Gradient heat preservation bag digital modeling simulation system for fresh food distribution and heat preservation bag

Abstract

The application relates to the technical field of cold chain logistics, and discloses a gradient heat preservation bag digital modeling simulation system for fresh food distribution and a heat preservation bag. The system comprises a heat preservation bag module, an edge computing module and a simulation calculation module. The edge computing module comprises a microcontroller module, a memory module, a temperature sensor module and a heat injection circuit module. The simulation calculation module collects boundary temperature data to calculate a temperature gradient vector and a change rate, and combines local temperature rise data of a heat injection pulse to calculate dynamic thermal impedance. Then, a clock stretching duration compensation value and a local electrical distortion index are calculated in combination with the above parameters. When the distortion index is greater than a safety threshold, the microcontroller module truncates the service data flow of a communication bus, marks the current physical sector as a heat isolation state, and redirects data to write into a standby physical sector. The application overcomes communication and storage failures caused by parasitic parameter drift under large temperature difference, and guarantees the integrity of cold chain traceability data.

Description

Technical Field

[0001] This invention relates to the field of cold chain logistics technology, specifically to a digital modeling and simulation system for gradient insulated bags and insulated bags for fresh food delivery. Background Technology

[0002] With the popularization of fresh food e-commerce and pharmaceutical cold chain logistics, temperature control solutions using insulated bags combined with built-in cold sources have become the mainstream logistics and distribution method. In order to achieve full-process traceability of the delivery process, edge computing devices are usually attached to the surface of the insulated bags to collect and persistently record data such as temperature, humidity, and geographical location in real time.

[0003] In actual delivery operations, especially when the external ambient temperature is high and cryogenic media such as dry ice are used internally, there is a significant absolute temperature difference between the inside and outside of the insulated bag. This temperature difference creates a nonlinear physical temperature gradient along the thickness of the insulated bag wall and on the attached edge computing device circuit board. Continuous thermodynamic transfer and localized condensation cause nonlinear thermal drift in the equivalent parasitic capacitance of the circuit board traces, while the charge tunneling effect inside the flash memory chip slows down significantly under low-temperature penetration.

[0004] Existing cold chain tracking devices generally employ static communication baud rate configurations and fixed memory programming timeout wait strategies. When the physical parameters of the underlying hardware change due to temperature gradients, the fixed communication bus timing cannot meet the setup and hold time requirements of digital level signals, leading to data transmission errors. Simultaneously, the increased flash memory write time can cause the microprocessor's read / write driver to timeout deadlock, resulting in corruption of the file system allocation table.

[0005] Because existing technologies lack feedforward modeling and compensation mechanisms for the coupling relationship between microscopic thermal field characteristics and electrical distortions of electronic components, systems can typically only passively restart via watchdog circuits after a substantial hardware crash. This delayed error handling approach cannot proactively mitigate risks before physical link deterioration, leading to frequent irreversible loss of core cold chain business tracking data, making it difficult to meet the stringent data integrity requirements for high-value goods. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a digital modeling and simulation system for gradient insulated bags and insulated bags for fresh food delivery. It solves the problem that existing cold chain tracking equipment lacks a dynamic compensation mechanism for thermal drift of hardware electrical parameters under large temperature difference conditions, which can easily lead to underlying communication and storage failures and result in the loss of traceability data throughout the entire process.

[0007] To achieve the above objectives, the present invention provides the following technical solution: The first aspect of the present invention provides a digital modeling and simulation system for gradient insulated bags for fresh food delivery, including an insulated bag body, an edge computing module and a simulation computing module. The edge computing module includes a microcontroller module, a memory module divided into physical sectors and connected to the microcontroller module via a serial communication bus, a temperature sensor module and a heat injection circuit module.

[0008] The simulation module collects boundary temperature data through the temperature sensor module and calculates the temperature gradient vector and the rate of change of temperature with respect to time. The microcontroller module controls the heat injection circuit module to generate DC pulses, and the simulation module collects local temperature rise data to calculate dynamic thermal impedance. The simulation module combines the temperature gradient vector and dynamic thermal impedance to calculate the clock stretch duration compensation value, and combines the clock stretch duration compensation value to calculate the local electrical distortion index. When the local electrical distortion index exceeds a pre-calibrated safety threshold, the microcontroller module truncates the service data stream of the serial communication bus and maintains the network heartbeat command, marks the physical sector as thermally isolated, and redirects newly written data to a spare physical sector.

[0009] Furthermore, the temperature sensor module includes a local temperature sensor located on the low-temperature side of the printed circuit board (PCB) and a reference temperature sensor located around the serial communication bus and memory module. The simulation calculation module uses numerical interpolation algorithms to establish a continuous thermal field model of the PCB's two-dimensional plane, solves for spatial partial derivatives to obtain the temperature gradient vector at specified coordinates of the PCB, and performs backward difference operations on continuously sampled data to obtain the rate of temperature change with time. The heat injection circuit module includes a copper trace module and a switching device module located in the inner layer area of ​​the PCB. The first end of the copper trace module is electrically connected to the ground network, and the second end of the copper trace module is connected to the system power rail through the switching device module. The control electrode of the switching device module is connected to a pin of the microcontroller module configured as a push-pull output mode.

[0010] Furthermore, the simulation module uses algebraic subtraction to calculate the relative difference between the temperature rise of the local temperature sensor and the temperature rise of the reference temperature sensor relative to the initial injection moment, extracting the effective temperature rise step peak. Combining the effective temperature rise step peak, the actual heating power parameters of the heat injection circuit module, and the pulse duration, the dynamic thermal impedance of the printed circuit board's physical medium is calculated. This differential processing, by eliminating common-mode thermal noise introduced by ambient airflow disturbances, allows the obtained thermal impedance value to characterize the degree of thermal conductivity attenuation caused by local condensation or thermal stress on the printed circuit board substrate.

[0011] To achieve the mapping transformation from physical fields to electronic control variables, under the condition of thermal gradient, the simulation calculation module obtains the physical direction unit vector of the microcontroller module to the target memory module on the printed circuit board. The temperature gradient vector is then multiplied by the physical direction unit vector to characterize the actual spatial projection component of the temperature gradient along the communication path. The simulation calculation module further combines the static thermal drift empirical coefficient (characterizing the weight of the influence of spatial temperature bias on the parasitic capacitance of the serial communication bus), the dynamic thermal impedance, the dynamic thermal drift empirical coefficient (characterizing the weight of temperature transient shocks on clock source frequency perturbation), and the rate of change of temperature with respect to time to calculate the clock stretching duration compensation value. This compensation value is used to adjust the bus signal level setup time to reduce the impact of hardware parameter drift on data transmission integrity.

[0012] Furthermore, the local electrical distortion index integrates the transient distortion weighting coefficient of the current communication cycle with the clock stretching duration compensation value, the cumulative thermal stress weighting coefficient with the cumulative clock stretching duration compensation value, and uses a discrete-time step summation to approximate the integral term. When the accumulated sum is detected to exceed the maximum limit allowed by the physical data type, the simulation calculation module actively freezes the accumulation step. This index integrates transient physical distortion and historical thermal fatigue variables to quantify the operational risks of semiconductor underlying communication.

[0013] Furthermore, the microcontroller module obtains the absolute temperature value of the physical coordinates of the memory module at the current moment and calculates the dynamic waiting timeout threshold matching the hardware temperature. The microcontroller module sequentially pushes write enable operation instructions, page programming control instructions containing the target address, and the actual operating condition data payload bitstream to the memory module via the serial communication bus, and reads the busy flag status of the memory module's internal status register. When the busy flag does not toggle before the dynamic waiting timeout threshold is exhausted, and the number of consecutive read / write retries initiated by the underlying communication driver module exceeds the set redundancy error tolerance limit, the microcontroller module releases control of the underlying hardware.

[0014] Furthermore, the microcontroller module internally runs an embedded operating system containing a file system layer. The microcontroller module queries the file system layer for the latest logical address-to-physical sector mapping table. Based on dynamically refreshed status labels, the file system layer bypasses coordinate intervals marked as thermally isolated and extracts the starting base address of a thermally stable spare physical sector as the target address. When the magnitude of the temperature gradient vector and the dynamic thermal impedance return to the set normal value range, the microcontroller module's underlying driver layer cancels the clock stretching duration compensation value, the file system layer releases the thermal isolation state of the corresponding physical sector, and the system restores normal communication bandwidth and the original address space mapping.

