Intelligent control method and execution system for adaptive optimization of fruit ripening parameters

Abstract

The invention discloses an intelligent control method and execution system for self-adaptive optimization of fruit ripening parameters, and the method comprises the steps: collecting local environment perception data of a fruit ripening space, and collecting fruit individual physiological parameters of a single fruit in the ripening space; and executing double-layer dynamic prediction operation fusing local environment difference and fruit individual physiological difference based on the data, outputting a prediction value, and generating a correction instruction of the initial ripening parameter based on the prediction value. Through the double-layer dynamic prediction core technology point fusing the local environment and the fruit individual difference, the problem of insufficient accuracy caused by the fact that the ripening control cannot adapt to the difference in the prior art is solved, the adaptability and accuracy of the ripening control are improved, and the method is suitable for the field of intelligent control of fruit ripening.

Description

Technical Field

[0001] This invention relates to the field of fruit ripening technology, and in particular to an intelligent control method and execution system for adaptive optimization of fruit ripening parameters. Background Technology

[0002] In the fruit ripening process, the precise control of ripening parameters directly affects the quality of fruit ripening. Current fruit ripening control technologies generally implement uniform parameter control based on global average environmental data, failing to fully consider local environmental differences within the ripening space, such as uneven ethylene concentration and temperature distribution in different areas. They also neglect individual fruit physiological differences, such as inherent variations in skin permeability and initial firmness. This control method results in ripening parameters that cannot be adapted to the actual needs of the local environment and individual fruits, easily leading to some fruits being overripe and others underripe. Insufficient precision in ripening control makes it difficult to ensure uniform fruit ripening.

[0003] Based on the above problems, there is an urgent need for a technical solution that can adapt to local environmental differences and individual fruit physiological differences, and improve the accuracy of ripening control. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and propose an intelligent control method for adaptive optimization of fruit ripening parameters, comprising the following steps: S1: Collect local environmental perception data of the fruit ripening space; S2: Collect individual physiological parameters of single fruits within the ripening space; S3: Based on the local environmental perception data and the individual physiological parameters of the fruit, perform a two-layer dynamic prediction operation that integrates local environmental differences and individual physiological differences of the fruit, and output the predicted value of single fruit maturity and the predicted value of regional maturity uniformity. S4: Based on the predicted maturity value of the single fruit and the predicted maturity uniformity of the region, generate a correction instruction for the initial ripening parameters.

[0005] Preferably, the local environmental sensing data in step S1 includes local ethylene gas concentration data, local air temperature data, local air humidity data, local air velocity data, and local carbon dioxide gas concentration data. The aforementioned data are all collected using a fixed-point continuous acquisition mode, with the acquisition time interval set as a preset time interval and the acquisition spatial location set as a preset spatial location. Both the preset time interval and the preset spatial location are determined by the three-dimensional structural parameters of the ripening space.

[0006] In a further preferred embodiment, the individual physiological parameters of the fruit in step S2 include the permeability of the single fruit skin, the initial firmness of the single fruit, the initial soluble solids content of the single fruit, the respiration intensity of the single fruit, and the temperature of the single fruit skin. The aforementioned parameters are all collected using a non-contact collection method, and the collection covers all individual fruits within the ripening space.

[0007] More preferably, the two-layer dynamic prediction operation that integrates local environmental differences and individual fruit physiological differences in step S3 includes a bottom-layer single-fruit maturity prediction operation and an upper-layer regional maturity uniformity prediction operation. The input data for the bottom-layer single-fruit maturity prediction operation are the individual physiological parameters of the fruit and the local environmental perception data corresponding to the single fruit. The input data for the upper-layer regional maturity uniformity prediction operation are the results of all single-fruit maturity prediction operations and the regional division data of the ripening space. The output data of the two layers of prediction operations are the single-fruit maturity prediction value and the regional maturity uniformity prediction value, respectively.

[0008] More preferably, the single-fruit maturity prediction calculation at the bottom layer adopts a two-layer dynamic prediction formula for fruit maturity.

[0009] More preferably, the generation of the correction instruction in step S4 adopts a ripening parameter collaborative correction formula.

[0010] More preferably, the method further includes a step of quantitatively evaluating the effect of the modified ripening parameters, wherein the quantitative evaluation adopts a quantitative evaluation formula for the ripening process.

[0011] Further preferably, the method further includes a closed-loop optimization step based on the quantitative evaluation value of the ripening process. The closed-loop optimization step includes: comparing the quantitative evaluation value of the ripening process with a preset evaluation threshold, adjusting the influence coefficient in the two-layer dynamic prediction formula for fruit maturity according to the comparison result, adjusting the correction coefficient and interference compensation coefficient in the collaborative correction formula for ripening parameters, adjusting the weight coefficient in the quantitative evaluation formula for the ripening process, and using the adjusted coefficients as input parameters for the next prediction, correction, and evaluation step.

[0012] An execution system for an intelligent control method for adaptive optimization of fruit ripening parameters, applied to any one of the aforementioned intelligent control methods for adaptive optimization of fruit ripening parameters, includes a local environment perception module, a fruit individual physiological parameter acquisition module, a two-layer dynamic prediction calculation module, a correction instruction generation module, a quantitative evaluation module, and a closed-loop optimization module. The local environment perception module is used to execute the step of acquiring local environment perception data of the fruit ripening space in the intelligent control method for adaptive optimization of fruit ripening parameters. The fruit individual physiological parameter acquisition module is used to execute the step of acquiring individual physiological parameters of a single fruit within the ripening space in the intelligent control method for adaptive optimization of fruit ripening parameters. The two-layer dynamic prediction calculation module is used to execute the step of performing a two-layer dynamic prediction calculation based on local environment perception data and fruit individual physiological parameters, fusing local environmental differences and fruit individual physiological differences, and outputting a single fruit maturity prediction value and a regional maturity uniformity prediction value. The correction instruction generation module is used to execute the intelligent control method for adaptive optimization of fruit ripening parameters. The method includes a step of generating initial ripening parameter correction instructions based on single fruit maturity prediction values ​​and regional maturity uniformity prediction values. The quantitative evaluation module is used to perform a step of quantitatively evaluating the effect of the corrected ripening parameters in the intelligent control method for adaptive optimization of fruit ripening parameters. The closed-loop optimization module is used to perform a step of closed-loop optimization based on the quantitative evaluation values ​​of the ripening process in the intelligent control method for adaptive optimization of fruit ripening parameters. The output of the local environment perception module is connected to the first input of the dual-layer dynamic prediction calculation module. The output of the individual fruit physiological parameter acquisition module is connected to the second input of the dual-layer dynamic prediction calculation module. The output of the dual-layer dynamic prediction calculation module is connected to the input of the correction instruction generation module. The output of the correction instruction generation module is connected to the input of the quantitative evaluation module. The output of the quantitative evaluation module is connected to the input of the closed-loop optimization module. The output of the closed-loop optimization module is connected to the parameter adjustment terminals of the dual-layer dynamic prediction calculation module, the correction instruction generation module, and the quantitative evaluation module, respectively.

