Cyclophosphamide and fludarabine conditioning regimen for allogeneic hematopoietic stem cell transplantation in patients with myelodysplastic syndrome

Abstract

The present application relates to the technical field of intelligent control and optimization of food sterilization process, and discloses a cooling water circulation parameter self-tuning optimization method and system for bird's nest sterilization kettle based on fuzzy logic, which iterates the center temperature of bird's nest through the dynamic time constant of cooling water flow, and calculates the instantaneous quality deterioration rate and cumulative quality damage index by coupling calculation of thermal damage and shear stress; the temperature deviation and change rate are substituted into the two-dimensional fuzzy rule base, and the flow and temperature correction values are obtained by defuzzification according to the rule activation degree and dynamic weight; the rule weight is updated by calculating the performance reward value based on the temperature deviation improvement and the quality deterioration rate; the flow and temperature proportional factors are adjusted based on the cumulative quality damage index, and the flow proportional factor is saturated and compressed based on the shear stress index; the dynamic weight and proportional factor at the end point are saved as the initial parameters of the next batch. The present application realizes the collaborative optimization of cooling efficiency, bird's nest quality maintenance and shear safety constraint, and improves the production control efficiency and product consistency of multiple batches.

Description

Technical Field

[0001] This invention relates to the field of intelligent control and optimization technology for food sterilization processes, specifically to a method and system for self-tuning and optimizing cooling water circulation parameters in bird's nest sterilization kettles based on fuzzy logic. Background Technology

[0002] During the cooling stage after high-temperature sterilization, the cooling rate of ready-to-eat bird's nest affects the final quality of the product. If the cooling rate is too low, the bird's nest remains at a high temperature, prolonging the protein denaturation and mucopolysaccharide dissolution process. If the cooling water flow rate is too high or the flow disturbance is too strong, the fluid shear force on the bird's nest inside the bottle increases, and the bird's nest strands may break or crumble.

[0003] Existing cooling water circulation control for sterilization autoclaves typically employs fixed-parameter proportional-integral-derivative (PID) control or fixed-rule fuzzy control. The tuning of circulating water pump frequency, hot-cold water mixing valve opening, cooling water flow rate, and cooling water temperature setpoints largely relies on manual experience. Some control methods adjust solely based on the temperature deviation within the autoclave, failing to incorporate the bird's nest center temperature, instantaneous quality deterioration rate, cumulative quality damage index, and shear stress index into the rule weight update and proportional factor adjustment process.

[0004] Therefore, this application provides a method and system for self-tuning and optimizing the cooling water circulation parameters of a bird's nest sterilization autoclave based on fuzzy logic to solve the above-mentioned technical problems. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for self-tuning and optimizing the cooling water circulation parameters of a bird's nest sterilization autoclave based on fuzzy logic, in order to solve the technical problem that the cooling water circulation control parameters in the prior art cannot be self-tuned and optimized based on the bird's nest center temperature, instantaneous quality deterioration rate, cumulative quality damage index and shear stress index.

[0006] To address the aforementioned technical problems, this invention provides a self-tuning optimization method for cooling water circulation parameters in a bird's nest sterilization autoclave based on fuzzy logic, comprising: Collect the temperature inside the reactor and the flow rate of the cooling water; Substituting the temperature inside the autoclave and the center temperature of the bird's nest in the previous cycle into the dynamic time constant of the flow rate, the center temperature of the bird's nest is obtained iteratively. The center temperature of the bird's nest is then substituted into the thermal damage calculation. The cooling water flow rate, the shelf structure inside the sterilization autoclave, and the bottle arrangement parameters are substituted into the shear stress calculation to obtain the instantaneous quality deterioration rate, the cumulative quality damage index, and the shear stress index. Input the temperature deviation and the rate of change of temperature deviation into the two-dimensional fuzzy rule base, defuzzify the flow correction value and temperature correction value by weighted average according to the rule activation degree and dynamic weight, and calculate the flow set value and temperature set value by the outer scale factor and output them. In the next cycle after the control is applied, the performance bonus value is calculated based on the absolute improvement of temperature deviation, the reference quality degradation rate and the measured instantaneous quality degradation rate, and the dynamic weights are updated, normalized and trimmed. The flow rate proportional factor is increased and the temperature proportional factor is decreased based on the cumulative quality damage index, and saturated compression is performed on the flow rate proportional factor based on the shear stress index. When the increase in the center temperature and cumulative quality damage index of the bird's nest meets the cooling endpoint conditions, the cooling water circulation is shut off and the dynamic weight and proportional factor are saved. If the cooling endpoint conditions are not met, the temperature inside the vessel and the cooling water flow rate are collected again.

[0007] In some specific embodiments, the temperature inside the pot and the center temperature of the bird's nest in the previous cycle are substituted into the dynamic time constant of the flow rate to iteratively obtain the center temperature of the bird's nest, further including: The cooling water flow rate is converted into a dynamic time constant of the flow rate using a negative exponential relationship; Calculate the temperature difference between the temperature inside the pot and the center temperature of the bird's nest in the previous cycle; Multiply the temperature difference by the negative exponential complement of the ratio obtained by dividing the control cycle by the flow dynamic time constant, and add it to the bird's nest center temperature of the previous cycle to obtain the bird's nest center temperature.

[0008] In some specific embodiments, the center temperature of the bird's nest is substituted into the thermal damage calculation, and the cooling water flow rate, the shelf structure inside the sterilization autoclave, and the bottle arrangement parameters are substituted into the shear stress calculation to obtain the instantaneous quality deterioration rate, the cumulative quality damage index, and the shear stress index, further including: The shear stress index is calculated based on the cooling water flow rate, resistance coefficient, cooling water density, and characteristic hydraulic diameter. The temperature of the bird's nest center was converted from Celsius to absolute temperature, and the absolute temperature, shear stress index, protein denaturation activation energy, pre-exponential factor, shear damage coupling coefficient and shear damage index were substituted into the thermal damage calculation to obtain the instantaneous quality deterioration rate. The instantaneous quality degradation rate of each cycle is multiplied by the control cycle and then recursively accumulated. An upper limit is applied to the accumulated value to obtain the cumulative quality damage index.

[0009] In some specific embodiments, the temperature deviation and the rate of change of temperature deviation are input into a two-dimensional fuzzy rule base, and the flow correction value and temperature correction value are defuzzified by weighted average according to the rule activation degree and dynamic weight, which further includes: The difference between the temperature inside the vessel and the target final temperature is defined as the temperature deviation, and the difference between the temperature deviations of adjacent cycles is divided by the control cycle to determine the temperature deviation change rate. The temperature deviation and the rate of change of temperature deviation are divided into multiple language ranges from negative to positive, and combined to form multiple rules. The activation degree of a rule is obtained by performing an algebraic product on the membership degrees of its two antecedents. By weighting the rule consequent level with both rule activation degree and dynamic weight, the traffic correction value and temperature correction value are obtained respectively.

