A Smart Control Method and System for Post-Harvest Processing of Fruits Based on Modified Atmosphere Preservation
By real-time monitoring of the respiration quotient and ethylene release rate of fruits, a safe envelope is generated. By using a Kalman filter to adjust the modified atmosphere equipment, the problems of physiological tolerance differences and dynamic changes in respiration during fruit storage are solved, thereby improving the preservation effect of fruits.
Patent Information
- Application Number
- CN202611217232.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-12
- Publication Date
- 2026-11-17
AI Technical Summary
Existing modified atmosphere storage technology fails to effectively consider the differences in physiological tolerance and dynamic changes in respiration of fruits during storage, resulting in poor preservation effects.
By collecting the respiration quotient and ethylene release rate of fruits in real time, a safe envelope is generated. A Kalman filter is used to determine the oxygen consumption rate and carbon dioxide production rate, identify envelope boundary crossing events, and adjust the processing strategy of the controlled atmosphere equipment according to the conflict mode.
It achieves a dynamic balance of the controlled atmosphere environment during fruit storage, improves preservation effect and control precision, prevents damage to fruit, and provides targeted treatment solutions.
Smart Images

Figure CN123113500A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of fruit preservation and control technology, and in particular to an intelligent control method and system for post-harvest fruit processing based on modified atmosphere preservation. Background Technology
[0002] Controlled atmosphere storage (CAS) extends the shelf life of fruits by regulating the concentrations of gases such as O2, CO2, and ethylene in the storage environment to inhibit the respiration and metabolism of the fruits. Early CAS relied on simple mechanical refrigeration and gas conditioning devices, which could only provide basic control over the temperature, humidity, and gas composition in the storage room. The gas concentration control precision was poor, resulting in unsatisfactory preservation effects. The gas composition of traditional CAS storage rooms is usually controlled at 2%–5% O2 and 2%–5% CO2. However, different varieties and fruits at different stages of maturity have significantly different tolerances to the gas environment. During long-term CAS storage, the tolerance of fruits to gases may change, which may lead to damage from low O2 or high CO2, affecting the storage effect. Existing cold storage control systems have shortcomings in the fruit preservation process, such as deviation from control objectives, insufficient process adaptability, lack of modeling of multi-parameter coupling relationships, and lack of dynamic collaborative strategies oriented towards the physiological characteristics of fruits.
[0003] Chinese Patent Application Publication No. CN121444958A discloses a gas regulation method, device, medium, product, and chip for a fruit and vegetable storage space. In this method, within a first time period, a first target concentration of oxygen, a first concentration of ethylene in the fruit and vegetable storage space, and a first temperature of the fruit and vegetable storage space are obtained; the fruit and vegetable storage space stores at least one type of fruit and vegetables; a first predicted respiration intensity is determined based on the first concentration, first temperature, first type, and first target concentration; a first power consumption of the nitrogen generator is obtained; a first balance value is determined based on the first predicted respiration intensity, the preset target respiration intensity, and the first power consumption; if the first balance value is less than the preset balance value, the oxygen concentration in the fruit and vegetable storage space is adjusted to the first target concentration. This avoids both excessive respiration intensity leading to accelerated ripening of the fruit and vegetables and excessive nitrogen input to the nitrogen generator resulting in higher costs.
[0004] However, existing technologies do not take into account the dynamic changes in the physiological tolerance of fruits and respiration to determine and maintain a comfortable controlled atmosphere environment, which leads to poor preservation results. Summary of the Invention
[0005] Therefore, this invention provides an intelligent control method for post-harvest processing of fruits based on modified atmosphere preservation, which overcomes the problem in the prior art that does not take into account the dynamic changes in the physiological tolerance differences of fruits and respiration to determine and maintain a comfortable modified atmosphere environment, thus leading to poor preservation effect.
[0006] To achieve the above objectives, the present invention provides an intelligent control method for post-harvest processing of fruits based on modified atmosphere preservation, comprising: The fruit was placed in a controlled atmosphere storage room, and the gas flow rate, oxygen concentration, and carbon dioxide concentration of the controlled atmosphere storage room were obtained to determine the current respiratory quotient in the storage room. The moving average slope of the respiratory quotient corresponding to the fruit and the relative change rate of ethylene release rate are collected to determine the envelope scaling factor. Based on the envelope scaling factor, the preset physiological tolerance envelope is adjusted to generate a safe envelope. The oxygen consumption rate of the fruit at the current moment and the rate of change of the oxygen consumption rate within the preset storage time are obtained. The oxygen consumption trend stage of the fruit is identified based on the rate of change of the oxygen consumption rate. The corresponding prediction mode is selected to extrapolate the trend of the oxygen consumption rate to obtain the predicted oxygen consumption rate sequence. The predicted oxygen consumption rate sequence is input into the controlled atmosphere storage environment dynamic simulator to obtain the environmental prediction sequence containing multiple future time steps of predicted oxygen concentration, predicted carbon dioxide concentration, predicted ethylene concentration and predicted temperature. Determine whether the fruit has experienced an envelope boundary crossing event based on the environmental prediction sequence, and determine the conflict mode of the envelope boundary crossing event; The classification and processing strategy of the modified atmosphere preservation equipment is determined according to the conflict mode, and the fruit is subjected to modified atmosphere preservation treatment based on the classification and processing strategy. The conflict modes include hypoxia stress conflict, carbon dioxide stress conflict, ethylene-induced stress conflict and mixed stress conflict; The classification and processing strategy includes raising the lower limit of the oxygen concentration control target, increasing the power of the carbon dioxide removal device, increasing the operating frequency of the ethylene removal device, and implementing the preset highest safety level control mode.
[0007] Furthermore, the process of determining the respiratory quotient in the current stock includes: Oxygen concentration, carbon dioxide concentration, and ethylene concentration were obtained to establish a dynamic gas mass balance equation; The dynamic gas mass balance equation is discretized, and the oxygen consumption rate and carbon dioxide production rate of each fruit variety are determined by Kalman filter. The respiratory quotient is determined based on the oxygen consumption rate and the carbon dioxide production rate.
[0008] Furthermore, the process of determining the scaling factor of the envelope surface includes: Obtain several respiratory quotients within a preset period to generate a respiratory quotient sequence, and perform univariate linear regression on the respiratory quotient sequence to obtain the moving average slope. The average ethylene release rate during the stabilization period after storage is obtained as the reference ethylene release rate. The average ethylene release rate over a short time window is calculated as the real-time ethylene release rate. The relative rate of change is determined based on the reference ethylene release rate and the real-time ethylene release rate; The envelope scaling factor is determined based on the relationship between the moving average slope and the reference moving average slope, and the relationship between the relative rate of change and the reference relative rate of change. The ethylene release rate is determined by the ethylene concentration balance equation and a Kalman filter.
[0009] Furthermore, the process of generating the security envelope includes: The physiological tolerance envelope is reduced based on the envelope scaling factor to generate the safety envelope.
[0010] Furthermore, the process of obtaining the predicted oxygen consumption rate sequence includes: When the absolute value of the rate of change of the oxygen consumption rate is less than or equal to the first rate of change threshold, the oxygen consumption trend stage is identified as a stable period, a stable prediction mode is selected, and the oxygen consumption rate at the current moment is determined as a constant value in the prediction time domain to generate the predicted oxygen consumption rate sequence. When the absolute value of the rate of change of the oxygen consumption rate is greater than the first rate of change threshold and less than or equal to the second rate of change threshold, the oxygen consumption trend stage is identified as a transition period. A linear prediction mode is selected, and the predicted oxygen consumption rate sequence is generated by calculating based on the oxygen consumption rate, the rate of change of the oxygen consumption rate, and the prediction time domain length. When the absolute value of the rate of change of the oxygen consumption rate is greater than the second rate of change threshold, the oxygen consumption trend stage is identified as the acceleration period. A modified linear prediction mode is selected, and the rate of change of the oxygen consumption rate is multiplied by a preset correction coefficient. Then, the predicted oxygen consumption rate sequence is generated based on the oxygen consumption rate, the modified rate of change of the oxygen consumption rate, and the predicted time domain length.
[0011] Furthermore, the process of determining whether a fruit has experienced an envelope boundary crossing event includes: Substitute several predicted points into the ellipse equation corresponding to the safe envelope to obtain several predicted distances; In response to any predicted distance being greater than the reference distance, an envelope boundary crossing event is determined for the fruit. The prediction point is determined based on the predicted oxygen consumption rate.
[0012] Furthermore, the process of determining the conflict mode of envelope boundary crossing events includes: Several contribution values are determined based on the distance and the ellipse equation corresponding to the safety envelope surface; The conflict mode is determined based on the maximum value among several of the stated contributions.
[0013] Furthermore, the process of determining the conflict mode of envelope boundary crossing events also includes: If the maximum value is the oxygen contribution, then the conflict mode is determined to be the hypoxia stress conflict; If the maximum value is the contribution of carbon dioxide, then the conflict mode is determined to be the carbon dioxide stress conflict; If the maximum value is the contribution of ethylene, then the conflict mode is determined to be the ethylene-induced stress conflict; If the maximum value is the temperature contribution, then the conflict mode is determined to be the mixed stress conflict.
[0014] Furthermore, the process of determining the classification and processing strategy includes: If the conflict mode is a hypoxia stress conflict, the strategy is to raise the lower limit of the oxygen concentration control target. If the conflict mode is a carbon dioxide stress conflict, the strategy is to increase the power of the carbon dioxide removal device. If the conflict mode is an ethylene-induced stress conflict, the strategy is to increase the operating frequency of the ethylene removal unit. If the conflict mode is a mixed coercive conflict, the strategy is to execute the preset highest security level control mode.
