A latent heat defrosting intelligent control method and system

Abstract

The present invention discloses an intelligent control method and system for latent heat defrost, which specifically relates to the technical field of latent heat defrost control, comprising the following steps: accurately judging the defrost start timing by establishing a multi-parameter environmental perception model, selecting carbon dioxide, Freon or air as a heat medium, and inputting latent heat into the heat exchange area through an independent heating cycle path; the heat exchange pipe dynamically judges the gravity direction according to the platform posture, and preferentially heats the pipe section in the natural frost stripping direction; an intermittent heating strategy based on thermal modeling is adopted to control energy consumption; when energy is insufficient, heating is suspended and the status is recorded, and defrosting is continued after energy is restored; the present invention realizes accurate judgment of defrost start, directional and efficient heating control and closed-loop management of the whole process, which significantly improves energy utilization efficiency, frost stripping effect and control reliability, has intelligence, adaptability and recoverability, and is suitable for stable defrosting needs in energy-constrained and complex environmental scenarios such as high-altitude platforms.

Description

Technical Field

[0001] The present invention relates to the technical field of latent heat defrosting control, and more particularly to a latent heat defrosting intelligent control method and system. Background Art

[0002] In operating environments such as high-altitude floating platforms, external condensation structures on aircraft, and unmanned heat transfer equipment, frost is prone to forming on heat exchange surfaces due to extremely low ambient temperatures and frequent humidity fluctuations. Once frost accumulates, it significantly reduces heat exchange efficiency, increases energy consumption, and can affect the stability of the platform's overall attitude control and the normal operation of mission equipment. In severe cases, it can even cause equipment failure. Therefore, ensuring that heat exchange structures remain clean and unobstructed during operation is one of the keys to ensuring the long-term stable operation of high-altitude equipment.

[0003] Existing technologies often use heating defrost, electric heating film heating, compression heat reflow, and mechanical scraping to treat frosted areas. However, these traditional methods suffer from the following shortcomings: high energy consumption, the inability to dynamically adjust the thermal control process, which can easily lead to overheating or ineffective heating; a lack of adaptive capabilities to environmental changes, with the defrost triggering logic relying on manually set thresholds or fixed cycles, unable to accurately respond to the dynamic changes in the complex high-altitude environment; and an inability to effectively coordinate defrost resources with other platform tasks during periods of limited energy or high-priority tasks. The defrost process is easily interrupted and lacks a recovery mechanism.

[0004] Furthermore, traditional defrosting methods mostly employ "global heating" or "localized periodic activation" strategies, failing to dynamically optimize the heating area based on factors such as gravity direction and frost thickness. This results in uneven heat distribution and low defrosting efficiency. Furthermore, the determination of defrosting completion lacks precision, often relying on time measurement or empirical judgment. This carries the risk of insufficient defrosting or redundant heating, compromising equipment safety. Therefore, the present invention proposes a latent heat defrosting intelligent control method and system to address these issues. Summary of the Invention

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A latent heat defrosting intelligent control method comprises the following steps:

[0007] Establish a multi-parameter environmental perception model that includes air pressure, temperature, humidity, heat exchange surface status, and platform attitude angle, and use this model to determine whether the critical conditions for initiating defrost have been met;

[0008] At least one of carbon dioxide, freon or air is selected as the heat medium, and the heat medium is guided to input latent heat to the predetermined heat exchange area through an independent heating circulation path isolated from the heat flow of the refrigeration system;

[0009] The heat exchange area is composed of independently arranged heat exchange tubes. The direction of gravity is dynamically determined based on the current platform posture angle, and the tube sections with the best position for natural frost stripping are selected for priority heating.

[0010] During the heating process, the target defrost time period is determined based on the predicted phase change latent heat transfer rate, and intermittent heating is performed on the selected heat exchange tube sections, ensuring that the average power in each heating cycle does not exceed the current platform's redundant energy capacity.

[0011] When insufficient energy is detected, the heating process is suspended and the completed defrost time and area information are recorded. After energy is restored, the remaining defrost process is continued until closed-loop defrost control is completed.

[0012] In a preferred embodiment, the construction process of the multi-parameter environmental perception model includes a dynamic coupling calculation of the high-altitude ambient air pressure and the relative motion state of the platform. The air pressure information is continuously collected from the platform's current altitude, latitude, and atmospheric temperature, and the empirical pressure-altitude formula and one of the moist air state equations are used to derive the instantaneous air pressure prediction value.

[0013] The platform attitude angle information is realized by averaging the filtered values ​​of multiple inertial measurement units and is corrected by combining the actual platform load offset to form a corrected angle model.

[0014] The heat exchange surface status information is continuously scanned and acquired by a micro-infrared temperature difference sensor, and the surface frost thickness is estimated based on the grayscale fitting function of the thermal map data.

[0015] This multi-parameter environmental perception model is based on the fitting curve of the environmental change trend in the historical operation cycle, and combines the current collection results to perform preset multi-target predictions to determine whether the temperature and pressure energy state of the critical point for triggering the first defrost has been reached.

