Tea storage bin management system and method

Abstract

The invention discloses a tea storage bin management system and method, and particularly relates to the field of tea intelligent storage. Comprising a storage environment modeling module, an information input classification module, a structure identification and authentication module, an environment gradient sensing module, a breathing parameter monitoring module, a tea quality sensing module, a storage environment regulation and control module, a storage opening and closing management module, a quality decay modeling module, a storage dynamic early warning module and a storage strategy generation module. According to the invention, environment blind areas can be effectively avoided, the accuracy and responsiveness of regulation and control are improved, quality deterioration signals can be extracted in real time, quality abnormity early warning can be given out in advance, manual detection errors and delay are greatly reduced, the storage safety is improved, dynamic gas interaction evaluation between tea leaves and the environment in the warehouse is realized, and the quality of the tea leaves is improved. The scientificity and effectiveness of regulation and control are improved, excessive storage or early failure is effectively avoided, differentiated and intelligent management is realized, and the automation and intelligent level is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent tea storage, and particularly to a tea storage bin management system and method. Background Art

[0002] As an important traditional special agricultural product, tea not only carries rich cultural value but also has significant economic benefits. Its quality directly affects consumers' sensory experience and market price. However, during the entire process from tea picking, primary processing, refining to circulation and storage, especially during the storage stage, it is extremely vulnerable to environmental factors such as temperature and humidity fluctuations, oxygen content changes, odor cross-contamination, and microbial activities. These factors will accelerate the oxidation of polyphenols in tea, the volatilization of aromatic substances, and the color variation, resulting in a significant decline in its sensory indicators such as color, aroma, and taste. In severe cases, it may even cause irreversible quality deterioration. Traditional tea storage methods mostly adopt constant temperature and humidity systems, using static control models, ignoring the physiological metabolic reactions and dynamic change characteristics of the tea itself. This method cannot respond in a timely manner according to the actual quality state of the tea and the trend of environmental evolution, lacking the ability of multi-point perception and personalized regulation of the microenvironment. In recent years, with the rapid development of technologies such as artificial intelligence, the Internet of Things, big data, and environmental control, it has provided the possibility for the intelligent upgrade of the tea storage process. In this context, there is an urgent need to construct a tea storage bin management system and method.

[0003] Existing tea storage bin management systems and methods have environmental blind spots, reducing the accuracy and responsiveness of regulation, unable to extract quality deterioration signals in real time, having high manual detection errors and delays, prone to over-storage or premature failure, and having a low level of automation and intelligence in storage. Therefore, we propose a tea storage bin management system and method. Summary of the Invention

[0004] The purpose of the present invention is to solve the defects existing in the prior art;

[0005] Therefore, a tea storage bin management system and method are proposed.

[0006] In the first aspect of the implementation of the present invention, a tea storage bin management system is first proposed. The system includes: a warehousing environment modeling module, an information entry and classification module, a structure recognition and authentication module, an environmental gradient perception module, a respiration parameter monitoring module, a tea quality perception module, a storage environment regulation module, a warehousing opening and closing management module, a quality decay modeling module, a warehousing dynamic warning module, a storage strategy generation module, an anti-counterfeiting traceability and security module, and a warehousing maintenance scheduling module;

[0007] The warehousing environment modeling module establishes a three-dimensional dynamic temperature and pressure field simulation model by collecting various climate parameters of the target tea production area;

[0008] The information input and classification module is used to record various information of each batch of tea leaves and manage the warehousing in different areas according to their characteristics;

[0009] The structure recognition and authentication module is used to uniquely identify the structural characteristics of each storage bin, bind it with the Internet of Things chip, and conduct digital identity authentication;

[0010] The environmental gradient perception module collects the micro-environment parameters in real time by arranging a multi-point high-precision sensor array in the bin and constructs an environmental gradient map in the bin;

[0011] The respiration parameter monitoring module monitors the changes in the respiration intensity and adsorption characteristics of tea leaves under different environmental conditions in real time to obtain the respiration entropy change of each tea leaf;

[0012] The tea quality perception module collects the surface characteristic spectra of tea leaves in real time through a micro fiber optic spectrometer array and dynamically monitors the tea quality;

[0013] The storage environment regulation module automatically adjusts the environmental parameters of the storage bin according to the environmental gradient map in the bin and the respiration entropy change of the tea leaves;

[0014] The warehousing opening and closing management module is used to automatically adjust the pressure difference between the inside and outside of the storage bin when opening and closing the warehouse door;

[0015] The quality decay modeling module is used to predict the best storage period and potential risks of tea leaves under the current conditions;