[0013] Further preferably, the local environment sensing module includes an ethylene concentration sensor group, a temperature sensor group, a humidity sensor group, a flow rate sensor group, and a carbon dioxide concentration sensor group. The ethylene concentration sensor group is used to collect local ethylene gas concentration data, the temperature sensor group is used to collect local air temperature data, the humidity sensor group is used to collect local air humidity data, the flow rate sensor group is used to collect local air flow rate data, and the carbon dioxide concentration sensor group is used to collect local carbon dioxide gas concentration data. The fruit individual physiological parameter acquisition module includes an air permeability detection unit, a hardness detection unit, and a soluble solids detection unit. The system includes a respiration intensity detection unit, a skin temperature detection unit, an air permeability detection unit for collecting air permeability parameters of a single fruit's skin, a hardness detection unit for collecting initial hardness parameters of a single fruit, a soluble solids detection unit for collecting initial soluble solids content parameters of a single fruit, a respiration intensity detection unit for collecting respiration intensity parameters of a single fruit, and a skin temperature detection unit for collecting skin temperature parameters of a single fruit. The dual-layer dynamic prediction calculation module, the correction instruction generation module, the quantitative evaluation module, and the closed-loop optimization module all employ dedicated digital signal processing chips, and the calculation logic of the aforementioned chips corresponds one-to-one with the algorithms of the corresponding steps.

[0014] The technical effects include: The core inventive technology of this invention is a two-layer dynamic prediction calculation that integrates local environmental differences and individual fruit physiological differences. This technology precisely solves the problem of insufficient accuracy in existing ripening control methods that cannot adapt to local environmental and individual differences. By using layered prediction, it achieves accurate prediction of the maturity of individual fruits and regions, providing a reliable basis for adjusting ripening parameters, effectively improving the adaptability and accuracy of ripening control, and ensuring the uniformity of fruit ripening. Attached Figure Description

[0015] Figure 1 This is a flowchart of an intelligent control method for adaptive optimization of fruit ripening parameters according to this application; Figure 2 This is a connection block diagram of an intelligent execution system for adaptive optimization of fruit ripening parameters according to this application. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0017] Existing fruit ripening control technologies have technical problems such as being unable to adapt to local environmental differences and individual fruit physiological differences, and insufficient precision in ripening control.

[0018] Based on this, please refer to Figures 1-2 This embodiment provides an intelligent control method for adaptive optimization of fruit ripening parameters, characterized by the following steps: First, local environmental perception data of the fruit ripening space is collected. This local environmental perception data reflects the environmental differences in different local areas within the fruit ripening space. The collection process must cover all local areas within the fruit ripening space to ensure that the acquired environmental perception data comprehensively reflects the environmental differences at different locations. Second, individual physiological parameters of single fruits within the ripening space are collected. These individual physiological parameters reflect the physiological differences of each individual fruit within the ripening space. The collection process must be performed separately for each individual fruit within the ripening space, and a non-contact collection method is used to avoid damaging the fruit, ensuring that the acquired individual physiological parameters accurately reflect the unique physiological state of each fruit. Next, based on the aforementioned collected local environmental perception data and individual fruit physiological parameters, a two-layer dynamic prediction operation is performed that integrates local environmental differences and individual fruit physiological differences. The calculation includes a bottom-level single-fruit maturity prediction calculation and an upper-level regional maturity uniformity prediction calculation. The bottom-level single-fruit maturity prediction calculation uses the local environmental perception data corresponding to a single fruit and the individual physiological parameters of the fruit as input data to predict the maturity status of each fruit. The upper-level regional maturity uniformity prediction calculation uses the maturity prediction results of all fruits and the regional division information of the ripening space as input data to predict the uniformity of fruit maturity in each region. Through this two-layer dynamic prediction calculation, the final output is the single-fruit maturity prediction value and the regional maturity uniformity prediction value. Finally, based on the aforementioned output single-fruit maturity prediction value and regional maturity uniformity prediction value, a correction instruction for the initial ripening parameters is generated. The correction instruction can make targeted adjustments to the initial ripening parameters according to the differences in the maturity status of different fruits and the differences in the uniformity of maturity in different regions, so that the ripening parameters can adapt to the differences in local environment and individual physiological differences of fruits, and achieve precise control of the fruit ripening process.

[0019] The core innovation of this technical solution lies in constructing a precise control logic that integrates local environmental differences with individual fruit physiological differences. From the data acquisition perspective, it breaks through the limitations of existing technologies that rely on global average data acquisition, constructing dedicated acquisition dimensions for both the environment and individual fruits to ensure that the data accurately reflects the differentiated characteristics of both the local environment and the individual fruit. From the computational perspective, it innovatively designs a two-layer dynamic prediction computational architecture to predict maturity at both the individual fruit and regional levels, taking into account both the personalized needs of individual fruits and the overall uniformity within the region. From the control perspective, it generates correction instructions based on the two-layer prediction results, enabling ripening parameters to be precisely adapted to different local environments and different individual fruits, forming a complete and precise control chain from data acquisition to computation to control instruction generation, effectively overcoming the core deficiencies of existing technologies.