[0016] In a preferred embodiment, the independent heating cycle path is configured to be physically isolated from the cooling cycle by setting up a closed pipe path, the heat medium filled therein is selected from one of carbon dioxide, freon or air, and a controllable variable speed flow is achieved by a built-in multi-stage micro pump;

[0017] The heating path is pre-set with multiple heat exchange branch sections, each of which has a one-way heat conduction performance to prevent heat from flowing back into the main heat exchange condensation side;

[0018] In the control strategy, the average rate of latent heat release per unit length of heat medium under the current environment is calculated, and the flow rate and heating time are dynamically selected by matching the current estimated frost thickness of the heat exchange surface.

[0019] The direction of heat medium flow is adjusted in real time according to the platform posture information, and is preferentially guided to the branch section in the favorable direction of gravity defrosting, so as to improve the frost peeling efficiency and reduce the peak energy consumption per unit time.

[0020] In a preferred embodiment, the gravity direction identification strategy for determining whether the heating target area is in a favorable position for natural frost stripping includes performing vector projection of the heat exchange tube layout direction under the platform's three-dimensional attitude angles, namely pitch, yaw, and roll, and using the gravity vector as a reference to determine the most favorable potential frost stripping direction by calculating the angle between the heat exchange tube axis and the gravity vector;

[0021] When the angle is less than the set first angle threshold, the heat exchange section is considered to have the ability to naturally defrost; if all areas do not meet the requirements, it enters a waiting state until the next cycle environmental parameters are updated and re-evaluated for activation.

[0022] In a preferred embodiment, the thermal modeling algorithm for predicting the latent heat transfer rate includes a linked calculation of the heat medium flow rate, initial temperature, ambient temperature-humidity ratio and current frost layer thickness, and is modeled using one of a group of classical heat transfer models, including a one-dimensional steady-state heat conduction model, a non-steady-state phase change model or a thermal resistance network model. By calculating the deviation between the latent heat release value per unit time and the total latent melting demand of the frost layer, the predicted defrost time period is dynamically adjusted.

[0023] The intermittent heating strategy uses a dynamic duty cycle control algorithm to set the on-off ratio within the heating cycle based on the current redundant energy capacity, mission load peaks and valleys, and the external air pressure fluctuation rate. The duty cycle is limited by the power control threshold and allows automatic adjustment at different stages.

[0024] The heating cycle is divided into two stages: activation period and cooling period. The activation period is carried out after the platform task management layer allocates available energy. The cooling period is adjusted according to the trend of external temperature changes to prevent condensation and refreezing during the defrosting process.

[0025] In a preferred embodiment, the defrost status recording strategy for pausing the heating process includes synchronously recording the currently heated area, the input heat energy, the average temperature rise of the heat exchange surface, and the timestamp of the current heating time point. This recorded information forms a defrost breakpoint status structure and is written into the platform task data storage area;

[0026] When energy is restored, the system first reads the previous breakpoint information and estimates the current required compensation heat based on the unfinished defrosting area, the heated energy consumption, and the remaining thickness of the frost layer. This is then compared with the latent heat release rate of the heat medium to determine whether the system has the capability to execute the function.

[0027] If the conditions for continuing execution are met, the heating process of the corresponding heat exchange tube section is directly resumed, otherwise it is delayed to enter the next cycle.

[0028] In a preferred embodiment, the method for obtaining the platform's redundant energy capacity includes comparing the platform's current total task power demand with the energy supply capacity in real time, determining whether to allow the activation of the heating operation based on the redundant capacity threshold, and the heating control sub-process is only executed under the condition that the platform task scheduling mechanism sends a non-interference signal.

[0029] In a preferred embodiment, the completion criteria of the closed-loop defrost control are jointly determined based on the heat exchange surface temperature recovery rate, the frost layer shedding detection results, and the statistical results of the remaining unprocessed area, and the termination threshold is met as the basis for determination;

[0030] The surface temperature is detected in real time by multiple bonded temperature sensors and compared with the initial unfrosted temperature to calculate whether the heating rate is stable and tends to zero;

[0031] The frost layer is removed by micro-vibration acceleration feedback to determine whether physical removal is completed;

[0032] The distribution of untreated areas is distinguished by comparing the temperature difference before and after treatment based on thermal image recognition;

[0033] When all three indicators reach the preset threshold, the defrost is automatically determined to be completed and the task record is cleared.

[0034] In a preferred embodiment, a latent heat defrosting intelligent control system includes:

[0035] An environmental perception unit, which is used to establish a multi-parameter environmental perception model including air pressure, temperature, humidity, heat exchange surface state, and platform attitude angle, and to determine whether the critical conditions for starting defrost have been met based on the model;

[0036] A heat medium guiding unit is used to select at least one of carbon dioxide, freon or air as a heat medium and guide the heat medium to input latent heat to a predetermined heat exchange area through an independent heating circulation path isolated from the heat flow of the refrigeration system;

[0037] The directional heating unit is used to dynamically determine the direction of gravity in the heat exchange area formed by independently arranged heat exchange tubes based on the current platform posture angle, and prioritize heating the tube sections that are in a favorable position for natural frost stripping.

