Digital logistics management system and method

Abstract

The invention relates to the technical field of logistics management, in particular to a digital logistics management system and method. The method comprises the following steps: collecting multi-dimensional environmental parameter field data around a cargo and biochemical index change data of a cargo body; constructing a permeability coefficient matrix of the cargo based on the multi-dimensional environmental parameter field data and the biochemical index change data according to a preset material library, wherein the permeability coefficient matrix is obtained by quantifying the interaction rate between different environmental factors and the material of the cargo; and calculating the current environmental influence intensity based on the permeability coefficient matrix and the multi-dimensional environmental parameter field data, and performing thermodynamic coupling analysis in combination with the biochemical index change data to obtain quality entropy increase rate data. According to the method, the problem that the cargo quality change is difficult to accurately monitor, predict and actively protect in the logistics process is solved through a method of combining the establishment of the permeability coefficient matrix and thermodynamic coupling analysis, and the whole-course digital management from environmental perception to quality guarantee is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics management, and in particular to a digital logistics management system and method. Background Art

[0002] Logistics management refers to the planning, organization, command, coordination, control, and supervision of logistics activities. The core of logistics management is to maximize resource utilization by optimizing every aspect of logistics activities. Digital logistics management is an indispensable component of the modern supply chain. By integrating information technology with traditional logistics operations, it improves logistics efficiency, reduces costs, and optimizes the entire supply chain. The foundation of digital logistics lies in comprehensive and accurate data collection. This includes inventory data, shipping status, delivery routes, time records, and other information.

[0003] However, traditional digital logistics management methods often suffer from the following issues: Traditional logistics management systems rely heavily on conventional protective measures and lack the flexibility to adapt to changing transportation environments. Digital logistics management methods dynamically adjust environmental factors and implement intelligent protective measures (such as temperature and humidity control, vapor phase protection, etc.), combined with real-time feedback mechanisms. This not only improves cargo protection but also ensures the precise and effective implementation of regulatory measures. Summary of the Invention

[0004] Based on this, it is necessary for the present invention to provide a digital logistics management system and method to solve at least one of the above technical problems.

[0005] To achieve the above objectives, a digital logistics management method includes the following steps:

[0006] Step S1: Collecting multi-dimensional environmental parameter field data around the cargo and biochemical index change data of the cargo itself;

[0007] Step S2: Constructing a cargo permeability coefficient matrix based on the multi-dimensional environmental parameter field data and biochemical indicator change data according to a preset material library, wherein the permeability coefficient matrix is ​​obtained by quantifying the interaction rate between different environmental factors and cargo materials;

[0008] Step S3: Calculate the current environmental impact intensity based on the permeability coefficient matrix and multidimensional environmental parameter field data, and perform thermodynamic coupling analysis in combination with the biochemical indicator change data to obtain quality entropy increase rate data, which represents the rate of increase in disorder of the quality state of the goods; generate a quality state prediction model based on the quality entropy increase rate data, and perform microenvironment compensation parameter combination analysis on the quality state prediction model to obtain a microenvironment compensation parameter combination;

[0009] Step S4: performing a difference analysis on the multi-dimensional environmental parameter field data using a combination of microenvironment compensation parameters to generate dynamic control instruction data; controlling the microenvironment control device based on the dynamic control instruction data, and making adjustments based on real-time feedback from biochemical indicators to obtain protection execution status data;

[0010] Step S5: The protection execution status data is used to generate a digital identification of the cargo protection history through quantum dot marking technology, and is written into the blockchain system to build a responsibility traceability proof chain for the cargo.

[0011] By collecting multidimensional environmental parameter field data surrounding the goods and biochemical indicator variation data of the goods themselves, this invention comprehensively understands the external environmental factors and internal biochemical characteristics that influence the goods' quality, ensuring the accuracy and pertinence of subsequent calculations. Based on this data, a permeability coefficient matrix for the goods is constructed using a pre-set material library, quantifying the interaction rates between different environmental factors and the goods' materials, providing a scientific basis for subsequent environmental impact assessments. By calculating the permeability coefficient matrix and multidimensional environmental parameter field data, the current environmental impact intensity can be accurately assessed. Combined with the biochemical indicator variation data, thermodynamic coupling analysis is performed to derive the rate of increase in disorder of the goods' quality state, i.e., the quality entropy increase rate. This data is used to construct a quality state prediction model, which not only identifies potential quality degradation trends in the goods in advance but also analyzes microenvironmental compensation parameter combinations based on the Le Chatelier equilibrium principle to optimize environmental control strategies. By comparing environmental differences under different compensation parameters, dynamic control instruction data is generated, enabling precise control of the microenvironmental control device. Combined with real-time feedback from biochemical indicators, the effectiveness and adaptability of control measures are ensured. Ultimately, through quantum dot tagging technology, protection execution status data is converted into a traceable digital identifier and written into the blockchain system, establishing a chain of accountability for product traceability. This mechanism not only enables proactive protection of goods during storage and transportation but also forms a complete data chain, providing strong support for quality traceability, accountability definition, and intelligent management.

[0012] The present invention further provides a digital logistics management system for executing the digital logistics management method described above, the digital logistics management system comprising:

[0013] Environmental perception and collection module, used to collect multi-dimensional environmental parameter field data around the cargo and biochemical indicator change data of the cargo itself;

[0014] The material interaction modeling module is used to construct a cargo permeability matrix based on a preset material library, multi-dimensional environmental parameter field data, and biochemical indicator change data. The permeability matrix is ​​obtained by quantifying the interaction rate between different environmental factors and cargo materials.

[0015] The quality prediction and analysis module is used to calculate the current environmental impact intensity based on the permeability coefficient matrix and multi-dimensional environmental parameter field data, and conduct thermodynamic coupling analysis in combination with biochemical indicator change data to obtain the quality entropy increase rate data, which represents the rate of increase in disorder of the quality state of the goods. The quality state prediction model is generated based on the quality entropy increase rate data, and a microenvironment compensation parameter combination analysis is performed to obtain the microenvironment compensation parameter combination.

[0016] The intelligent control execution module is used to perform differential analysis on multi-dimensional environmental parameter field data using a combination of microenvironment compensation parameters to generate dynamic control instruction data; the microenvironment control device is controlled based on the dynamic control instruction data, and adjustments are made based on real-time feedback from biochemical indicators to obtain protection execution status data;

[0017] The blockchain traceability module is used to generate a digital identification of the cargo protection history through quantum dot marking technology from the protection execution status data, and write it into the blockchain system to build a responsibility traceability proof chain for the cargo.

[0018] This invention provides comprehensive information on the external environment and internal biochemical state of a product through the real-time collection of multidimensional environmental parameter field data surrounding the product and biochemical indicator variation data within the product. This data collection provides a comprehensive foundation for subsequent analysis and decision-making, enabling all subsequent operations to respond to real, timely environmental changes. Building on this foundation, the material interaction modeling module, using a pre-set material library and combining multidimensional environmental parameter field data with biochemical indicator variation data, constructs a permeability matrix for the product, quantifying the interaction rate between different environmental factors and the product's material. This enables accurate assessment of the specific impacts of different environmental factors on the product, improving the accuracy of product status predictions. The combination of the permeability matrix and multidimensional environmental parameter field data enables real-time calculation of the current environmental impact intensity. Using the biochemical indicator variation data, thermodynamic coupling analysis is performed to obtain the rate of increase in disorder, or quality entropy increase rate, which characterizes the product's quality status. This data not only reveals the quality trend of the product under specific environmental conditions but also, based on thermodynamic principles, allows for the deduction of future quality degradation patterns. The quality status prediction model generated based on this data enables more accurate dynamic prediction of cargo quality changes. The Le Châtelier equilibrium principle plays a key role in analyzing microenvironmental compensation parameter combinations, helping to determine the most appropriate microenvironmental control scheme under varying environmental conditions. Using microenvironmental compensation parameter combinations, a differential analysis of multidimensional environmental parameter field data is performed to generate dynamic control instruction data. These instructions precisely adjust the microenvironmental control devices, enabling timely adjustment and optimization of environmental factors, thereby ensuring that cargo maintains the optimal protective environment during transportation. Combined with real-time feedback from biochemical indicators, the system dynamically optimizes control measures to ensure the continued effectiveness of protective measures, ultimately generating protection implementation status data. Quantum dot labeling technology generates a unique digital identifier for each item, and this protection implementation status data is written to the blockchain system. This ensures that every protective measure and every key node is recorded in the blockchain, ensuring data integrity and immutability. By establishing a chain of accountability for cargo, the system provides a transparent and traceable digital record of the entire cargo transportation process, accurately reflecting the responsibilities of all parties involved and the entire cargo protection process. This traceability not only enhances the safety of the cargo transportation process, but also provides a reliable data basis for accountability when problems arise. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Other features, objects and advantages of the present invention will become more apparent upon reading the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0020] Figure 1 This is a schematic diagram of the steps of the digital logistics management method of the present invention;

[0021] Figure 2 for Figure 1 Detailed step flow diagram of step S1;

[0022] Figure 3 for Figure 1 Detailed step flow chart of step S2 in FIG. DETAILED DESCRIPTION

[0023] To achieve this, please refer to Figures 1 to 3 The present invention provides a digital logistics management method, which includes the following steps:

[0024] Step S1: Collecting multi-dimensional environmental parameter field data around the cargo and biochemical index change data of the cargo itself;

[0025] In the embodiment of the present invention, a batch of seafood products is selected as the experimental object under the cold chain transportation environment, and a multi-dimensional environmental monitoring device integrating temperature and humidity, gas composition (such as oxygen, carbon dioxide, ammonia), light intensity, pressure and vibration sensors is used to collect data on the environmental parameter field of the cargo storage area. At the same time, biochemical sensors are used to monitor the biochemical indicators of seafood in real time, including volatile basic nitrogen (TVB-N), pH value, ATP degradation products (the ratio of IMP to HxR), etc., and the data is uploaded to the central data processing system at certain time intervals. During the data collection process, the system automatically eliminates outliers and performs data smoothing based on the sliding window technology to ensure the stability and continuity of the data.

[0026] Step S2: Constructing a cargo permeability coefficient matrix based on the multi-dimensional environmental parameter field data and biochemical indicator change data according to a preset material library, wherein the permeability coefficient matrix is ​​obtained by quantifying the interaction rate between different environmental factors and cargo materials;

[0027] The embodiment of the present invention is based on a preset material library, which contains the permeability characteristics of different seafood packaging materials (such as polyethylene, PET aluminized film, multi-layer composite film) under the action of various environmental factors (temperature, humidity, gas composition). First, for the packaging materials used in this batch of seafood, the system extracts its diffusion rate data under the action of different environmental parameters, and combines the environmental parameter field data collected in real time with the change data of seafood biochemical indicators, and uses the finite element numerical calculation method to solve the diffusion coefficient and mass transfer flux of different environmental factors in the packaging material, thereby constructing the permeability coefficient matrix of this batch of goods. Each element in the matrix represents the permeation rate of a specific environmental factor to the material of the goods, such as how much the oxygen permeation rate increases due to a temperature change of 1°C, how much the rate of water vapor passing through the film layer increases due to a 10% change in humidity, etc., providing a quantitative basis for subsequent analysis.

[0028] Step S3: Calculate the current environmental impact intensity based on the permeability coefficient matrix and multidimensional environmental parameter field data, and perform thermodynamic coupling analysis in combination with the biochemical indicator change data to obtain quality entropy increase rate data, which represents the rate of increase in disorder of the quality state of the goods; generate a quality state prediction model based on the quality entropy increase rate data, and perform microenvironment compensation parameter combination analysis on the quality state prediction model to obtain a microenvironment compensation parameter combination;

[0029] Based on the calculated permeability coefficient matrix and real-time multidimensional environmental parameter field data, this embodiment of the present invention uses numerical integration methods to calculate the comprehensive impact of the current environment on cargo quality. This involves calculating the combined forces exerted by factors such as temperature, humidity, and gas concentration on cargo. Furthermore, using dynamic nonequilibrium thermodynamics, the system analyzes the impact of environmental changes on biochemical indicators and calculates the cargo quality entropy increase rate. The quality entropy increase rate characterizes the rate of cargo quality deterioration; a higher value indicates a faster decline in quality. Based on this, the system constructs a quality status prediction model using machine learning algorithms (such as LSTM or random forest regression). This model is trained using historical data to predict quality status at future points in time. Furthermore, based on the Le Chatelier equilibrium principle, the system analyzes how to minimize the cargo quality entropy increase rate by manipulating environmental parameters such as temperature, humidity, and oxygen concentration. This system then generates microenvironmental compensation parameter combinations, such as lowering the ambient temperature by 1°C, increasing the oxygen concentration by 2%, and adjusting the humidity to 60%, to slow the degradation process.