[0020] Existing technologies lack targeted local environmental data collection, leading to inaccurate subsequent control measures.

[0021] Based on this, the local environmental sensing data in the first step includes local ethylene gas concentration data, local air temperature data, local air humidity data, local air velocity data, and local carbon dioxide gas concentration data. The aforementioned data are all collected using a fixed-point continuous acquisition mode, with the acquisition time interval set as a preset time interval and the acquisition spatial location set as a preset spatial location. Both the preset time interval and the preset spatial location are determined by the three-dimensional structural parameters of the ripening space.

[0022] In practice, a three-dimensional structural scan of the ripening space is first performed to obtain parameters such as the space's length, width, height, and the distribution of internal obstacles. Based on these parameters, the collection area is divided to ensure that each collection point covers a different local environmental region. The preset time interval is determined in conjunction with the fruit's physiological ripening cycle. For fruits with a short ripening cycle, a shorter collection interval is set; for fruits with a longer ripening cycle, the collection interval can be appropriately extended, but it must be ensured that dynamic changes in environmental parameters are captured. The fixed-point continuous collection mode deploys corresponding collection devices at preset spatial points to achieve uninterrupted data collection, ensuring that the collected data continuously and accurately reflects the changes in local environmental parameters, providing accurate and reliable environmental data support for subsequent two-layer dynamic prediction calculations.

[0023] Existing technologies do not take into account the individual physiological differences of fruits, and uniform parameter control cannot adapt to the needs of individual fruits.

[0024] Based on this, the physiological parameters of individual fruits in the second step include the permeability of the skin of a single fruit, the initial firmness of a single fruit, the initial soluble solids content of a single fruit, the respiration intensity of a single fruit, and the skin temperature of a single fruit. All of the aforementioned parameters are collected using a non-contact collection method, and the collection covers all individual fruits within the ripening space.

[0025] During implementation, the selection of non-contact data collection methods must be matched with corresponding data collection equipment based on the characteristics of different parameters. For example, the air permeability parameter of a single fruit peel is obtained through a non-contact gas permeability detection device. Utilizing the principle of gas molecule permeation, the gas permeation rate can be detected without contact with the fruit peel, thereby obtaining the air permeability parameter. The initial firmness parameter of a single fruit is obtained using ultrasonic detection technology. By emitting ultrasonic waves and receiving reflected waves, the internal firmness of the fruit is calculated based on the change in wave velocity, avoiding damage to the fruit peel caused by contact detection. The initial soluble solids content parameter of a single fruit is obtained through near-infrared spectroscopy non-contact detection technology. It utilizes the absorption characteristics of different soluble solids to infrared light of specific wavelengths to achieve accurate detection. The respiration intensity parameter of a single fruit is obtained by collecting changes in the gas composition around the fruit. A non-contact gas detection device is deployed to monitor the rate at which the fruit releases carbon dioxide in real time, thereby converting it into the respiration intensity parameter. The temperature parameter of a single fruit peel is obtained using an infrared thermometer to accurately capture the temperature value of the fruit peel. During the collection process, it is necessary to ensure that the detection range of the collection equipment covers all individual fruits within the ripening space, to uniquely identify each fruit, and to bind the collected individual physiological parameters with the corresponding fruit identifier, so as to provide an accurate individual data foundation for subsequent prediction calculations at the single-fruit level.

[0026] Existing technologies lack maturity prediction mechanisms that can adapt to local and individual differences, and the prediction results cannot support precise control.

[0027] Based on this, the third step integrates the two-layer dynamic prediction operation of local environmental differences and individual fruit physiological differences. This includes the bottom layer of single fruit maturity prediction operation and the top layer of regional maturity uniformity prediction operation. The input data for the bottom layer of single fruit maturity prediction operation are the individual physiological parameters of the fruit and the local environmental perception data corresponding to the single fruit. The input data for the top layer of regional maturity uniformity prediction operation are the results of all single fruit maturity prediction operations and the regional division data of the ripening space. The output data of the two layers of prediction operations are the single fruit maturity prediction value and the regional maturity uniformity prediction value, respectively.

[0028] In practical implementation, the bottom-level single-fruit maturity prediction calculation adopts a two-layer dynamic prediction formula for fruit maturity, the expression of which is: .

[0029] The core function of this formula is to accurately predict the ripeness of a specific fruit in a specific region at a specific time. Its parameter design and computational logic fully integrate local environmental differences and individual fruit physiological differences; each parameter corresponds to a key factor affecting fruit ripening. Among these, This represents the predicted maturity value of the j-th fruit in the i-th region at time t. It is a dimensionless parameter with a value between 0 and 1, where 0 represents the fruit is completely immature and 1 represents the fruit is completely mature. This parameter is the core basis for subsequent ripening parameter correction, and its accuracy directly determines the control effect. This represents the initial maturity value of the j-th fruit in the i-th region. It is also a dimensionless parameter, determined by collecting the initial physiological state of the fruit. The initial maturity of different fruits varies, and the introduction of this parameter ensures the personalized basis of the prediction. The coefficient represents the influence of ethylene concentration on the j-th fruit in the i-th region, with dimensions of cubic meters per second per milligram. This coefficient reflects the difference in sensitivity of different fruits to ethylene concentration. Different types of fruits, and even different individuals of the same type, have different sensitivities to ethylene. By obtaining this parameter through experimental measurement, it is possible to accurately match the characteristics of individual fruits. The value represents the local ethylene gas concentration of the j-th fruit in the i-th region at time t, with the dimension of milligrams per cubic meter. It is collected by the local environment sensing module. This parameter directly reflects the ripening gas conditions of the local environment in which the fruit is located, and reflects the influence of local environmental differences on ripening. This represents the epidermal permeability parameter value of the j-th fruit in the i-th region, with the dimension of negative first second. It reflects the permeability of the fruit epidermis to gases and is a core indicator of the individual physiological parameters of the fruit. Epidermal permeability determines the rate at which ethylene gas enters the fruit and directly affects the ripening process. The temperature influence coefficient of the j-th fruit in the i-th region is expressed in negative first power of M Kelvin. It reflects the degree of influence of temperature on the ripening process of different fruits. Temperature is an important driving factor in the fruit metabolism process. Different fruits respond differently to temperature. This parameter is determined through experimental calibration to achieve personalized adaptation of temperature influence. This represents the local air temperature value of the j-th fruit in the i-th region at time t, with the dimension of Kelvin. It is collected by the temperature sensor in the corresponding region and reflects the temperature conditions of the local environment where the fruit is located. This represents the surface area of ​​the epidermis of the j-th fruit in the i-th region, measured in square meters. It is calculated by measuring the fruit size using image recognition technology. The volume of the j-th fruit in the i-th region is expressed in cubic meters. It is calculated after obtaining the fruit size through image recognition technology. The ratio of surface area to volume reflects the heat exchange efficiency between the fruit and the environment, which directly affects the effect of temperature on the ripening process. The parameter represents the respiration intensity influence coefficient of the j-th fruit in the i-th region, with the dimension of negative first second. It reflects the degree to which fruit respiration promotes the ripening process. The energy released by respiration accelerates fruit ripening. Different fruits have different respiration intensity characteristics. This parameter was obtained through experimental measurement. The value of the respiration intensity parameter for the j-th fruit in the i-th region at time t is expressed in milligrams per kilogram per second. It is collected by the respiration intensity detection unit and reflects the real-time respiration status of the individual fruit. t represents the ripening time value, expressed in seconds, obtained through the system's timing module, reflecting the cumulative effect of the ripening process over time.