[0010] In some specific embodiments, calculating and outputting the flow setpoint and temperature setpoint using an outer scaling factor further includes: The flow rate setpoint is obtained by adding the product of the base cooling water flow rate, the flow rate ratio factor, and the flow rate correction value. The temperature setpoint is obtained by adding the product of the base cooling water temperature, the temperature scaling factor, and the temperature correction value. The flow rate setpoint is sent to the frequency converter of the circulating water pump, and the temperature setpoint is sent to the controller of the hot and cold water mixing valve.

[0011] In some specific embodiments, in the next cycle after the control variable is applied, a performance reward value is calculated based on the absolute improvement in temperature deviation, the reference quality degradation rate, and the measured instantaneous quality degradation rate. The dynamic weights are then updated, normalized, and trimmed. This further includes: For rules whose activation degree is greater than the preset activation threshold, calculate the difference between the absolute value of the temperature deviation between the current cycle and the next cycle, and determine the sign of the difference; Multiply the difference sign by the temperature deviation improvement weight, and subtract the instantaneous quality degradation rate of the next cycle from the reference quality degradation rate, then multiply by the quality degradation suppression weight, and add the two to obtain the performance bonus value. The updated weights are obtained by adding the dynamic weights to the product of the learning rate and the performance reward value; Normalize all update weights, and reset update weights below the preset pruning threshold to zero before normalizing them; Rules whose activation degree is not greater than the preset activation threshold retain their current dynamic weight.

[0012] In some specific embodiments, the flow rate scaling factor is increased and the temperature scaling factor is decreased based on the cumulative quality damage index, and saturation compression of the flow rate scaling factor is performed based on the shear stress index, further including: Input the ratio of the cumulative quality damage index to the cumulative damage threshold into the hyperbolic tangent function; Multiply the hyperbolic tangent function output by the cumulative damage gain coefficient and add it to the flow base scaling factor; The saturation function input is determined based on the excess amount of the shear stress index exceeding the critical shear stress index. When the shear stress index does not exceed the critical shear stress index, the excess amount is set to zero. The compression flow rate ratio factor is output based on the saturation function. The temperature scaling factor is obtained by multiplying the output of the hyperbolic tangent function by the temperature suppression coefficient and then subtracting it from the temperature scaling factor.

[0013] In some specific embodiments, when the central temperature of the bird's nest and the incremental cumulative quality damage index meet the cooling endpoint condition, the cooling water circulation is shut off while the dynamic weights and scaling factors are preserved, further including: The cooling endpoint is defined as the point where the center temperature of the bird's nest is lower than or equal to the sum of the target final temperature and the preset temperature difference, and the cumulative increase in the quality damage index over the past preset period is less than the preset increment threshold. Turn off the cooling water circulation at the end of the cooling process; Extract the normalized and trimmed dynamic weights, flow rate scaling factor, and temperature scaling factor of this batch, write them into the edge database, and read them as initial parameters when the next batch begins cooling.

[0014] In some specific embodiments, updating, normalizing, and pruning dynamic weights further includes: The current performance reward value is added to the performance reward value within the preset historical control period window according to the decreasing weight over time to obtain the cumulative reward value of the rule; When the cumulative reward value of a rule is negative and the shear stress exponent exceeds the preset shear stress warning value, the learning rate of the rule is reduced and the weights are calculated and updated. When the cumulative reward value of the rule is positive and the cumulative quality damage index exceeds the cumulative damage threshold, the update weight of the flow correction direction rule is increased and the update weight of the temperature correction direction rule is decreased according to the signs of the flow correction value and temperature correction value, respectively. Normalize the updated weights and perform pruning.

[0015] Based on the same concept, this invention also provides a fuzzy logic-based self-tuning and optimization system for cooling water circulation parameters in a bird's nest sterilization autoclave, comprising: The data acquisition module is configured to collect the temperature inside the vessel and the flow rate of the cooling water. The quality and shear calculation module is configured to substitute the temperature inside the autoclave and the center temperature of the bird's nest in the previous cycle into the dynamic time constant of the flow rate, iterate to obtain the center temperature of the bird's nest, substitute the center temperature of the bird's nest into the thermal damage calculation, and substitute the cooling water flow rate, the shelf structure inside the sterilization autoclave and the bottle arrangement parameters into the shear stress calculation to obtain the instantaneous quality deterioration rate, the cumulative quality damage index and the shear stress index. The fuzzy control module is configured to input the temperature deviation and the rate of change of temperature deviation into a two-dimensional fuzzy rule base, defuzzify the flow correction value and temperature correction value by weighted average according to the rule activation degree and dynamic weight, and calculate and output the flow setpoint and temperature setpoint by the outer scale factor. The weight tuning module is configured to calculate the performance reward value based on the absolute improvement of temperature deviation, the reference quality degradation rate, and the measured instantaneous quality degradation rate in the next cycle after the control quantity is applied, and update, normalize, and trim the dynamic weights. The proportional factor adjustment module is configured to increase the flow proportional factor and decrease the temperature proportional factor based on the cumulative quality damage index, and to perform saturated compression on the flow proportional factor based on the shear stress index. The batch management module is configured to shut down the cooling water circulation and save the dynamic weights and proportional factors when the central temperature of the bird's nest and the cumulative quality damage index increment meet the cooling endpoint conditions; otherwise, it will return to collecting the temperature inside the vessel and the cooling water flow rate.

[0016] Compared with existing technologies, its advantages are as follows: This invention discloses a self-tuning optimization method and system for cooling water circulation parameters in a bird's nest sterilization autoclave based on fuzzy logic. The method iteratively calculates the bird's nest center temperature by considering the autoclave temperature, cooling water flow rate, and flow rate dynamic time constant. The bird's nest center temperature, shear stress index, and thermal damage calculation relationship are then used together to form the instantaneous quality deterioration rate and cumulative quality damage index. This allows the cooling control process to obtain control inputs related to the thermal damage of the bird's nest and the fluid shearing action, thereby solving the problem in the prior art where adjustments are made solely based on the autoclave temperature deviation, which fails to reflect changes in the internal quality state of the bird's nest.

[0017] The activation degree of the rule is formed by a two-dimensional fuzzy rule base based on temperature deviation and the rate of change of temperature deviation. Dynamic weights are added to the weighted average defuzzification process of flow correction value and temperature correction value. The improvement of absolute value of temperature deviation after the control variable is applied, the reference quality degradation rate and the measured instantaneous quality degradation rate together form the performance reward value. The dynamic weights participate in the rule calculation of subsequent control cycles through the update, normalization and pruning process, so that the fuzzy rule weights participating in the current cooling state can be continuously updated and optimized with temperature changes and quality degradation state, thereby improving the adaptability of cooling water circulation parameters to different operating conditions.

