Intelligent control method and system for fresh-keeping environment of roxburghii fruit based on Internet of Things

Abstract

The invention discloses an intelligent control method and system for a fresh-keeping environment of roxburghii fresh fruit based on the Internet of Things, and specifically relates to the technical field of fresh-keeping environment control, and is used to solve the problem that traditional fresh-keeping technologies are often unable to effectively cope with physiological changes of fruits and dynamic changes of external environments during storage. A three-dimensional air flow model of a storage environment is constructed based on the influence of thorn-like structures on the surface of roxburghii fresh fruit on air flow by utilizing a computational fluid dynamics algorithm; an Internet of Things sensor network is deployed in the storage environment to monitor air flow data in real time, and to identify micro-turbulence areas and air stagnation points; based on real-time collected data and the three-dimensional model, a machine learning algorithm is used to adjust the operating parameters of an air circulation system; the abnormal degree of heat flux density change and respiratory jump phenomenon are comprehensively analyzed to establish a physiological state prediction model for roxburghii fresh fruit, and a control algorithm is used to intelligently adjust the temperature, humidity and gas composition in the storage environment.

Description

Technical Field

[0001] The present invention relates to the technical field of fresh-keeping environment regulation and control, and more specifically, to an intelligent regulation and control method for the fresh-keeping environment of roxburghii fresh fruit based on the Internet of Things and a system thereof. Background Art

[0002] Fresh seabuckthorn fruit is a seasonal fruit, mainly produced in Guizhou. It has rich nutritional value and wide application. Fresh seabuckthorn fruit is also known as the "king of vitamin C". The development of modern agriculture and cold chain logistics has made the preservation technology of fresh fruits more and more important. Fresh seabuckthorn fruit is a nutritious fruit, and intelligent control of its preservation environment is crucial to extending the shelf life and maintaining fruit quality.

[0003] However, traditional preservation technology mainly relies on static temperature and humidity control, and often cannot effectively cope with the physiological changes of fruits during storage and the dynamic changes of the external environment, which will lead to deterioration of fruit quality, loss of nutrients, and increased risk of decay, thereby shortening the shelf life and affecting market value and consumer experience.

[0004] In order to solve the above problems, a technical solution is now provided. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide a method and system for intelligently controlling the fresh-keeping environment of fresh roxburghii fruit based on the Internet of Things to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The method for intelligently controlling the fresh-keeping environment of roxburghii fresh fruit based on the Internet of Things comprises the following steps:

[0008] S1: Using computational fluid dynamics algorithms, a three-dimensional air flow model of the storage environment was constructed based on the interference of the thorny structure on the surface of fresh roxburghii fruit on air flow;

[0009] S2: Deploy an IoT sensor network in the storage environment to collect air flow data in real time and identify micro-turbulence areas and air stagnation points;

[0010] S3: Based on the collected real-time air flow data and three-dimensional air flow model, the machine learning algorithm is used to adjust the operating parameters of the air circulation system to determine whether to trigger the air flow path that optimizes the storage environment;

[0011] S4: When the air flow path of the storage environment is optimized: a heat flow model is constructed using the finite element analysis method to simulate the heat flow distribution under different air flow conditions, and a transient thermal analysis is performed to evaluate the abnormal degree of the heat flux density change in the storage environment of the fresh roxburghii fruit; by analyzing the entropy change of the respiration rate signal of the fresh roxburghii fruit, the degree of potential respiratory jump phenomenon in the fresh roxburghii fruit is identified;

[0012] S5: Comprehensively analyze the abnormal degree of heat flux density change in the storage environment of fresh roxburghii fruit and the degree of potential respiratory jump phenomenon in fresh roxburghii fruit, and establish a prediction model for the physiological state of fresh roxburghii fruit;

[0013] S6: Based on the output of the physiological state prediction model of fresh sea buckthorn fruit, the control algorithm is used to adjust the temperature, humidity and gas composition in the storage environment.

[0014] In a preferred embodiment, S1 specifically includes:

[0015] S101: According to the characteristics of the thorn-like structure on the surface of fresh roxburghii fruit, determine the key interference factors and main influencing areas of air flow; the main influencing areas include the area adjacent to the fruit surface, the air flow channel around the thorn-like structure, and the air flow outlet area;

[0016] S102: Select computational fluid dynamics equations applicable to small-scale flows, the computational fluid dynamics equations including the Navier-Stokes equations;

[0017] S103: setting boundary conditions of the three-dimensional model according to the storage environment and the fruit structure, including the air inlet, the air outlet, and the fruit surface;

[0018] S104: Adopting an iterative numerical solution method and combining a meshing technique, a three-dimensional air flow model of the storage environment is constructed, and air flow path data is output.

[0019] In a preferred embodiment, S2 specifically includes:

[0020] S201: Select an air flow sensor and reasonably arrange the sensor network according to the characteristics of the three-dimensional air flow model;

[0021] S202: Setting the sampling frequency and data storage strategy of the sensor to collect air speed and direction data in real time;

[0022] S203: The collected data is transmitted in real time through the Internet of Things, and noise is removed by using data filtering and preprocessing algorithms;

[0023] S204: Based on real-time air flow data, use fluid dynamics analysis methods to identify micro-turbulence areas and air stagnation points:

[0024] Based on real-time air flow data, fluid dynamics analysis methods are used to identify micro-turbulence areas and air stagnation points; a velocity distribution model of the air flow field is established to calculate the local turbulent kinetic energy of the fluid: ;in, is the local turbulent kinetic energy of the fluid, represents the velocity component at a specific point, is the average velocity component around that particular point, is the index, and the specific point is each computational grid point in the air flow field;

[0025] A preset threshold value of the local turbulent kinetic energy of the fluid is set, and when the local turbulent kinetic energy of the fluid is greater than the preset threshold value of the local turbulent kinetic energy of the fluid, it is judged to be a micro-turbulent area;

[0026] By analyzing the separation of the fluid in the velocity field, air stagnation points are identified.

