Adaptive cold chain thermal energy management device and control method

Abstract

This invention relates to the field of cold chain logistics technology, and provides an adaptive cold chain thermal energy management device and control method, comprising: a retractable insulated curtain module installed at the door opening for unfolding to close the opening during non-entry / exit periods; a dual-mode air curtain module installed above the door opening for activating when the insulated curtain is open to form a dynamic airflow barrier; and an intelligent sensing and control module, including a sensor group for real-time monitoring of the operational status and a main control unit. The main control unit receives real-time data collected by the sensor group, predicts the operational workload within a future time window based on the processed data, and generates control commands based on the prediction results to drive the retractable insulated curtain module and the dual-mode air curtain module to operate collaboratively. This invention achieves a synergy between physical isolation and airflow isolation, effectively reducing heat exchange through intelligent prediction and decision-making, lowering energy consumption, and improving operational efficiency.

Description

Technical Field

[0001] This invention relates to the field of cold chain logistics technology, and in particular to an adaptive cold chain thermal energy management device and control method. Background Technology

[0002] Cold chain transportation is a crucial link in ensuring the quality and safety of temperature-sensitive goods such as food and pharmaceuticals. During loading and unloading, container doors need to be kept open for extended periods, causing a large influx of hot outside air and a rapid loss of cold air from inside. This not only affects the quality of the goods but also significantly increases the energy consumption of the refrigeration system.

[0003] Several solutions have been proposed to address the above problems, but each has its limitations, failing to achieve high efficiency and intelligent adaptation while ensuring the continuity and efficiency of loading and unloading operations. One common method is to hang ordinary plastic or canvas curtains. While this provides some physical isolation, it requires manual opening for personnel or equipment access, which is inconvenient, and a tight seal between the curtain and the door frame is difficult to achieve, limiting the isolation effect. Another method is to install independent air curtains, forming an airflow barrier to block the exchange of air between the inside and outside. This method solves the inconvenience of manual operation, but the airflow angle and temperature of ordinary air curtains are usually fixed, making it difficult to optimize for cold chain environments. The airflow is also insufficient to effectively compress cold air at the bottom of the compartment, causing cold air to still escape from the bottom, resulting in poor isolation. Furthermore, air curtains typically require high-power continuous operation, making it impossible to coordinate with dynamically changing loading and unloading processes, leading to energy waste. Another solution is to use high-speed lifting doors, reducing door opening time through rapid opening and closing. While this method can reduce cold air loss, it is impractical in scenarios requiring frequent, continuous loading and unloading operations, and it cannot provide continuous protection while the door remains open. Alternatively, timed operation devices can be used to control the opening and closing of air curtains or doors. However, timed methods are only suitable for scenarios with fixed operational patterns and cannot perceive or adapt to the dynamic changes in the flow of goods and personnel during actual loading and unloading operations, lacking intelligence and flexibility.

[0004] In summary, existing technologies are insufficient to simultaneously meet the three core requirements of low energy consumption, high sealing performance, and intelligent adaptability. Specifically, existing solutions cannot intelligently sense and predict real-time loading and unloading operation status (such as the frequency and flow of personnel, equipment, and goods entering and leaving), thus failing to achieve precise on-demand start-up and shutdown in the "time domain" and the coordination of physical isolation and dynamic airflow isolation in the "spatial domain." Summary of the Invention

[0005] Therefore, it is necessary to provide an adaptive cold chain thermal energy management device and control method to address the problem that existing technologies cannot simultaneously meet the requirements of low energy consumption, high sealing performance, and intelligent adaptability. By combining physical isolation, aerodynamic isolation, and intelligent sensing control, precise suppression of heat exchange and synergistic optimization of energy consumption can be achieved.

[0006] This invention provides an adaptive cold chain thermal energy management device, installed at the door opening of a cold chain container, characterized in that it comprises:

[0007] A retractable heat-insulating curtain module is installed at the opening of the compartment door and is used to unfold and close the opening during non-entry / exit periods to form a physical isolation barrier.

[0008] A dual-mode air curtain module, installed above the door opening, is activated when the retractable thermal curtain module is opened to form a dynamic airflow barrier covering the opening; and

[0009] The intelligent sensing and control module includes a sensor group for real-time monitoring of the operating status of the door opening area, and a main control unit that is electrically connected to the retractable heat-insulating curtain module, the dual-mode air curtain module and the sensor group respectively.

[0010] The main control unit is configured to: receive and process real-time data collected by the sensor group, predict the workload level within a future time window based on the processed data, and generate control commands based on the prediction results; the control commands are used to drive the retractable thermal insulation curtain module and the dual-mode air curtain module to work together, so that when work activity is predicted, the retractable thermal insulation curtain module is controlled to open and the dual-mode air curtain module is activated simultaneously or subsequently; when no work activity is predicted, the dual-mode air curtain module is controlled to close and the retractable thermal insulation curtain module is subsequently controlled to close.

[0011] In one embodiment, the retractable thermal insulation curtain module includes a thermal insulation curtain roller, a retractable thermal insulation curtain, a retractable counterweight rod, and a thermal insulation curtain drive motor; the thermal insulation curtain roller houses the thermal insulation curtain drive motor; the retractable thermal insulation curtain is composed of multiple horizontally overlapping transparent material curtains; the retractable counterweight rod is located at the bottom of the compartment and is synchronously controlled by the thermal insulation curtain drive motor, automatically retracting to the bottom of the compartment when the retractable thermal insulation curtain is opened, and lifting up to lock the retractable thermal insulation curtain when the retractable thermal insulation curtain is closed.

