Meat cold chain transportation temperature and humidity intelligent monitoring and early warning system

Abstract

This invention belongs to the field of cold chain transportation technology, and specifically relates to an intelligent monitoring and early warning system for temperature and humidity in cold chain transportation of meat products. It includes a multi-dimensional sensing array for meat products, composed of several movable embedded sensing nodes. These nodes are evenly deployed inside refrigerated truck compartments, between meat stacks, on the surface of meat packaging, and inside the packaging. They are used to collect macroscopic environmental parameters, micro-environmental parameters, quality parameters, and packaging sealing parameters throughout the entire cold chain transportation process. An edge computing fusion gateway is also included, connected to the multi-dimensional sensing array via LoRaWAN / NB-IoT dual-mode communication. This invention significantly improves the accuracy of monitoring data, ensuring real-time and comprehensive capture of changes in the state of meat products during transportation. This provides reliable data support for subsequent adjustments, early warnings, and quality predictions, guaranteeing the comprehensiveness and accuracy of meat quality monitoring from the source.

Description

Technical Field

[0001] This invention belongs to the field of cold chain transportation technology, and in particular relates to an intelligent monitoring and early warning system for temperature and humidity in cold chain transportation of meat products. Background Technology

[0002] As a perishable fresh product, the quality of meat is closely related to environmental parameters such as temperature and humidity during cold chain transportation. The stability of these parameters throughout the entire cold chain transportation process directly determines the freshness, safety, and shelf life of the meat. Abnormal fluctuations in temperature and humidity, or changes in the meat's microenvironment, can easily lead to microbial growth and quality deterioration, causing economic losses and potentially posing food safety risks. Currently, temperature and humidity monitoring technology has been gradually introduced into the cold chain transportation of meat. Various sensing devices collect environmental parameters during transportation, and combined with communication technology, the data is transmitted to management terminals to achieve preliminary monitoring of the cold chain transportation process. Some systems have also added simple early warning functions, attempting to reduce meat transportation losses, ensure meat quality and safety, and provide basic data support for the management of cold chain logistics companies.

[0003] However, through research, we found that existing temperature and humidity monitoring and early warning systems for cold chain meat transportation still have some shortcomings and deficiencies: First, the monitoring dimensions are limited, mostly only monitoring the macroscopic temperature and humidity inside the refrigerated compartment, failing to accurately capture the internal microenvironment of the meat, the packaging sealing status, and core meat quality parameters, resulting in an inability to truly reflect the actual quality status of the meat; second, data processing and response are lagging, with most systems relying on centralized cloud data processing, lacking local emergency decision-making capabilities, unable to achieve real-time early warning and control when the network is interrupted, and data calibration accuracy is insufficient, easily leading to false alarms and missed alarms; third, the prevention and control mode is passive, only issuing warnings after parameters are abnormal, unable to proactively adjust cold chain equipment and the meat microenvironment, making it difficult to avoid the risk of meat deterioration in advance; fourth, the functions are incomplete, lacking functions such as meat quality trend prediction, equipment health assessment, data traceability, and transportation route optimization, while the early warning response lacks a multi-role collaborative mechanism, resulting in low efficiency in handling anomalies and failing to meet the needs of high-quality development in cold chain meat transportation. Therefore, there is an urgent need to improve the existing technical problems and related deficiencies. Based on research, we hereby provide an intelligent temperature and humidity monitoring and early warning system for cold chain meat transportation. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by providing an intelligent temperature and humidity monitoring and early warning system for cold chain transportation of meat products. This system combines comprehensive monitoring dimensions with efficient data processing, can dynamically adapt to various specifications of meat products and cold chain transportation conditions, improves the accuracy of anomaly identification and the ability to predict meat quality, and simultaneously achieves dual protection of proactive prevention and control throughout the entire process and data traceability. This system is intended to solve the problems existing in existing technologies.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A smart monitoring and early warning system for temperature and humidity in cold chain transportation of meat products, comprising:

[0007] The meat product full-dimensional sensing array consists of several movable embedded sensing nodes, which are evenly deployed inside the refrigerated compartment, in the gaps between meat stacks, on the surface of meat packaging, and inside the packaging. These nodes are used to collect macro-environmental parameters, meat micro-environmental parameters, meat quality parameters, and packaging sealing parameters throughout the entire cold chain transportation process.

