Multi-sensor network adaptive perception system for a gas-controlled refrigerated vehicle

Abstract

The present application relates to the technical field of intelligent control of refrigerated transport, and particularly relates to a multi-sensor network adaptive sensing system for a controlled atmosphere refrigerated vehicle, comprising: a loading identification module, a transport identification module, a data acquisition module, a gas regulating device, and a main sensor group baseline sampling frequency determined according to a transport stage; the dispatch controller further switches from the sensor group to a preheating standby state and increases the baseline sampling frequency according to the current time being in a respiratory jump early warning period; a change curve is generated according to CO2 data of the main sensor group, a threshold group is selected according to the transport stage to determine whether the respiratory jump condition is met, and the slave sensor group is switched to a working state to perform multi-point sampling in response to the respiratory jump condition being met, so as to locate a CO2 enrichment area and control corresponding independent control regulating units to perform partition intervention. The present application realizes predictive awakening of the sensor and partition intervention, and reduces system energy consumption.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for refrigerated transportation, and in particular to a multi-sensor network adaptive sensing system for controlled atmosphere refrigerated trucks. Background Technology

[0002] Modified atmosphere refrigerated trucks are core equipment for the cold chain transportation of fresh agricultural products. By regulating environmental parameters such as oxygen, carbon dioxide, temperature, and humidity inside the truck during transportation, they inhibit the respiration of fruits and vegetables, delaying ripening and senescence. To achieve precise control, modern modified atmosphere refrigerated trucks are typically equipped with a multi-parameter sensor network to monitor changes in the microenvironment inside the truck in real time, and to coordinate with modified atmosphere actuators such as nitrogen generators, carbon dioxide removers, and ethylene removers to maintain the preset gas ratio.

[0003] Currently, most mainstream controlled atmosphere refrigerated truck sensor systems adopt a centralized control architecture. All sensors (oxygen sensors, carbon dioxide sensors, temperature sensors, etc.) continuously collect data at a fixed frequency and upload it to the central controller, which then performs unified calculations and issues commands to the actuators. Under this architecture, the sensor network operates at full load for extended periods, resulting in high energy consumption and a continuous burden on the vehicle's power system. Some improved solutions introduce simple threshold control logic, which activates the actuators when environmental parameters exceed set ranges. However, the sensor network itself still operates in a constantly active or continuous monitoring mode, lacking fine-grained scheduling of sensor resources.

[0004] Furthermore, existing technologies do not adequately consider the dynamic changes during transportation. Controlled atmosphere refrigerated trucks operate under various conditions, including long-distance trunk transport and short-distance urban delivery, with significantly different monitoring requirements under each condition. However, existing sensor systems often employ a one-size-fits-all approach, resulting in energy waste or monitoring blind spots. On the other hand, different products (such as climacteric fungi and non-climacteric leafy vegetables) exhibit vastly different physiological characteristics, with varying sensitive periods and thresholds for environmental parameters. Existing systems typically rely on fixed preset values ​​for control, failing to dynamically adjust monitoring strategies based on the real-time physiological state of the products.

[0005] Chinese Patent Publication No. CN222418107U discloses a controlled atmosphere refrigeration device, including a cabinet with a refrigeration chamber inside. It also includes a refrigeration device for controlling the temperature of the refrigeration chamber, a nitrogen supply device for supplying nitrogen to the refrigeration chamber, a carbon dioxide supply device for supplying carbon dioxide to the refrigeration chamber, and a control device for executing commands. The nitrogen supply device and the carbon dioxide supply device are respectively connected to the refrigeration chamber. This utility model's controlled atmosphere refrigeration device is based on a vertical freezer with optimized design. It adds key components such as gas cylinders, gas monitoring sensors, gas supply devices, and humidifiers to the vertical freezer, resulting in a controlled atmosphere freezer that displays parameters such as internal temperature, humidity, oxygen concentration, and carbon dioxide concentration. Users can adjust the internal temperature, humidity, and gas concentration according to their needs and the displayed parameters, thereby creating a suitable environment for fruit and vegetable storage and improving the preservation effect. However, the following problems exist:

[0006] The sensor networks of existing controlled atmosphere refrigerated trucks generally adopt a fixed working mode of all-time monitoring and centralized control. This mode cannot perform differentiated scheduling and energy consumption optimization of sensor resources according to the dynamic changes of the goods' biological clock and transportation stages. This results in energy waste during non-critical periods and the inability to achieve accurate positioning and zoned intervention during critical periods. Summary of the Invention

[0007] To address this, the present invention provides a multi-sensor network adaptive perception system for modified atmosphere refrigerated trucks, which overcomes the problem in the prior art that it is impossible to differentiate and optimize the energy consumption of sensor resources according to the dynamic changes of the goods' biological clock and transportation stage, resulting in energy waste during non-critical periods and the inability to achieve accurate positioning and zonal intervention during critical periods.

[0008] To achieve the above objectives, the present invention provides a multi-sensor network adaptive perception system for controlled atmosphere refrigerated trucks, comprising:

[0009] A recognition module is installed to obtain physiological stage prediction parameters corresponding to the current goods. The physiological stage prediction parameters include respiratory jump warning period and jump judgment threshold group.

[0010] The transportation identification module is used to generate transportation stage identifiers based on vehicle speed and door opening / closing detection sensors;

[0011] The data acquisition module includes a main sensor group and several slave sensor groups, and each of the slave sensor groups constitutes a multi-point distributed sensor array.

[0012] The gas conditioning device includes several independent control and regulation units, each of which corresponds to an internal compartment of the carriage.

[0013] The scheduling controller is used to determine the baseline sampling frequency of the main sensor group according to the transportation stage, determine the working status of the slave sensor group according to the relationship between the current time and the respiratory jump warning period, generate a carbon dioxide change curve based on the carbon dioxide data collected by the main sensor group and select the corresponding jump judgment threshold group according to the transportation stage, and determine whether the respiratory jump condition is met according to the relationship between the carbon dioxide concentration value and the carbon dioxide concentration change rate and the jump judgment threshold group.

