A plastic fruit cup quality traceability system based on the internet of things

Abstract

The application relates to the technical field of Internet of Things and fresh agricultural product quality tracing, in particular to a plastic fruit cup quality tracing system based on the Internet of Things. The system comprises a multi-modal data acquisition unit configured to acquire microscopic physical stress correlation data, mesoscopic interface environment data, macroscopic biochemical state data and implicit environmental factor data in real time; a key parameter calculation unit configured to calculate an interface biochemical amplification coefficient and vibration energy and illumination energy; a positive feedback avalanche risk assessment model unit configured to solve normalized cumulative released stress and normalized ethylene release flux and generate an avalanche risk index; and an early warning and decision support unit configured to output a risk early warning based on the avalanche risk index. The system deeply reveals and quantifies the positive feedback avalanche mechanism between physical stress and biochemical reaction, and overcomes the limitation that traditional macroscopic parameter monitoring cannot predict sudden and group corruption.

Description

Technical Field

[0001] This invention relates to the fields of Internet of Things (IoT) technology and fresh agricultural product quality traceability technology, specifically to an IoT-based plastic fruit tray quality traceability system. Background Technology

[0002] In the cold chain logistics of fresh products, especially fruits, ensuring their quality and safety is of paramount importance. To monitor the condition of goods, the industry generally relies on tracking macro-environmental parameters, such as changes in temperature and humidity. These monitoring links form the basis of quality traceability, aiming to slow down the rate of natural product deterioration by maintaining a stable external environment. However, traditional quality control methods mostly rely on linear, time-integral-based evaluation models, which have limitations in predicting and mitigating sudden, large-scale group corruption risks and can easily lead to significant economic losses.

[0003] While existing technologies have made monitoring of macroscopic parameters such as temperature and humidity relatively mature, these methods generally fail to delve into the level of multi-physics coupling effects. Existing technologies neglect the nonlinear coupling effects of microscopic physical stress, mesoscopic interface environment, and implicit environmental factors such as vibration and light in transportation and storage. Therefore, when these factors interact and form a positive feedback loop, traditional methods cannot capture the avalanche-like transformation of the system from slow deterioration to exponentially accelerated decay, resulting in significant deficiencies in early warning and risk intervention.

[0004] Therefore, how to establish a technical method that can accurately monitor and proactively intervene in the risk of mass corruption caused by the positive feedback avalanche mechanism of biochemical and physical stress is a problem that urgently needs to be solved by those skilled in the art.

[0005] The information disclosed in the background section above is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention discloses a plastic fruit tray quality traceability system based on the Internet of Things. Specifically, the technical solution of this invention includes:

[0007] A multimodal data acquisition unit is configured to acquire microscopic physical stress correlation data, mesoscopic interface environment data, macroscopic biochemical state data, and implicit environmental factor data in real time.

[0008] The key parameter calculation unit is configured to calculate the interface biochemical amplification factor based on the mesoscopic interface environment data; and to calculate vibration energy and light energy based on the latent environmental factor data.

[0009] The positive feedback avalanche risk assessment model unit is configured to combine the interface biochemical amplification factor, the vibration energy, the light energy, and the macroscopic biochemical state data to calculate the normalized cumulative release stress and the normalized ethylene release flux; and to generate an avalanche risk index based on the normalized cumulative release stress and the normalized ethylene release flux.

[0010] An early warning and decision support unit is configured to output a risk warning based on the avalanche risk index. Preferably, the multimodal data acquisition unit specifically includes:

[0011] Microscopic physical stress correlation data acquisition module, configured to collect stress data at stress concentration points of plastic fruit holder structure;

[0012] Mesoscopic interface environment data acquisition module, configured to collect temperature and humidity data near the interface between the fruit tray and the fruit;

[0013] Macroscopic biochemical state data acquisition module, configured to collect ethylene concentration data of the gas environment inside the packaging box;

[0014] The latent environmental factor acquisition module is configured to collect vibration spectrum data of transport vehicles and spectral energy distribution data of warehouse lighting.

[0015] Preferably, the key parameter calculation unit is specifically configured for:

[0016] The vibration spectrum data acquired by the latent environmental factor acquisition module is integrated within a preset dangerous frequency band to determine the vibration energy;

[0017] The spectral energy distribution data acquired by the latent environmental factor acquisition module is integrated in the high-risk short-wavelength band to determine the illumination energy.

[0018] Preferably, the key parameter calculation unit is specifically configured for:

[0019] Based on the temperature and humidity data acquired by the mesoscopic interface environment data acquisition module, the microcondensation phase transition period is calculated using a peak-valley detection algorithm.