[0015] The second aspect of the present invention provides an insulated bag for fresh food delivery, which is applied to the digital modeling and simulation system for the gradient insulated bag for fresh food delivery provided in the first aspect. The bag includes an inner wall structure, an outer wall structure, and a heat-insulating solid layer disposed between the inner wall structure and the outer wall structure. The inner wall structure and the outer wall structure are spatially closed and define an internal cavity for containing goods.

[0016] This insulated bag restricts the rate of heat exchange between the inside and outside through an insulating solid layer, creating a non-linear, continuous physical temperature gradient distribution along the wall thickness. An externally attached digital modeling and simulation system utilizes the boundary conditions formed by the internal cold source medium and the external environment to perform thermal field reconstruction and underlying hardware timing protection, ensuring the integrity of data retention and traceability throughout the entire process of cryogenic delivery.

[0017] This invention provides a digital modeling and simulation system for gradient insulated bags and the insulated bag itself, designed for fresh produce delivery. It offers the following advantages: 1. This invention uses a microcontroller module to schedule the heat injection circuit to generate controlled pulses, and uses a simulation calculation module to perform differential calculations on the synchronous temperature rise data collected by the temperature sensor to calculate the dynamic thermal impedance of the printed circuit board. This mechanism uses differential logic to cancel the common-mode environmental thermal noise introduced by the cold convection airflow inside the insulation bag, thereby objectively quantifying the degree of thermal conductivity attenuation caused by local condensation or thermophysical stress of the physical substrate, providing accurate underlying physical field data support for subsequent hardware timing compensation of the system.

[0018] 2. This invention uses a microcontroller module to schedule the heat injection circuit to generate controlled pulses, and uses a simulation calculation module to perform differential calculations on the synchronous temperature rise data collected by the temperature sensor to calculate the dynamic thermal impedance of the printed circuit board. This mechanism uses differential logic to cancel the common-mode environmental thermal noise introduced by the cold convection airflow inside the insulation bag, thereby objectively quantifying the degree of thermal conductivity attenuation caused by local condensation or thermophysical stress of the physical substrate, providing accurate underlying physical field data support for subsequent hardware timing compensation of the system.

[0019] 3. This invention uses a logical remapping mechanism at the file system layer to trigger thermal isolation and data redirection of the physical sectors of the memory when the local electrical distortion index exceeds a safety threshold. This hardware and software collaborative blocking strategy combines a dynamic waiting timeout threshold to prevent deadlock, enabling the system to transfer core business data streams in advance before extreme cold penetration causes the underlying programming of the flash memory chip to fail. This avoids damage to the file system allocation table and ensures the complete retention of temperature and humidity traceability data throughout the fresh food delivery process. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the system architecture of the present invention; Figure 2This is a schematic diagram of the thermal insulation bag body structure of the present invention; Figure 3 This is a flowchart of the method of the present invention; Figure 4 This is a physical cross-section of the thermal insulation bag and a schematic diagram of the spatial topology of the printed circuit board of the present invention. Figure 5 This is a schematic diagram of the underlying hardware and software control architecture and the hot injection circuit topology of the present invention; Figure 6 This is a diagram showing the internal algorithm execution logic and cross-layer collaborative architecture of the simulation calculation module of the present invention; Figure 7 This is a schematic diagram illustrating the system state evolution and collaborative repair effect verification under extreme thermal conditions of the present invention.

[0021] The components are: 1. the insulated bag body; and 2. the heat-insulating solid layer. Detailed Implementation

[0022] 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.

[0023] Please see the appendix Figure 1 and attached Figure 2 This invention provides a digital modeling and simulation system for gradient insulated bags for fresh food delivery, which may include: an insulated bag body 1, an edge computing module, and a simulation computing module.

[0024] The edge computing module is located on the surface of the insulated bag body 1. The insulated bag body 1 contains a cold source medium, while its exterior is in contact with the ambient space. A physical temperature gradient exists between the inner and outer walls of the insulated bag body 1.

[0025] The edge computing module includes a microcontroller module, a memory module, a temperature sensor module, and a heat injection circuit module. All components of the edge computing module are integrated on the same printed circuit board.

[0026] The microcontroller module includes pins, timer modules, and a bus controller module with a clock stretching setting register. Internally, the microcontroller module runs an embedded operating system, which includes a file system layer.

[0027] The memory module is a flash memory chip with physical sector partitioning. The memory module is electrically connected to the bus controller module via a serial communication bus. The memory module receives and records operating data sent by the microcontroller module.

[0028] The temperature sensor module includes multiple temperature sensor nodes arranged at different physical coordinates on the printed circuit board. The temperature sensor module includes a local temperature sensor located on the low-temperature side of the printed circuit board, and a reference temperature sensor located around the serial communication bus and memory module.

[0029] The hot-injection circuit module includes a copper trace module and a switching device module. The copper trace module is located in the inner layer area of ​​the printed circuit board. The first terminal of the copper trace module is electrically connected to the ground network, and the second terminal of the copper trace module is connected to the system power rail through the switching device module. The control terminal of the switching device module is connected to the pin of the microcontroller module.

[0030] The simulation computing module is deployed inside the microcontroller module or in the communication gateway device. The simulation computing module establishes a data connection with the edge computing module, receives physical field data, and outputs control parameters to the microcontroller module.

[0031] See attached document Figure 3 This invention provides a method based on timing compensation and logic remapping, comprising the following steps: S100, the simulation calculation module collects the boundary temperature data of the inner and outer sides of the heat preservation bag body 1 through the temperature sensor module, establishes a two-dimensional thermal field model of the printed circuit board using the interpolation algorithm, and calculates the temperature gradient vector of the specified coordinates of the printed circuit board and the rate of change of temperature with time. S200, the microcontroller module outputs a drive level to control the switching device module of the heat injection circuit module to turn on, the copper trace module of the heat injection circuit module generates a DC pulse, the simulation calculation module collects the local temperature rise data of the temperature sensor module, and calculates the dynamic thermal impedance of the current physical medium of the printed circuit board through the differential function. S300, the simulation calculation module combines the temperature gradient vector and dynamic thermal impedance to calculate the clock stretching duration compensation value of the serial communication bus, and feeds the clock stretching duration compensation value back to the microcontroller module. S400, the simulation calculation module calculates the local electrical distortion index based on the clock stretching duration compensation value and its integral accumulation over time. The microcontroller module obtains the local electrical distortion index and compares it with the set safety threshold to make a judgment. When the local electrical distortion index exceeds the set safety threshold, the microcontroller module’s underlying driver layer cuts off the business data flow of the serial communication bus and maintains the network heartbeat command on the serial communication bus. The file system layer marks the physical sector of the memory module corresponding to the coordinate with the highest dynamic thermal impedance value as thermally isolated and redirects the newly written data to the spare physical sector. S600: When the magnitude of the temperature gradient vector and the dynamic thermal impedance return to the set normal value range, the underlying driver layer of the microcontroller module cancels the clock stretch duration compensation value, and the file system layer releases the thermal isolation state of the corresponding physical sector.

[0032] Please see the appendix Figure 4 In this embodiment, the insulated bag body 1 serves as the physical entity that carries fresh produce and edge computing hardware. It mainly includes an inner wall structure, an outer wall structure, and a heat insulation layer disposed between the two. The inner wall structure and the outer wall structure are spatially closed and define an internal cavity for accommodating the goods.

[0033] Based on this physical entity, the insulation layer constitutes the core physical barrier that blocks the exchange of heat between the inside and outside. To ensure that the system can establish temperature gradient characteristics that can be effectively measured by downstream algorithms, the insulation layer 2 is usually made of an industrial material with low thermal conductivity. This material can be polyurethane foam, polypropylene foam, or vacuum insulation board, and its equivalent thermal conductivity is usually limited to the range of 0.01 to 0.04 watts per meter Kelvin under normal delivery ambient temperatures. The physical reason for choosing this range is that if the thermal conductivity is higher than the upper limit of this range, the heat between the inside and outside will be quickly neutralized, and it will be impossible to provide a thermodynamic boundary with clear distinction for the system; if the thermal conductivity is too low, it will increase the manufacturing cost of the insulation equipment many times over. The outer wall structure is usually woven from a polymer waterproof and flame-retardant fabric, and the inner wall structure is made of aluminum foil fiberglass composite material with heat radiation reflection function. For the zipper closure structure and load-bearing strap sewing process of the insulation bag body 1, those skilled in the art can directly implement them in combination with conventional bag manufacturing processes. The mechanical forming and sewing sealing methods are well-known technologies in the field and will not be described in detail here.