[0015] This invention also provides an intelligent control system for post-harvest processing of fruits based on modified atmosphere preservation, comprising: The data acquisition module is used to obtain the gas flow rate, oxygen concentration and carbon dioxide concentration of the controlled atmosphere storage, collect the moving average slope of the respiratory quotient of the fruit and the relative change rate of ethylene release rate, and obtain the oxygen consumption rate and the change rate of oxygen consumption rate of the fruit. The envelope generation module is used to determine the respiratory quotient in the current controlled atmosphere storage based on the gas flow rate, oxygen concentration and carbon dioxide concentration, determine the envelope scaling factor based on the relative change rate of the moving average slope and the ethylene release rate, and adjust the preset physiological tolerance envelope based on the envelope scaling factor to generate a safe envelope. An environmental prediction module is used to determine an environmental prediction sequence by identifying oxygen consumption trend stages and selecting corresponding prediction modes based on the oxygen consumption rate and the rate of change of the oxygen consumption rate; The event analysis module is used to determine whether the fruit has experienced an envelope boundary crossing event based on the environmental prediction sequence, and to determine the conflict mode of the envelope boundary crossing event. The classification processing module is used to determine the classification processing strategy of the modified atmosphere preservation equipment according to the conflict mode, and to perform modified atmosphere preservation treatment on the fruit based on the classification processing strategy.
[0016] Compared with existing technologies, the beneficial effects of this invention are as follows: by considering that the physiological tolerance of fruits gradually decreases during long-term storage due to factors such as changes in maturity and cumulative stress, the invention collects in real time the moving average slope of the respiratory quotient and the relative change rate of ethylene release rate. The moving average slope of the respiratory quotient determines whether respiratory metabolism is slowly sliding towards anaerobic fermentation, and the relative change rate of ethylene release rate determines whether the ethylene production system is abnormally activated. The envelope scaling factor is determined by the moving average slope of the respiratory quotient and the relative change rate of ethylene release rate. The semi-axial length of each dimension of the physiological tolerance envelope is contracted to generate a safe envelope, so that the safety boundary can adapt and adjust according to the changes in the actual physiological state of the fruit. This achieves a dynamic balance between control precision and storage safety, thereby further improving the preservation capability of the intelligent control method for post-harvest fruit processing based on modified atmosphere preservation.
[0017] Furthermore, this invention obtains a predicted oxygen consumption rate sequence by linearly extrapolating the real-time oxygen consumption rate and its rate of change. This sequence is then transformed into an environmental prediction sequence containing multiple future time steps using a controlled atmosphere storage environment dynamic simulator. The environmental state at each prediction point is substituted into the elliptical equation of the safety envelope to calculate the dimensionless distance. When any distance exceeds the baseline distance, an envelope boundary crossing event is determined to have occurred. By decomposing the contribution of each dimension to the total boundary crossing distance, the dominant boundary crossing dimension is identified, and the boundary crossing event is classified into hypoxia stress conflict, carbon dioxide stress conflict, ethylene-induced stress conflict, or mixed stress conflict. This allows for early warning before the fruit is actually damaged and accurately identifies the root cause dimension of the problem, providing clear guidance for targeted treatment. This further improves the preservation capability of the intelligent control method for post-harvest fruit processing based on controlled atmosphere preservation.
[0018] Furthermore, this invention establishes a mapping relationship between four conflict modes and corresponding classification and processing strategies in advance: raising the lower limit of the oxygen concentration control target during low-oxygen stress conflict, increasing the power of the carbon dioxide removal device during carbon dioxide stress conflict, increasing the operating frequency of the ethylene removal device during ethylene-induced stress conflict, and executing the preset highest safety level control mode during mixed stress conflict. When the conflict mode of the envelope boundary crossing event is determined, the system calls the corresponding classification and processing strategy to perform directional correction on the control variables of the modified atmosphere preservation system, and re-inputs the corrected control target into the environmental dynamic simulator for verification. Execution is only initiated when the new predicted sequence is entirely within the safe envelope, thereby improving the targeting of modified atmosphere preservation and further enhancing the preservation capability of the intelligent control method for post-harvest fruit processing based on modified atmosphere preservation. Attached Figure Description
[0019] Figure 1 This is a flowchart of an intelligent control method for post-harvest processing of fruits based on modified atmosphere preservation, as described in an embodiment of the present invention. Figure 2 This is a logic diagram for determining whether a fruit has experienced an envelope boundary crossing event, according to an embodiment of the present invention. Figure 3 A logic decision diagram for determining the classification processing strategy of the security envelope surface in an embodiment of the present invention; Figure 4 This is a block diagram of the intelligent control system for post-harvest processing of fruits based on modified atmosphere preservation, according to an embodiment of the present invention. Detailed Implementation
[0020] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.
[0021] Please see Figure 1 The diagram shown is a flowchart of an intelligent control method for post-harvest processing of fruits based on modified atmosphere preservation, according to an embodiment of the present invention. The method in this embodiment includes: Step S1: Place the fruit in a controlled atmosphere storage room and obtain the gas flow rate, oxygen concentration, and carbon dioxide concentration of the controlled atmosphere storage room to determine the current respiration quotient in the storage room. Step S2: Collect the moving average slope of the respiratory quotient corresponding to the fruit and the relative change rate of ethylene release rate to determine the envelope scaling factor. Adjust the preset physiological tolerance envelope based on the envelope scaling factor to generate a safe envelope. Step S3: Obtain the oxygen consumption rate of the fruit at the current moment and the rate of change of the oxygen consumption rate within the preset storage time. Identify the oxygen consumption trend stage of the fruit based on the rate of change of the oxygen consumption rate, and select the corresponding prediction mode to extrapolate the trend of the oxygen consumption rate to obtain the predicted oxygen consumption rate sequence. Input the predicted oxygen consumption rate sequence into the controlled atmosphere storage environment dynamic simulator to obtain an environmental prediction sequence containing multiple future time steps of predicted oxygen concentration, predicted carbon dioxide concentration, predicted ethylene concentration and predicted temperature. Step S4: Determine whether the fruit has experienced an envelope boundary crossing event based on the environmental prediction sequence, and determine the conflict mode of the envelope boundary crossing event. Step S5: Determine the classification and processing strategy of the modified atmosphere preservation equipment according to the conflict mode, and perform modified atmosphere preservation treatment on the fruit based on the classification and processing strategy. Among them, the conflict modes include hypoxia stress conflict, carbon dioxide stress conflict, ethylene-induced stress conflict and mixed stress conflict; The classification and treatment strategies include raising the lower limit of the oxygen concentration control target, increasing the power of the carbon dioxide removal unit, increasing the operating frequency of the ethylene removal unit, and implementing the preset highest safety level control mode.
[0022] In this embodiment, the process of determining the respiratory quotient in the current library includes: Oxygen concentration, carbon dioxide concentration, and ethylene concentration were obtained to establish a dynamic gas mass balance equation; The dynamic gas mass balance equation is discretized, and the oxygen consumption rate and carbon dioxide production rate of each fruit variety are determined by Kalman filter. The respiratory quotient is determined based on the rate of oxygen consumption and the rate of carbon dioxide production.
[0023] Specifically, in this embodiment, the Kalman filter is used to estimate online the rates of oxygen consumption, carbon dioxide production, and ethylene production of the fruit, which cannot be directly measured, from real-time measurements of gas concentration in the storage facility.
[0024] The state equation of a Kalman filter describes the evolution of the variables to be estimated over time. In this embodiment, the state variables to be estimated are: the oxygen consumption rate of each fruit variety, the carbon dioxide production rate of each fruit variety, the ethylene production rate of each fruit variety, and the gas leakage rate of the storage tank. Since the respiration rate of the fruit usually changes slowly between adjacent sampling periods, the state equation adopts a random walk model, that is, the state variable at the next time step is equal to the state variable at the current time step plus a small amount of process noise.
[0025] The observation equation of the Kalman filter describes the mathematical relationship between state variables and measurable quantities. In this embodiment, the observed quantities are the real-time sensor measurements of oxygen, carbon dioxide, and ethylene concentrations within the storage facility. The observation equation is the discretized form of the dynamic gas mass balance equation. Specifically, the oxygen concentration at the next moment is equal to the current oxygen concentration plus the concentration change caused by nitrogen injection from the nitrogen generator, storage leakage, and fruit consumption. The fruit consumption term is equal to the sum of the oxygen consumption rates of each variety multiplied by their respective masses and then divided by the total volume of free gas within the storage facility. Similarly, the updates to carbon dioxide and ethylene concentrations are determined by fruit production, equipment removal, and leakage, respectively.
[0026] The process noise covariance matrix of a Kalman filter reflects the uncertainty of the random walk model in the state equation, namely the natural fluctuations in respiration rate and leakage rate per unit time. This matrix is diagonal, with each element on the diagonal set based on the physical characteristics and experience of the corresponding state variables. For example, the process noise variance of oxygen consumption rate can be set based on the historical fluctuation variance of the respiration rate of the fruit variety under steady-state storage conditions, typically taken as one to five percent of the square of the mean respiration rate. The process noise variance of leakage rate can be set based on the repeatability standard deviation of leakage rate measurements in empty storage pressure decay tests. Since the state variables evolve independently, all off-diagonal elements are zero.