[0016] The warehousing dynamic warning module constructs a warehousing process map based on the environmental parameters, tea quality and prediction results of the storage bin, and issues a warning for abnormal trends to prompt intervention;

[0017] The storage strategy generation module is used to dynamically adjust the storage conditions and storage period to form a self-optimizing warehousing management strategy;

[0018] The anti-counterfeiting traceability and security module stores the warehousing information, storage process, environmental data and quality information of each batch of tea leaves on the blockchain;

[0019] The warehousing maintenance scheduling module analyzes the probability of equipment failure, quality decline or environmental fluctuation according to the prediction results and the warehousing process map, and plans maintenance tasks and warehouse adjustment operations in advance.

[0020] As a further solution of the present invention, the specific steps for the warehousing environment modeling module to establish a three-dimensional dynamic temperature and pressure field simulation model are as follows:

[0021] S1.1: Collect the historical climate data of the origin area through meteorological station data, remote sensing satellite records, ground observations, and historical databases. Then, convert all the collected historical climate data into international standard units, and unify the time scales of the historical climate data from different data sources through linear interpolation method.

[0022] S1.2: Identify and remove the abnormal data points in the climate data by using the standard deviation judgment method, and then fill the abnormal data points in the climate data through the forward and backward mean filling method. Standardize each group of processed climate data, and then convert all the standardized meteorological data into a structured format.

[0023] S1.3: Based on each group of processed climate data, generate the basic static pressure data required for different "equivalent altitude layers" in the bin through the international standard atmospheric pressure model. Then, based on the heat conduction equation, simulate the evolution of the temperature distribution in three-dimensional space over time, and construct a distribution map of the temperature in the simulated space of the bin. After that, establish a humidity-temperature coupling field to calculate the target humidity value in the bin, and at the same time simulate the evolution of the temperature-pressure flow field of the air over time to obtain the change trend of the temperature-pressure field within a specified time in real time.

[0024] As a further solution of the present invention, the historical climate data of the origin area described in S1.1 includes annual average temperature, diurnal temperature difference, seasonal temperature and humidity changes, wind speed, sunshine intensity, precipitation, altitude, etc.

[0025] As a further solution of the present invention, the specific steps for the environmental gradient perception module to construct the environmental gradient map in the bin are as follows:

[0026] S2.1: Conduct a three-dimensional grid division of the internal space according to the geometric structure of the storage bin. According to the point selection principle of equidistant and uniform distribution and enhanced layout in key areas, various sensors are arranged in the storage bin. At the same time, mark the three-dimensional coordinates of each sensor, and based on the time sequence synchronization mechanism, unify the time stamps and recording frequencies to record the micro-environment parameters in the storage bin collected by each sensor in real time.

[0027] S2.2: Determine the space points not covered by the sensors according to the arrangement positions of the sensors, and use the inverse distance weighted interpolation method to reconstruct the continuous environmental variable distribution at the space points not covered, so as to reconstruct the environmental continuous field of the space points not covered in the storage bin. According to the micro-environment parameters collected in the storage bin, calculate the gradients of each micro-environment parameter in space.

[0028] S2.3: Integrate the reconstructed continuous environmental field and the gradient calculation results to generate a three-dimensional distribution map of each micro-environment parameter in the storage bin, that is, the in-bin environmental gradient map, and visually process the in-bin environmental gradient map in the form of color gradients. Then calculate the temporal standard deviation of each area of the storage bin. If the temporal standard deviation of the area exceeds the preset safety threshold, trigger the local warning mechanism. Otherwise, continuously detect the micro-environment parameters of each area of the storage bin.

[0029] As a further solution of the present invention, the micro-environment parameters described in S2.1 specifically include temperature, humidity, air pressure, oxygen concentration, CO2 concentration, light intensity, etc.

[0030] As a further solution of the present invention, the specific steps for the tea quality perception module to dynamically monitor the tea quality are as follows:

[0031] S3.1: Install miniature fiber optic spectrometers in the storage bin, and evenly irradiate the incoming tea with a standard white light source according to the preset irradiation angle and intensity. Each group of miniature fiber optic spectrometers records the spectral reflectance curves of the tea in each area, and at the same time, smooth and standardize the collected spectral reflectance curves;

[0032] S3.2: Extract the characteristic bands of each tea quality influencing component from the processed spectral reflectance curves by principal component analysis. Take the extracted characteristic bands as independent variables, and take the tea quality detection data at the corresponding time points as dependent variables, and perform mean normalization processing on each independent variable and dependent variable;