[0034] The internal cavity of the insulated bag body 1 contains a cold source medium, such as common solid carbon dioxide, phase change gel cold storage agent, or ordinary water ice pack. The cold source medium continuously absorbs heat from the surrounding environment within the internal cavity, forming a stable low-temperature thermodynamic field on the surface of the inner wall structure and in the surrounding air. This low-temperature thermodynamic field constitutes the internal cold-side temperature boundary for the entire system's operation, and its absolute temperature value is determined primarily by the critical temperature at which the selected cold source medium undergoes a physical phase change. Simultaneously, the outer wall structure of the insulated bag body 1 is directly exposed to the external atmospheric environment, absorbing heat from the ambient air or solar radiation, thus constituting the external hot-side temperature boundary of the system.

[0035] Because of the significant absolute temperature difference between the internal cold-side temperature boundary and the external hot-side temperature boundary under spatial physical isolation, heat will spontaneously be conducted from the outer wall structure to the inner wall structure through the insulation layer 2. The physical rate of heat transfer is directly limited by the thermal resistance of the medium itself and the spatial transmission distance. During this physical process, a nonlinear continuous physical temperature gradient distribution will form along the wall thickness direction of the insulation bag body 1. This temperature gradient directly acts on the edge computing device attached to the surface of the insulation bag, constituting the basic thermal conditions that cause parameter drift of electronic components.

[0036] The heat flux density of the internal and external environments of the insulation bag body 1 and the distribution of the physical temperature gradient follow the basic heat conduction equation. In the spatial dimension perpendicular to the surface of the insulation bag, to facilitate computational optimization by edge computing nodes at discrete time points, a continuous differential model or a discrete difference substitution model based on a finite spatial step size is typically used in practical engineering to characterize the physical field. This embodiment takes a continuous model as an example. Based on the basic heat conduction mechanism, the heat flux density along the wall thickness direction of the insulation bag body 1 is directly proportional to the equivalent thermal conductivity of the insulation layer 2 and the actual temperature gradient along this spatial physical direction. The specific value of the equivalent thermal conductivity is determined based on the standard thermophysical property test calibration parameters of the selected material at the time of manufacture.

[0037] When the system performs discretization difference operations to solve for the actual temperature gradient, an anti-boundary check control strategy must be activated simultaneously. If the difference between the spatial distance coordinates of two adjacent temperature measurement points approaches 0, the system will forcibly assign a preset minimum tolerance value for hardware physical thickness as a substitute parameter in the calculation, thereby effectively preventing system arithmetic overflow and crashes caused by a denominator of 0 in division operations. The heat transfer characteristics of the insulation bag body 1 in this spatial dimension provide basic physical boundary input data for the external digital modeling and simulation system.

[0038] In this embodiment, the edge computing module, serving as the system's data acquisition and control hub, is attached to a designated outer wall surface of the insulated bag body 1. To ensure that the edge computing module can effectively sense cold source infiltration inside the insulated bag and withstand real physical environmental changes, the module is typically fixed to the geometric center area of ​​the outer wall of the insulated bag body 1 using industrial-grade thermally conductive structural adhesive or mechanical fasteners. Regarding the specific injection molding and waterproof sealing methods for the equipment shell, those skilled in the art can directly implement conventional IP67-level industrial protection standards. The selection of the waterproof and breathable membrane and the assembly process of the shell sealing ring are well-known technologies in the field and will not be elaborated upon here.

[0039] As the physical substrate supporting various electronic components, the core carrier within the edge computing module is the printed circuit board (PCB). To support subsequent dynamic thermal impedance measurement and thermal field modeling, the substrate of this PCB needs to be a material with stable anisotropic thermal conductivity. This substrate can be a standard FR-4 epoxy glass cloth copper-clad laminate or a polyimide flexible copper-clad laminate. To ensure that the delay characteristics of the heat conduction process meet the system control requirements, the equivalent thermal conductivity of the PCB substrate in the direction perpendicular to the board thickness is typically limited to a range of 0.2 to 0.4 Kelvin per meter. If the substrate's thermal conductivity is too high, external heat will penetrate the board instantly, causing the system to be unable to capture transient thermal resistance; if it is too low, it cannot reflect the actual changes in coldness inside the insulation bag in a timely manner. This type of substrate exhibits physically different equivalent thermal conductivity in the planar direction and the direction perpendicular to the thickness. This physical property determines that the coldness transferred from the inner wall of the insulation bag and the heat introduced from the external environment will form a thermal diffusion distribution with a specific time constant on the PCB. This thermal delay characteristic provides the necessary and sufficient measurement time window for the microcontroller module to capture transient thermodynamic parameters.

[0040] Based on the aforementioned hardware foundation, the edge computing module integrates a microcontroller module, a memory module, a temperature sensor module, and a heat injection circuit module. In the two-dimensional spatial layout of the printed circuit board, to reduce the interference of the chip's own heat generation on the ambient temperature sampling, the microcontroller module and memory module are typically centrally located in the core thermal equilibrium area of ​​the printed circuit board. In this embodiment, this core thermal equilibrium area is specifically defined as the physical geometric center of the printed circuit board, or a gentle area confirmed by prior thermodynamic simulation calibration where the actual surface temperature gradient amplitude is less than 0.1 Kelvin per millimeter. Meanwhile, the multiple independent sampling nodes contained in the temperature sensor module are discretely arranged near the edge of the circuit board and the surrounding area of ​​the data serial communication bus traces. Regarding the electrical connection topology of the power and ground lines, the system adopts a wiring strategy of physical isolation between star ground and analog ground / digital ground. This topology effectively cuts off electromagnetic noise interference caused by the ground bounce generated by the heat injection circuit module when generating large current DC pulses, which affects the weak analog level signal acquired by the temperature sensor module.

[0041] Considering the characteristics of communication buses in practical engineering, since the temperature sensor module contains multiple independent nodes distributed at different physical coordinates, a hardware polling communication time difference will occur when the microcontroller module reads data from each node one by one through the serial communication bus. To eliminate the distortion of the physical field state time coordinate caused by this asynchronous sampling, the system performs forced time reference alignment on the multi-source heterogeneous data before constructing the thermal model. At the specific data processing level, the system uses the high-precision hardware timer inside the microcontroller module as the reference clock source to acquire the raw absolute temperature data read by each temperature sensor node at different absolute sampling times.

[0042] Subsequently, based on a pre-defined general numerical calculation principle, the system introduces a numerical interpolation algorithm to perform time projection transformation on the multi-source data. This numerical interpolation algorithm can be a linear interpolation algorithm based on a first-order Taylor expansion, or a quadratic polynomial interpolation algorithm or a cubic spline interpolation algorithm. Through the above mathematical projection calculation, the system can uniformly map the discretely acquired raw absolute temperature data to the same pre-defined reference timestamp, thereby generating temperature compensation values ​​after time reference alignment. The above data synchronization preprocessing mechanism based on the interpolation algorithm achieves accurate reconstruction of the absolute time profile of multi-source physical field data without the need for additional high-cost synchronization hardware sampling.

[0043] To prevent anomalies in the underlying driver software during computation, the system incorporates divisor boundary checks and defense logic before performing the aforementioned time-aligned interpolation. When the underlying logic performs any form of interpolation or differentiation, if the absolute value of the time interval between two adjacent sampling moments is detected to be less than a set time jitter threshold, it indicates that the system may be in a high-priority interrupt blocking state or a severely congested communication bus. In this case, the denominator in the conventional mathematical interpolation algorithm will approach 0, potentially triggering an arithmetic overflow in the underlying kernel. Under this specific condition, the system will bypass the current-value interpolation algorithm processing module based on a multi-dimensional fault-tolerance mechanism and forcibly assign the temperature compensation value, after time-base alignment, to the latest raw absolute temperature data collected by the corresponding node. This boundary-crossing fault-tolerance mechanism eliminates the risk of kernel crashes caused by division by 0, ensuring the robustness of the edge computing module under harsh communication conditions from the underlying coding level.

[0044] Please see the appendix Figure 5 In this embodiment, the microcontroller module serves as the data processing and peripheral scheduling unit of the edge computing module, responsible for coordinating the underlying hardware drivers and upper-layer data processing. To meet the system's requirements for real-time computing power and control of underlying hardware registers, the lower-level features of the microcontroller module are implemented as an industrial-grade 32-bit microprocessor chip with a floating-point arithmetic unit and a memory protection unit, such as a microcontroller based on the ARM Cortex-M4 instruction set or RISC-V instruction set architecture.

