Temperature control system of unmanned aerial vehicle airborne refrigeration box and refrigeration box

Abstract

This invention relates to the field of thermal management of refrigerated loads, and discloses a temperature control system and a refrigerated container for an airborne refrigerated container used by an unmanned aerial vehicle (UAV). The refrigerated container includes a UAV, an airborne refrigerated container, a microcontroller, internal and ambient temperature sensors, and a semiconductor cooling chip. The system runs on the microcontroller and includes: a data acquisition module, which acquires physical and boundary data and performs dead zone determination to generate an effective airspeed; a state observation module, which calculates heat leakage power based on the effective airspeed and verifies and generates an effective cold storage scalar; a load prediction module, which calculates the expected duration based on the flight path matrix and generates an evaluation result; and a control execution module, which performs mode arbitration and solves the command power, and verifies and generates a safe power to drive the cooling chip based on the instantaneous bus voltage. This invention overcomes temperature control lag by forward-looking cold storage and extrapolating the remaining cold storage capacity during phase change, while ensuring the stability of the bus power supply and flight safety.

Description

Technical Field

[0001] This invention relates to the field of thermal management of refrigerated loads, specifically to a temperature control system and a refrigerated container for an unmanned aerial vehicle (UAV) onboard refrigerated container. Background Technology

[0002] With the widespread application of drone technology, using drones to perform cold chain delivery of temperature-sensitive materials such as vaccines and biological products has become an industry trend. Existing drone-borne refrigerated boxes typically use semiconductor refrigeration components combined with phase change cold storage materials to maintain the temperature stability of the internal cavity.

[0003] However, existing airborne refrigerated container temperature control systems mainly rely on internal temperature sensors for single closed-loop feedback control. The adjustment mechanism has a lag, and the control system cannot sense the dynamic convection heat transfer disturbance caused by the instantaneous change in external airspeed under different flight conditions such as forward flight or hovering of the UAV. It cannot incorporate aerodynamic and thermodynamic boundary conditions into the control model, and it is difficult to perform cold storage in advance for severe heat dissipation conditions such as low-speed hovering that may occur in the future during the flight path.

[0004] Meanwhile, because the phase change material in the phase change storage box maintains a constant temperature within the latent heat release range of the solid-liquid phase change, the system that relies on temperature feedback cannot accurately obtain the actual physical residual cooling capacity inside the phase change material. This results in a blind spot in the system during the phase change transition period. Usually, the high-power cooling action is only triggered after the phase change material has completely melted and the internal temperature exceeds the critical value, thus missing the opportunity for stable regulation.

[0005] Furthermore, when faced with a surge in external load causing internal temperature rise, traditional control logic would instruct the refrigeration components to operate continuously at full load. Since the airborne refrigerated container is powered by the drone's main power network, the high-power refrigeration of the payload overlaps with the high-power output of the rotor required for the drone's large-scale flight maneuvers. This lack of coordinated power usage can cause voltage drops on the drone's power supply bus, leading to conflicts between the payload's power consumption and the flight control system's power supply, thus affecting the drone's flight safety. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a temperature control system and a refrigerator for an airborne refrigerated container for unmanned aerial vehicles (UAVs), solving the problems of lagging temperature control, inability to quantify and observe phase change cooling capacity, and the threat to the power supply safety of the flight platform posed by payload power consumption in existing technologies.

[0007] To achieve the above objectives, the present invention provides the following technical solution: The first aspect of this invention provides a temperature control system for an airborne refrigerated container for an unmanned aerial vehicle (UAV), comprising: The data acquisition module is used to acquire local physical data and flight boundary condition data in each control cycle, perform low-speed dead zone determination on the instantaneous airspeed and current flight condition identifier in the flight boundary condition data, and generate effective airspeed. The state observation module calculates the heat leakage power based on the effective airspeed and the local physical data, calls the cold storage state formula and performs boundary physical verification to generate the effective cold storage scalar. The load prediction module calculates the expected duration based on the route and waypoint matrix in the flight boundary condition data, converts the effective cold storage scalar into a cold storage maintenance time margin and compares it with the expected duration to generate a cold storage assessment result. The control execution module is used to combine the cold storage assessment results, the effective airspeed, the current flight condition identifier, and the local physical data to execute the working mode arbitration generation mode gate coefficient, call the power output formula to calculate the command power, check the command power according to the instantaneous bus voltage in the local physical data, generate safe power, and convert the safe power into a pulse width modulation signal to drive the semiconductor refrigeration chip to perform the cooling action.

[0008] In a preferred embodiment of the present invention, the data acquisition module includes a status acquisition unit and a data filtering unit; In each control cycle, the status acquisition unit acquires the internal temperature through an internal temperature sensor, the ambient temperature through an ambient temperature sensor, and the instantaneous bus voltage through a bus sampling circuit, and uses the internal temperature, the ambient temperature, and the instantaneous bus voltage as the local physical data. The status acquisition unit acquires the flight boundary condition data through an isolated communication bus. The flight boundary condition data includes the instantaneous airspeed measured by the Pitot airspeed meter, the current flight condition identifier, and the waypoint matrix. The data filtering unit determines whether the current flight condition identifier corresponds to the hovering condition and whether the instantaneous airspeed is less than a preset lower airspeed threshold. When the current flight condition identifier corresponds to the hovering condition, or the instantaneous airspeed is less than the lower airspeed threshold, the data filtering unit generates an effective airspeed with a value of zero; when the current flight condition identifier does not correspond to the hovering condition, and the instantaneous airspeed is greater than or equal to the lower airspeed threshold, the data filtering unit generates an effective airspeed with a value equal to the instantaneous airspeed; wherein, the lower airspeed threshold is preset based on the measurement dead zone characteristics of the Pitot airspeed meter in the low-speed region and the aerodynamic flow field distortion characteristics, and the waypoint matrix includes the planned arrival time and expected flight condition identifier of each waypoint.

[0009] In a preferred embodiment of the present invention, the state observation module includes a heat leakage calculation unit and a cold storage simulation unit; The heat leakage calculation unit calculates the real-time temperature difference between the ambient temperature and the internal temperature in the local physical data, calculates the equivalent heat transfer coefficient based on the effective space velocity and the preset heat transfer coefficient mapping relationship, and calculates the heat leakage power by multiplying the equivalent heat transfer coefficient, the preset total surface area of ​​the vacuum insulation shell, and the real-time temperature difference; the cold storage simulation unit calls the cold storage state formula to generate the initial cold storage scalar, specifically: The net energy difference between the active cooling capacity provided by the semiconductor refrigeration chip based on the coefficient of performance and the passive heat load caused by the heat leakage power is calculated. The net energy difference is accumulated within the sampling time interval and divided by the product of the preset phase change material mass and the preset latent heat constant to obtain the normalized change of cooling capacity. The normalized change of cooling capacity is superimposed on the effective cold storage scalar generated in the previous control cycle and the upper and lower limits are truncated using an interval constraint function to generate the initial cold storage scalar. The heat transfer coefficient mapping relationship is preset based on aerodynamic wind tunnel experimental data and empirical formulas for forced surface convection. The total surface area is preset based on the physical appearance dimensions of the vacuum insulation shell of the airborne refrigerated container. The mass of the phase change material and the latent heat constant are preset based on the inherent thermodynamic properties of the phase change material in the phase change cold storage box.

[0010] In a preferred embodiment of the present invention, the state observation module further includes a state verification unit; The state verification unit performs boundary physical verification on the initial cold storage scalar, specifically as follows: When the internal temperature is lower than the preset solidification critical point for a consecutive preset number of control cycles, the state verification unit overwrites the initial cold storage scalar with 1 to generate the effective cold storage scalar; when the internal temperature is higher than the preset melting critical point for a consecutive preset number of control cycles, the state verification unit overwrites the initial cold storage scalar with 0 to generate the effective cold storage scalar; if the aforementioned conditions are not triggered, the state verification unit determines that the effective cold storage scalar is equal to the initial cold storage scalar. The preset number of control cycles is preset based on the thermal response delay of the temperature sensor and the tolerance filtering length of transient temperature spikes. The solidification critical point is preset based on the lower limit of the physical transformation temperature at which the phase change material is completely crystallized and solidified. The melting critical point is preset based on the upper limit of the physical transformation temperature at which the phase change material is completely liquefied and the heat absorption ends.