[0033] S3.3: According to the normalized independent variables and dependent variables, establish a PLSR prediction model. Then, output the standardized quality prediction values of each tea through the PLSR prediction model, and restore the standardized prediction values to the actual predicted quality values of each detected tea, and record the predicted quality values of each tea;

[0034] S3.4: Automatically update the spectral prediction values every preset period, compare them with the historical spectral prediction values, obtain the quality change rate, and establish a corresponding quality trend curve. If the quality change rate is higher than the set threshold, it is judged that the tea has undergone significant deterioration, and an early warning of abnormal deterioration is given. At the same time, display each index of the tea quality in real time in the form of a chart or a heat map;

[0035] S3.5: Obtain the actual quality values of each tea in real time, and then jointly measure the performance of the PLSR prediction model based on the root mean square error and the coefficient of determination. If the performance of the PLSR prediction model is lower than the preset performance threshold, adjust the parameters of the PLSR prediction model through the Adam optimizer, and re-train and optimize the PLSR prediction model using the historical tea quality values until the performance of the PLSR prediction model meets the preset threshold.

[0036] In the second aspect of the implementation of the present invention, a method for managing a tea storage bin is proposed, and the method includes the following steps:

[0037] Ⅰ. Before the tea enters the bin, collect the climate parameters of the origin of the target tea, simulate the ecological environment of the target tea, and set it as the initial control target value of the storage bin;

[0038] Ⅱ. Collect the detailed information of each batch of tea, and allocate each batch of tea to the corresponding storage area or independent compartment;

[0039] Ⅲ. Bind each storage bin with the tea batch through the Internet of Things platform. After the tea officially enters the bin, collect the microenvironment data, tea physiological metabolism indexes and tea reflection spectrum in the bin in real time;

[0040] Ⅳ. Based on the data collected in real time, identify the local change trend in the storage bin, automatically adjust the environmental parameters in the bin, and at the same time predict the tea deterioration trend and automatically trigger an alarm;

[0041] Ⅴ. Dynamically formulate a tea storage strategy, and when the tea enters or exits the bin, control the gas flow through the buffer chamber, first adjust the internal and external pressure difference, and then perform the operation of entering or exiting the bin.

[0042] As a further solution of the present invention, the specific steps of automatically adjusting the environmental parameters in the bin in step Ⅳ are as follows:

[0043] S4.1: Collect the oxygen consumption and carbon dioxide release during the tea metabolism process in real time, establish a tea unit mass respiration rate model based on the temperature and oxygen partial pressure in the corresponding area of the storage bin, and then use the modified Langmuir model to establish an adsorption response model of tea to moisture and gas;

[0044] S4.2: According to the established respiration rate model and adsorption response model, obtain the respiration rate and adsorption flux of each area of the storage bin in real time, and calculate the influence of the respiration rate and adsorption flux on the local microenvironment to obtain the net gas interaction rate between the corresponding tea and the environment;

[0045] S4.3: According to the net gas interaction rate of each group of areas divided in the storage bin, adjust the environmental parameters of the corresponding area. If the net gas interaction rate > 0, it means that the storage bin is in the state of oxygen consumption and moisture discharge, and ventilation needs to be strengthened. If the net gas interaction rate < 0, it means that the storage bin is in the state of adsorption and gas storage, and the disturbance or humidity adjustment should be reduced;

[0046] S4.4: Based on the real-time net gas interaction rate of each area, establish a local control equation to dynamically simulate the flow, diffusion and feedback evolution of environmental factors in the bin, and combine the overall target environmental stability of the bin and the tea quality maintenance target to set an optimization target function, and adjust the overall environmental parameters of the storage bin in real time.

[0047] As a further solution of the present invention, the specific steps of dynamically formulating the tea storage strategy in step V are as follows:

[0048] S5.1: Real-time collect the metabolic intensity of each tea in the current silo area. According to the real-time collected metabolic intensity, calculate the physiological age of each tea. Then, obtain the optimal storage conditions of each tea based on expert experience and historical tea storage data, and obtain the current environmental stress index based on the deviation degree between the current environment and the optimal storage conditions of the tea;

[0049] S5.2: Jointly project the physiological age of the tea and the current environmental stress index onto a two-dimensional state space to form a tea storage state point, that is, the physiological maturity - environmental stress coordinates corresponding to the tea. Then, preset multiple response areas in the state space according to expert experience or historical experimental data;

[0050] S5.3: Dynamically predict the remaining storage period of each tea according to the current tea storage state point and the change trend of the physiological age of the corresponding tea. According to the predicted remaining storage period, dynamically adjust the storage strategy of each tea in the silo. At the same time, based on the adjusted storage strategy, update the physiological age of each tea and the current environmental stress index, and regenerate the corresponding tea storage state point.