[0014] In addition, in response to the fulfillment of the respiratory pulsation condition, the sensor group is switched to the working state and multi-point sampling is performed. Based on the multi-point sampling results, the carbon dioxide enrichment area is located and the corresponding independent control and regulation unit is controlled to perform zoned intervention.

[0015] As a preferred technical solution for a multi-sensor network adaptive perception system used in controlled atmosphere refrigerated trucks, the data acquisition module includes:

[0016] The main sensor group, which is integrated and installed in the central ceiling or return air duct area inside the transport compartment, includes at least one carbon dioxide sensor to collect the overall environmental parameters of the compartment.

[0017] Each sensor node in the carriage is equipped with a slave sensor group, and each slave sensor group includes at least one carbon dioxide sensor and one temperature sensor. The sensor nodes are fixed to the crossbeams between the shelf layers or the side wall of the carriage by detachable connectors to obtain microenvironmental parameters at different spatial locations in the carriage.

[0018] As a preferred technical solution for a multi-sensor network adaptive perception system for controlled atmosphere refrigerated trucks, the transportation identification module generates a transportation stage identifier based on the number of door openings and the average vehicle speed within a preset time window, including:

[0019] If the average vehicle speed is not less than the preset vehicle speed and the number of door openings is not greater than the preset number, then a long-distance steady-state stage identifier is generated.

[0020] Conversely, generate short-distance delivery stage identifiers;

[0021] The transportation phase includes a long-distance steady-state phase and a short-distance delivery phase.

[0022] As a preferred technical solution for a multi-sensor network adaptive perception system for controlled atmosphere refrigerated trucks, the dispatch controller determines the baseline sampling frequency of the main sensor group based on the transportation stage, including...

[0023] In response to the transportation phase identifier being identified as a long-haul steady-state phase identifier, the baseline sampling frequency is determined to be the default baseline sampling frequency;

[0024] In response to the transportation phase being identified as a short-distance delivery phase, the baseline sampling frequency is determined to be higher than the default baseline sampling frequency.

[0025] As a preferred technical solution for a multi-sensor network adaptive perception system used in controlled atmosphere refrigerated trucks, the dispatch controller determines the working status of the slave sensor group based on the current time and the breathing jump warning period, including:

[0026] In response to the fact that the current moment is not within the respiratory jump warning period, it is determined that the working state of the sensor group remains in a dormant state;

[0027] In response to the current time being in the respiratory pulsation warning period, the working state of the sensor group is switched to the preheating standby state and the baseline sampling frequency is increased;

[0028] The preheating standby state of the sensor group includes powering on the carbon dioxide sensor and maintaining preheating, and performing at least one of zero-point calibration or baseline calibration.

[0029] As a preferred technical solution for a multi-sensor network adaptive perception system for controlled atmosphere refrigerated trucks, the scheduling controller acquires the time-series values ​​of carbon dioxide concentration of the main sensor group within a default time window and generates the carbon dioxide change curve by connecting them in time sequence, and determines the carbon dioxide concentration change rate by the ratio of the concentration difference to the time difference between adjacent sampling times.

[0030] As a preferred technical solution for a multi-sensor network adaptive perception system used in controlled atmosphere refrigerated trucks, the dispatch controller selects a corresponding set of threshold judgments based on the transportation stage, including:

[0031] Based on the determination result that the transportation stage is identified as the long-distance steady-state stage, a low jump judgment threshold group is selected.

[0032] Based on the determination result that the transportation stage is identified as the short-distance delivery stage, a high jump judgment threshold group is selected.

[0033] The jump determination threshold group includes a concentration threshold and a change rate threshold;

[0034] Wherein, the threshold values ​​of the high transition determination threshold group are all not less than the corresponding threshold values ​​of the low transition determination threshold group.

[0035] As a preferred technical solution for a multi-sensor network adaptive perception system for controlled atmosphere refrigerated trucks, the scheduling controller determines whether the respiratory jump condition is met based on the relationship between the carbon dioxide concentration value and the rate of change of carbon dioxide concentration both being greater than the corresponding threshold of the jump judgment threshold group.

[0036] As a preferred technical solution for a multi-sensor network adaptive perception system for controlled atmosphere refrigerated trucks, the scheduling controller constructs a carbon dioxide spatial distribution matrix based on the carbon dioxide concentration values ​​of the sensor group at the same sampling time, and determines the set of sampling points in the carbon dioxide spatial distribution matrix whose concentration values ​​are greater than or equal to the enrichment threshold as carbon dioxide enrichment areas to determine the spatial center location and spatial range of the carbon dioxide enrichment areas.

[0037] As a preferred technical solution for a multi-sensor network adaptive perception system for controlled atmosphere refrigerated trucks, the scheduling controller controls the independent control and regulation unit corresponding to the spatial center position of the carbon dioxide enrichment zone to perform carbon dioxide control, and determines whether to control adjacent independent control and regulation units to cooperate in carbon dioxide control based on the spatial range.

[0038] Compared with existing technologies, the advantages of this invention are as follows: This invention provides a multi-sensor network adaptive perception system for modified atmosphere refrigerated trucks. By introducing a dual-modal driven intelligent scheduling mechanism based on the physiological clock of the goods and the disturbance of the transportation stage, it achieves refined energy consumption management of the sensor network and precise intervention in the modified atmosphere environment. This system has built-in physiological stage prediction parameters for the goods to switch the sensor group to a preheating standby state and increase the sampling frequency of the main sensor before the arrival of the respiratory leap warning period. During non-warning periods, it keeps the secondary sensors in deep sleep and the main sensor sampling at a low frequency to avoid energy waste caused by full-time full-load monitoring. At the same time, it identifies the transportation stage in real time by vehicle speed and door opening and closing, and dynamically selects the appropriate mode based on the stage. By using a threshold group for jump detection, the system avoids invalid wake-ups caused by false triggers while ensuring the accuracy of jump detection. When the respiratory jump condition is met, the system immediately wakes up to perform multi-point sampling from the sensor group and constructs a CO2 spatial distribution matrix to accurately locate the center position and spatial range of the enrichment area. It then controls the corresponding independent control and adjustment unit to perform targeted intervention. If the enrichment area is large, it coordinates with adjacent units for linkage adjustment to avoid energy waste and environmental fluctuations caused by global ventilation. This invention upgrades the sensor network of the modified atmosphere refrigerated truck from a passive response and all-time operation to an intelligent sensing system with active prediction and dynamic scheduling, achieving optimal monitoring energy consumption and maximum control accuracy while ensuring the freshness of the goods.