[0011] In a preferred embodiment of the present invention, the load prediction module includes a route analysis unit and a cold storage assessment unit; The route analysis unit extracts the expected flight condition identifiers within the corresponding time period from the route waypoint matrix based on a preset forward time window and the planned arrival time, generating an expected flight condition sequence. The route analysis unit determines the planned flight time for each segment by calculating the time difference between the planned arrival times of adjacent waypoints in the expected flight condition sequence, and sums the planned flight times corresponding to segments with hovering and climb condition identifiers in the expected flight condition sequence to calculate the expected duration. The cold storage assessment unit calculates the total physical latent heat by multiplying the effective cold storage scalar, the preset phase change material mass, and the preset latent heat constant. It then divides the total physical latent heat by the preset latent heat release rate to calculate the cold storage maintenance time margin. The cold storage assessment unit compares the cold storage maintenance time margin with the expected duration to generate the cold storage assessment result. The forward time window is preset based on the thermal inertia of the airborne refrigerated box and the flight time of the UAV performing a typical delivery task, and the latent heat release rate is preset based on the contact area of ​​the heat-conducting fins inside the phase change cold storage box and the thermal conductivity of the phase change material itself.

[0012] In a preferred embodiment of the present invention, the control execution module includes a mode switching unit; the mode switching unit performs working mode arbitration, specifically: When the cold storage assessment result indicates insufficient cooling capacity and the effective airspeed is greater than a preset optimal heat dissipation threshold, a mode gating coefficient with a value of 1 is generated; when the current flight condition identifier corresponds to a hovering condition or the effective airspeed is equal to zero, a mode gating coefficient with a value of 0 is generated; when the cold storage assessment result indicates sufficient cooling capacity and the internal temperature is less than or equal to a preset target refrigeration temperature, a mode gating coefficient with a value of 0 is generated; when the internal temperature is greater than the target refrigeration temperature and the effective airspeed is not equal to zero, a mode gating coefficient with a value of 1 is generated; when the current flight condition identifier, the effective airspeed, and the cold storage assessment result are not in the above-mentioned combination and the effective airspeed is not equal to zero, a mode gating coefficient with a value of 1 is generated. The optimal heat dissipation threshold is preset based on the optimal heat dissipation flow rate in aerodynamics, and the target refrigeration temperature is preset based on the preservation process standards for refrigerated items.

[0013] In a preferred embodiment of the present invention, the control execution module further includes an instruction processing unit; The instruction calculation unit extracts the target refrigeration temperature and the internal temperature and performs closed-loop proportional-integral-differential calculations to generate a feedback basis quantity; the instruction calculation unit calls the power output formula to calculate the instruction electrical power, specifically: Multiply the preset feedforward gain coefficient by the difference between the effective airspeed and the preset airspeed trigger threshold to obtain the product value; superimpose the feedback base quantity with the product value, and multiply the superposition result by the mode gating coefficient and the continuous suppression function in sequence to obtain the product result; use the power limiting function to process the product result based on the preset maximum electric power to generate the command electric power; Wherein, the continuous suppression function is a continuous function that monotonically decreases as the effective cold storage scalar increases, the feedforward gain coefficient is preset based on the slope of the linear influence of effective airspeed on heat loss, the airspeed trigger threshold is preset based on the minimum airflow velocity required to activate airspeed feedforward compensation, and the maximum electrical power is preset based on the upper limit of the rated electrical power of the semiconductor refrigeration chip.

[0014] In a preferred embodiment of the present invention, the control execution module further includes a bus verification unit; the bus verification unit determines whether the command power is greater than or equal to a preset limit cooling threshold, and determines whether the instantaneous bus voltage is less than a preset safety voltage lower limit; if both of the above judgment conditions are met at the same time, the bus verification unit multiplies the command power by the difference between the command power and a preset power attenuation ratio to generate the safety power. If the aforementioned conditions are not met, the bus verification unit will use the command power as the safe power; the control execution module will convert the safe power into the pulse width modulation signal to drive the semiconductor cooling chip to perform a cooling action; wherein, the extreme cooling threshold is preset based on the lower limit of heavy load power that may cause fluctuations in the UAV bus voltage, the lower limit of safe voltage is preset based on the power supply warning voltage of the UAV flight control system, and the power attenuation ratio is preset based on the empirical compensation rate for the bus voltage to recover to the safe range.

[0015] A second aspect of the present invention provides a refrigerated container, including a drone and an airborne refrigerated container, wherein the airborne refrigerated container is disposed below the drone. The airborne refrigerated container includes a vacuum-insulated shell and a phase change cold storage box. The phase change cold storage box is disposed inside the vacuum-insulated shell. A semiconductor refrigeration chip is installed inside the vacuum-insulated shell. A heat sink is installed inside the vacuum-insulated shell. The cold end of the semiconductor refrigeration chip faces the inside of the phase change cold storage box, and the hot end of the semiconductor refrigeration chip contacts the heat-conducting surface of the heat sink. An internal temperature sensor and an ambient temperature sensor are respectively installed on the inner and outer walls of the vacuum-insulated shell. A microcontroller is installed on the front side of the vacuum-insulated shell. The microcontroller is used to combine acquired flight boundary condition data and local physical data to deduce the effective cold storage scalar, calculate the safe electrical power and convert it into a pulse width modulation signal to drive the semiconductor refrigeration chip to perform a cooling action.

[0016] Furthermore, the UAV and the microcontroller establish a data interaction connection through an isolated communication bus. The UAV has a built-in flight control system, a pitot tube airspeed indicator, and a bus sampling circuit. The pitot tube airspeed indicator is used to acquire instantaneous airspeed in real time. The bus sampling circuit is used to collect instantaneous bus voltage. The flight control system is used to send the instantaneous airspeed, current flight condition identifier, and route waypoint matrix to the microcontroller through the isolated communication bus.

[0017] This invention provides a temperature control system and a refrigerator for an airborne refrigerated container used in unmanned aerial vehicles (UAVs). It offers the following advantages: 1. This invention obtains the effective airspeed of the UAV and calculates the expected duration of the severe heat dissipation conditions by combining the flight path and waypoint matrix; when it is determined that there is a risk of insufficient cooling in the future and the current external convection heat transfer conditions are good, the mode switching unit starts full-power cooling in advance to store cooling capacity, which can cope with the heat load changes under different flight conditions in advance and actively utilize the external flow field to store the latent heat of phase change. 2. This invention calculates the net energy difference between heat leakage power and active cooling capacity, and performs integral accumulation and boundary physical verification to deduce the effective cold storage scalar that characterizes the internal residual cold capacity. This deduction process solves the observation blind spot caused by the constant temperature of phase change materials during the solid-liquid coexistence period, and provides accurate data support for the closed-loop calculation of cooling power and the switching of working modes. 3. Before the output power drives the semiconductor cooling chip to perform the cooling action, the present invention simultaneously checks the command power and the instantaneous bus voltage. When it is determined that a high-power cooling request may cause the bus voltage to drop below the safe voltage lower limit, the system actively attenuates the power output according to a set ratio, effectively preventing the load cooling components and the UAV flight rotor motor from competing for power, and ensuring the physical safety of the UAV power system. Attached Figure Description

[0018] Figure 1 This is a perspective view of the UAV-borne refrigerated container according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the airborne refrigerated container structure according to an embodiment of the present invention; Figure 3 This is a structural diagram of a temperature control and regulation system for an unmanned aerial vehicle (UAV) airborne refrigerated box according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the workflow of the state observation module in an embodiment of the present invention; Figure 5 This is a schematic diagram of the workflow of the load forecasting module according to an embodiment of the present invention; Figure 6 This is a schematic diagram of the workflow of the control execution module according to an embodiment of the present invention; Figure 7 A comparative graph showing the changes in internal temperature and effective airspeed over time in an application embodiment of the present invention; Figure 8 A comparative graph showing the change in power consumption of the semiconductor cooling chip over time in an application embodiment of the present invention; Figure 9 This is a comparison curve of the instantaneous bus voltage of the UAV changing over time, representing an application embodiment of the present invention.

[0019] Among them, 1. UAV; 2. Vacuum insulation shell; 3. Phase change cold storage box; 4. Semiconductor cooling chip; 5. Heat sink; 6. Internal temperature sensor; 7. Ambient temperature sensor; 8. Microcontroller. Detailed Implementation

[0020] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Please see the appendix Figure 1 and attached Figure 2 This invention provides a refrigerator, including a drone 1 and an airborne refrigerator. The airborne refrigerator is located below the drone 1. The airborne refrigerator includes a vacuum-insulated shell 2 and a phase change cold storage box 3. The phase change cold storage box 3 is located inside the vacuum-insulated shell 2. A semiconductor cooling chip 4 is installed inside the vacuum-insulated shell 2. A heat sink 5 is installed inside the vacuum-insulated shell 2. The cold end of the semiconductor cooling chip 4 faces the inside of the phase change cold storage box 3, and the hot end of the semiconductor cooling chip 4 contacts the heat-conducting surface of the heat sink 5. An internal temperature sensor 6 and an ambient temperature sensor 7 are respectively installed on the inner and outer walls of the vacuum-insulated shell 2. A microcontroller 8 is installed on the front side of the vacuum-insulated shell 2.