[0027] The observation noise covariance matrix of the Kalman filter reflects the measurement noise characteristics of each gas concentration sensor. This matrix is also diagonal, with each element on the diagonal directly taken from the sensor's technical specifications or factory calibration data. For example, the measurement noise variance of an oxygen sensor can be calculated based on its nominal accuracy, such as 0.5% of full scale; the same applies to carbon dioxide and ethylene sensors. Since each sensor operates independently, there is no correlation between measurement noises, and all off-diagonal elements are zero. If zoned sampling is used in a multi-variety mixed storage scenario, the sensor noise characteristics at each sampling point can be set separately.
[0028] The Kalman filter requires a set of initial state estimates as the starting point for recursive calculations. In this embodiment, the initial estimates of the oxygen consumption rate and carbon dioxide production rate of each fruit variety can be set using the offline calibration value of the first respiration rate during the stabilization period after storage, and the initial estimate of the ethylene production rate is similarly set. The initial estimate of the storage leakage rate can be obtained from the results of an empty storage pressure decay test. Because the Kalman filter has convergence characteristics, as long as the initial values do not deviate significantly from the order of magnitude of the actual values, the filter can converge to near the true values within several sampling periods.
[0029] The initial error covariance matrix of the Kalman filter reflects the uncertainty of the initial state estimate. This matrix is diagonal, with each element on the diagonal set according to the confidence level of the initial estimate. If the initial estimate comes from a reliable offline calibration, the corresponding variance value is set smaller, indicating a higher confidence level in the initial value. If the initial estimate is merely an empirical guess, the corresponding variance value is set larger, indicating a lower confidence level in the initial value, so that the filter relies more on new measurements to quickly correct the state estimate in the initial stage. For example, if the initial estimate of oxygen consumption rate comes from offline measurements, its initial variance can be set to the square of 10 percent of the measured value; if it is merely a zero-value guess, its initial variance can be set to a larger value, such as the square of the average typical respiratory rate of the species.
[0030] Specifically, a dynamic gas mass balance equation is established, in which the rate of change of oxygen concentration is jointly determined by the nitrogen injection term from the nitrogen generator, the leakage term from the storage tank, and the oxygen consumption rate term of all fruit varieties; the rate of change of carbon dioxide concentration is jointly determined by the carbon dioxide production rate term of all fruit varieties, the removal term of the carbon dioxide removal device, and the leakage term from the storage tank; and the rate of change of ethylene concentration is jointly determined by the ethylene production rate term of all fruit varieties, the removal term of the ethylene removal device, and the leakage term from the storage tank.
[0031] Specifically, the dynamic gas mass balance equation is constructed with the gas in the storage chamber as the object of analysis, and the change in oxygen concentration per unit time is determined by the following three factors: The first item is the nitrogen injection factor. When the nitrogen generator injects high-purity nitrogen into the storage facility, it also introduces oxygen, causing a slight increase in the oxygen concentration inside the facility. The higher the nitrogen injection rate, the greater the contribution of this item.
[0032] The second item is the leakage factor. Due to the oxygen concentration difference between the inside and outside of the storage facility, oxygen from the outside air will seep into the facility through gaps, causing the oxygen concentration inside to tend to increase; conversely, if the oxygen concentration inside the facility is higher than that outside, oxygen will leak out. The magnitude of the leakage factor depends on the oxygen concentration difference between the inside and outside of the facility and the leakage rate of the facility.
[0033] The third item is the oxygen consumption due to fruit respiration. The respiration of all fruit varieties within the storage continuously consumes oxygen, causing a decrease in oxygen concentration. This consumption item equals the sum of the oxygen consumption rates of each fruit variety multiplied by their respective loading masses, divided by the volume of free gas within the storage.
[0034] In summary, the rate of change of oxygen concentration equals the contribution rate of nitrogen production and oxygen injection plus the oxygen exchange rate caused by leakage, minus the normalized value of the total oxygen consumption rate of all fruits relative to the volume of free gas in the storage.
[0035] Specifically, the change in carbon dioxide concentration per unit time is also determined by three factors: The first item is the fruit respiration production item, which is the sum of the carbon dioxide production rates of all varieties of fruit multiplied by their respective masses, and then divided by the volume of free gas in the storage, thus increasing the carbon dioxide concentration.
[0036] The second item is the removal item of the carbon dioxide removal device. The removal device extracts the gas in the storage room at a certain flow rate and removes the carbon dioxide from it before sending it back into the storage room. This item reduces the carbon dioxide concentration.
[0037] The third item is leakage from the storage facility. Since the concentration of carbon dioxide in the outside air is extremely low, leakage always causes the concentration of carbon dioxide inside the storage facility to tend to be at the outside level, which usually reduces the concentration of carbon dioxide inside the storage facility.
[0038] Specifically, the change in ethylene concentration per unit time is also determined by three factors: ethylene production from fruit, ethylene removal from the ethylene removal device, and leakage from the storage tank. The logic of this is consistent with the equation for the change in carbon dioxide concentration. The difference is that the source of ethylene production is the ethylene release rate of each variety of fruit, and the removal factor depends on the gas processing capacity and removal efficiency of the ethylene removal device.
[0039] The aforementioned dynamic gas mass balance equation can be expressed in the continuous time domain as three first-order differential equations, describing the rates of change of oxygen concentration, carbon dioxide concentration, and ethylene concentration over time. In practical digital control systems, these equations need to be discretized in time, i.e., the forward Euler method is used to transform the differential equations into recursive relationships between adjacent sampling times, so that the concentration value at the next time step can be calculated from the current concentration value and each input term.
[0040] Specifically, the gas mass balance equation is discretized over time to obtain the recursive relationship of gas concentration changes between adjacent sampling times. The variables to be estimated and the known quantities are set. The variables to be estimated include the oxygen consumption rate, carbon dioxide production rate and ethylene production rate of each fruit variety. The known quantities are the real-time sensor readings of oxygen concentration, carbon dioxide concentration and ethylene concentration in the storage.
[0041] Specifically, regarding the discretization and variable setting of the dynamic gas mass balance equations: In the continuous-time domain, the dynamic gas mass balance equations are a set of first-order differential equations, describing the rates of change of oxygen, carbon dioxide, and ethylene concentrations in the reservoir over time. To adapt to the sampling mechanism of the digital control system and construct a Kalman filter, these differential equations need to be transformed into discrete-time form. The discretization process is illustrated below using the oxygen concentration equation as an example; the discretization methods for the carbon dioxide and ethylene concentration equations are completely consistent.
[0042] The discretization process is implemented using the forward Euler method. The forward Euler method is a first-order numerical integration method whose core idea is to replace the differential with the difference, that is, to approximate the concentration change within a sampling period by multiplying the average rate of change over that period by the sampling period length. The forward Euler method is computationally simple and has good real-time performance, making it suitable for scenarios with high sampling frequencies and relatively gradual gas concentration changes in this embodiment. If the gas concentration in the controlled atmosphere storage changes drastically or the sampling period is long, the bilinear transform method or the zero-order hold method can be used to improve the discretization accuracy.
[0043] The specific derivation of the recursive relationship: Assume that the sampling period of the control system is a fixed time interval, and use the known quantity at the current moment to calculate the concentration value at the next sampling moment.
[0044] For oxygen concentration, the oxygen concentration at the next sampling time is equal to the oxygen concentration at the current sampling time, plus the contribution of the nitrogen generator injection term within this period, minus the contribution of the respiration consumption term of all fruit varieties within this period, and plus the contribution of the storage leakage term within this period. The contribution of the nitrogen generator injection term is equal to the current injection rate multiplied by the sampling period, multiplied by the air oxygen concentration, and divided by the free gas volume within the storage. This term is usually positive, indicating that the small amount of oxygen introduced during the nitrogen injection process slightly increases the oxygen concentration within the storage.
[0045] The contribution of fruit respiration consumption is equal to the sum of the oxygen consumption rates of each fruit variety multiplied by their respective masses, divided by the volume of free gas in the storage area, and then multiplied by the sampling period. This item has a negative contribution, causing a decrease in oxygen concentration.
[0046] The contribution of the leakage term is equal to the current leakage rate multiplied by the difference between the oxygen concentration outside the storage and the oxygen concentration inside the storage, divided by the volume of free gas inside the storage, and then multiplied by the sampling period. When the oxygen concentration inside the storage is lower than that outside, this term is positive, indicating that oxygen from the outside has seeped in.
[0047] For carbon dioxide concentration, the carbon dioxide concentration at the next sampling time is equal to the carbon dioxide concentration at the current sampling time, plus the contribution of the carbon dioxide production rate of all fruit varieties in this period, minus the contribution of the carbon dioxide removal device in this period, and plus the contribution of the storage leakage in this period. The carbon dioxide production rate is determined by the carbon dioxide production rate and mass of each variety; the removal rate is determined by the processing capacity and removal efficiency of the removal device and the current carbon dioxide concentration; and the leakage rate causes the carbon dioxide concentration inside the storage to tend towards extremely low levels compared to the outside environment.
[0048] For ethylene concentration, the recursive relationship is completely consistent with that of carbon dioxide concentration, except that the generation term is determined by the release rate of each type of ethylene, while the removal term is determined by the processing capacity and removal efficiency of the ethylene removal unit.