[0039] In particular, the loading identification module, through its built-in cargo physiological information database, makes sensor scheduling monitoring more targeted. Based on predictive knowledge of the post-harvest physiological patterns of the cargo, it provides the system with a respiratory pulsation warning period at the beginning of transportation, enabling the system to wake up from sensor preheating several hours before the actual arrival of the pulsation peak, achieving a leap from passive response to active prediction. At the same time, the stored pulsation judgment threshold group is independently set for each type of cargo, and a pulsation precursor threshold with a lower value than the conventional pulsation threshold is adopted based on the delay characteristics of the main sensor located on the top of the carriage. This ensures that the early pulsation signal can still be sensitively captured under single-point monitoring conditions at the top, while avoiding missed or false alarms caused by a uniform threshold. In addition, the module has a built-in temperature correction relationship so that the prediction parameters can adapt to the actual temperature changes during transportation, avoiding prediction deviations caused by temperature fluctuations. The loading identification module is the fundamental prerequisite for realizing predictive energy-saving scheduling and precise intervention.

[0040] In particular, the transportation identification module enables sensor scheduling to dynamically adjust according to the transportation scenario through the fusion of door magnetic switch and vehicle speed signal. By using dual judgment of vehicle speed and number of door openings, it accurately distinguishes between two typical transportation conditions: long-distance closed steady state and short-distance frequent disturbances, providing differentiated control basis for the dispatch controller. The statistical method based on time window effectively filters interference caused by instantaneous vehicle speed fluctuations or single accidental door openings, improving the robustness of stage identification. The direct physical detection method of door magnetic switch reliably records every door action, providing accurate condition labels for increasing the threshold for jump judgment in the delivery stage to filter door opening disturbances and lowering the threshold in the long-distance stage to sensitively capture jump signals. This achieves real-time matching between sensor scheduling strategy and transportation scenario, maximizing system energy consumption while ensuring monitoring accuracy.

[0041] In particular, the data acquisition module provides the system with dual-layer sensing capabilities through the differentiated layout and collaborative operation of the main sensor group and the slave sensor group. The main sensor group continuously monitors the overall environmental parameters of the carriage, providing a reliable global data foundation for the preliminary determination of respiratory change conditions. The main sensors capture overall trend changes, thus simplifying installation and maintenance while ensuring accuracy. The slave sensor group adopts a multi-point distributed sensor array, which not only achieves full coverage of micro-environmental parameters in different spatial locations within the carriage, but also avoids interference from fixed sensors during cargo loading and unloading. Furthermore, each sensor node is equipped with at least one carbon dioxide sensor and one temperature sensor. When the slave sensor group is activated to perform multi-point sampling, the CO2 concentration is compared with... Synchronous anomaly analysis of temperature can effectively verify whether CO2 enrichment originates from real physiological activity, eliminating gas stagnation caused by poor airflow or sensor drift interference, and providing dual credibility verification for subsequent location of carbon dioxide enrichment areas; the master-slave separation architecture of this module realizes the decoupling of global trend monitoring and local precise positioning functions, the multi-point distributed layout ensures that the spatial resolution is sufficient to support the location of enrichment areas, the detachable installation method takes into account both flexibility and equipment safety, and the dual-parameter verification mechanism significantly improves the reliability of enrichment area determination, enabling the system to accurately lock the spatial location and range that need intervention after the respiratory jump occurs, providing an accurate basis for subsequent zonal control, thereby minimizing ineffective intervention caused by misjudgment while ensuring the preservation effect;

[0042] In particular, the scheduling controller achieves coordinated optimization of sensor resources and controlled atmosphere execution through a multi-level intelligent scheduling algorithm: At the identification level during transportation, the controller dynamically sets the baseline sampling frequency of the main sensor group and controls the slave sensor group to remain dormant based on different operating conditions for long-distance steady-state and short-distance delivery. This satisfies the need for rapid capture of environmental disturbances during delivery while avoiding invalid wake-ups caused by door opening errors. At the jump warning level, based on the respiratory jump warning period provided by the loading identification module, the controller maintains deep dormancy of slave sensors and low-frequency sampling of the main sensor during non-critical periods to maximize energy saving. Upon entering the warning period, it immediately switches slave sensors to a preheating standby state and performs zero-point calibration on the CO2 sensor (additional baseline calibration is performed if the sensor has not been used for more than 7 days), while simultaneously increasing the sampling frequency of the main sensor to once every 5 minutes. At the jump judgment level, the controller calculates the CO2 concentration change rate in a 30-minute time window and dynamically selects a jump judgment threshold group based on the transportation stage. A low threshold is used for long-distance transport to sensitively capture early signals, while a high threshold is used for delivery to effectively filter door opening disturbances. Through the dual conditions of concentration and change rate, it ensures that the trigger only occurs when a true respiratory jump takes place. To prevent energy waste caused by false wake-ups, the controller responds immediately to sudden changes in conditions by waking up multiple points simultaneously from the sensor array. It constructs a CO2 spatial distribution matrix and identifies candidate enrichment points using enrichment thresholds. Through connected component analysis, it determines the spatial center and range of the enrichment area, then directs the nearest independent control unit to act as the master control unit for targeted intervention. If more than 25% of the enrichment area exceeds the master control unit's coverage, it coordinates with adjacent units in a gradient mode of full power master control and reduced power adjacent units. During intervention, it continuously monitors the CO2 concentration in each area until the maximum enrichment value drops below the termination threshold and stabilizes for 5 minutes before stopping. This controller deeply integrates four types of heterogeneous information: transportation stage conditions, cargo physiological clock, real-time CO2 data, and spatial distribution information. This forms a complete decision-making chain: stage-based baseline setting, early warning and pre-wake-up, threshold selection as needed, enrichment precision positioning, and zoned collaborative control. While ensuring the accuracy of respiratory change recognition and intervention precision, it minimizes sensor network energy consumption and maximizes atmospheric regulation execution efficiency, enabling the system to truly achieve an intelligent leap from passive response to proactive prediction. Attached Figure Description