[0022] Specifically, the UAV 1 and the microcontroller 8 on the airborne refrigerated container side establish a data interaction connection through an isolated communication bus. The airborne refrigerated container, as a relatively independent thermal management payload, performs temperature control actions.

[0023] The UAV 1 integrates a Flight Control System (FCS), a Pitot airspeed indicator, and a bus sampling circuit. The Pitot airspeed indicator is mounted externally on the nose of the UAV 1, and its signal output is connected to the sensor input interface of the flight control system. The bus interface of the flight control system is physically connected to the data terminal of the isolated communication bus. The voltage sampling terminal of the bus sampling circuit is connected in parallel between the positive and negative buses of the UAV 1's power supply bus to collect instantaneous bus voltage. The signal output of the bus sampling circuit is electrically connected to the analog signal acquisition channel of the microcontroller 8 inside the onboard refrigerated container.

[0024] The airborne refrigerated container includes a vacuum-insulated outer shell 2 and a phase change cold storage box 3. The vacuum-insulated outer shell 2 forms a closed internal cavity. A microcontroller 8, a power drive circuit, an internal temperature sensor 6, an ambient temperature sensor 7, a semiconductor cooling chip 4, a heat sink 5, and the phase change cold storage box 3 are all installed in their respective structural positions within the vacuum-insulated outer shell 2. The communication interface of the microcontroller 8 is connected to the other end of an isolated communication bus, enabling data exchange with the UAV 1.

[0025] The vacuum insulation shell 2 is made of low thermal conductivity microporous materials such as fumed silica or ultrafine glass fiber, and is wrapped with a high-barrier composite film. After being evacuated and sealed inside, it maximizes the prevention of heat conduction and convection leakage from the external high-temperature environment to the internal cavity, providing a stable and extremely low substrate static heat transfer coefficient for the calculation of heat leakage power in the condition observation module.

[0026] The phase change cold storage box 3 includes a high thermal conductivity metal shell, such as aluminum alloy. The internal cavity of the high thermal conductivity metal shell is encapsulated with a phase change material, and the inner wall of the high thermal conductivity metal shell is provided with micro-dense thermally conductive fins that are integrally formed and extend into the internal cavity of the phase change cold storage box 3. The solid-liquid phase change temperature range of the phase change material is matched with the preset target refrigeration temperature (for example, a tetradecane mixture or a composite hydrated salt material with a phase change point between 2.0℃ and 5.0℃ is selected).

[0027] The internal temperature sensor 6 is fixed to the inner wall of the internal cavity of the vacuum insulation shell 2, and the ambient temperature sensor 7 is fixed to the outer wall surface of the vacuum insulation shell 2. The signal output terminals of the internal temperature sensor 6 and the ambient temperature sensor 7 are respectively connected to the input channel of the microcontroller 8.

[0028] Among them, the internal temperature sensor 6 and the ambient temperature sensor 7 can be digital integrated chip temperature sensors, platinum resistance temperature sensors or negative temperature coefficient thermistor sensors, etc.

[0029] The semiconductor cooling chip 4 is embedded in the wall of the vacuum insulation shell 2. The cold end of the semiconductor cooling chip 4 extends into the internal cavity of the vacuum insulation shell 2, and the hot end of the semiconductor cooling chip 4 extends to the outside of the vacuum insulation shell 2 and contacts the heat-conducting surface of the heat sink 5 through silicone grease or a highly thermally conductive flexible phase change pad.

[0030] The control input terminal of the power drive circuit is connected to the pulse width modulation signal output interface of the microcontroller 8. The power supply input terminal of the power drive circuit is electrically connected to the load power supply terminal of the power supply bus of the UAV 1, and the power output terminal of the power drive circuit is connected to the power supply terminal of the semiconductor refrigeration chip 4.

[0031] See appendix Figure 3 The present invention also provides a temperature control system for an airborne refrigerated container of an unmanned aerial vehicle (UAV). The system operates in the microcontroller 8 of the airborne refrigerated container and includes a data acquisition module, a status observation module, a load prediction module, and a control execution module.

[0032] The data acquisition module is used to synchronously acquire heterogeneous data from multiple sources and perform physical credibility cleaning. The data acquisition module contains a status acquisition unit and a data filtering unit.

[0033] During each control cycle, the status acquisition unit acquires local physical data through local sensing components and acquires flight boundary condition data from the flight control system through an isolated communication bus. The data filtering unit acquires the flight boundary condition data acquired by the status acquisition unit and performs low-speed dead zone determination on the data to generate the effective airspeed.

[0034] The status observation module is used to simulate the unmeasurable physical residual cold energy inside the phase change cold storage box 3 in real time. The status observation module includes a heat leakage calculation unit, a cold storage simulation unit, and a status verification unit.

[0035] The heat leakage calculation unit acquires the local physical data obtained by the state acquisition unit and the effective air velocity generated by the data filtering unit, and calculates the transient heat leakage power. The cold storage simulation unit acquires the heat leakage power calculated by the heat leakage calculation unit and the safe electrical power fed back by the control execution module, and calls the cold storage state formula to perform transient integral calculation to generate the initial cold storage scalar. The state verification unit performs boundary physical verification on the initial cold storage scalar based on the local physical data to generate an effective cold storage scalar that characterizes the current physical remaining cold capacity of the system.

[0036] The load forecasting module is used to make forward-looking predictions of the future passive heat load of UAV 1 in terms of time and energy. The load forecasting module includes a flight path analysis unit and a cold storage assessment unit.

[0037] The flight path analysis unit acquires the flight boundary condition data extracted by the state acquisition unit, traverses and analyzes it to calculate the expected duration of the UAV1 under severe heat dissipation conditions. The cold storage assessment unit acquires the expected duration calculated by the flight path analysis unit and the effective cold storage scalar generated by the state observation module, converts the effective cold storage scalar into a cold storage maintenance time margin, compares the cold storage maintenance time margin with the expected duration, and finally generates the cold storage assessment result.

[0038] The control execution module is responsible for generating the operating mode based on the current physical state and future load forecast, and converting the operating mode into specific physical power control quantities. The control execution module includes a mode switching unit, an instruction calculation unit, and a bus verification unit.

[0039] The mode switching unit acquires local physical data, flight boundary condition data, effective airspeed, and cold storage assessment results, performs working mode arbitration, and generates corresponding mode gating coefficients. The command calculation unit acquires the mode gating coefficients, effective cold storage scalar, effective airspeed, and local physical data, performs closed-loop proportional-integral-differential operations to generate feedback basis quantities, and calls the power output formula to calculate the commanded electrical power. The bus verification unit acquires the commanded electrical power and local physical data, performs power verification and attenuation processing, and generates the final safe electrical power.

[0040] The adaptive temperature control system uses the microcontroller 8 to cyclically call the data acquisition module, status observation module, load prediction module, and control execution module according to a preset control cycle.

[0041] The data acquisition module continuously forms the baseline input of the system status. The status observation module and the load prediction module transform the sensed data into a metric for evaluating the system's cooling capacity. Based on the parameters generated by the above modules, the control execution module generates a pulse width modulation signal and applies it to the power drive circuit. The power drive circuit drives the semiconductor refrigeration chip 4 to perform the cooling action, forming a complete adaptive closed-loop temperature control workflow.

[0042] This invention also provides an adaptive temperature control method for an airborne refrigerated container for unmanned aerial vehicles (UAVs), comprising the following steps: S1, the data acquisition module synchronously acquires multi-source heterogeneous data and performs physical reliability cleaning within each control cycle. The data acquisition module obtains local physical data through local sensing components and flight boundary condition data through an isolated communication bus. The data acquisition module performs low-speed dead zone determination on the instantaneous airspeed and current flight condition identifier in the flight boundary condition data to generate the effective airspeed.

[0043] S2, the state observation module calculates the heat load and extrapolates the physical remaining cooling capacity inside the phase change cold storage box 3 in real time. Based on the effective space velocity and local physical data, the state observation module calculates the heat leakage power. Combining the heat leakage power with the safe electrical power from the previous control cycle, the state observation module uses the cold storage state formula to generate an initial cold storage scalar. The state observation module performs boundary physical verification on the initial cold storage scalar to generate an effective cold storage scalar.