[0049] After deriving the discretized recursive relationship, the variables to be estimated and the known quantities are defined. The variables to be estimated are the unknowns that the Kalman filter needs to solve online, including: the oxygen consumption rate of each fruit variety at the current time, the carbon dioxide production rate of each fruit variety at the current time, the ethylene production rate of each fruit variety at the current time, and the gas leakage rate of the storage tank. Among them, the respiration rate reflects the real-time metabolic intensity of the fruit, and the leakage rate reflects the airtightness of the storage tank. These quantities cannot be directly measured by sensors and must be obtained indirectly through state estimation.
[0050] The known quantities are real-time measurements directly acquired by sensors in each sampling period, including sensor readings for oxygen concentration, carbon dioxide concentration, and ethylene concentration in the storage area. These are directly input into a Kalman filter as observations to correct the predictions of the variables to be estimated.
[0051] In addition, parameters such as the volume of free gas in the storage, the loading mass of each type of fruit, the working flow rate and removal efficiency of each execution device in the corresponding sampling period are all substituted into the recursive relationship as known inputs and are not involved in the state estimation.
[0052] Specifically, a Kalman filter is used for recursive estimation. In each sampling period, the estimated value of the variable to be estimated at the previous time step and the known quantity at the current time step are used to predict and correct the value through the discretized recursive relationship, and the carbon dioxide production rate and oxygen consumption rate of each fruit variety at the current time step are output.
[0053] Specifically, the respiratory quotient is determined by dividing the current carbon dioxide production rate by the oxygen consumption rate.
[0054] In this embodiment, the process of determining the physiological tolerance envelope of the fruit includes: Maintain baseline carbon dioxide concentration and baseline temperature, gradually reduce oxygen concentration, continuously monitor respiratory quotient, record current oxygen concentration when respiratory quotient first reaches baseline respiratory quotient, and determine the coordinates of hypoxia critical point based on current oxygen concentration, baseline carbon dioxide concentration and baseline temperature. Maintain a baseline oxygen concentration and a baseline temperature, gradually increase the carbon dioxide concentration, continuously monitor the respiratory quotient, and when the respiratory quotient first reaches the baseline respiratory quotient, record the current carbon dioxide concentration, and determine the coordinates of the high carbon dioxide critical point based on the current carbon dioxide concentration, baseline oxygen concentration, and baseline temperature. Maintain baseline oxygen and carbon dioxide concentrations, gradually decrease the temperature, continuously monitor the respiratory quotient, and when the respiratory quotient reaches the baseline respiratory quotient for the first time, record the current temperature and determine the coordinates of the low temperature critical point based on the current temperature, baseline oxygen concentration, and baseline carbon dioxide concentration. The physiological tolerance envelope was determined by combining the hypoxia critical point, the high carbon dioxide critical point, and the low temperature critical point using an ellipsoidal fitting method.
[0055] Specifically, the physiological tolerance envelope adopts an ellipsoidal form, which defines a multidimensional closed surface centered on the safety center point. The safety center point is jointly determined by the reference oxygen concentration, reference carbon dioxide concentration, reference temperature, and reference ethylene concentration of the fruit variety, representing the environmental state of the fruit under optimal storage conditions.
[0056] Specifically, the process of determining the physiological tolerance envelope using the ellipsoid fitting method is as follows: The safety center point is taken as the geometric center of the ellipsoid, which is determined by three coordinate values: the reference oxygen concentration, the reference carbon dioxide concentration, and the reference temperature for this fruit variety. The low-oxygen critical point, high-carbon dioxide critical point, low-temperature critical point, and high-temperature critical point, previously measured through stepwise deviation experiments, are taken as the boundary points that the ellipsoid surface must pass through or encompass.
[0057] Using the ellipsoidal equation as the pre-defined mathematical form for the surface, the general form of this ellipsoidal equation in a three-dimensional space composed of oxygen concentration, carbon dioxide concentration, and temperature is: the square of the ratio of the oxygen concentration deviation term to the oxygen dimension's semi-axis length, plus the square of the ratio of the carbon dioxide concentration deviation term to the carbon dioxide dimension's semi-axis length, plus the square of the ratio of the temperature deviation term to the temperature dimension's semi-axis length, the sum of these three equals one. Substituting the coordinates of each critical point into the above ellipsoidal equation, a system of equations is constructed with the three semi-axis lengths as unknowns. The least squares method is used to solve this system of equations, minimizing the fitting error between the ellipsoidal surface and each critical point, thereby determining the tolerance semi-axis lengths in the three dimensions. Substituting the coordinates of the safety center point and the fitted three semi-axis lengths back into the ellipsoidal equation yields the physiological tolerance envelope equation for this fruit variety.
[0058] For any given environmental state point, the method for determining whether it is located inside the physiological tolerance envelope is as follows: Calculate the differences between the oxygen concentration and the reference oxygen concentration, the carbon dioxide concentration and the reference carbon dioxide concentration, the temperature and the reference temperature, and the ethylene concentration and the reference ethylene concentration at that point; divide each of these four differences by the tolerance semi-axis length of the corresponding dimension, which includes the semi-axis lengths of the oxygen concentration dimension, carbon dioxide concentration dimension, temperature dimension, and ethylene concentration dimension; square the resulting four ratios and sum them; if the sum is less than or equal to one, the environmental state point is located inside or on the boundary of the physiological tolerance envelope, belonging to the safe zone; if the sum is greater than one, the environmental state point is located outside the physiological tolerance envelope, belonging to the danger zone.
[0059] Among them, the tolerance semi-axis length of each dimension is determined by ellipsoid fitting through multiple critical points measured experimentally. Each critical point is the coordinate combination of oxygen concentration, carbon dioxide concentration and temperature corresponding to the first time the respiratory quotient of the fruit variety reaches the baseline respiratory quotient.
[0060] In this embodiment, the process of determining the envelope scaling factor includes: Obtain several respiratory quotients within a preset period to generate a respiratory quotient sequence, and perform univariate linear regression on the respiratory quotient sequence to obtain the moving average slope. The average ethylene release rate during the stabilization period after storage is obtained as the reference ethylene release rate. The average ethylene release rate over a short time window is calculated as the real-time ethylene release rate. The relative rate of change was determined based on the reference ethylene release rate and the real-time ethylene release rate. The envelope scaling factor is determined based on the relationship between the moving average slope and the reference moving average slope, as well as the relationship between the relative rate of change and the reference relative rate of change. The ethylene release rate is determined by the ethylene concentration balance equation and a Kalman filter.
[0061] Specifically, the ethylene concentration balance equation is a differential equation that describes the relationship between the ethylene concentration in the reservoir and time. Its construction logic is the same as that of the carbon dioxide concentration balance equation, the difference being the gas generation source and the removal device.
[0062] The first term in the ethylene concentration balance equation is the ethylene production term from the fruit. The respiration and metabolism of all fruit varieties within the storage facility release ethylene, increasing the ethylene concentration. This production term equals the sum of the ethylene release rates of each fruit variety multiplied by their respective loading masses, divided by the volume of free gas within the storage facility.
[0063] The second term in the ethylene concentration balance equation is the removal term from the ethylene removal unit. The ethylene removal unit draws gas from the storage tank, removes ethylene through catalytic oxidation or adsorption, and then returns the gas to the storage tank, thus reducing the ethylene concentration. The magnitude of this removal term depends on the gas throughput, removal efficiency, and current ethylene concentration in the storage tank of the ethylene removal unit.
[0064] The third term in the ethylene concentration balance equation is the storage leakage term. Because the ethylene concentration in the outside air is extremely low, leakage always causes the ethylene concentration inside the storage chamber to tend towards the outside level, meaning it usually decreases the ethylene concentration inside the chamber. The magnitude of the leakage term depends on the storage leakage rate, the ethylene concentration difference between inside and outside the chamber, and the volume of free gas inside the chamber.
[0065] In summary, the rate of change of ethylene concentration is equal to the normalized value of the total ethylene release rate of all fruits relative to the volume of free gas in the storage, minus the removal contribution rate of the ethylene removal device, and then minus the ethylene loss rate caused by leakage.
[0066] Specifically, to determine the ethylene release rate, the ethylene concentration balance equation is first discretized using the forward Eulerian method to obtain the recursive relationship of ethylene concentration between adjacent sampling times. The ethylene concentration at the next sampling time is equal to the ethylene concentration at the current sampling time, plus the total ethylene production of all fruits in this period minus the amount of ethylene removed and leaked. Here, the ethylene production is calculated by multiplying the ethylene release rate of each variety by its respective mass, dividing by the volume of free gas in the storage tank, and then multiplying by the sampling period.
[0067] The second step involves estimating the ethylene release rate of each fruit variety. Known quantities include the real-time measurements from the ethylene concentration sensor, the gas throughput and removal efficiency of the ethylene removal device, the storage leakage rate, the loading mass of each fruit variety, and the volume of free gas within the storage facility. The storage leakage rate can be obtained from previously established oxygen or carbon dioxide identification data.
[0068] The third step involves treating the ethylene release rate of each fruit variety as a state variable and employing a random walk model as the state equation. This means the ethylene release rate at the next time step equals the current ethylene release rate plus a small amount of process noise, reflecting the natural fluctuations in the ethylene release rate over a short period. The discretized recursive relationship of ethylene concentration is then used as the observation equation, and together with the real-time measurements from the ethylene concentration sensor, is input into a Kalman filter for recursive estimation. The filter performs prediction and correction steps in each sampling period, outputting an estimated ethylene release rate for each fruit variety at the current time.