[0030] Step S4: performing a difference analysis on the multi-dimensional environmental parameter field data using a combination of microenvironment compensation parameters to generate dynamic control instruction data; controlling the microenvironment control device based on the dynamic control instruction data, and making adjustments based on real-time feedback from biochemical indicators to obtain protection execution status data;

[0031] The embodiment of the present invention utilizes a combination of micro-environment compensation parameters to perform a differential analysis on the current multi-dimensional environmental parameter field data, calculates the deviation between the actual environmental parameters and the target compensation parameters, and generates dynamic control instruction data based on the deviation amplitude. For example, during transportation, if the ambient temperature is higher than the compensation parameter setting value, the control instruction will lower the set temperature of the cold chain refrigeration equipment. If the humidity is low, the humidification power of the humidity control device will be increased. If the oxygen concentration is insufficient, the micro-environment gas regulating device will be started for oxygen supplementation. During the dynamic control process, the system collects feedback data on the biochemical indicators of the goods in real time, and uses an adaptive PID control algorithm to adjust the operating parameters of the control device to ensure that the environmental parameters always approach the target value. All adjustment operations and equipment execution status are recorded to form protection execution status data.

[0032] Step S5: The protection execution status data is used to generate a digital identification of the cargo protection history through quantum dot marking technology, and is written into the blockchain system to build a responsibility traceability proof chain for the cargo.

[0033] After the protection execution status data in the embodiment of the present invention is compressed and encrypted, quantum dot marking technology is used to generate a digital identification of the cargo protection history. Specifically, through quantum dot coding technology, key environmental control data (such as the control history of temperature, humidity, and oxygen concentration) are embedded in the cargo packaging or shipping label in the form of optically readable nano-markers, so that the data is traceable. At the same time, the system writes the protection execution status data and its quantum dot identification information into the blockchain system to achieve full-chain tamper-proof evidence storage and build a responsibility traceability proof chain for the goods. For example, the blockchain evidence data of a batch of seafood includes storage temperature change curves, control device working records, biochemical indicator change trends, etc., to ensure traceability of the entire process from factory to terminal sales. Once a quality dispute occurs, the complete protection record of the goods can be queried through the blockchain to clarify the responsibility.

[0034] By collecting multidimensional environmental parameter field data surrounding the goods and biochemical indicator variation data of the goods themselves, this invention comprehensively understands the external environmental factors and internal biochemical characteristics that influence the goods' quality, ensuring the accuracy and pertinence of subsequent calculations. Based on this data, a permeability coefficient matrix for the goods is constructed using a pre-set material library, quantifying the interaction rates between different environmental factors and the goods' materials, providing a scientific basis for subsequent environmental impact assessments. By calculating the permeability coefficient matrix and multidimensional environmental parameter field data, the current environmental impact intensity can be accurately assessed. Combined with the biochemical indicator variation data, thermodynamic coupling analysis is performed to derive the rate of increase in disorder of the goods' quality state, i.e., the quality entropy increase rate. This data is used to construct a quality state prediction model, which not only identifies potential quality degradation trends in the goods in advance but also analyzes microenvironmental compensation parameter combinations based on the Le Chatelier equilibrium principle to optimize environmental control strategies. By comparing environmental differences under different compensation parameters, dynamic control instruction data is generated, enabling precise control of the microenvironmental control device. Combined with real-time feedback from biochemical indicators, the effectiveness and adaptability of control measures are ensured. Ultimately, through quantum dot tagging technology, protection execution status data is converted into a traceable digital identifier and written into the blockchain system, establishing a chain of accountability for product traceability. This mechanism not only enables proactive protection of goods during storage and transportation but also forms a complete data chain, providing strong support for quality traceability, accountability definition, and intelligent management.

[0035] Preferably, step S1 includes the following steps:

[0036] Step S11: Arranging a multi-parameter sensor array on the outer packaging of the goods to construct a spatially distributed sensor network, wherein the multi-parameter sensor array includes environmental parameter sensor elements for real-time acquisition of temperature gradient distribution, humidity fluctuation coefficient, light intensity, air pressure change rate, and vibration spectrum information;

[0037] In an embodiment of the present invention, a multi-parameter sensor array is arranged on cold chain transport packaging boxes for a batch of fresh fruits (such as blueberries) to construct a spatially distributed sensing network. In specific implementation, micro temperature and humidity sensors (with an accuracy of ±0.1°C and ±1% RH) are installed at different locations of the packaging box (such as the top, side walls, and bottom) to monitor the temperature gradient distribution and humidity fluctuation coefficient; a photoelectric sensor (with a response band of 400-700nm) is placed on the inner wall of the box to detect changes in light intensity; a MEMS pressure sensor (with a measurement range of 300-1100hPa) is installed at the four corners of the packaging box to record the rate of change of air pressure; and a high-sensitivity accelerometer (with a bandwidth of 0.1-500Hz) is attached to the junction of the bottom and side walls of the box to obtain vibration spectrum information. All sensor elements are connected to a central data acquisition terminal via a low-power wireless communication module (such as BLE or LoRa) to ensure stable transmission of sensor data. This spatially distributed sensing network can capture the dynamic changes of environmental parameters during transportation in real time, avoiding the distortion of environmental parameters caused by single-point monitoring, thereby providing more comprehensive environmental data support.

[0038] Step S12: using a spatially distributed sensor network to perform data collection and transmission according to a preset time interval to form multi-dimensional environmental parameter field data;

[0039] The embodiment of the present invention is based on the deployed spatial distributed sensor network, sets the data collection time interval to 5 minutes, and continuously performs automatic collection and transmission of multi-dimensional environmental parameters during the cold chain transportation process. In specific operation, each sensor is activated and records the current data according to the preset collection interval. The temperature and humidity sensor measures the ambient temperature and humidity, and calculates the humidity fluctuation coefficient (that is, the standard deviation of the humidity change per unit time); the photoelectric sensor detects the light intensity entering the package and calculates the light fluctuation rate; the air pressure sensor records the air pressure change and calculates its rate of change; the accelerometer collects vibration signals, and uses FFT (Fast Fourier Transform) to analyze the vibration spectrum and extract vibration characteristic values ​​in different frequency bands. All sensor data are aggregated to the data acquisition terminal through the wireless transmission module and timestamped in combination with GPS positioning information to ensure the time-space correspondence of the data. During the data collection process, the system will automatically detect abnormal data points (such as sudden change values, data that has not changed for a long time), and perform outlier removal and data smoothing processing, and finally form multi-dimensional environmental parameter field data, providing accurate environmental information input for subsequent cargo quality assessment.

[0040] Step S13: Install a non-invasive biochemical sensor array on the outer packaging surface of the goods to non-destructively detect changes in the biochemical state of the goods through the packaging material, thereby obtaining biochemical indicator change data of the goods themselves.

[0041] In an embodiment of the present invention, a non-invasive biochemical sensor array is attached to the inner wall of the blueberry cold chain packaging. This sensor uses surface-enhanced Raman spectroscopy (SERS) detection technology, which can penetrate the packaging material and non-destructively analyze changes in internal biochemical states. In specific implementation, the detection wavelength range of the biochemical sensor is set to 650-900nm, which can detect volatile organic compounds (VOCs) released by blueberries at the molecular level, such as acetaldehyde, ethanol, acetic acid, and hexenal. These indicators can be used to assess the freshness and spoilage tendency of blueberries. At the same time, the sensor array is also equipped with a pH detection unit, which uses fluorescence resonance energy transfer (FRET) technology to sense pH changes in the fruit microenvironment, thereby further reflecting the physiological state of the blueberries. All biochemical data are analyzed in real time by an embedded data processing module and uploaded to a cloud data center via wireless communication. Throughout the entire process, non-invasive detection methods ensure the integrity of the cargo packaging, avoid the pollution and damage caused by traditional sampling and testing, and at the same time achieve real-time, non-destructive biochemical status monitoring, providing a scientific basis for quality prediction and optimization of cargo preservation strategies.

[0042] Preferably, step S13 includes the following steps:

[0043] Step S131: installing a non-invasive biochemical sensor array on the surface of the outer packaging of the goods to form a multimodal biochemical sensor array, wherein the non-invasive biochemical sensor includes a near-infrared spectroscopy sensor, a Raman spectroscopy sensor, an electronic nose sensor, and a fluorescence spectroscopy sensor;

[0044] In an embodiment of the present invention, a multimodal biochemical sensor array is installed on the surface of the cold chain transportation packaging of a batch of high-end Wagyu beef, wherein the non-invasive biochemical sensor array is composed of a near-infrared spectral sensor (working wavelength 900-2500nm), a Raman spectral sensor (excitation wavelength 785nm), an electronic nose sensor (containing multiple gas sensitive elements, such as metal oxide semiconductor gas sensors), and a fluorescence spectral sensor (detection wavelength 350-600nm). The sensor is installed near the transparent detection window inside the packaging to ensure that the optical sensor can effectively penetrate the packaging material. At the same time, the electronic nose sensor is arranged at the vent of the packaging to efficiently collect volatile organic compounds (VOCs) released by the goods. All sensors are connected to the data acquisition terminal through a low-power wireless communication module (such as NB-IoT) to achieve real-time data acquisition and remote transmission. This multimodal sensor array can non-destructively monitor the microscopic biochemical changes in the quality of Wagyu beef and provide multi-dimensional data support for quality assessment.

[0045] Step S132: Using a near-infrared spectral sensor and a Raman spectral sensor to penetrate the cargo packaging material, collect reflection or transmission spectral data, and convert it into composition change data inside the cargo through multivariate spectral analysis technology;

[0046] The embodiment of the present invention utilizes near-infrared spectroscopy sensors and Raman spectroscopy sensors to penetrate the Wagyu beef packaging material, collect its reflection or transmission spectral data, and convert it into data on changes in the internal components of the beef through multivariate spectral analysis technology (such as partial least squares regression analysis (PLSR)). In specific implementation, the near-infrared spectroscopy sensor illuminates the surface of the beef and records the absorption spectrum to analyze changes in key components such as myoglobin, fat content, and moisture; the Raman spectroscopy sensor illuminates the surface of the beef with a laser and measures the scattered light signal to extract information such as the degree of myoglobin oxidation and changes in the secondary structure of the protein. Spectral feature extraction is optimized through cross-validation technology (such as principal component analysis (PCA)), and ultimately key biochemical parameters such as the beef fat oxidation rate and protein degradation rate are obtained, providing accurate data support for the assessment of the deterioration of the quality of the goods.

[0047] Step S133: using an electronic nose sensor to collect information on volatile organic compounds released by the cargo;

[0048] The electronic nose sensor of the embodiment of the present invention is composed of multiple gas-sensitive units, including metal oxide semiconductor (MOS) gas sensors, electrochemical gas sensors, etc., which can detect volatile organic compounds (VOCs) released by beef during storage and transportation. In specific implementation, the electronic nose sensor is placed near the vent of the packaging box and automatically activated every 5 minutes to inhale the gas sample in the package and record the resistance change of the gas sensor through a high-precision analog-to-digital conversion module (resolution 16-bit). Through pre-trained pattern recognition algorithms (such as support vector machines SVM or deep neural networks DNN), VOCs signals are analyzed and gas information such as amines and thiols released in the early stages of beef deterioration is extracted to provide data support for real-time monitoring.

[0049] Step S134: Perform gas molecule identification on the volatile organic compound information and analyze characteristic gas molecules generated during the oxidation, hydrolysis or microbial metabolism of the cargo;

[0050] The embodiment of the present invention performs gas molecule identification on the volatile organic compound (VOCs) data collected by the electronic nose sensor to analyze the characteristic gas molecules produced by beef during oxidation, hydrolysis or microbial metabolism. In specific implementation, the gas sensor array is combined with the principal component analysis (PCA) method to perform feature dimensionality reduction on the detection signal, and different gas signals are classified by K-means clustering. During the analysis process, if an increase in the concentration of volatile sulfides (such as dimethyl disulfide and hydrogen sulfide) is detected, it means that fat oxidation is intensified; if a high concentration of trimethylamine or indole gas is detected, it indicates that protein decomposition is active and there may be a risk of microbial contamination. Finally, based on the concentration change trend of the characteristic gas molecules, an oxidation, hydrolysis and microbial metabolism evaluation model for beef quality is established, providing a reliable basis for subsequent quality prediction.