[0020] By using a preset inverse proportional function relationship, the microcondensation phase transition period is mapped to the interface biochemical amplification coefficient.

[0021] Preferably, the positive feedback avalanche risk assessment model unit is specifically configured for:

[0022] The growth rate of the normalized cumulative released stress is determined to be driven by the basic thermal relaxation rate, the vibration energy, the light energy, and the normalized ethylene release flux.

[0023] Preferably, the positive feedback avalanche risk assessment model unit is specifically configured for:

[0024] The growth rate of the normalized ethylene release flux is determined to be driven by both the instantaneous rate of the normalized cumulative release stress and the interfacial biochemical amplification factor.

[0025] Preferably, the positive feedback avalanche risk assessment model unit generates the avalanche risk index by calculating the sum of the squares of the normalized cumulative released stress and the normalized ethylene release flux.

[0026] Preferably, the early warning and decision support unit is specifically configured to:

[0027] The growth rate of the avalanche risk index is compared with a preset warning threshold;

[0028] Wherein, when the growth rate exceeds the warning threshold, the risk warning is a warning message and intervention suggestion.

[0029] Preferably, the early warning and decision support unit is specifically configured to:

[0030] Compare the current value of the avalanche risk index with the pre-treatment threshold;

[0031] Wherein, when the current value exceeds the disposal threshold, the risk warning is a disposal alert and a business action recommendation.

[0032] Preferably, the early warning and decision support unit is further configured to:

[0033] Fault attribution is performed based on the contribution of each driving term to the normalized cumulative released stress growth rate to identify the main causes of increased risk.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] 1. This system profoundly reveals and quantifies the positive feedback avalanche mechanism between physical stress and biochemical reactions, overcoming the limitations of traditional macroscopic parameter monitoring in predicting sudden and widespread spoilage. By constructing a coupled dynamic model and integrating multi-dimensional data such as microscopic stress, mesoscopic interfaces, macroscopic biochemical processes, and latent environmental factors, quality traceability is elevated from post-event recording to precise pre-event prediction. This method can capture the transformation of the system from linear degradation to nonlinear accelerated spoilage at the mechanistic level, greatly improving the accuracy and reliability of predicting large-scale cargo damage events.

[0036] 2. This system innovatively incorporates implicit environmental factors such as transportation vibration and warehouse lighting into its monitoring. By integrating the vibration spectrum of specific hazardous frequency bands and the spectral energy of high-risk shortwave bands, it accurately quantifies their energy contribution to cargo damage. More importantly, after an early warning is triggered, the system can attribute the fault to the contribution of each driving factor to the risk increase, clarifying whether the dominant factor of the increased risk stems from vibration, lighting, or internal biochemical processes. This provides managers with intervention suggestions that have clear physical orientation, achieving a leap from risk warning to root cause location.

[0037] 3. This system constructs a dual-threshold, multi-level early warning decision-making system based on both the risk index growth rate and absolute value. By monitoring the risk index growth rate, the system can issue early warnings before significant quality deterioration occurs, buying valuable time for preventative intervention. Conversely, by monitoring the absolute value of the risk index, the system can issue alerts indicating that goods are nearing irreversible large-scale spoilage. This tiered response mechanism can output alerts and business recommendations of varying urgency levels according to different stages of risk evolution, achieving refined and intelligent risk management.

[0038] 4. This system achieves dynamic quantification of the mesoscale microenvironment by deploying high-frequency temperature and humidity sensors at the contact interface between the fruit tray and the fruit. The system employs a peak-valley detection algorithm to calculate the micro-condensation phase transition period and maps it to the interface biochemical amplification coefficient through an inverse proportional function, thereby parameterizing the catalytic effect of interface micro-condensation on biochemical reactions. This design not only provides key adjustment parameters for accurate prediction but also reveals the intrinsic mechanism of anomalous phenomena such as preferential decay of the underlying fruit from a physicochemical perspective, significantly enhancing the model's explanatory power and prediction accuracy. Attached Figure Description

[0039] The present invention will be further explained below with reference to the accompanying drawings and embodiments:

[0040] Figure 1 This is a system structure diagram of the present invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0042] Example 1:

[0043] Please see Figure 1 A quality traceability system for plastic fruit trays based on the Internet of Things includes:

[0044] A multimodal data acquisition unit is configured to acquire microscopic physical stress correlation data, mesoscopic interface environment data, macroscopic biochemical state data, and implicit environmental factor data in real time.