[0038] A power control unit is used to determine a target defrost time period based on the predicted phase change latent heat transfer rate during the heating process, and to perform intermittent heating on the selected heat exchange tube section, controlling the average power in each heating cycle to not exceed the current platform redundant energy capacity;

[0039] The status continuation control unit is used to suspend the heating process and record the completed defrosting time and area information when insufficient energy is detected, and continue the remaining defrosting process after energy is restored until the closed-loop defrost control is completed.

[0040] Technical effects and advantages of the present invention:

[0041] The present invention establishes a multi-parameter environmental perception model, combines air pressure, temperature and humidity, heat exchange surface status and platform posture information, and realizes accurate judgment of defrost triggering, which significantly improves the scientificity and timeliness of defrost startup. Traditional timed heating or single-parameter triggering mechanisms are prone to false starts or response lags in high-altitude dynamic environments, resulting in energy waste or difficult-to-control frost accumulation. The present invention determines whether the temperature and pressure energy critical point has been reached by integrating historical trends with real-time collection, thereby fundamentally avoiding the common problems of "early start, late start, and false start" in the defrost process, ensuring that heating control is only executed when necessary, and improving the energy scheduling efficiency and task responsiveness of the entire platform.

[0042] The present invention combines an independent heating cycle path with a gravity direction identification strategy to construct a directional defrosting and regional priority heating mechanism, significantly improving the thermal efficiency utilization and reliability of the defrosting process. Compared with the global uniform heating mode of the traditional defrosting method, the present invention guides the heat medium to flow preferentially to the heat exchange pipe section in the direction of natural frost stripping, and dynamically controls the flow rate and heating time, so that the heat is concentrated on the area that is more prone to defrosting, reducing ineffective heating and heat loss. At the same time, the multi-stage micropump and unidirectional heat conduction structure are used to avoid heat backflow, effectively improving the frost stripping rate per unit time, reducing the defrosting time, and ensuring the long-term operational stability of the equipment.

[0043] The present invention ensures that the defrost task has full-process closed-loop control capabilities with interruption recovery, process traceability, and execution termination by constructing a defrost breakpoint state structure and a multi-indicator closed-loop completion judgment mechanism. In the case of energy shortage or task conflict, the present invention can record core variables such as heating status, heat input, and temperature rise trend. After energy is restored, the control process can be accurately restored according to the remaining state of the frost layer to avoid repeated heating and resource waste. At the same time, when defrosting is completed, the termination time is jointly determined by the three indicators of temperature recovery rate, micro-vibration response, and thermal map recognition, effectively preventing overheating and incomplete defrosting, and realizing a truly intelligent, efficient, and controllable defrost closed-loop control process. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings;

[0045] Figure 1 This is a schematic diagram of the principle of a latent heat defrosting intelligent control method in the present invention.

[0046] Figure 2This is a schematic diagram of a latent heat defrosting intelligent control system in the present invention. DETAILED DESCRIPTION

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0048] Reference Figure 1 - Figure 2 The following examples were obtained:

[0049] Example 1:

[0050] The present invention relates to an intelligent control method for latent heat defrosting in a high-altitude floating platform environment, which aims to achieve an efficient, controllable and reliable defrosting process under conditions of extremely low air pressure, frequent attitude changes and limited energy. The method dynamically determines whether the defrost start-up conditions are met by constructing a multi-parameter environmental perception model based on air pressure, temperature, humidity, heat exchange surface state and platform attitude angle, and realizes predictive response to environmental changes in combination with historical trends. The system adopts an independent heating cycle path isolated from the heat flow of the refrigeration system, selects carbon dioxide, Freon or air as the heat medium, drives the heat medium into the heat exchange tubes independently arranged on the structure through a multi-stage controllable micropump, and dynamically selects the heating area in the natural frost stripping advantage direction according to the platform attitude, effectively improving the frost stripping efficiency.

[0051] During the defrosting process, the heat transfer model is used to predict the matching relationship between the latent heat transfer rate and the frost melting demand, the defrosting time range is set, and a dynamic duty cycle control strategy is used for intermittent heating based on the current redundant energy capacity. When energy is insufficient, the system records the executed defrosting status based on the timestamp and heat input, and writes it into the platform task data storage area; after the energy is restored, the platform task management layer will determine whether the conditions for continued execution are met, and restore the heating process based on the breakpoint information until the closed loop is completed. The entire method uses highly adaptive control logic throughout the perception, judgment, execution and feedback links, combined with gravity direction recognition, local energy consumption regulation and termination state multi-determination mechanism, to significantly improve the defrosting stability and resource utilization efficiency of the floating platform in complex operating environments.

