An internet of things-based cold chain transportation monitoring and early warning system

Abstract

The application relates to the field of Internet of Things and discloses a cold chain transportation monitoring and early warning system based on Internet of Things, which comprises a cloud control center, an end-side sensing node, an edge gateway and a local arrangement node; the edge gateway calculates equivalent external thermal load parameters and dynamic heat transfer coefficients in an offline working state, calculates thermal exhaustion time and generates a thermal vulnerability index; after receiving broadcast data, the local arrangement node calculates a scheduling priority score in combination with a time attenuation function and outputs a machine queue instruction for a cold chain container; in the flight stage, the edge gateway cuts off radio frequency power supply according to an absolute air pressure gradient to maintain silence, and the end-side sensing node performs timestamp binding and caching; and after the landing air pressure recovers, the data is transmitted again. Through the application, time quantization prior warning during network interruption is realized, on-site machine scheduling is closely matched with the thermal state of equipment, and radio frequency compliance and data traceability are considered.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) technology, specifically to an IoT-based cold chain transportation monitoring and early warning system. Background Technology

[0002] Air cold chain transportation is mainly used for goods that are highly sensitive to temperature conditions, such as pharmaceuticals, biological products, and high-end fresh produce. These goods typically undergo several stages during transport, including cold storage release, ground transfer, waiting on the tarmac for loading, flight, and unloading at the destination port. In these stages, the tarmac waiting and flight phases offer limited human intervention and are easily affected by communication conditions, ambient temperature, and aviation operating regulations. Therefore, this places high demands on the continuity, timeliness, and traceability integrity of the cold chain monitoring system.

[0003] In the existing technology, there are various IoT-based cold chain monitoring or early warning solutions. For example, CN120543064B discloses an IoT-integrated organic vegetable cold chain loss early warning system, which assesses risk by collecting cold chain environmental data and combining it with a loss calculation model; CN120013403B discloses an AI-based cold chain transportation route optimization method and system, focusing on optimizing cold chain transportation routes through predictive models; CN121836541A discloses an IoT-based intelligent monitoring system for the cold chain logistics environment, capable of collecting and monitoring environmental parameters such as temperature and humidity; CN119398641B discloses an AI-based big data-driven integrated smart cold chain management method and system; and IN202221064576A discloses an IoT-based cold chain management system and method. These solutions can play a role in data collection, status monitoring, route optimization, or platform management in ordinary cold chain transportation.

[0004] However, there are still shortcomings when applying the above technologies to the apron operations of aviation cold chain containers. First, existing cold chain monitoring systems often rely on wide area networks to upload data to the cloud for processing. When the container leaves the cold storage and enters the apron or is loaded into the aircraft, communication quality deteriorates or is restricted, making it difficult for cloud-based early warnings to be effective in a timely manner. Second, existing early warning methods mostly rely on fixed temperature thresholds, often triggering alarms only when the cargo temperature approaches or exceeds the safe temperature, making it difficult to determine in advance whether the container's current cooling capacity is sufficient to support the subsequent flight phase.

[0005] Furthermore, containers in the tarmac environment are affected not only by air temperature but also by solar radiation and heat absorption by the outer skin. Using only air temperature as an external thermal environment parameter can easily underestimate the actual thermal load on the container. Existing systems also typically do not incorporate the real-time thermal status of different containers into the loading sequence adjustment; ground operations still primarily rely on predetermined loading, hold arrangements, and manual scheduling procedures. For containers with reduced insulation capacity or higher thermal risks, continuing to wait for loading in a fixed sequence increases the risk of temperature exceeding limits.

[0006] Meanwhile, the active radio frequency transmission of the equipment during the aircraft loading and flight phases presents electromagnetic compatibility and operational management requirements. If existing equipment uses manual shutdown or a fixed countdown to shut down the communication module, it is easily affected by factors such as manual operation, flight delays, and taxiing waits; if monitoring is stopped directly, temperature data during the flight phase will be lost, which is not conducive to the traceability of the entire cold chain process.

[0007] Therefore, it is necessary to propose an IoT monitoring and early warning system suitable for aviation cold chain transportation scenarios, which can make localized thermal risk assessment when wide area network communication is limited, and provide a basis for apron loading scheduling and pre-loading early warning in combination with the actual thermal status of the container, while taking into account the radio frequency silence requirements during the flight phase and the integrity of temperature data traceability. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this invention provides an IoT-based cold chain transportation monitoring and early warning system, which solves the problems of lack of local pre-warning capabilities in wide area network disconnection scenarios, the disconnection of multiple equipment on-site installation and scheduling from the real-time thermal status of the container, and the existence of data monitoring blind spots during the flight radio frequency silence phase.

[0009] To achieve the above objectives, the present invention is implemented through the following technical solution: a cold chain transportation monitoring and early warning system based on the Internet of Things, including a cloud control center, end-side sensing nodes, edge gateways, and local orchestration nodes.

[0010] The cloud control center generates and sends task configuration parameters that include the upper limit of safe temperature and the estimated flight time. Specifically, upon receiving the pre-departure electronic instruction, the cloud control center initiates an initialization process, accesses external flight scheduling and cargo shipping databases, and obtains the target flight's schedule information and the attribute details of the cargo carried in the container. The cloud control center reads the historical average execution time of the same route from the flight scheduling database or the air traffic control plan for the day to determine the estimated flight time. The estimated flight time includes the target flight's airtime from takeoff to landing, the aircraft's estimated taxiing time before takeoff, and the estimated waiting time for the cargo to open after arrival at the destination. The cloud control center converts various types of cargo into corresponding Celsius limit points as the upper limit of safe temperature based on a mapping table.

[0011] The edge sensing node is used to collect internal load temperatures and transmit them to the edge gateway via a local wireless link. In one optional implementation, the edge sensing node includes a local storage module, a cache recording module, and a retransmission module. The cache recording module is used to determine that the local wireless link is disconnected when no handshake confirmation data packet is received from the edge gateway within several consecutive communication cycles, and then writes the collected internal load temperatures to the local storage module after binding them with the corresponding absolute timestamps. The retransmission module is used to supplement the uploaded cached internal load temperature sequence according to the absolute timestamp order after listening to the handshake beacon signal broadcast by the edge gateway.

[0012] The edge gateway is used to acquire surface temperature, ambient air temperature and internal load temperature in offline operation, calculate equivalent external heat load parameters, dynamic heat transfer coefficient, thermal depletion time and thermal vulnerability index, and send local broadcast data frames carrying thermal depletion time and thermal vulnerability index.

[0013] In one optional implementation, the edge gateway includes an equivalent external heat load parameter calculation module, a dynamic heat transfer coefficient determination and update module, a thermal depletion time boundary control module, a thermal vulnerability index generation module, an installation blockage judgment module, a barometric pressure sensor, and a radio frequency silence control module.

[0014] The equivalent external heat load parameter calculation module is used to obtain the difference between the surface temperature and the ambient air temperature, multiply the difference between the surface temperature and the ambient air temperature by a preset skin heat absorption coefficient to generate a temperature compensation value, and superimpose the temperature compensation value on the ambient air temperature to obtain the equivalent external heat load parameter.

[0015] As a further preferred option, when the surface temperature is less than or equal to the ambient air temperature, the equivalent external heat load parameter calculation module sets the temperature compensation value to zero.

[0016] The dynamic heat transfer coefficient determination and update module is used to calculate the dynamic heat transfer coefficient by comparing the rate of change of the internal load temperature with the equivalent internal and external temperature difference. The equivalent internal and external temperature difference is the difference between the equivalent external heat load parameter and the internal load temperature. In the initial stage of the calculation, the dynamic heat transfer coefficient determination and update module calls the pre-stored static thermal conductivity coefficient of the container as the initial value of the dynamic heat transfer coefficient, and continuously updates the dynamic heat transfer coefficient by introducing a sliding time window. When the absolute value of the difference between the equivalent external heat load parameter and the internal load temperature is less than the preset environmental noise fluctuation threshold, or when the dynamic heat transfer coefficient approaches zero, the dynamic heat transfer coefficient determination and update module pauses the update of the dynamic heat transfer coefficient and uses the value in the previous stable time window to participate in the current calculation.