[0045] The key parameter calculation unit is configured to calculate the interface biochemical amplification factor based on mesoscopic interface environmental data; and to calculate vibration energy and light energy based on latent environmental factor data.

[0046] The positive feedback avalanche risk assessment model unit is configured to combine interface biochemical amplification factor, vibration energy, light energy and macroscopic biochemical state data to calculate the normalized cumulative release stress and normalized ethylene release flux; and generate an avalanche risk index based on the normalized cumulative release stress and normalized ethylene release flux.

[0047] The early warning and decision support unit is configured to output risk warnings based on the avalanche risk index.

[0048] This embodiment discloses an IoT-based plastic fruit tray quality traceability system. The core purpose of this system is to solve the technical problem that existing technologies rely solely on macroscopic environmental parameters and cannot predict or avoid the risk of mass spoilage of fresh products, especially fruits, caused by the nonlinear coupling of multiple factors. This system, deployed within the packaging unit, constructs a complete technical closed loop from multi-dimensional data collection, key parameter extraction, nonlinear dynamic modeling to risk warning and decision support, thereby achieving precise monitoring and proactive intervention of the core degradation mechanism of biochemical-physical stress positive feedback avalanche.

[0049] The system includes:

[0050] The multimodal data acquisition unit aims to comprehensively and in real-time capture multi-source heterogeneous data directly related to the mechanism of group corruption, providing basic data support for subsequent risk assessment. In this embodiment, the unit is deployed inside the cargo packaging box and integrates a series of sensors. It not only monitors traditional macroscopic environmental parameters but also innovatively delves into the microscopic and mesoscopic levels, incorporating latent environmental factors into the monitoring scope. The data dimensions it collects include: microscopic physical stress correlation data, mesoscopic interface environmental data, macroscopic biochemical state data, and latent environmental factor data.

[0051] The purpose of the key parameter calculation unit is to process the raw, high-dimensional sensor data stream collected by the preceding unit in real time and transform it into key parameters with clear physical and biochemical significance that can be directly input into the core risk assessment model. In this embodiment, this unit acts as a data processing intermediate layer and is responsible for performing specific physical and mathematical operations to extract the raw data into interface biochemical amplification coefficient, vibration energy and light energy that can characterize the core dynamics of the system.

[0052] The positive feedback avalanche risk assessment model unit, which is the core technology of this invention, aims to quantitatively describe and predict the systemic corruption risk driven by the positive feedback loop between physical stress and biochemical reactions through a nonlinear dynamic model. In this embodiment, the unit embeds a set of coupled differential equations to simulate the dynamic evolution of two core state variables—normalized cumulative release stress and normalized ethylene release flux. This model can capture the avalanche-like transformation of the system from linear, slow deterioration to nonlinear, exponentially accelerated corruption. It receives real-time parameters provided by the key parameter calculation unit and, combined with macroscopic biochemical state data, continuously calculates the current values ​​of the aforementioned two core state variables. Subsequently, based on these two state variables, the unit further generates a comprehensive avalanche risk index, which constitutes the direct basis for risk judgment.

[0053] The purpose of the early warning and decision support unit is to transform the risk index output by the model into early warning information and intervention suggestions that have clear guiding significance for logistics managers. In this embodiment, the unit acts as an intelligent decision engine, which monitors the avalanche risk index and its rate of change in real time and compares it with a preset threshold that has been calibrated through a large amount of experimental data and simulation verification.

[0054] Through the collaborative work of the above four units, a new technical framework has been constructed that can understand, quantify and predict the risk of mass spoilage of fruit at the mechanistic level. It goes beyond the traditional linear model that relies on temperature-time integrals, reveals and utilizes the biochemical-physical stress positive feedback avalanche mechanism, and can accurately predict sudden and large-scale cargo damage events. It provides a powerful technical means for achieving proactive and refined risk management in cold chain logistics, thereby reducing economic losses and improving the overall resilience of the supply chain.

[0055] It should be noted that this system focuses on addressing the risk of sudden and widespread spoilage caused by the specific mechanism of biochemical-physical stress positive feedback avalanche. It is intended to supplement and deepen traditional temperature and humidity monitoring and is applicable to cargo damage scenarios dominated by latent factors in the transportation environment. For spoilage patterns dominated by other factors such as microbial contamination or initial physical damage, it is necessary to combine other detection methods for comprehensive evaluation.

