A monitoring device and a monitoring method for agricultural product storage

Abstract

The application discloses a kind of monitoring device and monitoring method for agricultural product storage, comprising: acquisition module, for by the multimodal sensor network deployed in storage warehouse, real-time acquisition includes temperature, humidity, gas composition and illumination intensity environmental data stream;Analysis module is used to determine the monitoring parameter that needs to be dynamically regulated in environmental data stream based on the preset agricultural product storage threshold condition and target agricultural product physiological characteristics;Control strategy module is used to obtain the preset storage control strategy library corresponding to the monitoring parameter;Intelligent control module is used to match corresponding control strategy according to the preset storage control strategy library, and real-time dynamic adjustment is carried out on storage environment based on control strategy.It can ensure that agricultural products are in the most suitable environmental conditions during the entire storage process through real-time monitoring and intelligent control.

Description

Technical Field

[0001] This invention relates to the field of monitoring technology, and in particular to a monitoring device and method for storing agricultural products. Background Technology

[0002] With the advancement of agricultural modernization, agricultural product storage technologies have gradually developed to ensure that agricultural products maintain their quality and safety during storage. However, existing agricultural product storage technologies still have many shortcomings in terms of automation and intelligence, resulting in insufficient precision and flexibility in the control of the storage environment, and relatively limited monitoring methods, which cannot effectively address the impact of environmental changes on the quality of agricultural products.

[0003] Most current agricultural product storage systems employ simple temperature and humidity control systems to regulate the storage environment. However, this traditional method typically focuses only on temperature and humidity, neglecting the synergistic effects of various environmental parameters such as gas composition and light intensity. For example, gaseous components in the storage environment, such as ethylene and carbon dioxide concentrations, significantly impact the freshness and prevent spoilage of agricultural products, and traditional systems often fail to monitor and adjust these factors in a timely manner. This singular approach to environmental regulation frequently increases the risk of spoilage, thereby reducing the quality and safety of agricultural products. Existing agricultural product storage systems often rely on manual intervention or pre-set fixed control strategies, lacking the ability to make real-time dynamic adjustments. The storage environment requirements vary depending on the type of agricultural product and its different growth stages, and traditional storage systems do not provide precise control based on the specific needs of each product. For instance, during the storage of certain agricultural products, changes in light intensity and gas composition have a crucial impact on their quality, but traditional systems often cannot dynamically adjust the storage environment based on real-time monitoring data.

[0004] Therefore, there is an urgent need for a monitoring device and method for agricultural product storage. Summary of the Invention

[0005] This invention provides a monitoring device and method for storing agricultural products to solve the aforementioned problems in the prior art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A monitoring device for storing agricultural products, comprising:

[0008] The data acquisition module is used to acquire environmental data streams, including temperature, humidity, gas composition and light intensity, in real time through a multimodal sensor network deployed in the storage warehouse.

[0009] The analysis module is used to determine the monitoring parameters in the environmental data stream that need to be dynamically controlled based on preset agricultural product storage threshold conditions and the physiological characteristics of the target agricultural product.

[0010] The control strategy module is used to obtain the preset stored control strategy library corresponding to the monitoring parameters;

[0011] The intelligent control module is used to match the corresponding control strategy according to the preset storage control strategy library, and to make real-time dynamic adjustments to the storage environment based on the control strategy.

[0012] The analysis module includes:

[0013] The time-segmentation submodule is used to segment environmental data streams into time periods to obtain multiple data time periods.

[0014] The detection submodule is used to perform component analysis on multiple data time periods based on gas chromatography analysis technology and spectral analysis algorithm to obtain the corresponding gas components and current illumination conditions.

[0015] The determination submodule is used to combine gas composition with current lighting conditions as monitoring parameters.

[0016] The intelligent control module includes:

[0017] The initial data acquisition submodule is used to acquire the first data period when the monitoring condition is first triggered based on the environmental data stream, and at the same time acquire the data type of the first monitoring parameter;

[0018] The gas matching submodule is used to identify the first gas concentration feature in the first data period if the data type is gas composition.

[0019] The first strategy matching submodule is used to match the first gas concentration feature with the standard gas threshold in the corresponding stored control strategy library to obtain the first gas matching strategy.

[0020] The ventilation control submodule is used to initiate a graded ventilation program if the concentration exceeds the standard in the first gas matching strategy, prioritize the low-speed circulation mode, and output the first gas control command.

[0021] The continuous monitoring submodule is used to continuously acquire the second data period after the first data period if the first matching strategy is within a safe range. The first matching strategy includes a first gas matching strategy and a first illumination matching strategy.

[0022] The data sequence construction submodule is used to integrate the first and second data periods into an environmental change trend stream according to the time series.

[0023] The quality prediction submodule is used to determine the first control strategy based on the environmental change trend flow. The first gas control strategy is an environmental parameter adjustment scheme based on the prediction of agricultural product maturity.

[0024] The intelligent control module also includes:

[0025] The illumination analysis submodule is used to identify the first illumination distribution characteristics in the first data period if the data type is illumination condition.

[0026] The second strategy matching submodule is used to match the first illumination distribution features with the standard illumination spectrum in the corresponding storage control strategy library to obtain the first illumination matching strategy.

[0027] The supplementary lighting control submodule is used to automatically adjust the LED lighting system and output the first lighting control command if the first lighting matching strategy is spectral deficiency.

[0028] The quality prediction submodule includes:

[0029] The trend analysis unit is used to analyze each data period in the environmental change trend stream in chronological order.

[0030] The metabolic feature extraction unit is used to extract the gas exchange features of the current data period based on the agricultural product respiratory metabolism model and integrate them into a metabolic feature sequence during each analysis.

[0031] The correlation assessment unit is used to analyze the correlation between the parameters of the current data period and the previous data period. When the rate of change of ethylene concentration and the amount of carbon dioxide accumulation are positively correlated within a preset time window, it is determined to be a correlated storage state.

[0032] The prediction model input unit is used to input the real-time acquired metabolic feature sequence into the preset quality prediction model after each parsing to obtain the quality prediction result;

[0033] The emergency control judgment unit is used to immediately trigger a temperature and humidity control command to adjust the temperature and humidity if the correlation exceeds the preset correlation threshold and the prediction result indicates a risk of quality deterioration.

[0034] The gradual control determination unit is used to output a gradual control strategy if the correlation is lower than a preset threshold and the prediction result continues to be normal maturity.

[0035] The correlation assessment unit includes:

[0036] The gas phase recognition subunit is used to identify the ethylene release phase in the current data period and the carbon dioxide absorption phase in the previous period.

[0037] The metabolic balance calculation subunit is used to calculate the metabolic balance coefficient of the two gas phases. When the coefficient is lower than the preset balance threshold, it is determined to be a metabolic imbalance.

[0038] The light intensity impact factor acquisition subunit is used to acquire the impact factor of the current light intensity on the photosynthesis of agricultural products;

[0039] The comprehensive correlation generation subunit is used to weight and fuse the metabolic balance coefficient with the light influence factor to generate the environmental correlation index;

[0040] The storage status classification subunit is used to reclassify the storage stage and update the control strategy library when the environmental correlation index exceeds the dynamic adjustment boundary.

[0041] The metabolic feature extraction unit includes:

[0042] The metabolic parameter calibration subunit is used to obtain standard respiratory metabolism curves for different agricultural products and establish metabolic parameter calibration templates.

[0043] The real-time metabolic tracking subunit is used to identify the deviation between the actual metabolic rate and the theoretical value for the current data period based on the calibration template.

[0044] The feature vector construction sub-unit is used to encode the gradient of ethylene concentration change, the slope of carbon dioxide accumulation, and the light response delay time into a multi-dimensional feature vector.