[0027] In a preferred embodiment, S3 specifically includes:

[0028] S301: Fusing the real-time collected air flow data with the three-dimensional air flow model to construct a comprehensive flow characteristic model:

[0029] Standardize the air flow data to obtain the standardized air flow data ;

[0030] Use the spatial interpolation method to map the real-time data to the model grid nodes, and set the interpolation function to ,but: ;in, is the value of the mapped flow velocity in the three-dimensional grid, is the interpolation function, is a spatial variable in a three-dimensional coordinate system;

[0031] Fill missing data through interpolation methods;

[0032] S302: Extract key features from the comprehensive flow characteristic model, including turbulent kinetic energy and flow velocity distribution, etc.:

[0033] Identify important parameters that affect the operation of air circulation systems;

[0034] Arrange the extracted features into a matrix form;

[0035] S303: Select a machine learning algorithm to train a model based on historical data to learn optimal operating parameters of the air circulation system;

[0036] S304: Using the trained model, calculate and adjust the operating parameters of the air circulation system in real time:

[0037] Input the newly collected feature data into the machine learning model; adjust the operating parameters of the air circulation system in real time according to the model output;

[0038] S305: Determine whether the current air flow state matches the optimal parameters output by the model, and decide whether to trigger further air flow path optimization:

[0039] Compare the difference between the current flow state and the model output parameters and calculate the matching index;

[0040] When the matching index is lower than its corresponding preset threshold, it is determined to trigger further air flow path optimization.

[0041] In a preferred embodiment, a heat flow model is constructed using a finite element analysis method to simulate the heat flow distribution under different air flow conditions, and a transient thermal analysis is performed to evaluate the abnormal degree of heat flux density change in the storage environment of fresh roxburghii fruit, specifically:

[0042] The heat conduction equation is used to describe the heat flow behavior: ;in, represents temperature, Represents time, is the thermal diffusivity;

[0043] Perform mesh division;

[0044] Set the time step and total analysis time;

[0045] The set heat flow model is solved using finite element analysis software; after the solution, key heat flow distribution data is extracted through post-processing tools;

[0046] Calculate the standard deviation of the heat flux: ;in, is the standard deviation of the heat flux, For the The heat flux density at each node is is the number of nodes, is the average value of the node heat flux density;

[0047] Collect the heat flux data of all nodes and integrate them into a heat flux data set, which includes the heat flux value of each node and the corresponding time step;

[0048] Calculate the rate of change of heat flux density, which is defined as the difference in heat flux density at consecutive time steps;

[0049] By analyzing the standard deviation and change rate of heat flux density, the evaluation value of heat flux anomaly is calculated, and its expression is: ;in, is the evaluation value of heat flow anomaly degree, is the threshold value corresponding to the rate of change of heat flux density, is the rate of change of heat flux, and They are and The weight coefficient, and and Both are greater than 0.

[0050] In a preferred embodiment, the degree of potential respiratory climacteric phenomenon in the fresh fruit of Rosa roxburghii is identified by analyzing the entropy change of the respiratory rate signal of the fresh fruit of Rosa roxburghii, specifically:

[0051] Real-time monitoring of the respiration rate of fresh roxburghii fruit, and collection of time series data of the respiration rate;

[0052] Preprocessing the respiratory rate signal of the respiratory rate time series data, the preprocessing steps including smoothing and standardization;

[0053] The respiratory rate signal is divided into multiple overlapping windows, each of which moves with a fixed step size;

[0054] For each window, calculate the entropy of the respiration rate signal: ;in, represents the entropy of the respiratory rate signal, Representation characteristics The probability distribution of represents the total number of features, Represents the feature index;

[0055] Among them, the characteristics including breathing rate, temperature changes, humidity levels, and gas concentrations;

[0056] The absolute value obtained by subtracting the entropy of the real-time respiratory rate signal from the entropy of the last monitored respiratory rate signal is marked as the respiratory rate entropy change index.

[0057] In a preferred embodiment, the abnormal degree of heat flux density change in the storage environment of roxburghii fresh fruit and the degree of potential respiratory climacteric phenomenon in roxburghii fresh fruit are comprehensively analyzed to establish a physiological state prediction model for roxburghii fresh fruit, specifically:

[0058] The heat flow abnormality evaluation value and the respiratory rate entropy change index are normalized, and the normalized heat flow abnormality evaluation value and the respiratory rate entropy change index are respectively assigned preset proportional coefficients to calculate the physiological prediction coefficient of fresh sea buckthorn fruit.

[0059] In a preferred embodiment, based on the output of the physiological state prediction model of fresh roxburghii fruit, a control algorithm is used to adjust the temperature, humidity and gas composition in the storage environment, specifically:

[0060] Real-time acquisition of the output value of the current physiological prediction coefficient of fresh roxburghii fruit;

[0061] Based on the physiological prediction coefficient of fresh roxburghii fruit, the corresponding control target is set, and the output of the physiological state prediction model of fresh roxburghii fruit is used to dynamically adjust the temperature target, humidity target and gas composition target;

[0062] Execute the control algorithm, the specific steps include:

[0063] Set current environmental parameters, including temperature, humidity, and gas composition; calculate the difference between the target and current values ​​to obtain the adjustment amount;

[0064] Temperature adjustment: If the adjustment amount is positive, the heater is started; if the adjustment amount is negative, the cooler is started;

[0065] Humidity adjustment: If the adjustment amount is positive, the humidifier is started; if the adjustment amount is negative, the dehumidifier is started;

[0066] Gas composition adjustment: Adjust the gas concentration according to the gas sensor data.

[0067] On the other hand, the present invention provides an intelligent control system for fresh-keeping environment of roxburghii fruit based on the Internet of Things, comprising a flow model construction module, a real-time data acquisition module, an operation parameter optimization module, a heat flow respiration analysis module, a physiological state prediction module and an environmental parameter control module;

[0068] Flow model building module: Using computational fluid dynamics algorithms, based on the interference of the thorny structure on the surface of fresh roxburghii fruit on air flow, a three-dimensional air flow model of the storage environment is constructed;

[0069] Real-time data collection module: deploy an IoT sensor network in the storage environment to collect air flow data in real time and identify micro-turbulence areas and air stagnation points;

[0070] Operation parameter optimization module: Based on the collected real-time air flow data and three-dimensional air flow model, the machine learning algorithm is used to adjust the operating parameters of the air circulation system to determine whether to trigger the air flow path to optimize the storage environment;

[0071] Thermal flow respiration analysis module: When the air flow path of the storage environment is optimized, the finite element analysis method is used to construct a thermal flow model to simulate the heat flow distribution under different air flow conditions, and transient thermal analysis is performed to evaluate the abnormal degree of heat flux density changes in the storage environment of the fresh roxburghii fruit; by analyzing the entropy changes of the respiration rate signal of the fresh roxburghii fruit, the degree of potential respiratory jump phenomenon in the fresh roxburghii fruit is identified;

[0072] Physiological state prediction module: Comprehensively analyze the abnormal degree of heat flux density change in the storage environment of fresh roxburghii fruit and the degree of potential respiratory jump phenomenon in fresh roxburghii fruit, and establish a physiological state prediction model for fresh roxburghii fruit;

[0073] Environmental parameter control module: Based on the output of the physiological state prediction model of fresh sea buckthorn fruit, the control algorithm is used to adjust the temperature, humidity and gas composition in the storage environment.