[0056] Example 2:

[0057] The multimodal data acquisition unit specifically includes:

[0058] Microscopic physical stress correlation data acquisition module, configured to collect stress data at stress concentration points of plastic fruit holder structure;

[0059] Mesoscopic interface environment data acquisition module, configured to collect temperature and humidity data near the interface between the fruit tray and the fruit;

[0060] Macroscopic biochemical state data acquisition module, configured to collect ethylene concentration data of the gas environment inside the packaging box;

[0061] The latent environmental factor acquisition module is configured to collect vibration spectrum data of transport vehicles and spectral energy distribution data of warehouse lighting.

[0062] Based on Example 1, this embodiment optimizes and limits the internal structure of the multimodal data acquisition unit. The purpose of this specific design is to ensure that the acquired data can most directly and comprehensively reflect the key aspects of the biochemical-physical stress positive feedback avalanche mechanism.

[0063] Specifically, the multimodal data acquisition unit includes:

[0064] The purpose of the microscopic physical stress correlation data acquisition module is to capture the changes in microscopic structural stress caused by external vibration and internal chemical erosion of the plastic fruit holder. The microscopic physical stress correlation data refers to the real-time state data of the internal stress of the fruit holder material. Its function is to serve as a direct measure of the accumulation and release of physical stress. It is obtained by sensors integrated at the stress concentration points of the fruit holder structure, such as edges and partitions. In this embodiment, this module is preferably a piezoelectric thin film sensor or a high-frequency accelerometer.

[0065] The purpose of the mesoscopic interface environment data acquisition module is to monitor the microenvironment dynamics of the key area where the fruit tray and fruit contact each other. Mesoscopic interface environment data refers to environmental parameters within a few millimeters of the fruit-vegetable tray interface. Its function is to provide a basis for calculating the catalytic effect of microcondensation on biochemical reactions. It is a sensor array deployed near the fruit tray and fruit contact interface. In this embodiment, the module consists of a high-precision, high-temporal-resolution temperature and humidity sensor array.

[0066] The macroscopic biochemical state data acquisition module aims to track the concentration of biochemical messengers, which are key indicators of fruit maturity. Macroscopic biochemical state data refers to the concentration of specific chemical substances in the gaseous environment inside the packaging box. Its function is to provide core biochemical state variables for the evaluation model. It is a gas sensor deployed inside the packaging box. In this embodiment, the module is specifically an electrochemical or photoacoustic spectroscopy ethylene sensor, which is used to monitor the concentration of ethylene gas, which is a messenger of fruit maturity, inside the packaging box in real time.

[0067] The latent environmental factor acquisition module aims to quantify two external energy inputs that are ignored by traditional traceability systems but are crucial for triggering positive feedback avalanches. Latent environmental factor data refers to vibration and light energy data in specific frequency bands in the transportation and storage environment. Its function is to serve as an external physical forcing term that drives stress release, and it is a specialized spectrum analysis device. In this embodiment, the module consists of a vibration spectrum analyzer and a spectral energy distribution sensor.

[0068] Through this four-in-one modular design, the multimodal data acquisition unit can simultaneously capture physical, chemical, and environmental data that trigger mass corruption with unprecedented precision and dimensionality. This multi-scale, cross-domain data acquisition scheme ensures that the input to the subsequent risk assessment model is comprehensive and directly addresses the core of the problem, thereby greatly improving the accuracy and reliability of the entire system's predictions.

[0069] Example 3:

[0070] The key parameter calculation unit is specifically configured for:

[0071] The vibration spectrum data acquired by the latent environmental factor acquisition module is integrated within a preset dangerous frequency band to determine the vibration energy;

[0072] The spectral energy distribution data acquired by the latent environmental factor acquisition module is integrated in the high-risk short-wavelength band to determine the light energy;

[0073] The key parameter calculation unit is specifically configured for:

[0074] Based on the temperature and humidity data acquired by the mesoscopic interface environment data acquisition module, the microcondensation phase transition period is calculated using the peak-valley detection algorithm.

[0075] By pre-setting an inverse proportional function relationship, the micro-condensation phase change period is mapped to the interface biochemical amplification factor;

[0076] Based on Example 1, this embodiment elaborates on the specific calculation logic of the key parameter calculation unit. Its design purpose is to transform the collected multi-source heterogeneous raw data into standardized, low-dimensional key parameters that can directly drive the core dynamic model through an algorithm with clear physical meaning.

[0077] Calculation of energy parameters for latent environmental factors:

[0078] The key parameter calculation unit receives raw spectral data from the latent environmental factor acquisition module and performs integral calculations to quantify its energy contribution.