[0045] Based on the aforementioned hardware, the microcontroller module integrates pins, a timer module, and a bus controller module. The general-purpose input / output pins of the microcontroller module are electrically interconnected to the control electrodes of the switching devices in the heat injection circuit module and the data bus interface of the temperature sensor module. To ensure that the switching devices can quickly overcome the Miller plateau and enter a deep saturation conduction state upon receiving control commands, the pins connected to the control electrodes of the switching devices are forcibly set to push-pull output mode during the low-level hardware initialization phase, and their transient source current drive capability is set to be greater than or equal to 8 mA.

[0046] The selection of these specific drive parameters is based on the principle that only by providing sufficient transient gate charge extraction and injection capabilities can the hot-injection pulse be guaranteed to have steep edge characteristics, thereby providing a clean step excitation signal for downstream dynamic thermal impedance measurement. The timer module includes at least one independent hardware timer peripheral, whose period setting parameters are set according to the time resolution requirements for acquiring dynamic thermal impedance. The clock prescaler and reload value of the timer module are set to an operating mode capable of generating microsecond-level interrupts to ensure that the duration of the pulse width and the sampling timestamp of the local temperature rise data have strict hardware-level accuracy guarantees during bypass controlled hot injection.

[0047] In the specific data interaction interface design, the data link between the microcontroller module and the external memory module relies on the bus controller module. In this embodiment, the bus controller module adopts a hardware peripheral controller that supports internal integrated circuit bus protocols or serial peripheral interface protocols. Its internal physical register group includes a clock stretching setting register and a baud rate prescaler register with read and write permissions. This bus controller module supports active intervention in bus communication timing at the hardware level.

[0048] In harsh operating conditions where there is a significant physical temperature gradient in the insulation bag, the parasitic capacitance of the traces on the printed circuit board will drift nonlinearly with the ambient temperature and the intrusion of cold sources. This will lead to a prolonged setup time of the digital level signal on the serial communication bus, inducing data errors. To quantify and compensate for this hardware drift effect from the underlying mechanism, the driver layer built into the microcontroller module performs equivalent modeling based on the transient charging physical characteristics of the bus digital signal. In specific implementation, the underlying driver extracts the set pull-up resistor value of the serial communication bus under the current operating conditions and obtains the equivalent parasitic capacitance bias of the physical medium by combining the changes in the external ambient temperature. Based on the above RC network parameters, the microcontroller module calculates the actual rise time parameter required for the digital level signal on the communication link to transition from a low threshold to a high threshold through a general polynomial mapping or a built-in lookup table.

[0049] It should be further noted that when the microcontroller module uses the aforementioned mathematical model or data table for low-level communication risk assessment, a strict arithmetic boundary check mechanism must be introduced because the computational logic typically includes a normalization process for the reference parameters. If the denominator value (such as a variable representing the reference temperature or reference time) approaches 0 due to sensor disconnection or memory overflow, the low-level exception handling mechanism will directly intercept the calculation and force the denominator to be assigned a preset non-zero minimal positive number (such as a machine precision constant) to avoid kernel hardware failure caused by division by zero. Based on the rise time parameter calculated by this robust model, the microcontroller module can dynamically calculate the required clock stretching duration compensation value and maintain signal integrity by rewriting the setting value of the clock stretching setting register in the event of deterioration of physical electrical characteristics through logic timing broadening.

[0050] At the software architecture level, an embedded operating system runs within the flash memory storage of the microcontroller module. The underlying implementation of this embedded operating system can be a real-time kernel supporting preemptive multitasking scheduling, such as FreeRTOS or RT-Thread. The internal logical stack of the embedded operating system includes a file system layer, which employs a journaling file system supporting wear leveling. During system initialization, the file system layer reads the physical block size and page count information of the memory module and constructs and dynamically maintains a mapping table from logical addresses to physical sectors within the microcontroller module's random access memory. This mapping table allows upper-layer application programs to avoid directly interacting with the specific physical coordinates of the underlying storage chip, thus providing the necessary system abstraction interface support for subsequent implementation of physical sector thermal isolation status marking and write data redirection.

[0051] For the task context switching interrupt triggering logic, mutex lock resource allocation mechanism, and direct memory access channel settings of peripheral controllers in embedded operating systems, those skilled in the art can implement them directly with reference to the factory development manual of the specific microcontroller chip. The underlying architecture scheduling mechanism of the system is a well-known technology in this field and will not be described in detail here.

[0052] Please see the appendix Figure 6In this embodiment, the memory module serves as a persistent storage medium for recording delivery status data such as temperature, humidity, and geographical location of the insulated bags. To meet the stringent power consumption limitations and physical shock resistance requirements of edge computing devices, the memory module employs a non-volatile solid-state flash memory chip. This memory module uses a NOR flash memory chip that supports the serial peripheral interface protocol. The flash memory chip adopts a hierarchical layout mechanism in its physical space. Its internal storage array consists of multiple independent physical blocks, each physical block is further subdivided into several physical sectors, and each physical sector consists of consecutively arranged storage pages. This hierarchical topology based on physical blocks and physical sectors defines the underlying hardware execution rules for erasing by sector and programming by storage page, while simultaneously providing the necessary physical media support boundaries for the system file system layer to implement fine-grained thermal isolation status marking and backup address redirection.

[0053] Based on the aforementioned physical storage architecture, a data interaction link is established between the memory module and the microcontroller module via a serial communication bus. In the specific pin electrical interconnection topology, the serial communication bus includes a chip select signal line, a serial clock signal line, a master output / slave input data line driven independently by the microcontroller module, and a master input / slave output data line driven by the memory module. This four-wire physical connection constitutes the hardware channel that meets the requirements of high-speed synchronous digital transmission. During data transmission, the microcontroller module extracts the clock stretching duration compensation value obtained from the previous algorithm derivation and dynamically adjusts the high and low level maintenance period of the serial clock signal line. By introducing this controlled clock level down-conversion mechanism, even if the physical parameters of the printed circuit board dielectric constant and trace parasitic capacitance drift due to temperature changes, the digital levels on the master and slave device pins can still meet the data setup time and data hold time margin requirements of the latch.

[0054] Considering the fundamental physical characteristics of semiconductors, the physical charge injection and tunneling penetration rate of the floating-gate transistors inside flash memory chips are highly sensitive to ambient absolute temperature. When a large amount of cold air enters from the insulation bag, causing the chip's ambient temperature to drop, the time required for actual memory page programming at the underlying level of the memory module will increase non-linearly. To prevent the microcontroller module's read / write drive thread from getting stuck indefinitely while polling and waiting for the memory chip to complete writing, the system's underlying driver layer constructs a dynamic programming timeout threshold evaluation model based on thermodynamic states. Based on the general principles of the Arrhenius empirical model in semiconductor physics, the physical time required for electrons to tunnel through the insulating oxide layer exhibits a non-linear, negatively correlated relationship with ambient absolute temperature. In specific implementation, the system incorporates preset exponential or polynomial approximation calculation logic, combined with the current absolute temperature value of the physical coordinates of the memory module fed back by the simulation calculation module at the current moment, to calculate a dynamic waiting timeout threshold that matches the current hardware temperature state. The system uses this dynamic waiting timeout threshold to replace the fixed software delay parameters in traditional driver code, thereby giving the underlying driver more flexible timing adaptability.

[0055] When the underlying logic performs the aforementioned timeout threshold evaluation, the system needs to guard against the risk of data anomalies introduced by physical sensor disconnection or communication interference. If the current absolute temperature value extracted by the system is equal to or lower than the set physically unreasonable extreme value (e.g., a default reading of 0 Kelvin due to sensor disconnection), this abnormally low value, as the denominator or polynomial input parameter of the temperature evaluation factor, will induce arithmetic overflow in the system. To address this computational boundary risk, the underlying driver engine sets up a threshold interception detector. When it detects that the acquired temperature data is lower than the minimum extreme value allowed by the hardware operating specifications, the system forcibly blocks the current normal operation, and uses a system-preset safe positive constant as the substitute input parameter for the current absolute temperature value, thereby avoiding the risk of kernel crash at the execution boundary of the underlying code.