[0051] Step S135: Detecting specific biomarkers in the cargo using a fluorescence spectrum sensor and obtaining characteristic gas molecules of the specific biomarkers, wherein the specific biomarkers include oxidation products, enzyme activity change products, and microbial metabolites;

[0052] The embodiment of the present invention uses a fluorescence spectrum sensor to detect specific biomarkers in beef and obtain corresponding characteristic gas molecules. In specific implementation, the fluorescence spectrum sensor uses ultraviolet excitation light (wavelength 350nm) to excite oxidation products (such as malondialdehyde), enzyme activity change products (such as superoxide dismutase SOD), and microbial metabolites (such as indole compounds) in beef, and records their fluorescence emission spectra (500-600nm). Through changes in fluorescence spectrum intensity, the degree of beef oxidation, changes in enzyme activity, and whether there is a risk of microbial contamination can be quantitatively analyzed. For example, if a significant increase in fluorescence intensity is detected at 520nm, it indicates that fat oxidation has intensified; if the fluorescence signal at 570nm is enhanced, it may be a sign of active microbial metabolism. Combined with the VOCs data collected by the electronic nose sensor, the quality change trend of Wagyu beef can be further verified.

[0053] Step S136: Perform data fusion and dimensionality reduction processing on the characteristic gas molecules, characteristic gas molecules, and component change data to generate biochemical indicator change data of the cargo itself.

[0054] The embodiment of the present invention performs data fusion and dimensionality reduction processing on the component change data obtained by characteristic gas molecules and spectral detection, and finally generates the biochemical index change data of the beef body. In specific implementation, first, a multimodal data fusion method (such as weighted average fusion) is used to align the data of different sensors to ensure the temporal consistency of each data type; secondly, principal component analysis (PCA) and linear discriminant analysis (LDA) are used to reduce the dimensionality of high-dimensional biochemical data and extract key characteristic variables, such as fat oxidation rate, protein degradation rate, microbial metabolic index, etc.; finally, a time series modeling method (such as LSTM neural network) is used to predict the changing trend of beef quality, and the biochemical index change data is uploaded to the cloud database. Through this data processing method, real-time evaluation of Wagyu beef quality can be achieved, providing a scientific basis for microenvironmental regulation during transportation and storage.

[0055] This invention utilizes a multimodal biochemical sensor system, constructed by installing a non-invasive biochemical sensor array on the outer packaging of goods, to accurately monitor the internal quality of goods. The non-invasive biochemical sensor array includes near-infrared spectroscopy, Raman spectroscopy, electronic nose, and fluorescence spectroscopy. These sensors work synergistically to comprehensively analyze the internal composition and biochemical reaction characteristics of goods using multiple detection methods. Near-infrared and Raman spectroscopy sensors penetrate the packaging material to collect reflectance or transmission spectral data from the interior of the goods. Combined with multivariate spectral analysis techniques, these spectral signals are converted into information on changes in the goods' internal composition, enabling highly sensitive detection of compositional fluctuations. The electronic nose sensor further collects information on volatile organic compounds (VOCs) released by the goods, enabling rapid identification of characteristic gases generated by oxidation, hydrolysis, or microbial metabolism during storage or transportation. Combined with gas molecule recognition technology, this system can accurately analyze oxidation products, metabolic byproducts, and potential signs of spoilage or deterioration involved in changes in the goods' quality. Fluorescence spectroscopy sensors detect specific biomarkers, identifying oxidation products, enzyme activity changes, and microbial metabolites in the goods, enabling comprehensive biochemical assessment of the goods' quality. Ultimately, through data fusion and dimensionality reduction, the system integrates characteristic gas molecule information, volatile organic compound data, and spectral component analysis results to generate data on changes in the biochemical indicators of the goods themselves, providing high-precision, real-time data support for subsequent quality assessment and environmental control. This method enables precise monitoring of the internal composition and quality status of goods without compromising packaging integrity, improving quality monitoring capabilities within supply chain management and enhancing early warning and traceability of changing product quality trends.

[0056] Preferably, step S2 includes the following steps:

[0057] Step S21: Retrieving reference material property data that matches the current cargo type from a preset material library, wherein the reference material property data includes cargo internal structure stability index, surface activity index, heat exchange coefficient, and gas phase exchange coefficient;

[0058] In the cold chain transportation process of high-end fresh seafood (such as deep-sea salmon) in the embodiment of the present invention, in order to evaluate the impact of the environment on the quality of the goods, it is necessary to first retrieve the benchmark material property data of this type of seafood from the preset material library. The specific operations are as follows: query the database to obtain the internal structure stability index of salmon (used to measure the muscle fiber bonding strength and cell wall integrity), surface activity index (characterizing the degree of protein denaturation and hydration layer stability), heat exchange coefficient (thermal conductivity and specific heat capacity of different parts of the fish body measured by experiment), gas phase exchange coefficient (reflecting the permeability of fish tissue to oxygen, carbon dioxide and other gases). For example, under standard storage conditions of 0°C and 85% RH, the internal structure stability index of salmon is 0.9 (1.0 in a completely fresh state), the surface activity index is 0.75, the heat exchange coefficient is 1.5W / (m·K), and the gas phase exchange coefficient is 2.3×10-3cm 2 / s.

[0059] Step S22: converting the multidimensional environmental parameter field data into a standardized environmental factor intensity matrix, wherein the environmental factor intensity matrix includes numerical representations of five environmental factors, namely, temperature, humidity, light, air pressure, and vibration, at different intensity levels;

[0060] In the embodiment of the present invention, during the cold chain transportation of salmon, the on-board environmental monitoring system collects parameters such as temperature (-2°C to 5°C), humidity (65% RH to 95% RH), light (0lx to 8000lx), air pressure (95kPa to 102kPa) and vibration (0g to 5g acceleration). Since the data units collected by different sensors are different, a normalization processing method is required to map the raw data to the [0,1] interval. For example, in an environment of 0°C, the standardized value of temperature is set to 0.2, the humidity is set to 0.8 at 85% RH, and the vibration intensity is set to 0.3 at 0.5g. Ultimately, these standardized data form an environmental factor intensity matrix for subsequent calculations.

[0061] Step S23: establishing a response relationship model between environmental factors and biochemical indicators based on the biochemical indicator change data and the standardized environmental factor intensity matrix;

[0062] In order to analyze the impact of environmental factors on the biochemical indicators of salmon, the embodiment of the present invention uses historical experimental data to establish a response relationship model between environmental factors and biochemical indicators. First, biochemical indicators such as tissue protein degradation rate, lipid oxidation index, and microbial growth rate are extracted from the data sets of different transport batches, and combined with environmental factor data such as temperature, humidity, and light, multiple regression analysis and random forest regression are used to construct a prediction model. For example, when the temperature rises by 5°C, the tissue protein degradation rate increases by about 0.08 / h; when the humidity rises by 10%RH, the lipid oxidation index increases by 0.05 / h. The model parameters are optimized through cross-validation to make it have a higher prediction accuracy.

[0063] Step S24: Determine the initial state value and stability range of each cargo characteristic parameter based on the reference material characteristic data, and establish a reference model for the cargo characteristic parameters, wherein the reference model includes standard values ​​of the cargo's internal structure stability index, surface activity index, heat exchange coefficient, and gas phase exchange coefficient under standard environmental conditions;

[0064] In order to establish a benchmark model for salmon, the present invention first determines the initial state values ​​and stability ranges of its characteristic parameters. For example, under a standard transportation environment (0°C, 85% RH, dark environment, no vibration), the internal structure stability index of salmon is 0.9, the surface activity index is 0.75, the heat exchange coefficient is 1.5 W / (m·K), and the gas phase exchange coefficient is 2.3×10-3cm 2 Then, using the experimentally measured variation trends under different environmental conditions, a baseline mathematical model of salmon characteristic parameters was established for subsequent impact assessment.

[0065] Step S25: Using the response relationship model, a dynamic framework for the influence of environmental factors is established to convert the effects of the five environmental factors (temperature, humidity, light, air pressure, and vibration) on cargo characteristics into reaction kinetic parameters;

[0066] The embodiment of the present invention uses the principle of the Arrhenius equation to construct a dynamic framework for environmental influences to calculate the rate of change of salmon characteristics due to factors such as temperature, humidity, and light. For example, in the laboratory, salmon were stored in environments of 0°C, 5°C, and 10°C, and their muscle fiber degradation rates were measured. The activation energy and reaction rate constant were calculated by exponential fitting. Experiments have shown that for every 5°C increase in temperature, the muscle fiber degradation rate increases by about 20%. Similarly, the effects of humidity and light on the surfactant index and lipid oxidation index can also be quantified using this framework.

[0067] Step S26: Calculating the theoretical impact rates of the five environmental factors on the four benchmark properties based on the reaction kinetic parameters, thereby obtaining theoretical penetration impact data;

[0068] Based on the aforementioned kinetic parameters, the present invention calculates the theoretical impact rate of environmental factors on baseline properties. For example, a 5°C temperature increase increases the muscle fiber degradation rate by 0.08 / h; a 20% RH increase in humidity increases the water evaporation rate by 0.04 / h; and a 2000 lx increase in light intensity increases the pigment degradation rate by 0.02 / h. The same method is used to calculate the effects of air pressure and vibration on the four baseline properties, ultimately yielding theoretical penetration impact data.

[0069] Step S27: calibrating and optimizing the theoretical penetration impact data and the actual impact data predicted by the response relationship model to obtain a penetration impact coefficient;

[0070] In order to calibrate and optimize the theoretical penetration impact data, the embodiment of the present invention compares it with actual transportation data. For example, during an actual transportation process, temperature monitoring data showed that the protein degradation rate of salmon was 0.09 / h, while the theoretical calculated value was 0.08 / h, indicating that the theoretical model slightly underestimated the impact of temperature. Therefore, the least squares method was used to adjust the temperature-sensitive parameters to make them more in line with actual conditions. Similarly, error analysis and model optimization were performed on humidity, light, air pressure, and vibration factors to improve calculation accuracy.

[0071] Step S28: Organize the permeability influence coefficient into a 5×4 order permeability coefficient matrix, where each element of the permeability coefficient matrix represents the permeability influence coefficient of a specific environmental factor on a specific cargo attribute.

[0072] The embodiment of the present invention organizes the optimized permeability influence coefficient into a 5×4 order permeability coefficient matrix. The rows of the matrix represent environmental factors (temperature, humidity, light, air pressure, vibration), and the columns represent cargo characteristics (internal structure stability index, surface activity index, heat exchange coefficient, gas phase exchange coefficient). Each element in the matrix represents the degree of permeability influence of a certain environmental factor on a certain characteristic. For example, the (1,1) element in the matrix = 0.85, indicating that temperature has a greater impact on internal structure stability; the (2,3) element = 0.92, indicating that humidity has a smaller impact on the heat exchange coefficient. Ultimately, the matrix can be used to predict the impact of different environmental factors on salmon quality, thereby optimizing cold chain transportation strategies.

[0073] This invention retrieves benchmark material property data from a pre-set material library to accurately match the characteristic parameters of the current cargo type, ensuring that the cargo's changing trends under different environmental conditions can be effectively predicted. The benchmark material property data covers the cargo's internal structural stability index, surface activity index, heat exchange coefficient, and gas exchange coefficient, forming a basic standard for measuring cargo stability. After standardization, multidimensional environmental parameter field data is transformed into an environmental factor intensity matrix containing numerical representations of five environmental factors: temperature, humidity, light, air pressure, and vibration. This provides high-precision environmental parameter input for subsequent analysis. Combined with biochemical indicator change data, a response relationship model between environmental factors and biochemical indicators is constructed, enabling a quantitative link between cargo quality changes and external environmental conditions. Based on the benchmark material property data, the initial state values ​​and stability ranges of the cargo characteristic parameters are determined, and a benchmark model is then established to ensure that the cargo characteristic parameters under standard environmental conditions have referenced standard values. The Arrhenius equation principle is incorporated to construct a kinetic framework for the influence of environmental factors. The effects of five environmental factors, namely temperature, humidity, light, air pressure, and vibration, on cargo properties are quantified as reaction kinetic parameters, providing a clear mathematical description of the impact of environmental changes on cargo. By calculating the theoretical impact rate of each environmental factor on the four benchmark properties, theoretical penetration impact data are obtained, providing a basis for further optimizing cargo quality predictions. The theoretical penetration impact data are calibrated and optimized with the actual impact data predicted by the response relationship model to eliminate potential calculation biases and ensure a more accurate quantitative description of environmental impacts. Ultimately, all penetration impact coefficients are integrated into a 5×4 order penetration coefficient matrix, in which each matrix element accurately represents the degree of influence of a specific environmental factor on a specific cargo attribute, allowing the interaction rate between cargo and the environment to be accurately quantified. This method constructs a comprehensive mathematical model that converts the impact of environmental factors on cargo characteristics into a calculable and optimizable numerical expression, allowing the dynamic changes in cargo quality to be more accurately predicted, providing a scientific basis for the formulation of subsequent microenvironmental regulation and quality maintenance strategies.