[0079] Vibrational energy Calculation: To quantify the accelerating effect of vibration at a specific frequency on the microstructure fatigue and residual stress release of the plastic fruit holder, this embodiment introduces vibration energy. ;

[0080] At time t, the total vibration energy of the transport vehicle within a preset hazardous frequency band, such as 15Hz to 25Hz, is calculated using the following formula:

[0081]

[0082] in, Instantaneous vibrational energy, with the dimension of energy E;

[0083] Vibration power spectral density function, with dimensions E / Hz, is acquired and output in real time by the vibration spectrum analyzer in the multimodal data acquisition unit;

[0084] Dangerous frequency band, with the dimension of frequency, is determined by dynamic mechanical analysis experiments on specific plastic fruit holder materials to identify the frequency range in which resonance and energy absorption are most likely to occur.

[0085] This calculation focuses broad vibration data into a highly destructive narrow band, transforming the complex spectrum into a single, dynamically changing energy parameter through integration. This improves the model's ability to capture the key physical mechanism of resonance-induced damage;

[0086] Light energy Calculation: To quantify the catalytic effect of specific spectral illumination on polymer photodegradation and fruit surface biochemical reactions, this embodiment introduces light energy. ;

[0087] At time t, the total light energy emitted by the storage or transportation environment in the high-risk short-wavelength band (e.g., the blue light band from 400nm to 480nm); its calculation formula is:

[0088]

[0089] in, Instantaneous light energy, measured in energy E;

[0090] : Spectral irradiance function, with dimensions E / nm, is acquired and output in real time by the spectral energy distribution sensor in the multimodal data acquisition unit;

[0091] High-risk short-wavelength band, with wavelength as its dimension, is determined based on existing photochemical research, which has shown that high-energy photons in this band have the strongest catalytic effect on polymer chain breakage and specific biochemical reactions.

[0092] This calculation precisely quantifies complex spectral distribution data into the energy value of the high-energy blue light component that contributes the most to system degradation. This allows the model to directly assess the hidden risks caused by improper lighting;

[0093] Calculation of the biochemical amplification factor for mesoscopic interfaces:

[0094] The key parameter calculation unit also calculates the interface biochemical amplification factor based on the temperature and humidity data acquired by the mesoscopic interface environment data acquisition module. ;

[0095] Interfacial biochemical amplification factor ( ) refers to a dimensionless coefficient used to quantify the amplification effect of micro-condensation at the fruit and vegetable-tray interface on the rate of biochemical reactions. Its function is to correct the rate of biochemical reactions in the core model. It is calculated based on the interface temperature and humidity data through a specific algorithm.

[0096] To achieve the above objectives, this unit uses a peak-valley detection algorithm on high-frequency temperature and humidity data to identify the period of temperature fluctuations, and then calculates the micro-condensation phase transition period based on this. The peak-valley detection algorithm specifically involves: smoothing the high-frequency humidity data acquired by the mesoscopic interface environment data acquisition module using a sliding window to eliminate noise; identifying all local maxima (peaks) and local minima (valleys) within each time window; defining the time span of a complete valley-peak-valley process as a micro-condensation phase transition cycle; and obtaining the instantaneous micro-condensation phase transition cycle by averaging multiple consecutive cycles. ;

[0097] Based on the above results, the period is mapped to an amplification factor using a pre-defined inverse proportional function relationship with a saturation effect; the calculation formula is as follows:

[0098]

[0099] Among them, the interfacial biochemical amplification factor was clarified. Phase change period with micro-condensation The inverse relationship between them is ensured, and the amplification factor approaches 1 when the period is extremely large, that is, there is no amplification effect;

[0100] : Interface biochemical amplification factor, dimensionless;

[0101] The microcondensation phase transition period, measured in units of time T, is calculated by this unit using a peak-valley detection algorithm based on temperature and humidity data acquired by the mesoscopic interface environment data acquisition module.

[0102] The characteristic time amplification constant, measured in units of time T, is determined through accelerated aging experiments conducted under controlled laboratory conditions, calibrated for a specific fruit-calyx combination. Specifically, it is obtained by setting multiple sets of different constant microcondensation phase change cycle experimental values. And measure the corresponding biochemical reaction acceleration rate. A set of calibration data was obtained. Then fit using the least squares method Relationship determination value;

[0103] This calculation quantifies the dynamics of the mesoscopic interface into a key regulatory parameter that directly affects the rate of macroscopic biochemical reactions. This reveals the physicochemical mechanism behind the abnormal phenomenon of lower-level fruits rotting before the upper-level fruits;

[0104] In addition, the key parameter calculation unit also has built-in preprocessing logic for the input data; for the calculated micro-condensation phase change period The system will set a minimum physical threshold. When the calculated value is less than this threshold, the system will use... Subsequent calculations are performed to prevent the denominator from becoming too small or zero due to sensor noise or extreme environmental conditions, thereby ensuring the numerical stability and robustness of the core risk assessment model.