[0045] The metabolic map update subunit is used to construct a dynamic metabolic map based on the feature vectors of three consecutive data periods, and compare the difference with the benchmark map in the regulatory strategy library. The comparison result is fed back to the correlation assessment unit in real time.

[0046] The storage control strategy library includes:

[0047] A safety threshold matrix for gas components of different agricultural products, where each matrix element corresponds to the extreme values ​​of ethylene / carbon dioxide concentration under specific temperature and humidity conditions;

[0048] A dynamic light regulation strategy tree, whose nodes contain the matching relationship between light intensity-wavelength-irradiation cycle combination parameters and photosynthetic requirements at different maturity stages;

[0049] An environmental parameter coupling rule base stores the influence function of temperature fluctuation on the gas diffusion coefficient and the synergistic effect model of humidity and light.

[0050] One of the monitoring methods for agricultural product storage includes:

[0051] S101: Real-time acquisition of environmental data streams including temperature, humidity, gas composition and light intensity through a multimodal sensor network deployed in the storage warehouse;

[0052] S102: Based on the preset agricultural product storage threshold conditions and the physiological characteristics of the target agricultural product, determine the monitoring parameters in the environmental data stream that need to be dynamically controlled;

[0053] S103: Obtain the preset storage control strategy library corresponding to the monitoring parameters;

[0054] S104: Match the corresponding control strategy according to the preset storage control strategy library, and make real-time dynamic adjustments to the storage environment based on the control strategy.

[0055] Step S102 includes:

[0056] S1021: Perform time-segmentation processing on the environmental data stream to obtain multiple data time segments;

[0057] S1022: Based on gas chromatography analysis technology and spectral analysis algorithm, the components of multiple data periods are analyzed to obtain the corresponding gas components and current illumination conditions.

[0058] S1023: Use gas composition and current illumination conditions together as monitoring parameters.

[0059] Compared with the prior art, the present invention has the following advantages:

[0060] A monitoring device for agricultural product storage includes: a data acquisition module for acquiring real-time environmental data streams containing temperature, humidity, gas composition, and light intensity through a multimodal sensor network deployed within the storage warehouse; an analysis module for determining monitoring parameters in the environmental data stream that require dynamic control based on preset agricultural product storage threshold conditions and the physiological characteristics of the target agricultural product; a control strategy module for acquiring a preset storage control strategy library corresponding to the monitoring parameters; and an intelligent control module for matching corresponding control strategies according to the preset storage control strategy library and dynamically adjusting the storage environment in real-time based on the control strategies. Through real-time monitoring and intelligent control, it is possible to ensure that agricultural products are in the most suitable environmental conditions throughout the entire storage process.

[0061] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention.

[0062] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0063] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0064] Figure 1 This is a structural diagram of a monitoring device for agricultural product storage according to an embodiment of the present invention;

[0065] Figure 2 This is a structural diagram of the analysis module in an embodiment of the present invention;

[0066] Figure 3This is a flowchart of a monitoring method for agricultural product storage according to an embodiment of the present invention. Detailed Implementation

[0067] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0068] The embodiments of the present invention provide, as follows Figure 1 As shown, a monitoring device for agricultural product storage includes:

[0069] The data acquisition module is used to acquire environmental data streams, including temperature, humidity, gas composition and light intensity, in real time through a multimodal sensor network deployed in the storage warehouse.

[0070] The analysis module is used to determine the monitoring parameters in the environmental data stream that need to be dynamically controlled based on preset agricultural product storage threshold conditions and the physiological characteristics of the target agricultural product.

[0071] The control strategy module is used to obtain the preset stored control strategy library corresponding to the monitoring parameters;

[0072] The intelligent control module is used to match the corresponding control strategy according to the preset storage control strategy library, and to make real-time dynamic adjustments to the storage environment based on the control strategy.

[0073] The working principle of the above technical solution is as follows: The device deploys a multimodal sensor network within the storage warehouse to monitor and collect various environmental data in real time, such as temperature, humidity, gas composition (e.g., oxygen, carbon dioxide), and light intensity. This information constitutes a data stream, which is continuously collected and transmitted to the system's analysis module. Temperature and humidity are key factors affecting the quality of agricultural products. For example, excessively high temperatures can lead to spoilage, while excessively low humidity can cause agricultural products to lose moisture, affecting their taste and nutritional content. Monitoring gas composition helps determine whether the storage environment is suitable for the respiration of agricultural products; excessively high carbon dioxide concentrations can accelerate the aging of some agricultural products. Light intensity also affects certain agricultural products, especially fruits and vegetables that are easily affected by light.

[0074] The collected environmental data stream is transmitted to the analysis module. This module analyzes the data based on preset storage thresholds for agricultural products and the physiological characteristics of each product. Different agricultural products have different environmental requirements; for example, apples and watermelons have different temperature requirements. The analysis module identifies which environmental parameters (such as temperature and humidity) need to be adjusted to maintain optimal storage conditions. For example, a certain agricultural product requires a temperature of 10°C and humidity of around 80%. If real-time data exceeds these ranges, the analysis module will respond promptly, informing the system that adjustments are necessary.

[0075] The analysis module selects appropriate control strategies based on monitored parameter changes. The control strategy module contains a library of preset control strategies, such as temperature control, humidity control, and gas composition adjustment. These strategies are optimized for different types of agricultural products, different storage environments, and different needs. For example, if the humidity is too low, the system will select a "humidity enhancement" strategy from the library to increase the humidity in the warehouse using a humidification device.

[0076] The intelligent control module automatically adjusts the storage environment dynamically based on strategies selected from a control strategy library. For example, when the temperature exceeds a predetermined range, the intelligent control module will activate the air conditioning system to lower the temperature; if the humidity is too low, it may activate the humidifier. This intelligent control system can respond to environmental changes in real time, ensuring that agricultural products are always stored under optimal conditions, reducing human intervention and improving storage efficiency.

[0077] The beneficial effects of the above technical solution are as follows: Through real-time monitoring and intelligent control, it ensures that agricultural products are kept in the most suitable environmental conditions throughout the entire storage process. This significantly reduces problems such as spoilage, deterioration, and nutrient loss in agricultural products, extends their shelf life, and maintains their taste and nutritional value. Because the system can automatically adjust the storage environment based on real-time data, it avoids excessive control and energy waste, providing an energy-efficient storage solution. In traditional storage, the control of the agricultural product storage environment often relies on manual inspection and adjustment, which may be untimely or inaccurate. This system reduces human intervention through automated adjustment, ensuring the accuracy and timeliness of environmental control.

[0078] In another embodiment, such as Figure 2 As shown, the analysis module includes:

[0079] The time-segmentation submodule is used to segment environmental data streams into time periods to obtain multiple data time periods.

[0080] The detection submodule is used to perform component analysis on multiple data time periods based on gas chromatography analysis technology and spectral analysis algorithm to obtain the corresponding gas components and current illumination conditions.

[0081] The determination submodule is used to combine gas composition with current lighting conditions as monitoring parameters.

[0082] The working principle of the above technical solution is as follows: To more accurately capture environmental changes, the environmental data stream is first segmented into time periods using a time-segmentation submodule. This module divides the real-time collected environmental data into multiple data periods. This time-segmentation helps the system identify environmental characteristics at different points in time, ensuring reasonable environmental analysis at different times. For example, temperature, humidity, and gas composition in the storage environment of agricultural products change over time. By segmenting the data into time periods, the dynamic changes in the environment within a specific time period can be reflected more accurately, helping the system take appropriate control measures at the right time.