[0069] Specifically, the scaling factor of the envelope surface is determined as follows: The comprehensive deviation index is determined based on the moving average slope and the relative rate of change. The comprehensive deviation index is calculated as follows: the difference between the moving average slope and the benchmark moving average slope is divided by a preset slope normalization factor to obtain the first term; the difference between the relative rate of change and the benchmark relative rate of change is divided by a preset rate of change normalization factor to obtain the second term; the larger of the first and second terms is taken as the comprehensive deviation index. When both differences are negative or zero, the comprehensive deviation index is set to zero.
[0070] The envelope scaling factor is calculated based on the comprehensive deviation index and a preset mapping function. The mapping function has the following characteristics: when the comprehensive deviation index is zero, the output is 1, indicating that the envelope does not need to shrink; as the comprehensive deviation index increases, the output continuously and monotonically decreases from 1; when the comprehensive deviation index reaches or exceeds a preset upper threshold, the output is 0.8, indicating that the envelope has shrunk to its most stringent state; within the intermediate range, the output value smoothly changes with the comprehensive deviation index between 1 and 0.8. The mapping function can adopt a linear transition, an S-curve, or other function forms that satisfy the above monotonicity and boundary conditions.
[0071] Specifically, the slope normalization factor is used to divide the moving average slope of the respiratory quotient by a reference scale, converting it into a dimensionless relative deviation. It is determined by taking the typical value of the moving average slope of the respiratory quotient when the fruit variety is deemed to have experienced significant stress during storage. This typical value can be determined through historical data; that is, in the hypoxia tolerance limit experiment of the fruit variety, the level reached by the moving average slope of the respiratory quotient before and after the first breakthrough of the critical value is recorded, and the average value or conservative estimate of this level is used as the slope normalization factor.
[0072] Specifically, the rate of change normalization factor is used to divide the relative rate of change of ethylene release rate by a reference scale, converting it into a dimensionless relative deviation. It is determined by taking the typical value of the relative rate of change corresponding to when the variety of fruit is deemed to have abnormally activated ethylene during storage. This typical value can be determined based on the typical increase in the ethylene release rate relative to the stable period baseline value when the variety of fruit undergoes a ripening climacteric or is subjected to stress. For example, if the ethylene release rate of a certain variety typically rises to 1.5 to 2 times the stable period baseline value when a climacteric initiation occurs, the rate of change normalization factor can be set to 100%, meaning that a deviation of 1 is considered to occur when the relative rate of change reaches 100%.
[0073] Specifically, the preset cycle should take into account both response speed and anti-interference ability, and should cover at least one complete gas regulation cycle of the controlled atmosphere storage, so that the respiratory quotient sequence can characterize the true physiological response of the fruit to environmental adjustment. At the same time, it should not exceed the typical time scale of significant changes in the physiological state of the fruit, which can be set for example to be 6 hours to 48 hours.
[0074] Specifically, the baseline moving average slope is a reference zero point used to determine whether the respiratory quotient trend is abnormal. Under ideal healthy aerobic respiration conditions, the respiratory quotient of fruit should remain stable, and its moving average slope should theoretically be zero. Any continuous positive deviation means that the respiratory quotient is monotonically rising, which is an early signal that the fruit is sliding towards anaerobic respiration or is under stress. Therefore, zero is used as the baseline moving average slope.
[0075] Specifically, the baseline relative change rate is the reference zero point for determining whether ethylene release is abnormally activated. When the real-time ethylene release rate is equal to the reference ethylene release rate, any sustained positive value means that the ethylene generation system has been additionally activated, which may be due to jump initiation, stress response or pathological reaction. Therefore, zero is used as the baseline relative change rate.
[0076] Specifically, the stabilization period after storage refers to the initial time required for the respiration metabolism and ethylene release of fruits to reach a relatively stable state under the set optimal storage environment after they are stored. Specifically, the timing starts after the environmental parameters reach the target set value and stabilize, and the respiration quotient and ethylene release rate are continuously monitored. When the fluctuation range of the respiration quotient and ethylene release rate remains within ±5% for 12 consecutive hours, it is determined that the fruit has entered the stabilization period.
[0077] Specifically, a short window refers to a relatively short time interval used to calculate the average real-time ethylene release rate before the current control cycle. In particular, its length should be sufficient to capture the natural fluctuations of sensor noise and breathing while capturing recent trends.
[0078] In this embodiment, the process of generating the safety envelope includes: The physiological tolerance envelope is reduced based on the envelope scaling factor to generate a safe envelope.
[0079] Specifically, the process of generating the safety envelope is as follows: The first step is to read the coordinates of the safety center point of the fruit and the basic tolerance semi-axis length of each dimension from the established physiological tolerance envelope data. The basic tolerance semi-axis length is the original semi-axis length of the physiological tolerance envelope when it is not scaled.
[0080] The second step is to obtain the envelope scaling factor calculated within the current preset period. The smaller the envelope scaling factor, the more severe the decline in the fruit's current physiological tolerance compared to its baseline state.
[0081] The third step is to multiply the basic tolerance semi-axis length of the oxygen concentration dimension by the envelope scaling factor to obtain the scaled oxygen concentration semi-axis length; multiply the basic tolerance semi-axis length of the carbon dioxide concentration dimension by the envelope scaling factor to obtain the scaled carbon dioxide concentration semi-axis length; multiply the basic tolerance semi-axis length of the temperature dimension by the envelope scaling factor to obtain the scaled temperature semi-axis length; and multiply the basic tolerance semi-axis length of the ethylene concentration dimension by the envelope scaling factor to obtain the scaled ethylene concentration semi-axis length.
[0082] The fourth step is to keep the coordinates of the safety center point unchanged, replace the original basic tolerance semi-axis length with the scaled semi-axis lengths of each dimension, and together with the safety center point, form a new ellipsoid equation. The new ellipsoid equation is the safety envelope surface at the current moment.
[0083] It should be noted that in the scenario of mixed storage of multiple varieties, this embodiment coordinates the differences between different varieties in the following way.
[0084] Regarding the generation and maintenance of the safety envelope: Different fruit varieties have their own independent physiological tolerance envelope and safety envelope. Their baseline envelope parameters, critical point coordinates, and tolerance semi-axis lengths are established and stored separately for each variety. During mixed storage, the system independently generates and dynamically maintains a dedicated safety envelope for each variety in the storage facility. The scaling factor of the envelope for each variety is also independently calculated based on the real-time monitored moving average slope of the respiratory quotient and the relative change rate of ethylene release rate.
[0085] Regarding the determination of envelope boundary crossing events: In each control cycle, the system compares each prediction point in the environmental prediction sequence with the safe envelope of each variety. If any variety's prediction point falls outside its safe envelope, an envelope boundary crossing event is determined to have occurred. At this time, the system records the identity of the crossing variety, its crossing time step, and its contribution in each dimension, and uses this variety as the subject of the current boundary crossing event for subsequent conflict mode analysis. If multiple varieties cross the boundary simultaneously, the variety with the largest crossing distance is prioritized for processing, and the information of the remaining crossing varieties is used as superimposed constraints.
[0086] Regarding the execution of the classification and processing strategy: The classification and processing strategy is issued to the execution equipment of the controlled atmosphere storage system, not to individual varieties. When a variety crossing the boundary and its conflict mode are identified, the system adjusts the equipment operating parameters according to the corresponding strategy. These adjustments affect the overall shared gas environment within the storage facility; therefore, while correcting the dangerous state of the crossing-boundary variety, it inevitably affects other varieties as well. To ensure that the adjustments do not trigger new crossings, the corrected control parameters are re-entered into the controlled atmosphere storage environment dynamic simulator for closed-loop verification before being issued for execution. The correction strategy is only executed when the simulation results show that the environmental state points of all varieties within the entire prediction time domain return to their respective safety envelopes, fundamentally guaranteeing the safety of multi-variety coordination.
[0087] Regarding special cases where coordination is impossible: If, after the above closed-loop verification, no set of control parameters can simultaneously ensure that all varieties are within their respective safety envelopes, the system determines that the current mixed storage combination has an irreconcilable physiological conflict under the existing equipment capabilities. In this case, the system triggers the highest-level alarm and outputs suggestions to the operator, including but not limited to: prioritizing the safety of high-value varieties and providing corresponding control parameter schemes.
[0088] This embodiment takes into account that the physiological tolerance of fruits gradually decreases during long-term storage due to factors such as changes in maturity and cumulative stress. It collects the moving average slope of the respiratory quotient and the relative change rate of ethylene release rate in real time. The moving average slope of the respiratory quotient determines whether respiratory metabolism is slowly sliding towards anaerobic fermentation, and the relative change rate of ethylene release rate determines whether the ethylene production system is abnormally activated. The envelope scaling factor is determined by the moving average slope of the respiratory quotient and the relative change rate of ethylene release rate. The semi-axial length of each dimension of the physiological tolerance envelope is contracted to generate a safe envelope, so that the safety boundary can be adapted and adjusted according to the changes in the actual physiological state of the fruit, thus achieving a dynamic balance between control precision and storage safety.
[0089] In this embodiment, the process of obtaining the predicted oxygen consumption rate sequence includes: A predicted oxygen consumption rate sequence is generated based on several predicted oxygen consumption rates calculated from the oxygen consumption rate, the rate of change of the oxygen consumption rate, the preset period, and the predicted time domain length.