[0044] S3, the load forecasting module performs forward-looking estimation of the future passive heat load of UAV 1. The load forecasting module extracts the waypoint matrix from the flight boundary condition data, analyzes and calculates the expected duration of severe heat dissipation conditions. The load forecasting module converts the effective cold storage scalar into a cold storage maintenance time margin, and generates a cold storage assessment result by comparing the cold storage maintenance time margin with the expected duration.

[0045] S4, the control execution module arbitrates the operating mode and calculates the physical power control quantity. The control execution module combines the cold storage assessment results, effective airspeed, current flight condition identifier, and local physical data to perform operating mode arbitration and generate mode gating coefficients. The control execution module calls the power output formula to calculate the commanded power. The control execution module verifies the commanded power based on local physical data and generates a safe power.

[0046] In the process of the above steps being executed cyclically, the control execution module converts the safe electrical power into a pulse width modulation signal to drive the semiconductor refrigeration chip 4 to perform the cooling action. This safe electrical power is recorded at the end of the current control cycle and used as a calculation parameter for the cold storage state deduction in the next control cycle.

[0047] The technical solutions in the embodiments of the present invention are described in detail below: In step S1 of this embodiment, the data acquisition module is mainly responsible for synchronously acquiring multi-source heterogeneous data and performing physical reliability cleaning in each control cycle, so as to provide an accurate data benchmark for subsequent calculations. The data acquisition module includes a state acquisition unit and a data filtering unit.

[0048] S101, the status acquisition unit performs acquisition actions within each control cycle. The control cycle is set based on the processing performance of the microcontroller 8 and the response time constant of the airborne refrigerated container's thermodynamic system, for example: 10ms to 50ms.

[0049] The status acquisition unit acquires local physical data through local sensing components. Specifically, the local sensing components include an internal temperature sensor 6, an ambient temperature sensor 7, and a bus sampling circuit. The status acquisition unit acquires the internal temperature through the internal temperature sensor 6, the ambient temperature through the ambient temperature sensor 7, and the instantaneous bus voltage through the bus sampling circuit.

[0050] For the analog-to-digital conversion of the physical signals of the internal temperature sensor 6 and the bus sampling circuit, those skilled in the art can use conventional analog-to-digital converter hardware circuits in conjunction with digital filtering algorithms to read the signals. The analog-to-digital conversion and physical quantity mapping process are well-known technologies in the field and will not be described in detail here.

[0051] To ensure the alignment of multi-source data across systems in the time domain, the status acquisition unit generates a local timestamp when triggering an acquisition action and binds the local physical data to the local timestamp.

[0052] Meanwhile, the status acquisition unit acquires flight boundary condition data from the flight control system through the isolated communication bus. The flight boundary condition data specifically includes instantaneous airspeed measured by the pitot tube airspeed meter, current flight condition identifier, and route waypoint matrix.

[0053] The isolated communication bus isolates the electrical interference between the UAV side and the airborne refrigerated box side in the communication signal path. The status acquisition unit reads the data frame carrying the communication timestamp broadcast by the flight control system from the isolated communication bus according to the predetermined communication protocol, and performs clock deviation compensation with the local timestamp to ensure that the input parameters during the physical model calculation are in the same time segment.

[0054] The flight path and waypoint matrix is ​​a three-dimensional path planning dataset for UAV 1 to perform flight missions. The flight path and waypoint matrix includes at least the waypoint location, waypoint altitude, planned arrival time, and expected flight condition identifiers for the corresponding flight segments.

[0055] For the encapsulation and data unpacking process of the underlying communication protocol of the isolated communication bus, those skilled in the art can configure the Controller Area Network Bus protocol or the Universal Asynchronous Transceiver Protocol. The protocol stack configuration and physical layer isolation circuit design are well-known technologies in the field and will not be described in detail here.

[0056] S102, the data filtering unit acquires the instantaneous airspeed and current flight condition identifier from the flight boundary condition data acquired by the status acquisition unit, and performs low-speed dead zone determination.

[0057] From the perspective of fluid dynamics measurement principles, in the actual operating environment of UAV 1, when UAV 1 is in a low-speed state such as hovering, the air pressure at the Pitot tube airspeed sensor probe is relatively weak, resulting in a low signal-to-noise ratio. Directly using this instantaneous airspeed would introduce broadband noise into the subsequent control logic. The data filtering unit performs physical truncation of the measurement blind zone by performing low-speed dead zone determination.

[0058] Specifically, the data filtering unit determines whether the current flight condition indicator corresponds to the hovering condition, and whether the instantaneous airspeed is less than the preset lower airspeed threshold. When the current flight condition identifier corresponds to the hovering condition, or when the instantaneous airspeed is less than the preset lower airspeed threshold, the data filtering unit determines the result of the low-speed dead zone determination as yes. When the current flight condition indicator does not correspond to the hovering condition, and the instantaneous airspeed is greater than or equal to the preset lower airspeed threshold, the data filtering unit determines the low-speed dead zone determination result as negative. The lower airspeed threshold is set based on the measurement dead zone characteristics of the pitot tube airspeed meter in the low-speed region and the aerodynamic flow field distortion characteristics, for example: 2.0 m / s to 3.5 m / s.

[0059] If the low-speed dead zone determination result is yes, it means that UAV 1 is currently in a hovering state or the external airflow is in the measurement blind zone of the Pitot airspeed meter. At this time, the physical reliability of the obtained airspeed is low, and the data filtering unit generates an effective airspeed with a value of zero. If the low-speed dead zone determination result is no, it means that UAV 1 is in a normal forward flight state and the airflow speed is sufficient to support reliable measurement. The data filtering unit generates an effective airspeed with a value of that instantaneous airspeed.

[0060] The data filtering unit eliminates aerodynamic noise interference under low-speed flight conditions by determining the low-speed dead zone, providing a physically reliable data benchmark for subsequent thermodynamic estimations.

[0061] See appendix Figure 4 In step S2 of this embodiment, the state observation module is mainly responsible for calculating the heat load and extrapolating the unmeasurable physical residual cold energy inside the phase change cold storage box 3 in real time. The state observation module includes a heat leakage calculation unit, a cold storage extrapolation unit, and a state verification unit.

[0062] S201, the heat leakage calculation unit obtains the internal temperature and ambient temperature from the local physical data obtained by the status acquisition unit of the data acquisition module, as well as the effective airspeed generated by the data filtering unit of the data acquisition module.

[0063] The heat leakage calculation unit calculates the real-time temperature difference between the ambient temperature and the internal temperature, and sets the real-time temperature difference used for heat leakage power calculation to zero when the ambient temperature is lower than or equal to the internal temperature. In actual flight conditions, the airborne refrigerated container is exposed to the external flow field, and its external heat exchange boundary conditions change with flight speed; static adiabatic parameters are difficult to accurately reflect the dynamic changes in external convective heat transfer. Based on the effective airspeed and the preset heat transfer coefficient mapping relationship, the heat leakage calculation unit calculates the equivalent heat transfer coefficient of the vacuum adiabatic shell 2 under the current flight conditions.

[0064] The heat transfer coefficient mapping relationship is a discrete data structure that is pre-calibrated and stored in the microcontroller 8 based on aerodynamic wind tunnel experimental data and empirical formulas for forced surface convection. It is used to map instantaneous air velocity in different ranges to the corresponding convective heat transfer coefficient.

[0065] The heat leakage calculation unit multiplies the equivalent heat transfer coefficient, the preset total surface area of ​​the vacuum insulation shell 2, and the real-time temperature difference to calculate the heat leakage power. This heat leakage power represents the total amount of heat that overcomes the thermal resistance of the vacuum insulation shell 2 and penetrates into the internal cavity from the external environment per unit time. The total surface area of ​​the vacuum insulation shell 2 is a structural parameter that is fixed based on the physical appearance dimensions of the airborne refrigerated container.

[0066] S202, the cold storage simulation unit obtains the heat leakage power calculated by the heat leakage calculation unit and obtains the safe electrical power generated by the bus verification unit of the control execution module in the previous control cycle. For the integral estimation model that relies on time series recursion, the accuracy of the state simulation is directly controlled by the initial boundary conditions.

[0067] When the system is in the initial control cycle, since there is no historical iteration data, the cold storage simulation unit assigns the safe power of the previous control cycle to the preset initial power constant and the effective cold storage scalar of the previous control cycle to the preset initial cold storage constant.

[0068] The initial power constant is set according to the default power supply strategy of the semiconductor cooling chip 4 during system cold start, for example, 0W; the initial cold storage constant is set according to the standard pre-cooling saturation state of the phase change cold storage box 3 before takeoff, for example, 1.