[0012] In one embodiment, the dual-mode air curtain module includes an air curtain fan, a cooling heat exchanger, and adjustable-angle guide vanes; the guide vanes are driven by a guide vane adjustment motor; the dual-mode air curtain module has a conventional isolation mode and a high-efficiency isolation mode; in the conventional isolation mode, the angle between the guide vanes and the horizontal plane is adjusted to 15~30°, and the air curtain fan operates at a first-level wind speed; in the high-efficiency isolation mode, the angle between the guide vanes and the horizontal plane is adjusted to 60~75°, and the air curtain fan operates at a second-level wind speed.

[0013] In one embodiment, the main control unit is further configured to: when an object is detected about to pass through the door opening, first control the retractable heat-insulating curtain module to open, and then activate the dual-mode air curtain module; when no work activity is detected, first control the dual-mode air curtain module to close, and then control the retractable heat-insulating curtain module to close.

[0014] In one embodiment, the main control unit incorporates a time-series prediction neural network and a reinforcement learning model. The main control unit is configured to: construct time-series features based on real-time data collected by the sensor array; predict the workload within a future time window using the time-series prediction neural network; and input the prediction results into the reinforcement learning model. The reinforcement learning model uses energy minimization and response efficiency as optimization objectives to generate control parameters. The main control unit controls the retractable thermal insulation curtain module and the dual-mode air curtain module based on the control parameters. The reward function of the reinforcement learning model includes a negative energy consumption index, a negative response delay index, and a positive temperature stability index.

[0015] In one embodiment, the temporal characteristics include at least one of the following: the frequency of object passage collected by an infrared sensor, the temperature difference between the inside and outside of the compartment collected by a temperature sensor, the cargo flow rate collected by a vision sensor, and periodic patterns in historical operation data.

[0016] In one embodiment, the main control unit is further configured to: when the prediction result indicates that a busy work period is about to begin, control the air curtain fan of the dual-mode air curtain module to start in advance for preheating or precooling, and put the drive motor of the retractable heat-insulating curtain module into standby mode; when the prediction result indicates that an idle period is about to begin, control the dual-mode air curtain module to completely power off, and put the drive motor of the retractable heat-insulating curtain module into low-power sleep mode.

[0017] In one embodiment, the main control unit is further configured to: apply a hysteresis mechanism and a smooth transition logic when switching working modes based on prediction results; the hysteresis mechanism is to forcibly maintain the current control mode when the predicted workload is within a preset critical range; the smooth transition logic is to gradually adjust the equipment parameters according to a preset S-curve change rate when the execution state is switched.

[0018] The present invention also provides a control method based on any of the above-described adaptive cold chain thermal energy management devices, comprising the following steps:

[0019] Real-time operational status data of the door area is collected using a sensor array;

[0020] Based on the collected data, predict the workload of tasks within the future time window;

[0021] Based on the predicted workload, control instructions are generated;

[0022] According to the control command, the retractable thermal insulation curtain module and the dual-mode air curtain module are driven to work together, so that when work activity is predicted, the retractable thermal insulation curtain module is controlled to open and the dual-mode air curtain module is activated, and when no work activity is predicted, the dual-mode air curtain module is controlled to close and the retractable thermal insulation curtain module is controlled to close.

[0023] In one embodiment, the prediction includes:

[0024] The time-series features are constructed based on the collected data, and the time-series features are input into a time-series prediction neural network to obtain the prediction result of the workload within the future time window; the step of generating control instructions includes: inputting the prediction result into a reinforcement learning model, and the reinforcement learning model, with the optimization objectives of minimizing energy consumption and maximizing response efficiency, deciding to generate the control instructions.

[0025] The aforementioned adaptive cold chain thermal energy management device and control method solves the core technical problems mentioned in the background by setting up a retractable heat-insulating curtain module to provide a physical isolation barrier during non-entry and exit periods, setting up a dual-mode air curtain module to provide a dynamic airflow barrier during entry and exit periods, and setting up an intelligent sensing and control module. Its main control unit predicts the future workload based on real-time data collected by the sensor group, and then generates control commands to drive the heat-insulating curtain module and the air curtain module to work together. Specifically, this solution solves the problem of insufficient sealing of a single heat-insulating curtain or cold air leakage at the bottom of a single air curtain by coordinating physical isolation and dynamic airflow isolation in the spatial domain; through intelligent control based on real-time sensing and future work prediction, it realizes on-demand precise start-stop and mode switching in the time domain, overcoming the defects of existing timed or fixed mode control that cannot adapt to dynamic work processes, resulting in high energy consumption or poor flexibility; finally, the overall technical solution achieves adaptive intelligent management of heat exchange at the opening of the cold chain compartment door while ensuring the continuity and efficiency of loading and unloading operations, achieving a unity of low energy consumption, high isolation and intelligent adaptation. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0027] Figure 1 This is a schematic diagram of the overall structure of the adaptive cold chain thermal energy management device installed at the door of a cold chain container according to an embodiment of the present invention.

[0028] Figure 2 This is a system principle block diagram of the adaptive cold chain thermal energy management device according to an embodiment of the present invention;

[0029] Figure 3 This is a schematic diagram of the working mode of the adaptive cold chain thermal energy management device according to an embodiment of the present invention;

[0030] Figure 4 This is a flowchart of the control method for the adaptive cold chain thermal energy management device according to an embodiment of the present invention;

[0031] Figure 5 This is a detailed flowchart of the control method of the adaptive cold chain thermal energy management device according to an embodiment of the present invention;

[0032] Figure 6 This is another structural schematic diagram of the adaptive cold chain thermal energy management device according to an embodiment of the present invention.