[0008] The edge computing fusion gateway is connected to the meat product full-dimensional sensing array via LoRaWAN / NB-IoT dual-mode communication. The edge computing fusion gateway has a built-in edge computing module, AI anomaly detection model, data calibration unit and fault self-diagnosis unit. It is used to perform real-time preprocessing, dynamic calibration, anomaly feature extraction and fault diagnosis on heterogeneous data collected from multiple nodes. The edge computing fusion gateway can independently trigger an early warning and execute preset emergency control commands when the network is interrupted. At the same time, it can detect equipment and communication failures and automatically start the backup mechanism to ensure continuous system operation.

[0009] The intelligent adaptive adjustment module is electrically connected to the edge computing fusion gateway. The intelligent adaptive adjustment module is also electrically connected to the refrigeration unit, fresh air system, humidification equipment, and packaging sealing compensation equipment of the refrigerated compartment. It is used to adaptively adjust the cooling power, fresh air volume, humidification volume, and packaging sealing pressure according to the fusion calibration data and abnormal characteristics output by the edge computing fusion gateway, so as to avoid the risks of temperature and humidity fluctuations, packaging leakage, and meat quality deterioration in advance and achieve proactive prevention and control.

[0010] The cloud-based intelligent management platform establishes bidirectional communication with the edge computing fusion gateway through a 5G / NB-IoT communication network. The cloud-based intelligent management platform has built-in meat deterioration trend prediction model, equipment health assessment model, dynamic threshold generation module, blockchain evidence storage unit, and transportation route optimization module. It is used to store full-process monitoring data, dynamically adjust early warning thresholds, accurately predict the remaining shelf life of meat products, assess the health status of cold chain equipment, and achieve tamper-proof data storage and full traceability. At the same time, the cloud-based intelligent management platform can dynamically optimize transportation routes based on real-time road conditions and quality prediction results.

[0011] The multi-terminal hierarchical collaborative early warning module establishes communication connections with the edge computing fusion gateway and the cloud-based intelligent management platform, respectively. The multi-terminal hierarchical collaborative early warning module includes a vehicle-mounted early warning terminal, a dispatch center early warning terminal, and a mobile early warning terminal for management personnel. It is used to push early warning information and emergency handling suggestions through multiple channels according to the four levels of anomaly, such as sound and light prompts, voice broadcasts, graphic pop-ups, SMS notifications, and emergency calls, so as to achieve rapid response by multiple roles and improve the efficiency of anomaly handling.

[0012] In a preferred embodiment, the mobile embedded sensing node includes an embedded temperature sensor, a dual-channel humidity sensor, an infrared CO2 concentration sensor, a miniature pressure sensor, an electrochemical volatile basic nitrogen sensor, a bioelectrochemical microbial sensor, and a low-power LoRaWAN / NB-IoT dual-mode communication module. The mobile embedded sensing node supports a sleep-wake mode, has a standby power consumption of ≤50mW, and can automatically wake up according to the acquisition frequency to ensure long-term stable communication.

[0013] The macro-environmental parameters include temperature, humidity, and airflow speed in different areas of the compartment; the meat micro-environmental parameters include internal temperature, surface humidity, and CO2 concentration inside the packaging; the meat quality parameters include volatile basic nitrogen concentration, oxygen concentration, and the number of microorganisms on the surface of the meat; and the packaging sealing parameters include internal air pressure and gas leakage rate.

[0014] As a preferred implementation, the AI ​​anomaly detection model adopts a CNN+Transformer fusion algorithm and is generated by optimizing and training an LSTM neural network. The AI ​​anomaly detection model can accurately identify sudden changes in temperature and humidity, sudden increases in CO2 concentration, packaging leakage, equipment malfunctions, frequent opening and closing of doors, and rapid deterioration of meat quality. The data calibration unit adopts the weighted least squares method and combines the position weights of several movable embedded sensing nodes, measurement accuracy, and environmental interference coefficients to dynamically calibrate the collected data.

[0015] As a preferred embodiment, the specific adjustment logic of the intelligent adaptive adjustment module is as follows:

[0016] When the internal temperature of the meat exceeds the preset threshold by 0.3℃, the refrigeration unit power is automatically increased by 10%-30%, while the fresh air volume is increased by 5%-15%.

[0017] When the CO2 concentration inside the packaging exceeds 5% VOL or a packaging leak is detected, the packaging sealing compensation device is activated to replenish the air pressure inside the packaging to the standard range.

[0018] When the airflow velocity inside the carriage is below 0.2 m / s, adjust the angle of the fresh air system outlet to ensure uniform airflow distribution.