[0069] The cold storage simulation unit calls the cold storage state formula to perform transient integral calculations on the cold storage state of phase change cold storage box 3, generating an initial cold storage scalar, which is then limited to between 0 and 1. The cold storage state formula is as follows: ; In the formula: This is the initial scalar quantity of cold storage; It is an interval constraint function; The effective cold storage scalar quantity of the previous control cycle; The preset mass of the phase change material; The latent heat constant is preset; This is the definite integral operator; This refers to the previous sampling time. This is the current sampling time; Coefficient of performance (COP); The safe electrical power for the previous control cycle; This refers to the heat leakage power. For integration variables; The sign for the differential of an integral.

[0070] in, and These are physical constants that have been pre-calibrated and stored based on the physical loading amount and inherent thermodynamic properties of the phase change material in the selected phase change cold storage box 3. It is a constant that is pre-calibrated based on the inherent energy efficiency ratio parameter of the semiconductor cooling chip 4, for example, The value range can be set from 200kJ / kg to 300kJ / kg. The value can be set to 0.5 to 1.5.

[0071] The integral operation calculates the net energy difference between the active cooling capacity provided by the semiconductor refrigeration chip 4 based on the coefficient of performance and the passive heat load caused by external heat leakage power. The difference is accumulated by the integral operation within the sampling time interval and divided by the inherent total latent heat capacity of the phase change cold storage box 3 to obtain the normalized change in cooling capacity.

[0072] By superimposing the effective cold storage scalar generated in the previous control cycle and truncating the upper and lower limits using an interval constraint function, scalar data that maps the current physical cold storage is formed. For the discretized numerical integration solution process corresponding to the definite integral operation symbol, those skilled in the art can use the Runge-Kutta method or the trapezoidal integral algorithm. The numerical integration process is a well-known technology in this field and will not be described in detail here.

[0073] S203, the status verification unit obtains the initial cold storage scalar generated by the cold storage simulation unit, and the internal temperature in the local physical data obtained by the status acquisition unit of the data acquisition module.

[0074] During long-term discrete operations of the microcontroller 8, sensor quantization errors and model parameter deviations introduced by airflow disturbances can easily cause integral drift, causing the purely mathematically derived cold storage state to gradually deviate from the actual physical laws. The state verification unit performs boundary physical verification and uses the inherent thermodynamic temperature plateau characteristics of phase change materials near the solid-liquid phase transition point to reset and align the integral results.

[0075] When the internal temperature is less than the preset solidification critical point for a continuous preset number of control cycles, the state verification unit will overwrite the initial cold storage scalar with 1 to generate an effective cold storage scalar.

[0076] When the internal temperature exceeds the preset melting critical point for a continuous preset number of control cycles, the state verification unit will overwrite the initial cold storage scalar with 0 to generate an effective cold storage scalar.

[0077] If the aforementioned conditions are not triggered, it indicates that the phase change material does not meet the boundary reset conditions and can be regarded as being in the phase change transition range or the range where the model can be continuously deduced. The integral estimation process is regarded as a reliable state, and the state verification unit determines that the effective cold storage scalar is equal to the initial cold storage scalar.

[0078] By establishing a verification mechanism that unifies data-driven estimation with physical phase transition boundaries, the system corrects accumulated errors.

[0079] The preset number of control cycles is set based on the thermal response delay of the temperature sensor and the tolerance filtering length of transient temperature spikes, for example, 100 to 200 control cycles; the solidification critical point is set based on the lower limit of the physical transformation temperature at which the phase change material is completely crystallized and solidified, for example, 2.0℃; the melting critical point is set based on the upper limit of the physical transformation temperature at which the phase change material is completely liquefied and the end of heat absorption, for example, 8.0℃.

[0080] See appendix Figure 5 In step S3 of this embodiment, the load prediction module is mainly used to make forward-looking predictions of the future passive heat load of UAV 1 in terms of time and energy dimensions. The load prediction module includes a flight path analysis unit and a cold storage assessment unit.

[0081] S301, the route analysis unit acquires the route waypoint matrix from the flight boundary condition data acquired by the status acquisition unit of the data acquisition module.

[0082] Since the flight conditions of UAV 1 change dynamically with waypoints during the mission, extracting the flight condition characteristics of future segments in advance can provide data support for the prediction of heat load. The route analysis unit extracts the expected flight condition identifiers within the corresponding time period from the route waypoint matrix according to the preset look-ahead time window and the planned arrival time, and splices them sequentially to generate the expected flight condition sequence. The preset look-ahead time window is set based on the thermal inertia of the airborne refrigerated box and the flight time of UAV 1 performing a typical delivery task, for example: 10 min to 20 min.

[0083] The route analysis unit compares the current system running time with the planned arrival time of each waypoint in the route waypoint matrix, filters out the set of time nodes that fall within the look-ahead time window, extracts the expected flight condition identifiers corresponding to each node in the set of time nodes, and arranges them in chronological order to form an expected flight condition sequence; at the same time, the route analysis unit determines the planned flight time of each corresponding flight segment by calculating the time difference between the planned arrival times of adjacent waypoints in the expected flight condition sequence.

[0084] Specifically, the route parsing unit initializes an empty sequence storage array and adds the current system running time to the preset look-ahead time window length to obtain the upper limit of the time window; the route parsing unit reads the records in the route waypoint matrix line by line and extracts the planned arrival time from each record; if the extracted planned arrival time is greater than or equal to the current system running time and less than or equal to the upper limit of the time window, the route parsing unit binds the expected flight condition identifier in that record with the planned arrival time and pushes it into the sequence storage array; the route parsing unit reorders the bound data in the sequence storage array according to the ascending order of planned arrival times from smallest to largest, and finally generates the expected flight condition sequence with time information.

[0085] The flight path analysis unit extracts the flight segment records corresponding to hovering or climbing conditions from the expected flight condition sequence, and calculates the expected duration of the UAV1 under severe heat dissipation conditions.

[0086] During hovering and climbing operations, the forward speed of UAV 1 is relatively low, which leads to a decrease in the efficiency of forced convection heat transfer outside the airborne refrigerated container. Heat tends to accumulate on the surface of the container. This type of operation is defined as a severe heat dissipation condition in the system.

[0087] The route analysis unit sums up the planned flight times corresponding to the segments marked with hovering and climb conditions in the expected flight condition sequence to obtain the expected duration. The expected duration reflects the scale of passive heat load accumulation that the system will face in the future.

[0088] S302, the cold storage assessment unit obtains the expected duration calculated by the route analysis unit and the effective cold storage scalar generated by the status verification unit of the status observation module.

[0089] The cold storage assessment unit calculates the refrigeration maintenance time margin based on the effective cold storage scalar, the preset mass of phase change material, the preset latent heat constant, and the preset latent heat release rate based on the structural parameters of the phase change cold storage box 3. The preset latent heat release rate is set based on the contact area of ​​the heat-conducting fins inside the phase change cold storage box 3 and the thermal conductivity of the phase change material itself. It is used to characterize the ability of the phase change material to stably absorb heat leakage and release cold energy per unit time, for example: 15W to 25W.

[0090] The cold storage assessment unit multiplies the effective cold storage scalar, the phase change material mass, and the latent heat constant to calculate the total remaining physical latent heat in the current phase change cold storage box 3. The cold storage assessment unit divides the total physical latent heat by the preset latent heat release rate to calculate the remaining refrigeration time margin that can be maintained by relying solely on the latent heat release of the phase change material itself under the condition of no active power input intervention from the semiconductor refrigeration chip 4.

[0091] The cold storage assessment unit compares the remaining cold storage time with the expected duration. If the remaining cold storage time is less than the expected duration, it indicates that the current physical cooling capacity is insufficient to support the system to smoothly pass through the future severe heat dissipation flight segment. The cold storage assessment unit generates a cold storage assessment result that is determined to be insufficient in cooling capacity. If the remaining refrigeration time is greater than or equal to the expected duration, it indicates that there is sufficient remaining refrigeration capacity, and the refrigeration assessment unit generates a refrigeration assessment result indicating that there is sufficient refrigeration capacity.

[0092] By comparing energy and time dimensions to calculate supply and demand, the load forecasting module anticipates the future thermal failure risk of the system, providing a quantitative basis for subsequent mode switching logic.

[0093] See appendix Figure 6 In step S4 of this embodiment, the control execution module is responsible for generating the working mode based on the current physical state and future load prediction, and converting the abstract logical mode into specific physical power. The control execution module includes a mode switching unit, an instruction calculation unit and a bus verification unit.

[0094] S401, the mode switching unit acquires the current flight condition identifier from the flight boundary condition data acquired by the status acquisition unit of the data acquisition module, the effective airspeed generated by the data filtering unit of the data acquisition module, the cold storage assessment result generated by the cold storage assessment unit of the load prediction module, and the internal temperature from the local physical data acquired by the status acquisition unit of the data acquisition module.