[0030] The theoretical design of this formula is based on the dynamic changes in fruit ripening in biophysiology, combined with the influence mechanism of local environmental factors on the ripening process. From a biophysiological perspective, fruit ripening is a dynamic process involving the synergistic effects of multiple factors such as ethylene, metabolic reactions, and respiration. Different factors have different mechanisms of influence on ripening, requiring the construction of targeted mathematical models. From a control logic perspective, it is necessary to quantify and integrate local environmental differences and individual physiological differences of fruits into the model to ensure that the prediction results can accurately reflect the ripening state under different conditions.

[0031] The logical derivation process is as follows: First, the change in fruit maturity is based on the initial maturity, therefore... As a fundamental parameter, ethylene is a key ripening gas. Its effect on ripening is concentration-dependent and time-cumulative, and is regulated by the permeability of the fruit peel. The better the permeability of the peel, the easier it is for ethylene gas to enter the fruit and exert its effect. Therefore, an exponential term is introduced. The negative exponential form reflects the driving effect of ethylene concentration and air permeability on ripeness improvement. As ethylene concentration increases and air permeability improves, the exponential term changes, thus driving the dynamic adjustment of the ripeness prediction value. Secondly, temperature is a crucial factor driving fruit metabolism. The ratio of fruit surface area to volume determines the heat exchange efficiency between the fruit and the environment; the larger the ratio, the more efficient the heat exchange, and the more significant the effect of temperature on metabolism. Therefore, a linear term is constructed... The effect of temperature on maturation is quantified by multiplying the temperature value by the surface area-to-volume ratio, and then combined with an individual-specific temperature influence coefficient. This allows for personalized adaptation to the effects of temperature. Finally, the energy released through respiration accelerates the maturation process; the greater the respiration intensity, the more energy is released, and the faster the maturation. This effect accumulates over time, therefore a linear term is included. The cumulative effect of respiration is quantified by multiplying respiratory intensity by time, combined with an individual-specific respiratory intensity influence coefficient. This method adapts to the respiratory characteristics of different fruits. Finally, the three factors are superimposed to obtain the predicted value of single fruit maturity, achieving a comprehensive consideration of local environmental differences and individual fruit physiological differences, thus ensuring the accuracy of the prediction results.

[0032] The implementation of this formula relies on a dedicated computing module. First, a data acquisition module obtains real-time data for all parameters, classifying and binding these parameters according to region and individual fruit. The computing module then calls a pre-defined formula algorithm, substituting the corresponding parameters to perform calculations and outputting a predicted maturity value for each fruit. Its core innovation lies in breaking away from the existing single-dimensional, averaged prediction model, achieving differentiated predictions based on both local environmental and individual physiological dimensions, providing a reliable predictive basis for subsequent precise control.

[0033] The existing technology for adjusting ripening parameters lacks scientific basis and cannot adapt to local and individual differences.

[0034] Based on this, the generation of the correction instruction in the fourth step adopts a formula for the coordinated correction of ripening parameters, the expression of which is: .

[0035] The core function of this formula is to generate precisely tailored ripening parameter correction instructions based on the prediction results of single fruit maturity and local environmental disturbance factors, enabling personalized and differentiated adjustments to the ripening parameters. Each parameter carries a specific function and significance. This represents the corrected ripening parameter value for the j-th fruit in the i-th region at time t. Its specific dimensions vary depending on the type of ripening parameter. If it is ethylene concentration, the dimension is mg / m³; if it is temperature, the dimension is Kelvin; and if it is humidity, the dimensionless dimension. This parameter is the final ripening control parameter and directly determines the ripening effect. This represents the initial ripening parameter value for the j-th fruit in the i-th region. Its dimensions are consistent with the corrected ripening parameters. It is set according to the fruit type and target ripening standard, providing a basic benchmark value for parameter correction. The parameter represents the maturity deviation correction coefficient for the j-th fruit in the i-th region. It is a dimensionless parameter that reflects the influence weight of maturity deviation on the ripening parameter correction. Different fruits have different tolerances and correction sensitivities to maturity deviation. This parameter is determined experimentally to ensure that the correction range can accurately match the fruit characteristics. This represents the target maturity value of the j-th fruit in the i-th region. It is dimensionless and ranges from 0 to 1. It is set according to market demand and fruit characteristics and serves as the benchmark for judging maturity deviation. This represents the deviation between the predicted maturity value of a single fruit and the target maturity value. A positive deviation indicates that the fruit maturity exceeds the target, while a negative deviation indicates that the target has not been reached. This deviation value is the core driving basis for parameter correction. The interference compensation coefficient for the j-th fruit in the i-th region is expressed in seconds per cubic meter. It is used to compensate for the influence of interference factors such as air velocity on the ripening parameter execution effect. Different fruits are affected by interference factors to varying degrees. This parameter is obtained through experimental measurement. The value of the local air velocity at time t represents the i-th region and j-th fruit. The dimension is m / s. It is collected by a flow velocity sensor. The air velocity affects the distribution and efficiency of ripening parameters around the fruit and is an important environmental disturbance factor. This represents the control feedback time lag value for the j-th fruit in the i-th region, measured in seconds. It reflects the time delay from detection to instruction execution. This delay can cause the actual effect of the ripening parameters to deviate from the expected effect, and the influence needs to be eliminated through parameter compensation.