[0074] The technical effects and advantages of the intelligent control method for fresh-keeping environment of roxburghii fruit based on the Internet of Things and the system thereof are as follows:

[0075] 1. By using computational fluid dynamics algorithms to construct a three-dimensional air flow model, the effect of the thorny structure on the surface of fresh sea buckthorn fruit on air flow can be effectively analyzed, thereby optimizing air flow in the storage environment, improving the ventilation of the fruit, and reducing the risk of preservation caused by air stagnation. Deploying an IoT sensor network in the storage environment can collect air flow data in real time, identify micro-turbulence areas and air stagnation points, ensure timely response to changes in the storage environment, and ensure optimal storage conditions for the fruit. Using machine learning algorithms to analyze real-time data, dynamically adjust the operating parameters of the air circulation system, and realize intelligent control of the storage environment. This method not only improves the efficiency of preservation, but also reduces the need for manual intervention and reduces the risk of human error.

[0076] 2. By combining the heat flux model with the entropy change of the respiration rate signal, the physiological state of fresh roxburghii fruit can be accurately assessed, potential respiratory jump phenomena can be identified, and a deep insight into the physiological changes of the fruit can be provided, which helps to timely detect the risk of fruit ripening or corruption. By comprehensively analyzing the abnormal degree of heat flux density changes and respiratory jump phenomena, the established physiological state prediction model can provide a scientific basis for the adjustment of the storage environment, thereby achieving more accurate control of temperature, humidity and gas composition, extending the shelf life of fresh roxburghii fruit, and maintaining its nutrition and taste. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 It is a schematic diagram of the intelligent control method of the fresh-keeping environment of roxburghii fresh fruit based on the Internet of Things of the present invention;

[0078] Figure 2 The present invention is a structural schematic diagram of an intelligent control system for preserving environment of fresh roxburghii fruit based on the Internet of Things. DETAILED DESCRIPTION

[0079] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0080] Example 1

[0081] Figure 1 The present invention provides an intelligent control method for the preservation environment of fresh roxburghii fruit based on the Internet of Things, which comprises the following steps:

[0082] S1: Using computational fluid dynamics algorithms, a three-dimensional air flow model of the storage environment was constructed based on the interference of the thorny structure on the surface of fresh sea buckthorn fruit on air flow.

[0083] S2: Deploy an IoT sensor network in the storage environment to collect air flow data in real time and identify micro-turbulence areas and air stagnation points.

[0084] S3: Based on the collected real-time air flow data and three-dimensional air flow model, the machine learning algorithm is used to adjust the operating parameters of the air circulation system to determine whether to trigger the air flow path that optimizes the storage environment.

[0085] S4: When the air flow path of the storage environment is optimized: the finite element analysis method is used to construct a heat flow model to simulate the heat flow distribution under different air flow conditions, and transient thermal analysis is performed to evaluate the abnormal degree of the heat flux density change in the storage environment of the seabuckthorn fresh fruit; by analyzing the entropy changes of the respiratory rate signal of the seabuckthorn fresh fruit, the degree of potential respiratory jump phenomenon in the seabuckthorn fresh fruit is identified.

[0086] S5: Comprehensively analyze the abnormal degree of heat flux density changes in the storage environment of fresh seabuckthorn fruit and the degree of potential respiratory jump phenomenon in fresh seabuckthorn fruit, and establish a physiological state prediction model for fresh seabuckthorn fruit.

[0087] S6: Based on the output of the physiological state prediction model of fresh sea buckthorn fruit, the control algorithm is used to adjust the temperature, humidity and gas composition in the storage environment.

[0088] S1 specifically includes:

[0089] S101: According to the characteristics of the thorny structure on the surface of fresh roxburghii fruit, determine the key interference factors and main influencing areas of air flow:

[0090] Based on the unique thorn-like structure on the surface of fresh sea buckthorn fruit, its interference characteristics on air flow are analyzed. These thorn-like structures produce tiny turbulence and air stagnation by changing the direction and speed of the airflow. By measuring and analyzing the geometric features of the fruit surface, key interference factors are identified, such as the height, spacing and arrangement angle of the thorns. The key interference factors directly affect the distribution pattern and flow characteristics of the airflow.

[0091] Identify key impact areas, including:

[0092] The area near the fruit surface is the area where air flow is most complex. The airflow is directly affected by the thorn-like structure, and micro-turbulence and stagnation zones are easily formed.

[0093] Airflow channels around thorn-like structures: Areas where air flows between thorns may produce local low-pressure and high-pressure areas, affecting the overall air flow characteristics.

[0094] Airflow outlet area: The area where air flows away from the fruit surface. Changes in flow rate and direction will affect the surrounding environment.

[0095] S102: Select computational fluid dynamics equations suitable for small-scale flows to ensure accurate simulation of micro-turbulence and air stagnation points:

[0096] To ensure accurate simulation of microscale flow, computational fluid dynamics (CFD) equations suitable for small-scale flow are selected, such as the Navier-Stokes equations, which can describe the movement of viscous fluids and are suitable for analyzing the behavior of air flowing on the surface of fruit. The specific form is: ;in, is the fluid velocity vector, It's time. is the fluid density, is the fluid pressure, is the kinematic viscosity, is the acceleration due to gravity.

[0097] in, Represents partial derivative, which is used to describe the change of a multivariable function with respect to a certain variable; is a gradient operator used to represent the gradient of a vector field.

[0098] The Navier-Stokes equations can effectively capture the formation of microturbulence and stagnation points caused by the thorny structures by describing the fluid dynamics behavior.

[0099] S103: According to the storage environment and fruit structure, the boundary conditions of the three-dimensional model are set:

[0100] According to the storage environment and the specific structure of the fruit, the boundary conditions of the 3D model are set, including the air inlet, air outlet and fruit surface; the specific parameters are set as follows:

[0101] Air inlet: Set the air flow speed usually between 0.1 m / s and 0.5 m / s to simulate the air flow in the actual storage environment.

[0102] Air outlet: Set the zero pressure condition, that is, the pressure at the outlet is atmospheric pressure, indicating free flow of fluid.

[0103] Fruit surface: The no-slip condition is set, which means that the velocity of the fluid on the fruit surface is zero.

