Aquatic product cold chain transportation monitoring system

Abstract

The invention discloses an aquatic product cold chain transportation monitoring system which is characterized in that the system comprises a hardware layer and a software layer, the hardware layer comprises a sensing control module, a structure bearing module and a communication interaction module, and the software layer comprises a data management module, an intelligent calculation module and a business application module; the sensing control module comprises an environment sensing sub-module and an equipment control sub-module, the structure bearing module comprises a compartment main body sub-module and a cargo protection sub-module, and the communication interaction module comprises an edge calculation sub-module and a vehicle-mounted interaction sub-module. According to the application, a real-time monitoring-dynamic prediction-intelligent regulation and control microbial inhibition system is constructed, a high-precision sensor is used for collecting environmental data in real time, and an LSTM-Attention model is combined for accurately pre-judging the growth trend of microorganisms, so that the effects of early warning before the colony count reaches a dangerous threshold value and automatically coordinating equipment to start targeted bacteriostatic measures are achieved, and the safety of the system is improved. And the loss rate of aquatic products caused by microbial pollution is reduced.

Description

Technical Field

[0001] This invention relates to the field of cold chain transportation technology, and in particular to a monitoring system for cold chain transportation of aquatic products. Background Technology

[0002] The cold chain transportation monitoring system collects environmental data such as temperature, humidity, and gas concentration in real time through sensors. After analysis by software algorithms, it intelligently adjusts hardware equipment to achieve visualized and intelligent management of the transportation process, thus ensuring the quality of fresh products.

[0003] Traditional cold chain monitoring systems rely heavily on fixed temperature control, lacking accurate prediction and comprehensive regulation of microbial growth dynamics. On the one hand, a single temperature control strategy cannot effectively cope with the impact of complex environmental factors such as temperature fluctuations and changes in gas composition on microbial reproduction; on the other hand, the lack of real-time microbial monitoring and prediction capabilities prevents the implementation of targeted antibacterial measures before the colony count exceeds the standard, resulting in significant losses of aquatic products due to microbial contamination during transportation. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a monitoring system for cold chain transportation of aquatic products.

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

[0006] A cold chain transportation monitoring system for aquatic products includes a hardware layer and a software layer. The hardware layer includes a sensing and control module, a structural support module, and a communication and interaction module. The software layer includes a data management module, an intelligent computing module, and a business application module.

[0007] The perception and control module includes an environmental perception submodule and an equipment control submodule; the structural support module includes a carriage body submodule and a cargo protection submodule; and the communication and interaction module includes an edge computing submodule and an in-vehicle interaction submodule.

[0008] The data management module includes a data acquisition submodule and a data storage submodule; the intelligent computing module includes a predictive analysis submodule and an optimization decision-making submodule; and the business application module includes a monitoring and early warning submodule and a collaborative management submodule.

[0009] The environmental perception submodule acquires data through sensors, the equipment control submodule dynamically adjusts the parameters of the refrigeration equipment based on the data from the software layer, the structural support module divides the interior of the compartment into multiple areas and provides protection for aquatic products, and the communication interaction module is used to transmit data. The vehicle interaction submodule is used to provide an operating interface.

[0010] The data management module is used for data preprocessing and storage. The predictive analysis submodule uses a model to predict the microbial reproduction trend in advance and calculates the remaining shelf life using the Arrhenius equation. The optimization decision submodule is used to generate temperature control strategy schemes.

[0011] Preferably, the collaborative management submodule synchronizes order information with the enterprise's ERP system through an API interface, and the collaborative management submodule automatically adjusts the order priority according to the product's shelf life.

[0012] Preferably, the main body sub-module of the carriage avoids temperature conflict between different types of aquatic products through a multi-temperature zone heat insulation structure, and the cargo protection sub-module forms a buffer through an airbag shock absorption system.

[0013] Preferably, the dashboard of the monitoring and early warning submodule is developed based on the Vue.js framework and the data is updated every second via the WebSocket protocol.

[0014] Preferably, the vehicle interaction submodule sends the airbag pre-inflation command via the CAN bus.

[0015] Preferably, the monitoring and early warning submodule is used to display monitoring data in real time and issue early warnings based on preset thresholds.

[0016] Preferably, the perception and control module collects environmental data and transmits it to the software layer data management module through the communication interaction module.