[0018] The flow rate proportional factor and temperature proportional factor are adjusted based on the cumulative quality damage index, and the flow rate proportional factor is saturated and compressed based on the shear stress index. This ensures that the formation process of the flow rate setpoint and temperature setpoint simultaneously includes the cooling process, quality damage, and shear stress conditions. As a result, the heat exchange regulation capability is enhanced when the cumulative quality damage increases, and the flow rate regulation range is automatically limited when the shear stress approaches the critical value, thus balancing cooling efficiency, bird's nest quality maintenance, and mechanical safety constraints.

[0019] When the temperature at the center of the bird's nest and the cumulative quality damage index increment meet the cooling endpoint conditions, the cooling water circulation is shut off, and the dynamic weight, flow rate ratio factor, and temperature ratio factor are saved. The saved parameters are read as the initial parameters for the next batch, so that a parameter inheritance process is formed between batches. This reduces the time and energy consumption of repeated tuning between batches, and improves the control efficiency and product consistency of multi-batch continuous production. Attached Figure Description

[0020] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a flowchart illustrating some specific embodiments of the self-tuning optimization method for cooling water circulation parameters of the bird's nest sterilization autoclave based on fuzzy logic of the present invention; Figure 2 This is a schematic diagram of the structure of the self-tuning optimization system for cooling water circulation parameters of the bird's nest sterilization autoclave based on fuzzy logic in some specific embodiments of the present invention; Detailed Implementation To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to limit the application. The singular forms "a," "the," and "the" as used in the embodiments of this application are also intended to include the plural forms, unless the context clearly indicates otherwise, and "multiple" generally includes at least two.

[0022] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0023] It should be understood that although the terms first, second, third, etc., may be used in the embodiments of this application, these descriptions should not be limited to these terms. These terms are only used to distinguish the descriptions. For example, first may also be referred to as second without departing from the scope of the embodiments of this application, and similarly, second may also be referred to as first.

[0024] Depending on the context, the words “if” or “suppose” as used here can be interpreted as “when” or “in response to determination” or “in response to detection.” Similarly, depending on the context, the phrases “if determination” or “if detection (of the stated condition or event)” can be interpreted as “when determination” or “in response to determination” or “when detection (of the stated condition or event)” or “in response to detection (of the stated condition or event).”

[0025] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or device that includes said element.

[0026] It should be noted that any symbols and / or numbers present in the specification that are not marked in the accompanying drawings are not reference numerals.

[0027] Reference Figure 1 A self-tuning optimization method for cooling water circulation parameters in a bird's nest sterilization autoclave based on fuzzy logic includes: S101, collects the temperature inside the vessel and the flow rate of cooling water; S102, substitute the temperature inside the reactor and the center temperature of the bird's nest in the previous cycle into the dynamic time constant of the flow rate, iterate to obtain the center temperature of the bird's nest, and substitute the center temperature of the bird's nest into the thermal damage calculation. Substitute the cooling water flow rate, the shelf structure inside the sterilization reactor and the bottle arrangement parameters into the shear stress calculation to obtain the instantaneous quality deterioration rate, the cumulative quality damage index and the shear stress index. S103, input the temperature deviation and the rate of change of temperature deviation into the two-dimensional fuzzy rule base, defuzzify the flow correction value and temperature correction value by weighted average according to the rule activation degree and dynamic weight, calculate the flow set value and temperature set value by the outer layer scaling factor and output them; S104, in the next cycle after the control quantity is applied, calculate the performance bonus value based on the absolute value improvement of temperature deviation, the reference quality degradation rate and the measured instantaneous quality degradation rate, and update, normalize and trim the dynamic weights. S105, based on the cumulative quality damage index, increases the flow rate proportional factor and decreases the temperature proportional factor, and performs saturated compression on the flow rate proportional factor based on the shear stress index. S106: When the increase in the center temperature of the bird's nest and the cumulative quality damage index meet the cooling endpoint conditions, the cooling water circulation is turned off and the dynamic weight and proportional factor are saved. If the cooling endpoint conditions are not met, the temperature inside the collection vessel and the cooling water flow rate are returned.

[0028] To provide a clearer explanation, the steps in the embodiments of the present invention are described in detail below: S101, collects the temperature inside the vessel and the flow rate of cooling water; In this embodiment, the temperature inside the vessel is collected by an internal temperature sensor. Cooling water flow rate is collected by an electromagnetic flow meter. The cooling water flow signal is subjected to a first-order low-pass filter with a time constant of 0.5 seconds to filter out short-term fluctuations caused by the pulsation of the circulating pump, and then written into the flow buffer of the current control cycle.

[0029] In this embodiment, the cooling water inlet temperature is also collected simultaneously. Cooling water outlet temperature and pressure inside the vessel The cooling water inlet temperature, cooling water outlet temperature, and reactor pressure are used to record the operating conditions of the cooling circuit, and trigger sensor data verification when the sensor readings exceed the preset process range. The verified reactor temperature is... and cooling water flow rate Proceed to subsequent processing.

[0030] S102, substitute the temperature inside the reactor and the center temperature of the bird's nest in the previous cycle into the dynamic time constant of the flow rate, iterate to obtain the center temperature of the bird's nest, and substitute the center temperature of the bird's nest into the thermal damage calculation. Substitute the cooling water flow rate, the shelf structure inside the sterilization reactor and the bottle arrangement parameters into the shear stress calculation to obtain the instantaneous quality deterioration rate, the cumulative quality damage index and the shear stress index. Furthermore, by substituting the temperature inside the pot and the center temperature of the bird's nest from the previous cycle into the dynamic time constant of the flow rate, the center temperature of the bird's nest is obtained iteratively, including: The cooling water flow rate is converted into a dynamic time constant of the flow rate using a negative exponential relationship; Calculate the temperature difference between the temperature inside the pot and the center temperature of the bird's nest in the previous cycle; Multiply the temperature difference by the negative exponential complement of the ratio obtained by dividing the control cycle by the flow dynamic time constant, and add it to the bird's nest center temperature of the previous cycle to obtain the bird's nest center temperature.

[0031] Furthermore, by substituting the center temperature of the bird's nest into the thermal damage calculation, and by substituting the cooling water flow rate, the shelf structure inside the sterilizer, and the bottle arrangement parameters into the shear stress calculation, the instantaneous quality deterioration rate, the cumulative quality damage index, and the shear stress index are obtained, including: The shear stress index is calculated based on the cooling water flow rate, resistance coefficient, cooling water density, and characteristic hydraulic diameter. The temperature of the bird's nest center was converted from Celsius to absolute temperature, and the absolute temperature, shear stress index, protein denaturation activation energy, pre-exponential factor, shear damage coupling coefficient and shear damage index were substituted into the thermal damage calculation to obtain the instantaneous quality deterioration rate. The instantaneous quality degradation rate of each cycle is multiplied by the control cycle and then recursively accumulated. An upper limit is applied to the accumulated value to obtain the cumulative quality damage index.