[0051] Advantages of the present invention:

[0052] Through three-dimensional grid design and multi-point sensor layout, the present invention can real-time collect the temperature, humidity, gas concentration and other micro-environment parameters of each area in the silo, and based on the inverse distance weighted interpolation method, reconstruct the spatial environment continuous field of the uncovered silo area, generate a complete three-dimensional environmental gradient map. Then, cooperate with the arranged microfiber spectrometer array to collect the spectral reflectance of the tea, extract the characteristic bands through PCA and PLSR models and predict the tea quality. If the quality change rate exceeds the limit, the early warning mechanism will be triggered. At the same time, collect the tea respiration and adsorption behavior data, combine the respiration rate and adsorption flux models, calculate the net gas interaction rate and dynamically regulate the local environment accordingly. Further, fuse the physiological age of the tea and the environmental stress index, project and generate the tea storage state point, and predict the remaining storage period based on its change trend, dynamically adjust the personalized storage strategy of each tea, which can effectively avoid environmental blind spots, improve the accuracy and responsiveness of regulation, can real-time extract the quality deterioration signal, issue an early warning of quality abnormality in advance, greatly reduce the manual detection error and delay, improve the storage safety, realize the dynamic gas interaction evaluation between the tea and the in-silo environment, improve the scientificity and effectiveness of regulation, effectively avoid over-storage or early failure, realize differential and intelligent management, and significantly improve the automation and intelligent level. Description of the Drawings

[0053] The present invention will be further described below in conjunction with the accompanying drawings.

[0054] Figure 1 It is a system block diagram of a tea storage bin management system provided by an embodiment of the present invention;

[0055] Figure 2 It is a flowchart of a tea storage bin management method provided by an embodiment of the present invention. Specific embodiments

[0056] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0057] Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0058] An embodiment of the present invention provides a tea storage bin management system. Refer to Figure 1 , Figure 1 It is a system block diagram of a tea storage bin management system provided by an embodiment of the present invention. The system includes the following modules: a warehousing environment modeling module, an information input and classification module, a structure recognition and authentication module, an environmental gradient perception module, a respiration parameter monitoring module, a tea quality perception module, a storage environment regulation module, a warehousing opening and closing management module, a quality decay modeling module, a warehousing dynamic warning module, a storage strategy generation module, an anti-counterfeiting traceability security module, and a warehousing maintenance scheduling module.

[0059] The warehousing environment modeling module establishes a three-dimensional dynamic temperature and pressure field simulation model by collecting various climate parameters of the target tea production area.

[0060] Specifically, historical climate data of the origin are collected through meteorological station data, remote sensing satellite records, ground observations, and historical databases. After that, all the collected historical climate data are converted into international standard units. Then, the time scales of the historical climate data from different data sources are unified through linear interpolation. The abnormal data points in the climate data are identified and removed using the standard deviation judgment method. The abnormal data points in the climate data are filled by the forward and backward mean filling method. The processed groups of climate data are standardized. After that, all the standardized meteorological data are converted into a structured format. Based on the processed groups of climate data, the basic static pressure data required for different "equivalent altitude layers" in the warehouse are generated through the international standard atmospheric pressure model. Then, based on the heat conduction equation, the evolution of the temperature distribution in three-dimensional space over time is simulated, and a distribution map of the temperature in the simulated space in the warehouse is constructed. After that, a humidity-temperature coupling field is established to calculate the target humidity value in the warehouse. At the same time, the evolution of the temperature-pressure flow field of the air over time is simulated to obtain the change trend of the temperature-pressure field within a specified time in real time.

[0061] It should be further noted that the historical climate data of the origin include annual average temperature, diurnal temperature difference, seasonal temperature and humidity changes, wind speed, sunshine intensity, precipitation, altitude, etc.

[0062] The information entry and classification module is used to record the information of each batch of tea and manage the warehousing according to its characteristics; the structure identification and authentication module is used to uniquely identify the structural characteristics of each storage bin, bind it with the Internet of Things chip, and conduct digital identity authentication; the environmental gradient perception module collects the microenvironment parameters in real time by deploying a multi-point high-precision sensor array in the warehouse and constructs an environmental gradient map in the warehouse.