[0019] When the surface humidity of meat products deviates from the target range of ±5%RH, start the humidification equipment or fresh air dehumidification function to adjust it.

[0020] As a preferred implementation, the meat deterioration trend prediction model employs an attention-enhanced LSTM algorithm, which integrates the Arrhenius equation and the Gompertz model for optimization. The input parameters include the meat microenvironment parameter change curve, meat quality parameters, transportation time, initial meat quality, and fluctuating data of the truck compartment environment. The equipment health assessment model generates a health score of 0-100 based on equipment operating parameters and historical fault data. When the score is below 60, targeted maintenance suggestions are pushed.

[0021] As a preferred implementation, the four anomaly levels and response methods of the multi-terminal hierarchical collaborative early warning module are as follows:

[0022] Level 1 (Slight): Data from a single movable embedded sensing node deviates from the preset threshold without a significant abnormal trend. Only the vehicle-mounted warning terminal issues an alert sound and records the abnormal data simultaneously.

[0023] Level 2 (General): Multiple adjacent movable embedded sensing nodes show abnormal data, the abnormal trend is gradual, the vehicle terminal voice broadcasts, the dispatch center terminal pops up a prompt, and the management personnel receive text reminders on their mobile terminals;

[0024] Level 3 (Severe): More than half of the mobile embedded sensing nodes show abnormal data, the meat microenvironment shows a significant deterioration trend, multiple terminals issue simultaneous early warnings, and the dispatch center can remotely issue adjustment instructions.

[0025] Level 4 (Emergency): Core parameters exceed safety thresholds, meat products are at risk of spoilage, triggering multi-terminal emergency warnings, automatically dialing emergency numbers for management personnel, and simultaneously activating the highest level of emergency response from the intelligent adaptive adjustment module.

[0026] As a preferred implementation, the blockchain evidence storage unit adopts a consortium blockchain architecture to upload monitoring data, early warning records, emergency response records, meat quality assessment reports, and route optimization records in real time. It also supports hierarchical queries by transportation companies, regulatory authorities, and consumers, achieving tamper-proof and traceable data throughout the entire process. The dynamic threshold generation module dynamically adjusts the early warning thresholds of various parameters based on meat type, transportation stage, transportation duration, and real-time monitoring data to adapt to the cold chain transportation needs of different meat products.

[0027] As a preferred embodiment, it also includes a wireless charging and energy storage module, which is used to power the meat product full-dimensional sensing array and the edge computing fusion gateway. The wireless charging and energy storage module adopts electromagnetic induction wireless charging and has a built-in high-capacity lithium battery that can provide power for more than 72 hours without external power supply. It also has real-time power monitoring and low power warning functions.

[0028] As a preferred embodiment, the movable embedded sensing node adopts a foldable snap-on installation structure to adapt to different specifications of meat packaging and vehicle structure, and fits tightly with the meat and vehicle after installation.

[0029] As a preferred implementation, the transportation route optimization module can dynamically adjust the transportation route and driving speed based on real-time road condition information, changes in ambient temperature along the route, meat quality prediction results, and transportation timeliness requirements, thereby reducing temperature fluctuations during transportation.

[0030] The cloud-based intelligent management platform also includes a user permission hierarchical management module and a report generation module. The user permission hierarchical management module has permissions divided into administrators, dispatchers, drivers, and supervisors. Users with different permissions can view monitoring data and reports within their respective scopes. The report generation module can automatically generate daily / weekly / monthly monitoring reports, meat quality analysis reports, and equipment operation reports, and supports exporting and printing, facilitating full-process management and traceability.

[0031] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0032] In the solution of this invention:

[0033] This invention utilizes several movable embedded sensing nodes in a multi-dimensional meat sensing array to simultaneously collect four core parameters: macroscopic environment, meat microenvironment, meat quality, and packaging sealing. These parameters cover key indicators such as temperature, humidity, volatile basic nitrogen concentration, microbial quantity, and CO2 concentration. Combined with weighted least squares data calibration, the accuracy of monitoring data is significantly improved, ensuring real-time and comprehensive capture of changes in the state of meat during transportation. This provides reliable data support for subsequent adjustments, early warnings, and quality predictions, guaranteeing the comprehensiveness and accuracy of meat quality monitoring from the source.