[0017] The thermal exhaustion time boundary control module is used to calculate the time required for the internal load temperature to reach the upper limit of the safe temperature based on the equivalent external heat load parameters and the dynamic heat transfer coefficient, and use this as the thermal exhaustion time; when the internal load temperature is greater than or equal to the upper limit of the safe temperature, the thermal exhaustion time is forcibly assigned to zero and a local alarm signal is output; when the equivalent external heat load parameters are less than or equal to the upper limit of the safe temperature, the thermal exhaustion time is assigned to a preset maximum safe constant value.

[0018] The thermal vulnerability index generation module is used to assign a preset maximum penalty extreme value to the thermal vulnerability index when the internal load temperature is greater than or equal to the upper limit of the safe temperature; when the internal load temperature is less than the upper limit of the safe temperature, the difference between the upper limit of the safe temperature and the internal load temperature is used as the current temperature safety margin. When the current temperature safety margin is greater than zero, the dynamic heat transfer coefficient, the reciprocal of the thermal depletion time, and the reciprocal of the current temperature safety margin are normalized by the maximum-minimum linear normalization algorithm, respectively. The thermal vulnerability index is obtained by weighted summation of the normalized indicators based on the pre-allocated weight coefficients.

[0019] The installation blocking judgment module is used to receive the estimated flight time issued by the cloud control center, continuously read the real-time updated thermal exhaustion time, and compare the thermal exhaustion time with the estimated flight time. When the thermal exhaustion time is less than or equal to the product of the safety margin coefficient and the estimated flight time, a high-level trigger signal is generated to drive the external audible and visual alarm component to emit an alarm sound, set the installation blocking flag bit in the local broadcast data frame, and issue an installation blocking command to the local orchestration node.

[0020] The radio frequency silence control module is used to temporarily shut down the radio frequency transmission function after receiving the loading completion signal or the cabin door closing signal. It uses a barometric pressure sensor to acquire the absolute air pressure data of the external environment and calculates the time gradient rate of change of the absolute air pressure data between adjacent sampling points. When a continuous negative gradient is detected in the absolute air pressure data or the current absolute air pressure data falls into the cruise air pressure range, the power supply circuit of the radio frequency transceiver front end used to send local broadcast data frames is cut off on the hardware circuit to maintain the radio frequency silence state. When the absolute air pressure data shows a positive gradient change and stably falls into the ground standard air pressure range, the radio frequency silence state is exited and a handshake beacon signal is broadcast.

[0021] The local orchestration node is used to calculate scheduling priority scores based on local broadcast data frames and generate installation queue instructions for the containers based on the scheduling priority scores. In one optional implementation, the local orchestration node includes a broadcast parsing module, a lifecycle management module, a priority scoring module, and a queue output module. The broadcast parsing module is used to capture local broadcast data frames sent by multiple edge gateways within the communication coverage radius based on a carrier sense multiple access mechanism with collision avoidance, and to perform decoding operations on the received local broadcast data frames to extract the hardware identification codes of each container.

[0022] The lifecycle management module is used to establish a lifecycle timer for each hardware identification code. When no local broadcast data frame of a certain hardware identification code is received within a series of set time intervals, the anti-jitter waiting mechanism is activated. If communication is not restored within an additional preset anti-jitter window period, the corresponding container is determined to have left the current working area, and the data of the corresponding container is removed from the local cache queue.

[0023] The priority scoring module is used to calculate the scheduling priority score based on the thermal vulnerability index and the thermal depletion time.

[0024] The queue output module is used to advance the order of containers with earlier expected departure times of their assigned flights when multiple containers in the exchangeable candidate queue have the same scheduling priority score. When the expected departure times of the assigned flights are the same, the module arbitrates and sorts them according to the value of the hardware identification code and outputs the loading queue instruction.

[0025] Preferably, the priority scoring module is used to multiply the first weight coefficient by the thermal vulnerability index to obtain a first product, multiply the time decay constant by the thermal depletion time to obtain a second product, use the negative of the second product as the exponent of the natural constant to obtain an exponential decay term, and add the first product to the product of the second weight coefficient and the exponential decay term to obtain a scheduling priority score.

[0026] This invention provides a cold chain transportation monitoring and early warning system based on the Internet of Things, which has the following beneficial effects: This invention uses an edge gateway to acquire surface temperature and ambient air temperature offline to calculate equivalent external heat load parameters. It then continuously updates the dynamic heat transfer coefficient by combining this with the internal load temperature change rate, thereby deriving the thermal depletion time and thermal vulnerability index. This mechanism enables the system to perform thermophysical calculations independently of the cloud when the cold chain container is disconnected from the wide area network, transforming traditional post-event temperature exceedance alarms into pre-event time-quantitative warnings. This allows on-site personnel to take intervention measures based on the thermal depletion time before damage to the internal goods, reducing the risk of temperature runaway in network-off environments.

[0027] This invention utilizes local orchestration nodes to capture local broadcast data frames within the coverage area. The extracted thermal vulnerability index and thermal depletion time are then substituted into a time decay function to calculate a scheduling priority score, which is used to generate loading queue instructions for containers. This solution directly uses the dynamic thermophysical state inside the cold chain equipment and the degree of thermal intrusion from the external environment as the decision-making basis for multi-equipment on-site scheduling. In resource-constrained loading operations, it prioritizes the rapid loading of containers with weak heat resistance and short thermal depletion times, effectively reducing the disorderly waiting time of high-risk containers on the tarmac.

[0028] This invention utilizes an edge gateway and a barometric pressure sensor to calculate the temporal gradient rate of change of absolute air pressure data. Upon detecting a continuous negative gradient, it maintains radio frequency (RF) silence by cutting off the power supply to the RF transceiver front-end, ensuring electromagnetic safety during aircraft operation. Simultaneously, in conjunction with the local storage mechanism of the edge sensing nodes, temperature data is bound to an absolute timestamp and written to a non-volatile storage array during link interruptions. Once the flight lands and the air pressure returns to the standard ground pressure range, RF communication is restored and a supplementary upload is performed. This achieves effective compatibility between RF silence requirements during flight and data integrity throughout the cold chain process. Attached Figure Description

[0029] Figure 1 This is a physical architecture topology diagram of an Internet of Things-based cold chain transportation monitoring and early warning system according to an embodiment of the present invention; Figure 2 This is an overall flowchart of a cold chain transportation monitoring and early warning method based on the Internet of Things according to an embodiment of the present invention; Figure 3 This is a flowchart of a method for performing offline feedforward evaluation by an edge gateway according to an embodiment of the present invention; Figure 4 This is a flowchart of a method for performing dynamic priority scheduling on a local orchestration node according to an embodiment of the present invention; Figure 5 This is a comparison diagram of the internal load temperature changes of a container during a tarmac waiting period according to an embodiment of the present invention; Figure 6 This is a schematic diagram of the logical structure of an edge gateway according to an embodiment of the present invention. Detailed Implementation

[0030] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] like Figure 1 As shown, the present invention provides a cold chain transportation monitoring and early warning system based on the Internet of Things. The system includes a cloud control center, an edge gateway, end-side sensing nodes, and local orchestration nodes.

[0032] The cloud control center is deployed on a remote server and is equipped with a task distribution module. The cloud control center generates flight task configuration parameters that include the estimated flight duration and the upper limit of safe temperature, and sends the flight task configuration parameters to the edge gateway via the wide area network.