[0063] Specifically, the internal space of the storage bin is divided into three-dimensional grids according to the geometric structure of the storage bin. According to the point selection principle of equidistant and uniform distribution and enhanced layout in key areas, various sensors are deployed in the storage bin. At the same time, the three-dimensional coordinates of each sensor are marked. Based on the time sequence synchronization mechanism, the time stamps and recording frequencies are unified to record the microenvironment parameters in the storage bin collected by each sensor in real time. According to the deployment positions of each sensor, the spatial points not covered by the sensors are determined. The inverse distance weighted interpolation method is used to reconstruct the continuous environmental variable distribution at the non-covered spatial points to reconstruct the environmental continuous field at the non-covered spatial points in the storage bin. According to the microenvironment parameters collected in the storage bin, the gradients of the microenvironment parameters in space are calculated. The reconstructed environmental continuous field and the gradient calculation results are fused to generate a three-dimensional distribution map of the microenvironment parameters in the storage bin, that is, the environmental gradient map in the warehouse. The environmental gradient map in the warehouse is visualized in the form of color gradient. After that, the time sequence standard deviation of each area of the storage bin is calculated. If the time sequence standard deviation of the area exceeds the preset safety threshold, the local warning mechanism is triggered. Otherwise, the microenvironment parameters of each area of the storage bin are continuously detected.

[0064] In addition, it should be noted that the microenvironment parameters specifically include temperature, humidity, air pressure, oxygen concentration, CO2 concentration, light intensity, etc.

[0065] The respiration parameter monitoring module monitors the changes in the respiration intensity and adsorption characteristics of tea leaves under different environmental conditions in real time to obtain the respiration entropy changes of each tea leaf; the tea quality perception module collects the surface characteristic spectra of tea leaves in real time through a microfiber spectrometer array and dynamically monitors the tea quality.

[0066] Specifically, microfiber spectrometers are arranged in the storage bin, and a standard white light source is used to uniformly irradiate the tea leaves entering the bin at a preset irradiation angle and intensity. Each group of microfiber spectrometers records the spectral reflectance curves of the tea leaves in each area. At the same time, the collected spectral reflectance curves are smoothed and standardized. The characteristic bands of the components affecting the quality of each tea leaf are extracted from the processed spectral reflectance curves through the principal component analysis method. The extracted characteristic bands are used as independent variables, and the tea quality detection data at the corresponding time points are used as dependent variables. The independent variables and dependent variables are subjected to mean normalization processing. According to the independent variables and dependent variables after normalization processing, a PLSR prediction model is established. Then, the standardized quality prediction values of each tea leaf are output through the PLSR prediction model. The standardized prediction values are restored to the actual prediction quality values of the detected tea leaves, and the prediction quality values of each tea leaf are recorded. Every preset period, the spectral prediction values are automatically updated, compared with the historical spectral prediction values, the quality change rate is obtained, and a corresponding quality trend curve is established. If the quality change rate is higher than the set threshold, it is judged that the tea leaves have undergone significant deterioration, and an early warning of abnormal deterioration is issued. At the same time, the quality indicators of each tea leaf are displayed in real time in the form of a chart or heat map to obtain the actual quality values of each tea leaf in real time. Then, the performance of the PLSR prediction model is jointly measured based on the root mean square error and the coefficient of determination. If the performance of the PLSR prediction model is lower than the preset performance threshold, the parameters of the PLSR prediction model are adjusted through the Adam optimizer, and the PLSR prediction model is retrained and optimized using the historical tea quality values until the performance of the PLSR prediction model meets the preset threshold.

[0067] The storage environment control module automatically adjusts the environmental parameters of the storage bin according to the environmental gradient map in the bin and the respiration entropy change of the tea leaves; the storage opening and closing management module is used to automatically adjust the pressure difference between the inside and outside of the storage bin when opening and closing the bin door; the quality decay modeling module is used to predict the best storage period and potential risks of the tea leaves under the current conditions.

[0068] The warehousing dynamic warning module constructs a warehousing process map based on the environmental parameters, tea quality, and prediction results of the storage bin, and issues a warning for abnormal trends to prompt intervention; the storage strategy generation module is used to dynamically adjust the storage conditions and storage cycle to form a self-optimizing warehousing management strategy; the anti-counterfeiting traceability security module stores the warehousing information, storage process, environmental data, and quality information of each batch of tea on the blockchain; the warehousing maintenance scheduling module analyzes the probability of equipment failure, quality decline, or environmental fluctuations based on the prediction results and the warehousing process map, and plans maintenance tasks and bin adjustment operations in advance.

[0069] Based on a tea storage bin management system provided by an embodiment of the present invention, through.

[0070] An embodiment of the present invention also provides a method for managing a tea storage bin, as Figure 2 shown, the method includes the following steps:

[0071] Before the tea is warehoused, collect the climate parameters of the origin of the target tea, simulate the ecological environment of the target tea, and set it as the initial control target value of the storage bin.