[0033] Figure label:

[0034] 110. Retractable thermal insulation curtain module; 112. Thermal insulation curtain roller; 114. Retractable thermal insulation curtain; 116. Retractable counterweight rod; 118. Thermal insulation curtain drive motor; 120. Dual-mode air curtain module; 122. Air curtain fan; 124. Adjustable angle guide vanes; 126. Guide vane adjustment motor; 130. Sensor group; 132. Infrared beam sensor; 134. Vision sensor; 136. Temperature sensor; 140. Main control unit. Detailed Implementation

[0035] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0036] It should be noted that when a component is referred to as being "fixed to" or "set on" another component, it can be directly on the other component or there may be an intermediate component. When a component is considered to be "connected to" another component, it can be directly connected to the other component or there may be an intermediate component present. The terms "vertical," "horizontal," "upper," "lower," "left," "right," and similar expressions used in this specification are for illustrative purposes only and do not represent the only possible implementation.

[0037] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0038] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature and the second feature are in indirect contact through an intermediate medium. Furthermore, "above," "over," and "on top" the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply indicates that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply indicates that the first feature is at a lower horizontal level than the second feature.

[0039] Unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used in this specification includes any and all combinations of one or more of the associated listed items.

[0040] The following is combined Figures 1-6 The present invention describes the adaptive cold chain thermal energy management device and control method.

[0041] like Figure 1 and Figure 2 As shown, in one embodiment, an adaptive cold chain thermal energy management device is installed at the door opening of a cold chain container, aiming to solve the problem that existing technologies cannot simultaneously meet the three core requirements of low energy consumption, high sealing performance, and intelligent adaptation. The device mainly includes a retractable insulated curtain module 110, a dual-mode air curtain module 120, and an intelligent sensing and control module.

[0042] The retractable thermal curtain module 110 is installed at the door opening and is used to close the opening during non-entry / exit periods, forming a physical isolation barrier. Specifically, the retractable thermal curtain module 110 serves as the core actuator for spatial isolation, and its main structure covers the cross-section of the door opening. When the refrigerated container is stationary or in a non-operational state, the retractable thermal curtain module 110 is in the deployed state, blocking air convection between the inside and outside of the container like a conventional curtain. However, unlike ordinary soft curtains, this module is controlled by a control system and can automatically retract or lower according to instructions. It should be understood that although this embodiment describes its installation at the door opening, its specific installation position can be adjusted according to the container structure, such as being embedded inside the door frame or attached to the outside of the door frame, as long as it can effectively close the opening.

[0043] The dual-mode air curtain module 120 is installed above the door opening and is activated when the retractable insulated curtain module 110 is opened to form a dynamic airflow barrier covering the opening. Specifically, when personnel or vehicles need to enter or exit, the retractable insulated curtain module 110 is retracted, and the door opening is directly exposed to the external environment. At this time, the dual-mode air curtain module 120 is activated, spraying high-speed airflow downwards. This airflow forms an invisible air door, using aerodynamic principles to isolate the cold air inside the compartment from the hot air outside. This dynamic airflow barrier complements the physical isolation barrier, allowing personnel and equipment to pass through while maintaining a temperature gradient.

[0044] The intelligent sensing and control module includes a sensor group 130 for real-time monitoring of the operating status of the door opening area, and a main control unit 140 electrically connected to the retractable heat-insulating curtain module 110, the dual-mode air curtain module 120, and the sensor group 130. (Refer to...) Figure 2 The sensor group 130 acts as the system's senses, responsible for collecting environmental data; the main control unit 140 acts as the system's brain, responsible for data processing and decision-making. The main control unit 140 is configured to: receive and process the real-time data collected by the sensor group 130, predict the workload within a future time window based on the processed data, and generate control commands based on the prediction results.

[0045] Control commands are used to drive the retractable thermal curtain module 110 and the dual-mode air curtain module 120 to work together. When work activity is predicted, the retractable thermal curtain module 110 is opened, and the dual-mode air curtain module 120 is activated simultaneously or subsequently. When no work activity is predicted, the dual-mode air curtain module 120 is closed, and the retractable thermal curtain module 110 is closed subsequently. Specifically, this coordinated action embodies the dual optimization of the invention in both the spatial and temporal domains. In the spatial domain, the retractable thermal curtain module 110 forms a physical isolation barrier, and the dual-mode air curtain module 120 forms a dynamic airflow barrier; the two complement each other and work together to achieve a highly airtight isolation effect. In the temporal domain, the main control unit 140 uses a predictive mechanism to achieve on-demand start and stop, activating the equipment only when needed, avoiding the energy waste caused by the long-term continuous operation of traditional air curtain machines. For example, when the main control unit 140 predicts based on sensor data that a forklift is about to enter, it will issue an instruction in advance or immediately to retract the insulation curtain and open the air curtain to ensure operational continuity; when it predicts that the operation is over or that there will be a long period of idle time, it will promptly close the air curtain and lower the insulation curtain to reduce ineffective energy consumption. Through this organic combination of physical isolation in the spatial domain and dynamic airflow isolation in the temporal domain, the minimization of cold energy loss and the maximization of operational efficiency are achieved.

[0046] In one embodiment, the specific mechanical structure and sensor layout of the adaptive cold chain thermal energy management device are described in detail.