[0090] When the absolute value of the rate of change of oxygen consumption rate is less than or equal to the first rate of change threshold, the oxygen consumption trend stage is identified as a stable period. The stable prediction mode is selected, and the oxygen consumption rate at the current moment is determined as a constant value in the prediction time domain to generate a predicted oxygen consumption rate sequence. When the absolute value of the rate of change of oxygen consumption rate is greater than the first rate of change threshold and less than or equal to the second rate of change threshold, the oxygen consumption trend stage is identified as the transition period. A linear prediction mode is selected, and the oxygen consumption rate, the rate of change of oxygen consumption rate and the prediction time domain length are used to calculate and generate a predicted oxygen consumption rate sequence. When the absolute value of the rate of change of oxygen consumption rate is greater than the second rate of change threshold, the oxygen consumption trend stage is identified as the acceleration period. The modified linear prediction mode is selected, and the rate of change of oxygen consumption rate is multiplied by the preset correction coefficient. Then, the oxygen consumption rate, the modified rate of change of oxygen consumption rate and the prediction time domain length are used to calculate and generate the predicted oxygen consumption rate sequence. Among them, the first rate of change threshold is less than the second rate of change threshold.
[0091] Specifically, the first rate of change threshold is used to distinguish between the stable period and the transition period of oxygen consumption trends. Its physical meaning is the upper limit of the natural fluctuation of the oxygen consumption rate of fruit under steady-state storage conditions. It is determined by taking the statistical upper limit of the absolute value of the rate of change of oxygen consumption rate for that fruit variety during the stable period after storage. Specifically, during the stable period after storage, the rate of change of oxygen consumption is collected continuously for multiple sampling periods, and the 95th percentile of the absolute value sequence of the rate of change is taken as the first rate of change threshold.
[0092] Specifically, the second rate of change threshold is used to distinguish between the transitional and accelerated phases of oxygen consumption trends. Physically, it represents the lower limit of the oxygen consumption rate change when the fruit has entered a state of significant stress or climax. It is determined by taking the lower limit of the typical level reached by the absolute value of the oxygen consumption rate change under known stress conditions or climax stages for that fruit variety. This can be determined through preliminary experiments: in hypoxia tolerance limit experiments or climax monitoring experiments for that fruit variety, the minimum absolute value of the oxygen consumption rate change is recorded when the fruit transitions from a normal state to a stress or accelerated climax stage, and the average value is taken as the second rate of change threshold.
[0093] Specifically, the process of obtaining the predicted oxygen consumption rate sequence is as follows: The first step is to obtain the oxygen consumption rate at the current sampling time to determine the real-time respiration intensity of the fruit under the current environment. At the same time, the rate of change of the oxygen consumption rate at the current sampling time is obtained to determine the trend and speed of increase or decrease of the oxygen consumption rate per unit time. This can be obtained directly through a filter.
[0094] The second step is to determine the current preset cycle.
[0095] The third step is to calculate from the current sampling time. For the first future time step, the predicted oxygen consumption rate is equal to the oxygen consumption rate at the current sampling time plus the rate of change of the oxygen consumption rate multiplied by a preset period. For the second future time step, the predicted oxygen consumption rate is equal to the oxygen consumption rate at the current sampling time plus the rate of change of the oxygen consumption rate multiplied by two preset periods. And so on. For any future time step, the predicted oxygen consumption rate is equal to the oxygen consumption rate at the current sampling time plus the rate of change of the oxygen consumption rate multiplied by the corresponding time step and the preset period.
[0096] Fourth, if any predicted oxygen consumption rate calculated above is negative, then replace that value with zero, because the oxygen consumption rate cannot be negative physically. The appearance of a negative value only means that the mathematical extrapolation has exceeded the reasonable range. In this case, zero should be used as its lower limit.
[0097] The fifth step involves arranging all the predicted oxygen consumption rates calculated for the first time step, the second time step, and up to the last time step corresponding to the predicted time domain length in chronological order. This forms the predicted oxygen consumption rate sequence for the fruit over the preset period. Each value in the predicted oxygen consumption rate sequence represents the predicted oxygen consumption intensity of the fruit at the corresponding future time.
[0098] Please see Figure 2 As shown, this is a logic diagram for determining whether a fruit has experienced an envelope boundary crossing event according to an embodiment of the present invention. The process for determining whether a fruit has experienced an envelope boundary crossing event includes: Substitute several predicted points into the ellipse equation corresponding to the safe envelope to obtain several predicted distances; In response to any predicted distance being greater than the reference distance, an envelope boundary crossing event is determined for the fruit. In response that all predicted distances are less than or equal to the baseline distance, it is determined that no envelope boundary crossing event has occurred for the fruit. The prediction point is determined based on the predicted oxygen consumption rate.
[0099] In practical implementation, the controlled atmosphere storage environment dynamic simulator operates based on the principle of gas mass balance. Its input parameters include: the total loading mass of each fruit variety, obtained through weighing upon entry into the storage (e.g., 50,000 kg of Fuji apples); the predicted oxygen consumption rate sequence, calculated in real-time by the upstream prediction step; the volume of free gas within the storage, calculated by subtracting the volume occupied by the fruit and packaging from the total storage volume (e.g., 800 m³); the gas leakage rate of the storage, obtained through empty storage pressure decay testing or online identification (e.g., 0.5 m³ / h); the maximum nitrogen injection rate of the nitrogen generator (e.g., 15 m³ / h); the maximum gas throughput and removal efficiency of the carbon dioxide removal machine (e.g., 10 m³ / h, 90%); the maximum gas throughput and removal efficiency of the ethylene removal machine (e.g., 5 m³ / h, 85%); and the standard concentrations of each gas component in the air. The simulator's initial conditions are set to the measured oxygen concentration, measured carbon dioxide concentration, measured ethylene concentration, and measured temperature within the storage at the current sampling time.
[0100] Specifically, the prediction point is the environmental state point calculated further by a controlled atmosphere storage environment dynamic simulator based on the predicted oxygen consumption rate sequence. Specifically, the predicted oxygen consumption rate sequence of each fruit variety obtained in the current control cycle is input into the controlled atmosphere storage environment dynamic simulator. The controlled atmosphere storage environment dynamic simulator operates according to the gas mass balance principle of the controlled atmosphere storage. Based on the known current gas concentration in the storage, the execution status of the current control variables, the leakage rate of the storage, and the mass of each fruit variety, the predicted oxygen consumption rate of each variety is substituted into the mass balance equation step by step to calculate the changes in oxygen concentration, carbon dioxide concentration, and ethylene concentration in the storage at each future time step, thereby obtaining the predicted oxygen concentration, predicted carbon dioxide concentration, and predicted ethylene concentration at each future time step. At the same time, the predicted temperature at each future time step is determined by combining the set value of the temperature control loop.
[0101] Then, the predicted oxygen concentration, predicted carbon dioxide concentration, predicted temperature and predicted ethylene concentration corresponding to each future time step are combined into an environmental state point for the corresponding time step, namely a prediction point. A prediction point corresponds to the predicted value of the multidimensional environmental state in the controlled atmosphere storage at a future time, including four dimensions: oxygen concentration, carbon dioxide concentration, temperature and ethylene concentration.
[0102] Specifically, the baseline distance is the critical value for determining whether a predicted point lies inside the safe envelope. Its value is one. According to the mathematical definition of an ellipsoid, when the distance is less than one, the predicted point lies inside the ellipsoid; when the distance is equal to one, the predicted point lies exactly on the ellipsoid; and when the distance is greater than one, the predicted point lies outside the ellipsoid. Therefore, using one as the baseline distance is a natural threshold for determining whether a predicted point falls inside the safe envelope and requires no further calibration.
[0103] Specifically, the process of determining whether a fruit has experienced an envelope boundary crossing event is as follows: For each prediction point in the environmental prediction sequence, extract its oxygen concentration, carbon dioxide concentration, temperature and ethylene concentration, calculate the difference between the oxygen concentration of the prediction point and the baseline oxygen concentration of the safety center point, divide the difference by the scaled tolerance half-axis length of the oxygen concentration dimension, and square the resulting quotient to obtain the first term. Calculate the difference between the carbon dioxide concentration at the predicted point and the baseline carbon dioxide concentration at the safety center point. Divide this difference by the scaled tolerance semi-axis length of the carbon dioxide concentration dimension, and square the resulting quotient to obtain the second term. Calculate the difference between the temperature at the predicted point and the reference temperature at the safety center point, divide the difference by the scaled tolerance semi-axis length of the temperature dimension, and square the resulting quotient to obtain the third term. Calculate the difference between the ethylene concentration at the predicted point and the baseline ethylene concentration at the safety center point. Divide this difference by the scaled tolerance semi-axis length of the ethylene concentration dimension, and square the resulting quotient to obtain the fourth term. Summing the above four items gives the predicted distance to the predicted point.
[0104] Specifically, if the predicted distance of any one of the predicted points is greater than the baseline distance, an envelope boundary crossing event is determined to have occurred, indicating that at least at some future moment, the environmental state inside the controlled atmosphere storage facility will break through the current safe envelope of the fruit and enter the danger zone. If the predicted distance of all predicted points is less than or equal to the baseline distance, an envelope boundary crossing event is determined to have occurred, indicating that in the current prediction results, the environmental state throughout the entire future prediction time domain is within the safe envelope, and the fruit is in a safe state.
[0105] In this embodiment, the process of determining the conflict mode of the envelope boundary crossing event includes: Several contribution levels are determined based on the elliptic equations corresponding to the distance and the safety envelope surface; The conflict mode is determined based on the maximum value among several contribution levels.
[0106] Specifically, the contribution level is used to quantify the relative responsibility of the four dimensions—oxygen, carbon dioxide, temperature, and ethylene—for the total out-of-bounds distance in an envelope out-of-bounds event, and is the core basis for determining the conflict mode.