[0104] The setting of boundary conditions will ensure that the actual situation of fluid flow is reasonably reflected during the simulation process and the key features of the flow can be effectively captured.

[0105] S104: Using iterative numerical solution method and meshing technology, a three-dimensional air flow model of the storage environment is constructed and air flow path data is output:

[0106] The meshing process includes the following steps:

[0107] According to the geometric shape of the fruit surface, an appropriate computational grid is generated, and the fruit surface and surrounding areas are refined to improve the analytical ability of local flow.

[0108] The Navier-Stokes equations are discretized using numerical solutions such as the finite volume method and solved iteratively using time steps. Each time step is set to 0.01 seconds to ensure convergence speed and computational stability.

[0109] By solving the velocity field and pressure field, the air flow path data can be output, which provides basic information for subsequent steps and helps to further optimize the storage environment.

[0110] S2 specifically includes:

[0111] S201: Select high-precision air flow sensors and rationally layout the sensor network based on the characteristics of the three-dimensional air flow model:

[0112] According to the characteristics of the three-dimensional air flow model, suitable high-precision air flow sensors (such as ultrasonic anemometers or hot wire anemometers) are selected. These sensors can accurately measure the speed and direction of the air. When laying out the sensor network, the thorny structure on the surface of the fruit and the expected air flow pattern are taken into consideration to ensure that the sensors cover all key areas, especially potential micro-turbulence areas and air stagnation points. Reasonable layout can ensure the comprehensiveness and representativeness of the data.

[0113] S202: Set the sampling frequency and data storage strategy of the sensor to ensure real-time collection of air speed and direction data, and ensure the timeliness and accuracy of the data:

[0114] To ensure real-time collection of air flow data, set the sampling frequency and data storage strategy of the sensor. For example, the sampling frequency is 10 times per second to capture rapidly changing flow characteristics. The data storage strategy should adopt a circular storage method to ensure that the latest data is kept in the case of limited memory, and regular backup is performed to avoid data loss. Through appropriate configuration, the timeliness and accuracy of the data can be effectively guaranteed.

[0115] S203: The collected data is transmitted in real time through the Internet of Things, and noise is removed by data filtering and preprocessing algorithms to ensure data quality:

[0116] The real-time collected air flow data is transmitted in real time through the Internet of Things, and data filtering and preprocessing algorithms are used to remove noise to ensure data quality. Specific algorithms may include Kalman filtering to reduce sensor measurement errors and improve data reliability.

[0117] S204: Based on real-time air flow data, use fluid dynamics analysis methods to identify micro-turbulence areas and air stagnation points:

[0118] Based on real-time air flow data, fluid dynamics analysis methods are used to identify micro-turbulence areas and air stagnation points. During the analysis, the velocity distribution model of the air flow field is first established, and the local turbulent kinetic energy of the fluid is calculated, which is defined as: ;in, is the local turbulent kinetic energy of the fluid, represents the velocity component at a specific point, is the average velocity component around that particular point, are indices representing the three velocity components of the fluid.

[0119] Among them, the specific point refers to each computational grid point obtained by discretization in the air flow field. These points have clear three-dimensional spatial coordinates and are used to calculate the fluid dynamics parameters at that position, such as local turbulent kinetic energy.

[0120] A preset threshold value of the local turbulent kinetic energy of the fluid is set, and when the local turbulent kinetic energy of the fluid is greater than the preset threshold value of the local turbulent kinetic energy of the fluid, it is determined to be a micro-turbulent area.

[0121] At the same time, by analyzing the separation phenomenon of the fluid in the velocity field, air stagnation points are identified, ensuring a comprehensive understanding of the air flow in the storage environment and providing basic data support for subsequent air flow optimization.

[0122] S3 specifically includes:

[0123] S301: Fusing the real-time collected air flow data with the three-dimensional air flow model to construct a comprehensive flow characteristic model:

[0124] The real-time collected air flow data is recorded as , the three-dimensional air flow model is recorded as , integrate air flow data and 3D air flow model to build a comprehensive flow characteristic model This process includes:

[0125] Standardize air flow data to eliminate the impact of different units and obtain standardized air flow data .

[0126] Use the spatial interpolation method to map the real-time data to the model grid nodes, and set the interpolation function to ,but: ;in, is the value of the mapped flow velocity in the three-dimensional grid; is the interpolation function used to map normalized data to specific grid points; is a spatial variable in a three-dimensional coordinate system and is used to represent the position in the air flow model.

[0127] The missing data are filled by interpolation to ensure the continuity of the model. The missing values ​​are assumed to be , the nearest neighbor interpolation method can be used: ;in, is the flow rate data after filling; and is the nearest neighbor velocity value, used to calculate the fill value.

[0128] The flow velocity and turbulent kinetic energy are obtained respectively. The calculation process of the flow velocity and turbulent kinetic energy will not be repeated here.

[0129] S302: Extract key features from the comprehensive flow feature model, including turbulent kinetic energy and velocity distribution, to provide input for the machine learning algorithm:

[0130] Identify important parameters that affect the operation of air circulation systems, such as standard deviation and mean flow rate.

[0131] Organize the extracted features into matrix form , specifically: ; is the mean flow velocity, is the standard deviation of the flow rate.

[0132] S303: Select a suitable machine learning algorithm, train the model based on historical data, and learn the optimal operating parameters of the air circulation system:

[0133] Select an appropriate machine learning algorithm (such as support vector machine or random forest) to train the model. The specific steps are:

[0134] S303: Select a suitable machine learning algorithm, train the model based on historical data, and learn the optimal operating parameters of the air circulation system:

[0135] Select an appropriate machine learning algorithm (such as support vector machine or random forest) to train the model. The specific steps are:

[0136] The historical data is divided into a training set and a test set. The training set contains samples ;in, is the training set, which contains the feature matrix and the optimal parameters; are the optimal operating parameters for the output of the machine learning model.

[0137] Use the selected algorithm to train the training set and obtain the best running parameter model For example, if we use random forest: .

[0138] S304: Using the trained model, calculate and adjust the operating parameters of the air circulation system in real time:

[0139] The newly collected feature data Input to the machine learning model. Adjust the operating parameters of the air circulation system in real time based on the model output: ;in, It is the operating parameter of real-time calculation.

[0140] S305: Determine whether the current air flow state matches the optimal parameters output by the model, and decide whether to trigger further air flow path optimization:

[0141] Compare the difference between the current flow state and the model output parameters and calculate the matching index : ;in, is the current flow state, is the total number of comparison parameters.