[0056] During the actual persistent recording of operational data, the data interaction between the microcontroller module and the memory module follows a strict underlying state machine polling confirmation rule. The system will invoke storage read / write phase instructions in the cross-module collaborative workflow, and its core write response mechanism is specifically manifested as follows: In the data retrieval pre-stage, after receiving a data write-to-disk request from the upper-layer business logic, the microcontroller module queries the internal file system layer for the latest logical address-to-physical sector mapping table. During this address retrieval process, the file system layer directly bypasses coordinate intervals marked as thermally isolated based on dynamically refreshed status labels, safely extracting the thermally stable backup physical sector's starting base address as the target address.

[0057] Then, the physical bus communication phase begins. The microcontroller module outputs a low-level active signal to the chip select signal line to wake up the hardware communication port of the memory module. Then, through the serial communication bus, it sequentially pushes the write enable operation instruction code, the page programming control instruction code containing the aforementioned target address, and the actual operating condition data payload bit stream to be written to the memory module according to the standard operating specifications.

[0058] After the microcontroller module removes the low-level state of the chip select signal line, the microcontroller module's communication driver layer continuously reads the dedicated busy flag bit status of the memory module's internal status register via register addressing. If the busy flag bit flips from high to low before the dynamic wait timeout threshold calculated by the evaluation model is exhausted, the system determines that the data burning operation of this physical sector has successfully closed the loop. Conversely, if the busy flag bit does not flip before the dynamic wait timeout threshold is exhausted, and the number of consecutive read / write retries initiated by the underlying communication driver module exceeds the set redundancy error threshold limit, the system, considering the above multi-dimensional judgment conditions, confirms that the currently selected backup physical sector has experienced physical communication blockage due to local transient thermal stress. In response to this joint judgment result, the system will immediately release the underlying hardware control authority to forcibly terminate the current write thread and report the fault event to the system task queue, thereby triggering a new round of physical sector redirection mapping scheduling tasks.

[0059] The microstructure of the charge pump high-voltage bias circuit used to implement erase and write operations inside the memory module, as well as the hardware logic implementation of the error correction code verification generator polynomial used to ensure the static maintenance life of data, can be directly implemented by those skilled in the art in conjunction with conventional integrated circuit memory design specifications. The internal semiconductor physical circuit and basic verification operation rules are well-known technologies in this field and will not be described in detail here.

[0060] In this embodiment, the temperature sensor module serves as the basic hardware interface for the system to perceive the microscopic thermophysical field. It is responsible for collecting the absolute temperature data of each key node on the printed circuit board in real time, thereby providing basic data support for subsequent thermal field modeling and dynamic thermal impedance calculation.

[0061] To meet the system's requirement for capturing minute temperature differences in space, the temperature sensor module employs a high-precision integrated digital temperature sensor chip with high resolution and a built-in analog-to-digital converter. This chip supports directly outputting the digital code corresponding to Celsius or Kelvin temperature via its internal integrated circuit bus. Its hardware measurement resolution is typically limited to between 0.01 Kelvin and 0.1 Kelvin. The physical reason for choosing this resolution range is that if the resolution exceeds the upper limit of this range, the system will be unable to effectively identify the transient temperature rise step response caused by subsequent weak thermal pulse injection; conversely, blindly pursuing ultra-high resolution below the lower limit of this range will introduce a large amount of low-frequency environmental thermal noise generated by external airflow disturbances into the sensor probe, resulting in the effective signal being overwhelmed and significantly increasing hardware selection costs.

[0062] Based on the aforementioned components, the coordinate layout of the temperature sensor module on the printed circuit board follows strict principles of thermodynamic isolation and focused monitoring. This module includes a localized temperature sensor located on the low-temperature side of the printed circuit board. This localized temperature sensor is physically positioned close to the inner wall of the insulation bag body 1, and its linear physical distance from core heat sources such as the microcontroller module must be greater than a set safety isolation threshold. It should be noted that this safety isolation threshold is not a fixed constant, but rather is calculated based on the nominal peak heat dissipation of the system's main control chip and the in-plane equivalent thermal conductivity of the selected printed circuit board substrate, combined with the basic steady-state heat conduction equation.

[0063] In practical engineering calibration, it is necessary to ensure that the static temperature rise caused by the power consumption of the microcontroller module being conducted to this coordinate point is strictly less than the physical measurement resolution of the sensor chip itself. This layout rule based on the thermal conductivity attenuation characteristics aims to enable the sensor to objectively respond to the boundary cold penetration caused by the cold source medium inside the insulation bag, without being affected by the heat crosstalk caused by the power consumption of the device itself.

[0064] Correspondingly, the temperature sensor module also includes a reference temperature sensor located around the serial communication bus and the memory module. Since the parasitic capacitance of the physical traces of the serial communication bus and the internal tunneling effect of the flash memory chip both exhibit physical correlations with the local ambient temperature, the temperature probe of the reference temperature sensor is positioned within a safe distance from the bus trace network and the memory chip package pins. The digital temperature stream output by the reference temperature sensor can accurately characterize the current physical thermal boundary state of the critical communication link and the persistent storage medium.

[0065] In the underlying data acquisition process, the microcontroller module assigns a fixed discrete sampling frequency to each temperature sensor node. This sampling frequency is set based on the thermal time constant of the printed circuit board substrate, and its value is typically more than twice the substrate's highest thermal response frequency. Following the general principles of the Nyquist sampling theorem, this oversampling mechanism prevents spectral aliasing distortion of the time-domain thermophysical signal during mapping to the digital conversion domain.

[0066] Considering that mechanical vibrations or transient electromagnetic pulse interference during the cold chain transportation of fresh produce may cause outliers in the sensor hardware output, the system introduces a data cleaning mechanism based on a multi-dimensional time window after acquiring the raw temperature data. This prevents irreversible driving errors caused by the system control logic relying solely on a single extreme value. Based on general principles in conventional digital signal processing, the system typically uses software filtering algorithms to remove and smooth the continuously acquired discrete temperature data stream. In practice, the system pre-allocates a sliding memory area to extract the temperature values ​​from the current moment and several past sampling periods. After removing physically abnormal extreme values ​​from this set, the system calculates the arithmetic mean or weighted average of the remaining valid data, using this as the true temperature output after filtering. Through this joint judgment logic based on the historical time sequence, the system effectively suppresses the interference of random high-frequency physical noise on subsequent calculations.

[0067] During the mean averaging calculations of the various filtering algorithms executed at the lower level, the system must guard against the risk of arithmetic operation overflows due to abnormal parameters. If the sliding time window length parameter in the algorithm logic becomes extremely small due to human error or abnormal modification of the memory address, it will trigger a division-by-zero interrupt or arithmetic overflow in the lower-level arithmetic logic unit when used as the denominator in a division operation. In response to this potential computational boundary risk, the lower-level driver injects a forced interception and verification instruction before calling the data cleaning module. When it is detected that the denominator term does not meet the safety setting lower limit, the system will automatically bypass the filtering function and force the latest acquired raw absolute temperature data within the hardware specification range to be returned as a substitute theoretical value to the upper-level calling interface. This defensive coding mechanism ensures the robustness of the perception data stream transmission under abnormal operating conditions from the underlying code execution level.

[0068] The bandgap reference voltage source circuit topology contained within the temperature sensor chip, and the microscopic working principle of the incremental accumulator modulator used to perform the conversion of tiny analog level signals to digital sequences, can be directly implemented by those skilled in the art by referring to integrated circuit design manuals. Its internal semiconductor circuit structure and basic analog voltage conversion mechanism are well-known technologies in the field and will not be described in detail here.

[0069] In this embodiment, the heat injection circuit module serves as the pre-excitation execution unit for the system to obtain the dynamic thermal characteristics of the underlying physical medium. It is responsible for generating controlled transient DC pulses within the target physical region, thereby providing a measurable step heat source signal for the subsequent simulation calculation module to calculate the physical heat transfer characteristics.

[0070] The core dissipation component of the heat injection circuit module is the copper trace module. In terms of physical implementation, the copper trace module is embedded within the inner layer of the printed circuit board. The physical purpose of using inner-layer wiring is to ensure that the injected Joule heat is conducted isotropically to the upper and lower surfaces of the printed circuit board, preventing the heat-generating traces from being directly exposed to the surface and rapidly dissipated by the convective cold air inside the insulation bag. This copper trace module uses a serpentine geometric pattern for continuous wiring. Through this folded spatial winding physical design, the system can increase the absolute length of the copper trace within a very small planar physical area, thereby stabilizing its equivalent room-temperature impedance within a set range of 1 ohm to 10 ohms. This impedance range is calculated based on the system's mainstream DC power supply voltage, aiming to ensure that hundreds of milliwatts of transient heating power can be generated at the moment the circuit is closed. This set power level constitutes the lower energy limit for the temperature sensor module to capture temperature rise step signals and overcome ambient thermal noise.