[0105] Through the parametric calculations of vibration, illumination, and interface microenvironment described above, the key parameter calculation unit successfully distilled complex, multi-domain physical phenomena into three core inputs that can directly drive the nonlinear dynamics model. This process of data dimensionality reduction and meaning extraction ensures that the model can accurately capture the key driving forces that trigger systemic risks.

[0106] Example 4:

[0107] The specific configuration of the positive feedback avalanche risk assessment model unit is used for:

[0108] The growth rate of normalized cumulative released stress was determined to be driven by the basic thermal relaxation rate, vibration energy, light energy, and normalized ethylene release flux.

[0109] The specific configuration of the positive feedback avalanche risk assessment model unit is used for:

[0110] The growth rate of normalized ethylene release flux was determined to be driven by both the instantaneous rate of normalized cumulative release stress and the interfacial biochemical amplification factor.

[0111] The positive feedback avalanche risk assessment model unit calculates the avalanche risk index by summing the squares of the normalized cumulative released stress and the normalized ethylene release flux.

[0112] Based on Example 1, this embodiment defines in detail the internal model structure and calculation method of the positive feedback avalanche risk assessment model unit. This model aims to reproduce the positive feedback closed loop between physical stress release and biochemical maturation process through mathematical means, and to quantify the risk based on this.

[0113] The core of the model consists of a set of coupled differential equations used to describe the dynamic evolution of two core state variables;

[0114] Core state variable definition:

[0115] Normalized cumulative released stress ( ): A dimensionless parameter representing the proportion of released polymer residual stress in the plastic fruit holder to the initial total stress at time t; its function is to quantify the degree of deterioration of the physical structure; its value range is [0,1].

[0116] Normalized ethylene release flux ( ): A dimensionless parameter representing the ratio of the ethylene release flux of a fruit to its peak respiratory climacteric flux at time t; its function is to quantify the process of biochemical maturation; its value range is [0,1]; where the peak respiratory climacteric flux used for normalization is a baseline parameter determined by publicly available literature data for a specific fruit variety or by previous biochemical experiments.

[0117] To ensure the model matches physical reality, the microscopic physical stress correlation data acquisition module will perform data acquisition at the initial moment. The collected data, after normalization, serves as the initial condition for the normalized cumulative released stress state variable. During system operation, the data collected by this module can be used to solve the model. Values ​​are periodically calibrated or data fused to correct model errors;

[0118] Coupled dynamic equations:

[0119] Before solving the model, the initial ethylene concentration data obtained by the macroscopic biochemical state data acquisition module is normalized and used as the initial condition for the normalized ethylene release flux state variable. ;

[0120] The positive feedback avalanche risk assessment model unit will normalize the cumulative released stress. The growth rate was determined to be driven by a combination of basal thermal relaxation, external physical energy input, and internal biochemical feedback; its calculation formula is:

[0121]

[0122] in, : The growth rate of normalized cumulative released stress, with dimensions of This is obtained by solving the differential equation in this unit;

[0123] Instantaneous vibrational energy and light energy, both with the dimension E;

[0124] Normalized ethylene emission flux, dimensionless, obtained by solving the coupled ethylene emission flux change rate equation in this unit;

[0125] The basic stress release rate constant, with dimensions of ;

[0126] The conversion coefficient from vibration and light energy to stress release rate, with dimensions of [missing value]. ;

[0127] Biochemical-physical coupling coefficient, with dimensions of ;

[0128] The above The model parameters were obtained through a series of calibration experiments conducted under controlled laboratory conditions, combined with nonlinear least squares fitting; for example, to determine... A series of experiments need to be conducted, with different constant vibrational energies applied to each experiment. And measure the corresponding initial stress release rate. To obtain the data point set By analyzing experimental data points within a pre-defined typical energy range... Linear regression analysis is used to simultaneously determine the effective range within that working interval. The value of is ; the calibration process for other parameters is similar;

[0129] The equation clearly reveals the multiple driving forces of physical stress release: inherent aging, external energy input, and internal chemical corrosion. It should be noted that the equation uses a linear superposition method to describe the multiple driving forces, which is an effective simplification under the assumption that there is no strong coupling effect between the driving terms in the normal operating range. This is sufficient to capture the main dynamic behavior of the system.