[0032] In this embodiment, the center temperature of the bird's nest is iteratively determined using a non-steady-state lumped parameter relationship of thermal conduction. The temperature inside the pot during the current control cycle is then... The core temperature of the bird's nest in the previous cycle Control cycle and flow dynamic time constant Substitute the values ​​into the center temperature calculation relationship.

[0033] Flow dynamic time constant Calculated based on the negative exponential relationship of cooling water flow rate: ; in, The flow dynamic time constant related to the cooling water flow rate. For cooling water flow rate, , , This is a constant obtained through a heat permeation test of bird's nest bottle packaging. In this embodiment, Take 12.5. Take 0.8, Take 6.2.

[0034] The core temperature of bird's nest is obtained using the following formula: ; in, For the first The core temperature of the bird's nest is controlled over a specific cycle. This refers to the center temperature of the bird's nest from the previous control cycle. The temperature inside the vessel during the current control cycle. To control the cycle, This is the dynamic time constant of the current cooling water flow rate.

[0035] At the end of sterilization and heat preservation, the initial temperature at the center of the bird's nest is taken as the temperature inside the pot at the end of the heat preservation process. Calculate the temperature inside the pot. Compared to the previous cycle, the center temperature of the bird's nest The temperature difference between them is used to calculate the control cycle. With the dynamic time constant of flow The ratio of the temperature difference to the negative exponent is used to supplement the difference. The temperature difference is then multiplied by the negative exponent and added to the center temperature of the bird's nest from the previous cycle to form the center temperature of the bird's nest for the current cycle. .

[0036] In this embodiment, heat penetration tests can also be conducted for different bottle types, filling volumes, and bird's nest solid content to obtain the corresponding process formulas. , , Parameter set. At the start of the batch, read the parameter set that matches the current bottle type and fill volume, and perform the same center temperature iteration relationship.

[0037] Shear stress index based on cooling water flow rate Calculated cooling water flow rate drag coefficient Cooling water density and characteristic hydraulic diameter calculate: ; in, For the first Shear stress index for each control cycle To reflect the flow resistance coefficient generated by the shelf structure and bottle arrangement inside the sterilizer, For the density of cooling water, The cooling water flow rate for the current control cycle is determined by the cooling water volumetric flow rate for the current control cycle. Divide by the cross-sectional area of ​​the pipe to obtain the result. The characteristic hydraulic diameter.

[0038] In this embodiment, Take 0.87, Take 1000 kg per cubic meter. Take 0.04 meters.

[0039] Convert the core temperature of bird's nest from Celsius to absolute temperature: ; Absolute temperature, shear stress index, and protein denaturation activation energy Pre-exponential factors Shear damage coupling coefficient and shear damage index Substituting into the following formula, we obtain the instantaneous rate of quality deterioration. : ; in, For the first The instantaneous quality degradation rate per control cycle Pre-exponential factor, The protein denaturation activation energy is obtained from differential scanning calorimetry of bird's nest protein. Let be the ideal gas constant. The core temperature of bird's nest is measured in Kelvin. The shear damage coupling coefficient is... This represents the shear damage index.

[0040] In this embodiment, Pick , Pick , Take 8.314, Take 0.034, Take 1.55.

[0041] The instantaneous quality degradation rate of each control cycle is multiplied by the control cycle number and then recursively accumulated. ; in, For the first The cumulative quality damage index at the end of each control cycle For the first Instantaneous quality degradation rate per control cycle. Cumulative quality damage index. The upper limit is set at 1. Shear stress exponent. and cumulative quality damage index During the scaling factor adjustment process, the instantaneous rate of quality deterioration Enter the weight update process.

[0042] In this embodiment, the drag coefficient can also be... The parameter table is configured to be related to the number of shelf layers, bottle diameter, and bottle arrangement density. The resistance coefficient is read according to the shelf loading information of the current batch, while keeping the calculation relationship of shear stress, instantaneous quality deterioration rate, and cumulative quality damage index unchanged.

[0043] S103, input the temperature deviation and the rate of change of temperature deviation into the two-dimensional fuzzy rule base, defuzzify the flow correction value and temperature correction value by weighted average according to the rule activation degree and dynamic weight, calculate the flow set value and temperature set value by the outer layer scaling factor and output them; Furthermore, the temperature deviation and the rate of change of temperature deviation are input into a two-dimensional fuzzy rule base. The flow correction value and temperature correction value are then defuzzified by a weighted average based on rule activation degree and dynamic weights, including: The difference between the temperature inside the vessel and the target final temperature is defined as the temperature deviation, and the difference between the temperature deviations of adjacent cycles is divided by the control cycle to determine the temperature deviation change rate. The temperature deviation and the rate of change of temperature deviation are divided into multiple language ranges from negative to positive, and combined to form multiple rules. The activation degree of a rule is obtained by performing an algebraic product on the membership degrees of its two antecedents. By weighting the rule consequent level with both rule activation degree and dynamic weight, the traffic correction value and temperature correction value are obtained respectively.

[0044] Furthermore, the flow rate setpoint and temperature setpoint are calculated and output using the outer scaling factor, including: The flow rate setpoint is obtained by adding the product of the base cooling water flow rate, the flow rate ratio factor, and the flow rate correction value. The temperature setpoint is obtained by adding the product of the base cooling water temperature, the temperature scaling factor, and the temperature correction value. The flow rate setpoint is sent to the frequency converter of the circulating water pump, and the temperature setpoint is sent to the controller of the hot and cold water mixing valve.

[0045] In this embodiment, the temperature inside the vessel is... With the target final temperature The difference between them is calculated as temperature deviation. The temperature deviation change rate is calculated by dividing the difference in temperature deviation between adjacent control cycles by the control cycle. .

[0046] Temperature deviation is calculated using the following formula: ; in, For temperature deviation, The temperature inside the vessel. The target final temperature.

[0047] The rate of change of temperature deviation is calculated according to the following formula: ; in, The rate of change of temperature deviation. This represents the temperature deviation difference between adjacent control cycles. To control the cycle.

[0048] Temperature deviation and the rate of change of temperature deviation As a precondition input to the two-dimensional fuzzy rule base, the cumulative quality impairment index is used. and shear stress index It participates in the calculation of rule performance reward value during the weight update process and in the adjustment of the proportional factor during the proportional factor adjustment process.