[0052] The specific steps include:

[0053] A multi-parameter environmental perception model, encompassing air pressure, temperature, humidity, heat exchange surface conditions, and platform attitude angles, is established. Based on this model, critical conditions for initiating defrost are determined. This step serves to build a comprehensive model for perceiving the platform's current operating environment. Due to the high uncertainty inherent in the floating platform's environment, only through multi-dimensional perception of meteorological conditions, frost conditions, and equipment attitude can the necessity and timing of defrost be accurately determined, preventing false triggering or delayed initiation and ensuring the control system's targeted and real-time judgment.

[0054] At least one of carbon dioxide, Freon, or air is selected as the heat medium. Through an independent heating cycle path isolated from the refrigeration system's heat flow, the heat medium is directed to the designated heat exchange area to transfer latent heat. The core significance of this step is to establish an independent heating channel that does not interfere with the refrigeration process, thereby achieving precise control of heat flow. This selectable heat medium allows the system to adapt to different environmental requirements, while also enhancing the system's thermal response efficiency and energy scheduling flexibility.

[0055] The heat exchange area is composed of independently arranged heat exchange tubes on the structure. The direction of gravity is dynamically determined based on the current platform posture angle, and the pipe section with an advantageous position in the natural frost stripping direction is selected for priority heating. The design significance here is to achieve directional optimization of the frost stripping process. By sensing the relationship between the platform posture and the direction of gravity, the heating is more concentrated in the position where the frost layer is most likely to fall off naturally, maximizing the use of gravity, improving defrosting efficiency and reducing energy waste.

[0056] During the heating process, the target defrost time period is determined based on the predicted latent heat transfer rate of phase change. Intermittent heating is then applied to selected heat exchange tube sections, ensuring that the average power consumed during each heating cycle does not exceed the platform's redundant energy capacity. This step emphasizes the precise matching of energy and heat to avoid overheating and energy waste. Thermal model predictions, optimal heating time scheduling, and intermittent control methods are employed to ensure effective defrosting while minimizing interference with other platform tasks, achieving a balance between thermal control and energy sustainability.

[0057] When energy shortage is detected, the heating process is suspended and the completed defrost duration and area information are recorded. Once energy is restored, the remaining defrost process continues until closed-loop defrost control is complete. This step is crucial for enabling system recovery from interruptions, ensuring continuity and fault tolerance in defrost tasks. By not forcing defrost execution during energy shortages, but retaining current progress data and resuming execution after conditions are restored, defrost execution not only ensures safe and stable system operation but also improves resource utilization.

[0058] The construction process of the multi-parameter environmental perception model includes the dynamic coupling calculation of the high-altitude ambient air pressure and the relative motion state of the platform. The air pressure information is continuously collected from the platform's current altitude, latitude, and atmospheric temperature. The empirical pressure-altitude formula and one of the moist air state equations are used to derive the instantaneous air pressure prediction value.

[0059] The platform attitude angle information is realized by averaging the filtered values ​​of multiple inertial measurement units and is corrected by combining the actual platform load offset to form a corrected angle model.

[0060] The heat exchange surface status information is continuously scanned and acquired by a micro-infrared temperature difference sensor, and the surface frost thickness is estimated based on the grayscale fitting function of the thermal map data.

[0061] This multi-parameter environmental perception model is based on the fitting curve of the environmental change trend in the historical operation cycle, and combines the current collection results to perform preset multi-target predictions to determine whether the temperature and pressure energy state of the critical point for triggering the first defrost has been reached.

[0062] In this paper, a multi-parameter environmental perception model was constructed to accurately identify defrost control triggering conditions during the operation of a high-altitude floating platform. This model's construction first involves calculating the dynamic coupling between the high-altitude ambient air pressure and the platform's relative motion state, reflecting the combined impact of ambient pressure changes on the platform's posture and external heat exchange conditions.

[0063] Specifically, the platform obtains its current operating altitude information through its altimeter. This information is combined with geographic latitude data provided by the positioning module and the current atmospheric temperature value from the ambient temperature sensor to form a basic set of input variables. These variables are highly correlated and together determine the thickness and density distribution of the air column in which the platform is located, thereby affecting the air pressure value.

[0064] When calculating air pressure, the empirical pressure-altitude formula is used for preliminary derivation. This formula is based on the standard atmospheric structure model in atmospheric physics and reflects the law that air pressure gradually decreases with increasing altitude. The specific calculation logic is:

[0065] First, the standard sea level pressure is set, typically 101,325 Pa. Using the currently measured altitude as a variable, the corresponding atmospheric temperature at that altitude is calculated based on the temperature gradient in the troposphere (generally, a temperature drop of approximately 6.5 degrees Celsius per kilometer of elevation). These parameters are then substituted into an exponential function to determine the corresponding static pressure.