[0130] Description of the rate of change in ethylene release flux:

[0131] The positive feedback avalanche risk assessment model unit will normalize the ethylene release flux. The growth rate was determined to be a result of the chemical stimulation associated with stress release, amplified through the interfacial microenvironment; its calculation formula is:

[0132]

[0133] in, : The growth rate of normalized ethylene release flux, in units of ;

[0134] : Interface biochemical amplification factor, dimensionless;

[0135] The instantaneous rate of stress release, with dimensions of ;

[0136] : Physicochemical coupling coefficient, dimensionless, used in experiments to control stress at different constant rates Release and measure the corresponding initial ethylene flux growth rate. Obtain the data point set Through the Perform linear regression analysis to determine The value;

[0137] This equation establishes a crucial bridge between physical structural degradation and accelerated biochemical reactions, and innovatively amplifies the effect of the interfacial microenvironment. As a key multiplier term;

[0138] Closed-loop positive feedback and risk quantification:

[0139] The above two equations are passed and This forms a complete positive feedback loop. To comprehensively quantify this systemic risk, the positive feedback avalanche risk assessment model unit calculates the avalanche risk index by summing the squares of the normalized cumulative released stress and the normalized ethylene release flux. The calculation formula is as follows:

[0140]

[0141] in, Avalanche risk index, dimensionless, derived by this unit based on real-time solution. and Calculated;

[0142] This index integrates two core state variables of different natures into a unified risk measure, which can sensitively reflect the deterioration of the overall health status of the system and provide a reliable quantitative basis for setting early warning thresholds.

[0143] By constructing this coupled dynamics model, this invention mathematically formalizes the core mechanism of biochemical-physical stress positive feedback avalanche; this model can not only explain the phenomenon, but also make quantitative predictions; and the introduction of the avalanche risk index transforms the complex system state into an intuitive and operable single risk indicator, laying a theoretical foundation for achieving accurate early warning and proactive intervention.

[0144] Example 5:

[0145] The early warning and decision support unit is specifically configured for:

[0146] The growth rate of the avalanche risk index is compared with the preset warning threshold;

[0147] Among them, when the growth rate exceeds the warning threshold, the risk warning includes warning information and intervention suggestions;

[0148] The early warning and decision support unit is specifically configured for:

[0149] Compare the current value of the avalanche risk index with the pre-response threshold;

[0150] Among them, when the current value exceeds the disposal threshold, the risk warning is a disposal alert and a business measure suggestion;

[0151] The early warning and decision support unit is also configured to:

[0152] Fault attribution is performed based on the contribution of each driving term to the normalized cumulative released stress growth rate in order to identify the main causes of increased risk.

[0153] Based on Example 1, this embodiment further specifies and hierarchically designs the functions and logic of the early warning and decision support unit; its purpose is to transform the abstract risk index output by the preceding unit into a set of early warning and decision-making schemes with different levels of urgency that can effectively guide actual operations.

[0154] Early warning based on growth rate:

[0155] The early warning and decision support unit compares the growth rate of the avalanche risk index with a preset early warning threshold to achieve early identification of risks.

[0156] An abnormally high growth rate of the risk index is a strong signal that the system is about to enter a phase of exponential growth avalanche.

[0157] This unit calculates the avalanche risk index over continuous time intervals. Perform difference operations to obtain its instantaneous growth rate. When the calculated growth rate Exceeding the preset warning threshold At that time, the system will output early warning information and intervention suggestions;

[0158] Warning threshold ( This refers to a critical growth rate used to determine whether a risk has begun to accelerate; it is based on receiver operating characteristic (ROC) curve analysis of a large amount of historical data and model simulation results to determine an inflection point value that can optimally balance the sensitivity and specificity of early warning.

[0159] Alarm handling based on absolute values:

[0160] The early warning and decision support unit also compares the current value of the avalanche risk index with a preset handling threshold to trigger the highest level of alarm;

[0161] When the risk index reaches a high level, it usually means that the goods have undergone significant and potentially irreversible quality deterioration.

[0162] When the current value of the risk index Exceeding the preset processing threshold In such cases, the system will output alerts and recommendations for appropriate business actions.

[0163] Disposal threshold ( This refers to a risk index threshold directly associated with large-scale, irreversible corruption; it is calculated by comparing the final cargo loss rate of a large number of historical transportation batches with the final loss rate calculated by the model. Regression analysis was performed on the values ​​to find a risk index boundary that could best distinguish between acceptable cargo damage and large-scale cargo damage.