[0071] The switching device module is responsible for connecting and disconnecting the aforementioned energy conversion circuits. Based on the overall power topology of the system, the first end of the copper trace module is hardwired to the system's ground network via vias, while the second end of the copper trace module needs to be connected to the system's high-level side. Based on this high-side drive topology, the lower-level characteristic implementation of the switching device module is typically a P-channel metal-oxide-semiconductor field-effect transistor with low on-resistance. The drain of this switching device module is connected to the second end of the copper trace module, its source is connected to the system power rail, and its control terminal is electrically connected to a dedicated pin of the microcontroller module configured for push-pull output mode.

[0072] To execute a detection process for acquiring the thermal impedance of a physical medium, the heat injection circuit module performs a thermal pulse excitation task based on timing instructions issued by the microcontroller module. Considering that the resistivity of copper drifts significantly with decreasing ambient absolute temperature, the system does not use an open-loop preset constant method to estimate the heating power, but instead introduces a closed-loop actual measurement calibration mechanism. The system calls the heat injection execution instructions in the cross-module collaborative workflow, and its underlying electro-thermal interaction sequence is specifically manifested as follows: During the pulse triggering phase, the timer inside the microcontroller module triggers a hardware interrupt, and the corresponding pin then outputs a transient toggling drive level. This drive level causes the bias voltage of the switching device module to exceed the semiconductor turn-on threshold, causing the switching device module to momentarily enter the saturation conduction region, thereby opening the energy release channel.

[0073] As the energy release channel is activated, the DC current from the system power rail flows into the copper trace module through the already activated switching device module. During the physical process of electrons passing through the conductor lattice, electrical energy is converted into heat energy and diffuses into the surrounding printed circuit board medium. Based on the general principles of the intersection of electromagnetism and thermodynamics, this physical process injects the effective heat generation power corresponding to the transient DC pulse into the surrounding medium. Instead of traditional resistance-based calculations, it uses Ohm's law of energy to calculate the effective heat generation power by multiplying the dynamic terminal voltage and real-time current. In specific engineering implementations, the microcontroller module synchronously acquires the actual DC bus voltage parameters output by the system power rail at the moment of activation, as well as the total loop feedback current parameters flowing through the switching device module and the copper trace module, via its internally integrated analog-to-digital converter hardware. The system's underlying logic performs real-time multiplication of the synchronously acquired voltage and current parameters, thereby dynamically calculating the actual transient effective physical heat generation power released by the copper trace module into the surrounding medium during this command activation cycle.

[0074] During the execution of the aforementioned closed-loop excitation actions, the system incorporates a strict physical boundary anti-crossing interception and multi-dimensional joint detection mechanism. Considering that strong condensation during fresh food cold chain transportation may cause localized micro-short circuits in the inner layers of the printed circuit board, its actual equivalent impedance may experience unexpected drops. To prevent large current injection from burning out the system, the microcontroller module issues a probe pulse command with a duration on the order of microseconds before releasing the full-width master control pulse. During this period, if the underlying control logic, based on the current parameters fed back from the analog-to-digital converter, determines that the current feedback current parameter of the current loop exceeds the maximum transient rated value allowed by the system power management chip, or if the sum of the impedance evaluation parameters participating in the closed-loop calculation does not meet the set safe physical lower limit and induces a division-by-zero overflow risk in the underlying processor's arithmetic logic unit, the system will forcibly bypass the subsequent full-width hot-injection action. The microcontroller module will continue to maintain a shutdown level to cut off the power rail release loop, thereby safely eliminating hardware-level short circuits and code-level crash risks at the system control boundary.

[0075] The mathematical equations for matching the etching impedance and copper layer thickness of the inner layer serpentine traces on the printed circuit board, as well as the hardware design for bypassing the gate pull-up resistor to prevent the microcontroller chip from generating glitches and causing mis-conduction in the switching device module during power-on reset, can be directly implemented by those skilled in the art in conjunction with conventional hardware circuit anti-interference reliability design specifications. The basic etching manufacturing process and semiconductor switch anti-jitter design are well-known technologies in the field and will not be elaborated here.

[0076] In this embodiment, the simulation calculation module, as the system's data processing and control variable calculation unit, is responsible for mapping the discrete physical quantities acquired by the underlying hardware to the system's state space parameters, and outputting hardware-software collaborative recovery instructions based on the calculation results. At the physical deployment level, the simulation calculation module is specifically implemented as a high-priority software task thread running inside the microcontroller module, or, if the communication bandwidth allows, deployed as an edge processing service in an external communication gateway device.

[0077] Based on the overall system control closed-loop architecture, the internal algorithm logic of the simulation calculation module unfolds according to the following process: After acquiring the thermodynamic boundary data of the inner and outer sides of the insulation bag body 1, the simulation calculation module initiates spatial-dimensional thermal field reconstruction. Based on the general principle of fitting a continuous physical field to multi-source discrete data points, the system uses a software program to call a numerical interpolation algorithm to establish a continuous thermal field model of the two-dimensional plane of the printed circuit board. This algorithm employs either the inverse distance weighted interpolation algorithm or the bilinear interpolation algorithm, which have lower computational overhead. After completing the continuous model construction, the system obtains the temperature gradient vector at specified coordinates of the printed circuit board by solving for spatial partial derivatives, and calculates the rate of temperature change with time by performing backward difference operations on the continuously sampled data.

[0078] Due to general engineering limitations in numerical computation, the underlying system logic must guard against arithmetic overflow risks induced by the mesh discretization process when performing the aforementioned spatial partial derivative and temporal difference operations. If the spatial Euclidean distance parameter between two adjacent temperature sampling nodes input to the algorithm approaches 0 due to incorrect coordinate settings, or if the time step of two sampling calculations is severely reduced due to internal clock jitter, this distance or time parameter, as the denominator in the differential calculation, will directly trigger a division-by-zero overflow interrupt in the floating-point arithmetic unit. In response to this potential boundary collapse risk, the simulation module incorporates defensive boundary overflow prevention logic. When the denominator parameter is detected to be below the set tolerance lower limit, the system forcibly replaces it with the minimum physical line tolerance constant allowed by the printed circuit board manufacturing process or the basic resolution equivalent of the system hardware timer. This replacement mechanism eliminates kernel-level crash risks while ensuring the continuity of mathematical calculations.

[0079] The simulation calculation module utilizes a differential function mechanism to solve for the dynamic thermal impedance of the physical medium. After the underlying heat injection circuit module generates a DC pulse, the system acquires synchronized temperature rise data from the local temperature sensor and the reference temperature sensor. Considering the actual cold chain transportation conditions for fresh produce, the complex and variable convective cold airflow inside the insulated bag will generate random thermal disturbances on the printed circuit board surface. Based on the general physical principle of differential signal processing to suppress common-mode interference, the simulation calculation module calculates the relative difference between the local sensor temperature rise and the reference sensor temperature rise relative to the initial injection moment through algebraic subtraction. This differential operation utilizes the correlation characteristics of background noise from adjacent spatial physical nodes, effectively canceling the low-frequency common-mode thermal noise introduced by environmental airflow disturbances, thereby extracting the undisturbed effective temperature rise step peak. Subsequently, based on the fundamental response law of heat transfer, the system divides this effective temperature rise step peak by the actual heating power parameter of the heat injection circuit module and the pulse duration to calculate the current dynamic thermal impedance of the physical medium on the printed circuit board. This impedance parameter directly maps the degree of attenuation of thermal conductivity caused by local condensation or thermal expansion and contraction within the physical medium.