[0034] The edge computing fusion gateway of this invention has local emergency decision-making, fault self-diagnosis and backup mechanisms. It can independently complete anomaly identification and emergency control when the network is interrupted, which solves the problems of data processing lag and reliance on the cloud in the existing system. Moreover, the intelligent adaptive adjustment module can actively adjust the cold chain equipment and packaging status according to monitoring data to avoid the risk of meat deterioration in advance. At the same time, the multi-terminal hierarchical collaborative early warning module realizes multi-role and multi-channel collaborative response. The four-level anomaly level accurately matches different anomaly scenarios, which greatly reduces the probability of false alarms and missed alarms, improves the timeliness and pertinence of anomaly handling, and effectively reduces meat transportation losses.

[0035] This invention integrates multiple technologies, including LoRaWAN / NB-IoT dual-mode communication, CNN+Transformer+LSTM fusion algorithm, wireless charging, and blockchain notarization, to solve pain points in existing systems such as unstable communication, insufficient battery life, unreliable data, and unreasonable routing. The foldable snap-fit ​​installation structure design of the movable embedded sensing node improves the system's adaptability and installation flexibility for cold chain transportation. Utilizing the quality prediction, equipment evaluation, route optimization, and access control functions of the cloud-based intelligent management platform, it not only provides cold chain logistics companies with intelligent management tools for the entire process but also achieves data immutability and full traceability through blockchain notarization, balancing enterprise management efficiency, consumer rights, and regulatory requirements, and possesses strong prospects for industrial application. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. The drawings are described as follows:

[0037] Figure 1 This is a schematic diagram of the overall architecture of the intelligent temperature and humidity monitoring and early warning system for cold chain transportation of meat products according to the present invention. Detailed Implementation

[0038] The embodiments described below are merely some embodiments of the present invention and do not represent all embodiments consistent with the present invention. Exemplary embodiments will now be described with reference to the accompanying drawings:

[0039] like Figure 1 As shown, the present invention provides an intelligent monitoring and early warning system for temperature and humidity in cold chain transportation of meat products, comprising:

[0040] The meat product full-dimensional sensing array consists of several movable embedded sensing nodes, which are evenly deployed inside the refrigerated compartment, in the gaps between meat stacks, on the surface of meat packaging, and inside the packaging. These nodes are used to collect macro-environmental parameters, meat micro-environmental parameters, meat quality parameters, and packaging sealing parameters throughout the entire cold chain transportation process.

[0041] The edge computing fusion gateway connects to the meat product full-dimensional sensing array via LoRaWAN / NB-IoT dual-mode communication. The edge computing fusion gateway has built-in edge computing modules, AI anomaly detection models, data calibration units, and fault self-diagnosis units. It is used to perform real-time preprocessing, dynamic calibration, anomaly feature extraction, and fault diagnosis on heterogeneous data collected from multiple nodes. The edge computing fusion gateway can independently trigger early warnings and execute preset emergency control commands when the network is interrupted. At the same time, it can detect equipment and communication faults and automatically start the backup mechanism to ensure continuous system operation.

[0042] The intelligent adaptive adjustment module is electrically connected to the edge computing fusion gateway, and is also electrically connected to the refrigeration unit, fresh air system, humidification equipment, and packaging sealing compensation equipment in the refrigerated compartment. It is used to adaptively adjust the cooling power, fresh air volume, humidification volume, and packaging sealing pressure based on the fusion calibration data and abnormal characteristics output by the edge computing fusion gateway, so as to avoid the risks of temperature and humidity fluctuations, packaging leakage, and meat quality deterioration in advance and achieve proactive prevention and control.

[0043] The cloud-based intelligent management platform establishes bidirectional communication with the edge computing gateway through a 5G / NB-IoT communication network. The platform has built-in meat deterioration trend prediction model, equipment health assessment model, dynamic threshold generation module, blockchain evidence storage unit, and transportation route optimization module. It is used to store full-process monitoring data, dynamically adjust early warning thresholds, accurately predict the remaining shelf life of meat products, assess the health status of cold chain equipment, and achieve tamper-proof data storage and full traceability. At the same time, the cloud-based intelligent management platform can dynamically optimize transportation routes based on real-time road conditions and quality prediction results.

[0044] The multi-terminal hierarchical collaborative early warning module establishes communication connections with the edge computing converged gateway and the cloud intelligent management platform, respectively. The multi-terminal hierarchical collaborative early warning module includes vehicle-mounted early warning terminals, dispatch center early warning terminals, and mobile early warning terminals for management personnel. It is used to push early warning information and emergency handling suggestions through multiple channels according to the four levels of anomaly, such as sound and light prompts, voice broadcasts, graphic pop-ups, SMS notifications, and emergency calls, so as to achieve rapid response by multiple roles and improve the efficiency of anomaly handling.