[0017] Preferably, after the data management module processes and stores the data, the intelligent computing module analyzes and predicts the data to generate a control strategy. The business application module sends instructions to the hardware layer according to the control strategy. After the hardware layer executes the instructions, it feeds back the execution results to the software layer through the perception control module and the communication interaction module.

[0018] The present invention has the following beneficial effects:

[0019] 1. By constructing a real-time monitoring-dynamic prediction-intelligent regulation microbial inhibition system, high-precision sensors are used to collect environmental data in real time, and the LSTM-Attention model is combined to accurately predict the growth trend of microorganisms. This achieves the effect of early warning before the colony count reaches the danger threshold and automatically coordinating equipment to start targeted antibacterial measures, thereby reducing the loss rate of aquatic products caused by microbial contamination.

[0020] 2. By adopting a hardware-precise sensing, software-intelligent decision-making, and module-efficient collaborative architecture, and utilizing multi-protocol access units to achieve rapid acquisition of sensor data, the edge computing submodule is used to preprocess the data and optimize instructions. Combined with the precise control of hardware devices, this significantly improves the accuracy of environmental parameter monitoring, significantly shortens equipment response time, effectively reduces the overall energy consumption of the system, and greatly enhances the stability and economy of cold chain transportation.

[0021] 3. By deeply integrating the real-time monitoring-dynamic prediction-intelligent regulation microbial inhibition system with the hardware precision perception-software intelligent decision-making-module efficient collaborative architecture, guided by the early warning of the microbial prediction model, and combined with the agile response of the hardware modules and the flexible strategy adjustment of the software algorithm, a seamless connection between risk prediction, strategy generation and execution is achieved. This significantly extends the shelf life of aquatic products and greatly reduces the overall loss cost during transportation, creating a highly intelligent and efficient solution for the cold chain transportation industry. Attached Figure Description

[0022] Figure 1 This is an architecture diagram of a cold chain transportation monitoring system for aquatic products proposed in this invention;

[0023] Figure 2 The flowchart illustrates the process of inhibiting microorganisms in a cold chain transportation monitoring system for aquatic products proposed in this invention. Detailed Implementation

[0024] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0025] Example 1: Overall System Framework

[0026] This system is divided into a hardware layer and a software layer. The hardware layer consists of a perception and control module, a structural support module, and a communication and interaction module. The software layer consists of a data management module, an intelligent computing module, and a business application module.

[0027] Hardware Layer: The perception and control module is used to realize real-time perception of the transportation environment and precise control of equipment, providing raw data for microbial prediction and preservation strategies, and executing control commands issued by the software layer. It includes an environmental perception submodule and an equipment control submodule.

[0028] The environmental perception submodule uses sensors for temperature, humidity, gas composition, vibration, and other sensors to collect multi-dimensional data inside the carriage in real time, accurately capturing environmental changes and providing key inputs for the microbial prediction model in the software layer.

[0029] The equipment control submodule dynamically adjusts the parameters of the refrigeration equipment according to the instructions of the software layer. The gas regulation unit maintains a high oxygen environment of 21%-23% in the live fresh mode to inhibit anaerobic bacteria, or reduces the O2 concentration to 5%-8% in the chilled mode to inhibit the reproduction of aerobic bacteria, thus directly inhibiting the growth of microorganisms from the environmental level.

[0030] The structural load-bearing module is used to optimize the physical structure of the transport vehicle, improving preservation and cargo safety through zoning design and shock absorption protection. The structural load-bearing module includes a main body sub-module and a cargo protection sub-module.

[0031] The main body module of the carriage achieves zoned control for -5℃ freezing, 0℃ chilled, and 4℃ live seafood through a multi-temperature zone insulation structure (100mm thick polyurethane insulation layer), avoiding preservation failure caused by conflicting temperature requirements of different types of aquatic products; the live seafood breeding system maintains stable water quality (pH 7.0-8.5) through circulating water filtration and constant temperature heating, reducing the risk of decreased immunity and microbial infection caused by stress response in live seafood.

[0032] The cargo protection submodule's airbag shock absorption system automatically inflates when it detects severe vibrations, forming a thick buffer layer to reduce surface damage to fish and shrimp caused by collisions, thereby reducing the risk of bacteria entering through wounds.

[0033] The communication and interaction module is used to build a data transmission bridge between hardware devices and software systems, and to provide a human-machine interface. It includes an edge computing submodule and an in-vehicle interaction submodule.