[0066] If you need to improve accuracy, especially in areas with high relative humidity or frequent platform fluctuations, you can switch to using one of the moist air state equations for correction calculations. This equation treats air as a mixture of dry air and water vapor, calculates the partial pressures of the two separately, and then combines them to obtain the current mixed pressure. In the calculation, the water vapor partial pressure can be indirectly obtained from the air temperature and relative humidity through the Clausius-Clapeyron relationship. Based on the theoretical logarithmic relationship of the saturated water vapor pressure with temperature in the Clausius-Clapeyron relationship, an exponential empirical function is used in practical applications to fit it to improve computational simplicity and engineering adaptability. Specifically:

[0067] Saturated vapor pressure is calculated using an empirical exponential function with temperature as the variable. The formula is: saturated vapor pressure equals a constant of 6.112 times an exponential function, where the exponent is 17.67 times the temperature in degrees Celsius divided by the temperature plus 243.5. Actual vapor pressure equals saturated vapor pressure times relative humidity (percentage converted to a decimal between 0 and 1). Subsequently, the dry air partial pressure is calculated by subtracting the actual vapor pressure from the total air pressure. This constitutes the partial pressure input parameter for the moist air equation of state, ultimately yielding the predicted instantaneous air pressure.

[0068] The platform will ultimately select one of the two methods described above, based on whether the current environment exhibits significant humidity fluctuations, as measured by the data standard deviation meeting a preset requirement, or whether the system enters a sensitive altitude range. Under normal stable flight conditions, an empirical formula will be used, while the humid air state equation will be automatically activated in the event of a sudden change in humidity, a jump in altitude, or when the defrost boundary is approaching.

[0069] The prediction results ultimately generate a real-time, updated instantaneous air pressure value, which serves as one of the core criteria for subsequent models to determine whether to initiate defrost. This ensures that the platform can achieve stable, high-precision air pressure modeling at various altitudes and atmospheric structures, providing scientific support for latent heat defrost timing.

[0070] During platform operation, attitude measurement is often affected by factors such as platform load asymmetry and motion inertia, which can cause deviations in the raw angle data collected by the IMU. To improve the accuracy of determining the direction of the heat exchange tube and gravity, a center of mass offset correction function is required.

[0071] The center of mass offset correction function is a functional model used to compensate for the impact of the platform's current structural load distribution on attitude angles. The function's input parameters include the standard center of mass position of the platform structure, the current load mass distribution (such as batteries, communication equipment, thermal management components, etc.), and the corresponding installation location coordinates. Based on the principle of conservation of static torque, the three-dimensional offset vector between the current true center of mass and the standard center of mass is calculated. This offset vector is then corrected by superimposing the IMU-collected angles using a direction cosine matrix or quaternion attitude transformation to derive the true pitch, yaw, and roll angles, thereby improving the accuracy of gravity direction determination for defrost zone positioning.

[0072] To determine the heat exchange surface condition, a miniature infrared temperature difference sensor, mounted directly opposite the heat exchange tube array, continuously scans the surface. The sensor outputs an infrared heat distribution map as a frame image, with grayscale values ​​corresponding to different surface temperatures. Using a grayscale-temperature calibration function, the grayscale values ​​are converted to actual temperature distributions. A spatial fitting algorithm for the thermal distribution gradient is then used to estimate the frost thickness in each region.

[0073] After visualizing the temperature distribution on the heat exchange surface, in order to estimate the thickness of the frost layer, it is necessary to analyze the gradient variation characteristics of the surface heat distribution. The spatial fitting algorithm for the heat distribution gradient is implemented through the following steps:

[0074] First, obtain the two-dimensional grayscale matrix of the thermal image and convert the grayscale value of each pixel into a temperature value (through the sensor calibration curve);

[0075] Then, the temperature difference between adjacent pixels is calculated in the matrix in row or column direction to construct a temperature gradient map;

[0076] Then select the target area (such as the entire row or column) for fitting modeling, and you can choose linear function, polynomial function, piecewise fitting or spline interpolation;

[0077] Areas with sudden temperature gradient changes or forming a "temperature platform" structure are judged to be covered by frost;

[0078] A preliminary estimate of frost thickness is based on an empirical function. For example, when the gradient amplitude reaches a specific threshold and persists for more than a few pixels, combined with the condition that the average surface temperature in the area is lower than the dew point temperature, it can be determined that frost exists in the area. Calibration experiments are used to obtain a corresponding function between temperature difference and frost thickness, such as: frost thickness (in millimeters) = a×ΔT+b (a and b are experimental fitting coefficients), where ΔT is the temperature difference, to complete the numerical calculation of thickness.

[0079] During the model's comprehensive judgment phase, environmental parameter trends from historical operating cycles are incorporated. Temperature, humidity, and pressure curves are constructed using polynomial regression or spline interpolation methods and compared with the current real-time collected parameters. A comprehensive temperature and pressure judgment mechanism is established in conjunction with threshold rules. This mechanism determines that the platform's current pressure forecast, combined temperature and humidity, and estimated heat exchange surface frost thickness all meet the preset startup threshold requirements, indicating that the current run has reached the trigger condition for the first defrost.