[0095] Specifically, the mode switching unit performs working mode arbitration and executes the following logic branches sequentially according to preset priorities to address dynamic thermodynamic boundary changes during flight. When the cooling assessment result indicates insufficient cooling capacity and the effective airspeed exceeds the preset optimal heat dissipation threshold, the mode switching unit generates a mode gating coefficient with a value of 1; the optimal heat dissipation threshold is set based on the aerodynamically optimal heat dissipation flow rate, for example, 10 m / s to 15 m / s. When the system detects a future risk of insufficient cooling capacity, it utilizes the current external airflow range that can provide good convective heat transfer efficiency to initiate cooling to store latent heat of phase change.

[0096] When the current flight condition indicator corresponds to the hovering condition or the effective airspeed is zero, the risk of heat accumulation at the hot end of the semiconductor cooler 4 increases due to the limitation of external forced convection heat transfer. The mode switching unit generates a mode gating coefficient with a value of 0 and implements a passive heat unloading strategy to protect the hardware.

[0097] When the cold storage assessment result indicates that the cold capacity is sufficient and the internal temperature is less than or equal to the preset target cold storage temperature, the mode switching unit generates a mode gate coefficient with a value of 0; the target cold storage temperature is set according to the preservation process standards of refrigerated items, for example: 2.0℃ to 6.0℃; this branch triggers the energy-saving sleep mechanism to avoid additional power consumption.

[0098] When the internal temperature is greater than the preset target refrigeration temperature and the effective air velocity is not equal to zero, it indicates that the system needs to perform conventional closed-loop cooling regulation, and the mode switching unit generates a mode gating coefficient with a value of 1.

[0099] When the system state is not in any of the above known conditions and the effective airspeed is not equal to zero, the mode switching unit generates a mode gating coefficient with a value of 1. This logic ensures that, under the state combination where low-speed heat dissipation limiting conditions are not triggered, the system prioritizes the temperature safety of the internal refrigerated items, preventing the cooling function from stalling due to algorithm logic dead zones.

[0100] S402, the instruction calculation unit acquires the mode gating coefficient generated by the mode switching unit, the effective cold storage scalar generated by the state verification unit of the state observation module, the effective air velocity generated by the data filtering unit of the data acquisition module, and the internal temperature in the local physical data acquired by the state acquisition unit of the data acquisition module.

[0101] The instruction calculation unit extracts the set target refrigeration temperature and internal temperature and performs closed-loop proportional-integral-differential calculation to generate the feedback basis quantity. The closed-loop proportional-integral-differential calculation calculates the deviation ratio, deviation integral and deviation derivative between the target refrigeration temperature and the internal temperature, and then sums them by weight to obtain the feedback basis quantity used to eliminate temperature error.

[0102] In the specific calculation steps, the instruction calculation unit calculates the real-time difference between the preset target refrigeration temperature and the internal temperature to generate a temperature deviation. The instruction calculation unit extracts the temperature deviation from the previous control cycle and the historical deviation sum, multiplies the temperature deviation by a preset proportional gain to obtain the deviation proportional term, adds the multiplication of the temperature deviation and the control cycle to the historical deviation sum, multiplies it by a preset integral gain to obtain the deviation integral term, and subtracts the temperature deviation from the temperature deviation of the previous control cycle from the temperature deviation, divides it by the control cycle, and multiplies it by a preset differential gain to obtain the deviation differential term. The instruction calculation unit mathematically sums the deviation proportional term, the deviation integral term, and the deviation differential term to obtain the feedback base quantity.

[0103] The instruction processing unit calls the power output formula to calculate the initial electrical power, and then performs non-negative and maximum power limiting on the initial electrical power to generate the instruction electrical power. The power output formula is as follows: ; In the formula: Command power; This is the power limiting function; The preset maximum power; The mode gating coefficient; This serves as the basic quantity for feedback; The preset feedforward gain coefficient; Effective airspeed; The preset airspeed trigger threshold; It is a continuous suppression function; This refers to the effective cold storage scalar generated within the current control cycle.

[0104] The maximum power is set based on the upper limit of the rated power of the semiconductor cooling chip 4, for example, 50W to 100W; the feedforward gain coefficient is set based on the slope of the linear effect of effective airspeed on heat dissipation, for example, 1.5W / (m / s) to 3.0W / (m / s); the airspeed trigger threshold is set based on the minimum airflow speed required to activate airspeed feedforward compensation, for example, 5.0m / s to 8.0m / s.

[0105] In the calculation of the power output formula, the system superimposes a feedforward term based on the effective space velocity on the basis of the traditional temperature closed loop to counteract the dynamic disturbance of external convective heat transfer in advance; the continuous suppression function is a continuous function with a value between 0 and 1, and it decreases monotonically with the increase of the effective cold storage scalar; when the effective cold storage scalar approaches 1, the continuous suppression function approaches 0; when the effective cold storage scalar approaches 0, the continuous suppression function approaches 1.

[0106] As a specific implementation method, the continuous suppression function can adopt a linear decreasing model, for example, expressed by the formula: The introduction of the continuous suppression function enables the system to smoothly reduce the commanded electrical power when the phase change cold storage box 3 tends to solidify and saturate, thus avoiding power oscillations and overshoot caused by switching control.

[0107] S403, the bus verification unit obtains the command power calculated by the command calculation unit and the instantaneous bus voltage in the local physical data obtained by the status acquisition unit of the data acquisition module. When the airborne refrigerated box requests high power, it may cause the power supply bus voltage of UAV 1 to drop, thereby affecting the normal operation of the flight rotor motor.

[0108] The bus verification unit determines whether the command power is greater than or equal to the preset limit cooling threshold, and whether the instantaneous bus voltage is less than the preset safety voltage lower limit. If both of the above judgment conditions are met simultaneously, the bus verification unit reduces the command power according to the preset power attenuation ratio to generate safe power; the specific calculation logic is as follows: The commanded power is multiplied by the difference between the commanded power and the preset power attenuation ratio to obtain the safe power. If the above conditions are not met, it means that the current power supply is sufficient to support the cooling load requested by the system, and the bus verification unit will take the commanded power as the safe power.

[0109] The extreme cooling threshold is set based on the lower limit of heavy load power that may cause fluctuations in the bus voltage of UAV 1, for example: 40W to 80W; the safe voltage lower limit is set based on the power supply warning voltage of the flight control system of UAV 1, for example: 20V to 22V; the power attenuation ratio is set based on the empirical compensation rate for the bus voltage to recover to the safe range, for example: 0.1 to 0.3.

[0110] The bus verification unit converts safe electrical power into a pulse width modulation signal through a power drive circuit, and the semiconductor refrigeration chip 4 performs a cooling action based on the pulse width modulation signal.

[0111] For the specific generation process of converting electrical power into pulse width modulation signals, those skilled in the art can use the built-in timer peripheral of the microcontroller 8 to configure the duty cycle mapping register. Its underlying driving logic is a well-known technology in the field and will not be described in detail here.

[0112] The safe electrical power is recorded at the end of the current control cycle and used as the electrical power parameter for the cold storage state integration calculation in the cold storage simulation unit of the state observation module in the next control cycle, forming a complete link in which the execution result participates in the closed-loop iteration of state observation.

[0113] The following detailed description of the practical application effects of the technical solutions of the present invention, in conjunction with specific application examples, will illustrate these effects in detail.

[0114] This application example uses a certain model of medical logistics drone 1 to transport thermosensitive vaccines in a high-temperature summer environment. The onboard refrigerated container contains vaccines that require a storage environment between 2.0℃ and 6.0℃, with a target refrigeration temperature set at 4.0℃. The key parameters preset by the system are as follows: The initial cold storage constant of the phase change cold storage box 3 after complete solidification is 1, and the continuous suppression function adopts a linear decreasing model. The maximum power is set to 100W, the feedforward gain coefficient is set to 2.0W / (m / s), the airspeed trigger threshold is set to 5.0m / s, the extreme cooling threshold is set to 40W, the lower limit of the safe voltage is set to 22.0V, and the power attenuation ratio is set to 0.25.

[0115] During a level flight cruise of UAV 1, the data acquisition module obtained an ambient temperature of 32.0℃, while the internal temperature rose to 5.5℃ due to heat leakage. At this time, the flight boundary condition data indicated that the current flight condition was forward flight, with an instantaneous airspeed of 15.0 m / s. This instantaneous airspeed was greater than the preset lower airspeed threshold, and the data filtering unit generated an effective airspeed of 15.0 m / s. The cold storage assessment unit of the load prediction module determined that the cold storage maintenance time margin was greater than the expected duration, and generated a cold storage assessment result indicating sufficient cold capacity. Since the effective airspeed was not equal to zero, the mode switching unit of the control execution module performed working mode arbitration and generated a mode gating coefficient with a value of 1.