[0083] The detection submodule is responsible for in-depth analysis of data within each time period. This module utilizes gas chromatography analysis technology and spectral analysis algorithms to analyze the gas composition and light conditions of each time period, obtaining specific gas components and light intensity. Gas chromatography analysis technology: This is a method for separating and quantitatively measuring gas components. By injecting gas samples into a chromatographic column, the separation characteristics of different gases within the column are utilized to analyze the proportion of each gas component. Gas components such as oxygen and carbon dioxide in the storage environment can be accurately detected using this technology, helping to determine whether they meet the respiration requirements of agricultural products. Spectral analysis algorithm: This algorithm analyzes light data to determine characteristics such as light intensity and wavelength, thereby assessing the impact of light conditions on agricultural products. For example, some agricultural products are prone to yellowing or spoilage under excessive light, while insufficient light affects their ripeness. Through this algorithm, the system can accurately understand the current light conditions.

[0084] After analyzing gas composition and light conditions, the determination submodule uses gas composition and current light conditions as monitoring parameters for joint evaluation. This module combines these two factors to more comprehensively reflect the impact of the current environment on agricultural products and provides data support for subsequent regulation. For example, some agricultural products (such as apples) require lower carbon dioxide concentrations and specific light intensities for storage. By comprehensively evaluating these two parameters, the system can more accurately determine whether the current storage environment is suitable for the storage needs of agricultural products.

[0085] The beneficial effects of the above technical solution are as follows: By dividing environmental data into time periods and analyzing it in conjunction with gas composition and lighting conditions, the system can monitor environmental changes more precisely and achieve more accurate control. This time-based data processing method helps the system detect environmental fluctuations in a timely manner, avoid delayed responses, and improve the safety of agricultural product storage. The combination of gas chromatography analysis technology and spectral analysis algorithms provides in-depth analysis of gas composition and lighting conditions in the environment, enabling more accurate capture of the impact of environmental changes on agricultural products. Through automated time-based data analysis and environmental monitoring, the system can not only adjust the storage environment in real time but also make comprehensive judgments based on multiple parameters, improving the level of intelligent management. Managers do not need to intervene excessively; the system automatically identifies and adjusts environmental conditions, saving labor costs and improving efficiency.

[0086] In another embodiment, the intelligent control module includes:

[0087] The initial data acquisition submodule is used to acquire the first data period when the monitoring condition is first triggered based on the environmental data stream, and at the same time acquire the data type of the first monitoring parameter;

[0088] The gas matching submodule is used to identify the first gas concentration feature in the first data period if the data type is gas composition.

[0089] The first strategy matching submodule is used to match the first gas concentration feature with the standard gas threshold in the corresponding stored control strategy library to obtain the first gas matching strategy.

[0090] The ventilation control submodule is used to initiate a graded ventilation program if the concentration exceeds the standard in the first gas matching strategy, prioritize the low-speed circulation mode, and output the first gas control command.

[0091] The continuous monitoring submodule is used to continuously acquire the second data period after the first data period if the first matching strategy is within a safe range. The first matching strategy includes a first gas matching strategy and a first illumination matching strategy.

[0092] The data sequence construction submodule is used to integrate the first and second data periods into an environmental change trend stream according to the time series.

[0093] The quality prediction submodule is used to determine the first control strategy based on the environmental change trend flow. The first gas control strategy is an environmental parameter adjustment scheme based on the prediction of agricultural product maturity.

[0094] The working principle of the above technical solution is as follows: the initial data acquisition submodule accesses the environmental data stream, acquires and records the data period when the monitoring condition is first triggered in real time, and extracts the data type of the first monitoring parameter. For example, in an agricultural product storage environment, when the temperature, humidity, or gas concentration reaches the set monitoring conditions, the environmental data stream will trigger the first recording. Suppose that during apple storage, the monitoring condition is that the CO2 concentration exceeds the set threshold, then the first record in the environmental data stream is the first data period, and the first monitoring parameter can be "gas concentration".

[0095] When the data type is gas composition, the gas matching submodule identifies and processes the first gas concentration feature in the first data period. For example, if the first triggered gas in the gas concentration monitoring data stream of an apple storage environment is CO2, and the CO2 concentration is 10%, the system will identify this data as the first gas concentration feature. This gas concentration feature is crucial to the storage of agricultural products.

[0096] The first strategy matching submodule matches the first gas concentration characteristic with the standard gas thresholds stored in the control strategy library to determine which gas control strategy to adopt. For example, in the case of apple storage, the standard gas threshold library specifies that when the CO2 concentration exceeds 8%, the apples ripen faster, thus requiring ventilation. If the first obtained CO2 concentration is 10%, the system will match this data with the standard gas thresholds and determine the first gas matching strategy as "concentration exceeds the standard".

[0097] When the first gas matching strategy indicates that the gas concentration exceeds the standard, the ventilation control submodule initiates a tiered ventilation program, prioritizing the use of low-speed circulation mode as needed, and outputs relevant gas control commands. If the CO2 concentration in the Apple storage warehouse reaches 10%, exceeding the standard threshold, the system will initiate a ventilation program, activating low-speed circulation mode to gradually reduce the CO2 concentration in the warehouse to maintain a suitable storage environment.

[0098] If the initial matching strategy determines that the current environment is within a safe range, the system will continue to acquire and record data for subsequent time periods. The initial matching strategy includes gas concentration matching strategies and other possible monitoring parameters, such as light intensity. For example, assuming that during apple storage, the CO2 concentration is below 10% and within a safe range, the system will continuously monitor other environmental data, such as light intensity and humidity, and adjust the control strategy accordingly.

[0099] The data sequence construction submodule integrates the first and second data periods in chronological order to form an environmental change trend stream for subsequent analysis and decision-making. For example, the system records that the CO2 concentration is 10% in the first data period, and after a period of time, the CO2 concentration drops to 8% in the second data period. Then, the system integrates the data from these two periods in chronological order to obtain the trend of CO2 concentration changes in the Apple storage environment.

[0100] Based on environmental change trends, the quality prediction submodule allows the system to adjust environmental parameters to predict the maturity of agricultural products and implement corresponding control strategies. For example, after a period of monitoring, the CO2 concentration in the apple storage environment has dropped to a safe level. Based on previous environmental data and gas concentration trends, the system predicts the ripeness of the apples and formulates an environmental parameter adjustment plan accordingly. For instance, if apples require a lower CO2 concentration to maintain freshness, the system will reduce ventilation and adjust the storage environment.

[0101] First Data Period: This refers to the data recorded during the first time period when monitoring conditions are first triggered. First Monitoring Parameter: The first monitored item recorded, such as gas concentration, temperature, and humidity. First Gas Concentration Characteristic: The gas concentration characteristic obtained during the first monitoring, for example, the CO2 concentration in the apple storage environment is 10%. Standard Gas Threshold: The system's preset gas concentration standard used to determine whether control measures are needed, for example, ventilation is required if the CO2 concentration exceeds 8%. First Gas Matching Strategy: The control strategy derived by comparing the standard gas threshold with the actual gas concentration, for example, the concentration exceeding strategy when the CO2 concentration exceeds the threshold.

[0102] The beneficial effects of the above technical solution are as follows: Through precise environmental monitoring and control, the storage period of agricultural products can be effectively extended, losses reduced, and the quality of agricultural products ensured. The system can automatically adjust environmental parameters based on real-time data without human intervention, reducing the complexity and error of manual operation. The system can monitor and respond quickly to environmental changes in real time, ensuring that the environment is within the most suitable range. Through data sequence construction and quality prediction, the system can make scientific decisions and optimize the storage environment. Whether it is apples, strawberries, or other agricultural products, the system can flexibly adjust environmental parameters according to their different storage needs to ensure optimal preservation results.