[0033] The edge gateway is installed in a pre-defined location on the outer skin of the container. The edge gateway is equipped with a microprocessor, a local area broadcast module, a barometric pressure sensor, a surface temperature probe, an ambient temperature probe, and an audible and visual alarm component. An industrial microservice component is deployed within the edge gateway to execute local computation and logical scheduling instructions when disconnected from the wide area network. In one optional implementation, the industrial microservice component includes a module for calculating equivalent external thermal load parameters, a module for determining and updating dynamic heat transfer coefficients, a module for controlling thermal depletion time boundaries, a module for generating thermal vulnerability indices, a module for determining installation blockages, and a radio frequency silence control module. Each functional module is invoked and executed by the microprocessor. The edge gateway and its associated sensors are powered by battery packs compliant with air cargo transport regulations.

[0034] like Figure 6 As shown, the surface temperature probe and ambient temperature probe collect surface temperature and ambient air temperature respectively, and input the collected results into the equivalent external heat load parameter calculation module; the end-side sensing node sends the internal load temperature to the dynamic heat transfer coefficient determination and update module; the equivalent external heat load parameter calculation module, the dynamic heat transfer coefficient determination and update module, the thermal depletion time boundary control module, and the thermal vulnerability index generation module sequentially output the equivalent external heat load parameters, dynamic heat transfer coefficient, thermal depletion time, and thermal vulnerability index; the barometric pressure sensor sends the absolute barometric pressure data to the radio frequency silence control module, which controls the radio frequency silence or resumption of the local broadcast module; the local broadcast module sends local broadcast data frames to the local orchestration node.

[0035] The surface temperature probe is mounted close to the outer skin of the container, or multiple temperature sampling points are distributed on the outer skin of the container and output representative surface temperature values ​​to the edge gateway. The ambient temperature probe is used to collect the ambient air temperature on the apron.

[0036] The edge sensing node is deployed within the load area of ​​the container and is equipped with a microcontroller, a local wireless communication module, a temperature sensor, and a local storage module. The edge sensing node uses the microcontroller to collect the internal load temperature at set time intervals and transmits the temperature data to the edge gateway via the local wireless link. In one optional implementation, the edge sensing node is also equipped with a buffer recording module and a retransmission module. The buffer recording module binds the internal load temperature with an absolute timestamp and writes it to the local storage module when the local wireless link is disconnected. The retransmission module replenishes and uploads the buffered data according to the absolute timestamp order after listening to the handshake beacon signal.

[0037] Local orchestration nodes are deployed on apron loading vehicles or ground support terminal equipment. Each local orchestration node is equipped with a radio frequency receiving module and a display control module, used to listen for local broadcast data frames emitted by the edge gateway and output loading queue instructions for the containers based on the data frame content. In one optional implementation, the local orchestration node further includes a broadcast parsing module, a lifecycle management module, a priority scoring module, and a queue output module; the broadcast parsing module parses local broadcast data frames, the lifecycle management module maintains the validity of data corresponding to each container, the priority scoring module calculates scheduling priority scores, and the queue output module outputs loading queue instructions.

[0038] See Figure 2 This invention provides a method for monitoring and early warning of cold chain transportation based on the Internet of Things, comprising the following steps: S1, the edge gateway connects to the cloud control center via the wide area network and receives flight mission configuration parameters. After the container leaves the cold storage area, the edge gateway disconnects the communication link with the cloud control center and enters offline working state. S2, the edge gateway acquires the internal load temperature uploaded by the end-side sensing node, the air ambient temperature collected by the ambient temperature probe, and the surface temperature collected by the surface temperature probe through a preset time period. The microprocessor uses the surface temperature and the air ambient temperature to calculate the equivalent external heat load parameters, and calculates the dynamic heat transfer coefficient by combining the rate of change of the internal load temperature. S3, the microprocessor calculates the time required for the internal load temperature to reach the upper limit of the safe temperature based on the equivalent external heat load parameters and the dynamic heat transfer coefficient, defines the time value as the heat depletion time, and outputs a local alarm signal when the internal load temperature is greater than or equal to the upper limit of the safe temperature. S4, the microprocessor calculates and generates a thermal vulnerability index by combining the dynamic heat transfer coefficient, thermal depletion time and current temperature safety margin. The edge gateway drives the local broadcast module to send local broadcast data frames containing the thermal vulnerability index and thermal depletion time in a loop. S5, the local orchestration node receives local broadcast data frames sent by multiple containers corresponding to the same flight. Under the premise of maintaining flight load balance and cabin allocation rules, it regenerates the loading queue instructions for each container in the exchangeable candidate queue according to the scheduling priority score calculated by the comprehensive thermal vulnerability index and thermal exhaustion time. S6, the edge gateway continuously reads the thermal exhaustion time and compares it with the expected flight time. When the thermal exhaustion time is less than or equal to the product of the safety margin coefficient and the expected flight time, the edge gateway triggers the external audible and visual alarm component and issues a blocking installation command. S7. After receiving the loading completion signal or the cabin door closing signal, the edge gateway controls the local broadcast module to enter the radio frequency silence state. When the barometric pressure sensor detects a continuous negative pressure gradient or enters the preset cruise pressure range, the edge gateway locks and maintains the radio frequency silence state. The end-side sensing node calls the local storage module to save the internal load temperature data. When the barometric pressure sensor detects that the pressure has risen and meets the ground landing pressure characteristics, the local broadcast module resumes the radio frequency transceiver function, and the end-side sensing node uploads the stored internal load temperature data to the edge gateway.

[0039] In this embodiment, the cloud control center is deployed on a remote server, which specifically adopts a public cloud server, a private cloud server, or a hybrid cloud data center. The cloud control center serves as the initial data configuration endpoint for the cold chain transportation monitoring and early warning system, used to preset local task parameters before the container enters the tarmac and disconnects from the network. The process of the cloud control center issuing tasks includes the following steps: S101, triggering the system initialization process and extracting basic data. The system initialization process begins when the container completes the physical packing of cold chain goods inside the cold storage, or when the cloud control center receives a pre-shipment electronic instruction from the warehouse management system. As a preferred method, the task distribution module configured within the cloud control center connects to external flight scheduling and cargo shipping databases to obtain the target flight's schedule information and the attribute details of the goods carried in the container, providing data support for subsequent parameter generation.

[0040] S102, Generate Flight Mission Configuration Parameters. The mission issuance module generates flight mission configuration parameters, including the estimated flight duration and upper limit of safe temperature, based on the acquired basic data. Considering the actual air transport scenario, the estimated flight duration covers the target flight's airtime from takeoff to landing, the aircraft's estimated taxiing time before takeoff, and the estimated waiting time for cargo hold opening after arrival at the destination, thus covering the closed periods in the cargo hold where manual intervention is difficult. These duration parameters can be determined by reading the historical average execution time of the same route from the flight scheduling database or the daily air traffic control plan.

[0041] The upper limit of safe temperature is a physical threshold set based on the specific category of goods inside the container. Since different categories of goods have varying tolerances to thermal shock, these categories include medical vaccines, fresh food, and temperature-sensitive chemicals. In this embodiment, the task distribution module converts each type of goods into corresponding Celsius temperature limits using a preset mapping table. This mapping table stores pre-entered empirical values ​​or industry standard values ​​for the maximum permissible ambient temperature corresponding to each type of goods.

[0042] S103, Establish a communication link and distribute flight mission configuration parameters. Before the container leaves the physical area of ​​the cold storage, the cloud control center establishes a data connection link with the edge gateway installed outside the container via a wide area network (WAN). The specific communication entity of the WAN can be a fourth-generation mobile communication network, a fifth-generation mobile communication network, or a narrowband Internet of Things (IoT). The cloud control center packages and sends the flight mission configuration parameters to the edge gateway. After receiving and parsing the data, the edge gateway writes it into its local storage unit as the basic constraint condition for subsequent offline calculations.