[0074] Preferably, step S23 includes the following steps:

[0075] Step S231: performing time series analysis on the biochemical indicator change data, extracting the periodic change pattern, trend component and noise component therein, and obtaining the biochemical indicator time series feature data;

[0076] In order to analyze the changing patterns of biochemical indicators during the cold chain transportation of deep-sea salmon, the present invention first obtains biochemical indicator data collected at different time points during transportation, such as tissue protein degradation rate (unit: h-1), lipid oxidation index (dimensionless), and microbial growth rate (unit: cfu / g / h), and constructs a time series dataset. A sliding window smoothing method is used to eliminate high-frequency noise. Then, the empirical mode decomposition (EMD) method is used to decompose the time series to obtain high-frequency noise components, periodic change patterns, and trend components. For example, during a cold chain transportation process, the tissue protein degradation rate is stable at 0.02 / h at an initial temperature of 0°C. After 12 hours of transportation, the temperature rises to 5°C, and the degradation rate rapidly increases to 0.08 / h. During the subsequent periodic temperature fluctuations, the degradation rate exhibits periodic fluctuations of 0.06 / h to 0.09 / h. This trend can be extracted and quantified using empirical mode decomposition. At the same time, an autoregressive moving average model (ARMA) is used for fitting to predict the changing trends of biochemical indicators over a period of time. Finally, the time series characteristic data of biochemical indicators were obtained, including periodic components (such as protein degradation fluctuations with a cycle of 6 hours), trend components (overall linear increase) and random noise components (small fluctuations of about 0.005 / h).

[0077] Step S232: performing time alignment processing on the standardized environmental factor intensity matrix according to the biochemical indicator time series characteristic data, thereby obtaining time-aligned environmental factor data;

[0078] In the embodiment of the present invention, since the change of environmental factors has a lag effect on biochemical indicators, it is necessary to perform time alignment processing on the environmental factor data. First, the temperature, humidity, light, air pressure and vibration data recorded during transportation are compared with the sampling time points of the biochemical indicators, and the dynamic time warping (DTW) algorithm is used to calculate the optimal alignment path. For example, the temperature is stable at 0°C during the first 6 hours of transportation and then rises to 5°C, while the obvious change in tissue protein degradation rate occurs about 3 hours after the temperature change, so the temperature data needs to be shifted back 3 hours for alignment. Similarly, through DTW analysis, it was found that the lag time of humidity change is 2 hours, and the effect of vibration on protein degradation rate is almost real-time, so no adjustment is required. After time alignment processing, time-aligned environmental factor data are obtained to ensure that the environmental factor data at each time point can accurately correspond to the changes in biochemical indicators affected by it.

[0079] Step S233: using a multivariate statistical analysis method to calculate the correlation matrix between the time-aligned environmental factor data and the biochemical indicator time series feature data, and extracting environmental factor-biochemical indicator pairs;

[0080] In order to analyze the influence of different environmental factors on biochemical indicators, the embodiment of the present invention uses Pearson correlation coefficient, partial correlation analysis and canonical correlation analysis to calculate the correlation matrix of time-aligned environmental factor data and biochemical indicator time series feature data. Specifically, the values ​​of environmental factors (such as temperature, humidity, light, air pressure, vibration) at different time points are extracted, and the correlation coefficients between them and biochemical indicators (such as tissue protein degradation rate, lipid oxidation index, and microbial growth rate) are calculated. For example, the correlation coefficient between temperature and protein degradation rate is calculated to be 0.87, the correlation coefficient between humidity and lipid oxidation index is 0.76, and the correlation coefficient between vibration and microbial growth rate is 0.92, which indicates that there is a strong linear relationship between the three. In addition, after using partial correlation analysis to eliminate the influence of temperature on humidity, it was found that the independent contribution of humidity to lipid oxidation was reduced to 0.58, indicating that there may be a certain interaction between temperature and humidity on the lipid oxidation index. Finally, based on these correlation analysis results, environmental factor-biochemical indicator pairs such as (temperature-protein degradation rate), (humidity-lipid oxidation index), and (vibration-microbial growth rate) are extracted for subsequent modeling.

[0081] Step S234: establishing a single factor response function of each environmental factor corresponding to the biochemical index based on nonlinear regression analysis according to the environmental factor-biochemical index, thereby obtaining a single factor response model set;

[0082] The embodiment of the present invention constructs a single factor response function for each pair of environmental factor-biochemical index. First, the extracted environmental factor-biochemical index pairs are fitted with data. Since most biochemical processes have exponential growth or exponential decay characteristics, nonlinear regression analysis is used for fitting. For example, the (temperature-protein degradation rate) relationship is analyzed, and after observing the data distribution, an exponential growth function is selected for fitting. It is found that the relationship between protein degradation rate and temperature can be expressed as follows: when the temperature rises from 0°C to 5°C, the protein degradation rate increases from 0.02 / h to 0.08 / h, showing a nonlinear upward trend. Finally, the response function is obtained by exponential fitting: for every 1°C increase in temperature, the protein degradation rate increases by about 20%. Similarly, a power function is fitted to the (humidity-lipid oxidation index) relationship. It is found that when the humidity rises from 70%RH to 90%RH, the lipid oxidation index increases from 0.4 to 0.9. The fitted power function shows that for every 10%RH increase in humidity, the lipid oxidation index increases by about 30%. Logistic regression analysis of the vibration-microbial growth rate relationship revealed that when the vibration intensity increased from 0.2 g to 1.0 g, the microbial growth rate increased from 100 cfu / g / h to 1200 cfu / g / h, exhibiting an S-shaped growth curve. Ultimately, all single-factor response models were stored as a single-factor response model set for subsequent multifactor interaction modeling.

[0083] Step S235: constructing a response relationship model of the interaction of multiple environmental factors based on the single-factor response model set, wherein the response relationship model is used to characterize the changing pattern of the biochemical characteristics of the cargo under different combinations of environmental conditions.

[0084] In the embodiment of the present invention, since multiple environmental factors may jointly affect the same biochemical index, it is necessary to construct an interaction model of multiple environmental factors. First, based on the single-factor response model set, the multivariate nonlinear regression method is adopted to calculate the influence of the combination of multiple factors on the biochemical index. For example, when analyzing the comprehensive influence of temperature, humidity and light on the protein degradation rate, it is found that the temperature and protein degradation rate are exponentially related, and the influence of humidity is enhanced when the temperature is higher. Therefore, a quadratic interaction model is established, in which the influence weight of temperature is 0.6, the influence weight of humidity is 0.3, and the influence weight of light is 0.1. In addition, in order to take into account nonlinear interactions, support vector regression (SVR) and Bayesian network are used for modeling to calculate the changes in biochemical indicators under different combinations of environmental factors. For example, in an experimental simulation, when the temperature was set to 3°C, the humidity was set to 80% RH, and the vibration was set to 0.5g, the model predicted a protein degradation rate of 0.045 / h, a lipid oxidation index of 0.65, and a microbial growth rate of 600 cfu / g / h. The actual measured values ​​were 0.048 / h, 0.63, and 590 cfu / g / h, respectively. The errors were all within 5%, indicating that the model has high prediction accuracy. Ultimately, this multi-environmental factor response relationship model can be used to predict changes in biochemical characteristics under different transportation environments and provide a reference for optimizing cold chain transportation solutions.

[0085] The present invention performs time series analysis on biochemical indicator change data, extracts the periodic change pattern, trend component and noise component therein, and ensures that the dynamic evolution characteristics of the biochemical indicators are accurately portrayed. In this way, the evolution law of the biochemical characteristics of the goods under different storage conditions can be identified, providing a reliable data basis for subsequent analysis. Subsequently, based on the biochemical indicator time series characteristic data, the standardized environmental factor intensity matrix is ​​time-aligned so that the environmental factor data and the biochemical change data are consistent in the time dimension, thereby ensuring the timeliness and accuracy of the data and improving the effectiveness of multivariate analysis. On this basis, a multivariate statistical analysis method is adopted to calculate the correlation matrix between the environmental factor data and the biochemical indicator time series characteristic data after time alignment, and extract the environmental factor-biochemical indicator pairs to clarify the degree of influence of different environmental factors on the biochemical characteristics. Subsequently, for each environmental factor-biochemical indicator pair, a nonlinear regression analysis method is used to establish a single factor response function to form a single factor response model set so that the influence law of each environmental factor on the biochemical indicator can be quantitatively described. Finally, based on the set of single-factor response models, a response relationship model for the interaction of multiple environmental factors was further constructed. This model is used to characterize the changing trends of the biochemical properties of goods under different combinations of environmental conditions. This method can accurately characterize the influence of environmental factors on the biochemical properties of goods and quantify the interactions between these factors, making the prediction of changes in goods quality more accurate and providing a scientific basis for environmental optimization and regulation. Furthermore, this model can be used to simulate the biochemical evolution of goods under different storage conditions, helping to optimize storage plans, extend the shelf life of goods, and reduce the risk of quality deterioration due to environmental changes.

[0086] Preferably, the calculation of the current environmental impact intensity in step S3 specifically includes:

[0087] Perform matrix multiplication on the permeability coefficient matrix and the multidimensional environmental parameter field data to calculate the environment-cargo interaction intensity vector;

[0088] Calculate the comprehensive intensity index of the overall impact of the environment on the goods based on the environment-goods interaction intensity vector;

[0089] Calculate the distribution of environmental impact intensity in different regions and time points by using the comprehensive intensity index to obtain an environmental impact intensity distribution map;

[0090] Based on the environmental impact intensity distribution map, environmental impact hotspots and key time windows are identified to obtain key environmental impact node data;

[0091] The current environmental impact intensity data is generated based on the environmental impact key node data and the environment-goods interaction intensity vector.