[0103] In another embodiment, the intelligent control module further includes:

[0104] The illumination analysis submodule is used to identify the first illumination distribution characteristics in the first data period if the data type is illumination condition.

[0105] The second strategy matching submodule is used to match the first illumination distribution features with the standard illumination spectrum in the corresponding storage control strategy library to obtain the first illumination matching strategy.

[0106] The supplementary lighting control submodule is used to automatically adjust the LED lighting system and output the first lighting control command if the first lighting matching strategy is spectral deficiency.

[0107] The working principle of the above technical solution is as follows: The role of the light analysis submodule is to identify the light conditions in the storage environment. Suppose we are storing tomatoes, an agricultural product with high light requirements, especially during the fruit growth stage. If the tomatoes are stored in an area with insufficient light, they will turn yellow or grow slowly. This submodule will collect light data in the first data period and analyze the current environment's light intensity, distribution, frequency, and other information through sensors. If the sensor detects that the current light intensity is 500 lux and the light distribution is relatively uniform, the system will record this light data and analyze the light distribution characteristics to determine whether it meets the growth requirements of tomatoes.

[0108] The second strategy matching submodule compares the illumination data with the standard illumination spectrum in the storage control strategy library to obtain a control strategy suitable for the current storage environment. The standard illumination spectrum contains the optimal illumination distribution for various agricultural products, helping the system understand the impact of different illumination conditions on various agricultural products. For example, the system queries the standard illumination spectrum and finds that the optimal illumination distribution for tomatoes should be in the range of 200 lux to 800 lux, with relatively uniform illumination and avoidance of strong direct sunlight. Based on the previously collected 500 lux data, the system determines that the current illumination conditions are relatively suitable and there is no serious deviation.

[0109] If, during the matching process, the system determines that the lighting conditions do not meet the optimal requirements for agricultural product growth (e.g., spectral deficiencies or excessive light), the supplemental lighting control submodule will automatically adjust the LED lighting system. At this time, the system will send control commands to the LED lighting equipment to supplement or adjust the intensity and distribution of light. If the light intensity in the tomato storage environment is insufficient, the system will automatically activate the LED lighting system to supplement the missing light frequency and intensity, bringing it to the standard range. For example, when the detected light intensity is 300 lux, the system will increase the light intensity to 600 lux via the LED lighting system to ensure that the tomatoes receive sufficient light to promote ripening.

[0110] The beneficial effects of the above technical solution are as follows: By intelligently controlling lighting, agricultural products, such as tomatoes and cucumbers, are ensured to be in an optimal lighting environment, effectively extending their shelf life and preventing premature spoilage or quality decline. Precise lighting control also avoids excessive energy waste. The system only supplements lighting when needed, avoiding unnecessary energy consumption and contributing to energy conservation and cost reduction.

[0111] In another embodiment, the quality prediction submodule includes:

[0112] The trend analysis unit is used to analyze each data period in the environmental change trend stream in chronological order.

[0113] The metabolic feature extraction unit is used to extract the gas exchange features of the current data period based on the agricultural product respiratory metabolism model and integrate them into a metabolic feature sequence during each analysis.

[0114] The correlation assessment unit is used to analyze the correlation between the parameters of the current data period and the previous data period. When the rate of change of ethylene concentration and the amount of carbon dioxide accumulation are positively correlated within a preset time window, it is determined to be a correlated storage state.

[0115] The prediction model input unit is used to input the real-time acquired metabolic feature sequence into the preset quality prediction model after each parsing to obtain the quality prediction result;

[0116] The emergency control judgment unit is used to immediately trigger a temperature and humidity control command to adjust the temperature and humidity if the correlation exceeds the preset correlation threshold and the prediction result indicates a risk of quality deterioration.

[0117] The gradual control determination unit is used to output a gradual control strategy if the correlation is lower than a preset threshold and the prediction result continues to be normal maturity.

[0118] The working principle of the above technical solution is as follows: the trend analysis unit is responsible for analyzing the changing trends of the storage environment in chronological order, including key parameters such as temperature, humidity, oxygen (O2), carbon dioxide (CO2), and ethylene (C2H4) concentration. Assuming apples are stored in a cold storage, the system collects data every hour, obtaining temperature (3℃), humidity (90%), CO2 concentration (0.5%), and ethylene concentration (0.1ppm). This data is recorded and forms an environmental change trend stream for analysis of its changes over time.

[0119] Agricultural products undergo respiration during storage, consuming oxygen, releasing carbon dioxide, and producing gases such as ethylene. The metabolic feature extraction unit, based on a respiratory metabolism model of agricultural products, extracts the gas exchange characteristics under the current storage environment and integrates them into a metabolic feature sequence. For example, the ethylene released by apples during storage promotes their ripening, while simultaneously accumulating carbon dioxide.

[0120] The metabolic feature extraction unit records the changes in CO2 and C2H4 concentrations within each time period, such as:

[0121] After 1 hour: CO2 = 0.6%, C2H2 = 0.12 ppm

[0122] After 2 hours: CO2 = 0.7%, C2H4 = 0.15 ppm

[0123] After 3 hours: CO2 = 0.9%, C2H4 = 0.2 ppm

[0124] These data will be integrated into metabolic signature sequences for further analysis of the apple's ripeness.

[0125] The correlation assessment unit analyzes the correlation between parameters in the current data period and the previous data period to determine whether the apples are in a correlated storage state. If the rate of change in ethylene concentration is positively correlated with the amount of carbon dioxide accumulation within a certain time window, it indicates that the apples are ripening rapidly and may even be entering a deterioration stage. For example, if calculations show that ethylene concentration increases by 20% per hour, carbon dioxide accumulation increases by 0.2% simultaneously, and the two show a positive correlation trend over the past 4 hours, the system determines that the apples are currently in a correlated storage state, indicating strong metabolic activity.

[0126] The predictive model input unit feeds the integrated metabolic feature sequences into the quality prediction model, which predicts future quality trends of agricultural products based on historical data and machine learning algorithms. For example, after analysis, the predictive model determines the ripeness of apples and predicts that they may enter a quality deterioration stage within 48 hours, requiring intervention. The prediction results show that under current environmental conditions, the apples are ripening faster than standard storage requirements, leading to accelerated spoilage.

[0127] The emergency control judgment unit, if the prediction results indicate a risk of quality deterioration and the change rate of ethylene and carbon dioxide exceeds the preset correlation threshold, will trigger an emergency temperature and humidity control command to quickly adjust the storage environment and slow down the metabolic rate. For example, due to the rapid respiration and metabolism of apples, the system decides to lower the temperature (from 3°C to 1°C), increase the oxygen supply (from 18% to 21%), and reduce ethylene accumulation (by activating the ethylene adsorption device). This can effectively extend the storage time of apples and prevent premature ripening or rotting.

[0128] The gradual control judgment unit, if the correlation is below a threshold and the prediction results indicate that the apples are still in a normal ripening state, will not immediately take emergency measures. Instead, it will output a gradual control strategy to maintain the optimal storage environment. For example, if the prediction model shows that the apples are still in a normal ripening stage and there is no risk of quality deterioration in the short term, the system will slowly adjust the storage environment, including: lowering the temperature by 0.5°C per day to gradually reduce the metabolic rate; controlling the humidity between 85% and 90% to prevent moisture loss; and maintaining the CO2 concentration below 0.5% to avoid excessive accumulation that could affect quality. Through this gradual control, both excessive intervention and optimal storage conditions for the apples are maintained.