[0047] like Figure 1 and Figure 3 As shown, where, Figure 3 (a) is the heat preservation mode. Figure 3 (b) is the standard isolation mode. Figure 3(c) High-efficiency isolation mode. The retractable thermal insulation curtain module 110 includes a thermal insulation curtain roller 112, a retractable thermal insulation curtain 114, a retractable counterweight rod 116, and a thermal insulation curtain drive motor 118. The thermal insulation curtain roller 112 has a built-in thermal insulation curtain drive motor 118. This built-in design effectively reduces the complexity of the external linkage mechanism, improves transmission efficiency, and saves installation space. The retractable thermal insulation curtain 114 is composed of multiple horizontally overlapping transparent material curtains. Specifically, the transparent material is preferably high-strength PVC or polycarbonate (PC) soft strips. The multi-layered horizontally overlapping structure forms tiny air layers and a labyrinthine sealing path between adjacent curtains, significantly improving thermal insulation performance and physical sealing effect. When the curtain is unfolded, this overlapping structure can effectively block the convective heat exchange between the air inside and outside the compartment. A retractable counterweight bar 116 is located at the bottom of the cargo box and is synchronously controlled by the insulated curtain drive motor 118. It automatically retracts into the bottom of the cargo box when the retractable insulated curtain 114 is opened, and lifts up to lock the curtain when it is closed. This design solves the problem of traditional door curtain bottom counterweights easily tripping workers or being damaged by forklifts. When the insulated curtain is open, the counterweight bar automatically retracts into the groove at the bottom of the cargo box, keeping the ground flat and ensuring unobstructed passage for forklifts and other handling equipment. When the insulated curtain is closed, the counterweight bar lifts up and presses down on the bottom of the curtain, forming a bottom seal to prevent cold air from escaping through the bottom gaps.

[0048] Furthermore, flexible heating wires are embedded within the transparent curtain. In the low-temperature environment of the cold chain, the surface of the curtain in the door area is prone to frost or fogging due to temperature differences, obstructing visibility and posing a safety hazard. The flexible heating wires are evenly distributed inside the curtain, generating a small amount of heat when energized, which effectively eliminates frost and fog on the curtain surface, ensuring that workers can clearly observe the situation inside and outside the compartment through the insulated curtain, thus improving the safety and efficiency of the operation.

[0049] The dual-mode air curtain module 120 includes an air curtain fan 122, a refrigeration heat exchanger (not shown), and adjustable-angle guide vanes 124; the guide vanes 124 are driven by a guide vane adjustment motor 126. The air curtain fan 122 generates a high-speed airflow, which is cooled by the refrigeration heat exchanger and then sprayed downwards through the air outlet. The guide vanes 124 are located at the air outlet, and their angle directly determines the range and intensity of the airflow coverage. The dual-mode air curtain module 120 has a conventional isolation mode and a high-efficiency isolation mode. In the conventional isolation mode, the angle between the guide vanes 124 and the horizontal plane is adjusted to 15~30°, and the air curtain fan 122 operates at a first-level wind speed. In this mode, the airflow is relatively gentle, mainly covering the upper and middle areas of the opening, suitable for scenarios with small temperature differences between the inside and outside of the chamber or low operating frequency, reducing fan energy consumption while ensuring basic isolation effect. In the high-efficiency isolation mode, the angle between the guide vanes 124 and the horizontal plane is adjusted to 60-75°, and the air curtain fan 122 operates at a secondary wind speed. In this mode, the airflow impacts downwards at a large angle and high speed, which can more effectively penetrate the hot air layer at the opening and reach the bottom of the carriage, forming a strong air wall. This large-angle, high-speed airflow pattern is particularly suitable for busy periods with large temperature differences between the inside and outside of the carriage, strong buoyancy of external hot air, or frequent entry and exit, effectively suppressing the intrusion of external hot air and preventing the leakage of internal cool air.

[0050] Sensor group 130 includes at least one of an infrared beam sensor 132, a vision sensor 134, and a temperature sensor 136. Specifically, the infrared beam sensor 132 is mounted on the columns or rails on both sides of the door. By emitting and receiving infrared beams, it can accurately detect whether objects (such as personnel, forklifts, or goods) pass through the door frame plane and preliminarily determine the object's height and speed. The vision sensor 134 is located in the air curtain assembly area above the door, and its monitoring field covers the entire door opening area. It is used to assist in identifying object types (such as distinguishing between personnel and forklifts) and monitoring the operational status of the door area. The temperature sensor 136 is located on the inner wall of the compartment near the door. It is used to monitor the temperature changes in the door area and the temperature difference between the inside and outside of the compartment in real time, providing environmental parameters for the control system's mode decision-making. Through the combination of the above-mentioned multiple sensors, the system can comprehensively perceive the operational status and environmental parameters of the door area, providing reliable data support for precise control.

[0051] In one embodiment, the control timing and strategy optimization of the adaptive cold chain thermal energy management device are described in detail. (Refer to...) Figure 5 The main control unit 140 solves the problems of slow response, high energy consumption and rapid equipment wear in traditional cold chain access control systems through refined timing control and strategy scheduling.

[0052] Firstly, regarding the timing of actions during the switching of operating states, the main control unit 140 is further configured as follows: when an object is detected about to pass through the door opening, the retractable insulated curtain module 110 is opened first, followed by the activation of the dual-mode air curtain module 120; when no operational activity is detected, the dual-mode air curtain module 120 is closed first, followed by the closure of the retractable insulated curtain module 110. Specifically, this specific timing logic is designed based on an in-depth analysis of the cold air escape mechanism. When an object is about to pass through, if the air curtain is activated first and then the curtain is opened, the high-speed airflow of the air curtain will directly impact the closed insulated curtain, causing airflow turbulence and even drawing in external hot air into the compartment; however, by adopting the sequence of opening the curtain first and then activating the air curtain, the insulated curtain is quickly retracted, eliminating physical obstruction, and the airflow barrier formed by the subsequent activation of the air curtain can smoothly cover the entire opening, avoiding the instantaneous hot and cold convection when the curtain is opened. Conversely, if the curtain is closed before the air curtain is closed when the operation is finished, the insulation curtain will compress the cold air inside the compartment during the closing process, and the residual airflow after the air curtain is closed may be locked outside the curtain or cause the curtain to sway. By adopting the order of closing the air curtain first and then closing the curtain, the airflow inside and outside the compartment tends to stabilize after the air curtain stops, and the insulation curtain then slowly falls down, which can smoothly close the opening like a piston, effectively preventing the cold air from overflowing due to the air pressure fluctuation generated at the moment the curtain is closed.