[0092] The present invention performs time series analysis on biochemical indicator variation data. First, biochemical indicator data over a specific time span must be collected, such as the pH value, dissolved oxygen, and microbial concentration of a food or drug. This data is typically recorded on an hourly or daily basis. During implementation, the data is preprocessed, including missing value filling, outlier removal, and data smoothing. Time series decomposition methods, such as empirical mode decomposition (EMD) or Hodrick-Prescott filtering, are then used to separate the cyclical variation patterns, long-term trend components, and noise components in the biochemical indicator data. The cyclical variation patterns can be used to identify regular fluctuations in biochemical indicators under environmental influences, the trend components can be used to analyze long-term variation trends, and the noise components can be used to assess the impact of random disturbances. Finally, these feature data are extracted and stored as a time series feature dataset. Environmental factor data related to the biochemical indicators, such as temperature, humidity, gas concentrations (oxygen, carbon dioxide), and light intensity, are collected, and the sampling frequency of the environmental data is ensured to match the biochemical indicator data. During implementation, the dynamic time warping (DTW) algorithm was used to time-align the temporal feature data of biochemical indicators and the environmental factor intensity matrix, ensuring one-to-one correspondence between data at the same time point. Furthermore, linear interpolation or spline interpolation methods were used to complement unevenly sampled data to ensure the integrity of the time series data. Ultimately, time-aligned environmental factor data were generated, providing a foundation for subsequent analysis. Methods such as the Pearson correlation coefficient, Spearman rank correlation coefficient, or mutual information analysis were used to calculate the correlation between each environmental factor and biochemical indicator, and a correlation matrix was constructed. During implementation, a threshold (e.g., an absolute value of the correlation coefficient greater than 0.5) was set to screen for significantly correlated environmental factor-biochemical indicator pairs. Furthermore, to further analyze potential relationships between variables, principal component analysis (PCA) or factor analysis (FA) was used to reduce the data dimensionality, remove redundant information, and extract key environmental factor-biochemical indicator pairs to provide input data for subsequent modeling. Typical nonlinear regression models, such as the logistic regression model, exponential decay model, and hyperbolic model, were selected to accommodate the influence of different environmental factors on biochemical indicators. During the implementation process, the least squares method or Bayesian optimization method was used to fit the relationship between each environmental factor and the corresponding biochemical index, and the goodness of fit (R 2) to evaluate the accuracy of the model. Specific application scenarios include: for the temperature-enzyme activity relationship, the Arrhenius equation can be used to model it; for humidity-food moisture content changes, the exponential decay model can be used to describe the moisture loss rate. Ultimately, the established set of single-factor response models will be used to further analyze the combined effects of multiple environmental factors. Using generalized additive models (GAMs), random forest regression (RF), or neural network-based regression models, multiple environmental factors are used as independent variables and biochemical indicators as dependent variables to train multi-factor interaction models. Specific steps include: first, selecting the output values ​​of multiple single-factor models as input features, and optimizing the weight coefficients between different environmental factors using grid search methods or genetic algorithms; second, cross-validation methods are used to evaluate the model's generalization ability, and simulations are conducted under multiple environmental scenarios to verify the model, such as simulating the biochemical changes of food under different storage conditions. Ultimately, the constructed response relationship model can be used to predict the impact of different environmental combinations on the biochemical properties of goods, providing decision support for environmental control. A permeability coefficient matrix is ​​constructed to characterize the effects of different environmental factors on cargo. Each row of the matrix represents a specific cargo type, and each column represents an environmental factor, such as temperature, humidity, and air pressure. The matrix elements represent the intensity of each environmental factor's impact on the cargo's biochemical properties. During implementation, permeability coefficients are extracted from experimental measurements or historical data, and the data is supplemented using numerical interpolation. Next, multidimensional environmental parameter field data, such as time series data on temperature and humidity measured at different storage locations, is obtained as the input matrix. Matrix multiplication is performed with the permeability coefficient matrix to obtain the environmental-cargo interaction intensity vector for each cargo under different environmental conditions, representing the degree of environmental influence on the cargo's biochemical properties. A weighted summation method is used to weight the impact of each environmental factor according to its relative importance. For example, assuming that the weight of temperature on cargo deterioration is 0.4, humidity is 0.3, and light intensity is 0.3, the comprehensive intensity index is calculated as the sum of the intensity values ​​of each environmental factor multiplied by the corresponding weight. Furthermore, a normalization method can be used to map the comprehensive intensity index to a range of 0–1 to facilitate comparison across different cargoes. During the implementation process, typical goods, such as fresh food or chemical reagents, are selected to calculate their comprehensive intensity index under different storage environments, and the calculation results are verified to be consistent with the actual situation. Environmental data for each storage area and time point are collected, and the corresponding comprehensive intensity index is calculated. During the implementation process, a geographic information system (GIS) or interpolation algorithm (such as Kriging interpolation or inverse distance weighted interpolation) is used to visualize the comprehensive intensity index at different spatial locations to generate an environmental impact intensity distribution map. Specific applications include: in food cold chain logistics, the impact of temperature and humidity on the shelf life of food is calculated for different cold storage areas, and an environmental impact distribution map is drawn to identify areas with greater environmental impact.Set a high threshold for the intensity of environmental impact, for example, define areas greater than 0.8 as hot spots, and use cluster analysis methods (such as DBSCAN or K-means) to identify hot spots in spatial distribution. During implementation, for a certain food storage system, you can analyze storage areas with large temperature and humidity anomalies, and combine historical data to identify key time windows, such as high temperature periods in summer. Ultimately, you can obtain data on key nodes of environmental impact, including the coordinates of hot spot areas, key time points, and the degree of their impact. Specific applications include: in a pharmaceutical storage system, by combining the environmental data of the past month with the currently measured temperature and humidity data, calculate the intensity of the current impact of the pharmaceutical on the environment, and provide environmental control recommendations, such as adjusting the storage temperature or humidity to reduce environmental impacts.

[0093] By performing matrix multiplication on the permeability coefficient matrix and multidimensional environmental parameter field data, the present invention can convert complex environmental impact factors into a quantifiable environment-cargo interaction intensity vector, thereby accurately describing the combined effects of environmental factors on various cargo properties. This vector not only reflects the combined impact of different environmental factors but also ensures that the dynamic changes of environmental variables are effectively captured during the analysis process. Subsequently, a comprehensive intensity index of the overall environmental impact of the cargo is calculated based on the environment-cargo interaction intensity vector, thereby obtaining the overall stability of the cargo under current storage or transportation conditions and providing a quantitative basis for environmental adaptability assessment. Based on this, by calculating the distribution of the comprehensive intensity index at different regions and time points, an environmental impact intensity distribution map can be generated, which intuitively shows the degree of impact of environmental factors on the cargo status under different storage or transportation conditions, facilitating the analysis of potential environmental risks. Further, using the environmental impact intensity distribution map, hotspots and critical time windows of environmental impact can be accurately identified, thereby determining key environmental impact node data. This data can help identify situations where cargo properties change significantly at specific time points or spatial regions, providing a reference for environmental optimization strategies. Finally, based on the environmental impact key node data and the environment-cargo interaction intensity vector, current environmental impact intensity data is generated, enabling dynamic monitoring and accurate assessment of environmental impacts. This method not only improves the accuracy of environmental impact assessments but also provides data support for environmental management during cargo storage and transportation, enabling managers to promptly adjust environmental control strategies to ensure the stability of cargo quality. Furthermore, this method is applicable to different types of cargo, and by accurately characterizing environmental impacts, it provides scientific guidance for quality management under various storage and transportation conditions.

[0094] Preferably, the thermodynamic coupling analysis in step S3 specifically includes:

[0095] Draw a corresponding relationship curve between environmental factors and biochemical responses of goods based on environmental impact intensity data and biochemical indicator change data;

[0096] According to the corresponding relationship curve, the changing pattern of the internal order of the goods under environmental stress is quantified, thereby constructing an entropy function model of the goods quality status;

[0097] Substitute the biochemical index change data into the entropy function model to calculate the initial value of the entropy growth rate of the goods under the current environmental conditions;

[0098] Establish an acceleration effect model of environment-matter interaction based on the permeability coefficient matrix;

[0099] The amplification factor of environmental fluctuations on the entropy growth rate is calculated based on the acceleration effect model. The amplification factor is calculated by detecting the fluctuation of environmental parameters in the acceleration effect model. When the fluctuation amplitude exceeds 3 times the mean standard deviation within 15 minutes, the instantaneous fluctuation impact value is calculated and the amplification factor is increased by 0.3. The humidity change rate is monitored. When it exceeds 5% RH / min, the humidity mutation impact coefficient is calculated and the amplification factor is further increased by 0.2.

[0100] Multiply the initial value of the entropy growth rate by the amplification coefficient to obtain the quality entropy growth rate data that represents the rate of increase of disorder in the quality state of the goods.

[0101] In this embodiment of the present invention, typical goods, such as fresh food or chemicals, are selected in a laboratory environment and placed in a controlled environment chamber. Multiple environmental variables (such as temperature, humidity, and oxygen concentration) are set, and real-time data under different environmental conditions is recorded using precise sensors. Simultaneously, key biochemical indicators of the goods (such as redox potential, pH, and protein degradation rate) are collected to form a time series dataset. Data fitting methods, such as multivariate regression analysis or machine learning regression models (such as random forest regression and LSTM), are used to establish a functional relationship between environmental factors and the biochemical responses of the goods, and a curve is plotted. For example, in aquatic product preservation research, the effect of temperature fluctuations between 0°C and 20°C on the ATP decomposition rate in fish muscle tissue can be recorded, and a nonlinear relationship curve between ambient temperature and ATP degradation rate can be plotted. Based on the second law of thermodynamics, the evolution of the quality state of a product can be described by changes in entropy. First, the trends of biochemical indicators with environmental changes, such as the degree of protein denaturation and microbial growth rate, are analyzed, and these indicators are normalized to ensure data comparability. Then, information entropy calculation methods, such as Shannon entropy or Kolmogorov entropy, are used to define the quality state entropy of the goods. An entropy function model is constructed using experimental data. For example, the order of the initial quality state is set to the maximum, and the quality state gradually evolves under environmental stress. A curve showing the change in entropy over time is fitted. For example, in fruit and vegetable storage research, changes in vitamin C content and polyphenol oxidase activity can be used to construct an entropy function reflecting the quality decay of fruits and vegetables. The validity of this model can then be verified using field data. Based on the entropy function model, biochemical indicator data at a specific moment are selected to calculate the entropy value of the product's quality state at that moment. The initial value of the entropy growth rate per unit time is then calculated using numerical differentiation. Assuming the entropy value is a function of the product's quality over time, the forward difference or five-point difference method can be used to calculate its rate of change over time. For example, in the preservation of fresh meat, volatile basic nitrogen (TVB-N) content is selected as a quality indicator. The entropy growth rate is calculated based on experimentally measured TVB-N growth data to determine the rate of quality degradation under the current environment. A permeability matrix is ​​used to analyze the impact of environmental factors (such as temperature, humidity, and gas concentration) on changes in the internal material of the product, establishing an acceleration effect model for environmental-material interactions. First, based on the experimentally measured diffusion coefficients of goods under different environmental conditions, the impact of environmental variables on material penetration is calculated. For example, during the food drying process, the water loss rate is calculated using Fick's diffusion law, and the functional relationship between temperature, humidity, and water diffusion rate is established in combination with experimental data to form an environment-material interaction model. Based on the acceleration effect model, the degree of influence of different environmental variables (such as sudden temperature changes and drastic fluctuations in humidity) on the entropy growth rate is analyzed, and the amplification coefficient of environmental fluctuations on the entropy growth rate is calculated. Sensitivity analysis methods, such as local sensitivity analysis or the Sobol method, are used to evaluate the impact of changes in environmental parameters on the entropy growth rate.For example, in cold chain logistics research, the situation where the temperature of the refrigerated compartment fluctuates from -2°C to 4°C is simulated, and the amplification effect of temperature changes on the entropy growth rate of dairy product quality is calculated to quantify its accelerated deterioration effect. The initial value of the entropy growth rate is multiplied by the amplification coefficient to obtain the final quality entropy growth rate data. This step requires quantifying the impact of environmental fluctuations on the entropy growth rate, combining it with the initial entropy growth rate, and calculating the disorder growth rate of the goods under the current environmental conditions. For example, in drug stability research, the decomposition rate data of specific drugs are selected, and the impact of the storage environment (temperature, light) on its stability is combined to calculate the quality entropy increase rate of the drug under different environmental conditions, and predict its shelf life change trend.

[0102] By plotting the corresponding relationship curves between environmental factors and the biochemical responses of goods, this method can intuitively demonstrate the impact of different environmental factors on the biochemical state of goods, clarifying how environmental changes affect the evolution of goods quality. Quantitative analysis of this correspondence curve using the second law of thermodynamics establishes a mathematical relationship between the internal order of goods and changes in environmental stress, thereby constructing an entropy function model that allows the goods' quality state to be described using thermodynamic parameters. Substituting biochemical indicator change data into the entropy function model, the initial value of the entropy growth rate of the goods under current environmental conditions can be calculated, providing basic data for assessing quality change trends. Furthermore, a model of the accelerating effect of environmental-material interactions is established using a permeability coefficient matrix, quantifying the exacerbating effect of different environmental factors on changes in goods' quality. Based on this model, the amplification coefficient of environmental fluctuations on the entropy growth rate is calculated, accurately measuring how drastic environmental changes accelerate the deterioration of goods' quality, providing key data support for environmental optimization and risk prediction. Finally, by multiplying the initial entropy growth rate by the amplification coefficient, the quality entropy increase rate data, representing the rate of increase in disorder in the goods' quality state, is obtained, allowing the changing trends of goods' quality to be precisely quantified using the principle of entropy increase. This method can effectively identify the evolution of cargo quality stability under different environmental conditions, providing a scientific basis for the dynamic adjustment of environmental parameters during storage and transportation, ensuring the long-term stability of cargo quality. Furthermore, the method is adaptable to different types of cargo, providing a systematic analysis method based on thermodynamic theory, making environmental management and quality control more targeted and refined.