[0129] The beneficial effects of the above technical solution are as follows: It achieves efficient storage management of agricultural products through trend analysis, metabolic monitoring, intelligent prediction, and hierarchical regulation. It can not only predict the quality change trends of agricultural products, but also take emergency or gradual regulatory measures according to specific circumstances to ensure stable quality, reduced losses, and maximized storage time.

[0130] In another embodiment, the correlation assessment unit includes:

[0131] The gas phase recognition subunit is used to identify the ethylene release phase in the current data period and the carbon dioxide absorption phase in the previous period.

[0132] The metabolic balance calculation subunit is used to calculate the metabolic balance coefficient of the two gas phases. When the coefficient is lower than the preset balance threshold, it is determined to be a metabolic imbalance.

[0133] The light intensity impact factor acquisition subunit is used to acquire the impact factor of the current light intensity on the photosynthesis of agricultural products;

[0134] The comprehensive correlation generation subunit is used to weight and fuse the metabolic balance coefficient with the light influence factor to generate the environmental correlation index;

[0135] The storage status classification subunit is used to reclassify the storage stage and update the control strategy library when the environmental correlation index exceeds the dynamic adjustment boundary.

[0136] The working principle of the above technical solution is as follows: the task of the gas phase recognition subunit is to identify the gas phase of the current period by detecting changes in the concentrations of ethylene (C2H2) and carbon dioxide (CO2) in the air. Ethylene is an important hormone in the ripening and senescence process of plants and has a significant impact on the post-ripening process of agricultural products. Carbon dioxide, on the other hand, is a gas absorbed by plants during photosynthesis. Assuming a vegetable storage environment, a high ethylene concentration indicates that vegetables such as tomatoes and bananas are ripening, while a high carbon dioxide concentration indicates active photosynthesis. The system compares the ethylene release phase of the current period with the carbon dioxide absorption phase of the previous period to determine whether the current environment is suitable for storing these agricultural products.

[0137] The task of the metabolic balance calculation subunit is to assess the metabolic status of agricultural products in the current environment by calculating the metabolic balance coefficient. When the metabolic balance coefficient is lower than a preset threshold, the system determines that there is a metabolic imbalance, leading to a decline in the quality of the agricultural products. For example, when storing tomatoes, the system calculates a metabolic balance coefficient based on the concentrations of ethylene and carbon dioxide. If this coefficient is lower than a preset balance threshold, it indicates that there is a problem with the balance between plant respiration (consuming oxygen and releasing carbon dioxide) and photosynthesis (absorbing carbon dioxide and releasing oxygen), resulting in a decline in the quality of the tomatoes. At this time, the system will issue an instruction to adjust the storage conditions.

[0138] The task of the light influence factor acquisition subunit is to obtain the influence factors on the photosynthesis of agricultural products based on the current light intensity. During the storage of agricultural products, light intensity directly affects the efficiency of photosynthesis, and thus affects the metabolic balance of the agricultural products. For example, when storing apples, if the light intensity is too low, the photosynthesis of apples will be inhibited, leading to metabolic imbalance. The system calculates the influence factors by detecting the current light intensity and combining it with the plant's light requirements, so as to further adjust the storage environment.

[0139] The comprehensive correlation generation subunit weights and fuses the metabolic balance coefficient and the light influence factor to generate a comprehensive environmental correlation index. This index reflects whether the current storage environment is suitable for the optimal storage of agricultural products. Assuming that the metabolic balance coefficient and light influence factor are 0.75 and 0.80 respectively when storing bananas, a comprehensive environmental correlation index is generated after weighted calculation. Assuming the weight of the metabolic balance coefficient is 0.6 and the weight of the light influence factor is 0.4, the final correlation index is 0.78. Based on this index, the system can determine whether the current environment needs adjustment.

[0140] The storage status classification subunit reclassifies storage stages and updates control strategies based on changes in the comprehensive environmental correlation index. Storage stages include different temperature and humidity controls, gas concentrations, and other parameters to ensure optimal preservation of agricultural products. For example, if the comprehensive environmental correlation index exceeds the dynamic adjustment boundary (e.g., exceeding 0.85) while storing apples, the system will determine that the current environment is no longer suitable for storing apples and will automatically adjust environmental parameters to enter a new storage stage. This might involve lowering the temperature, increasing humidity, or adjusting gas composition to prevent the apples from ripening or rotting too quickly.

[0141] The beneficial effects of the above technical solution are as follows: By monitoring gas phase, metabolic balance, and the effects of light in real time, the system can precisely adjust storage conditions, thereby extending the shelf life of agricultural products and reducing losses. The system can dynamically adjust storage strategies based on a comprehensive correlation index, avoiding excessive or insufficient intervention and improving storage efficiency. Through automated monitoring and adjustment of the storage environment, the system reduces human intervention, minimizes errors caused by human factors, and improves the accuracy of storage management. By precisely controlling temperature, humidity, and gas composition, excessive energy consumption (such as excessive cooling or humidification) and resource waste are avoided, improving energy utilization and meeting the requirements of sustainable development.

[0142] In another embodiment, the metabolic feature extraction unit includes:

[0143] The metabolic parameter calibration subunit is used to obtain standard respiratory metabolism curves for different agricultural products and establish metabolic parameter calibration templates.

[0144] The real-time metabolic tracking subunit is used to identify the deviation between the actual metabolic rate and the theoretical value for the current data period based on the calibration template.

[0145] The feature vector construction sub-unit is used to encode the gradient of ethylene concentration change, the slope of carbon dioxide accumulation, and the light response delay time into a multi-dimensional feature vector.

[0146] The metabolic map update subunit is used to construct a dynamic metabolic map based on the feature vectors of three consecutive data periods, and compare the difference with the benchmark map in the regulatory strategy library. The comparison result is fed back to the correlation assessment unit in real time.

[0147] The working principle of the above technical solution is as follows: The main task of the metabolic parameter calibration subunit is to obtain standard respiratory metabolic curves for different agricultural products and establish metabolic parameter calibration templates. Agricultural products undergo respiration during storage, consuming oxygen and releasing carbon dioxide. Different types of agricultural products have different respiratory metabolic rates; therefore, each agricultural product needs to have its standard metabolic curve obtained experimentally. Assuming we choose apples as the research object, we first monitor the respiratory metabolism of apples under different storage conditions using experimental equipment, such as measuring hourly oxygen consumption and carbon dioxide release. Then, we fit these experimental data to a mathematical model to obtain the standard respiratory metabolic curve of apples and form a metabolic parameter calibration template.

[0148] The real-time metabolic tracking subunit acquires metabolic data of agricultural products in real time and compares it with a standard template to identify the deviation between the actual metabolic rate and the theoretical value for the current data period. Taking apples as an example, during actual storage, we monitor the respiratory metabolic data of apples in real time, such as oxygen consumption and carbon dioxide release. Then, by comparing these real-time data with the standard template (the respiratory metabolic curve of apples), we can calculate the deviation between the current actual metabolic rate and the theoretical value. If a large deviation is found between the metabolic rate and the theoretical value for a certain period, it means that there is a problem with the storage environment of the apples, such as excessively high temperature or excessively low humidity.

[0149] The task of the feature vector construction sub-unit is to transform multiple metabolic features (such as the gradient of ethylene concentration change, the slope of carbon dioxide accumulation, and the light response delay time) into a multi-dimensional feature vector to describe the comprehensive characteristics of agricultural product metabolism. Taking apples as an example, during storage, apples produce ethylene, a hormone that affects fruit ripening. We can measure multiple indicators such as the gradient of ethylene concentration change, the slope of carbon dioxide accumulation, and the response delay time of apples under different light conditions in the apple storage environment. These indicators can be transformed into a multi-dimensional feature vector, for example: [ethylene concentration gradient, carbon dioxide accumulation slope, light response delay time]. This feature vector can be used to comprehensively describe the metabolism of apples.