[0045] As a preferred embodiment, based on the above structure, the mobile embedded sensing node further includes an embedded temperature sensor, a dual-channel humidity sensor, an infrared CO2 concentration sensor, a miniature pressure sensor, an electrochemical volatile basic nitrogen sensor, a bioelectrochemical microbial sensor, and a low-power LoRaWAN / NB-IoT dual-mode communication module. The mobile embedded sensing node supports a sleep-wake mode, has a standby power consumption of ≤50mW, and can automatically wake up according to the acquisition frequency to ensure long-term stable communication.

[0046] Macro-environmental parameters include temperature, humidity, and airflow speed in different areas of the compartment; meat micro-environmental parameters include internal temperature, surface humidity, and CO2 concentration inside the packaging; meat quality parameters include volatile basic nitrogen concentration, oxygen concentration, and the number of microorganisms on the surface of the meat; and packaging sealing parameters include internal air pressure and gas leakage rate.

[0047] As a preferred implementation, based on the above structure, the AI ​​anomaly detection model further adopts a CNN+Transformer fusion algorithm, combined with LSTM neural network optimization training generation. The AI ​​anomaly detection model can accurately identify sudden changes in temperature and humidity, sudden rise in CO2 concentration, packaging leakage, equipment operation failure, frequent opening and closing of doors, and rapid deterioration of meat quality. The data calibration unit adopts the weighted least squares method, combined with the position weights of several movable embedded sensing nodes, measurement accuracy, and environmental interference coefficient, to dynamically calibrate the collected data.

[0048] As a preferred embodiment, based on the above structure, the specific adjustment logic of the intelligent adaptive adjustment module is further as follows:

[0049] When the internal temperature of the meat exceeds the preset threshold by 0.3℃, the refrigeration unit power is automatically increased by 10%-30%, while the fresh air volume is increased by 5%-15%.

[0050] When the CO2 concentration inside the packaging exceeds 5% VOL or a packaging leak is detected, the packaging sealing compensation device is activated to replenish the air pressure inside the packaging to the standard range.

[0051] When the airflow velocity inside the carriage is below 0.2 m / s, adjust the angle of the fresh air system outlet to ensure uniform airflow distribution.

[0052] When the surface humidity of meat products deviates from the target range of ±5%RH, start the humidification equipment or fresh air dehumidification function to adjust it.

[0053] As a preferred implementation, based on the above structure, the meat deterioration trend prediction model further employs an attention-enhanced LSTM algorithm, which integrates the Arrhenius equation and the Gompertz model for optimization. The input parameters include the meat microenvironment parameter change curve, meat quality parameters, transportation time, initial meat quality, and fluctuating data of the compartment environment. The equipment health assessment model generates a health score of 0-100 based on equipment operating parameters and historical fault data. When the score is below 60, targeted maintenance suggestions are pushed.

[0054] As a preferred implementation, based on the above structure, the four levels of anomalies and their response methods of the multi-terminal hierarchical collaborative early warning module are further as follows:

[0055] Level 1 (Slight): Data from a single movable embedded sensing node deviates from the preset threshold without a significant abnormal trend. Only the vehicle-mounted warning terminal issues an alert sound and records the abnormal data simultaneously.

[0056] Level 2 (General): Multiple adjacent movable embedded sensing nodes show abnormal data, the abnormal trend is gradual, the vehicle terminal voice broadcasts, the dispatch center terminal pops up a prompt, and the management personnel receive text reminders on their mobile terminals;

[0057] Level 3 (Severe): More than half of the mobile embedded sensing nodes show abnormal data, the meat microenvironment shows a significant deterioration trend, multiple terminals issue simultaneous early warnings, and the dispatch center can remotely issue adjustment instructions.

[0058] Level 4 (Emergency): Core parameters exceed safety thresholds, meat products are at risk of spoilage, triggering multi-terminal emergency warnings, automatically dialing emergency numbers for management personnel, and simultaneously activating the highest level of emergency response from the intelligent adaptive adjustment module.

[0059] As a preferred implementation, based on the above structure, the blockchain evidence storage unit further adopts a consortium blockchain architecture to upload monitoring data, early warning records, emergency response records, meat quality assessment reports, and route optimization records in real time. It also supports hierarchical queries by transportation companies, regulatory authorities, and consumers, achieving tamper-proof and traceable data throughout the entire process. The dynamic threshold generation module dynamically adjusts the early warning thresholds of each parameter based on meat type, transportation stage, transportation duration, and real-time monitoring data to adapt to the cold chain transportation needs of different meat products.