[0049] In this embodiment, temperature deviation Divided into multiple language ranges including negative large, negative medium, negative small, zero, positive small, positive medium, and positive large, with a domain of discourse ranging from negative five degrees Celsius to positive five degrees Celsius; temperature deviation change rate. The system is divided into multiple language regions, including negative large, negative medium, negative small, zero, positive small, positive medium, and positive large, with a domain ranging from -0.5 degrees Celsius per second to +0.5 degrees Celsius per second. Adjacent membership functions intersect at a membership degree of 0.5. Specifically, this embodiment uses seven language regions, forming forty-nine rules.

[0050] Temperature deviation The following membership functions are used: large negative values ​​use trapezoidal intervals, medium negative values ​​use triangular intervals, small negative values ​​use triangular intervals, zero negative values ​​use triangular intervals, small positive values ​​use triangular intervals, medium positive values ​​use triangular intervals, and large positive values ​​use trapezoidal intervals. Temperature deviation change rate. The same language range arrangement is used to cover temperatures from -0.5 degrees Celsius per second to +0.5 degrees Celsius per second.

[0051] The two-dimensional fuzzy rule base consists of temperature deviation Language range and temperature deviation change rate Multiple rules are formed by the cross-combination of language intervals. Each rule includes a consequent level for flow correction and a consequent level for temperature correction, with consequent levels ranging from -3 to +3, representing correction directions from negative to positive. The algebraic product of the membership degrees of the two antecedents of each rule is calculated to obtain the rule activation degree. .

[0052] The flow correction value is calculated using the following formula: ; in, For the first Flow correction value for each control cycle For the first The dynamic weight of each rule For rule activation degree, For flow rate correction, the part level is adjusted. Temperature correction values ​​are calculated using the same weighted average relationship. Replace with temperature-corrected part grade .

[0053] In this embodiment, when the temperature deviation is positive and the rate of change of the temperature deviation is negative, the component grade after flow correction is set to positive and the component grade after temperature correction is set to negative, thereby increasing the cooling water flow rate and decreasing the correction direction of the cooling water temperature setting.

[0054] The flow correction value and temperature correction value are converted into flow setpoint and temperature setpoint by an outer scaling factor, wherein the outer scaling factor includes a flow scaling factor. and temperature scaling factor .

[0055] The flow rate setpoint is calculated using the following formula: ; in, Set a value for the flow rate. Based on cooling water flow rate, For flow rate ratio factor, This is a flow rate correction value. In this embodiment, the basic cooling water flow rate is... Take 25.

[0056] The temperature setpoint is calculated using the following formula: ; in, Set the temperature value. Based on the cooling water temperature, This is the temperature scaling factor. This is the temperature correction value.

[0057] In this embodiment, the basic cooling water temperature Take 25.

[0058] The flow rate setpoint is sent to the frequency converter of the circulating water pump, and the temperature setpoint is sent to the controller of the hot and cold water mixing valve. The frequency converter of the circulating water pump adjusts the cooling water circulation flow rate, and the hot and cold water mixing valve adjusts the cooling water temperature entering the cooling circuit of the sterilizer.

[0059] S104, in the next cycle after the control quantity is applied, calculate the performance bonus value based on the absolute value improvement of temperature deviation, the reference quality degradation rate and the measured instantaneous quality degradation rate, and update, normalize and trim the dynamic weights. Furthermore, in the next cycle after the control variable is applied, the performance bonus value is calculated based on the absolute improvement in temperature deviation, the reference quality degradation rate, and the measured instantaneous quality degradation rate. The dynamic weights are then updated, normalized, and trimmed, including: For rules whose activation degree is greater than the preset activation threshold, calculate the difference between the absolute value of the temperature deviation between the current cycle and the next cycle, and determine the sign of the difference; Multiply the difference sign by the temperature deviation improvement weight, and subtract the instantaneous quality degradation rate of the next cycle from the reference quality degradation rate, then multiply by the quality degradation suppression weight, and add the two to obtain the performance bonus value. The updated weights are obtained by adding the dynamic weights to the product of the learning rate and the performance reward value; Normalize all update weights, and reset update weights below the preset pruning threshold to zero before normalizing them; Rules whose activation degree is not greater than the preset activation threshold retain their current dynamic weight.

[0060] Furthermore, updating, normalizing, and pruning dynamic weights also includes: The current performance reward value is added to the performance reward value within the preset historical control period window according to the decreasing weight over time to obtain the cumulative reward value of the rule; When the cumulative reward value of a rule is negative and the shear stress exponent exceeds the preset shear stress warning value, the learning rate of the rule is reduced and the weights are calculated and updated. When the cumulative reward value of the rule is positive and the cumulative quality damage index exceeds the cumulative damage threshold, the update weight of the flow correction direction rule is increased and the update weight of the temperature correction direction rule is decreased according to the signs of the flow correction value and temperature correction value, respectively. Normalize the updated weights and perform pruning.

[0061] In this embodiment, after the circulating water pump inverter and the hot and cold water mixing valve execute the flow rate setpoint and temperature setpoint, they wait for one control cycle to obtain the reactor temperature and cooling water flow rate again, and calculate the instantaneous quality degradation rate for the next cycle. The absolute change in temperature deviation between the current cycle and the next cycle, and the rate of deterioration of reference quality. and the instantaneous quality degradation rate in the next cycle Input performance reward value calculation process.

[0062] The ideal cooling curve is: ; in, After cooling begins The ideal core temperature at any time To cool the initial center temperature, To achieve the target final temperature, For the ideal cooling attenuation coefficient, This refers to the time after cooling begins. In this embodiment, Pick , In seconds. At the... One control cycle, take ,Will Convert Celsius temperature to absolute temperature, and compare it with the first... Shear stress index per control cycle Substituting all the values ​​into the thermal damage calculation relationship, we obtain the reference quality degradation rate. The results of the dynamic weight update are used in the calculation of the flow correction and temperature correction values ​​for the next control cycle.

[0063] First, rules with activation values ​​greater than a preset activation threshold are selected. In this embodiment, the preset activation threshold is set to 0.05. For each rule with an activation value greater than the preset activation threshold, the difference between the absolute values ​​of the temperature deviations between the current cycle and the next cycle is calculated: ; in, This represents the improvement in the absolute value of the temperature deviation. This represents the temperature deviation for the current cycle. This is the temperature deviation for the next cycle. A sign determination is performed on the difference in the absolute value of the temperature deviation.

[0064] Performance Bonus Value Calculate according to the following formula: ; in, For the first The performance bonus value of each activation rule. To improve the weighting of temperature deviation, Weighting for quality degradation suppression For symbolic functions, For reference, the rate of quality deterioration, The instantaneous quality degradation rate is measured for the next cycle.

[0065] In this embodiment, Take 0.4, Take 0.6.

[0066] Combine the current dynamic weights with the learning rate and performance bonus value Adding the products together, we get the updated weights: ; in, For the updated dynamic weights, For the current dynamic weight, The learning rate. In this embodiment, the learning rate is... Take 0.15.