[0053] Furthermore, to improve system response speed and reduce energy consumption during non-operational periods, the main control unit 140 is further configured to employ a pre-response mechanism. When the prediction indicates an upcoming busy operating period, the air curtain fan 122 of the dual-mode air curtain module 120 is started in advance for preheating or precooling, and the drive motor of the retractable insulation curtain module 110 is put into standby mode. When the prediction indicates an upcoming idle period, the dual-mode air curtain module 120 is completely powered off, and the drive motor of the retractable insulation curtain module 110 enters a low-power sleep mode. Specifically, before the busy period arrives (e.g., based on historical data predicting the imminent arrival of a forklift), the main control unit 140 controls the air curtain fan 122 to operate at low speed in advance, ensuring that the air curtain outlet temperature is close to the set temperature inside the compartment, eliminating the thermal inertia delay during equipment startup, and ensuring an efficient isolation barrier is provided the instant the insulation curtain opens. Simultaneously, the drive motor is powered on in advance and put into standby mode, eliminating the delay of the mechanical brake and achieving a millisecond-level response speed. During off-peak hours (such as at night or during work breaks), the system not only shuts off the air curtain but also cuts off the fan power and puts the motor into sleep mode. Compared to traditional standby mode, this deep energy-saving strategy can significantly reduce ineffective energy consumption. It should be understood that preheating or precooling here refers not only to temperature regulation but also to preparatory actions such as pre-ventilation of the air ducts.

[0054] Furthermore, to avoid frequent mode switching due to environmental disturbances or sensor data fluctuations, the main control unit 140 is further configured to apply a hysteresis mechanism and smooth transition logic when switching operating modes based on prediction results. The hysteresis mechanism forces the current control mode to be maintained when the predicted workload is within a preset critical range. The smooth transition logic gradually adjusts equipment parameters according to a preset S-curve change rate when performing state switching. Specifically, the hysteresis mechanism is implemented by setting a preset critical range (e.g., the predicted workload is between 40% and 60%). When the predicted value falls within this range, regardless of its slight fluctuations, the system forces the maintenance of the control mode from the previous moment, thereby avoiding frequent equipment start-ups and shutdowns (oscillations) due to sensor noise or brief interference. The smooth transition logic is applied when state switching must be performed. For example, when adjusting the wind curtain speed or the 124-degree angle of the guide vanes, a direct step adjustment would cause mechanical shock to the fan motor and the blade adjustment motor, and sudden airflow changes would affect the isolation effect. This embodiment employs an S-curve rate of change control parameter adjustment, meaning that control parameters (such as wind speed and angle) gradually change over time according to an S-curve trajectory, rather than undergoing abrupt nonlinear changes. This control method exhibits a relatively gentle rate of change at the beginning and end stages, accelerating in the intermediate stage. This ensures rapid adjustment while significantly reducing mechanical wear and system inrush current, thus extending the device's service life.

[0055] In one embodiment, a control method based on the above-described adaptive cold chain thermal energy management device is also provided. For example... Figure 4 As shown, the control method specifically includes the following steps:

[0056] Step S410: Real-time data collection of operational status in the door area using sensor array.

[0057] Step S420: Based on the collected data, predict the workload level within the future time window. Specifically, this prediction step includes: constructing time-series features based on the collected data, and inputting the time-series features into a time-series prediction neural network to obtain the predicted workload level within the future time window.

[0058] Step S430: Based on the predicted workload, control commands are generated. Specifically, this step of generating control commands includes: inputting the prediction results into a reinforcement learning model, which then generates control commands with the optimization objectives of minimizing energy consumption and maximizing response efficiency. This process achieves closed-loop control from data perception, feature extraction, trend prediction to intelligent decision-making, enabling the device to predict workload and adjust equipment status in advance with the same accuracy as an experienced operator, thereby achieving optimal energy management while ensuring cold chain quality.

[0059] Step S440: According to the control command, drive the retractable thermal insulation curtain module and the dual-mode air curtain module to work together, so that when it is predicted that there is work activity, control the retractable thermal insulation curtain module to open and start the dual-mode air curtain module, and when it is predicted that there is no work activity, control the dual-mode air curtain module to close and control the retractable thermal insulation curtain module to close.

[0060] More specifically, in combination Figure 5 The detailed process, specifically the execution of steps S410 to S440 above, is as follows:

[0061] In step S410, the system is powered on and initialized, and is in the heat preservation mode by default. The sensor group begins to collect the operation status data of the door area in real time.

[0062] In step S420, the main control unit predicts the workload level within a future time window based on a time-series prediction neural network. Specifically, the main control unit constructs time-series features (such as object passage frequency, temperature difference between inside and outside the compartment, and cargo flow) based on real-time data collected by the sensor group, and inputs these time-series features into the time-series prediction neural network to obtain the prediction result of the workload level within the future time window.

[0063] In step S430, the main control unit determines whether the prediction result indicates that a busy period is about to begin. If yes, proceed to step S431; otherwise, proceed to step S440.

[0064] Step S431, perform pre-response operation: control the air curtain fan to start preheating or precooling in advance, and drive the motor to enter standby state.

[0065] Step S432: When an object is detected to be about to pass through, execute the action timing control: first control the retractable heat preservation curtain module to open, and then start the dual-mode air curtain module.

[0066] Step S433: Determine the workload based on real-time monitoring data and select either the normal isolation mode or the strong isolation mode for operation.