[0034] The edge computing submodule performs real-time calibration of sensor data through the Node-RED platform and stores preset control strategies to ensure the reliability of the system in complex network environments. At the same time, it converts industrial protocol data such as Modbus into MQTT format and uploads it to the software layer once per second with a latency of less than 100ms.

[0035] The vehicle-mounted interactive submodule provides the driver with an intuitive operating interface, allowing them to view the microbial prediction curve and equipment operating status in real time, and manually adjust the temperature setpoint. The audible and visual alarm unit uses different frequencies of sound and light colors, with a high-frequency red sound corresponding to the first-level warning, to quickly transmit abnormal information and help operators respond in a timely manner. Emergency cooling will be activated within 5 minutes if the temperature exceeds the limit.

[0036] Software Layer: The data management module is responsible for the collection, storage, and preprocessing of data throughout the entire process, building the system's data foundation. The data management module includes a data acquisition submodule and a data storage submodule.

[0037] The data acquisition submodule is compatible with various hardware devices through a multi-protocol driver pool, uses a circular buffer design to ensure high-frequency data acquisition without loss, and performs anti-interference filtering on vibration data to improve data quality; it also supports offline data retransmission to ensure data integrity during network interruptions.

[0038] The time-series database cluster of the data storage submodule can store several years of historical data and supports querying and trend analysis; the model training database independently stores microbial measured values ​​and environmental parameters, providing training samples for the model and continuously optimizing prediction accuracy.

[0039] The intelligent computing module is used to implement microbial prediction, strategy optimization, and path planning using algorithmic models, outputting intelligent decisions. It includes a predictive analysis submodule and an optimization decision submodule.

[0040] The predictive analysis submodule's model can predict microbial reproduction trends 24 hours in advance by learning historical temperature and humidity sequences and colony count variations. Combined with the Arrhenius equation, it calculates the remaining shelf life, providing a scientific basis for transportation scheduling.

[0041] The NSGA-II algorithm in the optimization decision-making submodule can generate temperature control strategy schemes within 2 hours, compare and analyze spoilage time and energy consumption costs, and finally recommend the optimal scheme that balances efficiency and cost; the path planning unit combines real-time traffic data to plan the shortest route for goods with tight shelf life, avoiding the risk of microbial contamination due to delays.

[0042] The business application module is used to transform algorithm results into actionable business functions, enabling monitoring, early warning, and cross-departmental collaboration. It includes a monitoring and early warning submodule and a collaborative management submodule.

[0043] The real-time dashboard of the monitoring and early warning submodule displays transportation data in visual charts, allowing operators to drill down and view control records at any time point; the multi-level early warning engine notifies relevant personnel through multiple channels according to preset rules.

[0044] The collaborative management submodule synchronizes order information with the enterprise ERP system through an API interface. When the software layer detects that a batch of products has insufficient shelf life, it automatically adjusts the order priority. The equipment operation and maintenance unit predicts the wear trend of the compressor based on fault tree analysis and generates maintenance work orders in advance, turning passive repair into proactive maintenance and reducing the equipment failure rate.

[0045] Example 2: Interaction between Software and Hardware Layers

[0046] The data acquisition submodule maintains the Modbus RTU slave protocol driver pool and establishes a communication link with the sensor through the COM port. A query frame is sent every 500ms (frame format: slave address 0x01 + function code 0x03 + register address 0x0000 + register quantity 0x0002 + CRC checksum). The sensor returns a response frame containing the raw temperature and humidity values ​​(2 bytes temperature + 2 bytes humidity, two's complement format).

[0047] The signal conditioning circuit driver linearizes the raw value: the temperature value is calculated using the formula T = -45 + (175.72 × $raw t Convert emp$ / 65536 to actual temperature (unit: °C), humidity value is calculated using RH=-6+(125×$raw) h The value (um$ / 65536) is converted to relative humidity (unit: %RH), and a second-order Butterworth low-pass filter is applied to eliminate onboard power supply ripple interference.

[0048] For gas sensors, the Alphasense O2-A2 transmits data via the UART1 channel (9600 baud rate, 8 data bits, 1 stop bit), outputting data in the format "O2:XX.X%"; the SenseAir S8 transmits data via the UART2 channel (19200 baud rate), outputting JSON data containing CO2 concentration (e.g., {"co2":1200}). The data acquisition submodule uses multi-threading technology to read the two data streams in parallel, employing a timestamp alignment algorithm (based on the system's high-precision clock) to ensure synchronization of the two types of data, with the error controlled within 5ms.