[0067] Normalize all update weights, reset update weights below the preset pruning threshold to zero, and then normalize update weights that are not set to zero.

[0068] In this embodiment, the preset pruning threshold is set to 0.1. Rules whose activation degree is not greater than the preset activation threshold retain their current dynamic weight.

[0069] In this embodiment, the current performance reward value can be accumulated with the performance reward values ​​within a preset historical control period window using a time-decreasing weight to obtain the cumulative reward value for the rule. When the cumulative reward value for the rule is negative and the shear stress index exceeds the preset shear stress warning value, the learning rate of the rule is reduced before calculating the update weight. When the cumulative reward value for the rule is positive and the cumulative quality damage index exceeds the cumulative damage threshold, the update weight of the flow correction direction rule is increased and the update weight of the temperature correction direction rule is decreased according to the signs of the flow correction value and temperature correction value, respectively.

[0070] S105, based on the cumulative quality damage index, increases the flow rate proportional factor and decreases the temperature proportional factor, and performs saturated compression on the flow rate proportional factor based on the shear stress index. Furthermore, based on the cumulative quality damage index, the flow rate proportionality factor is increased and the temperature proportionality factor is decreased; based on the shear stress index, saturation compression is applied to the flow rate proportionality factor, including: Input the ratio of the cumulative quality damage index to the cumulative damage threshold into the hyperbolic tangent function; Multiply the hyperbolic tangent function output by the cumulative damage gain coefficient and add it to the flow base scaling factor; The saturation function input is determined based on the excess amount of the shear stress index exceeding the critical shear stress index. When the shear stress index does not exceed the critical shear stress index, the excess amount is set to zero. The compression flow rate ratio factor is output based on the saturation function. The temperature scaling factor is obtained by multiplying the output of the hyperbolic tangent function by the temperature suppression coefficient and then subtracting it from the temperature scaling factor.

[0071] Furthermore, based on the cumulative quality damage index, the flow rate proportionality factor is increased while the temperature proportionality factor is decreased; based on the shear stress index, saturation compression is applied to the flow rate proportionality factor, which also includes: Subtract the reference quality degradation rate from the current instantaneous quality degradation rate to obtain the quality degradation deviation. When the deviation of quality degradation is positive and the shear stress index is lower than the preset shear stress warning value, the deviation of quality degradation is multiplied by the preset deviation gain and then added to the flow rate proportional factor, and the preset temperature adjustment amount is deducted from the temperature proportional factor. When the deviation of quality degradation is not positive or the shear stress index is not lower than the preset shear stress warning value, maintain the temperature proportional factor and perform saturated compression on the flow proportional factor.

[0072] In this embodiment, the flow rate proportionality factor is calculated according to the following formula: ; in, For flow rate ratio factor, As the basic proportional factor for traffic, This is the cumulative damage gain coefficient. To accumulate the quality damage index, The cumulative damage threshold, This is the shear stress proportional suppression coefficient. The shear stress index is... The critical shear stress index. It is a saturation function. In this embodiment, Take 0.8, Take 2.0, Take 1.5, Take 0.65, Pick .

[0073] The temperature scaling factor is calculated using the following formula: ; in, This is the temperature scaling factor. As the temperature-based scaling factor, This is the temperature suppression coefficient. In this embodiment, Take 1.0, Take 1.0.

[0074] As the cumulative quality damage index increases, the hyperbolic tangent function output increases, the cumulative damage gain term in the flow rate proportional factor increases, and the temperature suppression term in the temperature proportional factor increases. When the shear stress index exceeds the critical shear stress index, the flow rate proportional factor is compressed based on the ratio of the excess to the critical shear stress index, thus entering the saturation function.

[0075] In this embodiment, the current instantaneous quality degradation rate can be subtracted from the reference quality degradation rate to obtain the quality degradation deviation. When the quality degradation deviation is positive and the shear stress index is lower than the preset shear stress warning value, the quality degradation deviation is multiplied by the preset deviation gain and then added to the flow rate proportional factor, and the preset temperature adjustment amount is subtracted from the temperature proportional factor.

[0076] S106: When the increase in the center temperature of the bird's nest and the cumulative quality damage index meet the cooling endpoint conditions, the cooling water circulation is turned off and the dynamic weight and proportional factor are saved. If the cooling endpoint conditions are not met, the temperature inside the collection vessel and the cooling water flow rate are returned.

[0077] Furthermore, when the increase in the center temperature and cumulative quality damage index of the bird's nest meets the cooling endpoint conditions, the cooling water circulation is shut off while the dynamic weights and scaling factors are preserved, including: The cooling endpoint is defined as the point where the center temperature of the bird's nest is lower than or equal to the sum of the target final temperature and the preset temperature difference, and the cumulative increase in the quality damage index over the past preset period is less than the preset increment threshold. Turn off the cooling water circulation at the end of the cooling process; Extract the normalized and trimmed dynamic weights, flow rate scaling factor, and temperature scaling factor of this batch, write them into the edge database, and read them as initial parameters when the next batch begins cooling.

[0078] In this embodiment, the center temperature of the bird's nest is calculated in each control cycle. and cumulative quality damage index The change in quantity.

[0079] In this embodiment, the cooling endpoint condition is: the center temperature of the bird's nest is lower than or equal to the sum of the target endpoint temperature and the preset temperature difference, and the cumulative quality damage index increment within the preset time period is less than the preset increment threshold. Specifically, the preset temperature difference is 2, the preset time period is 30, and the preset increment threshold is 0.003. When the cooling endpoint condition is met, a shutdown command is sent to the circulating water pump inverter and the hot and cold water mixing valve controller to stop the cooling water circulation. The normalized and trimmed dynamic weights, flow rate proportional factors, and temperature proportional factors of this batch are extracted and written into the edge database. When the next batch starts cooling, the dynamic weights, flow rate proportional factors, and temperature proportional factors of the most recent valid batch are read from the edge database as the initial parameters of the two-dimensional fuzzy rule base and the outer proportional factor.

[0080] The following describes this embodiment in conjunction with application scenarios: This embodiment was implemented on a 150-liter rotary sterilizer for bird's nest. The cooling piping was equipped with a DN25 electromagnetic flowmeter, a PT100 temperature sensor, and a pressure transmitter. The control platform used a programmable controller with edge computing capabilities. At the start of cooling, the initial center temperature was... Take 121, target final temperature Take 40 as the base cooling water flow rate. Take 25, base cooling water temperature Take 25.