[0043] Figure 1 This is a connection diagram of a multi-sensor network adaptive sensing system for a controlled atmosphere refrigerated truck according to an embodiment of the present invention;

[0044] Figure 2 This is a flowchart illustrating the workflow of the scheduling controller according to an embodiment of the present invention. Detailed Implementation

[0045] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0046] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0047] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0048] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0049] Please see Figure 1 The diagram shown is a connection diagram of a multi-sensor network adaptive sensing system for a controlled atmosphere refrigerated truck according to an embodiment of the present invention. This embodiment of the present invention provides a multi-sensor network adaptive sensing system for a controlled atmosphere refrigerated truck.

[0050] Specifically, the loading identification module of this system is a built-in database containing various goods and their corresponding physiological stage prediction parameters, including: 1. The respiratory climacteric warning period refers to the time window from warehousing to the appearance of the respiratory peak. For climacteric fruits, the respiratory peak marks the beginning of accelerated aging, and their storage tolerance decreases sharply after the peak. In this invention, the respiratory climacteric warning period is used to allow the system to wake up the sensor group in advance before the peak arrives; 2. The climacteric judgment threshold group includes the concentration threshold and the rate of change threshold. It can be understood that: the concentration threshold reflects the absolute value of respiratory intensity, and the CO2 tolerance thresholds of different goods vary significantly; the rate of change threshold reflects the severity of respiratory acceleration. During the respiratory climacteric period, the CO2 concentration usually shows a three-stage pattern of decrease-sudden increase-sudden decrease. The rate of change threshold is used to identify the starting point of the sudden increase;

[0051] In practice, the starting point of the respiratory jump warning period is the zero point when the goods are loaded into the refrigerated truck; the length of the period takes into account the typical occurrence time of the respiratory peak of the jump-type goods and the influence of the transportation temperature. In one implementation, the warning period can be dynamically adjusted according to the real-time temperature. For every 5°C increase in temperature, the warning period may be shortened by 20% to 30%; this system aims to wake up the sensor group in advance to preheat before the actual arrival of the jump peak.

[0052] In practice, this invention is applicable to fresh produce that exhibits climacteric behavior, including mushrooms (button mushrooms and shiitake mushrooms), climacteric fruits (Fuji apples, bananas, kiwis, mangoes, and peaches), and climacteric vegetables (tomatoes and avocados). The respiratory climacteric warning periods and climacteric determination threshold groups for these fresh produce are typically shown in Table 1.

[0053] Table 1

[0054]

[0055] It is understood that in the embodiments of the present invention, the main sensor is mostly located at the top (gas stratification may cause CO2 accumulation to start from inside the cargo pile first, and there is a delay in the response at the top). Therefore, the threshold group for determining the jump is lower than the threshold for the conventional jump that has occurred, that is, the value of the threshold group for determining the jump is equivalent to the threshold for the precursor of the jump.

[0056] Specifically, the door opening and closing detection sensor of the transportation identification module is a door magnetic switch. The permanent magnet of the door magnetic switch is installed on the edge of the door panel (moving part, rear door and side door) of the modified atmosphere refrigerated truck, and the reed switch is installed at the corresponding position on the door frame (fixed part) to ensure that the two are aligned when the door is closed, and the distance between them is less than the action distance (usually ≤10mm). It can be understood that when the door is closed, the magnet approaches the reed switch contact and closes, outputting a low level (or a high level, depending on the circuit design). When the door is opened, the magnet moves away from the contact and opens, outputting the opposite level. The sensor signal is connected to the dispatch controller and records the occurrence time and duration of each door opening event.

[0057] In implementation, the transportation identification module generates transportation stage identifiers based on the number of door openings and average vehicle speed within a preset time window, including:

[0058] If the average vehicle speed is not less than the preset vehicle speed and the number of door openings is not greater than the preset number, then a long-distance steady-state stage identifier is generated.

[0059] Conversely, generate short-distance delivery stage identifiers;

[0060] The transportation phase includes a long-distance steady-state phase and a short-distance delivery phase.

[0061] In practice, the transportation identification module uses a preset time window to count the average vehicle speed and number of door openings within that window to comprehensively determine the current transportation stage. The preset time window is usually 30 minutes (which can be adjusted within the range of 15 minutes to 60 minutes depending on the actual transportation scenario). The window data (i.e., the number of door openings and the average vehicle speed) within the last 30 minutes is updated every minute.

[0062] Understandably, the statistical window contains the arithmetic mean of all vehicle speed samples (vehicle speed signals come from the vehicle's CAN bus or GPS) and the number of door opening events detected within the statistical window (one door opening refers to the complete process from door opening to door closing, but in practice it is usually counted at the moment the door opens).