[0080] After obtaining the dynamic thermal impedance, the simulation module transforms the macroscopic physical field characteristics into microscopic electronic control variables, calculating the clock stretching duration compensation value of the serial communication bus. In microelectronics engineering practice, the parasitic capacitance of circuit board traces and the resonant frequency of chip crystal oscillators are highly susceptible to nonlinear interference from non-uniform temperature fields and transient thermal shocks. Based on the physical correlation mechanism of this thermoelectric coupling phenomenon, the simulation module establishes the following core control mathematical mapping formula: ; In the formula, This indicates the clock stretch duration compensation value; The static thermal drift empirical coefficient represents the weight of the effect of space temperature bias on bus parasitic capacitance. This represents the temperature gradient vector obtained through thermal field modeling. This represents the physical direction unit vector from the microcontroller module to the target memory module on the printed circuit board. The dot product operation between the two represents the actual spatial projection component of the temperature gradient vector along a specific communication path. This represents the dynamic thermal impedance calculated using differential methods. This represents the dynamic thermal drift empirical coefficient that characterizes the weight of the clock source frequency disturbance caused by the temperature transient shock. This represents the rate of change of temperature with respect to time.

[0081] The criteria for determining the aforementioned empirical coefficients typically need to be preset by a hardware calibration engineer based on the frequency sweep test results of the constant temperature and humidity test chamber. In this embodiment, The value is typically mapped to a fine-tuning range of 1% to 10% of the nominal operating clock cycle of the digital signal.

[0082] To mitigate the signal integrity degradation trend at the system's underlying layer through multi-dimensional quantification, the simulation calculation module further extracts the calculated clock stretching duration compensation value as a feedforward input, and introduces transient response and aging fatigue accumulation mechanisms to calculate the local electrical distortion index. The basic calculation formula is expressed as follows: ; In the formula, Indicates the local electrical distortion index; This represents the transient distortion weighting coefficient for the current communication cycle; This indicates the clock stretch duration compensation value; This represents the cumulative thermal stress weighting coefficient. The upper and lower limits of the integral term in the formula correspond to the system's self-startup time, respectively. relative to current absolute time The overall value represents the cumulative clock stretching duration compensation value of the system under continuous harsh thermal conditions. This local electrical distortion index avoids the one-sided evaluation caused by relying solely on the current temperature extreme value. By integrating transient physical distortion and historical thermal fatigue variables, it achieves a smooth prediction of the underlying communication risks of semiconductors.

[0083] In practical microprocessor digital engineering implementations, the system cannot perform ideal continuous-time integration. Instead, it approximates the integral term by using the sum of discrete time steps. To prevent numerical divergence or integer overflow of this sum during long-term cold chain transportation, the system incorporates anti-saturation limiting logic at the software level. When the sum is detected to exceed the maximum limit allowed by the physical data type, or when the temperature difference between the inner and outer boundaries of the insulation bag approaches zero for several consecutive cycles (i.e., thermal equilibrium), the system actively freezes the summation step or resets and releases the integral term according to the exponential decay coefficient.

[0084] After the above parameter calculations are completed, the microcontroller module obtains the local electrical distortion index output by the simulation calculation module and compares it with the pre-calibrated safety threshold for judgment. The safety threshold is determined mainly based on the bus eye diagram opening test data extracted using a high-speed oscilloscope during the early hardware development phase and the signal-to-noise ratio margin baseline. In this judgment process, the system adopts a multi-condition joint triggering strategy. If the local electrical distortion index exceeds the set safety threshold, and the bus read / write failure count value fed back by the microcontroller module's underlying state machine reaches the set redundancy error tolerance limit, the system confirms that the current communication link and the target physical sector face actual physical connectivity risks.

[0085] In response to this joint determination, the underlying driver layer suspends the service data flow on the serial communication bus, instead maintaining only low-frequency network heartbeat commands on the serial communication bus. Simultaneously, the system's internal file system layer marks the physical sector of the memory module corresponding to the coordinate with the highest dynamic thermal impedance value as thermally isolated, terminating subsequent data overwrite and erase operations, and collaboratively initiates an address redirection mechanism to map newly written data from the source address to a spare physical sector unaffected by the heatwave. This hardware-software collaborative blocking strategy proactively avoids the permanent loss of core service data during the early stages of physical link deterioration.

[0086] Specific application examples: To aid in understanding the highly abstract hardware and software collaborative repair technology described above, this section uses a cryogenic fresh food transportation task in the height of summer as an example, combined with the appendix... Figure 7 The system state evolution law and fault interception mechanism of the present invention in actual working conditions are described in detail.

[0087] Application scenarios and initial state settings: The current task is to deliver a batch of high-value, cryogenically frozen fresh seafood (such as ultra-low temperature bluefin tuna) from a cold storage facility to an end customer during the height of summer (when the ambient air temperature is approximately 35°C). The insulated bag module contains a large amount of solid carbon dioxide (dry ice, sublimation temperature approximately -78.5°C) as the cooling medium. An edge computing module is attached to the outer surface of the insulated bag, responsible for recording temperature and humidity throughout the process and maintaining a communication link with the cloud.

[0088] Anomaly Triggering and Physical Field Evolution: When the delivery vehicle travels over bumpy roads, the dry ice inside the insulation bag undergoes physical displacement and comes into direct contact with the inner wall of the insulation bag, causing a rapid outward permeation of localized cold energy. Based on fundamental heat conduction mechanisms, the heat flux density along the wall thickness of the insulation bag module is directly proportional to the equivalent thermal conductivity of the insulation layer and the actual temperature gradient along that physical direction. Based on this physical law, the simulation module, through a temperature sensor matrix calculation, discovered the temperature gradient vector on the printed circuit board surface. The temperature rises sharply. Simultaneously, the microcontroller module schedules the heat injection circuit to release a probe pulse, and calculates the dynamic thermal impedance of the region using a differential function. The abnormal changes objectively indicate that the printed circuit board substrate has undergone microscopic physical deformation or internal condensation due to rapid cooling.

[0089] Feedforward evaluation and electrical distortion index evolution: After acquiring the aforementioned rapidly changing physical quantities, the simulation module transforms them into microelectronic control variables and calculates the clock stretching duration compensation value required for the serial communication bus based on the following core control mathematical mapping formula. : ; As the cold stress impact continues, in order to prevent the signal integrity from deteriorating, the simulation calculation module further extracts the calculated clock stretching duration compensation value as the feedforward input, and introduces transient response and aging fatigue accumulation mechanisms to calculate the local electrical distortion index. The basic calculation formula is expressed as follows: ; See appendix Figure 7 Subgraph (a) shows the evolution of the electrical distortion index and the threshold decision diagram. This diagram visually represents the quantitative assessment process of the degradation of the electrical characteristics of the underlying physical channel. The horizontal axis in the diagram represents the absolute physical time (delivery operation time / minute) experienced since the fresh food insulated bag was packaged and the delivery task was started. The vertical axis corresponds to the local electrical distortion index calculated by the aforementioned formula. This value integrates the transient value of the current clock compensation with the historical accumulated fatigue. The horizontal gray dashed line in the figure represents the safety protection limit (system preset safety threshold) pre-calibrated based on the hardware oscilloscope eye diagram opening and signal-to-noise ratio margin.

[0090] As the cold air from the dry ice inside the insulation bag continues to penetrate, the black solid line representing the distortion index in the figure exhibits a non-linear upward evolution trajectory. When the distortion index is below the horizontal gray dashed line, the system can maintain communication with only routine timing fine-tuning; however, when running for approximately 44 minutes, the black solid line and the horizontal gray dashed line spatially intersect, forming the gray circular marker in the figure (cooperative blocking trigger point). This trigger point is the critical decision moment for the system control logic to make a state transition, based on which the system formally determines that the underlying channel faces a substantial physical breakdown risk.

[0091] Comparison of collaborative blocking, logical remapping, and effects: Upon reaching the trigger point represented by the aforementioned gray circle, even though the memory chip has not yet experienced an actual read / write crash, the system has already triggered a hardware-software collaborative defense mechanism. The underlying driver immediately takes over the serial communication bus, lengthens its clock cycle to adapt to the harsh electrical characteristics, and truncates large data packets, maintaining only a minimal number of network heartbeat commands. At the same time, the file system layer urgently marks the flash memory physical sector (sector A) that is close to the cold source and has the highest dynamic thermal resistance as thermally isolated, and redirects subsequent fresh food temperature tracking data to a spare sector (sector B) on the printed circuit board that is farther from the cold source and has a stable state.

[0092] For the actual preservation effect of the above defense mechanisms, please refer to the appendix. Figure 7Subgraph (b) in the figure compares the cumulative system bus errors under extreme temperature gradients. The horizontal axis of this subgraph is strictly synchronized with the time axis of subgraph (a), and the vertical axis represents the frequency of accumulated errors of the underlying communication bus (i.e., the total number of hardware-level failures caused by insufficient data setup time or memory page programming timeout). The vertical gray dotted line running through the figure serves as a time-aided alignment line, projecting the cooperative blocking trigger point in subgraph (a) onto subgraph (b), defining the time watershed between the system's deterioration from a safe operating condition to a catastrophic operating condition.