[0080] The specific operation process is as follows:

[0081] Historical data collection and storage: The platform records environmental parameters such as outside temperature, relative humidity, air pressure, heat exchange surface temperature and frost layer estimation during each operation cycle to form a time series data set.

[0082] Trend fitting modeling method: The system performs regression modeling on the above multidimensional environmental variables. Optional methods include: polynomial regression (usually second-order or third-order) is used to fit the fluctuation trend;

[0083] Spline interpolation (such as cubic spline) is used to smooth continuously changing curves;

[0084] Moving average or weighted sliding average is used to suppress abnormal fluctuations.

[0085] Comparative analysis of real-time data and fitting curves: The currently collected temperature, humidity, air pressure and other values ​​will be compared with the trend curve to extract the rate of change, direction of change and deviation value to identify whether there is an abnormal sudden change or continuous deterioration of the environment.

[0086] Critical condition setting and identification: The system presets a temperature and pressure energy combined critical start threshold, which is a parameter combination judgment boundary established based on a large number of actual measurements and simulation experiments. Specifically including:

[0087] The air pressure is lower than a certain absolute value (e.g. 5,000 Pa);

[0088] The relative humidity exceeds a certain percentage (such as 70);

[0089] The surface frost thickness is estimated to be more than one millimeter;

[0090] The temperature difference in the frost area exceeds a certain value (such as three degrees Celsius);

[0091] Any two of the above three conditions are met continuously for more than a certain period of time (for example, twenty minutes).

[0092] If all conditions jointly meet the threshold, it is determined that the first defrost start condition of this operation cycle is met, and the system will automatically enter the main defrost control process.

[0093] For example, the platform is operating at an altitude of 11,000 meters, the measured outside temperature is minus 40 degrees Celsius, the humidity is 50 percent, and the platform attitude maintains a small pitch angle. However, because the temperature of the condensing heat exchange surface has been below minus 5 degrees Celsius for 30 consecutive minutes and the estimated frost thickness is greater than one millimeter, and the predicted air pressure is close to one-quarter of an atmosphere, the system determines based on the model that it has entered a high-risk range for frost formation and automatically determines that the critical point for starting defrost has been reached.

[0094] The independent heating cycle is designed to be physically isolated from the cooling cycle by a closed pipe path. The heat transfer medium filled inside is selected from carbon dioxide, freon, or air, and a built-in multi-stage micro pump is used to achieve controllable variable speed flow.

[0095] The heating path is pre-set with multiple heat exchange branch sections, each of which has a one-way heat conduction performance to prevent heat from flowing back into the main heat exchange condensation side;

[0096] In the control strategy, the average rate of latent heat release per unit length of heat medium under the current environment is calculated, and the flow rate and heating time are dynamically selected by matching the current estimated frost thickness of the heat exchange surface.

[0097] The direction of heat medium flow is adjusted in real time according to the platform posture information, and is preferentially guided to the branch section in the favorable direction of gravity defrosting, so as to improve the frost peeling efficiency and reduce the peak energy consumption per unit time.

[0098] To ensure thermal isolation, directional heating, and energy efficiency during the latent heat defrost process, a separate heating circuit layout was proposed. This circuit is physically isolated from the cooling heat cycle, preventing heat backflow and energy interference during operation, thereby improving the thermal management stability of the entire platform.

[0099] The core of the structure consists of a closed pipe path, which serves as the main channel of the heating circuit and is filled with a pre-selected heat medium. This heat medium can be selected from at least one of carbon dioxide, freon, or air. The choice of heat medium is determined by the specific platform operating environment. For example, in high-altitude, low-pressure environments, carbon dioxide is more suitable for short-term, high-efficiency heat release due to its high phase change latent heat and density regulation capabilities. Freon is suitable for low-temperature and low-vibration environments, and air is suitable for high-redundancy environments.

[0100] The heat medium flows through this closed path via a built-in multi-stage micropump with controllable variable speed regulation. This regulation, based on platform control logic input signals, allows for varying flow rates during different heating phases, meeting refined thermal control requirements. For example, a higher flow rate can be set during the initial defrost phase for rapid heating, while a lower flow rate can be set during the later maintenance phase for energy-efficient operation.

[0101] This heating path is pre-configured with multiple heat exchange branch segments. Each branch segment is an independent structure with unidirectional heat conduction performance. This can be achieved by integrating heat flow rectifiers into the branch channels or using a heterogeneous material interface structure. For example, a branch segment can use a high thermal conductivity material (such as copper or aluminum) at one end and a low thermal conductivity blocking layer (such as a ceramic coating or aerogel lining) at the other end. This creates a unidirectional heat flow effect, effectively preventing heat from flowing back to the main heat exchange condensation side during the defrost process, ensuring heat exchange stability.

[0102] In terms of control strategy, in order to achieve precise heat release regulation, it is necessary to calculate the average rate at which latent heat is released per unit length of heat medium under the current environment.