[0107] Specifically, the process for determining contribution is as follows: When an envelope boundary crossing event is determined, the prediction point with the largest distance is selected as the boundary crossing analysis point. The boundary crossing distance of the boundary crossing analysis point is obtained by summing the four components. The first component is the square of the difference between the oxygen concentration at the out-of-bounds analysis point and the baseline oxygen concentration at the safety center point, divided by the square of the length of the half-axis after scaling the oxygen dimension. The second component is the square of the difference between the carbon dioxide concentration at the out-of-bounds analysis point and the baseline carbon dioxide concentration at the safety center point, divided by the square of the half-axis length after scaling the carbon dioxide dimension. The third component is the square of the difference between the temperature at the boundary analysis point and the reference temperature at the safety center point, divided by the square of the length of the half-axis after scaling the temperature dimension. The fourth component is the square of the difference between the ethylene concentration at the out-of-bounds analysis point and the baseline ethylene concentration at the safety center point, divided by the square of the half-axis length after scaling the ethylene dimension. Divide the first component by the distance of the out-of-bounds analysis point, and the resulting quotient is the oxygen contribution. Divide the second component by the distance of the outbound analysis point, and the resulting quotient is the carbon dioxide contribution. Divide the third component by the distance of the out-of-bounds analysis point, and the resulting quotient is the temperature contribution. Divide the fourth component by the distance of the out-of-bounds analysis point, and the resulting quotient is the ethylene contribution.
[0108] Specifically, the contribution values are all between zero and one, and the sum of the contributions of oxygen, carbon dioxide, temperature, and ethylene is equal to one. The larger the contribution, the more dominant the degree of deviation of the corresponding environmental parameter from the safety center is in the total boundary crossing, which is the primary cause of this boundary crossing event.
[0109] This embodiment extrapolates the real-time oxygen consumption rate and its rate of change using linear trends to obtain a predicted oxygen consumption rate sequence. This sequence is then transformed into an environmental prediction sequence containing multiple future time steps using a controlled atmosphere storage environment dynamic simulator. The environmental state at each predicted point is substituted into the elliptical equation of the safety envelope to calculate the dimensionless distance. When any distance exceeds the baseline distance, an envelope boundary crossing event is determined to have occurred. By decomposing the contribution of each dimension to the total boundary crossing distance, the dominant boundary crossing dimension is identified, and the boundary crossing event is classified into hypoxia stress conflict, carbon dioxide stress conflict, ethylene-induced stress conflict, or mixed stress conflict. This allows for early warning before the fruit is actually damaged and accurately identifies the root cause dimension of the problem, providing clear guidance for targeted treatment.
[0110] In this embodiment, the process of determining the conflict mode of the envelope boundary crossing event further includes: Compare the contributions of oxygen, carbon dioxide, temperature, and ethylene, and find the maximum value among them; If the maximum value is the oxygen contribution, then the conflict mode is determined to be a hypoxia stress conflict. This mode indicates that in this envelope boundary crossing event, the deviation of oxygen concentration from the safety center is the main cause, which means that the ambient oxygen concentration has dropped below the lower limit allowed by the current dynamic tolerance capacity of the fruit, and the fruit faces the risk of anaerobic respiration stress.
[0111] If the maximum value is the contribution of carbon dioxide, then the conflict mode is determined to be a carbon dioxide stress conflict. This mode indicates that the deviation of carbon dioxide concentration from the safety center is the main cause, which means that the environmental carbon dioxide concentration has exceeded the upper limit allowed by the current dynamic tolerance capacity of the fruit, and the fruit faces the stress risk of high carbon dioxide inhibiting normal respiratory metabolism.
[0112] If the maximum value is the contribution of ethylene, then the conflict mode is determined to be ethylene-induced stress conflict. This mode indicates that the deviation of ethylene concentration from the safety center is the most important inducing factor, which means that the accumulation of ethylene in the storage has reached the level that constitutes independent stress on the fruit, and the fruit faces the risk of being accelerated ripening and aging.
[0113] If the maximum value is the temperature contribution, the conflict mode is determined to be a mixed stress conflict. This mode indicates that the temperature deviation from the safety center is the primary cause of this boundary violation. However, in this embodiment, the significant deviation of the temperature dimension does not occur independently, but appears as a synergistic factor that exacerbates gas stress. Therefore, when the temperature contribution is at its maximum value, regardless of the contribution of other dimensions, the conflict mode is determined to be a mixed stress conflict, so as to trigger the highest safety level control response in a timely manner and comprehensively adjust all environmental parameters.
[0114] Understandably, before determining the conflict pattern according to the above rules, it can be first determined whether the maximum value of the contribution is significantly greater than the other contributions. If the maximum value of the contribution exceeds 0.5, that is, the contribution of a single dimension exceeds the sum of the other three dimensions, it indicates that there is a clear dominant cause, and it is classified according to the above rules. If all contributions are less than 0.5, that is, no dimension occupies an absolute dominant position, then regardless of which dimension the maximum value of the contribution belongs to, it is directly determined to be a mixed coercive conflict, indicating that this boundary crossing event was caused by multiple environmental parameters deviating from the security center at the same time.
[0115] Please see Figure 3 As shown, this is a logic decision diagram for determining the classification processing strategy of the security envelope surface in an embodiment of the present invention. The process of determining the classification processing strategy includes: If the conflict mode is hypoxia stress conflict, the classification and treatment strategy is to raise the lower limit of the oxygen concentration control target. If the conflict mode is carbon dioxide stress conflict, then the classification and treatment strategy is to increase the power of the carbon dioxide removal device. If the conflict mode is ethylene-induced stress conflict, then the classification and handling strategy is to increase the operating frequency of the ethylene removal unit. If the conflict mode is a mixed coercive conflict, then the classification and handling strategy is to execute the preset highest security level control mode.
[0116] Specifically, the classification and treatment strategies corresponding to hypoxia stress conflict include: Based on the deviation between the oxygen concentration value at the boundary analysis point and the baseline oxygen concentration value at the safety center point, determine the required adjustment range of the oxygen concentration. Convert the required callback magnitude into a command to reduce the nitrogen injection volume of the nitrogen generator; The lower limit of the oxygen concentration control target is determined based on the reduction instruction.
[0117] It should be noted that, under hypoxia stress, the method for determining the oxygen concentration correction range is as follows: the absolute value of the difference between the oxygen concentration value at the out-of-bounds analysis point and the baseline oxygen concentration value at the safety center point is added to a preset safety margin. The result is the required correction range for the oxygen concentration. The safety margin is calculated by multiplying the current oxygen dimension's scaled half-axis length by a preset scaling factor, with the scaling factor ranging from 0.1 to 0.3.
[0118] Specifically, the classification and treatment strategies corresponding to carbon dioxide stress conflicts include: The required additional amount of carbon dioxide removal is determined based on the extent to which the carbon dioxide concentration at the out-of-bounds analysis point exceeds the baseline carbon dioxide concentration at the safety center point. This embodiment uses a membrane separation desorption machine, and the compressor power is increased to improve the gas processing capacity according to the required increase in carbon dioxide removal.
[0119] Specifically, the classification and treatment strategies for ethylene-induced stress conflicts include: This embodiment uses a high-energy ultraviolet catalytic oxidation ethylene removal machine, which increases the working frequency of the air circulation fan, thereby increasing the volume of gas flowing through the reaction chamber per unit time and thus accelerating the oxidative decomposition of ethylene.
[0120] Specifically, the classification and handling strategies for mixed coercive conflicts include: The target for controlling the oxygen concentration in the storage facility is set as the maximum value among the baseline oxygen concentrations at the safety center point for each variety, ensuring that none of the varieties are stressed by low oxygen. The target for controlling carbon dioxide concentration in the storage facility is set as the minimum value among the baseline carbon dioxide concentrations at the safety center point for each variety, ensuring that none of the varieties are stressed by high carbon dioxide levels. Set the temperature inside the storage chamber to the lowest safe temperature that each variety can withstand under the current envelope scaling factor; switch the ethylene removal unit to full-power continuous operation mode; At the same time, the highest level alarm is issued to the operators, indicating that the physiological state of the current mixed storage combination has entered a high-risk zone, and it is recommended to check the equipment operation status and assess whether it is necessary to terminate the storage of some varieties in advance or to carry out separate storage.
[0121] This embodiment establishes a mapping relationship between four conflict modes and corresponding classification and processing strategies in advance: low oxygen stress conflict corresponds to raising the lower limit of the oxygen concentration control target; carbon dioxide stress conflict corresponds to increasing the power of the carbon dioxide removal device; ethylene-induced stress conflict corresponds to increasing the operating frequency of the ethylene removal device; and mixed stress conflict corresponds to executing the preset highest safety level control mode. When the conflict mode of the envelope boundary event is determined, the system calls the corresponding classification and processing strategy to perform directional correction on the control variables of the modified atmosphere preservation system, and re-inputs the corrected control target into the environmental dynamic simulator for verification. The system only issues the execution when the new predicted sequence is entirely within the safety envelope, thereby improving the targeting of modified atmosphere preservation.