[0063] In practice, the preset vehicle speed is usually 30km / h to 45km / h. Long-distance transport vehicles usually travel on highways or national roads, and the speed is generally higher than 40km / h. However, urban delivery is affected by traffic lights and congestion, and the average vehicle speed is mostly lower than 40km / h. Therefore, it is preferable to set the preset vehicle speed to 40km / h. The preset number of door openings is usually 1 to 3 times per minute. During long-distance transport, vehicles may only open their doors when they take a short break at service areas, usually once every 2 to 4 hours, that is, the number of door openings is ≤ 1 time within 30 minutes. However, short-distance delivery requires frequent stops to unload goods, and the number of door openings is usually ≥ 2 times per 30 minutes. Therefore, it is preferable to set the preset number of door openings to a threshold of 1 time per 30 minutes.

[0064] Specifically, the data acquisition module includes a main sensor group and several slave sensor groups, wherein:

[0065] The main sensor group is integrated and installed in the central ceiling or return air duct area inside the transport compartment. It includes at least one carbon dioxide sensor and may also be equipped with an oxygen sensor and a temperature sensor to collect the overall environmental parameters of the compartment. In this invention, the data collected by the carbon dioxide sensor in the main sensor group is mainly used to determine whether the current breathing state of the transported goods meets the breathing jump condition in order to control its carbon dioxide.

[0066] Each sensor node in the carriage is equipped with a slave sensor group, and each of the sensor nodes is equipped with a slave sensor group. Each slave sensor group includes at least one carbon dioxide sensor and one temperature sensor. The sensor nodes are fixed to the crossbeams between the shelf layers or the side wall of the carriage by disassembly connectors to obtain microenvironmental parameters at different spatial locations in the carriage.

[0067] It is understandable that when the carriage is equipped with fixed or adjustable shelves, the sensor nodes are evenly distributed on the crossbeams between all shelf layers. The sensor nodes are fixed at the lower edge of the crossbeam or the side of the longitudinal beam of each shelf layer by the above-mentioned disassembly connectors. The nodes need to face the direction of the stack of goods or the gap between goods in order to collect the micro-environmental parameters around the goods.

[0068] Understandably, when goods are transported in bulk or stacked on pallets without fixed shelving, sensor nodes are arranged at intervals along the length of the truck bed, on the side walls, front wall, or inside the rear door. The installation height is adjusted according to the stacking height of the goods, typically choosing the middle or upper part of the goods (e.g., 1.2m to 1.5m from the truck bed floor). In practice, when the truck bed length is ≤4m, the number of sensor nodes is usually 3 to 4, with a spacing of approximately 1m to 1.2m between adjacent nodes (sensor nodes are arranged alternately along the central axis or sides of the truck bed). When the length of the carriage is between 1 and 7 meters, the number of sensor nodes is usually 5 to 7, with a spacing of about 1 to 1.5 meters between adjacent nodes (symmetrically arranged along the two side walls of the carriage); when the length of the carriage is ≥ 7 meters, the number of sensor nodes is usually 8 to 12, with a spacing of about 1.2 to 1.8 meters between adjacent nodes (arranged along the two side walls of the carriage and the top return air duct); in practice, when transporting goods that require fine monitoring, such as button mushrooms, additional nodes can be installed on each beam of each shelf according to the number of shelves to form a three-dimensional monitoring network. After transportation, the nodes can be removed from the mounting base.

[0069] In implementation, each sensor node is equipped with at least one carbon dioxide sensor and one temperature sensor. The temperature sensor is used to verify the carbon dioxide enrichment area. It is understood that when the sensor group is activated to perform multi-point sampling, each sensor node synchronously collects the CO2 concentration and temperature values ​​at its location. When the cargo enters the respiratory climacteric phase, respiration intensifies dramatically, releasing not only a large amount of CO2 but also generating respiratory heat, resulting in local cargo pile temperatures significantly higher than the average ambient temperature. Therefore, if a node detects an abnormally high CO2 concentration, and its temperature value is significantly higher than the node's historical baseline... If the temperature rises synchronously (e.g., the temperature rise exceeds 0.5℃ and lasts for more than 10 minutes), then the area is determined to be a real respiratory hotspot, meaning that the CO2 enrichment originates from the physiological activity of the goods (the respiratory pulsation condition is truly met at this sensor node); conversely, if there is no obvious temperature abnormality in the high CO2 area, it may be due to local gas stagnation caused by poor airflow or sensor drift (the respiratory pulsation condition is falsely met at this sensor node), which requires analysis of wind field data or triggering sensor self-check (it should be understood that the sensor group will be calibrated when it is in the preheating standby state, so at this time it is mostly due to local gas stagnation caused by poor airflow).

[0070] The gas conditioning device includes several independent control and conditioning units corresponding to the compartment sections. In implementation, each compartment corresponds to one independent control and conditioning unit. Each independent control and conditioning unit includes at least a compartment air supply branch and an electrically controlled regulating valve. The compartment air supply branch is a branch pipe led out from the main air supply pipe, and its end is located at the top or side wall of each compartment for supplying conditioned gas to that area. The electrically controlled regulating valve is installed on the compartment air supply branch and can be a proportional regulating valve or an on / off type solenoid valve for controlling the gas flow rate entering the compartment. The opening degree of the regulating valve is independently controlled by the dispatch controller.

[0071] It is understood that the air intake pipes of each independent control and regulation unit are connected in parallel to the gas generating device (including nitrogen generator, CO2 remover, and ethylene remover; this invention is mainly for goods with rapid breathing, so the main gas generating device is the CO2 remover) of the modified atmosphere refrigerated truck. The gas generating device makes coarse adjustments based on the average gas concentration requirements of the whole vehicle, while the independent control and regulation unit is responsible for distributing the adjusted gas to each zone as needed.

[0072] Please see Figure 2 As shown, it is a flowchart of the scheduling controller in an embodiment of the present invention.