[0150] The metabolic map update subunit constructs a dynamic metabolic map based on feature vectors from three consecutive data time periods. This map is then compared with a baseline map in the regulatory strategy library, and the results are fed back to the correlation assessment unit in real time. Assume we have data from three time periods: the first period (T1) shows the metabolic characteristics of apples immediately upon arrival at storage; the second period (T2) shows data after one week of storage; and the third period (T3) shows data after two weeks of storage. Each time period has a corresponding feature vector: [ethylene concentration change gradient, carbon dioxide accumulation slope, light response delay time]. These three feature vectors are combined to construct a dynamic metabolic map. This map can be used to display the metabolic change trend of apples. This metabolic map is then compared with a baseline map in the regulatory strategy library. For example, the baseline map shows how the metabolic characteristics of apples should change under normal storage conditions, while the actual map may show abnormal fluctuations in ethylene concentration and carbon dioxide release. Based on the comparison results, the system will report the differences and adjust storage conditions, such as adjusting temperature, humidity, or ventilation, to optimize the apple storage environment.

[0151] The beneficial effects of the above technical solution are as follows: By updating the metabolic profile and comparing it with the baseline profile, the system can diagnose potential anomalies during storage in real time, ensuring the optimal storage environment for agricultural products and improving their quality and shelf life. It enables precise tracking, evaluation, and control of the metabolic state of agricultural products, ensuring the quality stability of agricultural products during storage.

[0152] In another embodiment, the storage control strategy library includes:

[0153] A safety threshold matrix for gas components of different agricultural products, where each matrix element corresponds to the extreme values ​​of ethylene / carbon dioxide concentration under specific temperature and humidity conditions;

[0154] A dynamic light regulation strategy tree, whose nodes contain the matching relationship between light intensity-wavelength-irradiation cycle combination parameters and photosynthetic requirements at different maturity stages;

[0155] An environmental parameter coupling rule base stores the influence function of temperature fluctuation on the gas diffusion coefficient and the synergistic effect model of humidity and light.

[0156] The working principle of the above technical solution is as follows: The gas composition safety threshold matrix stores the safe concentration thresholds of gas components (such as ethylene and carbon dioxide) for different agricultural products under different temperature and humidity conditions. These thresholds are derived through experiments or research and are used to ensure that agricultural products are not affected by excessive ethylene or carbon dioxide during storage, thereby avoiding rotting or over-ripening. Suppose we are storing bananas. Bananas are fruits with high respiration rates and are very sensitive to ethylene and carbon dioxide. During storage, temperature and humidity directly affect gas concentrations. For example, the temperature is set at 12℃ and the humidity at 85%. At this time, the ethylene concentration threshold for bananas is 0.5 ppm (parts per million), and the carbon dioxide concentration threshold is 1.0%. If, under these temperature and humidity conditions, the ethylene concentration exceeds 0.5 ppm or the carbon dioxide concentration exceeds 1.0%, the system will trigger an alarm and adjust the storage environment to prevent the bananas from over-ripening. During storage, gas sensors monitor the concentrations of ethylene and carbon dioxide in real time and compare them with the values ​​in the safety threshold matrix to ensure that the gas concentrations are always kept within safe ranges.

[0157] The dynamic light control strategy tree is used to adjust parameters such as light intensity, wavelength, and irradiation period according to the different ripening stages of agricultural products to meet their photosynthetic needs. Each strategy tree node represents a combination of light parameters and is matched with the photosynthetic requirements of a specific ripening stage. For example, if we are storing tomatoes, as a fruit, the ripening process requires appropriate light to promote photosynthesis and color change. If the tomato is in an unripe stage, its light intensity requirement is low, and it is suitable for a shorter light cycle. At this time, the light intensity is 3000 lux, the wavelength is concentrated in the red light range (620-750 nm), and the irradiation period is 6 hours per day. As the tomato gradually enters the ripening stage, the photosynthetic demand increases, and the system will adjust the light parameters through the dynamic light control strategy tree. For example, the light intensity may be increased to 4000 lux, the irradiation period extended to 12 hours, and the wavelength may be shifted towards blue or violet light to promote fruit ripening. Each control strategy tree node corresponds to specific light conditions and can be dynamically adjusted according to the ripening stage of the tomato to ensure it grows or ripens in the optimal environment.

[0158] The environmental parameter coupling rule base stores coupling rules related to temperature and humidity, describing the variation of gas diffusion coefficients under different temperature conditions and the synergistic effect of humidity and light. These rules allow for precise adjustment of environmental factors such as temperature, humidity, and light to optimize the storage conditions of agricultural products. For example, citrus fruits are highly sensitive to temperature and humidity changes; excessively high or low temperatures affect respiration and water evaporation. The system uses influence functions in the coupling rule base to determine the impact of temperature fluctuations on the gas diffusion coefficient. For instance, an increase in temperature from 10°C to 15°C leads to an increase in the gas diffusion coefficient, meaning that gases like carbon dioxide diffuse faster, requiring corresponding adjustments to the concentration thresholds of ethylene and carbon dioxide. Furthermore, humidity and light also interact. For example, under high humidity conditions, changes in light intensity significantly affect water evaporation. Therefore, the system adjusts light intensity using a synergistic model to ensure a balance between humidity and light. When the system detects a temperature change, it adjusts the gas concentration thresholds, humidity, and light conditions in real time according to the influence functions in the coupling rule base, thus ensuring the optimal storage environment for citrus fruits.

[0159] The beneficial effects of the above technical solution are as follows: the storage control strategy library can dynamically adjust and optimize the storage environment in real time. The gas composition safety threshold matrix ensures that the gas concentration of agricultural products remains within a safe range, the dynamic light control strategy tree adjusts light parameters according to the growth and maturity stages of agricultural products, and the environmental parameter coupling rule library ensures that agricultural products are stored in the optimal environment through precise coordinated control of temperature, humidity, and light.

[0160] In another embodiment, such as Figure 3 As shown, a monitoring method for agricultural product storage includes:

[0161] S101: Real-time acquisition of environmental data streams including temperature, humidity, gas composition and light intensity through a multimodal sensor network deployed in the storage warehouse;

[0162] S102: Based on the preset agricultural product storage threshold conditions and the physiological characteristics of the target agricultural product, determine the monitoring parameters in the environmental data stream that need to be dynamically controlled;

[0163] S103: Obtain the preset storage control strategy library corresponding to the monitoring parameters;

[0164] S104: Match the corresponding control strategy according to the preset storage control strategy library, and make real-time dynamic adjustments to the storage environment based on the control strategy.

[0165] The working principle of the above technical solution is as follows: Step S104 includes:

[0166] Based on the environmental data stream, the first data period when the monitoring condition is first triggered is obtained, and the data type of the first monitoring parameter is also obtained.

[0167] If the data type is gas composition, identify the first gas concentration feature in the first data period;

[0168] The first gas concentration feature is matched with the standard gas threshold in the corresponding stored control strategy library to obtain the first gas matching strategy;

[0169] If the first gas matching strategy is that the concentration exceeds the standard, the graded ventilation procedure is activated, the low-speed circulation mode is given priority, and the first gas control command is output.

[0170] If the first matching strategy is within a safe range, the second data period after the first data period is continuously acquired. The first matching strategy includes a first gas matching strategy and a first illumination matching strategy.

[0171] The first and second data periods are integrated into an environmental change trend stream by time series.

[0172] Based on the trend of environmental change, the first regulation strategy is determined. The first gas regulation strategy is an environmental parameter adjustment scheme based on the prediction of agricultural product maturity.

