutive invalid volumetric moisture content acquisition cycles, the latent heat of vaporization is calculated using the most recent valid evaporation rate. In this embodiment, the number of consecutive invalid cycles is determined to be 3, corresponding to 15 minutes. The variation in the evaporation process within 15 minutes is limited, and the calculation deviation introduced by this strategy is within an acceptable range. In response to the duration of invalid volumetric moisture content reaching the degradation limit, the latent heat of vaporization is forcibly set to 0, and a sensor maintenance prompt is triggered. The degradation limit is set to 12 acquisition cycles corresponding to 1 hour. After 1 hour, the most recent valid evaporation rate can no longer represent the current true evaporation state. Setting the latent heat of vaporization to zero degenerates the heat balance equation into a simplified mode without evaporation correction, resulting in a decrease in the accuracy of heat production rate identification. However, basic closed-loop control can still be maintained based on temperature feedback.

[0075] After each control cycle is completed, the core temperature, volumetric water content, evaporation rate, latent heat of vaporization, heat storage, convective heat dissipation, heat production rate equivalent, heat production rate trend, fermentation stage identifier, adaptive gain coefficient, feedforward component, feedback component, and target output power of the current cycle are written into the data record. After recording is completed, the system waits for the next acquisition cycle to begin, and then restarts the acquisition of core temperature and volumetric water content, forming a complete control closed loop with a cycle of 5 minutes.

[0076] like Figure 2As shown, the horizontal axis represents fermentation time, and the vertical axes represent the heat production rate equivalent and the target output power, respectively. Two vertical dashed lines divide the fermentation process into three stages: the heating activation period, the high-temperature maintenance period, and the cooling decline period. The heat production rate equivalent curve, represented by a solid line, rapidly increases from a low value to approximately 3.8 kW during the heating activation period, fluctuates slightly between 3.5 and 4.0 kW during the high-temperature maintenance period, and gradually declines during the cooling decline period. The target output power curve, represented by a dashed line, shows a complementary trend to the heat production rate equivalent: during the heating activation period, the target output power rapidly decreases from approximately 1.8 kW to around 0.3 kW; during the high-temperature maintenance period, it maintains a low output level between 0.3 and 0.5 kW; and during the cooling decline period, it gradually increases from 0.5 kW to approximately 1.5 kW. The inverse relationship between the two curves throughout the entire cycle indicates that the adaptive gain coefficient enables dynamic complementarity between the heating equipment output power and the heat production from microbial metabolism.

[0077] The output of the heating equipment dynamically complements the heat production of microorganisms at all stages of the fermentation cycle. The safety boundary constraint, as the final protection layer independent of the control algorithm, keeps the pile temperature within the range where microorganisms can survive. The sensor degradation strategy ensures that basic control capabilities are still available even when some monitoring devices fail.

[0078] The above are all preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Therefore, all equivalent changes made in accordance with the structure, shape and principle of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A temperature and humidity adaptive and coordinated control system for a medicinal herb fermentation tank, characterized in that, include: The parameter acquisition module collects the core temperature and volumetric water content of the reactor body, and obtains the actual output power of the heating equipment. The latent heat calculation module calculates the average rate of decrease in volumetric water content within a preset time window as the evaporation rate, and multiplies the evaporation rate by a preset vaporization constant to obtain the latent heat of evaporation. The heat generation identification module obtains the heat storage and convective heat dissipation items corresponding to the core temperature of the reactor body. After adding the heat storage, convective heat dissipation, and latent heat of vaporization items, the actual output power is subtracted to obtain the heat generation rate equivalent. The heat generation rate equivalent is used to reflect the heat power generated by microbial metabolism. The gain determination module performs linear fitting on the heat production rate equivalent of multiple historical sequences, uses the slope obtained from the fitting as the heat production rate trend value, and determines the current fermentation stage based on the heat production rate trend value; it obtains the quotient of the heat production rate equivalent and the preset full load equivalent, and subtracts the quotient from the constant 1 to obtain the adaptive gain coefficient. The component calculation module subtracts the heat generation rate equivalent from the convective heat dissipation term to obtain the feedforward component, and uses a proportional-integral controller to process the temperature deviation between the preset target temperature and the core temperature of the reactor to obtain the feedback component. The power output module sums the feedforward and feedback components, multiplies them by an adaptive gain coefficient to generate the target output power, and uses the target output power to control the heating equipment.

2. The adaptive and coordinated temperature and humidity control system for the medicinal herb fermentation tank according to claim 1, characterized in that, The core temperature and volumetric water content of the reactor were collected, including: Obtain the first detected value of the core temperature and the second detected value of the volumetric water content of the reactor body for the current cycle; In response to the first detection value being within a first preset range and the second detection value being within a second preset range, the first detection value is determined as the core temperature of the reactor body, and the second detection value is determined as the volumetric water content; In response to the first detection value exceeding the first preset range, or the second detection value exceeding the second preset range, the first and second detection values ​​of the current period are removed and an abnormal alarm is triggered.