[0072] Collect detailed information on each batch of tea and allocate each batch of tea to the corresponding storage area or independent bin.

[0073] Bind each storage bin to the tea batch through the Internet of Things platform. After the tea is officially warehoused, collect the micro-environment data, tea physiological metabolism indicators, and tea reflection spectrum in the bin in real time.

[0074] Based on the data collected in real time, identify the local change trend in the storage bin, automatically adjust the environmental parameters in the bin, and at the same time predict the tea deterioration trend and automatically trigger a warning.

[0075] Specifically, the oxygen consumption and carbon dioxide release during the tea metabolism process are collected in real time. Based on the temperature and oxygen partial pressure in the corresponding area of the storage bin, a respiration rate model per unit mass of tea is established. Then, using the modified Langmuir model, an adsorption response model of tea to moisture and gas is established. According to the established respiration rate model and adsorption response model, the respiration rate and adsorption flux of each area of the storage bin are obtained in real time, and the influence of the respiration rate and adsorption flux on the local microenvironment is calculated to obtain the net gas interaction rate between the corresponding tea and the environment. According to the net gas interaction rate of each group of areas divided in the storage bin, the environmental parameters of the corresponding area are adjusted. If the net gas interaction rate > 0, it means that the storage bin is in the state of oxygen consumption and moisture discharge, and ventilation needs to be strengthened. If the net gas interaction rate < 0, it means that the storage bin is in the state of adsorption and gas storage, and the disturbance or humidity adjustment should be reduced. Based on the real-time net gas interaction rate of each area, a local control equation is established to dynamically simulate the flow, diffusion, and feedback evolution of environmental factors in the bin. Combining the overall target environmental stability of the bin and the target of maintaining tea quality, an optimization objective function is set to adjust the overall environmental parameters of the storage bin in real time.

[0076] Dynamically formulate tea storage strategies. When tea is loaded into or unloaded from the bin, control the gas flow through the buffer chamber, first adjust the internal and external pressure difference, and then perform the loading and unloading operations.

[0077] Specifically, the metabolic intensity of each tea in the current storage bin area is collected in real time. According to the metabolically intensity collected in real time, the physiological age of each tea is calculated. Then, based on expert experience and historical tea storage data, the optimal storage conditions for each tea are obtained. Based on the deviation degree between the current environment and the optimal storage conditions of the tea, the current environmental stress index is obtained. The physiological age of the tea and the current environmental stress index are jointly projected onto a two-dimensional state space to form a tea storage state point, that is, the physiological maturity–environmental stress coordinate corresponding to the tea. Then, according to expert experience or historical experimental data, multiple response areas are preset in the state space. According to the current tea storage state point and the change trend of the physiological age of the corresponding tea, the remaining storage period of each tea is dynamically predicted. According to the predicted remaining storage period, the storage strategies of each tea in the storage bin are dynamically adjusted. At the same time, based on the adjusted storage strategies, the physiological age of each tea and the current environmental stress index are updated, and the corresponding tea storage state point is regenerated.

[0078] The above has described a specific embodiment of the present invention in detail, but the content is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the present invention application should still fall within the scope covered by the patent of the present invention.

Claims

1. A tea storage bin management system, characterized in that, include: Warehouse environment modeling module, information entry classification module, structure recognition and authentication module, environmental gradient perception module, respiratory parameter monitoring module, tea quality perception module, storage environment control module, warehouse opening and closing management module, quality decay modeling module, warehouse dynamic early warning module, storage strategy generation module, anti-counterfeiting traceability security module and warehouse maintenance scheduling module; The storage environment modeling module establishes a three-dimensional dynamic temperature and pressure field simulation model by collecting various characteristic data of the target tea producing area; The information entry classification module is used to record various information of each batch of tea and carry out zoning warehousing management according to its characteristics; The structural identification and authentication module is used to uniquely identify the structural features of each storage silo and bind it to the Internet of Things chip to perform digital identity authentication; The environmental gradient perception module collects various micro-environmental parameters in real time and constructs an environmental gradient map in the warehouse by deploying a multi-point high-precision sensor array in the warehouse; The respiratory parameter monitoring module monitors the changes in the respiratory intensity and adsorption characteristics of tea leaves under different environmental conditions in real time to obtain the respiratory entropy changes of each tea leaf; The tea quality sensing module collects the characteristic spectrum of the tea surface in real time through a micro-fiber spectrometer array, and dynamically monitors the quality of the tea; The storage environment control module automatically adjusts the environmental parameters of the storage bin according to the internal environment gradient map and the tea respiration entropy change; The storage opening and closing management module is used to automatically adjust the pressure difference between the inside and outside of the storage bin when opening and closing the bin door; The quality decay modeling module is used to predict the optimal storage period and potential risks of tea under current conditions; The storage dynamic warning module constructs a storage process map based on the environmental parameters of the storage bin, the quality of tea leaves and the prediction results, and issues warnings for abnormal trends, prompting intervention; The storage strategy generation module is used to dynamically adjust storage conditions and storage cycles to form a self-optimizing storage management strategy; The anti-counterfeiting and traceability security module stores the warehouse entry information, storage process, environmental data and quality information of each batch of tea on the chain; The warehouse maintenance scheduling module analyzes the probability of equipment failure, quality degradation or environmental fluctuation based on the prediction results and the warehouse process map, and plans maintenance tasks and warehouse adjustment operations in advance.