[0116] The cold storage simulation unit of the status observation module calculates that the current effective cold storage capacity has decreased to 0.2; the command calculation unit of the control execution module calculates the real-time difference between the set target cold storage temperature of 4.0℃ and the internal temperature of 5.5℃, and after closed-loop proportional-integral-differential calculation, obtains the feedback basis quantity used to eliminate temperature error as 50.0W; the command calculation unit substitutes the above specific data into the power output formula for calculation, and the derivation process is as follows: ; The feedforward compensation is calculated as 2.0 multiplied by 10.0, resulting in 20.0W. Adding the feedback base value and the feedforward compensation value together gives 70.0W. Substituting this into the continuous suppression function yields 70.0 multiplied by 0.8, which is 56.0W. After processing with the power limiting function, the generated command power is 56.0W.

[0117] The bus verification unit obtains the command power of 56.0W calculated by the command calculation unit and the instantaneous bus voltage obtained by the data acquisition module. At this time, the instantaneous bus voltage drops to 21.0V due to the additional force of the UAV's rotor in the headwind. The bus verification unit determines that the command power is greater than the preset limit cooling threshold of 40W and the instantaneous bus voltage is less than the preset safe voltage lower limit of 22.0V. The bus verification unit reduces the command power according to the preset power attenuation ratio and calculates the safe power by multiplying the command power by the difference between the command power and the preset power attenuation ratio. The process is as follows: ; The final generated safe power is 42.0W. The bus verification unit converts the safe power into a pulse width modulation signal through the power drive circuit, and the semiconductor cooling chip 4 performs the cooling action based on the pulse width modulation signal.

[0118] To further verify the actual performance of the present invention, a semi-physical simulation bench experiment was conducted. The experiment included a complete logistics route encompassing ground standby, high-speed cruising, low-speed hovering, and landing. The experimental subjects were divided into two groups: the experimental group used the adaptive temperature control system of the present invention; the control group used the traditional fixed-parameter PID temperature control strategy.

[0119] In the high-speed cruising zone, the load prediction module of the experimental group's route analysis unit detected in advance that there was a low-speed hovering zone of up to 9 minutes in the later stage, and thus determined the expected duration. The cold storage assessment unit judged that the refrigeration maintenance time margin was less than the expected duration, and generated a cold storage assessment result of insufficient cold capacity.

[0120] In the high-speed cruising zone where heat dissipation conditions are good, the mode switching unit generates a mode gate coefficient of 1 in advance, starts cooling to store latent heat of phase change, and increases the effective cold storage capacity.

[0121] When the UAV1 enters the low-speed hovering zone, its effective airspeed drops to zero, external forced convection heat transfer is limited, the mode switching unit generates a mode gating coefficient with a value of 0, implements a passive heat unloading strategy, cuts off the power input of the semiconductor cooling chip 4, and relies on the release of latent heat from the previously stored phase change material to maintain the internal temperature.

[0122] The control group lacked route forecasting and airspeed coupling logic, and did not make any cooling reserves in the high-speed cruise area. After entering the low-speed hovering area, the internal temperature rose rapidly. The traditional PID strategy suppressed the temperature and output at full load, which increased the risk of heat accumulation at the hot end of the semiconductor cooling chip.

[0123] From the perspective of instantaneous bus voltage distribution, the control group experienced multiple instances of instantaneous bus voltage dropping below the safe lower voltage limit due to full load demand in the low-speed hovering zone, affecting the normal operation of the rotor motor. The experimental group, based on the forward scheduling of the load prediction module and the power attenuation processing of the bus verification unit, shifted the power peak to the high-speed cruise zone where the bus power supply was sufficient and the external forced convection heat transfer was good, and the instantaneous bus voltage was maintained above the safe lower voltage limit throughout the entire process.

[0124] See appendix Figure 7 Based on the time scale on the horizontal axis, the effective airspeed remains stable at a high level between the 4th and 15th minutes, which is the high-speed cruising zone; the effective airspeed returns to zero between the 16th and 24th minutes, which is the low-speed hovering zone. Figure 7 The main body shows the internal temperature change over time. The thick solid line marked with a hollow circle represents the internal temperature change using the adaptive temperature control strategy of this invention, while the thin dashed line marked with a cross represents the internal temperature change using the traditional PID strategy.

[0125] Before the arrival of the low-speed hovering zone, the thick solid line undergoes an active downward pre-cooling and bottoming process; after entering the low-speed hovering zone, the thick solid line maintains a gentle, slight rise by relying on the release of latent heat, and is always constrained within the safe range of 2.0℃ to 6.0℃; the thin dashed line shows a steep upward trend in the low-speed hovering zone, and eventually breaks through the temperature boundary of 6.0℃.

[0126] See appendix Figure 8 , Figure 8 The cooling power output trajectory under different strategies is shown; the thick solid line with diamond markings represents the safe electrical power under the control of this invention, and the thin dashed line with square markings represents the output power of the traditional strategy; the power peak of the thick solid line is mainly concentrated in the high-speed cruise area, and quickly drops back to near zero when the air speed drops sharply to enter the low-speed hovering area, thus implementing passive thermal unloading; the thin dashed line shows a continuous full-load high-width pulse in the low-speed hovering area.

[0127] See appendix Figure 9The thick solid line represents the instantaneous bus voltage trajectory of the present invention, which fluctuates smoothly above the set safety voltage lower limit (22.0V) baseline. The thin dashed line represents the voltage trajectory of the traditional PID strategy, which drops in the corresponding low-speed hovering zone and crosses the safety voltage lower limit baseline.

[0128] Comparative studies have confirmed that the adaptive temperature control system of the present invention has a protective function in terms of heat load management and overall power safety scheduling.

[0129] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A temperature control system for an unmanned aerial vehicle (UAV) airborne refrigerated container, characterized in that, include: The data acquisition module is used to acquire local physical data and flight boundary condition data in each control cycle, perform low-speed dead zone determination on the instantaneous airspeed and current flight condition identifier in the flight boundary condition data, and generate effective airspeed. The state observation module calculates the heat leakage power based on the effective airspeed and the local physical data, calls the cold storage state formula and performs boundary physical verification to generate the effective cold storage scalar. The load prediction module calculates the expected duration based on the route and waypoint matrix in the flight boundary condition data, converts the effective cold storage scalar into a cold storage maintenance time margin and compares it with the expected duration to generate a cold storage assessment result. The control execution module is used to combine the cold storage assessment results, the effective airspeed, the current flight condition identifier and the local physical data to execute the working mode arbitration generation mode gate coefficient, call the power output formula to calculate the command power, check the command power according to the instantaneous bus voltage in the local physical data, generate safe power, and convert the safe power into a pulse width modulation signal to drive the semiconductor refrigeration chip (4) to perform the refrigeration action; The data acquisition module includes a status acquisition unit that acquires the internal temperature through an internal temperature sensor (6), the ambient temperature through an ambient temperature sensor (7), and the instantaneous bus voltage through a bus sampling circuit in each control cycle, and uses the internal temperature, the ambient temperature, and the instantaneous bus voltage as the local physical data. The state observation module includes a heat leakage calculation unit and a cold storage simulation unit; The heat leakage calculation unit calculates the real-time temperature difference between the ambient temperature and the internal temperature in the local physical data, calculates the equivalent heat transfer coefficient according to the effective air velocity and the preset heat transfer coefficient mapping relationship, and calculates the heat leakage power by multiplying the equivalent heat transfer coefficient, the total surface area of ​​the vacuum insulation shell (2) and the real-time temperature difference. The cold storage simulation unit calls the cold storage state formula to generate the initial cold storage scalar quantity, specifically: The net energy difference between the active cooling capacity provided by the semiconductor cooling chip (4) based on the coefficient of performance and the passive heat load caused by the heat leakage power is calculated. The net energy difference is accumulated over the sampling time interval and divided by the product of the preset phase change material mass and the preset latent heat constant to obtain the normalized change in cooling capacity. The normalized change in the cold energy is superimposed on the effective cold storage scalar generated in the previous control period, and the upper and lower limits are truncated using an interval constraint function to generate the initial cold storage scalar. The heat transfer coefficient mapping relationship is preset based on aerodynamic wind tunnel experimental data and surface forced convection empirical formula; the total surface area is preset based on the physical appearance dimensions of the vacuum insulation shell (2) of the airborne cold storage box; and the mass of the phase change material and the latent heat constant are preset based on the inherent thermodynamic properties of the phase change material in the phase change cold storage box (3). The status observation module also includes a status verification unit; The state verification unit performs boundary physical verification on the initial cold storage scalar, specifically as follows: When the internal temperature is less than the preset solidification critical point for a continuous preset number of control cycles, the state verification unit overwrites the initial cold storage scalar with 1 to generate the effective cold storage scalar. When the internal temperature is greater than the preset melting critical point for a continuous preset number of control cycles, the state verification unit overwrites the initial cold storage scalar with 0 to generate the effective cold storage scalar. If the aforementioned conditions are not triggered, the status verification unit determines that the effective cold storage scalar is equal to the initial cold storage scalar; The preset number of control cycles is preset based on the thermal response delay of the temperature sensor and the tolerance filtering length of transient temperature spikes. The solidification critical point is preset based on the lower limit of the physical transformation temperature at which the phase change material is completely crystallized and solidified. The melting critical point is preset based on the upper limit of the physical transformation temperature at which the phase change material is completely liquefied and the heat absorption ends.