[0142] When the matching index When it is lower than the corresponding preset threshold, it is determined to trigger further air flow path optimization.

[0143] Among them, the matching index The setting of the corresponding preset threshold should be based on historical data analysis and system performance requirements. First, by statistically analyzing the historical operating data, the difference between the flow state and the optimal parameters is identified, and a reasonable initial threshold is determined. Secondly, the initial threshold is adjusted based on the prediction accuracy of the fluid mechanics model and the machine learning algorithm to ensure that it can effectively reflect the flow state changes in actual applications. Finally, multiple experimental tests are carried out to verify the system response effect under different thresholds, and the final threshold that can balance the system sensitivity and stability is selected to ensure that there will be no overly frequent adjustments when triggering the optimization, thereby improving the overall system efficiency.

[0144] When a decision is made to trigger further air flow path optimization, the actions include:

[0145] Based on the current flow data, the 3D air flow model is recalculated to ensure that the model can reflect the changes in the actual flow state. A finer grid division can be used to increase the calculation accuracy.

[0146] Based on the identified micro-turbulence areas and air stagnation points, the sensor layout is adjusted, the number of sensors is increased or their positions are optimized to achieve more comprehensive air flow monitoring and collect more accurate data.

[0147] Based on the new flow model, the operating parameters of the air circulation system, including wind speed, wind direction and fan speed, are recalculated to ensure that the air can flow efficiently and cover every corner.

[0148] Apply an iterative optimization algorithm (such as genetic algorithm or particle swarm optimization) to perform multiple calculations and adjustments based on real-time data to achieve continuous improvement of the flow path until the desired match threshold is reached.

[0149] Establish an evaluation and feedback mechanism to continuously monitor the effectiveness of the optimized flow path and make further adjustments based on the feedback results to ensure the stability and uniformity of air flow.

[0150] The finite element analysis method was used to construct a heat flow model to simulate the heat flow distribution under different air flow conditions, and transient thermal analysis was performed to evaluate the abnormal degree of heat flux density changes in the storage environment of fresh roxburghii fruit, specifically:

[0151] In finite element analysis, it is necessary to select a suitable physical model to describe the heat flow behavior. For the heat flow analysis of fresh roxburghii fruit, the heat conduction equation can be used: ;in, represents temperature, Represents time, is the thermal diffusivity.

[0152] After building the model, mesh generation is required. Select the appropriate mesh type (such as tetrahedral or hexahedral mesh) to ensure that the model has sufficient resolution in areas with high geometric complexity to ensure the accuracy of the analysis. The density of the mesh generation should be appropriately adjusted according to the flow characteristics to capture the thermal flow characteristics of the key areas.

[0153] When performing transient thermal analysis, set the time step and total analysis time , ensuring that the changing process of heat flux density can be accurately captured. The choice of time step should be based on the frequency of change of fruit respiration rate and heat flux characteristics in order to obtain a fine dynamic response.

[0154] The set heat flow model is solved using finite element analysis software (such as ANSYS, COMSOL, etc.). After solving, key heat flow distribution data are extracted through post-processing tools, including temperature changes at each node, heat flux density, and heat flux in a specific area.

[0155] In the post-processing stage, heat flow distribution diagrams and temperature change diagrams can be generated to facilitate the analysis of the thermal conduction characteristics of fresh sea buckthorn fruit under different air flow conditions.

[0156] After the heat flow model was constructed and transient thermal analysis was performed, the abnormal degree of heat flux density changes in the storage environment of fresh sea buckthorn fruit was evaluated.

[0157] Calculate the standard deviation of the heat flux: ;in, is the standard deviation of the heat flux, For the The heat flux density at each node is is the number of nodes, is the average heat flux density at the node.

[0158] After the heat flow analysis, collect the heat flux data of all nodes and integrate them into a heat flux dataset for subsequent analysis. The heat flux dataset should include the heat flux value of each node and the corresponding time step. .

[0159] The heat flux dataset is in the following form: ;in, is the heat flux data set.

[0160] To further evaluate the change in heat flux density, it is necessary to calculate the heat flux density change rate , It is defined as the difference in heat flux at successive time steps: .

[0161] This change rate can reflect the dynamic changes of heat flux density. If the change rate exceeds the set threshold, it may indicate an abnormal situation.

[0162] By analyzing the standard deviation and change rate of heat flux density, the evaluation value of heat flux anomaly is calculated, and its expression is: ;in, is the evaluation value of heat flow anomaly degree, is the threshold value corresponding to the rate of change of heat flux density, and They are and The weight coefficient, and and Both are greater than 0.

[0163] The greater the heat flow anomaly assessment value, the higher the abnormality of the heat flux density change in the storage environment of fresh sea buckthorn fruit, which means that the heat flux density fluctuates more and there may be more significant preservation risks.

[0164] The setting of the threshold corresponding to the rate of change of heat flux density should be based on the statistical analysis of historical data, combined with the actual volatility of the storage environment and the physiological characteristics of fresh roxburghii fruit. By calculating the standard deviation and mean of the previous monitoring data, the normal fluctuation range can be determined. Abnormal conditions can be defined as the rate of change beyond this range. Usually, a multiple of the standard deviation greater than the normal fluctuation is selected as the threshold to ensure timely and accurate identification of potential risks and the taking of corresponding measures.

[0165] By analyzing the entropy changes of the respiratory rate signal of fresh roxburghii fruit, the potential respiratory jump phenomenon of fresh roxburghii fruit was identified, specifically:

[0166] The respiration rate of fresh sea buckthorn fruit is monitored in real time, and the time series data of the respiration rate is collected through a high-precision sensor.

[0167] A series of preprocessing steps are performed on the respiratory rate signal of the respiratory rate time series data to remove noise and improve the quality of the signal. The preprocessing steps include smoothing and standardization.

[0168] After the preprocessing of the respiratory rate signal is completed, the respiratory rate signal is divided into multiple time windows for subsequent entropy calculation. The specific steps are as follows:

[0169] The respiratory rate signal is divided into multiple overlapping windows. For example, the window length is set to 30 seconds, each window moves with a fixed step length, and data is extracted from each window.

[0170] For each window, the entropy of the respiratory rate signal is calculated to evaluate the complexity of the respiratory rate signal. The entropy of the respiratory rate signal is calculated as: ;in, The entropy of the respiratory rate signal reflects the uncertainty or information content of the system; Representation characteristics The probability distribution of , which indicates the frequency of occurrence of the feature in the sample; Indicates the total number of features, which refers to the number of different features considered in the entropy calculation; Represents the feature index, from 1 to An integer that iterates over all features.