[0073] Specifically, the dispatch controller determines the baseline sampling frequency of the main sensor group based on the transportation phase identifier, including:

[0074] In response to the long-distance steady-state phase marker, the baseline sampling frequency is determined to be the default baseline sampling frequency. It is understood that at this stage, the carriage is well-sealed, there is no door opening disturbance, and temperature and gas concentration changes slowly. Therefore, capturing the overall trend of environmental parameters is sufficient, and the default baseline sampling frequency (i.e., low-frequency sampling) is adequate to determine its slow changing trend. This significantly reduces the power consumption of the main sensor, and the slave sensor does not need to operate when there is no disturbance, maximizing energy savings. It should be understood that under long-distance steady-state conditions, the CO2 concentration change inside the carriage is typically less than 0.1% / hour. According to the Nyquist sampling theorem, the sampling frequency must be greater than the highest frequency of the signal. The rate of change is twice that of the standard rate. A change rate of 0.1% / hour corresponds to an extremely low frequency component. A sampling rate of 1 time / 30 minutes (0.033 times / minute) is sufficient to restore the sample without distortion and can maximize energy saving. Considering that for some abruptly changing goods, the CO2 change rate may rise to 0.3% / hour to 0.5% / hour when approaching the respiratory climax, in order to ensure timely capture in the early stage of the climax, the sampling interval should not be too long. 1 time / 5 minutes can obtain 3 sampling points within 15 minutes, which meets the statistical requirements for the change rate calculation. Therefore, in implementation, the default baseline sampling frequency is 1 time / 5 minutes to 1 time / 30 minutes, preferably set to 1 time / 12 minutes.

[0075] In response to the transportation phase being identified as a short-distance delivery phase, the baseline sampling frequency is determined to be higher than the default baseline sampling frequency. It is understood that frequent door opening and closing at this time leads to the entry of hot and humid air, causing rapid fluctuations in gas composition. It is necessary to quickly capture environmental disturbances to provide data support for possible emergency adjustments. Therefore, it is necessary to increase the sampling frequency to capture rapid changes in the gas. However, the main sensor is still responsible for monitoring at this time. In practice, the default baseline sampling frequency is 1 time / 1 minute to 1 time / 15 minutes, preferably set to 1 time / 8 minutes.

[0076] It should be understood that the transportation phase identifier only determines the baseline sampling frequency, while whether the sensor is woken up is controlled by the subsequent transition warning period and CO2 data. This ensures rapid response capability during the delivery phase and avoids energy waste caused by frequent false wake-ups of the sensor due to door opening disturbances.

[0077] Specifically, the scheduling controller determines the operating status of the sensor group based on the relationship between the current time and the respiratory pulsation warning period, including:

[0078] When the respiratory jump warning period has not yet begun, the slave sensor group remains in a dormant state and the baseline sampling frequency of the master sensor group is maintained. It should be understood that this is a non-critical period during which respiratory jumps do not occur. The slave sensor group remains in a deep dormant state (i.e., completely powered off), and the master sensor group operates according to the baseline sampling frequency determined during the transportation phase, which can save energy to the maximum extent. In practice, newly loaded jump-type goods are not in the respiratory jump period, so the slave sensors remain in a dormant state when they are first loaded, and the baseline sampling frequency corresponding to the transportation phase is maintained on this basis.

[0079] When the respiratory pulsation warning period begins, the sensor group will switch from its working state to a preheating standby state and increase the baseline sampling frequency of the main sensor group. It should be understood that at this time, the respiratory pulsation is approaching, and the scheduling controller will switch the sensor group from its working state to a preheating standby state and increase the sampling frequency of the main sensor group from the baseline value (to once every 5 minutes) to capture early signals of CO2 changes more frequently.

[0080] The preheating standby state of the sensor group includes powering on the carbon dioxide sensor and maintaining preheating, and performing at least one of zero-point calibration or baseline calibration. In practice, most CO2 sensors (especially non-dispersive infrared absorption type, NDIR) require a certain preheating time to output stable and accurate readings. Therefore, the carbon dioxide sensor is selectively powered on to allow its internal optical components (such as infrared light sources) or electrochemical components to reach stable operating temperature and voltage. During the preheating period, the CO2 sensor is in a known clean air environment (or through a built-in reference gas chamber), and the zero-point output of the sensor is automatically adjusted to eliminate measurement deviations caused by temperature drift and component aging (i.e., zero-point calibration is a mandatory calibration item during preheating). For CO2 sensors that have not been used for a long time (more than 7 days), their current output is compared with the factory baseline or the last calibration value stored in memory. If the deviation exceeds the allowable range, offset correction is automatically performed.

[0081] Specifically, the scheduling controller acquires the time-series values ​​of carbon dioxide concentration of the main sensor group within the default time window and generates the carbon dioxide change curve by connecting them in time sequence, and determines the carbon dioxide concentration change rate by the ratio of the concentration difference to the time difference between adjacent sampling times.

[0082] It should be understood that the default time window refers to the length of time used to collect carbon dioxide concentration time series values ​​and generate change curves. In the scenario of controlled atmosphere refrigerated truck transportation, the respiratory jump is a relatively slow process (lasting several hours to more than ten hours). The time window needs to take into account both trend capture and noise suppression. It is usually set to 30 minutes and updated every 5 minutes in a sliding window manner. This value can smooth the sampling noise and reflect the rising trend of CO2 concentration in a timely manner, avoiding misjudgment due to instantaneous fluctuations and avoiding delayed warnings.

[0083] In practice, the unit of carbon dioxide concentration change rate is usually % / hour, and the units need to be consistent when calculating it; carbon dioxide concentration change rate = (current CO2 concentration - CO2 concentration at the previous sampling time) / (current sampling time - previous sampling time).

[0084] Specifically, the scheduling controller selects the corresponding threshold group for jump judgment based on the transportation stage, including:

[0085] Based on the determination results of the transportation stage being identified as the long-distance steady-state stage, a low jump threshold group was selected. In practice, the concentration threshold was set at 1.2%, and the change rate threshold was set at 0.15% / hour. During the long-distance stage, the carriage is well-sealed, environmental parameters are stable, and there is no frequent door opening disturbance. A lower threshold can be used to sensitively capture early respiratory jump signals for early warning.