2. The tea storage bin management system according to claim 1, characterized in that, The specific steps of establishing the three-dimensional dynamic temperature and pressure field simulation model by the storage environment modeling module are as follows: S1.1: Collect historical climate data of the origin through meteorological station data, remote sensing satellite records, ground observations and historical databases, then convert all collected historical climate data into international standard units, and then unify the time scales of historical climate data from different data sources through linear interpolation; S1.2: Use the standard deviation judgment method to identify and remove abnormal data points in the climate data, then fill in the abnormal data points in the climate data through the front and back mean completion method, standardize each group of processed climate data, and then convert all standardized meteorological data into a structured format; S1.3: Based on the processed climate data of each group, generate the basic static pressure data required for different "equivalent altitude layers" in the silo through the international standard atmospheric pressure model. Then, based on the heat conduction equation, simulate the evolution of the temperature distribution in three-dimensional space over time, and construct a distribution map of the temperature in the simulated space of the silo. After that, establish a humidity-temperature coupling field to calculate the target humidity value in the silo, and simultaneously simulate the evolution of the temperature-pressure flow field of the air over time to obtain the change trend of the temperature-pressure field within a specified time in real time.

3. The tea storage bin management system according to claim 2, characterized in that, The specific steps for the environmental gradient perception module to construct the environmental gradient map in the silo are as follows: S2.1: Conduct three-dimensional grid division on the internal space according to the geometric structure of the storage silo. According to the point selection principle of equidistant uniform distribution and enhanced layout in key areas, deploy various sensors in the storage silo, mark the three-dimensional coordinates of each sensor at the same time, and based on the time series synchronization mechanism, unify the time stamps and recording frequencies to record the micro-environment parameters in the storage silo collected by each sensor in real time. S2.2: Determine the spatial points not covered by the sensors according to the deployment positions of the sensors. Use the inverse distance weighted interpolation method to reconstruct the continuous environmental variable distribution at the non-covered spatial points to reconstruct the environmental continuous field of the non-covered spatial points in the storage silo. Calculate the gradients of each micro-environment parameter in space according to the micro-environment parameters collected in the storage silo. S2.3: Integrate the reconstructed environmental continuous field and the gradient calculation results to generate a three-dimensional distribution map of each micro-environment parameter in the storage silo, that is, the environmental gradient map in the silo. Visualize the environmental gradient map in the silo in the form of color gradient. Then calculate the time series standard deviation of each area in the storage silo. If the time series standard deviation of the area exceeds the preset safety threshold, trigger the local warning mechanism. Otherwise, continuously detect the micro-environment parameters of each area in the storage silo.

4. A tea storage bin management system according to claim 1, characterized in that, The specific steps for the tea quality perception module to dynamically monitor the tea quality are as follows: S3.1: Deploy a micro fiber optic spectrometer in the storage silo, and use a standard white light source to uniformly irradiate the tea entering the silo according to the preset irradiation angle and intensity. Each group of micro fiber optic spectrometers records the spectral reflectance curves of the tea in each area, and at the same time, smooth and standardize the collected spectral reflectance curves. S3.2: Extract the characteristic bands of each tea quality influencing component from the processed spectral reflectance curves through the principal component analysis method. Take the extracted characteristic bands as independent variables, and take the tea quality detection data at the corresponding time points as dependent variables, and perform mean normalization processing on each independent variable and dependent variable. S3.3: Establish a PLSR prediction model according to the normalized independent variables and dependent variables. Then, output the standardized quality prediction values of each tea through the PLSR prediction model, and restore the standardized prediction values to the actual prediction quality values of each detected tea, and record the prediction quality values of each tea. S3.4: Automatically update the spectral prediction value at preset intervals, compare it with the historical spectral prediction value, obtain the quality change rate, and establish the corresponding quality trend curve. If the quality change rate is higher than the set threshold, it is determined that the tea has significantly deteriorated, and an early warning of abnormal deterioration is given. At the same time, display each index of the tea quality in real time in the form of a chart or heat map; S3.5: Obtain the actual quality value of each tea in real time, and then jointly measure the performance of the PLSR prediction model based on the root mean square error and the coefficient of determination. If the performance of the PLSR prediction model is lower than the preset performance threshold, adjust the parameters of the PLSR prediction model through the Adam optimizer, and retrain and optimize the PLSR prediction model using the historical tea quality values until the performance of the PLSR prediction model meets the preset threshold.