[0043] For the information encryption and network handshake verification process during wide area network (WAN) data transmission, those skilled in the art can implement it using standard transmission control protocols and advanced encryption standard algorithms. The underlying data packet encapsulation and parsing mechanisms are well-known technologies in the field and will not be elaborated upon here. Once the parameters are configured and the container is detected moving out of the cold storage area, the edge gateway actively disconnects the WAN communication link and officially enters offline independent operation mode to adapt to the working environment of the apron without fixed network coverage. To achieve accurate physical sensing of the cold storage's departure, the edge gateway is also equipped with an RFID reader module or a wireless LAN receiver module. Specifically, the detection logic is triggered by the edge gateway's RFID reader module scanning the RFID access control tag deployed at the cold storage exit, or by the edge gateway's wireless LAN receiver module detecting the drop or loss of the dedicated wireless LAN signal inside the cold storage for status determination.

[0044] In this embodiment, the edge gateway, acting as an offline computing node, is installed at a predetermined location on the outer skin of the container. To avoid compromising the structural integrity of the aviation equipment, the edge gateway is installed using a non-intrusive fixing structure. Specifically, this fixing structure includes aviation-grade strap components or high-strength magnetic clips, enabling secure attachment of the equipment without altering the container's external dimensions or loading locking relationship. The edge gateway and its associated sensors are equipped with low-power batteries. As a preferred option, the low-power batteries are lithium thionyl chloride batteries or flame-retardant solid-state batteries that meet the United Nations standards for safe air transport of dangerous goods, in order to satisfy the relevant regulations on electromagnetic emission and physical safety for air cargo transport.

[0045] See Figure 3When the container leaves the cold storage area and the WAN signal is interrupted, the industrial microservice component deployed within the edge gateway initiates local control logic. This component isolates the WAN communication process at the underlying operating system level, centrally allocating microprocessor computing power to local data acquisition and algorithm inference tasks. The microprocessor executes a feedforward evaluation mechanism in an offline environment, specifically including the following steps: S201, Extracting Equivalent External Heat Load Parameters. The microprocessor synchronously reads surface temperature data collected by the surface temperature probe and ambient air temperature data collected by the ambient temperature probe according to a preset time period. In the high-temperature, direct sunlight environment of the tarmac, a single air temperature reading is insufficient to accurately reflect the radiative thermal shock experienced by the container skin. To quantify the non-standard thermal environment characteristics experienced by the container, the microprocessor obtains the difference between the surface temperature and the ambient air temperature and multiplies this difference by a preset skin heat absorption coefficient to generate a temperature compensation value. The skin heat absorption coefficient is pre-calibrated based on the optical reflectivity and thermal conductivity of the container's outer surface material, with a value range set between 0.1 and 0.8. The microprocessor superimposes the calculated temperature compensation value onto the ambient air temperature to obtain the equivalent external heat load parameters. Specifically, the equivalent external heat load parameters are calculated according to the following formula: ; in, Indicates the first Equivalent external thermal load parameters at each sampling time. Indicates the first The ambient air temperature at each sampling time. Indicates the first Surface temperature at each sampling time Indicates the heat absorption coefficient of the skin. This indicates that the larger of the two values ​​within the parentheses is used. When the surface temperature is higher than the ambient air temperature, the temperature compensation value is calculated using the difference between the two values. When the surface temperature is less than or equal to the ambient air temperature, the temperature compensation value is zero.

[0046] S202, Fitting the Dynamic Heat Transfer Coefficient. The microprocessor acquires the internal load temperature data uploaded by the end-side sensing node and calculates the rate of change of the internal load temperature by combining it with the historical temperature sequence of the previous time period. Based on the basic principle of heat conduction, the rate of temperature change of the system is positively correlated with the internal and external temperature difference, and this proportional relationship reflects the overall heat transfer capacity of the system. To avoid errors caused by static factory insulation parameters that fail due to aging, the microprocessor calculates the ratio of the aforementioned rate of change of the internal load temperature to the equivalent internal and external temperature difference at the current moment to obtain the dynamic heat transfer coefficient. This equivalent internal and external temperature difference is defined as the difference between the equivalent external heat load parameter and the internal load temperature at the current moment. Specifically, the first... The rate of change of internal load temperature at each sampling time is calculated according to the following formula: ; No. The equivalent internal and external temperature difference at each sampling time is calculated using the following formula: ; No. The dynamic heat transfer coefficient at each sampling time is calculated according to the following formula: ; To reduce the impact of single-point sampling noise, the sliding time window includes When there are 10 valid sampling points, the dynamic heat transfer coefficient is updated according to the following formula: ; in, Indicates the first The rate of change of internal load temperature at each sampling time Indicates the first Internal load temperature at each sampling time Indicates the first Each sampling time, Indicates the first The equivalent internal and external temperature difference at each sampling time. Indicates the first The dynamic heat transfer coefficient at each sampling time. This indicates the number of valid sampling points within the sliding time window. This indicates the sampling point number within the sliding time window, and The range of values ​​is to , Indicates the first The rate of change of internal load temperature at each sampling time Indicates the first Equivalent external thermal load parameters at each sampling time. Indicates the first Internal load temperature at each sampling time This represents the summation of the ratios corresponding to each valid sampling point within the sliding time window. Indicates the time window within the sliding time window The average value of the calculated heat transfer coefficients corresponding to each of the effective sampling points is calculated. When the ambient noise fluctuation is less than the preset threshold, or when the dynamic heat transfer coefficient approaches zero, the update of the dynamic heat transfer coefficient is paused, and the value from the previous stable time window is used in the current calculation.

[0047] In the initial calculation phase after the system starts up and leaves the cold storage, due to the lack of sufficient time-series samples, the microprocessor uses the pre-stored static thermal conductivity coefficient of the container as the initial value for the dynamic heat transfer coefficient. By introducing a sliding time window to continuously update the dynamic heat transfer coefficient, the system can adaptively reflect the actual physical insulation degradation of the container.

[0048] S203, calculate the thermal depletion time and execute boundary convergence control. After obtaining the dynamic heat transfer coefficient, the microprocessor calculates the time required for the internal load temperature to reach the upper limit of the safe temperature based on the feedforward evaluation model. This embodiment uses the following formula to calculate the thermal depletion time: ; in, Indicates the first Thermal depletion time at each sampling moment Indicates the first The dynamic heat transfer coefficient at each sampling time. Indicates the first Equivalent external thermal load parameters at each sampling time. Indicates the first Internal load temperature at each sampling time This indicates the upper limit of the safe temperature range.

[0049] To address the complexity of the apron operating environment and ensure the mathematical convergence and logical robustness of the feedforward evaluation model under extreme conditions, the microprocessor applies specific boundary convergence rules to the calculation process. When the internal load temperature exceeds or equals the upper limit of the safe temperature, it is determined that the cargo has exceeded the safe range, the microprocessor forcibly assigns a thermal depletion time of 0, and outputs a local alarm signal.

[0050] When the equivalent external heat load parameter is less than or equal to the upper limit of the safe temperature, the external thermal environment is determined to be in a safe state. The microprocessor assigns a preset maximum safety constant value to the heat depletion time. This maximum safety constant value can be set as an empirical multiple of the expected flight duration to characterize the safety time margin. When the absolute value of the equivalent internal and external temperature difference is less than the preset environmental noise fluctuation threshold or the calculated dynamic heat transfer coefficient approaches zero, to prevent the denominator of the formula from approaching zero, causing algorithm divergence or division by zero anomalies, the microprocessor pauses updating the dynamic heat transfer coefficient and uses the value from the previous stable time window in the current feedforward calculation. The specific value of the aforementioned environmental noise fluctuation threshold can be preset by those skilled in the art based on the inherent quantization error amplitude of the temperature sensor. The calibration process is well-known in the art and will not be elaborated here.