[0067] Step S440: Determine if the prediction result indicates an upcoming idle period. If yes, proceed to step S441; otherwise, return to step S410 to continue monitoring.

[0068] Step S441: Apply hysteresis mechanism and smooth transition logic: Determine whether the workload has deviated from the preset critical range. If so, gradually reduce the fan speed and adjust the blade angle according to the S-curve change rate.

[0069] Step S442, execute action timing control: first control the dual-mode air curtain module to close, then control the retractable heat preservation curtain module to close.

[0070] Step S443: Power off the air curtain module and drive the motor into a low-power sleep mode.

[0071] Through the above control process, this embodiment achieves a leap from passive response to active prediction, maximizing the balance between operational efficiency and energy consumption while ensuring the stability of the cold chain environment.

[0072] In one embodiment, the intelligent decision-making core of the adaptive cold chain thermal energy management device is described in detail. (Refer to...) Figure 2 and Figure 5 The main control unit 140 incorporates a time-series prediction neural network and a reinforcement learning model. As the core processing hardware of the system, the main control unit 140 deploys a deep time-series prediction neural network based on the Transformer architecture and a reinforcement learning model based on the DQN (Deep-Q Network) algorithm. It should be understood that "integrated" here includes both firmware programs directly burned into the microcontroller or embedded chip, and software modules deployed on edge computing gateways or cloud servers and connected to the main control unit 140 via communication interfaces. The main control unit 140 is configured to: construct time-series features based on real-time data collected by the sensor group 130; predict the workload level within future time windows using the time-series prediction neural network; and input the prediction results into the reinforcement learning model. The reinforcement learning model, with the optimization objectives of minimizing energy consumption and maximizing response efficiency, generates control parameters. The main control unit 140 controls the retractable thermal curtain module 110 and the dual-mode air curtain module 120 according to the control parameters.

[0073] Specifically, the data flow and decision-making process of the entire algorithm architecture is as follows: First, the main control unit 140 receives raw data streams from the infrared beam sensor 132, the vision sensor 134, and the temperature sensor 136. This raw data has temporal continuity, and the main control unit 140 preprocesses and constructs features from it. Temporal features include at least one of the following: the frequency of object passage collected by the infrared sensor, the temperature difference between the inside and outside of the compartment collected by the temperature sensor 136, the cargo flow rate collected by the vision sensor 134, and periodic patterns in historical operation data. For example, the main control unit 140 counts the number of times the infrared sensor triggers within the sliding time window as the object passage frequency, which directly reflects the operational density of the current period; it calculates the temperature difference between the readings of the internal temperature sensor 136 and the external ambient temperature sensor 136 as the temperature difference between the inside and outside of the compartment, which characterizes the driving force of heat exchange; it identifies the movement trajectory of the cargo outline in the image through the vision sensor 134 and counts the cargo flow rate per unit time; simultaneously, the main control unit 140 calls up the stored historical operation data to extract periodic patterns, such as the loading and unloading peaks at specific times of the day or the operational troughs on specific days of the week. These multi-dimensional temporal features are constructed into feature vectors and used as input to the temporal prediction neural network.

[0074] The temporal prediction neural network utilizes the self-attention mechanism of the Transformer architecture to capture the dependencies of the aforementioned temporal features over long time series, thereby making rolling predictions of job workload within the next time window (such as the next 10 or 30 minutes). The prediction results are quantified as workload scores or job probabilities and passed as state input to the reinforcement learning model. The reward function of the reinforcement learning model includes negative energy consumption indicators, negative response latency indicators, and positive temperature stability indicators. Specifically, the reinforcement learning model learns the optimal strategy through trial and error with the environment. After each decision action (such as adjusting the wind speed of the air curtain, changing the angle of the guide vanes, or starting and stopping the insulation curtain), the environment provides a reward value. The reward value is composed of three weighted components: the first is an energy consumption index, where the system monitors the power consumption of equipment such as the air curtain fan 122 and drive motor in real time. The higher the energy consumption, the lower the reward value (i.e., negative penalty), guiding the model to reduce power while meeting isolation requirements. The second is a response delay index, where the system records the time difference from sensor triggering to equipment action completion. The longer the delay, the lower the reward value, forcing the model to optimize control timing and improve operational efficiency. The third is a temperature stability index, where the system calculates the variance of temperature fluctuation in the door area. The smaller the temperature fluctuation, the higher the reward value (i.e., positive reward), encouraging the model to maintain a stable cold chain environment. Through this multi-objective optimization reward mechanism, the reinforcement learning model can converge to the optimal control strategy that balances energy saving and efficiency, generating specific control parameters (such as the specific value of the air curtain speed, the precise angle of the guide vanes, and the start and stop times of the insulation curtain).

[0075] In a specific embodiment, taking a particular cold chain logistics loading and unloading scenario as an example, the dynamic switching process of the adaptive cold chain thermal energy management device is described in detail. The scenario is set as a fresh food cold chain distribution center, with operating hours from 8:00 to 10:00 AM during peak hours and subsequent off-peak hours.

[0076] Before the peak period arrives, the time-series prediction neural network inside the main control unit 140 predicts the upcoming busy operating period based on the periodic patterns in historical operation data. Specifically, the main control unit 140 analyzes the operation records of the same period in the past week and finds that the frequency of forklift entry and exit increases exponentially between 8:00 and 10:00, and the temperature difference between inside and outside the compartment is usually maintained above 25°C (-18°C inside, 7°C outside). Based on this prediction result, the main control unit 140 executes a pre-response mechanism: controlling the air curtain fan 122 of the dual-mode air curtain module 120 to start in advance for pre-cooling, and putting the drive motor of the retractable heat-insulating curtain module 110 into standby mode. For example, at 7:55, the main control unit 140 controls the air curtain fan 122 to run at low speed, pre-cooling the air curtain outlet temperature to -15°C, close to the set temperature inside the compartment; at the same time, the heat-insulating curtain drive motor 118 is powered on, the brake is released, and it is in a high-sensitivity standby state. This pre-response mechanism eliminates the thermal inertia delay and mechanical delay of equipment startup, ensuring that when the first forklift arrives on time at 8:00, the system can complete the curtain opening and air curtain switching within milliseconds, avoiding the peak of cold air escape at the moment the door is opened.