[0103] Preferably, the generation of the quality status prediction model in step S3 specifically includes:

[0104] Perform time series analysis on the quality entropy increase rate data to extract the trend pattern and periodic characteristics of quality attenuation, thereby obtaining the quality change trend characteristics;

[0105] Pattern matching is performed between the quality change trend characteristics and the pre-acquired historical quality change database to screen out similar quality decay paths and obtain reference decay pattern data;

[0106] Based on the reference decay pattern data and the quality entropy increase rate data, the time evolution function of the goods quality status is constructed to form a basic model for quality status prediction;

[0107] Input the environmental impact intensity data into the quality status prediction basic model, calculate the quality status prediction values ​​at different time points, and obtain the quality change prediction result data;

[0108] The biochemical index change data is used to make real-time corrections to the quality change prediction result data, and a quality status prediction model is constructed to make real-time predictions on the quality change trend of goods.

[0109] The embodiment of the present invention performs time series analysis on the quality entropy increase rate data, adopts methods such as moving average, exponential smoothing or wavelet transform to eliminate noise, and extracts the long-term trend and periodic change characteristics of the data. The time series is decomposed by methods such as ARIMA (autoregressive integrated moving average model) and STL (seasonal trend decomposition), and the trend pattern of quality decay is identified, such as linear decay, exponential decay or step-by-step change, and the quality change law in different cycles is analyzed at the same time. For example, in the quality analysis of refrigerated fruits and vegetables, the dissolved solids content (SSC) and respiration rate data of the past month can be used to identify the trend pattern of fruit and vegetable quality decline, and to find the quality periodic fluctuations that may exist under low temperature storage, such as accelerated decay every 7 days. The extracted quality change trend characteristics are pattern matched with the data in the historical quality change database to screen out similar quality decay paths. Methods such as DTW (dynamic time warping), cosine similarity or K-means clustering are used to compare the current quality change trend with the decay pattern in the historical data, and select several paths with the highest similarity as reference decay pattern data. For example, in shelf life prediction for dairy products, the growth rate of total bacterial counts and lactose degradation rates over the past three months can be compared with the decay paths of historical storage batches to identify the decay pattern that best matches the current quality trend of the dairy product, providing a reference for subsequent predictions. Based on the reference decay pattern data and the current quality entropy increase rate data, a time-evolution function of the product's quality status is constructed, forming a basic model for quality status prediction. First, an appropriate mathematical modeling method, such as logistic regression, exponential decay model, or LSTM neural network, is selected to fit the historical quality decay data to obtain a functional expression for the time-evolution of the product's quality status. Then, the current entropy increase rate data is input into the model, and the parameters are adjusted to accurately describe the current quality status changes. For example, in the storage prediction of aquatic products, a water loss model based on the Fick diffusion principle is used in combination with historical data to fit the time-evolution function of the water loss rate of the aquatic product. The current entropy increase rate data is then used to adjust the model parameters to adapt to the current storage environment. Environmental impact intensity data is input into the basic quality status prediction model, and predicted quality status values ​​at different future time points are calculated to obtain quality change prediction results. During implementation, factors such as ambient temperature, humidity, and gas composition are first converted into numerical values ​​and input into the model. By simulating environmental changes over a period of time, the corresponding quality change trends are calculated. For example, in drug stability studies, predicted temperature and humidity changes over the next seven days are input. Combined with the current drug's oxidative degradation rate, the residual rate of the active ingredient over the next week is calculated to predict the drug's quality change trend and ensure that it meets quality standards within its shelf life.In order to improve the prediction accuracy, the quality change prediction result data is dynamically corrected using the real-time monitored biochemical indicator change data, and a quality status prediction model is constructed that can predict the quality change trend of goods in real time. Using Kalman filtering, Bayesian updating or adaptive error correction methods, the latest biochemical indicator data (such as pH value, microbial growth rate, etc.) are input into the model, and the deviation of the prediction curve is adjusted to keep it consistent with the actual quality change trend. For example, during the cold chain meat transportation process, the TVB-N content on the meat surface is tested every 12 hours and input into the prediction model. The prediction curve is adjusted so that it can accurately reflect the real-time changes in meat quality, thereby providing accurate support for cold chain logistics decision-making.

[0110] By performing time series analysis on quality entropy increase rate data, the present invention can extract trend patterns and cyclical characteristics of goods quality degradation, thereby accurately identifying the long-term evolution of quality changes. This method can effectively separate short-term fluctuations from long-term trends, making quality evolution predictions more stable and reliable. Subsequently, the extracted quality change trend characteristics are pattern-matched with a historical quality change database to screen for similar quality degradation paths, forming reference degradation pattern data. This process can provide multiple possible quality evolution trajectories based on large-scale historical data, improving the accuracy and applicability of predictions. Combining the reference degradation pattern data with the current quality entropy increase rate data, a time evolution function of the goods quality status can be constructed, forming a basic model for quality status prediction, providing a more theoretical basis for deducing quality degradation trends. Furthermore, by inputting environmental impact intensity data into this basic model, the goods quality status at different time points can be calculated, resulting in a complete quality change prediction data set, allowing for a quantitative estimation of the future state of goods quality under different environmental conditions. This method not only predicts quality change trends but also identifies the degree of influence of different environmental factors on quality evolution, providing data support for environmental optimization. Furthermore, using real-time biochemical indicator change data to dynamically calibrate quality change predictions significantly improves the real-time performance and accuracy of the prediction model, making the predictions more closely aligned with actual changes in cargo quality. Ultimately, a quality status prediction model capable of predicting cargo quality trends in real time has been constructed, enabling proactive adjustments to cargo quality management based on scientific predictive analysis. This optimizes storage and transportation strategies, reduces the risk of quality loss, and enhances the intelligence of supply chain management.

[0111] Preferably, the microenvironment compensation parameter combination analysis in step S5 specifically includes:

[0112] According to the quality status prediction model, the correlation analysis between the quality entropy increase rate data and environmental factors was carried out to obtain the data of the dominant environmental stress factors;

[0113] Calculate the reverse adjustment parameters based on adverse effects on the dominant environmental stress factor data to obtain the environmental factor balance adjustment data;

[0114] Conduct feasibility analysis on environmental factor balance regulation data, optimize parameters within the actual regulation range, and generate feasible candidate sets of regulation parameters;

[0115] The quality status prediction model is used to simulate and verify the candidate set of control parameters, and the protection effect of different parameter combinations is evaluated to obtain protection effect evaluation data;

[0116] According to the protection effect evaluation data, a comprehensive score based on energy consumption and physical feasibility is performed to obtain a combination of microenvironment compensation parameters.

[0117] Based on the identified dominant environmental stress factor data, the present invention utilizes the Le Chatelier equilibrium principle to calculate inverse adjustment parameters for these environmental factors. In practice, the Le Chatelier equilibrium principle can be used to describe how, when a system is subjected to external stress, it tends to react and adjust itself to mitigate the external influence. For example, assuming temperature and humidity are dominant factors, if excessively high temperatures cause quality degradation during fruit storage, cooling is necessary to correct the situation. In this process, environmental factor balance adjustment data is generated by calculating inverse adjustment parameters for temperature and humidity control (such as the cooling rate and humidity control range). For example, through model calculation, under high summer storage temperatures, by setting the temperature adjustment range to within 10°C and combining the upper and lower limits of humidity control, adjustment parameters for different environmental stress factors are obtained. Based on the environmental factor balance adjustment data, a feasibility analysis is performed to ensure that the control range is achievable in actual application and to optimize the relevant parameters. In this process, it is first necessary to limit each adjustment parameter based on the actual capacity of the environmental facilities (such as air conditioners, humidifiers, dehumidifiers, etc.) to ensure that the adjustment plan is within a feasible range. For example, when regulating the storage environment for pharmaceuticals, the feasibility of the control parameters is calculated based on the temperature and humidity requirements of the pharmaceuticals to ensure that the equipment can complete the necessary temperature and humidity adjustments without exceeding power limits. If certain control parameters are too demanding and may exceed the performance limits of the equipment, they need to be optimized and adjusted to select parameters that can be achieved under the equipment and environmental conditions. This ultimately results in a candidate set of optimized, implementable control parameters. This candidate set of implementable control parameters is simulated and verified using a quality status prediction model. By simulating different control parameter combinations, their protective effectiveness is evaluated. Specifically, the optimized control parameters (such as the adjusted temperature and humidity range) are first input into the quality status prediction model. Multiple simulations are then performed, combined with real-time biochemical data and entropy increase rate data, to predict the resulting changes in product quality under different parameter combinations. For example, in a dairy product storage scenario, the effects of different temperature and humidity control combinations on dairy product quality (such as lactose degradation rate and bacterial growth) are simulated. The quality preservation effect of different control combinations is evaluated, thereby generating protective effectiveness evaluation data to identify which parameter combinations can optimize quality while reducing the negative impact of environmental fluctuations on quality. Based on the protection effect evaluation data, a comprehensive score based on energy consumption and physical feasibility is performed, and the optimal microenvironment compensation parameter combination is finally selected. This process first requires combining the protection effect evaluation data with actual conditions such as energy consumption and the feasibility of physical facilities. For example, when adjusting the temperature and humidity combination, it is necessary to consider the energy consumption of each parameter combination (such as air conditioning power, humidifier consumption, etc.), as well as factors such as the response time and stability of the equipment. Then, based on these factors, each parameter combination is scored.Scoring criteria can be comprehensively evaluated based on factors such as protection effectiveness (quality preservation), energy efficiency (lowest energy consumption solution), and equipment load (load stability). Ultimately, this comprehensive scoring process determines the optimal combination of microenvironmental compensation parameters, providing the best solution for quality preservation and energy management during actual cargo storage.

[0118] By analyzing the quality status prediction model, the present invention can identify the correlation between the quality entropy increase rate data and environmental factors, thereby extracting the dominant environmental stress factors that play a key role in quality changes. This process can accurately locate the core variables that affect the stability of goods quality and provide a scientific basis for subsequent regulation. On this basis, based on the Le Chatelier equilibrium principle, the dominant environmental stress factors are analyzed, and the reverse regulation parameters for adverse effects can be calculated, thereby generating environmental factor balance regulation data, so that environmental regulation has pertinence and directionality. Subsequently, a feasibility analysis is performed on the environmental factor balance regulation data, and parameter optimization is performed in combination with the regulation range of the actual application scenario. A candidate set of regulation parameters with actual operability can be screened out to ensure the feasibility and adaptability of the regulation scheme. In order to verify the effectiveness of the regulation scheme, the candidate set of regulation parameters is simulated and tested using the quality status prediction model. The influence of different parameter combinations on the stability of goods quality can be analyzed, and its protective effect can be quantified, thereby obtaining scientific protection effect evaluation data. Finally, based on the protection effect evaluation data, comprehensive consideration of energy consumption and physical feasibility is given to calculate the optimal combination of microenvironment compensation parameters, so that the control strategy can not only effectively improve the stability of cargo quality, but also take into account energy utilization efficiency and actual application needs, and ultimately achieve a coordinated improvement in quality assurance, resource optimization and economy.

[0119] Preferably, step S4 includes the following steps:

[0120] Step S41: Comparing and analyzing the microenvironment compensation parameter combination with the multi-dimensional environmental parameter field data, calculating the compensation difference of each environmental factor, and generating environmental compensation demand data;

[0121] The embodiment of the present invention needs to compare and analyze the microenvironment compensation parameter combination with the current multi-dimensional environmental parameter field data (such as temperature, humidity, oxygen concentration, etc.) to calculate the compensation difference. Specifically, it is assumed that the microenvironment compensation parameter combination has been calculated through the aforementioned steps, and it is assumed that the current environmental parameter field data is obtained by real-time monitoring. If the set temperature in the microenvironment compensation scheme is 20°C and the current actual temperature is 25°C, the temperature compensation difference is -5°C, and the humidity compensation difference may be 3%. Then, by calculating the compensation difference of each environmental factor, the environmental compensation demand data is obtained. This data shows the difference between the current environment and the ideal storage conditions, thereby providing a basis for subsequent control strategies.