[0051] S204, Generating the Thermal Vulnerability Index. To achieve cross-equipment apron collaborative scheduling, the microprocessor transforms the extracted multidimensional physical parameters into dimensionless scheduling criteria. The microprocessor extracts the current dynamic heat transfer coefficient, thermal exhaustion time, and current temperature safety margin. The current temperature safety margin is the difference between the upper limit of the safe temperature and the internal load temperature. The microprocessor converts the above three parameters into a thermal vulnerability index through a preset mapping algorithm. The higher the index value, the weaker the container's ability to resist external thermal shock. To avoid low-level arithmetic overflow anomalies, the microprocessor performs a pre-check of the safety boundary: when the internal load temperature is greater than or equal to the upper limit of the safe temperature (i.e., the thermal exhaustion time is 0, and the current temperature safety margin is less than or equal to 0), the microprocessor directly assigns the thermal vulnerability index to the system's preset maximum penalty extreme value and skips the subsequent reciprocal conversion calculation.

[0052] Within the normal temperature range, the dynamic heat transfer coefficient is positively correlated with the thermal vulnerability index, while the time to heat depletion and the current temperature safety margin are negatively correlated with the thermal vulnerability index. The current temperature safety margin is calculated using the following formula: ; The maximum-minimum linear normalization function is calculated according to the following formula: ; The thermal vulnerability index is calculated using the following formula: ; The weighting coefficients satisfy the following relationship: ; in, Indicates the first The current temperature safety margin at each sampling time. Indicates the upper limit of safe temperature. Indicates the first Thermal vulnerability index at each sampling time Indicates the first The dynamic heat transfer coefficient at each sampling time. Indicates the first Thermal depletion time at each sampling moment Represents the maximum-minimum linear normalization function. This represents the index value to be normalized. and These represent the minimum and maximum values ​​of the corresponding indicators within the preset statistical range, respectively. , , This represents the pre-assigned weighting coefficients.

[0053] The microprocessor normalizes the values ​​of the dynamic heat transfer coefficient, the reciprocal of the thermal depletion time, and the reciprocal of the current temperature safety margin using the maximum-minimum linear normalization algorithm. Based on the pre-assigned weight coefficients, it performs a weighted summation of the normalized indicators, thereby outputting a thermal vulnerability index that can comprehensively characterize multidimensional thermophysical features.

[0054] In this embodiment, the surface temperature probe and ambient temperature probe configured on the edge gateway are responsible for capturing multidimensional physical and thermal data of the container in the tarmac environment. The probe components provide basic physical parameters for thermodynamic calculations in the offline state of the system by performing the following data acquisition and preprocessing process: S301, Physical Deployment and Multi-Point Sampling Fusion of the Surface Temperature Probe. The surface temperature probe is mounted close to the outer skin of the container to directly sense the real-time physical thermal state of the container surface. To capture the skin heat accumulation effect caused by solar radiation, as a preferred method, the surface temperature probe is deployed in the heat-sensitive area of ​​the container's outer skin, specifically covering the top panel and large-area side metal walls. In specific engineering implementations, thermally conductive silicone grease is applied between the surface temperature probe and the outer skin to eliminate air gaps, and it is physically fixed by covering it with weather-resistant aluminum foil tape, thereby reducing the thermal resistance of the measurement contact.

[0055] When multiple temperature sampling points are distributed across the outer skin of the container, each probe node synchronously collects local surface temperatures and transmits them to the edge gateway via the underlying bus. The edge gateway's built-in data preprocessing logic mathematically fuses the multiple collected values. To ensure the safety and redundancy of the thermodynamic assessment, the edge gateway compares all probe data collected at the same time and extracts the highest temperature value as the representative surface temperature value for that moment; alternatively, the edge gateway performs a weighted average calculation on multiple temperature data points based on pre-set panel area weights for each sampling point, thereby outputting a representative surface temperature value that reflects the overall heating state of the container. In this weighted calculation process, the top panel, with its larger heated area, has a higher calculation weight for its corresponding probes than the side panels.

[0056] S302, Radiation-shielded deployment and periodic data acquisition of the ambient temperature probe. The ambient temperature probe is configured independently of the container skin surface and is specifically designed to collect the ambient air temperature on the apron. To prevent interference from heat conduction from the container housing on air temperature measurement, the ambient temperature probe can be deployed on the bottom cantilever bracket of the edge gateway housing or in a shaded area on the side of the container. To avoid inflated air temperature readings due to direct sunlight, the external cover of the ambient temperature probe's sensing end is equipped with a miniature louvered radiation-shielding structure. This structure blocks direct sunlight while allowing surrounding air to pass through and convectively exchange heat with the sensing end. A low-level wired communication link is established between the ambient temperature probe and the edge gateway's microprocessor via an internal integrated circuit bus or serial peripheral interface. Considering the gradual changes in ambient thermodynamics, the ambient temperature probe periodically wakes up and reads the current ambient air temperature at a set time sampling interval. This time sampling interval is configured to be 1 to 5 minutes to maintain the edge gateway's low-power operation while meeting the real-time requirements of temperature monitoring.

[0057] S303 provides a joint output and compensation benchmark for multi-source thermal data. In an airstrip exposed environment, the container simultaneously experiences convective heat transfer from the surrounding air and radiative heat transfer from the sun. If only air temperature is provided at the sensing level, subsequent heat transfer calculations will underestimate the rate of temperature rise of the cargo inside the container. Surface temperature probes and ambient temperature probes are strictly time-stamped, synchronously outputting the extracted representative surface temperature values ​​and ambient air temperature to the microprocessor. These two temperature parameters, with clearly defined physical differences, form the basis for calculating the equivalent external heat load parameters. The difference between surface temperature and ambient air temperature quantifies the degree to which solar radiation is converted into skin heat energy.

[0058] After receiving the combined data, the microprocessor can use the difference in conjunction with the skin's heat absorption coefficient to calculate the temperature compensation value. As a robust logical determination mechanism, when the surface temperature value is greater than the ambient air temperature, the microprocessor executes the above compensation deduction normally; when the surface temperature value is detected to be less than or equal to the ambient air temperature, such as at night, on a rainy day, or in a cold storage transfer passage, the microprocessor determines that there is no significant radiative heat enhancement effect at that moment, directly sets the temperature compensation value to zero, and only uses the ambient air temperature in subsequent calculations.

[0059] The analog-to-digital conversion instructions and hardware interface driver programming logic for the underlying temperature sensor can be implemented by those skilled in the art based on conventional microcontroller development manuals. The sensor electrical signal reading and digital conversion process is a well-known technology in this field and will not be described in detail here.

[0060] In this embodiment, the edge gateway, through its built-in local broadcast module and barometric pressure sensor, performs communication data interaction and physical interception control in scenarios without wide area network coverage and requiring compliance with aviation electromagnetic silence requirements. Specifically, this includes the following steps: S401, Assembly and transmission of local area broadcast (LAN) data frames. During the tarmac exposure phase, when disconnected from the wide area network (WAN), the LAN module is used to transmit dynamic scheduling data. As a preferred method, the LAN module employs Bluetooth Low Energy (BLE) or long-range low-power wireless LAN (WLAN) technology. The microprocessor extracts the currently calculated thermal depletion time and thermal vulnerability index, combines them with the container's hardware identification code and the assigned flight number, and encapsulates them into standard data frames according to a preset communication protocol. The LAN module cyclically broadcasts this data frame to the surrounding physical space at set time intervals, allowing local orchestration nodes on the tarmac to listen and parse parameters. To balance the real-time nature of data updates with the device's battery life, this set time interval can be configured between 5 and 30 seconds, depending on battery capacity and tarmac scheduling frequency.