[0077] As operations commence, sensor array 130 monitors a surge in the frequency of object passage in real time, and vision sensor 134 identifies a large volume of goods entering and exiting. Based on decisions made by the reinforcement learning model, main control unit 140 controls the dual-mode air curtain module 120 to enter a high-efficiency isolation mode. In this mode, the angle between the guide vanes 124 and the horizontal plane is adjusted to 60-75°, and the air curtain fan 122 operates at a secondary wind speed. Due to the significant temperature difference between the inside and outside of the compartment, and the strong buoyancy of external hot air, it easily intrudes from the bottom of the opening. The high-angle, high-velocity airflow effectively penetrates the hot air layer at the opening, reaching the bottom of the compartment and forming a powerful "air wall" that blocks external hot air from entering. Experimental data shows that in high-efficiency isolation mode, the average cold air loss within 30 minutes of the compartment door opening is reduced by approximately 35% compared to a traditional fixed-angle air curtain, and the temperature fluctuation inside the compartment is controlled within ±1.5℃, effectively ensuring the quality of fresh goods.

[0078] As the operation continued until around 10:00 AM, data collected by sensor group 130 showed a gradual decrease in the frequency of object passage, and the time-series prediction neural network predicted a reduction in the workload within the future time window. At this point, the main control unit 140 applied a hysteresis mechanism and smooth transition logic to switch modes. Specifically, when the predicted workload decreased from 80% to around 50% (within the preset critical range of 40%-60%), the hysteresis mechanism forced the system to maintain the current strong isolation mode, preventing frequent mode switching due to occasional brief absences of forklifts. Only when the predicted value remained below 40% for more than 5 minutes did the system determine that it had truly entered the idle transition period. At this time, the main control unit 140 gradually adjusted the equipment parameters according to the preset S-curve change rate: first, it smoothly reduced the air curtain speed from level two to level one, and then gradually reduced the guide vane angle from 60° to 30°. This smooth transition avoided current surges and sudden changes in pipeline pressure caused by a sudden drop in fan speed, extending the fan's lifespan.

[0079] When the prediction indicates that an idle period is about to begin, the main control unit 140 completely de-energizes the dual-mode air curtain module 120 and puts the drive motor of the retractable thermal curtain module 110 into a low-power sleep mode. After confirming that there are no activity signals in the doorway area, the system executes the timing logic of "closing the air curtain first and then closing the curtain," controlling the dual-mode air curtain module 120 to close, and then controlling the retractable thermal curtain module 110 to close. At this time, the device enters a deep energy-saving state. Compared with the traditional air curtain machine's continuous 24 / 7 operation, the device in this embodiment reduces standby power consumption by more than 95% during idle periods. Through the dynamic switching between busy and idle modes, this embodiment maximizes the reduction of cold energy loss and equipment energy consumption while ensuring the efficiency of cold chain operations, demonstrating significant commercial application value.

[0080] In one embodiment, an alternative implementation of an adaptive cold chain thermal energy management device is provided, the main difference from the aforementioned embodiment being the specific structural form of the retractable thermal insulation curtain module 110. (Refer to...) Figure 6 The retractable heat-insulating curtain 114 is composed of multiple longitudinally arranged soft strips, rather than the transversely overlapping structure described in the previous embodiments.

[0081] Specifically, multiple longitudinal flexible strips are arranged sequentially along the width of the door opening, with each strip independently suspended from a guide rail or roller mechanism above the door frame. The strips are preferably made of a flexible material with magnetic adsorption properties, or have magnetic strips embedded in their sides. When the retractable thermal curtain 114 is closed, adjacent longitudinal flexible strips adhere to each other through magnetic attraction, forming a continuous physical barrier at the door opening. This magnetic adsorption principle allows adjacent strips to close tightly, effectively blocking air convection between the inside and outside of the compartment. Simultaneously, they can quickly separate when subjected to external forces (such as personnel or forklifts), and automatically reset and close after passage, exhibiting excellent self-sealing and passageability.

[0082] Compared to the aforementioned horizontally overlapping transparent material curtain, the vertical flexible strip structure has unique advantages in dealing with frequent entry and exit scenarios. The horizontally overlapping structure is rolled up and down, with a counterweight rod 116 retractable at the bottom, providing good sealing and suitability for overall opening; while the vertical flexible strip structure allows for partial opening, meaning that when personnel or equipment pass through, only the corresponding width of the flexible strip needs to be pushed open, while the rest remains closed. This can further reduce the area of ​​cold air loss in certain small-batch, high-frequency operation scenarios. However, the overall sealing performance of the vertical flexible strip structure may be slightly inferior to that of the bottom-locked horizontal curtain, and it requires a higher degree of flatness of the bottom ground. Therefore, those skilled in the art can flexibly choose between the horizontally overlapping structure or the vertical flexible strip structure according to the actual cold chain operation scenario (such as passage frequency, cargo size, and sealing level requirements). Both are within the protection scope of this invention and can work in conjunction with the dual-mode air curtain module 120 and the intelligent sensing and control module to achieve effective management of heat and cold exchange.