3. The adaptive and coordinated temperature and humidity control system for the medicinal herb fermentation tank according to claim 1, characterized in that, Before collecting the core temperature and volumetric water content of the reactor, the following steps are also included: With the core temperature of the reactor body in the initial temperature range before biological heat generation occurs, the heating equipment is driven by constant test power to run for a calibration time. The temperature rise curve within the calibration time is obtained. Based on the correspondence between the slope of the temperature rise curve and the constant test power, the equivalent heat capacity and heat dissipation coefficient corresponding to the core temperature of the reactor are calibrated.

4. The adaptive and coordinated temperature and humidity control system for the medicinal herb fermentation tank according to claim 1, characterized in that, The calculation of the average rate of decrease in volumetric water content within a preset time window as the evaporation rate includes: Obtain a sliding smooth window containing a specific number of acquisition cycles; Calculate the difference in volumetric water content between the first and last periods within the sliding smoothing window; Divide the difference by the time span of the sliding smoothing window to obtain the evaporation rate.

5. The adaptive and coordinated temperature and humidity control system for the medicinal herb fermentation tank according to claim 3, characterized in that, The heat storage and convective heat dissipation terms corresponding to the core temperature of the reactor include: Obtain the temperature change of the core temperature of the reactor body between two adjacent acquisition cycles; Divide the temperature change by the data acquisition period, and multiply the result of the division by the equivalent heat capacity to obtain the heat storage term; Obtain the actual temperature difference between the core temperature of the reactor and the current ambient temperature, and multiply the actual temperature difference by the heat dissipation coefficient to obtain the convective heat dissipation term.

6. The adaptive and coordinated temperature and humidity control system for the medicinal herb fermentation tank according to claim 1, characterized in that, Adding the heat storage term, convective heat dissipation term, and latent heat of vaporization term, and subtracting the actual output power, yields the equivalent heat production rate, which includes: Calculate the sum of the values ​​of the heat storage term, the convective heat dissipation term, and the latent heat of vaporization term; Subtract the actual output power from the sum of the numerical values ​​to obtain the instantaneous heat generation rate; The moving average of multiple instantaneous heat production rates continuously acquired within the historical data smoothing window is calculated, and the mean calculation result is determined as the heat production rate equivalent.

7. The adaptive and coordinated temperature and humidity control system for the medicinal herb fermentation tank according to claim 1, characterized in that, The current fermentation stage is determined based on the trend value of heat production rate, including: When the heat production rate trend value is greater than the first slope threshold, the fermentation stage is determined to be the temperature rise activation period. If the absolute value of the heat production rate trend value is not greater than the first slope threshold, the fermentation stage is determined to be a high-temperature maintenance period. When the heat production rate trend value is less than the second slope threshold, the fermentation stage is determined to be the cooling decline period, where the second slope threshold is a negative number.

8. The adaptive and coordinated temperature and humidity control system for the medicinal herb fermentation tank according to claim 1, characterized in that, Subtracting the quotient from the preset value of 1 yields the adaptive gain coefficient, which includes: The initial gain is determined by subtracting the quotient from the preset value of 1. The initial gain is subjected to a limiting process based on a set lower limit and a set upper limit to obtain a limited gain; During the transition period when the fermentation stage is switched, a linear interpolation operation is performed on the historical limiting gain of the previous fermentation stage and the current limiting gain, and the interpolation result is determined as the adaptive gain coefficient.

9. The adaptive and coordinated temperature and humidity control system for the medicinal herb fermentation tank according to claim 1, characterized in that, The feedback components obtained by processing the temperature deviation between the preset target temperature and the core temperature of the reactor using a proportional-integral controller include: The temperature deviation is obtained by subtracting the core temperature of the reactor from the preset target temperature. Multiply the temperature deviation by the set proportional coefficient to obtain the proportional control term; The temperature deviation is integrally accumulated, and the integral accumulation result is multiplied by a set integral coefficient to obtain the integral control term; When the target output power reaches the rated upper limit and the temperature deviation is positive, or when the target output power reaches the power output lower limit and the temperature deviation is negative, stop accumulating the temperature deviation to freeze the integral control term; The feedback component is obtained by summing the proportional control term and the integral control term.

10. The adaptive and coordinated temperature and humidity control system for the medicinal herb fermentation tank according to claim 1, characterized in that, After controlling the heating equipment using the target output power, it also includes: In response to the reactor core temperature not being lower than the set safe upper limit temperature, the pre-calculation is ignored, and the control command sent to the heating equipment is forcibly reset to zero until the reactor core temperature drops back to the safe recovery range. In response to multiple consecutive invalid sampling cycles of volumetric water content, the latent heat of vaporization is calculated using the most recent valid evaporation rate. In response to the duration of invalid volumetric moisture content reaching the degradation limit, the latent heat of vaporization is forcibly set to 0, and a sensor maintenance prompt is triggered.

Images