[0173] Step S104 also includes:

[0174] If the data type is illumination conditions, identify the first illumination distribution characteristics in the first data period;

[0175] The first illumination distribution feature is matched with the standard illumination spectrum in the corresponding storage and control strategy library to obtain the first illumination matching strategy;

[0176] If the first illumination matching strategy is spectral missing, the LED illumination system is automatically adjusted and the first illumination control command is output.

[0177] The primary regulatory strategy includes:

[0178] Analyze each data period in the environmental change trend stream in chronological order;

[0179] During each analysis, gas exchange characteristics for the current data period are extracted based on the agricultural product respiratory metabolism model and integrated into a metabolic feature sequence.

[0180] Analyze the correlation between the parameters of the current data period and the previous data period. When the rate of change of ethylene concentration and the amount of carbon dioxide accumulation are positively correlated within the preset time window, it is determined to be a correlated storage state.

[0181] After each parsing is completed, the real-time acquired metabolic feature sequence is input into the preset quality prediction model to obtain the quality prediction result;

[0182] If the correlation exceeds the preset correlation threshold and the prediction result indicates a risk of quality deterioration, a temperature and humidity control command will be immediately triggered to adjust the temperature and humidity.

[0183] If the correlation is lower than the preset threshold and the prediction results continue to be normal and mature, a gradual control strategy will be output.

[0184] Analyze the correlation between the parameters of the currently parsed data period and the previous data period, including:

[0185] Identify the ethylene release phase in the current data period and the carbon dioxide absorption phase in the previous period;

[0186] Calculate the metabolic balance coefficient for the two gas phases, and determine metabolic imbalance when the coefficient is lower than a preset balance threshold;

[0187] To obtain the factors affecting the current light intensity on the photosynthesis of agricultural products;

[0188] The metabolic balance coefficient and the light influence factor are weighted and fused to generate an environmental correlation index;

[0189] When the environmental correlation index exceeds the dynamic adjustment boundary, the storage phase is redefined and the control strategy library is updated.

[0190] Extract gas exchange characteristics for the current data period, including:

[0191] Obtain standard respiratory metabolism curves for different agricultural products and establish templates for metabolic parameter calibration.

[0192] The deviation between the actual metabolic rate and the theoretical value for the current data period is identified based on the calibration template.

[0193] The gradient of ethylene concentration change, the slope of carbon dioxide accumulation, and the light response delay time are encoded into multidimensional feature vectors.

[0194] A dynamic metabolic map is constructed based on the feature vectors of three consecutive data periods, and the difference is compared with the benchmark map in the regulatory strategy library. The comparison results are fed back to the correlation assessment unit in real time.

[0195] The beneficial effects of the above technical solution are as follows: it can ensure precise control of environmental conditions during the storage of agricultural products, thereby optimizing the preservation quality of agricultural products and avoiding quality loss caused by environmental fluctuations. The data provided by multimodal sensors provides a precise basis for intelligent regulation, while the regulation strategy library ensures the scientific nature and practicality of the regulation measures, ultimately improving storage efficiency, extending shelf life, and reducing resource waste.

[0196] In another embodiment, step S102 includes:

[0197] S1021: Perform time-segmentation processing on the environmental data stream to obtain multiple data time segments;

[0198] S1022: Based on gas chromatography analysis technology and spectral analysis algorithm, the components of multiple data periods are analyzed to obtain the corresponding gas components and current illumination conditions.

[0199] S1023: Use gas composition and current illumination conditions together as monitoring parameters.

[0200] The working principle of the above technical solution is as follows: First, the environmental data stream collected by the sensors is segmented into multiple time periods. Each time period contains environmental data for a specific time, such as temperature, humidity, gas composition, and light intensity. Assuming that the temperature, humidity, and gas concentration (e.g., ethylene concentration) data streams in a storage warehouse are continuously collected throughout 24 hours, the system will segment this data according to certain time intervals (e.g., every hour, every half hour, or every 10 minutes), generating multiple data periods (e.g., 00:00-01:00, 01:00-02:00, etc.). During apple storage, certain gases, such as ethylene, affect the ripening process of apples at specific times. If the gas concentration is too high at night (e.g., 21:00-22:00), it will accelerate apple ripening, while during the day (e.g., 09:00-10:00), the higher temperature will cause different changes. Therefore, segmenting the data into time periods helps to accurately analyze the impact of environmental changes at different times on agricultural products.

[0201] Gas chromatography (GC) is used to accurately determine the gas composition for each time period, while spectral analysis algorithms are combined to analyze light intensity and its wavelength distribution. Gas chromatography is a commonly used technique for analyzing gas components. It separates different gases using a chromatographic column and then detects the concentration of each gas using sensors. For example, gas chromatography can be used to detect the concentrations of ethylene, carbon dioxide, oxygen, and other volatile substances in a warehouse. Spectral analysis algorithms are used to analyze light intensity and its wavelength distribution to determine the impact of light conditions in the storage environment on agricultural products. For example, some vegetables, such as leafy greens (e.g., lettuce, spinach), require less light to delay ripening, while fruits (e.g., strawberries) may require more light to extend their shelf life. For example, the system uses gas and light data for each time period to obtain the concentrations of various gas components through chromatographic analysis and to obtain the light intensity for different time periods through spectral algorithms. For example, a higher carbon dioxide concentration may be detected between 00:00 and 01:00, while stronger light is expected between 09:00 and 10:00. For grape storage, grapes are highly sensitive to ethylene, especially at night when light intensity is low; increased ethylene concentration can cause grapes to ripen prematurely. Therefore, through this technical analysis, the system can identify and adjust ethylene concentration and light conditions to prevent grapes from spoiling during storage.

[0202] The gas concentrations obtained after component analysis are comprehensively considered in conjunction with light conditions, forming a unified monitoring parameter for subsequent dynamic regulation. The concentration data of each gas (such as ethylene and carbon dioxide) obtained from gas analysis are combined with the light intensity at the current time. For example, if the ethylene concentration is too high and the light conditions are suitable at a certain time, fruits such as apples and bananas will ripen faster. Therefore, these environmental parameters are integrated as a basis for future regulation. For the storage environment of bananas, if the ethylene concentration is too high and the light intensity is strong, the ripening speed of bananas will be accelerated. The system will use this data to prompt managers to reduce the ethylene concentration and adjust the light intensity to delay the ripening of bananas and extend their shelf life.

[0203] The beneficial effects of the above technical solution are as follows: by combining time-based processing and analysis of gas composition and light conditions, it helps to provide a more precise storage environment control solution, extend the shelf life of agricultural products, improve storage efficiency, and at the same time reduce resource waste and reduce the negative environmental impact on agricultural products.