2. The temperature control system for an unmanned aerial vehicle (UAV) airborne refrigerated container according to claim 1, characterized in that, The data acquisition module includes the status acquisition unit and the data filtering unit; The status acquisition unit acquires the flight boundary condition data through an isolated communication bus. The flight boundary condition data includes the instantaneous airspeed measured by the Pitot airspeed meter, the current flight condition identifier, and the route waypoint matrix. The data filtering unit determines whether the current flight condition identifier corresponds to the hovering condition, and whether the instantaneous airspeed is less than the preset lower airspeed threshold. When the current flight condition identifier corresponds to the hovering condition, or when the instantaneous airspeed is less than the lower airspeed threshold, the data filtering unit generates the effective airspeed with a value of zero. When the current flight condition identifier does not correspond to the hovering condition, and the instantaneous airspeed is greater than or equal to the lower airspeed threshold, the data filtering unit generates the effective airspeed assigned to the instantaneous airspeed. The lower airspeed threshold is preset based on the measurement dead zone characteristics of the Pitot airspeed meter in the low-speed region and the aerodynamic flow field distortion characteristics. The waypoint matrix includes the planned arrival time and expected flight condition identifiers for each waypoint.

3. The temperature control system for an unmanned aerial vehicle (UAV) airborne refrigerated container according to claim 2, characterized in that, The load prediction module includes a route analysis unit and a cold storage assessment unit; The route analysis unit extracts the expected flight condition identifiers within the corresponding time period from the route waypoint matrix based on the preset forward time window and the planned arrival time, and generates the expected flight condition sequence. The route analysis unit determines the planned flight time of each segment by calculating the time difference between the planned arrival times of adjacent waypoints in the expected flight condition sequence, and sums up the planned flight times corresponding to the segments with hovering and climb conditions in the expected flight condition sequence to calculate the expected duration. The cold storage assessment unit calculates the total physical latent heat by multiplying the effective cold storage scalar, the preset phase change material mass, and the preset latent heat constant, and then calculates the cold storage maintenance time margin by dividing the total physical latent heat by the preset latent heat release rate. The cold storage assessment unit compares the cold storage maintenance time margin with the expected duration and generates the cold storage assessment result. The forward time window is preset based on the thermal inertia of the airborne refrigerated box and the flight time of the UAV (1) to perform a typical delivery task. The latent heat release rate is preset based on the contact area of ​​the heat-conducting fins inside the phase change cold storage box (3) and the thermal conductivity of the phase change material itself.

4. The temperature control system for an unmanned aerial vehicle (UAV) airborne refrigerated container according to claim 1, characterized in that, The control execution module includes a mode switching unit; The mode switching unit performs working mode arbitration, specifically as follows: When the cold storage assessment result is insufficient cold capacity and the effective air velocity is greater than the preset optimal heat dissipation threshold, the mode gating coefficient with a value of 1 is generated. When the current flight condition identifier corresponds to a hovering condition or the effective airspeed is equal to zero, the mode gating coefficient with a value of 0 is generated; When the cold storage assessment result is that the cold capacity is sufficient and the internal temperature is less than or equal to the preset target cold storage temperature, the mode gating coefficient with a value of 0 is generated. When the internal temperature is greater than the target refrigeration temperature and the effective air velocity is not equal to zero, the mode gating coefficient with a value of 1 is generated. When the current flight condition identifier, the effective airspeed, and the cold storage assessment result are not in the above-mentioned combination and the effective airspeed is not equal to zero, the mode gating coefficient with a value of 1 is generated. The optimal heat dissipation threshold is preset based on the optimal heat dissipation flow rate in aerodynamics, and the target refrigeration temperature is preset based on the preservation process standards for refrigerated items.

5. The temperature control system for an unmanned aerial vehicle (UAV) airborne refrigerated container according to claim 4, characterized in that, The control execution module also includes an instruction processing unit; The instruction calculation unit extracts the target refrigeration temperature and the internal temperature and performs closed-loop proportional-integral-differential calculations to generate feedback basic quantities. The instruction calculation unit calls the power output formula to calculate the instruction power, specifically as follows: The product value is obtained by multiplying the preset feedforward gain coefficient by the difference between the effective airspeed and the preset airspeed trigger threshold. The feedback base quantity and the product value are superimposed, and the superposition result is multiplied by the mode gating coefficient and the continuous suppression function in sequence to obtain the product result; The product result is processed using a power limiting function based on a preset maximum power to generate the command power. Wherein, the continuous suppression function is a continuous function that monotonically decreases as the effective cold storage scalar increases, the feedforward gain coefficient is preset based on the slope of the linear influence of effective airspeed on heat dissipation, the airspeed trigger threshold is preset based on the minimum airflow speed for initiating airspeed feedforward compensation, and the maximum electric power is preset based on the upper limit of the rated electric power of the semiconductor refrigeration chip (4).

6. The temperature control system for an unmanned aerial vehicle (UAV) airborne refrigerated container according to claim 1, characterized in that, The control execution module also includes a bus verification unit; The bus verification unit determines whether the commanded power is greater than or equal to a preset limit cooling threshold, and determines whether the instantaneous bus voltage is less than a preset safety voltage lower limit. If both judgment conditions are met simultaneously, the bus verification unit multiplies the command power by the difference between the command power and the preset power attenuation ratio to generate the safe power. If both judgment conditions are not met simultaneously, the bus verification unit will take the command power as the safe power. The control execution module converts the safety power into the pulse width modulation signal to drive the semiconductor refrigeration chip (4) to perform a cooling action; The extreme cooling threshold is preset based on the heavy load power lower limit that causes the bus voltage fluctuation of the UAV (1), the safe voltage lower limit is preset based on the power supply warning voltage of the flight control system of the UAV (1), and the power attenuation ratio is preset based on the empirical compensation rate for the bus voltage to recover to the safe range.

7. A refrigerator, characterized in that, A temperature control system for an airborne refrigerator for a drone, as described in any one of claims 1-6, comprises a drone (1) and an airborne refrigerator, wherein the airborne refrigerator is disposed below the drone (1), the airborne refrigerator comprising a vacuum-insulated shell (2) and a phase change cold storage box (3), the phase change cold storage box (3) being disposed inside the vacuum-insulated shell (2), a semiconductor cooling chip (4) being installed inside the vacuum-insulated shell (2), and a heat sink (5) being installed inside the vacuum-insulated shell (2), wherein the cold end of the semiconductor cooling chip (4) is... Facing the interior of the phase change cold storage box (3), the hot end of the semiconductor refrigeration chip (4) contacts the heat-conducting surface of the heat sink (5). An internal temperature sensor (6) and an ambient temperature sensor (7) are respectively installed on the inner and outer walls of the vacuum insulation shell (2). A microcontroller (8) is installed on the front side of the vacuum insulation shell (2). The microcontroller (8) is used to combine the acquired flight boundary condition data with local physical data to deduce the effective cold storage scalar, calculate the safe electrical power and convert it into a pulse width modulation signal to drive the semiconductor refrigeration chip (4) to perform the cooling action.

8. A refrigerator according to claim 7, characterized in that, The UAV (1) and the microcontroller (8) establish a data interaction connection through an isolated communication bus. The UAV (1) has a built-in flight control system, a pitot tube airspeed meter and a bus sampling circuit. The pitot tube airspeed meter is used to obtain instantaneous airspeed in real time; The bus sampling circuit is used to collect instantaneous bus voltage; The flight control system is used to send the instantaneous airspeed, current flight condition identifier, and route waypoint matrix to the microcontroller (8) via the isolated communication bus.

Images