[0086] Based on the determination results of the transportation stage being identified as the short-distance delivery stage, a high jump threshold group was selected. In practice, the concentration threshold was set at 1.5%, and the change rate threshold was set at 0.2% / hour. Frequent door openings during the short-distance stage led to the entry of outside air, causing instantaneous fluctuations in CO2 concentration. If a low threshold was used, the jump determination might be falsely triggered. Therefore, the threshold needs to be slightly increased to filter out door opening interference and ensure that subsequent actions are triggered only when a respiratory jump actually occurs.

[0087] Among them, the threshold group for determining the climax includes the concentration threshold and the rate of change threshold. It can be understood that the climax is a sign of a sharp increase in the breathing intensity of goods, which is manifested as a rapid and continuous increase in CO2 concentration. A high concentration alone may be caused by long-term accumulation, while a high rate of change alone may be caused by short-term fluctuations. Therefore, only when both conditions are met simultaneously can it be indicated that CO2 has not only accumulated to a certain level but is also in a rapid upward trend, which is the typical manifestation of the climax.

[0088] Among them, the thresholds of the high jump determination threshold group are not less than the corresponding thresholds of the low jump determination threshold group; it should be understood that the environmental disturbances in the short-distance delivery stage are large, requiring higher thresholds to ensure the robustness of the determination and avoid CO2 fluctuations caused by opening and closing doors being misjudged as respiratory jumps.

[0089] Specifically, the scheduling controller determines whether the respiratory raphe criterion is met based on the relationship between the carbon dioxide concentration value and the rate of change of carbon dioxide concentration and the corresponding threshold group, including:

[0090] When the carbon dioxide concentration is greater than the concentration threshold of the corresponding jump threshold group and the rate of change of carbon dioxide concentration is greater than the rate of change threshold of the corresponding jump threshold group, the respiratory jump condition is determined to be met.

[0091] Conversely, it is determined that the respiratory pulsatile condition is not met.

[0092] Understandably, the dispatch controller receives real-time transport stage identifiers and CO2 data from the main sensor. During the respiratory pulsation warning period, it increases the sampling frequency of the main sensor from the baseline value and switches the secondary sensor to a preheating standby state. Then, it calculates the current CO2 concentration and its rate of change using a default time window, and compares it with the corresponding pulsation judgment threshold group selected according to the transport stage. If the respiratory pulsation condition is met, it determines that a respiratory pulsation has occurred and immediately wakes up the secondary sensor group (i.e., switches to working state) to control it to perform multi-point sampling. Then, it locates the CO2 enrichment area and directs the gas regulation device to intervene in the zone. If the respiratory pulsation condition is not met, it remains in preheating standby and returns to dormancy after the respiratory pulsation warning period has expired.

[0093] Specifically, in response to the respiratory pulsation condition, the scheduling controller switches the sensor group to the working state and performs multi-point sampling. Based on the carbon dioxide concentration values ​​of the sensor group at the same sampling time, a carbon dioxide spatial distribution matrix is ​​constructed. The set of sampling points in the carbon dioxide spatial distribution matrix with concentration values ​​greater than or equal to the enrichment threshold is determined as the carbon dioxide enrichment area to determine the spatial center location and spatial range of the carbon dioxide enrichment area.

[0094] In implementation, (1) CO2 concentration values ​​at the current location of each node in the sensor group are collected at the same sampling time, and the concentration of node i is denoted as Ci, and the total number of nodes is N; (2) According to the spatial coordinates of each node (pre-calibrated, such as x in the length direction of the carriage, y in the height direction, and z in the width direction), the concentration values ​​are mapped to a three-dimensional spatial distribution matrix M(x,y,z)=Ci; (3) All nodes are traversed, and nodes that satisfy Ci≥enrichment threshold are marked as candidate enrichment points; (4) Spatial connectivity judgment is performed on the candidate enrichment points (based on Euclidean distance threshold, (5) Consider adjacent nodes with a spacing of ≤1.5 times the spacing between nodes as connected), and merge the spatially connected candidate enrichment point set into the same carbon dioxide enrichment region; (6) Calculate the geometric center or concentration weighted center of each enrichment region (i.e. calculate the weighted average coordinate with the concentration value of each node as the weight) and record it as the spatial center position; (7) The spatial range of the enrichment region is represented by the convex hull or minimum bounding box of the set of nodes it contains (the determination of the convex hull and minimum bounding box are existing technologies and will not be elaborated here) or directly record the coordinate interval of the boundary node and record it as the spatial range;

[0095] It should be understood that the respiratory climacteric threshold is an early warning line for the overall environmental concentration, while the enrichment zone is a localized peak concentration area, and therefore should be higher than this warning line. In addition, different products have different tolerance thresholds to high CO2. Agaricus bisporus can tolerate 2.5% in the short term, while Fuji apples may turn brown if the concentration exceeds 2%. Therefore, the threshold needs to be dynamically adjusted according to the product's physiological parameter library. Therefore, in practice, the enrichment threshold ranges from 1.5% (sensitive products) to 2.5% (products with strong tolerance). In one implementation, during the transportation of Agaricus bisporus, after detecting a respiratory climacteric during the long-distance steady-state phase, the enrichment threshold was set at 1.8%, and then adjusted to 2.2% during the delivery phase.

[0096] Specifically, the scheduling controller controls the independent control and regulation unit corresponding to the spatial center of the carbon dioxide enrichment area to perform carbon dioxide control, and determines whether to control adjacent independent control and regulation units to perform carbon dioxide control in conjunction with the spatial range.

[0097] It should be understood that the spatial center location (geometric center or weighted center) of the enrichment area is mapped to the car body coordinates, and the air outlet of the zone air supply branch closest to the coordinates is selected and its corresponding independent control and adjustment unit is used as the main control unit. At this time, the main control unit should increase the extraction air force of the corresponding CO2 removal branch. In addition, the valve opening can be adjusted to introduce low CO2 gas for dilution (including nitrogen-rich air produced by the nitrogen generator).