[0093] After crossing the vertical gray dotted line, the thick black dashed line representing conventional cold chain equipment (traditional solution: fixed baud rate / static addressing) exhibits an exponentially diverging upward trend. This is because the traditional solution does not introduce dynamic timing compensation and sector risk avoidance mechanisms. Thermal drift of hardware electrical parameters leads to bus timing disorder, a sharp increase in error frequency, and ultimately file system corruption and permanent loss of core business data. In contrast, the dark gray solid line representing the system of this invention (the solution of this invention: timing compensation / logical remapping), after crossing the trigger line, maintains a near-zero horizontal baseline because the data flow is forwarded to a safe backup physical sector, and the retained low-frequency heartbeat command matches the stretched safety clock. This demonstrates that the solution of this invention successfully achieves zero-error hardware-level fault interception and data preservation under severe thermal shock.

[0094] Closed-loop recovery: Once the fresh produce reaches the end customer, the insulated bag is opened and the dry ice inside is removed. The temperature difference between the inside and outside gradually neutralizes, and the temperature gradient vector approaches zero. The microcontroller module's underlying driver then removes the bus clock stretching configuration, and the file system layer removes the thermal isolation marker from sector A. Finally, the system uploads the complete, lossless historical delivery data to the cloud, completing the closed loop.

[0095] Experimental verification design and qualitative conclusions: To further verify the rigor of the above curve evolution law, a high-low temperature alternating chamber comparative experiment was introduced during the research and development phase. A traditional cold chain edge computing device (control group) and the device of this invention (experimental group) were simultaneously attached to the surface of a simulated insulated bag. The alternating chamber simulated a constant ambient temperature of 30°C, and liquid nitrogen was injected into the insulated bag to simulate an extreme cold source, causing the device's printed circuit board to withstand a spatial temperature difference impact exceeding 80°C for a short period.

[0096] The experimental phenomena qualitatively verified the logic of the aforementioned charts: Communication link status: Due to the use of a fixed baud rate, the rising edge of the bus signal was severely distorted in the control group, resulting in continuous read failures and frequent watchdog hardware resets, and the communication link was completely broken (corresponding to the divergence of the thick black dashed line in sub-figure b); the experimental group actively extended the bus level hold time before the electrical distortion index exceeded the threshold, and the oscilloscope test showed that its signal eye diagram opening was always maintained within the safety margin, avoiding bus deadlock.

[0097] Persistent storage integrity: In the control group, data was forcibly written to a fixed physical sector under extremely cold conditions. The electron tunneling effect was blocked, resulting in write timeouts and damage to the sector allocation table, leading to a large number of ECC check errors. In the experimental group, before the cold energy penetrated to the flash memory chip, the data flow was redirected to a safe sector (corresponding to the stable dark gray solid line in sub-figure b) by relying on the feedforward mechanism. After the experiment, the exported data verification showed that the file system was intact and undamaged, which fully confirmed the high reliability of the present invention under extreme physical field conditions.

Claims

1. A digital modeling and simulation system for gradient insulated bags for fresh food delivery, comprising an insulated bag body (1), an edge computing module, and a simulation computing module, wherein the edge computing module includes a microcontroller module, a memory module divided into physical sectors and connected to the microcontroller module via a serial communication bus, a temperature sensor module, and a heat injection circuit module, characterized in that, The simulation calculation module collects boundary temperature data through the temperature sensor module and calculates the temperature gradient vector and the rate of change of temperature with respect to time. The microcontroller module controls the heat injection circuit module to generate DC pulses, and the simulation calculation module collects local temperature rise data to calculate dynamic thermal impedance. The simulation calculation module calculates the clock stretch duration compensation value by combining the temperature gradient vector and the dynamic thermal impedance, and calculates the local electrical distortion index by combining the clock stretch duration compensation value. When the local electrical distortion index exceeds a pre-defined safety threshold, the microcontroller module truncates the service data stream of the serial communication bus and maintains the network heartbeat command, marks the physical sector as thermally isolated, and redirects newly written data to a spare physical sector.

2. The system according to claim 1, characterized in that, The temperature sensor module includes a local temperature sensor disposed on the low-temperature side of the printed circuit board, and a reference temperature sensor disposed around the serial communication bus and the memory module. The simulation calculation module uses a numerical interpolation algorithm to establish a continuous thermal field model of the two-dimensional plane of the printed circuit board, solves the spatial partial derivatives to obtain the temperature gradient vector at a specified coordinate of the printed circuit board, and performs backward difference operation on the continuous sampled data to obtain the rate of change of temperature with respect to time.

3. The system according to claim 1, characterized in that, The heat injection circuit module includes a copper-clad trace module and a switching device module disposed in the inner layer area of ​​the printed circuit board. The first end of the copper trace module is electrically connected to the ground network, and the second end of the copper trace module is connected to the system power rail through the switching device module. The control electrode of the switching device module is connected to the pin of the microcontroller module set to push-pull output mode.

4. The system according to claim 1, characterized in that, The simulation calculation module uses algebraic subtraction to calculate the relative difference between the temperature rise of the local temperature sensor and the temperature rise of the reference temperature sensor relative to the initial injection moment, extracts the effective temperature rise step peak value, and combines the effective temperature rise step peak value, the actual heating power parameters of the heat injection circuit module and the pulse duration to calculate the dynamic thermal impedance of the physical medium of the printed circuit board.

5. The system according to claim 1, characterized in that, The simulation calculation module obtains the physical direction unit vector of the microcontroller module to the target memory module on the printed circuit board, and performs a dot product operation between the temperature gradient vector and the physical direction unit vector to represent the actual spatial projection component in the communication direction. The clock stretch duration compensation value is calculated by combining the static thermal drift empirical coefficient characterizing the influence of spatial temperature bias on the parasitic capacitance of the serial communication bus, the dynamic thermal impedance, the dynamic thermal drift empirical coefficient characterizing the influence of temperature transient shock on the clock source frequency disturbance, and the rate of change of temperature with time.

6. The system according to claim 1, characterized in that, The local electrical distortion index integrates the transient distortion weight coefficient of the current communication cycle with the clock stretch duration compensation value, the cumulative thermal stress weight coefficient and the cumulative amount of the clock stretch duration compensation value, and uses the summation of discrete time steps to approximate the integral term. When the accumulated sum is detected to exceed the maximum limit allowed by the physical data type, the simulation calculation module actively freezes the accumulation step size.

7. The system according to claim 1, characterized in that, The microcontroller module obtains the absolute temperature value of the physical coordinates of the memory module at the current moment, and calculates the dynamic waiting timeout threshold that matches the hardware temperature state. The microcontroller module sequentially pushes write enable operation instruction code, page programming control instruction code containing the target address and actual operating condition data payload bit stream to the memory module through the serial communication bus, and reads the busy flag status of the internal status register of the memory module. When the busy flag does not flip before the dynamic wait timeout threshold is exhausted, and the number of consecutive read / write retries initiated by the underlying communication driver module exceeds the set redundancy error tolerance limit, the microcontroller module releases the underlying hardware control authority.

8. The system according to claim 1, characterized in that, The microcontroller module runs an embedded operating system that includes a file system layer. The microcontroller module queries the file system layer for the latest logical address to physical sector mapping table; The file system layer bypasses the coordinate range that has been marked as the thermally isolated state based on the dynamically refreshed status label, and extracts the starting base address of the thermally stable spare physical sector as the target address.

9. The system according to claim 1, characterized in that, The microcontroller module runs an embedded operating system that includes a file system layer. When the magnitude of the temperature gradient vector and the dynamic thermal impedance return to the set normal value range, the underlying driver layer of the microcontroller module cancels the clock stretch duration compensation value, and the file system layer releases the thermal isolation state of the corresponding physical sector.

10. An insulated bag for fresh produce delivery, characterized in that: The gradient insulated bag digital modeling and simulation system for fresh food delivery as described in any one of claims 1-9 includes an inner wall structure, an outer wall structure, and a heat-insulating solid layer (2) disposed between the inner wall structure and the outer wall structure, wherein the inner wall structure and the outer wall structure are spatially closed and define an internal cavity for accommodating goods.

Images