[0204] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A monitoring device for agricultural product storage, characterized in that, include: The data acquisition module is used to acquire environmental data streams, including temperature, humidity, gas composition and light intensity, in real time through a multimodal sensor network deployed in the storage warehouse. The analysis module is used to determine the monitoring parameters in the environmental data stream that need to be dynamically controlled based on preset agricultural product storage threshold conditions and the physiological characteristics of the target agricultural product. The control strategy module is used to obtain the preset stored control strategy library corresponding to the monitoring parameters; The intelligent control module is used to match the corresponding control strategy according to the preset storage control strategy library, and to make real-time dynamic adjustments to the storage environment based on the control strategy. The intelligent control module includes: The initial data acquisition submodule is used to acquire the first data period when the monitoring condition is first triggered based on the environmental data stream, and at the same time acquire the data type of the first monitoring parameter; The gas matching submodule is used to identify the first gas concentration feature in the first data period if the data type is gas composition. The first strategy matching submodule is used to match the first gas concentration feature with the standard gas threshold in the corresponding stored control strategy library to obtain the first gas matching strategy. The ventilation control submodule is used to initiate a graded ventilation program if the concentration exceeds the standard in the first gas matching strategy, prioritize the low-speed circulation mode, and output the first gas control command. The continuous monitoring submodule is used to continuously acquire the second data period after the first data period if the first matching strategy is within a safe range. The first matching strategy includes a first gas matching strategy and a first illumination matching strategy. The data sequence construction submodule is used to integrate the first and second data periods into an environmental change trend stream according to the time series. The quality prediction submodule is used to determine the first control strategy based on the environmental change trend flow. The first gas control strategy is an environmental parameter adjustment scheme based on the prediction of agricultural product maturity. The quality prediction submodule includes: The trend analysis unit is used to analyze each data period in the environmental change trend stream in chronological order. The metabolic feature extraction unit is used to extract the gas exchange features of the current data period based on the agricultural product respiratory metabolism model and integrate them into a metabolic feature sequence during each analysis. The correlation assessment unit is used to analyze the correlation between the parameters of the current data period and the previous data period, and to determine whether it is in a correlated storage state. When the rate of change of ethylene concentration and the amount of carbon dioxide accumulation are positively correlated within a preset time window, it is determined to be in a correlated storage state. The prediction model input unit is used to input the real-time acquired metabolic feature sequence into the preset quality prediction model after each parsing to obtain the quality prediction result; The emergency control judgment unit is used to immediately trigger a temperature and humidity control command to adjust the temperature and humidity if the correlation exceeds the preset correlation threshold and the prediction result indicates a risk of quality deterioration. The gradual control determination unit is used to output a gradual control strategy if the correlation is lower than a preset threshold and the prediction result continues to be normal maturity.

2. The monitoring device for agricultural product storage according to claim 1, characterized in that, The analysis module includes: The time-segmentation submodule is used to segment environmental data streams into time periods to obtain multiple data time periods. The detection submodule is used to perform component analysis on multiple data time periods based on gas chromatography analysis technology and spectral analysis algorithm to obtain the corresponding gas components and current illumination conditions. The determination submodule is used to combine gas composition with current lighting conditions as monitoring parameters.

3. The monitoring device for agricultural product storage according to claim 1, characterized in that, The intelligent control module also includes: The illumination analysis submodule is used to identify the first illumination distribution characteristics in the first data period if the data type is illumination condition. The second strategy matching submodule is used to match the first illumination distribution features with the standard illumination spectrum in the corresponding storage control strategy library to obtain the first illumination matching strategy. The supplementary lighting control submodule is used to automatically adjust the LED lighting system and output the first lighting control command if the first lighting matching strategy is spectral deficiency.

4. The monitoring device for agricultural product storage according to claim 1, characterized in that, The relevance assessment unit includes: The gas phase recognition subunit is used to identify the ethylene release phase in the current data period and the carbon dioxide absorption phase in the previous period. The metabolic balance calculation subunit is used to calculate the metabolic balance coefficient of the two gas phases. When the coefficient is lower than the preset balance threshold, it is determined to be a metabolic imbalance. The light intensity impact factor acquisition subunit is used to acquire the impact factor of the current light intensity on the photosynthesis of agricultural products; The comprehensive correlation generation subunit is used to weight and fuse the metabolic balance coefficient with the light influence factor to generate the environmental correlation index; The storage status classification subunit is used to reclassify the storage stage and update the control strategy library when the environmental correlation index exceeds the dynamic adjustment boundary.

5. The monitoring device for agricultural product storage according to claim 1, characterized in that, The metabolic feature extraction unit includes: The metabolic parameter calibration subunit is used to obtain standard respiratory metabolism curves for different agricultural products and establish metabolic parameter calibration templates. The real-time metabolic tracking subunit is used to identify the deviation between the actual metabolic rate and the theoretical value for the current data period based on the calibration template. The feature vector construction sub-unit is used to encode the gradient of ethylene concentration change, the slope of carbon dioxide accumulation, and the light response delay time into a multi-dimensional feature vector. The metabolic map update subunit is used to construct a dynamic metabolic map based on the feature vectors of three consecutive data periods, and compare the difference with the benchmark map in the regulatory strategy library. The comparison result is fed back to the correlation assessment unit in real time.

6. The monitoring device for agricultural product storage according to claim 1, characterized in that, The storage control strategy library includes: A safety threshold matrix for gas components of different agricultural products, where each matrix element corresponds to the extreme values ​​of ethylene / carbon dioxide concentration under specific temperature and humidity conditions; A dynamic light regulation strategy tree, whose nodes contain the matching relationship between light intensity-wavelength-irradiation cycle combination parameters and photosynthetic requirements at different maturity stages; An environmental parameter coupling rule base stores the influence function of temperature fluctuation on the gas diffusion coefficient and the synergistic effect model of humidity and light.

7. A monitoring method for agricultural product storage, characterized in that, include: S101: Real-time acquisition of environmental data streams including temperature, humidity, gas composition and light intensity through a multimodal sensor network deployed in the storage warehouse; S102: Based on the preset agricultural product storage threshold conditions and the physiological characteristics of the target agricultural product, determine the monitoring parameters in the environmental data stream that need to be dynamically controlled; S103: Obtain the preset storage control strategy library corresponding to the monitoring parameters; S104: Match the corresponding control strategy according to the preset storage control strategy library, and make real-time dynamic adjustments to the storage environment based on the control strategy; Based on the environmental data stream, the first data period when the monitoring condition is first triggered is obtained, and the data type of the first monitoring parameter is also obtained. If the data type is gas composition, identify the first gas concentration feature in the first data period; The first gas concentration feature is matched with the standard gas threshold in the corresponding stored control strategy library to obtain the first gas matching strategy; If the first gas matching strategy is that the concentration exceeds the standard, the graded ventilation procedure is activated, the low-speed circulation mode is given priority, and the first gas control command is output. If the first matching strategy is within a safe range, the second data period after the first data period is continuously acquired. The first matching strategy includes a first gas matching strategy and a first illumination matching strategy. The first and second data periods are integrated into an environmental change trend stream by time series. Based on the trend of environmental change, the first regulation strategy is determined. The first gas regulation strategy is an environmental parameter adjustment scheme based on the prediction of agricultural product maturity. Analyze each data period in the environmental change trend stream in chronological order; During each analysis, gas exchange characteristics for the current data period are extracted based on the agricultural product respiratory metabolism model and integrated into a metabolic feature sequence. Analyze the correlation between the parameters of the current data period and the previous data period to determine whether it is in a correlated storage state. When the rate of change of ethylene concentration and the amount of carbon dioxide accumulation are positively correlated within a preset time window, it is determined to be in a correlated storage state. After each parsing is completed, the real-time acquired metabolic feature sequence is input into the preset quality prediction model to obtain the quality prediction result; If the correlation exceeds the preset correlation threshold and the prediction result indicates a risk of quality deterioration, a temperature and humidity control command will be immediately triggered to adjust the temperature and humidity. If the correlation is lower than the preset threshold and the prediction results continue to be normal and mature, a gradual control strategy will be output.

8. The monitoring method for agricultural product storage according to claim 7, characterized in that, Step S102 includes: S1021: Perform time-segmentation processing on the environmental data stream to obtain multiple data time segments; S1022: Based on gas chromatography analysis technology and spectral analysis algorithm, the components of multiple data periods are analyzed to obtain the corresponding gas components and current illumination conditions. S1023: Use gas composition and current illumination conditions together as monitoring parameters.

Images