[0036] The theoretical design of this formula is based on the principle of adaptive correction of deviation in the field of control, combined with the need to compensate for interference factors in the ripening process. The principle of adaptive correction of deviation is to dynamically adjust the control parameters by detecting the deviation between the actual value and the target value, so that the system output approaches the target value. It is suitable for dynamic processes that require precise control. In the ripening process, in addition to the maturity deviation, there are also interference factors such as air velocity and time lag. Therefore, it is necessary to add interference compensation on the basis of deviation correction to form a collaborative correction mechanism.

[0037] The logical derivation process is as follows: First, using the initial ripening parameters... Based on this, a proportional correction term based on maturity bias is constructed. Maturity bias It reflects the degree of deviation between the actual maturity state and the target state, and is corrected by the maturity deviation coefficient. The deviation is weighted to obtain the correction magnitude, and then a proportional correction term is constructed. When a positive deviation occurs, the correction term value is greater than 1, resulting in a downward adjustment of the ripening parameter and slowing down the ripening process; when a negative deviation occurs, the correction term value is less than 1, resulting in an upward adjustment of the ripening parameter, accelerating the ripening process, and ensuring that the maturity approaches the target value. Secondly, considering that local airflow velocity affects the effectiveness of the ripening parameter—for example, excessive airflow velocity will accelerate the diffusion of ripening gas and reduce local gas concentration—and that there is a time lag in control feedback, which may cause correction commands to not take effect immediately, an exponential compensation term is introduced. This index term compensates for the proportionally corrected parameters by quantifying the synergistic interference effect of airflow velocity and time lag. When the interference factor is strong, the compensation term changes, decreasing or increasing the corrected parameter value to eliminate the interference effect. Finally, the basic correction term and the compensation term are multiplied to obtain the corrected ripening parameters. This ensures that the correction instructions can accurately adapt to the needs of individual fruits and local environmental interference, thereby achieving synergistic optimization of ripening parameters.

[0038] The implementation of this formula relies on a correction instruction generation module. This module first receives the single-fruit maturity prediction value output by the dual-layer dynamic prediction calculation, calculates the maturity deviation by combining it with the preset target maturity value, and simultaneously acquires interference parameters such as local airflow velocity and control feedback time lag. It then calls a preset ripening parameter collaborative correction formula, substitutes the corresponding parameters for calculation, obtains the corrected ripening parameters, and generates a corresponding correction instruction to send to the actuator. Its core innovation lies in constructing a collaborative correction mechanism that combines maturity deviation correction with interference factor compensation, breaking through the limitations of existing technologies that rely solely on single deviation correction, and achieving precise and personalized adjustment of ripening parameters.

[0039] Existing technologies lack a quantitative evaluation mechanism for the effects of ripening parameters, making it impossible to achieve control optimization.

[0040] Based on this, the method also includes a step of quantitatively evaluating the effect of the modified ripening parameters. This quantitative evaluation uses a quantitative evaluation formula for the ripening process, the expression of which is: .

[0041] The core purpose of this formula is to comprehensively and quantitatively evaluate the ripening effect of a specific region at a specific time, providing a scientific basis for subsequent closed-loop optimization. Each parameter corresponds to a key evaluation dimension of the ripening effect. Among them, This represents the quantitative evaluation value of the ripening process in the i-th region at time t. It is a dimensionless parameter, and the larger the value, the better the ripening effect. This parameter is the core indicator for judging whether the ripening process needs to be optimized. This represents the maturity uniformity weighting coefficient of the i-th region at time t, which is dimensionless and satisfies... According to the ripening priority setting, if the uniformity of maturity is the primary goal in the ripening requirement, then the weight coefficient is set to a larger value. This represents the uniformity of maturity in the i-th region at time t. It is dimensionless and is obtained by calculating the dispersion of the predicted maturity values ​​of all individual fruits in the region. The smaller the dispersion, the larger the uniformity value, reflecting the better consistency of the fruit maturity status in the region. This represents the ethylene concentration control precision weighting coefficient for the i-th region at time t. It is dimensionless and is set according to the importance of ethylene concentration control during the ripening process. This represents the ethylene concentration control accuracy value of the i-th region at time t. It is dimensionless and is the ratio of the actual ethylene concentration to the ethylene concentration in the corrected ripening parameters. The closer this ratio is to 1, the higher the ethylene concentration control accuracy and the better the execution effect of the ripening parameters. This represents the humidity control stability weighting coefficient for the i-th region at time t. It is dimensionless and is set according to the importance of humidity control in the ripening process. This represents the humidity control stability value of the i-th region at time t. It is dimensionless and is obtained by calculating the humidity fluctuation amplitude. The smaller the fluctuation amplitude, the larger the stability value, reflecting the better the stability of humidity control.

[0042] The theoretical design of this formula is based on the principle of multi-objective quantitative evaluation, combined with the core evaluation indicators of ripening effect. The principle of multi-objective quantitative evaluation is to select multiple key evaluation indicators, assign different weights, and perform weighted summation to obtain a comprehensive evaluation value, which can comprehensively and objectively reflect the overall performance of the system. The evaluation of ripening effect needs to take into account multiple dimensions such as maturity uniformity, ethylene concentration control accuracy, and humidity control stability. A single indicator cannot fully reflect the ripening quality, so a multi-objective quantitative evaluation mechanism is adopted.