[0049] Vibration data analysis:

[0050] The sensor communicates with the FPGA interface of the data acquisition submodule via an SPI bus (5MHz clock frequency), outputting raw triaxial acceleration values ​​(16-bit two's complement) every 10ms. A sliding window filtering algorithm (window size 10 sampling points, i.e., 100ms) is used to calculate the root mean square (RMS) of the data, using the following formula:

[0051] Where n=10, if the result of three consecutive calculations exceeds 1.5g, it is judged as a severe vibration event, triggering the following operation:

[0052] 1. Send a pre-inflation signal (3.3V high level for 200ms) to the hardware-layer airbag shock absorption system via GPIO pin;

[0053] 2. Record the event time, vibration intensity, and corresponding carriage location in the data buffer queue (associated with the sensor installation location tag).

[0054] Data storage submodule:

[0055] The time-series database unit is built on a distributed cluster based on InfluxDB and uses a three-level tagging system of "carriage-temperature zone-sensor" to store data. Taking the temperature data of the chilled zone as an example, the storage structure is as follows: Measurement name: temperature, Tags: car_id=Car01, zone=FreshZone, sensor_type=SHT45, Fields: value=2.28, unit="℃".

[0056] The data compression strategy uses the LZ4 algorithm, achieving a compression ratio of 3:1. Queries are performed using the statement `SELECT mean(value) FROM temperature WHERE time>now()-24hGROUPBYz one`, which returns the average temperature curve for the temperature zone within 10ms.

[0057] Command Log Management: The command log management unit uses a relational database (PostgreSQL) to store command records. The table structure includes:

[0058] instruction_id(UUID, primary key)

[0059] hardware_address(VARCHAR, such as "0x01")

[0060] command(TEXT, such as "SET_TEMP-2.0")

[0061] issue_time(TIMESTAMPWITHTIMEZONE)

[0062] expected_completion_time(TIMESTAMPWITHTIMEZONE)

[0063] actual_completion_time(TIMESTAMPWITHTIMEZONE)

[0064] status(ENUM, with values ​​of "success", "failed", or "timeout")

[0065] After the instruction is issued, if the actual completion time exceeds the expected execution time + 500ms, the status will be automatically marked as "timeout" and the timeout alarm process of the monitoring and early warning submodule will be triggered.

[0066] Predictive Analysis Submodule: The microbial prediction model is deployed based on the TensorFlow framework to implement the LSTM-Attention model. The input layer receives the temperature and humidity sequence of the past 12 hours (a total of 144 time steps, one sample every 5 minutes). The LSTM layer extracts the temporal features, the Attention layer calculates the contribution of each time step to the prediction result, and the output layer generates the predicted colony count for the next 24 hours through a fully connected layer (one prediction point per hour).

[0067] The model training data includes over 100,000 historical transportation records. Feature engineering includes: environmental features: temperature, humidity, O2 concentration, and CO2 concentration.

[0068] Product characteristics: aquatic product type (one-hot encoding), initial colony count, packaging type. Transportation characteristics: travel speed, vibration intensity. The confidence score of the prediction results is calculated using an ensemble learning method, based on the mean predictions of five models with different initial weights. The confidence score formula is as follows:

[0069] Where σ is the standard deviation of the five model predictions, and μ is the mean. When the confidence level is <80%, the manual sampling process of the hardware-layer ATP detector is triggered: after the driver confirms through the vehicle terminal, the instrument samples the surface of the goods and returns the test results (CFU / g) within 10 minutes. The data is transmitted to the data acquisition submodule via Bluetooth.

[0070] The RST calculation unit maintains a database of activation energies for different aquatic products (e.g., shrimp Ea = 55 kJ / mol, salmon Ea = 45 kJ / mol), and automatically retrieves the corresponding parameters based on the product type in the order. The calculation results are corrected using exponential smoothing, with the formula: RST final =α×RST calc +(1-α)×RST prev Where α = 0.8, RST prev The value is from the previous calculation to avoid drastic changes in the RST result due to instantaneous temperature fluctuations.

[0071] The optimization decision submodule uses the NSGA-II algorithm. Individuals are encoded as 24-dimensional vectors (each element represents the temperature setpoint for the corresponding hour, ranging from -5℃ to 5℃). The fitness function is: f_1=-\text{predicted_shelf_life}\quad(\text{maximize preservation time}) f_2=\sum_{t=1}^{24}\text{compressor_power}(t)\quad(\text{minimize energy consumption}).