[0083] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0084] The above-described embodiments are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. An adaptive cold chain thermal energy management device installed at a door opening of a cold chain container, characterized in that, include: A retractable heat-insulating curtain module is installed at the opening of the compartment door and is used to unfold and close the opening during non-entry / exit periods to form a physical isolation barrier. A dual-mode air curtain module, installed above the door opening, is activated when the retractable thermal curtain module is opened to form a dynamic airflow barrier covering the opening; and The intelligent sensing and control module includes a sensor group for real-time monitoring of the operating status of the door opening area, and a main control unit that is electrically connected to the retractable heat-insulating curtain module, the dual-mode air curtain module and the sensor group respectively. The main control unit is configured to: receive and process real-time data collected by the sensor group, predict the workload level within a future time window based on the processed data, and generate control commands based on the prediction results; the control commands are used to drive the retractable heat-insulating curtain module and the dual-mode air curtain module to work together, so that when work activity is predicted, the retractable heat-insulating curtain module is controlled to open and the dual-mode air curtain module is activated simultaneously or subsequently; when no work activity is predicted, the dual-mode air curtain module is controlled to close and the retractable heat-insulating curtain module is subsequently controlled to close. The main control unit is further configured to: when the prediction result indicates that a busy work period is about to begin, control the air curtain fan of the dual-mode air curtain module to start in advance for preheating or precooling, and put the drive motor of the retractable heat-insulating curtain module into standby mode; when the prediction result indicates that an idle period is about to begin, control the dual-mode air curtain module to completely power off, and put the drive motor of the retractable heat-insulating curtain module into low-power sleep mode.

2. The adaptive cold chain thermal energy management device according to claim 1, characterized in that, The retractable thermal insulation curtain module includes a thermal insulation curtain roller, a retractable thermal insulation curtain, a retractable counterweight rod, and a thermal insulation curtain drive motor; the thermal insulation curtain roller houses the thermal insulation curtain drive motor; the retractable thermal insulation curtain is composed of multiple horizontally overlapping transparent material curtains; the retractable counterweight rod is located at the bottom of the compartment and is synchronously controlled by the thermal insulation curtain drive motor, automatically retracting to the bottom of the compartment when the retractable thermal insulation curtain is opened, and lifting up to lock the retractable thermal insulation curtain when it is closed.

3. The adaptive cold chain thermal energy management device according to claim 1, characterized in that, The dual-mode air curtain module includes an air curtain fan, a cooling heat exchanger, and adjustable-angle guide vanes; the guide vanes are driven by a guide vane adjustment motor; the dual-mode air curtain module has a conventional isolation mode and a high-efficiency isolation mode; In the conventional isolation mode, the angle between the guide vanes and the horizontal plane is adjusted to 15~30°, and the air curtain fan operates at a first-level wind speed; in the high-efficiency isolation mode, the angle between the guide vanes and the horizontal plane is adjusted to 60~75°, and the air curtain fan operates at a second-level wind speed.

4. The adaptive cold chain thermal energy management device according to claim 1, characterized in that, The main control unit is further configured to: when an object is detected about to pass through the door opening, first control the retractable heat-insulating curtain module to open, and then activate the dual-mode air curtain module; when no work activity is detected, first control the dual-mode air curtain module to close, and then control the retractable heat-insulating curtain module to close.

5. The adaptive cold chain thermal energy management device according to claim 1, characterized in that, The main control unit incorporates a time-series prediction neural network and a reinforcement learning model. The main control unit is configured to: construct time-series features based on real-time data collected by the sensor array; predict the workload within a future time window using the time-series prediction neural network; and input the prediction results into the reinforcement learning model. The reinforcement learning model uses energy minimization and response efficiency as optimization objectives to generate control parameters. The main control unit controls the retractable heat-insulating curtain module and the dual-mode air curtain module based on these control parameters. The reward function of the reinforcement learning model includes negative energy consumption indicators, negative response delay indicators, and positive temperature stability indicators.

6. The adaptive cold chain thermal energy management device according to claim 5, characterized in that, The time-series characteristics include at least one of the following: the frequency of object passage collected by infrared sensors, the temperature difference between the inside and outside of the compartment collected by temperature sensors, the cargo flow rate collected by vision sensors, and periodic patterns in historical operation data.

7. The adaptive cold chain thermal energy management device according to claim 5, characterized in that, The main control unit is further configured to: apply a hysteresis mechanism and a smooth transition logic when switching working modes based on prediction results; the hysteresis mechanism is to forcibly maintain the current control mode when the predicted workload is within a preset critical range; the smooth transition logic is to gradually adjust the equipment parameters according to a preset S-curve change rate when switching execution states.

8. A control method based on the adaptive cold chain thermal energy management device according to any one of claims 1 to 7, characterized in that, Includes the following steps: Real-time operational status data of the door area is collected using a sensor array; Based on the collected data, predict the workload of tasks within the future time window; Based on the predicted workload, control instructions are generated; According to the control command, the retractable thermal insulation curtain module and the dual-mode air curtain module are driven to work together, so that when work activity is predicted, the retractable thermal insulation curtain module is controlled to open and the dual-mode air curtain module is activated, and when no work activity is predicted, the dual-mode air curtain module is controlled to close and the retractable thermal insulation curtain module is controlled to close.

9. The control method of the adaptive cold chain thermal energy management device according to claim 8, characterized in that, The predictions include: The time-series features are constructed based on the collected data, and the time-series features are input into a time-series prediction neural network to obtain the prediction result of the workload within the future time window; the step of generating control instructions includes: inputting the prediction result into a reinforcement learning model, and the reinforcement learning model, with the optimization objectives of minimizing energy consumption and maximizing response efficiency, deciding to generate the control instructions.

Images