[0171] Among them, the characteristics It is a specific attribute or index extracted when analyzing the respiration rate signal of fresh roxburghii fruit. It can represent a variety of biological or physical properties, such as:

[0172] Respiration rate: refers to the amount of oxygen consumed or carbon dioxide released by the fruit per unit time, reflecting the metabolic state of the fruit.

[0173] Temperature changes: reflects the impact of storage environment on fruit respiration rate.

[0174] Humidity level: affects the evaporation and respiration process of the fruit.

[0175] Gas concentration: Changes in the concentration of oxygen or carbon dioxide affect the respiratory efficiency of the fruit.

[0176] The absolute value obtained by subtracting the entropy of the real-time respiratory rate signal from the entropy of the last monitored respiratory rate signal is marked as the respiratory rate entropy change index.

[0177] The larger the respiratory rate entropy change index, the higher the degree of potential respiratory jump phenomenon in fresh sea buckthorn fruit, which may indicate a mutation in the physiological state of the fruit, which may point to an increased risk of ripening or corruption.

[0178] The abnormal degree of heat flux density change in the storage environment of fresh roxburghii fruit and the degree of potential respiratory jump phenomenon in fresh roxburghii fruit were comprehensively analyzed to establish a physiological state prediction model for fresh roxburghii fruit, specifically:

[0179] The specific steps for establishing the physiological state prediction model of fresh roxburghii fruit are as follows:

[0180] The heat flow abnormality evaluation value and the respiratory rate entropy change index are normalized, and the normalized heat flow abnormality evaluation value and the respiratory rate entropy change index are assigned preset proportional coefficients respectively, and then the physiological prediction coefficient of fresh roxburgh fruit is calculated. The physiological state prediction model of fresh roxburgh fruit is reflected by the physiological prediction coefficient of fresh roxburgh fruit.

[0181] The specific implementation method of calculating the physiological prediction coefficient of fresh sea buckthorn fruit is not specifically limited here. Any method that can calculate the physiological prediction coefficient of fresh sea buckthorn fruit by assigning the normalized heat flow abnormality evaluation value and the respiratory rate entropy change index to preset proportional coefficients respectively is acceptable. In order to implement the technical solution of the present invention, the present invention provides a specific method for calculating the physiological prediction coefficient of fresh sea buckthorn fruit.

[0182] The expression of the physiological prediction coefficient of fresh fruit of Rosa roxburghii is: ;in, is the physiological prediction coefficient of fresh fruit of roxburghii, is the normalized heat flow anomaly evaluation value, is the normalized entropy change index of respiratory rate, and are respectively the preset proportional coefficients of the normalized heat flow abnormality evaluation value and the normalized respiratory rate entropy change index, satisfying ,and and Both are greater than 0.

[0183] A higher physiological prediction coefficient of fresh seabuckthorn fruit usually indicates that the physiological state of fresh seabuckthorn fruit has undergone more significant changes, which may indicate that the fruit is close to maturity or has an increased risk of corruption. It reflects the combined influence of heat flux density fluctuations and respiratory jump phenomena in the storage environment, which means that the current preservation environment may have greater fluctuations and instability. A higher physiological prediction coefficient of fresh seabuckthorn fruit suggests that the storage environment needs to be adjusted in a timely manner to maintain suitable temperature, humidity and gas composition, thereby extending the shelf life of fresh seabuckthorn fruit.

[0184] Based on the output of the physiological state prediction model of fresh roxburghii fruit, the control algorithm is used to adjust the temperature, humidity and gas composition in the storage environment, specifically:

[0185] According to the established prediction model for the physiological state of fresh roxburghii fruit, the output value of the current physiological prediction coefficient of fresh roxburghii fruit is obtained in real time.

[0186] Based on the physiological prediction coefficient of fresh roxburghii fruit, the corresponding control targets are set. The control targets are determined based on literature and experimental data to ensure that the shelf life of fresh roxburghii fruit can be effectively extended. The output of the physiological state prediction model of fresh roxburghii fruit is used to dynamically adjust the following targets:

[0187] Temperature target: Adjust according to the physiological prediction coefficient of fresh sea buckthorn fruit. If the physiological prediction coefficient of fresh sea buckthorn fruit is high, lower the temperature target to slow down the respiration rate.

[0188] Humidity target: Humidity is also dynamically adjusted according to the value of the physiological prediction coefficient of fresh sea buckthorn fruit to prevent the fruit from drying out.

[0189] Gas composition target: Adjust the concentration of oxygen and carbon dioxide to optimize the storage conditions of the fruit, especially to increase the concentration of carbon dioxide and inhibit respiration when the physiological prediction coefficient of fresh sea buckthorn fruit is high.

[0190] Select the appropriate control algorithm for efficient regulation:

[0191] PID control algorithm: Dynamically adjust PID parameters according to changes in prediction coefficients to improve control flexibility.

[0192] Adaptive control algorithm: Combined with the physiological state prediction model, the control strategy is adjusted in real time to adapt to environmental changes.

[0193] Execute the control algorithm, the specific steps include:

[0194] Calculate adjustment amount: Set the current environmental parameters, including temperature, humidity and gas composition. Calculate the difference between the target and current values ​​to get the adjustment amount.

[0195] To perform adjustments:

[0196] Temperature adjustment: If the adjustment amount is positive, the heater is started; if the adjustment amount is negative, the cooler is started.

[0197] Humidity adjustment: If the adjustment amount is positive, the humidifier is started; if the adjustment amount is negative, the dehumidifier is started.

[0198] Gas composition adjustment: According to the gas sensor data, the gas concentration is adjusted to achieve the target by increasing or decreasing the proportion of oxygen or carbon dioxide.

[0199] Establish a real-time monitoring and feedback mechanism to ensure the effectiveness of environmental control and reflect the important role of the physiological state prediction model:

[0200] Continuously monitor the temperature, humidity and gas composition in the storage environment.

[0201] Compare the prediction coefficients output by the model with environmental parameters in real time, and use the control algorithm to automatically adjust the environmental parameters.

[0202] If the monitored environmental parameters exceed the preset range, an alarm will be issued immediately and corresponding adjustments will be made. At the same time, based on the feedback information of the physiological state model, the future control strategy will be adjusted.

[0203] Example 2

[0204] The difference between Example 2 of the present invention and Example 1 is that this example introduces an intelligent control system for the preservation environment of fresh roxburghii fruit based on the Internet of Things.