[0072] During the algorithm iteration process, population diversity is maintained through crowding calculation. The selection operation adopts binary tournament selection (tournamentsize=2), the crossover operation adopts simulated binary crossover (SBX, distribution index=20), and the mutation operation adopts multinomial mutation (distribution index=20). 50 individuals are generated in each generation, and after 100 generations of iteration, the strategy with the best overall performance is selected from the Pareto front.

[0073] The route planning unit integrates the OpenRouteService API, supporting custom truck parameters to avoid restricted road sections. When congestion is detected, the following joint optimization steps are performed: 1. Calculate the additional travel time Δt and additional fuel consumption Δf of the detour route;

[0074] Calculate the temperature loss per unit time based on the compressor's current cooling power P (kW) and ambient temperature T_env (°C); ΔT = k × P × Δt, (k is the heat dissipation coefficient, 0.05°C / (kW·h));

[0075] If ΔT > RST remaining time × allowable temperature rise rate (e.g., 0.5℃ / hour), then the emergency cooling mode of the hardware layer is activated: the compressor runs at 110% overload power (duration ≤ 2 hours), while non-essential hardware (e.g., vehicle lighting) is turned off, prioritizing temperature control.

[0076] The monitoring and early warning submodule: The real-time dashboard unit is developed based on the Vue.js framework and uses the ECharts heatmap component to display the temperature distribution of temperature zones. The data is updated every second via the WebSocket protocol.

[0077] The early warning grading control logic is divided into blue and red warnings. Blue warnings are used to address vibration situations; when vibration > 2g lasts for 5 minutes: 1. The early warning rule engine analyzes the vibration data.

[0078] After parsing the vibration data, the early warning rule engine generates an early warning event object {type:'vibration_alert',level:'blue',zone:'LeftShelf',time:'2025-06-05T14:35:00Z'};

[0079] The vehicle interaction submodule calls the Text-to-Speech engine to generate voice prompts; it sends an airbag pre-inflation command via the CAN bus: CAN frame ID = 0x123, data field = 0x01 (pre-inflation mode). After the hardware layer receives the command, the airbag pressure increases from 0.1MPa to 0.2MPa.

[0080] Red alert: Colony count ≥ 8 × 10⁻⁶ 5 CFU / g, Emergency Response Procedure Generates Command Sequence:

[0081] Instruction 1: Modbus write register (address 0x4000, value = -3.0), target temperature -3℃;

[0082] Command 2: RS485 sends "N2_ON50", nitrogen flow rate 50L / min;

[0083] Command 3: Send an HTTP POST request to the SMS gateway with parameters {phone:'138XXXX1234',content:'XX carriage microbial warning...'};

[0084] Hardware layer instruction execution timeout handling: If no temperature arrival feedback from the compressor is received within 15 minutes (the current temperature is queried via Modbus register 0x3002), the system automatically switches to the backup compressor and generates a fault work order {type:'compressor_failure',priority:'high'}.

[0085] The order matching program in the collaborative management submodule uses a fuzzy matching algorithm to calculate a matching score (0-100 points) between order requirements and vehicle hardware. The calculation formula is: score$ = 40 × δ zone +30×δ temp +20×δ gas +10×δ freshness Where: δzone: whether there is a corresponding temperature zone (yes = 1, no = 0);

[0086] δ temp Temperature control accuracy of the temperature zone is ≤ ±0.5℃ (Yes = 1, No = 0);

[0087] δ gas : Whether the target gas mode is supported (e.g., live oxygen-enriched = 1, otherwise = 0);

[0088] δ freshness : Is the remaining RST time for the vehicle greater than the required time for the order (Yes = 1, No = 0).

[0089] Priority S-level orders require a matching score of ≥90. Once a match is successful, the vehicle is automatically locked, and the electronic fence function of the in-vehicle interaction submodule is used to limit the vehicle from deviating from the preset route by ±5 kilometers.

[0090] The data acquisition submodule acts as an NTP client, sending a synchronization request to the server every 10 minutes. It adjusts its local clock using the symmetric NTP protocol (RFC 5905), achieving a synchronization accuracy of ±10ms. The hardware-layer sensors ensure that the data acquisition timestamp matches the software-layer clock through a hardware timestamp marking circuit (FPGA-based PTP clock synchronization).