[0122] Step S42: Design a micro-environment control strategy based on the environmental compensation demand data, and convert it into an instruction set executable by the device to obtain dynamic control instruction data. The micro-environment control strategy is specifically designed as follows:

[0123] When the temperature compensation difference exceeds ±3°C, the temperature quick response coefficient is calculated, the response priority of the temperature adjustment unit is set to the highest, and the temperature priority adjustment data is obtained;

[0124] Humidity gradient analysis is performed on areas where the humidity compensation difference exceeds ±15% RH to obtain the humidity distribution unevenness index;

[0125] The UV intensity in the calculated illumination compensation difference exceeds 3000 μW / cm 2 The exposure risk of UV light exposure is determined and the UV light exposure risk assessment data is obtained;

[0126] Detect the fluctuation characteristics of the air pressure compensation difference with a fluctuation frequency exceeding 0.5 Hz to obtain air pressure fluctuation characteristic data;

[0127] Perform spectrum decomposition on the shock waveform with peak acceleration exceeding 2g in the vibration compensation difference to obtain vibration spectrum characteristic data;

[0128] Generate a coordinated control strategy for environmental factors based on temperature priority adjustment data, humidity distribution unevenness index, UV exposure risk assessment data, air pressure fluctuation characteristics data, and vibration spectrum feature data;

[0129] The embodiment of the present invention designs a suitable micro-environment control strategy based on the environmental compensation demand data. First, it is necessary to determine the target value range that needs to be adjusted for each environmental factor (such as temperature, humidity, etc.). Assuming that the temperature in the current environment is too high and the humidity is low, and the compensation difference is a negative number, the control strategy may be to achieve the ideal environmental requirements by lowering the temperature and increasing the humidity. Then, these control strategies are converted into an instruction set that can be executed by the device. For example, instructions may include "start the air conditioner to 18°C", "increase the humidity to 70%", and include specific action parameters (such as adjustment rate, target value, etc.). Finally, these instructions are organized into dynamic control instruction data to form a set of control commands that can be executed immediately.

[0130] Step S43: Sending the dynamic control instruction data to the microenvironment control device, activating the corresponding environmental control unit, monitoring the operating status of the microenvironment control device and changes in environmental parameters, and recording real-time control effect data, wherein the environmental control unit includes a temperature control unit, a humidity control unit, a gas phase protection unit, and a physical barrier unit;

[0131] The dynamic control instruction data of the embodiment of the present invention will be sent to the control system of the microenvironment control device. Specifically, the system will activate the corresponding environmental adjustment units, such as the temperature adjustment unit, the humidity control unit, the gas phase protection unit and the physical barrier unit, etc. according to the instructions sent. For example, assuming that after the temperature adjustment unit is activated, the air-conditioning system starts working, the humidity control unit starts the humidifier, and the gas phase protection unit starts the oxygen removal system, etc. Next, the monitoring device will continue to track the operating status of the microenvironment control device (such as the power usage of the air conditioner, the humidity change of the humidifier, etc.) and the real-time changes of environmental parameters (temperature, humidity, gas concentration, etc.). Real-time data will continuously update and record the control effect, and finally form real-time control effect data, including changes in environmental parameters and equipment operation status.

[0132] Step S44: Collect real-time biochemical index data based on the real-time control effect data, evaluate the protective effect of microenvironment control on the quality of goods, and optimize the control parameters in real time based on the protection effect to obtain protection execution status data.

[0133] The real-time control effect data of the embodiment of the present invention will be combined with the real-time biochemical indicator data to evaluate the actual effect of micro-environment control on the protection of the quality of goods. First, the real-time biochemical indicator data is collected by sensors (for example, for the scene of storing fruits, the color change, sugar content, decay index and other biochemical indicators of the fruits can be collected), and compared with the real-time control effect data (such as the adjustment of temperature and humidity) to evaluate whether the control strategy is effective. For example, assuming that after the temperature and humidity reach the target value during the control process, the sugar content of the fruit remains stable, and the decay index does not rise further, indicating that the control measures are effective in protecting the quality. Based on the evaluation results, the micro-environment control strategy will be optimized in real time. Optimization can be achieved by adjusting the control parameters (for example, further refining the temperature control or humidity adjustment rate) to achieve a better protection effect. Finally, the protection execution status data is generated by the optimized control parameters to ensure that the goods can get the best environmental protection during the storage process.

[0134] By comparing and analyzing the combination of microenvironment compensation parameters with multi-dimensional environmental parameter field data, the present invention can accurately calculate the compensation difference of each environmental factor and then generate environmental compensation demand data. This process ensures that the control strategy can flexibly respond to specific environmental changes, accurately meet the environmental conditions required for the goods during transportation, and avoid unnecessary waste of resources or excessive control. Based on the environmental compensation demand data, the designed microenvironment control strategy can make detailed adjustments to different environmental factors and convert these strategies into a set of instructions executable by the device, ensuring that the control system can efficiently and accurately perform the corresponding operations and ensure the real-time and effectiveness of the control strategy. By sending dynamic control instructions to the microenvironment control device and activating environmental control units such as temperature control, humidity control, gas phase protection and physical barrier units, multi-dimensional precise adjustment can be performed in the actual environment according to the characteristics of the goods. In this process, the operating status of the microenvironment control device and the changes in environmental parameters are monitored in real time, so that the control strategy can be adjusted in time and real-time control effect data can be recorded, thereby dynamically tracking and evaluating the control effect. Combined with real-time control effect data, the collected real-time biochemical indicator data can comprehensively evaluate the protective effect of microenvironmental control on cargo quality, providing a scientific basis for real-time optimization of control parameters. By continuously optimizing the control strategy, it is possible to accurately control the changing trends of cargo quality, thereby ensuring that the cargo is always maintained at the optimal quality during transportation, while also achieving efficiency and resource conservation in the control process. Ultimately, the protection execution status data obtained not only effectively guarantees cargo quality but also enables dynamic adjustment of the control plan according to actual needs, enhancing the adaptability and reliability of the system.

[0135] Preferably, step S5 includes the following steps:

[0136] Step S51: Extract features from the protection execution status data to identify the triggering time, execution effect, and parameter changes of key protection measures, thereby obtaining key cargo protection event data;

[0137] The embodiment of the present invention identifies key protection measures and their corresponding triggering times by analyzing the time series data in the protection execution status data. For example, the temperature adjustment unit may start when the temperature reaches a certain preset value, and the humidity control unit may start working when the humidity is lower than a certain threshold. By recording these triggering times, the start time of each protection measure can be determined. Then, the execution effect of each protection measure is analyzed to determine whether the expected environmental condition adjustment target has been achieved, and the corresponding environmental parameter changes (such as temperature, humidity, air pressure, etc.) are recorded. Through these data, key event data in the cargo protection process can be obtained, including the triggering time, execution effect and corresponding parameter change data of each key protection measure, forming cargo protection key event data.

[0138] Step S52: Encoding the cargo protection key event data into a unique spectral signature using quantum dot labeling technology, thereby generating digital identification data of the cargo protection history;

[0139] The embodiment of the present invention utilizes quantum dot labeling technology to encode key event data for cargo protection. Quantum dot labeling technology utilizes the characteristics of nanomaterials (quantum dots) to convert information into spectral characteristics of specific wavelengths. The specific operation is to associate the identified key event data for cargo protection (such as the triggering time and execution effect of the protection measures) with a specific quantum dot label. Each quantum dot represents a unique data code based on its different spectral characteristics, thereby generating digital identification data for each protection event. These identification data can be quickly identified and read through specific spectral reactions. By matching all key event data in the cargo protection process with the spectral characteristics of quantum dots, a unique digital identification can be generated for the entire cargo protection process, which is convenient for subsequent tracing and management.

[0140] Step S53: collecting timestamp information and geographic location data during the logistics transportation process and associating them with the digital identification data of the cargo protection history to form a cargo protection data packet;

[0141] The embodiment of the present invention needs to collect timestamp information and geographic location data during the logistics transportation process. Specifically, timestamp information refers to the actual time record of each logistics transportation link, and geographic location data includes location information at each moment in the transportation process, such as GPS coordinates. After collecting these data, they are associated with the digital identification data of the cargo protection history. By embedding the corresponding timestamp and location information in the data of each key event in the cargo protection history, a complete cargo protection data packet can be formed. This data packet includes detailed information on each protection link of the cargo during the entire transportation process, including the implementation of protection measures, the time and geographic location during transportation, and provides a unique identification for this information through digital identification data.

[0142] Step S54: Processing the cargo protection data packet through a distributed encryption algorithm to generate a unique blockchain transaction record;

[0143] In this embodiment of the present invention, cargo protection data packets are processed using a distributed encryption algorithm to generate blockchain transaction records. Distributed encryption algorithms, such as hashing and symmetric encryption algorithms, ensure data security and privacy. Specifically, the generated cargo protection data packets are encrypted, converted into unique cryptographic hash values, and registered as transaction records on the blockchain network. Due to the decentralized and tamper-proof nature of blockchain technology, each transaction record is unique, ensuring the authenticity and credibility of the cargo protection data. The encrypted data can be recorded as transaction content on the blockchain, ensuring that the data cannot be tampered with during subsequent transactions and storage.

[0144] Step S55: Building a blockchain storage node network and writing blockchain transaction records into the blockchain storage node network;

[0145] This embodiment of the present invention requires the construction of a blockchain storage node network. A blockchain storage node is a unit within a distributed system that stores and manages blockchain transaction records. Specifically, generated blockchain transaction records are written to a storage node within the blockchain network. Each storage node stores the transaction records in its distributed ledger, ensuring that data can be simultaneously backed up and managed by multiple nodes within the blockchain, thereby improving data security and fault tolerance. This way, cargo protection data is permanently stored within the blockchain and can be queried and verified across the blockchain network.

[0146] Step S56: Construct a responsibility intelligent analysis model based on the logistics participants according to the blockchain storage node network, thereby forming a responsibility traceability certification chain for the goods.

[0147] This embodiment of the present invention uses data from a blockchain storage node network to construct an intelligent analysis model for the responsibilities of logistics stakeholders. Specifically, logistics stakeholders include transportation companies, warehousing companies, and distribution personnel, each of whom is responsible for different tasks at each stage of the cargo transportation process. By analyzing transaction data recorded in the blockchain, the responsibility and contribution of each logistics participant in the cargo protection process can be traced. The intelligent analysis model analyzes the behavior of each entity using transaction records and key event data in the blockchain data, and forms a chain of proof for traceability of responsibility. This chain of proof clearly records the responsibilities of each logistics link, ensuring that the protective measures and conditions for cargo are strictly implemented throughout the transportation process, providing a basis for subsequent quality tracing and accountability determination.

[0148] By extracting features from protection execution status data, the present invention can accurately identify the triggering time, execution effectiveness, and parameter changes of key protection measures, thereby generating key event data for cargo protection. This key event data provides a detailed record of cargo protection measures during transportation, helping to track key milestones in the protection process and ensure effective implementation of protection measures at every stage. Using quantum dot labeling technology, this key event data is encoded into a unique spectral signature, giving the cargo protection history a unique digital identifier, allowing the protection process of each item to be clearly recorded and traced. This digital identifier data provides a comprehensive, visual record of the cargo protection process, enhancing transparency and traceability. During transportation, the collected timestamp and geolocation data is linked to the digital identifier data of the cargo protection history to form a cargo protection data packet. This data packet not only contains real-time information on protection measures but also accurately records various environmental changes, time points, and locations during transportation, providing multi-dimensional contextual information on the cargo protection process. These cargo protection data packets are then processed using a distributed encryption algorithm to generate a unique blockchain transaction record. Leveraging the immutability and high security of blockchain technology, these records are encrypted and stored, permanently preserved within the network, ensuring the authenticity, integrity, and security of cargo protection data. To better trace responsibilities during cargo protection, an intelligent responsibility analysis model based on logistics stakeholders has been established. This model automatically analyzes and determines the responsibilities assumed by each party during the transportation process and their fulfillment of these responsibilities based on transaction records stored in the blockchain node network, forming a chain of proof for cargo responsibility traceability. This chain of proof not only helps accurately identify responsibility in the event of disputes but also provides data support for optimizing logistics management and enhancing cargo protection measures, ultimately ensuring more efficient, transparent, and reliable cargo safety and quality management during transportation.

[0149] The present invention further provides a digital logistics management system for executing the digital logistics management method described above, the digital logistics management system comprising:

[0150] Environmental perception and collection module, used to collect multi-dimensional environmental parameter field data around the cargo and biochemical indicator change data of the cargo itself;

[0151] The material interaction modeling module is used to construct a cargo permeability matrix based on a preset material library, multi-dimensional environmental parameter field data, and biochemical indicator change data. The permeability matrix is ​​obtained by quantifying the interaction rate between different environmental factors and cargo materials.

[0152] The quality prediction and analysis module is used to calculate the current environmental impact intensity based on the permeability coefficient matrix and multi-dimensional environmental parameter field data, and conduct thermodynamic coupling analysis in combination with biochemical indicator change data to obtain quality entropy increase rate data, which represents the rate of increase in disorder of the quality state of the goods. The quality state prediction model is generated based on the quality entropy increase rate data, and the microenvironment compensation parameter combination analysis of the quality state prediction model is performed to obtain the microenvironment compensation parameter combination.