[0205] Figure 2A structural schematic diagram of an intelligent control system for the preservation environment of fresh roxburghii fruit based on the Internet of Things of the present invention is given. The intelligent control system for the preservation environment of fresh roxburghii fruit based on the Internet of Things comprises a flow model building module, a real-time data acquisition module, an operation parameter optimization module, a heat flow respiration analysis module, a physiological state prediction module and an environmental parameter control module.

[0206] Flow model construction module: Using computational fluid dynamics algorithms, a three-dimensional air flow model of the storage environment is constructed based on the interference of the thorny structure on the surface of fresh sea buckthorn fruit on air flow.

[0207] Real-time data collection module: Deploy an IoT sensor network in the storage environment to collect air flow data in real time and identify micro-turbulence areas and air stagnation points.

[0208] Operation parameter optimization module: Based on the collected real-time air flow data and three-dimensional air flow model, the machine learning algorithm is used to adjust the operating parameters of the air circulation system to determine whether to trigger the air flow path to optimize the storage environment.

[0209] Thermal flow respiration analysis module: When the air flow path optimization of the storage environment is triggered: the finite element analysis method is used to construct a thermal flow model to simulate the heat flow distribution under different air flow conditions, and transient thermal analysis is performed to evaluate the abnormal degree of the heat flux density change in the storage environment of the sea buckthorn fresh fruit; by analyzing the entropy changes of the respiration rate signal of the sea buckthorn fresh fruit, the degree of potential respiratory jump phenomenon in the sea buckthorn fresh fruit is identified.

[0210] Physiological state prediction module: The abnormal degree of heat flux density change in the storage environment of fresh seabuckthorn fruit and the degree of potential respiratory jump phenomenon in fresh seabuckthorn fruit are comprehensively analyzed to establish a physiological state prediction model for fresh seabuckthorn fruit.

[0211] Environmental parameter control module: Based on the output of the physiological state prediction model of fresh sea buckthorn fruit, the control algorithm is used to adjust the temperature, humidity and gas composition in the storage environment.

[0212] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0213] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or may be transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium may be a solid-state hard disk.

[0214] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0215] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0216] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0217] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0218] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0219] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0220] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0221] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent control method for the preservation environment of fresh roxburghii fruit based on the Internet of Things, characterized in that: The steps include: S1: Using computational fluid dynamics algorithms, a three-dimensional air flow model of the storage environment was constructed based on the interference of the thorny structure on the surface of fresh roxburghii fruit on air flow; S2: Deploy an IoT sensor network in the storage environment to collect air flow data in real time and identify micro-turbulence areas and air stagnation points; S3: Based on the collected real-time air flow data and three-dimensional air flow model, the machine learning algorithm is used to adjust the operating parameters of the air circulation system to determine whether to trigger the air flow path that optimizes the storage environment; S4: When the air flow path of the storage environment is optimized: a heat flow model is constructed using the finite element analysis method to simulate the heat flow distribution under different air flow conditions, and a transient thermal analysis is performed to evaluate the abnormal degree of the heat flux density change in the storage environment of the fresh roxburghii fruit; by analyzing the entropy change of the respiration rate signal of the fresh roxburghii fruit, the degree of potential respiratory jump phenomenon in the fresh roxburghii fruit is identified; S5: Comprehensively analyze the abnormal degree of heat flux density change in the storage environment of fresh roxburghii fruit and the degree of potential respiratory jump phenomenon in fresh roxburghii fruit, and establish a prediction model for the physiological state of fresh roxburghii fruit; S6: Based on the output of the physiological state prediction model of fresh roxburghii fruit, the control algorithm is used to adjust the temperature, humidity and gas composition in the storage environment; S2 specifically includes: S201: Select an air flow sensor and reasonably arrange the sensor network according to the characteristics of the three-dimensional air flow model; S202: Setting the sampling frequency and data storage strategy of the sensor to collect air speed and direction data in real time; S203: The collected data is transmitted in real time through the Internet of Things, and noise is removed by using data filtering and preprocessing algorithms; S204: Based on real-time air flow data, use fluid dynamics analysis methods to identify micro-turbulence areas and air stagnation points: Based on real-time air flow data, fluid dynamics analysis methods are used to identify micro-turbulence areas and air stagnation points; a velocity distribution model of the air flow field is established to calculate the local turbulent kinetic energy of the fluid: ;in, is the local turbulent kinetic energy of the fluid, represents the velocity component at a specific point, is the average velocity component around that particular point, is the index, and the specific point is each computational grid point in the air flow field; A preset threshold value of the local turbulent kinetic energy of the fluid is set, and when the local turbulent kinetic energy of the fluid is greater than the preset threshold value of the local turbulent kinetic energy of the fluid, it is judged to be a micro-turbulent area; By analyzing the separation of the fluid in the velocity field, air stagnation points are identified.

2. The method for intelligently controlling the fresh-keeping environment of fresh roxburghii fruit based on the Internet of Things according to claim 1, characterized in that: S1 specifically includes: S101: According to the characteristics of the thorn-like structure on the surface of fresh roxburghii fruit, determine the key interference factors and main influencing areas of air flow; the main influencing areas include the area adjacent to the fruit surface, the air flow channel around the thorn-like structure, and the air flow outlet area; S102: Select computational fluid dynamics equations applicable to small-scale flows, the computational fluid dynamics equations including the Navier-Stokes equations; S103: setting boundary conditions of the three-dimensional model according to the storage environment and the fruit structure, including the air inlet, the air outlet, and the fruit surface; S104: Adopting an iterative numerical solution method and combining a meshing technique, a three-dimensional air flow model of the storage environment is constructed, and air flow path data is output.

3. The method for intelligently controlling the fresh-keeping environment of roxburghii fresh fruit based on the Internet of Things according to claim 2, characterized in that: S3 specifically includes: S301: Fusing the real-time collected air flow data with the three-dimensional air flow model to construct a comprehensive flow characteristic model: Standardize the air flow data to obtain the standardized air flow data ; Use the spatial interpolation method to map the real-time data to the model grid nodes, and set the interpolation function to ,but: ;in, is the value of the mapped flow velocity in the three-dimensional grid, is the interpolation function, is a spatial variable in a three-dimensional coordinate system; Fill missing data through interpolation methods; S302: Extract key features from the comprehensive flow characteristic model, including turbulent kinetic energy and flow velocity distribution: Identify important parameters that affect the operation of air circulation systems; Arrange the extracted features into a matrix form; S303: Select a machine learning algorithm to train a model based on historical data to learn optimal operating parameters of the air circulation system; S304: Using the trained model, calculate and adjust the operating parameters of the air circulation system in real time: Input the newly collected feature data into the machine learning model; adjust the operating parameters of the air circulation system in real time according to the model output; S305: Determine whether the current air flow state matches the optimal parameters output by the model, and decide whether to trigger further air flow path optimization: Compare the difference between the current flow state and the model output parameters and calculate the matching index; When the matching index is lower than its corresponding preset threshold, it is determined to trigger further air flow path optimization.