[0061] S402, offline interception control logic based on feedforward results. While the container is waiting on the tarmac or being loaded into the aircraft cargo hold, the system needs to prevent the cargo from exceeding temperature limits during subsequent flight periods when manual intervention is not possible. The edge gateway continuously reads the real-time updated thermal exhaustion time value and compares it with the locally stored estimated flight duration parameter. To cope with sudden delays or long taxiing times during actual flight operations, the system introduces a safety margin coefficient to weight and amplify the estimated flight duration. The microprocessor executes the specific interception decision mechanism, whose decision conditions are expressed by the following formula: ; in, Indicates the first Thermal depletion time at each sampling moment This represents the safety margin coefficient. This indicates the estimated flight time.

[0062] The safety margin coefficient is greater than 1, and those skilled in the art can set it to a range of 1.2 to 1.5 based on historical on-time performance data of the target route. When the above criteria are met, it indicates that the container's existing cooling capacity is insufficient to support the completion of the entire flight mission. In this case, the edge gateway's underlying controller generates a high-level trigger signal, driving the external audible and visual alarm component to emit a strobe light and an alarm sound. Simultaneously, the edge gateway sets the blocking loading flag in the next cyclic broadcast data frame, sending a blocking loading command to the local orchestration node deployed on the ground operations terminal, prompting on-site personnel to stop the loading operation of the current high-risk cargo. Conversely, when the thermal depletion time is greater than the weighted estimated flight time, it indicates that the cooling capacity reserve is safe, and the edge gateway maintains the normal data frame broadcast state.

[0063] S403, a physical dual-confirmation triggered RF silence mechanism. To comply with electromagnetic compatibility (EMC) safety standards during aircraft flight, the edge gateway employs a dual-confirmation strategy that combines operational commands and physical status to control RF transmission permissions. The edge gateway receives loading completion signals from ground crew equipment via external input ports, or determines the cabin door closure status by receiving signals from dedicated RFID tags inside the aircraft cargo hold. Upon receiving these signals, the edge gateway sends an initial silence command to the local broadcast module, temporarily disabling RF transmission. Due to the possibility of accidental signal transmission by tarmac personnel or prolonged flight delays on the ground, the edge gateway utilizes a barometric pressure sensor for secondary physical status confirmation.

[0064] The barometric pressure sensor continuously acquires absolute air pressure data of the external environment, and the microprocessor calculates the rate of change of the air pressure data over time between adjacent sampling points. When the aircraft takes off and is in the climb phase, the air pressure inside the cargo hold will show a continuous downward trend, eventually stabilizing within the set cruise pressure range. This cruise pressure range is set between 700 hPa and 850 hPa. When the microprocessor detects a continuous negative gradient in the air pressure data curve (indicating the aircraft is in takeoff and climb), or when the current air pressure value falls within the cruise pressure range, the edge gateway confirms entry into real flight mode. At this time, the microprocessor cuts off the power supply circuit of the local broadcast module's RF transceiver front-end in the hardware circuit, locking and maintaining RF silence throughout the entire flight phase. If, within the preset waiting time after temporarily disabling the RF transmission function, the edge gateway neither detects a continuous negative gradient in the air pressure data nor detects the air pressure value falling within the cruise pressure range, the microprocessor determines that the aircraft is still on the ground, immediately cancels the initial silence command, restores the RF transmission function of the local broadcast module, and re-enters the loop broadcast and anomaly interception monitoring state to prevent monitoring dead zones. Taking into account the regular taxiing and queuing time of aircraft from the apron to the takeoff runway, the preset waiting time is configured to be 30 to 60 minutes.

[0065] For the RF enable pin control logic and register configuration process of the underlying wireless communication chip, those skilled in the art can refer to the hardware development manual of the corresponding microprocessor for operation. The hardware register-level configuration and pin level control process are well-known technologies in the field and will not be described in detail here.

[0066] In this embodiment, the end-side sensing node and its storage module are used to monitor the thermal state of the load area inside the container and cache data in the in-flight network interruption environment after flight takeoff. To achieve continuous monitoring of important physical parameters of cold chain cargo, the end-side sensing node performs the following monitoring and recovery steps: S501: Acquisition and Local Wireless Transmission of Internal Load Temperature. End-side sensing nodes are deployed within the load area of ​​the container. As a preferred approach, considering the thermodynamic distribution gradient of the container's internal space, the end-side sensing nodes are fixed to the surface of the cargo packaging near the inside of the container doors or near the top, heat-prone areas to obtain the most representative internal load temperature. The end-side sensing nodes incorporate high-precision temperature sensors, acquiring the internal load temperature at set time intervals and transmitting the temperature data to the edge gateway via a local wireless link. In practical engineering, this set time interval is configured to be 1 to 10 minutes. To address the physical attenuation of radio waves by the container's metal walls, the specific implementation of this local wireless link can utilize the highly penetrating 433MHz or 868MHz industrial radio frequency bands, or the interference-resistant low-power Bluetooth communication technology.

[0067] S502 employs an offline buffer triggering mechanism based on communication link status. When the edge gateway enters RF silence due to aircraft takeoff, its hardware circuitry disconnects the receive and transmit loops, ceasing to respond to wireless signals from the internal payload area. After sending internal payload temperature data to the edge gateway, the end-side sensing node starts a local wait timer. If no handshake confirmation data packet is received from the edge gateway within 3 to 5 consecutive communication cycles, the microcontroller within the end-side sensing node determines that the current local wireless link has been disconnected.

[0068] To prevent physical loss of temperature monitoring data during flight, the underlying controller of the edge sensing node triggers offline caching logic, calling the local storage module to save internal payload temperature data. This local storage module specifically employs a low-power electrically erasable programmable read-only memory (EEPROM) or flash memory chip. Throughout the RF silent phase, the edge sensing node maintains its original temperature sampling frequency, binding and encapsulating each collected internal payload temperature data with its corresponding system absolute timestamp to form a structured log record, which is then sequentially written to the non-volatile storage array of the local storage module. During this process, if the available storage space of the local storage module reaches a preset limit, the edge sensing node employs a cyclic overwrite strategy, based on timestamp order, to overwrite the oldest historical data with newer data, ensuring the effective retention of critical temperature records for the current and near-term moments.

[0069] S503, a communication recovery and data archiving process based on air pressure recovery characteristics. As a flight arrives at its destination and lands, the air pressure inside the aircraft cargo hold gradually increases. When the air pressure sensor on the edge gateway continuously detects a positive gradient change in external air pressure, and the current air pressure value recovers and stabilizes within the standard ground pressure range, the edge gateway determines that the flight mission has ended. This standard ground pressure range is set between 950 hPa and 1050 hPa based on the physical parameters of typical low-altitude airports. Due to the special geographical environment of high-altitude airports, as a safety redundancy design, the edge gateway can adaptively adjust this standard pressure range downward based on pre-sent destination airport elevation data; alternatively, when the edge gateway detects that the positive air pressure gradient change approaches zero and remains so for a preset time, it also determines that the flight mission has ended. After landing is determined, the edge gateway's local broadcast module exits the radio frequency silent state, resumes radio frequency transceiver functions, and rebroadcasts the handshake beacon signal to the local area.

[0070] After the edge sensing node detects the beacon signal via its radio frequency antenna, it confirms the re-establishment of the local wireless link. Subsequently, the edge sensing node initiates a supplementary upload mechanism, reading the cached internal load temperature data sequence from its local storage module in chronological order of timestamps, packaging it, and uploading it to the edge gateway. Upon receiving and verifying the data, the edge gateway sends an archiving confirmation command to the edge sensing node. Upon receiving this command, the edge sensing node marks or erases the uploaded data blocks from its local storage module to free up storage space, thereby achieving complete archiving of the cargo hold temperature data chain throughout the entire cold chain transportation cycle.