[0043] The logical derivation process is as follows: First, the core evaluation dimensions of ripening effect are clarified. Combining the technical requirements of the ripening process and the physiological characteristics of fruit ripening, maturity uniformity, ethylene concentration control precision, and humidity control stability are identified as the three core evaluation dimensions. These three dimensions correspond to the consistency of fruit ripening status, the control precision of key ripening parameters, and the stability of environmental parameters, respectively, all of which directly affect the final ripening quality. Second, corresponding quantitative indicators are selected for each evaluation dimension. Maturity uniformity corresponds to... Ethylene concentration control precision corresponds to Humidity control stability corresponds to Furthermore, quantitative values ​​for each indicator are obtained through scientific calculation methods to ensure that the indicators accurately reflect the effects of the corresponding dimensions. Then, considering the different levels of importance of each evaluation dimension under different ripening scenarios, corresponding weighting coefficients are introduced. , , The weighting coefficients were determined through expert evaluation and experimental verification, satisfying the constraint that the sum of the three factors equals 1, ensuring that the evaluation results match the actual ripening priority requirements. Finally, a weighted summation method was used to integrate the quantitative indicators of the three dimensions into a comprehensive quantitative evaluation value for the ripening process. This allows for a comprehensive and objective quantification of the ripening effect, providing a clear basis for subsequent closed-loop optimization.

[0044] The implementation of this formula relies on a quantitative evaluation module, which first obtains the predicted maturity values ​​of all individual fruits within the region and then calculates the maturity uniformity value. Simultaneously, actual ethylene concentration and humidity data within the area were collected, and the ethylene concentration control accuracy value was calculated accordingly. and humidity control stability value Then, the preset quantitative evaluation formula for the ripening process is called, and the corresponding parameters and preset weight coefficients are substituted to calculate the quantitative evaluation value of the ripening process in the region. Its core innovation lies in the construction of a multi-dimensional and adjustable quantitative evaluation mechanism, which breaks through the limitation of existing technologies lacking scientific evaluation standards. It can comprehensively and objectively reflect the ripening effect and provide reliable support for closed-loop optimization.

[0045] Existing technologies lack a closed-loop optimization mechanism, making it impossible to continuously improve ripening control.

[0046] Based on this, the method also includes a closed-loop optimization step based on the quantitative evaluation value of the ripening process. The closed-loop optimization step includes: comparing the quantitative evaluation value of the ripening process with a preset evaluation threshold, adjusting the influence coefficient in the two-layer dynamic prediction formula for fruit maturity according to the comparison result, adjusting the correction coefficient and interference compensation coefficient in the collaborative correction formula for ripening parameters, adjusting the weight coefficient in the quantitative evaluation formula for the ripening process, and using the adjusted coefficients as input parameters for the next prediction, correction, and evaluation step.

[0047] In practice, the preset evaluation thresholds need to be determined based on the fruit type and ripening quality requirements, and are divided into qualified thresholds and optimized thresholds. When the quantitative evaluation value of the ripening process is greater than or equal to the qualified threshold, it indicates that the current ripening effect meets the basic requirements; when the evaluation value is greater than or equal to the optimized threshold, it indicates that the ripening effect is excellent and no significant parameter adjustment is needed; when the evaluation value is less than the qualified threshold, it indicates that the ripening effect is poor and parameter optimization adjustment is required. After the comparison is completed, corresponding parameter adjustment strategies are formulated for different comparison results. For the influence coefficients in the two-layer dynamic prediction formula for fruit maturity, including the influence coefficients of ethylene concentration, temperature, and respiration intensity, if the evaluation value is low due to a large deviation in maturity prediction, the corresponding influence coefficients are adjusted according to the direction and magnitude of the deviation. For example, if the actual maturity is higher than the predicted value, the influence coefficient of ethylene concentration is appropriately increased to make the subsequent predicted value closer to the actual value. For the correction coefficients and interference compensation coefficients in the formula for the coordinated correction of ripening parameters, if the evaluation value is too low due to insufficient correction of ripening parameters or inadequate interference compensation, the maturity deviation correction coefficient and interference compensation coefficient should be adjusted. For example, if the maturity deviation persists, the maturity deviation correction coefficient should be increased to enhance the correction magnitude. For the weighting coefficients in the quantitative evaluation formula for the ripening process, if the evaluation process finds that the impact of a certain dimension indicator on the ripening effect exceeds expectations, the corresponding weighting coefficients should be adjusted to make the evaluation results more in line with the actual ripening needs. All adjusted coefficients must be stored and updated in the corresponding formula parameter library as input parameters for the next prediction, correction, and evaluation steps, realizing a closed-loop cycle of prediction, correction, evaluation, and optimization, ensuring that the ripening control effect can continuously adapt to the dynamic changes of the actual ripening process.

[0048] The lack of a dedicated execution system that can adapt to the above methods in the current technology makes it impossible to implement the technical solutions.

[0049] Based on this, this embodiment provides an execution system for the intelligent control method of adaptive optimization of fruit ripening parameters described above. The system includes a local environment perception module, a fruit individual physiological parameter acquisition module, a two-layer dynamic prediction calculation module, a correction instruction generation module, a quantitative evaluation module, and a closed-loop optimization module. The local environment perception module performs the first step, the fruit individual physiological parameter acquisition module performs the second step, the two-layer dynamic prediction calculation module performs the third step, the correction instruction generation module performs the fourth step, the quantitative evaluation module performs the quantitative evaluation step, and the closed-loop optimization module performs the closed-loop optimization step. The output of the local environment perception module is connected to the first input of the two-layer dynamic prediction calculation module, the output of the fruit individual physiological parameter acquisition module is connected to the second input of the two-layer dynamic prediction calculation module, the output of the two-layer dynamic prediction calculation module is connected to the input of the correction instruction generation module, the output of the correction instruction generation module is connected to the input of the quantitative evaluation module, the output of the quantitative evaluation module is connected to the input of the closed-loop optimization module, and the output of the closed-loop optimization module is connected to the parameter adjustment terminals of the two-layer dynamic prediction calculation module, the correction instruction generation module, and the quantitative evaluation module, respectively.