4. The method for intelligently controlling the fresh-keeping environment of roxburghii fresh fruit based on the Internet of Things according to claim 3, characterized in that: The finite element analysis method was used to construct a heat flow model to simulate the heat flow distribution under different air flow conditions, and transient thermal analysis was performed to evaluate the abnormal degree of heat flux density changes in the storage environment of fresh roxburghii fruit, specifically: The heat conduction equation is used to describe the heat flow behavior: ;in, represents temperature, Represents time, is the thermal diffusivity; Perform mesh division; Set the time step and total analysis time; The set heat flow model is solved using finite element analysis software; after the solution, key heat flow distribution data is extracted through post-processing tools; Calculate the standard deviation of the heat flux: ;in, is the standard deviation of the heat flux, For the The heat flux density at each node is is the number of nodes, is the average value of the node heat flux density; Collect the heat flux data of all nodes and integrate them into a heat flux data set, which includes the heat flux value of each node and the corresponding time step; Calculate the rate of change of heat flux density, which is defined as the difference in heat flux density at consecutive time steps; By analyzing the standard deviation and change rate of heat flux density, the evaluation value of heat flux anomaly is calculated, and its expression is: ;in, is the evaluation value of heat flow anomaly degree, is the threshold value corresponding to the rate of change of heat flux density, is the rate of change of heat flux, and They are and The weight coefficient, and and Both are greater than 0.

5. The method for intelligently controlling the fresh-keeping environment of fresh roxburghii fruit based on the Internet of Things according to claim 4, characterized in that: By analyzing the entropy changes of the respiratory rate signal of fresh roxburghii fruit, the potential respiratory jump phenomenon of fresh roxburghii fruit was identified, specifically: Real-time monitoring of the respiration rate of fresh roxburghii fruit, and collection of time series data of the respiration rate; Preprocessing the respiratory rate signal of the respiratory rate time series data, the preprocessing steps including smoothing and standardization; The respiratory rate signal is divided into multiple overlapping windows, each of which moves with a fixed step size; For each window, calculate the entropy of the respiration rate signal: ;in, represents the entropy of the respiratory rate signal, Representation characteristics The probability distribution of represents the total number of features, Represents the feature index; Among them, the characteristics including breathing rate, temperature changes, humidity levels, and gas concentrations; The absolute value obtained by subtracting the entropy of the real-time respiratory rate signal from the entropy of the last monitored respiratory rate signal is marked as the respiratory rate entropy change index.

6. The method for intelligently controlling the fresh-keeping environment of roxburghii fresh fruit based on the Internet of Things according to claim 5, characterized in that: The abnormal degree of heat flux density change in the storage environment of fresh roxburghii fruit and the degree of potential respiratory jump phenomenon in fresh roxburghii fruit were comprehensively analyzed to establish a physiological state prediction model for fresh roxburghii fruit, specifically: The heat flow abnormality evaluation value and the respiratory rate entropy change index are normalized, and the normalized heat flow abnormality evaluation value and the respiratory rate entropy change index are respectively assigned preset proportional coefficients to calculate the physiological prediction coefficient of fresh sea buckthorn fruit.

7. The method for intelligently controlling the fresh-keeping environment of roxburghii fresh fruit based on the Internet of Things according to claim 6, characterized in that: Based on the output of the physiological state prediction model of fresh roxburghii fruit, the control algorithm is used to adjust the temperature, humidity and gas composition in the storage environment, specifically: Real-time acquisition of the output value of the current physiological prediction coefficient of fresh roxburghii fruit; Based on the physiological prediction coefficient of fresh roxburghii fruit, the corresponding control target is set, and the output of the physiological state prediction model of fresh roxburghii fruit is used to dynamically adjust the temperature target, humidity target and gas composition target; Execute the control algorithm, the specific steps include: Set current environmental parameters, including temperature, humidity, and gas composition; calculate the difference between the target and current values ​​to obtain the adjustment amount; Temperature adjustment: If the adjustment amount is positive, the heater is started; if the adjustment amount is negative, the cooler is started; Humidity adjustment: If the adjustment amount is positive, the humidifier is started; if the adjustment amount is negative, the dehumidifier is started; Gas composition adjustment: Adjust the gas concentration according to the gas sensor data.

8. An intelligent control system for the preservation environment of fresh roxburghii fruit based on the Internet of Things, used to implement the intelligent control method for the preservation environment of fresh roxburghii fruit based on the Internet of Things according to any one of claims 1 to 7, characterized in that: It includes flow model building module, real-time data acquisition module, operation parameter optimization module, thermal flow and respiration analysis module, physiological state prediction module and environmental parameter control module; Flow model building module: Using computational fluid dynamics algorithms, based on the interference of the thorny structure on the surface of fresh roxburghii fruit on air flow, a three-dimensional air flow model of the storage environment is constructed; Real-time data collection module: deploy an IoT sensor network in the storage environment to collect air flow data in real time and identify micro-turbulence areas and air stagnation points; Operation parameter optimization module: Based on the collected real-time air flow data and three-dimensional air flow model, the machine learning algorithm is used to adjust the operating parameters of the air circulation system to determine whether to trigger the air flow path to optimize the storage environment; Thermal flow respiration analysis module: When the air flow path of the storage environment is optimized, the finite element analysis method is used to construct a thermal flow model to simulate the heat flow distribution under different air flow conditions, and transient thermal analysis is performed to evaluate the abnormal degree of heat flux density changes in the storage environment of the fresh roxburghii fruit; by analyzing the entropy changes of the respiration rate signal of the fresh roxburghii fruit, the degree of potential respiratory jump phenomenon in the fresh roxburghii fruit is identified; Physiological state prediction module: Comprehensively analyze the abnormal degree of heat flux density change in the storage environment of fresh roxburghii fruit and the degree of potential respiratory jump phenomenon in fresh roxburghii fruit, and establish a physiological state prediction model for fresh roxburghii fruit; Environmental parameter control module: Based on the output of the physiological state prediction model of fresh sea buckthorn fruit, the control algorithm is used to adjust the temperature, humidity and gas composition in the storage environment.

Images