[0071] See Figure 4 In this embodiment, the local orchestration node serves as the decision-making center for the collaborative scheduling of multiple containers on the apron, undertaking the role of data aggregation and task distribution in exposed operations outside the wide area network. To adapt to the physical environment of the airport airside, the local orchestration node is mounted on special ground support vehicles such as apron towing vehicles or lifting platform vehicles in the form of an industrial-grade ruggedized tablet computer, or equipped by ground support personnel in the form of a ruggedized mobile smart terminal. The local orchestration node achieves dynamic allocation of apron resources by executing the following decentralized monitoring and evaluation scheduling process: S601, Concurrent Listening and Parameter Parsing of Local Broadcast Data Frames. In routine flight support operations, multiple containers belonging to different flights may be parked within the same physical area of ​​the apron. The RF receiving module inside the local orchestration node is in a scanning listening state, capturing local broadcast data frames sent by various edge gateways within its communication coverage radius. Since simultaneous broadcasting by multiple devices can easily cause signal collisions in the spatial channel, as a preferred approach, the underlying local orchestration node employs a carrier sense multiple access (CSMA) mechanism with collision avoidance for signal discrimination and reception.

[0072] The local orchestration node decodes received valid local broadcast data frames, extracting the hardware identification code, flight number, thermal exhaustion time, and thermal vulnerability index of each container. To prevent data stagnation caused by edge gateways moving out of communication range, the local orchestration node establishes a lifecycle timer for each hardware identification code. If no new local broadcast data frame for a certain hardware identification code is received within three consecutive set time intervals, the local orchestration node activates a de-jitter waiting mechanism due to temporary radio frequency obstruction caused by large metal vehicles shuttling across the apron. If communication is still not restored within an additional preset de-jitter window, the local orchestration node determines that the container has left the current operating area and removes its corresponding data from the local buffer queue, thereby maintaining the real-time performance and accuracy of scheduling data.

[0073] S602 performs multi-dimensional dynamic priority evaluation calculations. After obtaining real-time parameters from multiple containers within the region, the local orchestration node needs to determine the operational sequence of ground support vehicles to control the overall risk of heat buildup. The local orchestration node calculates a scheduling priority score for each container in the buffer queue based on a built-in overall scheduling algorithm. This algorithm comprehensively considers the container's current structural heat resistance and the urgency of the time dimension; its mathematical calculation process is expressed by the following formula: ; ; in, Indicates scheduling priority score, Indicates the first weighting coefficient; This represents the second weighting coefficient. Represents the natural constant. Indicates thermal vulnerability index, Indicates the time of heat depletion. This represents the time decay constant.

[0074] In the above formula, by introducing an exponential decay term with a natural base, the algorithm can assign a larger calculation proportion to containers facing imminent thermal depletion, thereby prioritizing processing under time-critical conditions. The specific values ​​of the first and second weighting coefficients can be empirically calibrated by those skilled in the art based on the median daily ambient temperature of the local airport apron, with their sum set to 1. The time decay constant is used to adjust the sensitivity to time urgency, and its value can be configured between 0.1 and 0.5.

[0075] S603, Generation and Sorting Conflict Handling of Scheduling Instructions. After calculating the scheduling priority scores of all containers to be processed, the local orchestration node, under the constraints of imported external flight load balancing and cabin allocation rules, extracts containers with the same load attributes and meeting the safety exchange conditions as exchangeable candidate queues. Within this exchangeable candidate queue, the local orchestration node sorts the containers in descending order of their scores, generating optimized loading queue instructions for the current region. Containers with higher scores have a higher priority in the scheduling operation sequence, such as physical loading or transfer to a cool, sheltered area. In actual queuing calculations, extreme cases may occur where two or more containers have the same scheduling priority score, causing the sorting algorithm to enter a logical dead zone that cannot be resolved.

[0076] To resolve this sorting conflict, the local orchestration node introduces the estimated flight departure time as a secondary criterion in such cases, prioritizing containers belonging to flights with earlier estimated departure times. If the estimated departure times are still the same, to avoid disorderly oscillations in the queue during interface refreshes, the local orchestration node performs a final arbitration sort based on the lexicographical order or numerical value of the container hardware identification codes. After sorting, the local orchestration node outputs the loading queue instructions and the corresponding container hardware identification codes to the screen interface for visualization, directly guiding ground personnel to drive the tractor unit to carry out targeted physical loading and unloading operations.

[0077] For the rendering logic of the local orchestration node screen interface, the process scheduling of the underlying operating system, and the underlying channel switching instructions of the radio frequency receiving module, those skilled in the art can rely on the conventional smart terminal development framework to write programs. The rendering of the graphical interface and the driving process of the underlying hardware are well-known technologies in this field and will not be described in detail here.

[0078] In a specific application example, two containers belonging to the same flight (defined as Container A and Container B) are assumed to be present simultaneously, exiting the cold storage and awaiting loading in the same high-temperature exposure environment on the tarmac. Container A carries temperature-sensitive medical supplies, while Container B carries general fresh produce. The estimated flight time is set at 5 hours.

[0079] Task assignment and parameter configuration: After container A and container B are physically packed inside the cold storage, the task distribution module of the cloud control center is activated. Based on the acquired cargo attributes, the cloud control center generates a task distribution module that includes the estimated flight time. ) and the upper limit of safe temperature ( The flight mission configuration parameters for container A. Taking container A as an example, its upper limit of safe temperature ( The temperature is configured to 8.0℃. The cloud control center sends these parameters to the edge gateways of container A and container B via the wide area network. After the container leaves the cold storage area, the edge gateway disconnects its communication link with the cloud control center and enters offline operation. Figure 5 The starting point of the 0-minute interval for apron exposure time on the horizontal axis.

[0080] Offline feedforward evaluation and data broadcasting (corresponding) Figure 5 (0 to 25 minutes) During the tarmac exposure period, the edge gateway of container A acquires the internal load temperature uploaded by the end-side sensing node at a preset time period. The microprocessor uses the surface temperature and ambient air temperature collected by external probes to calculate the equivalent external heat load parameters. ).

[0081] Combination Figure 5 The curve shown indicates that, within the interval of 0 to 25 minutes, the internal load temperature of container A gradually increased from an initial 2.0°C to approximately 4.6°C. During this process, the microprocessor calculated the dynamic heat transfer coefficient based on the rate of change of the internal load temperature. And calculate the time required for the internal load temperature to reach the upper limit of the safe temperature according to the formula, that is, the thermal depletion time. ): ; As the tarmac exposure time increases, the difference between the equivalent external heat load parameter and the internal load temperature decreases, resulting in a shorter calculated thermal depletion time. Then, the microprocessor integrates the dynamic heat transfer coefficient, thermal depletion time, and current temperature safety margin to generate a thermal vulnerability index. The edge gateway drives the local broadcast module to cyclically send local broadcast data frames containing the thermal vulnerability index and the thermal depletion time.

[0082] Dynamic scheduling calculation and priority installation intervention (corresponding) Figure 5 (25th minute) At this time, the local orchestration node deployed on the ground handling terminal equipment receives local broadcast data frames from container A and container B. If a traditional First-In-First-Out (FIFO) scheduling strategy is adopted, the queuing order of the containers depends only on their physical arrival time at the apron, without considering differences in internal thermal states. For example... Figure 5 As shown by the solid line (marked with a triangle), if the conventional strategy is followed and waiting continues, the internal load temperature of container A will continue to rise and will cross the "safe upper temperature limit (8.0°C)" reference line in the figure between approximately 40 and 45 minutes of exposure time, resulting in overheating damage to the cargo.

[0083] In this invention, the local orchestration node recalculates the scheduling priority score for each container within the exchangeable candidate queue based on a built-in algorithm. The calculation formula is as follows: ; in, Indicates the first weighting coefficient; This represents the second weighting coefficient; This represents the time decay constant.

[0084] Because container A has a lower upper limit for safe temperature and a shorter thermal depletion time, its exponential decay term has a weighting amplification effect, resulting in a higher scheduling priority score for container A compared to container B. Based on this, the local orchestration node advances container A's position in the installation queue instruction. Figure 5 The star-shaped marker at the 25th minute (corresponding to the legend "Priority Installation Intervention Point") and the text label "Trigger Priority Installation" represent the time node when the above scheduling instructions were issued and the physical mounting operation was executed.