[0081] The system collects the internal temperature and cooling water flow rate at one-second intervals, and simultaneously records the cooling water inlet temperature, cooling water outlet temperature, and internal pressure. Based on the collected data, it obtains the bird's nest center temperature, instantaneous quality deterioration rate, cumulative quality damage index, and shear stress index. It performs fuzzification, rule activation calculation, and weighted average defuzzification on the temperature deviation and temperature deviation change rate, generating flow rate setpoints and temperature setpoints for output. In the next control cycle, it calculates the performance bonus value, updates the dynamic weights, and performs pruning. Simultaneously, it adjusts the flow rate proportional factor and temperature proportional factor based on the cumulative quality damage index and shear stress index. This process is continuously repeated. When the bird's nest center temperature meets the preset temperature difference condition and the cumulative quality damage index increment meets the preset stability condition, the cooling water circulation is shut off, and the final dynamic weights, flow rate proportional factor, and temperature proportional factor are written to the edge database.

[0082] For the purpose of simplicity, the method steps disclosed in the above embodiments are described as a series of actions. However, those skilled in the art should understand that the embodiments of the present invention are not limited to the described order of actions, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.

[0083] like Figure 2 As shown, the present invention also provides a self-tuning optimization system for cooling water circulation parameters of a bird's nest sterilization autoclave based on fuzzy logic, comprising: Data acquisition module 201 is configured to acquire the temperature inside the vessel and the flow rate of cooling water; The quality and shear calculation module 202 is configured to substitute the temperature inside the reactor and the center temperature of the bird's nest in the previous cycle into the dynamic time constant of the flow rate, iterate to obtain the center temperature of the bird's nest, substitute the center temperature of the bird's nest into the thermal damage calculation, and substitute the cooling water flow rate, the shelf structure inside the sterilization reactor and the bottle arrangement parameters into the shear stress calculation to obtain the instantaneous quality deterioration rate, the cumulative quality damage index and the shear stress index. The fuzzy control module 203 is configured to input the temperature deviation and the rate of change of temperature deviation into a two-dimensional fuzzy rule base, defuzzify the flow correction value and temperature correction value by weighted average according to the rule activation degree and dynamic weight, and calculate and output the flow set value and temperature set value by the outer scale factor. The weight tuning module 204 is configured to calculate the performance reward value based on the absolute improvement of temperature deviation, the reference quality degradation rate and the measured instantaneous quality degradation rate in the next cycle after the control quantity is applied, and update, normalize and trim the dynamic weights. The proportional factor adjustment module 205 is configured to increase the flow proportional factor and decrease the temperature proportional factor based on the cumulative quality damage index, and to perform saturated compression on the flow proportional factor based on the shear stress index. The batch management module 206 is configured to shut down the cooling water circulation and save the dynamic weight and proportional factor when the central temperature of the bird's nest and the cumulative quality damage index increment meet the cooling endpoint conditions; otherwise, it will return to collect the temperature inside the vessel and the cooling water flow rate.

[0084] It is worth noting that although only some basic functional modules are disclosed in the embodiments of this invention, it does not mean that the composition of this system is limited to the above-mentioned basic functional modules. On the contrary, based on the above-mentioned basic functional modules, those skilled in the art can arbitrarily add one or more functional modules in combination with existing technology to form an infinite number of embodiments or technical solutions. That is to say, this system is open rather than closed. The fact that this embodiment only discloses a few basic functional modules should not be considered as the scope of protection of this invention being limited to the disclosed basic functional modules. At the same time, for the convenience of description, the above devices are described separately according to their functions as various units and modules. Of course, in implementing this invention, the functions of each unit and module can be implemented in one or more software and / or hardware.

[0085] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A self-tuning optimization method for cooling water circulation parameters in a bird's nest sterilization autoclave based on fuzzy logic, characterized in that, include: Collect the temperature inside the reactor and the flow rate of the cooling water; Substituting the temperature inside the autoclave and the center temperature of the bird's nest in the previous cycle into the dynamic time constant of the flow rate, the center temperature of the bird's nest is obtained iteratively. The center temperature of the bird's nest is then substituted into the thermal damage calculation. The cooling water flow rate, the shelf structure inside the sterilization autoclave, and the bottle arrangement parameters are substituted into the shear stress calculation to obtain the instantaneous quality deterioration rate, the cumulative quality damage index, and the shear stress index. Input the temperature deviation and the rate of change of temperature deviation into the two-dimensional fuzzy rule base, defuzzify the flow correction value and temperature correction value by weighted average according to the rule activation degree and dynamic weight, and calculate the flow set value and temperature set value by the outer scale factor and output them. In the next cycle after the control is applied, the performance bonus value is calculated based on the absolute improvement of temperature deviation, the reference quality degradation rate and the measured instantaneous quality degradation rate, and the dynamic weights are updated, normalized and trimmed. The flow rate proportional factor is increased and the temperature proportional factor is decreased based on the cumulative quality damage index, and saturated compression is performed on the flow rate proportional factor based on the shear stress index. When the increase in the center temperature and cumulative quality damage index of the bird's nest meets the cooling endpoint conditions, the cooling water circulation is shut off and the dynamic weight and proportional factor are saved. If the cooling endpoint conditions are not met, the temperature inside the vessel and the cooling water flow rate are collected again.

2. The self-tuning optimization method for cooling water circulation parameters of a bird's nest sterilization autoclave based on fuzzy logic according to claim 1, characterized in that, Substituting the temperature inside the pot and the center temperature of the bird's nest from the previous cycle into the dynamic time constant of the flow rate, the center temperature of the bird's nest is obtained iteratively, further including: The cooling water flow rate is converted into a dynamic time constant of the flow rate using a negative exponential relationship; Calculate the temperature difference between the temperature inside the pot and the center temperature of the bird's nest in the previous cycle; Multiply the temperature difference by the negative exponential complement of the ratio obtained by dividing the control cycle by the flow dynamic time constant, and add it to the bird's nest center temperature of the previous cycle to obtain the bird's nest center temperature.

3. The self-tuning optimization method for cooling water circulation parameters of a bird's nest sterilization autoclave based on fuzzy logic according to claim 1, characterized in that, Substituting the center temperature of the bird's nest into the thermal damage calculation, and substituting the cooling water flow rate, the shelf structure inside the sterilizer, and the bottle arrangement parameters into the shear stress calculation, the instantaneous quality deterioration rate, the cumulative quality damage index, and the shear stress index are obtained, further including: The shear stress index is calculated based on the cooling water flow rate, resistance coefficient, cooling water density, and characteristic hydraulic diameter. The temperature of the bird's nest center was converted from Celsius to absolute temperature, and the absolute temperature, shear stress index, protein denaturation activation energy, pre-exponential factor, shear damage coupling coefficient and shear damage index were substituted into the thermal damage calculation to obtain the instantaneous quality deterioration rate. The instantaneous quality degradation rate of each cycle is multiplied by the control cycle and then recursively accumulated. An upper limit is applied to the accumulated value to obtain the cumulative quality damage index.