[0091] The redundancy control program monitors the status of the master device through a heartbeat detection mechanism (sending a PING command every 2 seconds). The master-slave switchover process is as follows: 1. If the master device fails to respond to the heartbeat 3 times in a row, it is determined to be faulty;

[0092] 2. Send a TAKE_OVER command (Modbus function code 0x0F) to the standby device, including the current control parameters (such as target temperature and gas concentration);

[0093] 3. After receiving the instruction, the standby device will take over control within 500ms, and the main device will enter maintenance mode;

[0094] 4. Record the switch log to the data storage submodule and trigger the main equipment maintenance process of the equipment operation and maintenance unit.

[0095] When the path planning unit detects a rainstorm warning within 50 kilometers ahead (obtained via the weather API), the following contingency plan is executed: 1. Call the predictive analysis submodule to simulate temperature fluctuations under the rainstorm scenario;

[0096] 2. Based on the simulation results, the temperature in the carriage was lowered from 0℃ to -1℃ two hours in advance to establish a temperature buffer zone;

[0097] 3. Send automatic wiper control commands to the hardware layer (control the wiper motor via GPIO) and prompt the driver to turn on the low beam headlights through the vehicle interaction submodule;

[0098] 4. If heavy rain causes road closures, activate the cold start plan: shut down non-core hardware, maintain only the minimum cooling power, and maximize battery life.

[0099] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can explore the technical aspects disclosed in this invention.

[0100] Within the scope of the invention, equivalent substitutions or modifications may be made to the technical solution and inventive concept of the present invention.

[0101] All changes should be covered within the scope of protection of this invention.

Claims

1. A monitoring system for cold chain transportation of aquatic products, characterized in that, It includes a hardware layer and a software layer. The hardware layer includes a perception and control module, a structural support module, and a communication and interaction module. The software layer includes a data management module, an intelligent computing module, and a business application module. The perception and control module includes an environmental perception submodule and an equipment control submodule; the structural support module includes a carriage body submodule and a cargo protection submodule; and the communication and interaction module includes an edge computing submodule and an in-vehicle interaction submodule. The data management module includes a data acquisition submodule and a data storage submodule; the intelligent computing module includes a predictive analysis submodule and an optimization decision-making submodule; and the business application module includes a monitoring and early warning submodule and a collaborative management submodule. The environmental perception submodule acquires data through sensors, the equipment control submodule dynamically adjusts the parameters of the refrigeration equipment based on the data from the software layer, the structural support module divides the interior of the compartment into multiple areas and provides protection for aquatic products, and the communication interaction module is used to transmit data. The vehicle interaction submodule is used to provide an operating interface. The data management module is used for data preprocessing and storage. The predictive analysis submodule uses a model to predict the microbial reproduction trend in advance and calculates the remaining shelf life using the Arrhenius equation. The optimization decision submodule is used to generate temperature control strategy schemes.

2. The aquatic product cold chain transportation monitoring system according to claim 1, characterized in that, The collaborative management submodule synchronizes order information with the enterprise's ERP system through an API interface, and automatically adjusts the order priority based on the product's shelf life.

3. The aquatic product cold chain transportation monitoring system according to claim 1, characterized in that, The main body sub-module of the carriage avoids temperature conflicts between different types of aquatic products through a multi-temperature zone heat insulation structure, and the cargo protection sub-module forms a buffer through an airbag shock absorption system.

4. The aquatic product cold chain transportation monitoring system according to claim 1, characterized in that, The dashboard of the monitoring and early warning submodule is developed based on the Vue.js framework and the data is updated every second via the WebSocket protocol.

5. A cold chain transportation monitoring system for aquatic products according to claim 1, characterized in that, The vehicle interaction submodule sends airbag pre-inflation commands via the CAN bus.

6. The aquatic product cold chain transportation monitoring system according to claim 1, characterized in that, The monitoring and early warning submodule is used to display monitoring data in real time and issue early warnings based on preset thresholds.

7. The aquatic product cold chain transportation monitoring system according to claim 1, characterized in that, The perception and control module collects environmental data and transmits it to the software layer data management module through the communication interaction module.

8. A cold chain transportation monitoring system for aquatic products according to claim 1, characterized in that, After the data management module processes and stores the data, the intelligent computing module analyzes and predicts it to generate control strategies. The business application module sends instructions to the hardware layer according to the control strategy. After the hardware layer executes the instructions, it feeds back the execution results to the software layer through the perception control module and the communication interaction module.

Images