[0103] One of the two typical thermal power expressions in the existing thermal technology field can be used:

[0104] The first method is the total heat model consisting of the sum of sensible heat and latent heat. This model assumes that when the heat medium flows in the pipeline, it will release heat in two ways: one is the release of sensible heat through temperature drop, and the other is the release of latent heat if a phase change occurs (such as gas to liquid or vice versa).

[0105] The specific heat calculation method is as follows: First, measure the mass flow rate of the heat medium flowing through the heating section per unit time (for example, using a flow meter), in kilograms per second. Then, measure the temperature difference between the heat medium entering and leaving the pipe section (i.e., inlet temperature minus outlet temperature). Consult a thermodynamic property table for each heat medium (such as carbon dioxide, Freon, or air) to determine the specific heat capacity (the amount of heat required per unit mass to increase by one degree). Finally, if the heat medium undergoes a gas-liquid phase transition (such as supercritical carbon dioxide condensation), consult a table to determine the latent heat per unit mass of the heat medium. Adding these two heat components yields the total heat released per unit time and per unit length of the pipe (thermal power). This method is suitable for scenarios where the heat medium may experience both temperature changes and phase changes within the heating section, and is particularly well-suited for operating scenarios using carbon dioxide or heat media with condensing properties.

[0106] The second approach considers only the sensible heat model. If the heat medium in the control strategy primarily changes temperature without phase change (such as using air or sealed gaseous Freon), the latent heat term can be omitted, and only the sensible heat transfer from high temperature to low temperature can be calculated. This approach is more suitable for resource-constrained applications, short pipeline space, or limited heat medium flow rate, and is particularly suitable for lightweight platforms or temporary defrost strategies.

[0107] The first approach is recommended when the heat medium has significant phase change potential (such as supercritical carbon dioxide); the second approach is recommended when the heat medium is gaseous and remains single-phase throughout the defrost process. Both approaches are currently available in the field of fundamental thermodynamics. The choice of approach depends on the type of heat medium and the platform's thermodynamic objectives. Both approaches are supported by standardized data sources and readily available calculation tables.

[0108] To estimate the total energy required to melt the current frost layer on the heat exchange surface, the mass-latent heat model used for phase change calculations in existing thermodynamic knowledge can be used based on the frost layer thickness and the thermophysical properties of ice. This model is a standard calculation method in both physics and engineering thermodynamics.

[0109] The specific steps are as follows: First, use the infrared thermal imaging fitting algorithm to estimate the average thickness of the frost layer in different areas of the heat exchange surface, generally in millimeters or meters; then, based on the standard physical properties, use the density value of ice near zero degrees Celsius, which is generally about 917 kilograms per cubic meter; then look up the latent heat of melting of ice, that is, the energy required for unit mass of ice to change from solid to liquid, which is usually about 334,000 joules per kilogram; multiply the frost layer thickness (unit volume) by the density to obtain the mass, and then multiply it by the latent heat of melting per unit mass to obtain the energy required for the entire defrost area.

[0110] For example, if the average frost thickness in a certain area is one millimeter and the area is one square meter, the total volume is one liter; multiplied by the density of ice, which is about 0.917 kilograms, the theoretical minimum heat required to melt the frost layer in the area is about 300,000 joules.

[0111] The system control logic uses the average rate at which latent heat is released per unit length of heat medium under the current environment and the corresponding relationship between the total melting energy requirement of the current heat exchange cream layer as the judgment basis. It dynamically adjusts the micropump flow rate and heat medium heating time according to preset rules in different time periods to ensure that the heat supply meets the defrosting requirements without causing energy waste.

[0112] For example, the micropump flow rate is selected by the control logic based on the balance between the current required heat and the allowed heat loss:

[0113] If the current frost thickness is large and the ambient temperature remains low, the frost layer is expected to be difficult to melt. The system will increase the flow rate, increase the heat medium flux per unit time, and improve the heat transfer rate.

[0114] If the current energy status is sufficient and the platform has no high-priority tasks, the high flow rate mode can be set;

[0115] If the frost layer is thin and the platform energy is limited, the flow rate will be lowered to maintain only low intensity heating to prevent heat waste.

[0116] The specific flow rate adjustment level can be set to three gears: high, medium and low. The mass flow rate change between each gear can be controlled in the range of tens to hundreds of grams per second (depending on the heat medium and pipe size).

[0117] The control logic formula is as follows: Judgment condition 1: required melting heat; Judgment condition 2: heat medium release capacity per unit time; if condition 1 > condition 2 → increase flow rate; if condition 1 < condition 2 → maintain or reduce flow rate.

[0118] The setting of the heating time window depends on whether the system estimates that the total heat required for melting can be released within one heating cycle after the flow rate per unit time is determined.