[0050] The core of this system lies in constructing a complete automated control chain through the precise division of labor and orderly connection of each module. The local environment perception module is responsible for collecting local environmental data within the ripening space and transmitting the data to the dual-layer dynamic prediction and calculation module, providing environmental data support for the calculation. The individual fruit physiological parameter acquisition module is responsible for collecting the physiological parameters of individual fruits and transmitting the data to the dual-layer dynamic prediction and calculation module, providing individual data support for the calculation. The dual-layer dynamic prediction and calculation module receives data transmitted from the two acquisition modules, performs dual-layer dynamic prediction calculation, and transmits the obtained single-fruit maturity prediction value and regional maturity uniformity prediction value to the correction instruction generation module. The correction instruction generation module generates correction instructions based on the received prediction values ​​and transmits the correction instructions to the quantitative evaluation module, while simultaneously sending the correction instructions to the ripening execution mechanism. The quantitative evaluation module receives the correction instructions, collects actual data during the ripening process, performs quantitative evaluation, obtains the quantitative evaluation value of the ripening process, and transmits the evaluation value to the closed-loop optimization module. The closed-loop optimization module receives the evaluation value, compares it with a preset threshold, generates parameter adjustment instructions, and sends them to the dual-layer dynamic prediction and calculation module, the correction instruction generation module, and the quantitative evaluation module, respectively, to achieve dynamic optimization of the parameters of each module. The connections between modules are made using wired or wireless communication to ensure the real-time performance and reliability of data transmission. Through the collaborative work between modules, the entire system ensures that each step of the above-mentioned intelligent control method can be executed smoothly, realizing full-process automation of data acquisition, calculation, control, evaluation, and optimization.

[0051] The existing technology has an unclear structure of execution system modules, and the functions of the core acquisition and processing modules cannot be guaranteed.

[0052] Based on this, this embodiment provides an execution system for an intelligent control method for adaptive optimization of fruit ripening parameters, comprising: a local environment sensing module including an ethylene concentration sensor group, a temperature sensor group, a humidity sensor group, a flow rate sensor group, and a carbon dioxide concentration sensor group. The ethylene concentration sensor group is used to collect local ethylene gas concentration data; the temperature sensor group is used to collect local air temperature data; the humidity sensor group is used to collect local air humidity data; the flow rate sensor group is used to collect local air flow rate data; and the carbon dioxide concentration sensor group is used to collect local carbon dioxide gas concentration data. The fruit individual physiological parameter acquisition module includes an air permeability detection unit, a firmness detection unit, a soluble solids detection unit, a respiration intensity detection unit, and a skin temperature detection unit. The air permeability detection unit is used to collect the air permeability parameters of a single fruit skin; the firmness detection unit is used to collect the initial firmness parameters of a single fruit; the soluble solids detection unit is used to collect the initial soluble solids content parameters of a single fruit; the respiration intensity detection unit is used to collect the respiration intensity parameters of a single fruit; and the skin temperature detection unit is used to collect the skin temperature parameters of a single fruit. The dual-layer dynamic prediction operation module, correction instruction generation module, quantization evaluation module, and closed-loop optimization module all use dedicated digital signal processing chips, and the operation logic of the aforementioned chips corresponds one-to-one with the algorithm of the corresponding step.

[0053] In practical implementation, each sensor group in the local environment sensing module needs to be deployed according to the three-dimensional structural parameters of the ripening space to ensure that each collection point has a corresponding sensor. The ethylene concentration sensor group uses a high-precision electrochemical ethylene sensor with a low detection limit and high response speed, which can accurately capture minute changes in local ethylene concentration; the temperature sensor group uses a platinum resistance temperature sensor with high measurement accuracy and a wide temperature range, adapting to the temperature environment of the ripening space; the humidity sensor group uses a capacitive humidity sensor, which can stably detect local air humidity; the flow rate sensor group uses a thermal anemometer to achieve accurate measurement of local air flow rate; and the carbon dioxide concentration sensor group uses an infrared carbon dioxide sensor with high stability and low drift characteristics. Each detection unit in the fruit individual physiological parameter acquisition module needs to be deployed in the fruit collection area. The air permeability detection unit uses a non-contact gas permeability detection device, the hardness detection unit uses an ultrasonic hardness detection device, the soluble solids detection unit uses a near-infrared spectroscopy detection device, the respiration intensity detection unit uses a non-contact gas detection device, and the skin temperature detection unit uses an infrared thermometer. All detection units must have non-contact detection capabilities to avoid damaging the fruit. The dual-layer dynamic prediction computation module, correction instruction generation module, quantization evaluation module, and closed-loop optimization module utilize dedicated digital signal processing chips. These chips must possess high-speed computing and multi-tasking capabilities. Algorithm programs for the corresponding steps are pre-programmed into the chips, with a one-to-one correspondence between the computational logic and the algorithm, ensuring that each module can execute its corresponding computational and control tasks quickly and accurately. Chips are connected via a data bus to achieve rapid data transmission and interaction. The entire system, by clearly defining the specific composition and function of each acquisition module and employing dedicated digital signal processing chips, ensures the computational efficiency and accuracy of the computational modules, guaranteeing stable and accurate system operation.

[0054] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A smart control method for adaptive optimization of fruit ripening parameters, characterized in that, Includes the following steps: S1: Collect local environmental perception data of the fruit ripening space; S2: Collect individual physiological parameters of single fruits within the ripening space; S3: Based on the local environmental perception data and the individual physiological parameters of the fruit, perform a two-layer dynamic prediction operation that integrates local environmental differences and individual physiological differences of the fruit, and output the predicted value of single fruit maturity and the predicted value of regional maturity uniformity. S4: Based on the predicted maturity value of the single fruit and the predicted maturity uniformity of the region, generate a correction instruction for the initial ripening parameters.

2. The intelligent control method for adaptive optimization of fruit ripening parameters according to claim 1, characterized in that, The local environmental sensing data in step S1 includes local ethylene gas concentration data, local air temperature data, local air humidity data, local air velocity data, and local carbon dioxide gas concentration data. The aforementioned data are all collected using a fixed-point continuous acquisition mode, with the acquisition time interval set as a preset time interval and the acquisition spatial location set as a preset spatial location. Both the preset time interval and the preset spatial location are determined by the three-dimensional structural parameters of the ripening space.