[0122] Please see Figure 4 The diagram shown is a block diagram of a smart control system for post-harvest processing of fruit based on modified atmosphere preservation, according to an embodiment of the present invention. The smart control system for post-harvest processing of fruit based on modified atmosphere preservation includes: The data acquisition module is used to obtain the gas flow rate, oxygen concentration, and carbon dioxide concentration of the controlled atmosphere storage, collect the moving average slope of the respiratory quotient of the fruit and the relative change rate of ethylene release rate, and obtain the oxygen consumption rate and the change rate of oxygen consumption rate of the fruit. The envelope generation module, which is connected to the data acquisition module, is used to determine the respiratory quotient in the current storage room based on the gas flow rate, oxygen concentration and carbon dioxide concentration of the controlled atmosphere storage room, determine the envelope scaling factor based on the relative change rate of the moving average slope and ethylene release rate, and adjust the preset physiological tolerance envelope based on the envelope scaling factor to generate a safe envelope. The environmental prediction module, which is connected to the data acquisition module, is used to determine the environmental prediction sequence by identifying the oxygen consumption trend stage and selecting the corresponding prediction mode based on the oxygen consumption rate and the rate of change of the oxygen consumption rate. The event analysis module, which is connected to the envelope generation module and the environment prediction module respectively, is used to determine whether the fruit has experienced an envelope boundary crossing event based on the environment prediction sequence, and to determine the conflict mode of the envelope boundary crossing event. The classification and processing module, which is connected to the event analysis module, is used to determine the classification and processing strategy of the modified atmosphere preservation equipment based on the conflict mode, and to perform modified atmosphere preservation treatment on the fruit based on the classification and processing strategy.
[0123] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.
Claims
1. An intelligent control method for postharvest handling of fruits based on controlled atmosphere storage, characterized in that, include: The fruit was placed in a controlled atmosphere storage room, and the gas flow rate, oxygen concentration, and carbon dioxide concentration of the controlled atmosphere storage room were obtained to determine the current respiratory quotient in the storage room. The moving average slope of the respiratory quotient corresponding to the fruit and the relative change rate of ethylene release rate are collected to determine the envelope scaling factor. Based on the envelope scaling factor, the preset physiological tolerance envelope is adjusted to generate a safe envelope. The oxygen consumption rate of the fruit at the current moment and the rate of change of the oxygen consumption rate within the preset storage time are obtained. The oxygen consumption trend stage of the fruit is identified based on the rate of change of the oxygen consumption rate. The corresponding prediction mode is selected to extrapolate the trend of the oxygen consumption rate to obtain the predicted oxygen consumption rate sequence. The predicted oxygen consumption rate sequence is input into the controlled atmosphere storage environment dynamic simulator to obtain the environmental prediction sequence containing multiple future time steps of predicted oxygen concentration, predicted carbon dioxide concentration, predicted ethylene concentration and predicted temperature. Determine whether the fruit has experienced an envelope boundary crossing event based on the environmental prediction sequence, and determine the conflict mode of the envelope boundary crossing event; The classification and processing strategy of the modified atmosphere preservation equipment is determined according to the conflict mode, and the fruit is subjected to modified atmosphere preservation treatment based on the classification and processing strategy. The conflict modes include hypoxia stress conflict, carbon dioxide stress conflict, ethylene-induced stress conflict and mixed stress conflict; The classification and processing strategy includes raising the lower limit of the oxygen concentration control target, increasing the power of the carbon dioxide removal device, increasing the operating frequency of the ethylene removal device, and implementing the preset highest safety level control mode.
2. The intelligent control method for postharvest treatment of fruits based on controlled atmosphere preservation according to claim 1, characterized in that, The process of determining the respiratory quotient in the current library includes: Oxygen concentration, carbon dioxide concentration, and ethylene concentration were obtained to establish a dynamic gas mass balance equation; The dynamic gas mass balance equation is discretized, and the oxygen consumption rate and carbon dioxide production rate of each fruit variety are determined by Kalman filter. The respiratory quotient is determined based on the oxygen consumption rate and the carbon dioxide production rate.
3. The intelligent control method for postharvest treatment of fruits based on controlled atmosphere preservation according to claim 2, characterized in that, The process of determining the envelope scaling factor includes: Obtain several respiratory quotients within a preset period to generate a respiratory quotient sequence, and perform univariate linear regression on the respiratory quotient sequence to obtain the moving average slope. The average ethylene release rate during the stabilization period after storage is obtained as the reference ethylene release rate. The average ethylene release rate over a short time window is calculated as the real-time ethylene release rate. The relative rate of change is determined based on the reference ethylene release rate and the real-time ethylene release rate; The envelope scaling factor is determined based on the relationship between the moving average slope and the reference moving average slope, and the relationship between the relative rate of change and the reference relative rate of change. The ethylene release rate is determined by the ethylene concentration balance equation and a Kalman filter.
4. The intelligent control method for postharvest treatment of fruits based on controlled atmosphere preservation according to claim 3, characterized in that, The process of generating the security envelope includes: The physiological tolerance envelope is reduced based on the envelope scaling factor to generate the safety envelope.
5. The intelligent control method for post-harvest processing of fruits based on modified atmosphere preservation according to claim 4, characterized in that, The process of obtaining the predicted oxygen consumption rate sequence includes: When the absolute value of the rate of change of the oxygen consumption rate is less than or equal to the first rate of change threshold, the oxygen consumption trend stage is identified as a stable period, a stable prediction mode is selected, and the oxygen consumption rate at the current moment is determined as a constant value in the prediction time domain to generate the predicted oxygen consumption rate sequence. When the absolute value of the rate of change of the oxygen consumption rate is greater than the first rate of change threshold and less than or equal to the second rate of change threshold, the oxygen consumption trend stage is identified as a transition period. A linear prediction mode is selected, and the predicted oxygen consumption rate sequence is generated by calculating based on the oxygen consumption rate, the rate of change of the oxygen consumption rate, and the prediction time domain length. When the absolute value of the rate of change of the oxygen consumption rate is greater than the second rate of change threshold, the oxygen consumption trend stage is identified as the acceleration period. A modified linear prediction mode is selected, and the rate of change of the oxygen consumption rate is multiplied by a preset correction coefficient. Then, the predicted oxygen consumption rate sequence is generated based on the oxygen consumption rate, the modified rate of change of the oxygen consumption rate, and the predicted time domain length.
6. The intelligent control method for post-harvest processing of fruits based on modified atmosphere preservation according to claim 5, characterized in that, The process of determining whether a fruit has experienced an envelope boundary crossing event includes: Substitute several predicted points into the ellipse equation corresponding to the safe envelope to obtain several predicted distances; In response to any predicted distance being greater than the reference distance, an envelope boundary crossing event is determined for the fruit. The prediction point is determined based on the predicted oxygen consumption rate.
7. The intelligent control method for post-harvest processing of fruits based on modified atmosphere preservation according to claim 6, characterized in that, The process of determining the conflict mode of an envelope boundary overflow event includes: Several contribution values are determined based on the distance and the ellipse equation corresponding to the safety envelope surface; The conflict mode is determined based on the maximum value among several of the stated contributions.
8. The intelligent control method for post-harvest processing of fruits based on modified atmosphere preservation according to claim 7, characterized in that, The process of determining the conflict mode of envelope boundary crossing events also includes: If the maximum value is the oxygen contribution, then the conflict mode is determined to be the hypoxia stress conflict; If the maximum value is the contribution of carbon dioxide, then the conflict mode is determined to be the carbon dioxide stress conflict; If the maximum value is the contribution of ethylene, then the conflict mode is determined to be the ethylene-induced stress conflict; If the maximum value is the temperature contribution, then the conflict mode is determined to be the mixed stress conflict.
9. The intelligent control method for post-harvest processing of fruits based on modified atmosphere preservation according to claim 8, characterized in that, The process of determining the classification and processing strategy includes: If the conflict mode is a hypoxia stress conflict, the strategy is to raise the lower limit of the oxygen concentration control target. If the conflict mode is a carbon dioxide stress conflict, the strategy is to increase the power of the carbon dioxide removal device. If the conflict mode is an ethylene-induced stress conflict, the strategy is to increase the operating frequency of the ethylene removal unit. If the conflict mode is a mixed coercive conflict, the strategy is to execute the preset highest security level control mode.
10. An intelligent control system applied to the intelligent control method for post-harvest processing of fruits based on modified atmosphere preservation as described in any one of claims 1-9, characterized in that, include: The data acquisition module is used to obtain the gas flow rate, oxygen concentration and carbon dioxide concentration of the controlled atmosphere storage, collect the moving average slope of the respiratory quotient of the fruit and the relative change rate of ethylene release rate, and obtain the oxygen consumption rate and the change rate of oxygen consumption rate of the fruit. The envelope generation module is used to determine the respiratory quotient in the current controlled atmosphere storage based on the gas flow rate, oxygen concentration and carbon dioxide concentration, determine the envelope scaling factor based on the relative change rate of the moving average slope and the ethylene release rate, and adjust the preset physiological tolerance envelope based on the envelope scaling factor to generate a safe envelope. An environmental prediction module is used to determine an environmental prediction sequence by identifying oxygen consumption trend stages and selecting corresponding prediction modes based on the oxygen consumption rate and the rate of change of the oxygen consumption rate; The event analysis module is used to determine whether the fruit has experienced an envelope boundary crossing event based on the environmental prediction sequence, and to determine the conflict mode of the envelope boundary crossing event. The classification processing module is used to determine the classification processing strategy of the modified atmosphere preservation equipment according to the conflict mode, and to perform modified atmosphere preservation treatment on the fruit based on the classification processing strategy.
Citation Information
Patent Citations
Gas adjusting method and device for fruit and vegetable storage space, medium, product and chip
CN121444958A