[0060] As a preferred embodiment, based on the above structure, it further includes a wireless charging and energy storage module. The wireless charging and energy storage module is used to power the meat product full-dimensional sensing array and edge computing fusion gateway. The wireless charging and energy storage module adopts electromagnetic induction wireless charging method, has a built-in large-capacity lithium battery, and can provide power for more than 72 hours without external power supply. It also has real-time power monitoring and low power warning functions.

[0061] As a preferred embodiment, based on the above structure, the movable embedded sensing node further adopts a foldable snap-on installation structure to adapt to different specifications of meat packaging and vehicle structure, and fits tightly with the meat and vehicle after installation.

[0062] As a preferred embodiment, based on the above structure, the transportation route optimization module can further adjust the transportation route and driving speed based on real-time road condition information, changes in ambient temperature along the route, meat quality prediction results and transportation timeliness requirements, so as to reduce temperature fluctuations during transportation.

[0063] The cloud-based intelligent management platform also includes a user permission hierarchical management module and a report generation module. The user permission hierarchical management module has permissions divided into administrators, dispatchers, drivers, and supervisors. Users with different permissions can view monitoring data and reports within their corresponding scope. The report generation module can automatically generate daily / weekly / monthly monitoring reports, meat quality analysis reports, and equipment operation reports, and supports exporting and printing, facilitating full-process management and traceability.

[0064] The working principle of this invention is as follows:

[0065] In operation, this invention first deploys several movable embedded sensing nodes of the meat product full-dimensional sensing array at a preset density inside the refrigerated truck compartment, in the gaps between meat stacks, on the surface of meat packaging, and inside the packaging. After completing node installation and communication pairing, during transportation, each movable embedded sensing node synchronously collects macroscopic environmental parameters, meat micro-environment parameters, meat quality parameters, and packaging sealing parameters at a set frequency. The data is then transmitted in real time to the edge computing fusion gateway via LoRaWAN / NB-IoT dual-mode communication. After receiving the data, the edge computing fusion gateway first performs dynamic calibration and noise reduction filtering using the weighted least squares method, and then uses the CNN+Transformer+LSTM fusion algorithm to perform abnormal feature extraction and fault self-diagnosis, identifying abnormal patterns such as sudden temperature and humidity changes, packaging leakage, equipment failure, and meat quality deterioration. When the network is normal, the processed data is synchronously uploaded to the cloud. When the network is interrupted, local emergency warning and control commands are executed independently to ensure the continuity and reliability of monitoring.

[0066] Subsequently, the cloud-based intelligent management platform receives data uploaded from the edge gateway. Through the meat deterioration trend prediction model, equipment health assessment model, and dynamic threshold generation module, it completes the prediction of the remaining shelf life of meat, the health assessment of cold chain equipment, and the dynamic adjustment of early warning thresholds. Combined with real-time road condition information, the transportation route optimization module generates the optimal transportation route. At the same time, the intelligent adaptive adjustment module adaptively adjusts the power of the refrigeration unit, the fresh air volume, the humidification volume, and the packaging sealing pressure based on the analysis results of the edge gateway and the cloud platform, actively avoiding the risks of temperature and humidity fluctuations and meat deterioration. Moreover, through the multi-terminal hierarchical collaborative early warning module, it can push hierarchical early warning information and emergency handling suggestions to vehicle terminals, dispatch centers, and management personnel mobile terminals according to four levels of anomalies. All data in the entire process is uploaded to the consortium blockchain through the blockchain storage unit, realizing closed-loop management of the entire chain of monitoring, adjustment, early warning, and traceability.

[0067] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any equivalent changes, modifications, substitutions, and variations made by those skilled in the art based on the concept of the present invention and on the basis of the prior art through logical analysis, reasoning, or limited experiments shall be within the scope of protection defined by the claims.