5. A method for managing a tea storage bin, which is used to implement the functions of the tea storage bin management system described in any one of claims 1-4, characterized in that, The method includes the following steps: Ⅰ. Before the tea is stored in the warehouse, collect the climate parameters of the origin of the target tea, simulate the ecological environment of the target tea, and set it as the initial control target value of the storage bin; Ⅱ. Collect the detailed information of each batch of tea, and allocate each batch of tea to the corresponding storage area or independent compartment; Ⅲ. Bind each storage bin with the tea batch through the Internet of Things platform. After the tea is officially stored in the warehouse, collect the microenvironment data, tea physiological metabolism indicators, and tea reflection spectrum in the warehouse in real time; Ⅳ. Based on the data collected in real time, identify the local change trend in the storage bin, automatically adjust the environmental parameters in the bin, and predict the tea deterioration trend at the same time; Ⅴ. Dynamically formulate the tea storage strategy, and when the tea is stored in and out of the warehouse, control the gas flow through the buffer chamber, first adjust the internal and external pressure difference, and then perform the storage and out-of-warehouse operations.

6. The method for managing a tea storage bin according to claim 5, characterized in that, The specific steps of automatically adjusting the environmental parameters in the bin in step Ⅳ are as follows: S4.1: Collect the oxygen consumption and carbon dioxide release during the tea metabolism process in real time, establish a tea unit mass respiration rate model based on the temperature and oxygen partial pressure in the corresponding area of the storage bin, and then use the modified Langmuir model to establish an adsorption response model of tea to moisture and gas; S4.2: According to the established respiration rate model and adsorption response model, obtain the respiration rate and adsorption flux of each area of the storage bin in real time, and calculate the influence of the respiration rate and adsorption flux on the local microenvironment to obtain the net gas interaction rate between the corresponding tea and the environment; S4.3: Adjust the environmental parameters of the corresponding area according to the net gas interaction rate of each group of areas divided in the storage bin. If the net gas interaction rate > 0, it means that the storage bin is in the state of oxygen consumption and moisture discharge, and ventilation needs to be strengthened. If the net gas interaction rate < 0, it means that the storage bin is in the state of adsorption and gas storage, and the disturbance or humidity adjustment should be reduced; S4.4: Based on the real-time net gas interaction rate of each area, establish a local control equation to dynamically simulate the flow, diffusion, and feedback evolution of environmental factors in the bin, and combine the overall target environmental stability of the bin and the tea quality maintenance target to set the optimization objective function, and adjust the overall environmental parameters of the storage bin in real time.

7. A method for managing a tea storage bin according to claim 5, characterized in that, The specific steps of dynamically formulating the tea storage strategy in step Ⅴ are as follows: S5.1: Real-time collect the metabolic intensity of each tea leaf in the current storage bin area. According to the real-time collected metabolic intensity, calculate the physiological age of each tea leaf. Then, obtain the optimal storage conditions for each tea leaf based on expert experience and historical tea leaf storage data, and based on the deviation degree between the current environment and the optimal storage conditions of the tea leaf, obtain the current environmental stress index; S5.2: Jointly project the physiological age of the tea leaf and the current environmental stress index onto a two-dimensional state space to form a tea leaf storage state point, that is, the physiological maturity - environmental stress coordinate corresponding to the tea leaf. Then, preset multiple response regions in the state space according to expert experience or historical experimental data; S5.3: Dynamically predict the remaining storage period of each tea leaf according to the current tea leaf storage state point and the change trend of the physiological age of the corresponding tea leaf. According to the predicted remaining storage period, dynamically adjust the storage strategy of each tea leaf in the storage bin. At the same time, based on the adjusted storage strategy, update the physiological age of each tea leaf and the current environmental stress index, and regenerate the corresponding tea leaf storage state point.

Images