[0153] The intelligent control execution module is used to perform differential analysis on multi-dimensional environmental parameter field data using a combination of microenvironment compensation parameters to generate dynamic control instruction data; the microenvironment control device is controlled based on the dynamic control instruction data, and adjustments are made based on real-time feedback from biochemical indicators to obtain protection execution status data;

[0154] The blockchain traceability module is used to generate a digital identification of the cargo protection history through quantum dot marking technology from the protection execution status data, and write it into the blockchain system to build a responsibility traceability proof chain for the cargo.

Claims

1. A digital logistics management method, characterized in that: The following steps are involved: Step S1: Collecting multi-dimensional environmental parameter field data around the cargo and biochemical index change data of the cargo itself; Step S2: Constructing a cargo permeability coefficient matrix based on a preset material library, multi-dimensional environmental parameter field data, and biochemical indicator change data. The permeability coefficient matrix is ​​obtained by quantifying the interaction rate between different environmental factors and cargo materials. Step S3: Calculate the current environmental impact intensity based on the permeability coefficient matrix and multi-dimensional environmental parameter field data, and perform thermodynamic coupling analysis in combination with the biochemical indicator change data to obtain the quality entropy increase rate data representing the disorder increase rate of the goods quality state; A quality state prediction model is generated based on the quality entropy increase rate data, and a microenvironment compensation parameter combination analysis is performed on the quality state prediction model to obtain a microenvironment compensation parameter combination; Step S4: performing difference analysis on the multi-dimensional environmental parameter field data using the micro-environment compensation parameter combination to generate dynamic control instruction data; The microenvironment control device is controlled based on dynamic control instruction data, and is adjusted according to the real-time feedback of biochemical indicators to obtain protection execution status data; Step S5: The protection execution status data is used to generate a digital identification of the cargo protection history through quantum dot marking technology, and is written into the blockchain system to build a responsibility traceability proof chain for the cargo.

2. The digital logistics management method according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: Arranging a multi-parameter sensor array on the outer packaging of the goods to construct a spatially distributed sensor network, wherein the multi-parameter sensor array includes environmental parameter sensor elements for real-time acquisition of temperature gradient distribution, humidity fluctuation coefficient, light intensity, air pressure change rate, and vibration spectrum information; Step S12: using a spatially distributed sensor network to perform data collection and transmission according to a preset time interval to form multi-dimensional environmental parameter field data; Step S13: Install a non-invasive biochemical sensor array on the outer packaging surface of the goods to non-destructively detect changes in the biochemical state of the goods through the packaging material, thereby obtaining biochemical indicator change data of the goods themselves.

3. The digital logistics management method according to claim 2, characterized in that: Step S2 includes the following steps: Step S21: Retrieving reference material property data that matches the current cargo type from a preset material library, wherein the reference material property data includes cargo internal structure stability index, surface activity index, heat exchange coefficient, and gas phase exchange coefficient; Step S22: converting the multidimensional environmental parameter field data into a standardized environmental factor intensity matrix, wherein the environmental factor intensity matrix includes numerical representations of five environmental factors, namely, temperature, humidity, light, air pressure, and vibration, at different intensity levels; Step S23: establishing a response relationship model between environmental factors and biochemical indicators based on the biochemical indicator change data and the standardized environmental factor intensity matrix; Step S24: Determine the initial state value and stability range of each cargo characteristic parameter based on the reference material characteristic data, and establish a reference model for the cargo characteristic parameters, wherein the reference model includes standard values ​​of the cargo's internal structure stability index, surface activity index, heat exchange coefficient, and gas phase exchange coefficient under standard environmental conditions; Step S25: Using the response relationship model, a dynamic framework for the influence of environmental factors is established to convert the effects of the five environmental factors (temperature, humidity, light, air pressure, and vibration) on cargo characteristics into reaction kinetic parameters; Step S26: Calculating the theoretical impact rates of the five environmental factors on the four benchmark properties based on the reaction kinetic parameters, thereby obtaining theoretical penetration impact data; Step S27: calibrating and optimizing the theoretical penetration impact data and the actual impact data predicted by the response relationship model to obtain a penetration impact coefficient; Step S28: Organize the permeability influence coefficient into a 5×4 order permeability coefficient matrix, where each element of the permeability coefficient matrix represents the permeability influence coefficient of a specific environmental factor on a specific cargo attribute.

4. The digital logistics management method according to claim 3, characterized in that: The calculation of the current environmental impact intensity in step S3 specifically includes: Perform matrix multiplication on the permeability coefficient matrix and the multidimensional environmental parameter field data to calculate the environment-cargo interaction intensity vector; Calculate the comprehensive intensity index of the overall impact of the environment on the goods based on the environment-goods interaction intensity vector; Calculate the distribution of environmental impact intensity in different regions and time points by using the comprehensive intensity index to obtain an environmental impact intensity distribution map; Identify environmental impact hotspots and key time windows based on the environmental impact intensity distribution map, and obtain key environmental impact node data; The current environmental impact intensity data is generated based on the environmental impact key node data and the environment-goods interaction intensity vector.

5. The digital logistics management method according to claim 4, characterized in that: The thermodynamic coupling analysis described in step S3 specifically includes: Draw a curve of the corresponding relationship between environmental factors and biochemical responses of goods based on the environmental impact intensity data and biochemical indicator change data; According to the corresponding relationship curve, the changing pattern of the internal order of the goods under environmental stress is quantified, thereby constructing an entropy function model of the goods quality status; Substitute the biochemical index change data into the entropy function model to calculate the initial value of the entropy growth rate of the goods under the current environmental conditions; Establish an acceleration effect model of environment-matter interaction based on the permeability coefficient matrix; The amplification factor of environmental fluctuations on the entropy growth rate is calculated based on the acceleration effect model. The amplification factor is calculated by detecting the fluctuation of environmental parameters in the acceleration effect model. When the fluctuation amplitude exceeds 3 times the mean standard deviation within 15 minutes, the instantaneous fluctuation impact value is calculated and the amplification factor is increased by 0.

3. The humidity change rate is monitored. When it exceeds 5% RH / min, the humidity mutation impact coefficient is calculated and the amplification factor is further increased by 0.

2. Multiply the initial value of the entropy growth rate by the amplification coefficient to obtain the quality entropy growth rate data that represents the rate of increase of disorder in the quality state of the goods.

6. The digital logistics management method according to claim 5, characterized in that: The generation of the quality status prediction model described in step S3 specifically includes: Perform time series analysis on the quality entropy increase rate data to extract the trend pattern and periodic characteristics of quality attenuation, thereby obtaining the quality change trend characteristics; Pattern matching is performed between the quality change trend characteristics and the pre-acquired historical quality change database to screen out similar quality decay paths and obtain reference decay pattern data; Based on the reference decay pattern data and the quality entropy increase rate data, the time evolution function of the goods quality status is constructed to form a basic model for quality status prediction; Input the environmental impact intensity data into the quality status prediction basic model, calculate the quality status prediction values ​​at different time points, and obtain the quality change prediction result data; The biochemical index change data is used to make real-time corrections to the quality change prediction result data, and a quality status prediction model is constructed to make real-time predictions on the quality change trend of goods.

7. The digital logistics management method according to claim 6, characterized in that: The microenvironment compensation parameter combination analysis described in step S3 specifically includes: According to the quality status prediction model, the correlation analysis between the quality entropy increase rate data and environmental factors was carried out to obtain the data of the dominant environmental stress factors; Calculate the reverse adjustment parameters based on adverse effects on the dominant environmental stress factor data to obtain the environmental factor balance adjustment data; Conduct feasibility analysis on environmental factor balance regulation data, optimize parameters within the actual regulation range, and generate feasible candidate sets of regulation parameters; The quality status prediction model is used to simulate and verify the candidate set of control parameters, and the protection effect of different parameter combinations is evaluated to obtain protection effect evaluation data; According to the protection effect evaluation data, a comprehensive score based on energy consumption and physical feasibility is performed to obtain a combination of microenvironment compensation parameters.

8. The digital logistics management method according to claim 7, characterized in that: Step S4 includes the following steps: Step S41: Comparing and analyzing the microenvironment compensation parameter combination with the multi-dimensional environmental parameter field data, calculating the compensation difference of each environmental factor, and generating environmental compensation demand data, wherein the environmental compensation demand data includes temperature compensation difference, humidity compensation difference, light compensation difference, air pressure compensation difference, and vibration compensation difference; Step S42: Design a micro-environment control strategy based on the environmental compensation demand data, and convert it into an instruction set executable by the device to obtain dynamic control instruction data. The micro-environment control strategy is specifically designed as follows: When the temperature compensation difference exceeds ±3°C, the temperature quick response coefficient is calculated, the response priority of the temperature adjustment unit is set to the highest, and the temperature priority adjustment data is obtained; Humidity gradient analysis is performed on areas where the humidity compensation difference exceeds ±15% RH to obtain the humidity distribution unevenness index; The UV intensity in the calculated illumination compensation difference exceeds 3000 μW / cm 2 The exposure risk of UV light exposure is determined and the UV light exposure risk assessment data is obtained; Detect the fluctuation characteristics of the air pressure compensation difference with a fluctuation frequency exceeding 0.5 Hz to obtain air pressure fluctuation characteristic data; Perform spectrum decomposition on the shock waveform with peak acceleration exceeding 2g in the vibration compensation difference to obtain vibration spectrum characteristic data; Generate a coordinated control strategy for environmental factors based on temperature priority adjustment data, humidity distribution unevenness index, UV exposure risk assessment data, air pressure fluctuation characteristics data, and vibration spectrum feature data; Step S43: Sending the dynamic control instruction data to the microenvironment control device, activating the corresponding environmental control unit, monitoring the operating status of the microenvironment control device and changes in environmental parameters, and recording real-time control effect data, wherein the environmental control unit includes a temperature control unit, a humidity control unit, a gas phase protection unit, and a physical barrier unit; Step S44: Collect real-time biochemical index data based on the real-time control effect data, evaluate the protective effect of microenvironment control on the quality of goods, and optimize the control parameters in real time based on the protection effect to obtain protection execution status data.

9. The digital logistics management method according to claim 8, characterized in that: Step S5 includes the following steps: Step S51: Extract features from the protection execution status data to identify the triggering time, execution effect, and parameter changes of key protection measures, thereby obtaining key cargo protection event data; Step S52: Encoding the cargo protection key event data into a unique spectral signature using quantum dot labeling technology, thereby generating digital identification data of the cargo protection history; Step S53: collecting timestamp information and geographic location data during the logistics transportation process and associating them with the digital identification data of the cargo protection history to form a cargo protection data packet; Step S54: Processing the cargo protection data packet through a distributed encryption algorithm to generate a unique blockchain transaction record; Step S55: Building a blockchain storage node network and writing blockchain transaction records into the blockchain storage node network; Step S56: Construct a responsibility intelligent analysis model based on the logistics participants according to the blockchain storage node network, thereby forming a responsibility traceability certification chain for the goods.

10. A digital logistics management system, characterized in that: For executing the digital logistics management method according to claim 1, the digital logistics management system comprises: Environmental perception and collection module, used to collect multi-dimensional environmental parameter field data around the cargo and biochemical indicator change data of the cargo itself; The material interaction modeling module is used to construct a cargo permeability matrix based on a preset material library, multi-dimensional environmental parameter field data, and biochemical indicator change data. The permeability matrix is ​​obtained by quantifying the interaction rate between different environmental factors and cargo materials. The quality prediction and analysis module is used to calculate the current environmental impact intensity based on the permeability coefficient matrix and multi-dimensional environmental parameter field data, and conduct thermodynamic coupling analysis in combination with biochemical indicator change data to obtain quality entropy increase rate data, which represents the rate of increase in disorder of the quality state of the goods. The quality state prediction model is generated based on the quality entropy increase rate data, and the microenvironment compensation parameter combination analysis of the quality state prediction model is performed to obtain the microenvironment compensation parameter combination. The intelligent control execution module is used to perform differential analysis on multi-dimensional environmental parameter field data using a combination of microenvironment compensation parameters to generate dynamic control instruction data; the microenvironment control device is controlled based on the dynamic control instruction data, and adjustments are made based on real-time feedback from biochemical indicators to obtain protection execution status data; The blockchain traceability module is used to generate a digital identification of the cargo protection history through quantum dot marking technology from the protection execution status data, and write it into the blockchain system to build a responsibility traceability proof chain for the cargo.

Images