Claims

1. A smart monitoring and early warning system for temperature and humidity in cold chain transportation of meat products, characterized in that, include: The meat product full-dimensional sensing array consists of several movable embedded sensing nodes, which are evenly deployed inside the refrigerated compartment, in the gaps between meat stacks, on the surface of meat packaging, and inside the packaging. These nodes are used to collect macro-environmental parameters, meat micro-environmental parameters, meat quality parameters, and packaging sealing parameters throughout the entire cold chain transportation process. The edge computing fusion gateway is connected to the meat product full-dimensional sensing array via LoRaWAN / NB-IoT dual-mode communication. The edge computing fusion gateway has a built-in edge computing module, AI anomaly detection model, data calibration unit and fault self-diagnosis unit. It is used to perform real-time preprocessing, dynamic calibration, anomaly feature extraction and fault diagnosis on heterogeneous data collected from multiple nodes. The edge computing fusion gateway can independently trigger an early warning and execute preset emergency control commands when the network is interrupted. At the same time, it can detect equipment and communication failures and automatically start the backup mechanism to ensure continuous system operation. The intelligent adaptive adjustment module is electrically connected to the edge computing fusion gateway. The intelligent adaptive adjustment module is also electrically connected to the refrigeration unit, fresh air system, humidification equipment, and packaging sealing compensation equipment of the refrigerated compartment. It is used to adaptively adjust the cooling power, fresh air volume, humidification volume, and packaging sealing pressure according to the fusion calibration data and abnormal characteristics output by the edge computing fusion gateway, so as to avoid the risks of temperature and humidity fluctuations, packaging leakage, and meat quality deterioration in advance and achieve proactive prevention and control. The cloud-based intelligent management platform establishes bidirectional communication with the edge computing fusion gateway through a 5G / NB-IoT communication network. The cloud-based intelligent management platform has built-in meat deterioration trend prediction model, equipment health assessment model, dynamic threshold generation module, blockchain evidence storage unit, and transportation route optimization module. It is used to store full-process monitoring data, dynamically adjust early warning thresholds, accurately predict the remaining shelf life of meat products, assess the health status of cold chain equipment, and achieve tamper-proof data storage and full traceability. At the same time, the cloud-based intelligent management platform can dynamically optimize transportation routes based on real-time road conditions and quality prediction results. The multi-terminal hierarchical collaborative early warning module establishes communication connections with the edge computing fusion gateway and the cloud-based intelligent management platform, respectively. The multi-terminal hierarchical collaborative early warning module includes a vehicle-mounted early warning terminal, a dispatch center early warning terminal, and a mobile early warning terminal for management personnel. It is used to push early warning information and emergency handling suggestions through multiple channels according to the four levels of anomaly, such as sound and light prompts, voice broadcasts, graphic pop-ups, SMS notifications, and emergency calls, so as to achieve rapid response by multiple roles and improve the efficiency of anomaly handling.

2. The intelligent monitoring and early warning system for temperature and humidity in cold chain transportation of meat products according to claim 1, characterized in that: The mobile embedded sensing node includes an embedded temperature sensor, a dual-channel humidity sensor, an infrared CO2 concentration sensor, a miniature pressure sensor, an electrochemical volatile basic nitrogen sensor, a bioelectrochemical microbial sensor, and a low-power LoRaWAN / NB-IoT dual-mode communication module. The mobile embedded sensing node supports a sleep-wake mode, with standby power consumption ≤50mW, and can automatically wake up according to the acquisition frequency to ensure long-term stable communication. The macro-environmental parameters include temperature, humidity, and airflow speed in different areas of the compartment; the meat micro-environmental parameters include internal temperature, surface humidity, and CO2 concentration inside the packaging; the meat quality parameters include volatile basic nitrogen concentration, oxygen concentration, and the number of microorganisms on the surface of the meat; and the packaging sealing parameters include internal air pressure and gas leakage rate.

3. The intelligent monitoring and early warning system for temperature and humidity in cold chain transportation of meat products according to claim 1, characterized in that: The AI ​​anomaly detection model uses a CNN+Transformer fusion algorithm, combined with LSTM neural network optimization training to generate the model. The AI ​​anomaly detection model can accurately identify sudden changes in temperature and humidity, sudden increases in CO2 concentration, packaging leakage, equipment malfunctions, frequent opening and closing of doors, and rapid deterioration of meat quality. The data calibration unit uses the weighted least squares method, combined with the position weights of several movable embedded sensing nodes, measurement accuracy, and environmental interference coefficients, to dynamically calibrate the collected data.