4. The self-tuning optimization method for cooling water circulation parameters of a bird's nest sterilization autoclave based on fuzzy logic according to claim 1, characterized in that, Temperature deviation and its rate of change are input into a two-dimensional fuzzy rule base. The flow correction value and temperature correction value are then defuzzified by a weighted average based on rule activation and dynamic weights. Further steps include: The difference between the temperature inside the vessel and the target final temperature is defined as the temperature deviation, and the difference between the temperature deviations of adjacent cycles is divided by the control cycle to determine the temperature deviation change rate. The temperature deviation and the rate of change of temperature deviation are divided into multiple language ranges from negative to positive, and combined to form multiple rules. The activation degree of a rule is obtained by performing an algebraic product on the membership degrees of its two antecedents. By weighting the rule consequent level with both rule activation degree and dynamic weight, the traffic correction value and temperature correction value are obtained respectively.

5. The self-tuning optimization method for cooling water circulation parameters of a bird's nest sterilization autoclave based on fuzzy logic according to claim 1, characterized in that, The flow rate setpoint and temperature setpoint are calculated and output using the outer scaling factor, and further include: The flow rate setpoint is obtained by adding the product of the base cooling water flow rate, the flow rate ratio factor, and the flow rate correction value. The temperature setpoint is obtained by adding the product of the base cooling water temperature, the temperature scaling factor, and the temperature correction value. The flow rate setpoint is sent to the frequency converter of the circulating water pump, and the temperature setpoint is sent to the controller of the hot and cold water mixing valve.

6. The self-tuning optimization method for cooling water circulation parameters of a bird's nest sterilization autoclave based on fuzzy logic according to claim 1, characterized in that, In the next cycle after the control variable is applied, the performance bonus value is calculated based on the absolute improvement in temperature deviation, the reference quality degradation rate, and the measured instantaneous quality degradation rate. The dynamic weights are then updated, normalized, and trimmed, further including: For rules whose activation degree is greater than the preset activation threshold, calculate the difference between the absolute value of the temperature deviation between the current cycle and the next cycle, and determine the sign of the difference; Multiply the difference sign by the temperature deviation improvement weight, and subtract the instantaneous quality degradation rate of the next cycle from the reference quality degradation rate, then multiply by the quality degradation suppression weight, and add the two to obtain the performance bonus value. The updated weights are obtained by adding the dynamic weights to the product of the learning rate and the performance reward value; Normalize all update weights, and reset update weights below the preset pruning threshold to zero before normalizing them; Rules whose activation degree is not greater than the preset activation threshold retain their current dynamic weight.

7. The self-tuning optimization method for cooling water circulation parameters of a bird's nest sterilization autoclave based on fuzzy logic according to claim 1, characterized in that, Increasing the flow rate proportionality factor and decreasing the temperature proportionality factor based on the cumulative quality damage index, and performing saturated compression on the flow rate proportionality factor based on the shear stress index, further includes: Input the ratio of the cumulative quality damage index to the cumulative damage threshold into the hyperbolic tangent function; Multiply the hyperbolic tangent function output by the cumulative damage gain coefficient and add it to the flow base scaling factor; The saturation function input is determined based on the excess amount of the shear stress index exceeding the critical shear stress index. When the shear stress index does not exceed the critical shear stress index, the excess amount is set to zero. The compression flow rate ratio factor is output based on the saturation function. The temperature scaling factor is obtained by multiplying the output of the hyperbolic tangent function by the temperature suppression coefficient and then subtracting it from the temperature scaling factor.

8. The self-tuning optimization method for cooling water circulation parameters of a bird's nest sterilization autoclave based on fuzzy logic according to claim 1, characterized in that, When the increase in the center temperature and cumulative quality damage index of the bird's nest meets the cooling endpoint condition, the cooling water circulation is shut off while the dynamic weights and scaling factors are preserved, further including: The cooling endpoint is defined as the point where the center temperature of the bird's nest is lower than or equal to the sum of the target final temperature and the preset temperature difference, and the cumulative increase in the quality damage index over the past preset period is less than the preset increment threshold. Turn off the cooling water circulation at the end of the cooling process; Extract the normalized and trimmed dynamic weights, flow rate scaling factor, and temperature scaling factor of this batch, write them into the edge database, and read them as initial parameters when the next batch begins cooling.

9. The self-tuning optimization method for cooling water circulation parameters of a bird's nest sterilization autoclave based on fuzzy logic according to claim 6, characterized in that, Updating, normalizing, and pruning dynamic weights also includes: The current performance reward value is added to the performance reward value within the preset historical control period window according to the decreasing weight over time to obtain the cumulative reward value of the rule; When the cumulative reward value of a rule is negative and the shear stress exponent exceeds the preset shear stress warning value, the learning rate of the rule is reduced and the weights are calculated and updated. When the cumulative reward value of the rule is positive and the cumulative quality damage index exceeds the cumulative damage threshold, the update weight of the flow correction direction rule is increased and the update weight of the temperature correction direction rule is decreased according to the signs of the flow correction value and temperature correction value, respectively. Normalize the updated weights and perform pruning.

10. A self-tuning optimization system for cooling water circulation parameters in a bird's nest sterilization autoclave based on fuzzy logic, characterized in that, include: The data acquisition module is configured to collect the temperature inside the vessel and the flow rate of the cooling water. The quality and shear calculation module is configured to substitute the temperature inside the autoclave and the center temperature of the bird's nest in the previous cycle into the dynamic time constant of the flow rate, iterate to obtain the center temperature of the bird's nest, substitute the center temperature of the bird's nest into the thermal damage calculation, and substitute the cooling water flow rate, the shelf structure inside the sterilization autoclave and the bottle arrangement parameters into the shear stress calculation to obtain the instantaneous quality deterioration rate, the cumulative quality damage index and the shear stress index. The fuzzy control module is configured to input the temperature deviation and the rate of change of temperature deviation into a two-dimensional fuzzy rule base, defuzzify the flow correction value and temperature correction value by weighted average according to the rule activation degree and dynamic weight, and calculate and output the flow setpoint and temperature setpoint by the outer scale factor. The weight tuning module is configured to calculate the performance reward value based on the absolute improvement of temperature deviation, the reference quality degradation rate, and the measured instantaneous quality degradation rate in the next cycle after the control quantity is applied, and update, normalize, and trim the dynamic weights. The proportional factor adjustment module is configured to increase the flow proportional factor and decrease the temperature proportional factor based on the cumulative quality damage index, and to perform saturated compression on the flow proportional factor based on the shear stress index. The batch management module is configured to shut down the cooling water circulation and save the dynamic weights and proportional factors when the central temperature of the bird's nest and the cumulative quality damage index increment meet the cooling endpoint conditions; otherwise, it will return to collecting the temperature inside the vessel and the cooling water flow rate.

Images