[0164] Risk attribution and root cause identification:

[0165] To make intervention recommendations more targeted, the early warning and decision support unit is also equipped with a fault attribution function;

[0166] When a warning or alarm is triggered, the unit attributes the fault to the cause based on the contribution of each driving term in the normalized cumulative released stress growth rate equation.

[0167] The implementation method is real-time comparison. , , and The instantaneous values ​​of these four items; if at a certain moment, The item with the highest percentage in the total will be identified by the system as the dominant factor in the current risk increase, and corresponding intervention suggestions with clear direction will be generated.

[0168] Through this multi-level, dual-indicator threshold logic, the system effectively distinguishes between early warning and final handling of risks. Its built-in fault attribution function enables the system not only to report the existence of risks but also to reveal their sources, thereby elevating the effectiveness and operability of intervention recommendations to a new level and truly achieving intelligent and precise risk management.

[0169] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An Internet of Things based plastic fruit cup quality traceability system characterized in that, The method comprises the following steps: a multi-modal data acquisition unit configured to acquire micro-physical stress-related data, mesoscopic interface environment data, macro-biochemical state data, and implicit environmental factor data in real time; a key parameter calculation unit configured to calculate an interface biochemical amplification coefficient based on the mesoscopic interface environment data; and calculate vibration energy and illumination energy based on the implicit environmental factor data; a positive feedback avalanche risk assessment model unit configured to combine the interface biochemical amplification coefficient, the vibration energy, the illumination energy, and the macro-biochemical state data to solve normalized cumulative release stress and normalized ethylene release flux; and generate an avalanche risk index based on the normalized cumulative release stress and the normalized ethylene release flux; a warning and decision support unit configured to output a risk warning based on the avalanche risk index. The multi-modal data acquisition unit specifically comprises: a micro-physical stress-related data acquisition module configured to acquire stress data of stress concentration points of a plastic fruit cup structure; a mesoscopic interface environment data acquisition module configured to acquire temperature and humidity data near the interface between the fruit cup and the fruit; a macro-biochemical state data acquisition module configured to acquire ethylene concentration data of the gas environment in the packaging box; an implicit environmental factor acquisition module configured to acquire vibration frequency spectrum data of the transport vehicle and spectral energy distribution data of the warehouse lighting; The key parameter calculation unit is specifically configured to: calculate a micro-condensation phase change period through a peak-valley detection algorithm based on the temperature and humidity data obtained by the mesoscopic interface environment data acquisition module; map the micro-condensation phase change period to the interface biochemical amplification coefficient through a pre-set inverse function relationship; The positive feedback avalanche risk assessment model unit is specifically configured to: determine the growth rate of the normalized cumulative release stress as being jointly driven by the basic thermal relaxation rate, the vibration energy, the illumination energy, and the normalized ethylene release flux; The positive feedback avalanche risk assessment model unit is specifically configured to: determine the growth rate of the normalized ethylene release flux as being jointly driven by the instantaneous rate of the normalized cumulative release stress and the interface biochemical amplification coefficient.

2. The plastic fruit cup quality tracing system based on the Internet of Things according to claim 1, characterized in that, The key parameter calculation unit is specifically configured to: integrate the vibration frequency spectrum data obtained by the implicit environmental factor acquisition module in a pre-set dangerous frequency band to determine the vibration energy; integrate the spectral energy distribution data obtained by the implicit environmental factor acquisition module in a high-risk short-wave band to determine the illumination energy.

3. The plastic fruit cup quality tracing system based on the Internet of Things according to claim 1, characterized in that, The positive feedback avalanche risk assessment model unit generates the avalanche risk index by calculating the square sum of the normalized cumulative release stress and the normalized ethylene release flux.

4. The plastic fruit cup quality tracing system based on the Internet of Things according to claim 1, characterized in that, The warning and decision support unit is specifically configured to: compare the growth rate of the avalanche risk index with a pre-set warning threshold; wherein when the growth rate exceeds the warning threshold, the risk warning is warning information and intervention suggestions.

5. The plastic fruit cup quality tracing system based on the Internet of Things according to claim 1 or 4, characterized in that, The warning and decision support unit is specifically configured to: compare the current value of the avalanche risk index with a pre-disposal threshold; wherein the risk alert is a treatment alert and a business measure recommendation when the current value exceeds the treatment threshold.

6. The plastic pun quality traceability system based on the Internet of Things according to claim 1, characterized in that, The alert and decision support unit is further configured for: According to the contribution degree of each driving item to the normalized accumulated released stress growth rate, failure attribution is performed to clarify the main reason for the risk increase.

Images