4. The intelligent monitoring and early warning system for temperature and humidity in cold chain transportation of meat products according to claim 1, characterized in that: The specific adjustment logic of the intelligent adaptive adjustment module is as follows: When the internal temperature of the meat exceeds the preset threshold by 0.3℃, the refrigeration unit power is automatically increased by 10%-30%, while the fresh air volume is increased by 5%-15%. When the CO2 concentration inside the packaging exceeds 5% VOL or a packaging leak is detected, the packaging sealing compensation device is activated to replenish the air pressure inside the packaging to the standard range. When the airflow velocity inside the carriage is below 0.2 m / s, adjust the angle of the fresh air system outlet to ensure uniform airflow distribution. When the surface humidity of meat products deviates from the target range of ±5%RH, start the humidification equipment or fresh air dehumidification function to adjust it.

5. The intelligent monitoring and early warning system for temperature and humidity in cold chain transportation of meat products according to claim 1, characterized in that: The meat deterioration trend prediction model adopts the attention mechanism enhanced LSTM algorithm, which integrates the Arrhenius equation and the Gompertz model for optimization. The input parameters include the meat microenvironment parameter change curve, meat quality parameters, transportation time, initial meat quality and compartment environment fluctuation data. The equipment health assessment model generates a health score of 0-100 based on equipment operating parameters and historical fault data. When the score is below 60, targeted maintenance suggestions are pushed.

6. The intelligent monitoring and early warning system for temperature and humidity in cold chain transportation of meat products according to claim 1, characterized in that: The four levels of anomalies and their response methods of the multi-terminal hierarchical collaborative early warning module are as follows: Level 1 (Slight): Data from a single movable embedded sensing node deviates from the preset threshold without a significant abnormal trend. Only the vehicle-mounted warning terminal issues an alert sound and records the abnormal data simultaneously. Level 2 (General): Multiple adjacent movable embedded sensing nodes show abnormal data, the abnormal trend is gradual, the vehicle terminal voice broadcasts, the dispatch center terminal pops up a prompt, and the management personnel receive text reminders on their mobile terminals; Level 3 (Severe): More than half of the mobile embedded sensing nodes show abnormal data, the meat microenvironment shows a significant deterioration trend, multiple terminals issue simultaneous early warnings, and the dispatch center can remotely issue adjustment instructions. Level 4 (Emergency): Core parameters exceed safety thresholds, meat products are at risk of spoilage, triggering multi-terminal emergency warnings, automatically dialing emergency numbers for management personnel, and simultaneously activating the highest level of emergency response from the intelligent adaptive adjustment module.

7. The intelligent monitoring and early warning system for temperature and humidity in cold chain transportation of meat products according to claim 1, characterized in that: The blockchain evidence storage unit adopts a consortium blockchain architecture, uploading monitoring data, early warning records, emergency response records, meat quality assessment reports, and route optimization records in real time. It also supports hierarchical queries by transportation companies, regulatory authorities, and consumers, achieving tamper-proof and traceable data throughout the entire process. The dynamic threshold generation module dynamically adjusts the early warning thresholds of various parameters based on meat type, transportation stage, transportation duration, and real-time monitoring data to adapt to the cold chain transportation needs of different meat products.

8. The intelligent monitoring and early warning system for temperature and humidity in cold chain transportation of meat products according to claim 1, characterized in that: It also includes a wireless charging and energy storage module, which is used to power the meat product full-dimensional sensing array and the edge computing fusion gateway. The wireless charging and energy storage module adopts electromagnetic induction wireless charging and has a built-in high-capacity lithium battery that can provide power for more than 72 hours without external power supply. It also has real-time power monitoring and low power warning functions.

9. The intelligent monitoring and early warning system for temperature and humidity in cold chain transportation of meat products according to claim 1, characterized in that: The movable embedded sensing node adopts a foldable snap-on installation structure to adapt to different specifications of meat packaging and vehicle structure, and fits tightly with the meat and vehicle after installation.

10. The intelligent monitoring and early warning system for temperature and humidity in cold chain transportation of meat products according to claim 1, characterized in that: The transportation route optimization module can dynamically adjust the transportation route and speed based on real-time road condition information, changes in ambient temperature along the route, meat quality prediction results, and transportation timeliness requirements, thereby reducing temperature fluctuations during transportation. The cloud-based intelligent management platform also includes a user permission hierarchical management module and a report generation module. The user permission hierarchical management module has permissions divided into administrators, dispatchers, drivers, and supervisors. Users with different permissions can view monitoring data and reports within their respective scopes. The report generation module can automatically generate daily / weekly / monthly monitoring reports, meat quality analysis reports, and equipment operation reports, and supports exporting and printing, facilitating full-process management and traceability.

Images