3. The intelligent control method for adaptive optimization of fruit ripening parameters according to claim 1, characterized in that, In step S2, the individual physiological parameters of the fruit include the permeability of the single fruit skin, the initial firmness of the single fruit, the initial soluble solids content of the single fruit, the respiration intensity of the single fruit, and the temperature of the single fruit skin. All of the aforementioned parameters are collected using a non-contact collection method, and the collection covers all individual fruits within the ripening space.

4. The intelligent control method for adaptive optimization of fruit ripening parameters according to claim 1, characterized in that, The two-layer dynamic prediction operation in step S3, which integrates local environmental differences and individual fruit physiological differences, includes a bottom-layer single-fruit maturity prediction operation and an upper-layer regional maturity uniformity prediction operation. The input data for the bottom-layer single-fruit maturity prediction operation are the individual physiological parameters of the fruit and the local environmental perception data corresponding to the single fruit. The input data for the upper-layer regional maturity uniformity prediction operation are the results of all single-fruit maturity prediction operations and the regional division data of the ripening space. The output data of the two prediction operations are the single-fruit maturity prediction value and the regional maturity uniformity prediction value, respectively.

5. The intelligent control method for adaptive optimization of fruit ripening parameters according to claim 4, characterized in that, The underlying single-fruit maturity prediction calculation adopts a two-layer dynamic prediction formula for fruit maturity.

6. The intelligent control method for adaptive optimization of fruit ripening parameters according to claim 1, characterized in that, The generation of the correction instruction in step S4 adopts a formula for coordinated correction of ripening parameters.

7. The intelligent control method for adaptive optimization of fruit ripening parameters according to claim 6, characterized in that, The method further includes a step of quantitatively evaluating the effect of the modified ripening parameters, wherein the quantitative evaluation adopts a quantitative evaluation formula for the ripening process.

8. The intelligent control method for adaptive optimization of fruit ripening parameters according to claim 7, characterized in that, The method further includes a closed-loop optimization step based on the quantitative evaluation value of the ripening process. The closed-loop optimization step includes: comparing the quantitative evaluation value of the ripening process with a preset evaluation threshold, adjusting the influence coefficient in the two-layer dynamic prediction formula for fruit maturity according to the comparison result, adjusting the correction coefficient and interference compensation coefficient in the collaborative correction formula for ripening parameters, adjusting the weight coefficient in the quantitative evaluation formula for the ripening process, and using the adjusted coefficients as input parameters for the next prediction, correction, and evaluation step.

9. An execution system for an intelligent control method for adaptive optimization of fruit ripening parameters, applied to the intelligent control method for adaptive optimization of fruit ripening parameters as described in any one of claims 1-8, characterized in that, The system includes a local environment perception module, a fruit individual physiological parameter acquisition module, a two-layer dynamic prediction calculation module, a correction instruction generation module, a quantitative evaluation module, and a closed-loop optimization module. The local environment perception module is used to collect local environment perception data of the fruit ripening space in the intelligent control method for adaptive optimization of fruit ripening parameters. The fruit individual physiological parameter acquisition module is used to collect individual physiological parameters of a single fruit within the ripening space in the intelligent control method for adaptive optimization of fruit ripening parameters. The two-layer dynamic prediction calculation module is used to perform a two-layer dynamic prediction calculation based on local environment perception data and individual fruit physiological parameters, integrating local environmental differences and individual fruit physiological differences, and outputting a single fruit maturity prediction value and a regional maturity uniformity prediction value. The correction instruction generation module is used to generate an initial value based on the single fruit maturity prediction value and the regional maturity uniformity prediction value in the intelligent control method for adaptive optimization of fruit ripening parameters. The process includes steps for correcting ripening parameters, including: a quantitative evaluation module for evaluating the effect of the corrected ripening parameters in the intelligent control method for adaptive optimization of fruit ripening parameters; a closed-loop optimization module for performing closed-loop optimization based on the quantitative evaluation value of the ripening process in the intelligent control method for adaptive optimization of fruit ripening parameters; an output of the local environment perception module connected to the first input of the dual-layer dynamic prediction calculation module; an output of the fruit individual physiological parameter acquisition module connected to the second input of the dual-layer dynamic prediction calculation module; an output of the dual-layer dynamic prediction calculation module connected to the input of the correction instruction generation module; an output of the correction instruction generation module connected to the input of the quantitative evaluation module; an output of the quantitative evaluation module connected to the input of the closed-loop optimization module; and an output of the closed-loop optimization module connected to the parameter adjustment terminals of the dual-layer dynamic prediction calculation module, the correction instruction generation module, and the quantitative evaluation module.

10. The execution system of the intelligent control method for adaptive optimization of fruit ripening parameters according to claim 9, characterized in that, The local environment sensing module includes an ethylene concentration sensor group, a temperature sensor group, a humidity sensor group, a flow rate sensor group, and a carbon dioxide concentration sensor group. The ethylene concentration sensor group is used to collect local ethylene gas concentration data; the temperature sensor group is used to collect local air temperature data; the humidity sensor group is used to collect local air humidity data; the flow rate sensor group is used to collect local air flow rate data; and the carbon dioxide concentration sensor group is used to collect local carbon dioxide gas concentration data. The fruit individual physiological parameter acquisition module includes an air permeability detection unit, a hardness detection unit, a soluble solids detection unit, and a respiration rate detection unit. The system includes a temperature detection unit, a skin temperature detection unit, an air permeability detection unit for collecting air permeability parameters of a single fruit's skin, a hardness detection unit for collecting initial hardness parameters of a single fruit, a soluble solids detection unit for collecting initial soluble solids content parameters of a single fruit, a respiration intensity detection unit for collecting respiration intensity parameters of a single fruit, and a skin temperature detection unit for collecting skin temperature parameters of a single fruit. The dual-layer dynamic prediction calculation module, the correction instruction generation module, the quantitative evaluation module, and the closed-loop optimization module all employ dedicated digital signal processing chips, and the calculation logic of the aforementioned chips corresponds one-to-one with the algorithms of the corresponding steps.

Images