[0098] In implementation, if the enriched area is completely contained within the coverage area of ​​the main control unit, only the main control unit will operate; if more than 25% of the enriched area is outside the coverage area of ​​the main control unit, adjacent independent control and regulation units are required to coordinate carbon dioxide control, and the intervention intensity will be reduced outward from the enriched area by the main control unit operating at full power and the adjacent units operating at reduced power.

[0099] During implementation, the control process involves continuous monitoring of CO2 concentration in each area from the sensor group. When the maximum concentration value in the enrichment area drops below the termination threshold (usually the CO2 concentration tolerance value of the transported respiratory climacteric goods, which is a known value that can be manually entered or obtained online during transport) and remains stable for more than 5 minutes, the scheduling controller stops intervening and controls the sensor group to continue analyzing the CO2 enrichment area.

[0100] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0101] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A multi-sensor network adaptive sensing system for controlled atmosphere refrigerated trucks, characterized in that, include: A recognition module is installed to obtain physiological stage prediction parameters corresponding to the current goods. The physiological stage prediction parameters include respiratory jump warning period and jump judgment threshold group. The transportation identification module generates a transportation stage identifier based on the average vehicle speed and the number of door openings monitored by the door opening and closing detection sensors within a preset time window. This includes: If the average vehicle speed is not less than the preset vehicle speed and the number of door openings is not greater than the preset number, then a long-distance steady-state stage identifier is generated. Conversely, generate short-distance delivery stage identifiers; The transportation phase includes a long-distance steady-state phase and a short-distance delivery phase. The data acquisition module includes a main sensor group and several slave sensor groups; Gas conditioning device, comprising several independent control and regulation units; A scheduling controller for determining the baseline sampling frequency of the main sensor group based on the transportation phase, including: determining the baseline sampling frequency as a default baseline sampling frequency in response to a transportation phase identifier being a long-distance steady-state phase identifier, and determining the baseline sampling frequency as higher than the default baseline sampling frequency in response to a transportation phase identifier being a short-distance delivery phase identifier; The scheduling controller determines the working state of the slave sensor group based on the current time and the respiratory jump warning period, including: maintaining the working state of the slave sensor group in a dormant state in response to the current time not being in the respiratory jump warning period, and switching the working state of the slave sensor group to a preheating standby state and increasing the baseline sampling frequency in response to the current time being in the respiratory jump warning period; The preheating standby state of the sensor group includes powering on the carbon dioxide sensor and maintaining preheating, and performing at least one of zero-point calibration or baseline calibration. The scheduling controller generates a carbon dioxide change curve based on the carbon dioxide data collected by the main sensor group and selects a corresponding threshold group for jump judgment according to the transportation stage, including: selecting a low jump judgment threshold group based on the judgment result that the transportation stage is identified as a long-distance steady-state stage, and selecting a high jump judgment threshold group based on the judgment result that the transportation stage is identified as a short-distance delivery stage. The jump determination threshold group includes a concentration threshold and a change rate threshold; Wherein, the thresholds of the high jump determination threshold group are all not less than the corresponding thresholds of the low jump determination threshold group; Furthermore, the system determines whether the respiratory jump condition is met based on the relationship between the carbon dioxide concentration value, the rate of change of carbon dioxide concentration, and the jump judgment threshold group. In response to the meeting of the respiratory jump condition, the system switches from the sensor group to the working state and performs multi-point sampling. Based on the multi-point sampling results, the system locates the carbon dioxide enrichment area and controls the corresponding independent control and regulation unit to perform zonal intervention.

2. The multi-sensor network adaptive sensing system for controlled atmosphere refrigerated trucks according to claim 1, characterized in that, The data acquisition module includes: The main sensor group, which is integrated and installed in the central ceiling or return air duct area inside the transport compartment, includes at least one carbon dioxide sensor to collect the overall environmental parameters of the compartment. Each sensor node in the carriage is equipped with a slave sensor group, and each slave sensor group includes at least one carbon dioxide sensor and one temperature sensor. The sensor nodes are fixed to the crossbeams between the shelf layers or the side wall of the carriage by detachable connectors to obtain microenvironmental parameters at different spatial locations in the carriage.

3. The multi-sensor network adaptive sensing system for controlled atmosphere refrigerated trucks according to claim 1, characterized in that, The scheduling controller acquires the time-series values ​​of carbon dioxide concentration of the main sensor group within the default time window and generates the carbon dioxide change curve by connecting them in time sequence. The carbon dioxide concentration change rate is determined by the ratio of the concentration difference to the time difference between adjacent sampling times.

4. The multi-sensor network adaptive sensing system for controlled atmosphere refrigerated trucks according to claim 1, characterized in that, The scheduling controller determines whether the respiratory jump condition is met based on the relationship between the carbon dioxide concentration value and the rate of change of carbon dioxide concentration both being greater than the corresponding threshold of the jump judgment threshold group.

5. The multi-sensor network adaptive sensing system for controlled atmosphere refrigerated trucks according to claim 1, characterized in that, The scheduling controller constructs a carbon dioxide spatial distribution matrix based on the carbon dioxide concentration values ​​of the sensor group at the same sampling time, and determines the set of sampling points in the carbon dioxide spatial distribution matrix whose concentration values ​​are greater than or equal to the enrichment threshold as carbon dioxide enrichment areas to determine the spatial center location and spatial range of the carbon dioxide enrichment areas.

6. The multi-sensor network adaptive sensing system for controlled atmosphere refrigerated trucks according to claim 5, characterized in that, The scheduling controller controls the independent control and regulation unit corresponding to the spatial center of the carbon dioxide enrichment area to perform carbon dioxide control, and determines whether to control adjacent independent control and regulation units to perform carbon dioxide control in conjunction with the spatial range.

Images