[0085] Post-intervention effect comparison and radiofrequency silencing archiving: Container A responded to the loading command at 25 minutes and was prioritized for transfer to the temperature-controlled aircraft cargo hold. (As...) Figure 5 As shown by the dashed line (with a circular mark, corresponding to the legend "Dynamic Priority Scheduling Strategy of the Invention"), after the intervention was executed, the internal load temperature of container A stopped rising due to being removed from the extreme thermal environment, and then gradually dropped and remained within a stable range. It did not reach the upper limit of the safe temperature during the entire subsequent monitoring period.

[0086] After loading is complete, the edge gateway receives the cabin door closing signal and controls the local broadcast module to enter radio frequency silence mode, meeting aviation safety requirements. During flight, the edge sensing nodes continuously access the local storage module to save internal payload temperature data until the aircraft lands and the air pressure rises. Afterward, the edge sensing nodes supplement the stored internal payload temperature data and upload it to the edge gateway for archiving, thus constituting... Figure 5 The data recording loop is complete within the area indicated by the dashed line. By implementing the above method, the system effectively avoids the risk of temperature-limit exceeding failure caused by static queuing on the apron.

[0087] The above description is merely some specific implementations of this application and is not intended to limit the scope of protection of this application. Any variations or substitutions easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A cold chain transportation monitoring and early warning system based on the Internet of Things, characterized in that, include: The cloud-based control center is used to generate and send mission configuration parameters, including the upper limit of safe temperature and the expected flight duration. End-side sensing nodes are used to collect internal load temperatures and transmit them to the edge gateway via a local wireless link. The edge gateway is used to acquire surface temperature, ambient air temperature and internal load temperature in offline working state, calculate equivalent external heat load parameters, dynamic heat transfer coefficient, thermal depletion time and thermal vulnerability index, and send local broadcast data frames carrying the thermal depletion time and thermal vulnerability index. The local orchestration node is used to calculate the scheduling priority score based on the local broadcast data frame, and generate the loading queue instruction for the container based on the scheduling priority score.

2. The cold chain transportation monitoring and early warning system based on the Internet of Things as described in claim 1, characterized in that, The edge gateway includes an equivalent external heat load parameter calculation module, which is used to obtain the difference between the surface temperature and the ambient air temperature, multiply the difference between the surface temperature and the ambient air temperature by a preset skin heat absorption coefficient to generate a temperature compensation value, and superimpose the temperature compensation value on the ambient air temperature to obtain the equivalent external heat load parameter.

3. The cold chain transportation monitoring and early warning system based on the Internet of Things as described in claim 2, characterized in that, The edge gateway further includes a dynamic heat transfer coefficient determination and update module, used for: The dynamic heat transfer coefficient is obtained by calculating the ratio of the rate of change of the internal load temperature to the equivalent internal and external temperature difference. The equivalent internal and external temperature difference is the difference between the equivalent external heat load parameter and the internal load temperature. In the early stage of the calculation, the pre-stored static thermal conductivity coefficient of the container is called as the initial value of the dynamic heat transfer coefficient. The dynamic heat transfer coefficient is continuously updated by introducing a sliding time window. And when the absolute value of the difference between the equivalent external heat load parameter and the internal load temperature is less than the preset environmental noise fluctuation threshold or the dynamic heat transfer coefficient approaches zero, the update of the dynamic heat transfer coefficient is paused, and the value in the previous stable time window is used in the current calculation.

4. The cold chain transportation monitoring and early warning system based on the Internet of Things as described in claim 3, characterized in that, The edge gateway also includes a thermal exhaustion time boundary control module, which is used to force the thermal exhaustion time to be zero and output a local alarm signal when the internal load temperature is greater than or equal to the upper limit of the safe temperature; and to assign a preset maximum safe constant value to the thermal exhaustion time when the equivalent external thermal load parameter is less than or equal to the upper limit of the safe temperature.

5. The cold chain transportation monitoring and early warning system based on the Internet of Things as described in claim 1, characterized in that, The edge gateway includes a thermal vulnerability index generation module, which is used to assign a preset maximum penalty extreme value to the thermal vulnerability index when the internal load temperature is greater than or equal to the upper limit of the safe temperature; when the internal load temperature is less than the upper limit of the safe temperature, the difference between the upper limit of the safe temperature and the internal load temperature is used as the current temperature safety margin, and when the current temperature safety margin is greater than zero, the values ​​of the dynamic heat transfer coefficient, the reciprocal of the heat depletion time and the reciprocal of the current temperature safety margin are normalized respectively, and the thermal vulnerability index is obtained by weighted summation of the normalized indicators based on the pre-allocated weight coefficients.

6. The cold chain transportation monitoring and early warning system based on the Internet of Things as described in claim 1, characterized in that, The edge gateway includes an installation blocking detection module, used for: Receive the estimated flight time from the cloud control center, continuously read the real-time updated thermal exhaustion time, and compare the thermal exhaustion time with the estimated flight time; When the thermal depletion time is less than or equal to the product of the safety margin coefficient and the expected flight time, a high-level trigger signal is generated to drive the external audible and visual alarm component to emit an alarm sound, set the blocking installation flag bit in the local broadcast data frame, and issue a blocking installation command to the local orchestration node.

7. The cold chain transportation monitoring and early warning system based on the Internet of Things as described in claim 1, characterized in that, The edge gateway includes a barometric pressure sensor and an RF silence control module. The RF silence control module is used to temporarily disable the RF transmission function after receiving a loading completion signal or a cabin door closing signal. It uses the barometric pressure sensor to acquire absolute barometric pressure data of the external environment and calculates the time gradient rate of change of the absolute barometric pressure data between adjacent sampling points. When a continuous negative gradient is detected in the absolute pressure data or the current absolute pressure data falls into the cruise pressure range, the power supply circuit of the radio frequency transceiver front end used to send the local broadcast data frame is cut off on the hardware circuit to maintain radio frequency silence.

8. The cold chain transportation monitoring and early warning system based on the Internet of Things as described in claim 7, characterized in that, The end-side sensing node includes: The cache recording module is used to determine that the local wireless link is disconnected when no handshake confirmation data packet is received from the edge gateway within multiple consecutive communication cycles, and to bind the collected internal load temperature with the corresponding absolute timestamp and write it into the local storage module. The retransmission module is used to supplement the uploaded internal load temperature sequence according to the absolute timestamp order after listening to the handshake beacon signal broadcast by the edge gateway.

9. The cold chain transportation monitoring and early warning system based on the Internet of Things as described in claim 1, characterized in that, The local orchestration nodes include: The broadcast parsing module is used to capture local broadcast data frames sent by multiple edge gateways within the communication coverage radius based on a carrier sense multiple access mechanism with collision avoidance, and to perform decoding operations on the received local broadcast data frames to extract the hardware identification code of each unit. The lifecycle management module is used to establish a lifecycle timer for each hardware identification code. When no local broadcast data frame of a certain hardware identification code is received within a series of set time intervals, the anti-jitter waiting mechanism is activated. If communication is not restored within an additional preset anti-jitter window period, it is determined that the corresponding container has left the current working area, and the data of the corresponding container is removed from the local cache queue.

10. The cold chain transportation monitoring and early warning system based on the Internet of Things as described in claim 9, characterized in that, The local orchestration nodes include: The priority scoring module is used to determine the scheduling priority score based on the thermal vulnerability index and the thermal depletion time. The queue output module is used to advance the order of containers with earlier expected departure times of their assigned flights when multiple containers in the exchangeable candidate queue have the same scheduling priority score. When the expected departure times of the assigned flights are the same, the modules are arbitrated and sorted according to the value of the hardware identification code, and the loading queue instruction is output.

Images