[0119] If the micro pump flow rate has been adjusted to the maximum and still cannot meet the heat demand, the heating time can be extended;

[0120] If the platform's energy supply is at a medium level, staged heating is adopted, and after each heating, the "cooling-feedback-judgment" stage is entered;

[0121] If the predicted remaining frost layer can be defrosted within a short period of high flow rate, the heating time is set to the minimum value for quick processing.

[0122] During the specific execution process, the control unit sets an initial time window (such as 30 seconds) before each heating starts. If it is judged that it is not completed after the cycle ends, it enters the extension phase or continues with the next cycle.

[0123] For example, during a platform's operating cycle, the current frost layer thickness is 2 mm; the ambient temperature is -20°C; the redundant energy source is operating normally; the micropump's current flow rate is 100 grams of carbon dioxide per second; and according to calculations, the heat released per second is approximately 5,000 joules, resulting in a total defrosting heat requirement of approximately 300,000 joules. The control logic will determine that continuous heating is required for approximately 60 seconds. However, to avoid thermal shock and energy waste, the system will perform two heating cycles, each 30 seconds long, with a 10-second rest in between, forming an intermittent "add-stop-add" thermal control strategy. If, at the end of the second cycle, the sensor detects that the surface has met the preset requirements, the defrost task is considered complete and heating is discontinued.

[0124] Furthermore, the direction of heat transfer flow is not fixed but is controlled in real time based on the platform's attitude information. The platform's attitude is determined by corrected angle information, which indicates its current angle with the direction of gravity in both pitch and roll. The control logic calculates the angle by multiplying the gravity vector with the direction vectors of each branch heat exchange tube, prioritizing activation of the branch with the smallest angle with gravity. This ensures that heat is preferentially input to areas that are optimally positioned for natural defrosting, improving frost removal efficiency.

[0125] To illustrate, let's assume the platform is currently tilted slightly forward (at a pitch angle of five degrees), and the heat map shows a thick layer of frost on the front heat exchange surface. The control logic determines that the angle between the gravity vector and the direction of the front heat exchange tubes is only 20 degrees, while the angle in the rest of the area is over 40 degrees. Therefore, it prioritizes heating the front branches. This dynamic control approach not only improves defrosting efficiency but also significantly reduces peak energy consumption per unit time, ensuring energy balance across the entire platform.

[0126] In summary, the structural layout and control strategy achieves the comprehensive goals of heat path isolation, heat flow orientation, and load adaptive regulation, ensuring that the latent heat defrosting process is efficient, stable, and energy-efficient in a complex platform environment.

[0127] The gravity direction identification strategy used to determine whether the heating target area is in a favorable position for natural frost stripping involves vector projection of the heat exchange tube layout direction under the platform's three-dimensional attitude angles (pitch, yaw, and roll). Using the gravity vector as a reference, the angle between the heat exchange tube axis and the gravity vector is calculated to determine the most favorable potential frost stripping direction.

[0128] When the angle is less than the set first angle threshold, the heat exchange section is considered to have the ability to naturally defrost; if all areas do not meet the requirements, it enters a waiting state until the next cycle environmental parameters are updated and re-evaluated for activation.

[0129] During operation, the platform will continuously undergo attitude changes such as pitch, yaw, and roll. These angle information is collected by multiple inertial measurement units (IMUs). After averaging filtering and center of mass offset correction, the current platform's attitude angle data set in three-dimensional space is obtained, including: pitch angle (the elevation or depression angle in the front-to-back direction); yaw angle (horizontal rotation angle); and roll angle (left-right tilt angle).

[0130] Heat exchange tubes are arranged in a structure with their own spatial orientations. These orientations can be preset as a set of fixed direction vectors, often defined in the platform's coordinate system, such as the forward X-axis, the left Y-axis, and the upward Z-axis. In the control logic, the axial direction of each heat exchange tube segment is represented by a unit vector, denoted as the "heat exchange axis vector." After each attitude angle update, the spatial projection of these axis vectors relative to the direction of gravity at the current platform attitude is calculated.

[0131] The direction of gravity is always toward the center of the Earth in the Earth's coordinate system. However, its representation in the platform's local coordinate system changes as the platform's attitude changes. Based on the current pitch and roll angles, the global gravity vector (normally pointing vertically downward) is transformed into the gravity vector in the platform's local coordinate system using a direction cosine transformation matrix or quaternion conversion.

[0132] At this point, the angle between the axis vector of each heat exchange tube segment and the transformed gravity vector can be calculated using the vector angle formula: cosine of the vector angle = dot product of the two vectors / product of the two vector modules. Since both vectors are unit vectors, simply calculating their dot product yields the cosine of the angle, which can then be calculated using the inverse cosine function to determine the actual angle. This angle is the angle between the direction of the heat exchange tube segment and the direction of gravity, and reflects whether the frost layer will naturally slide off under the influence of gravity.

[0133] The control logic presets a first angle threshold, for example, 30 degrees, as the boundary for determining whether natural defrosting is possible. When the angle between a heat exchange tube segment and the direction of gravity is less than this threshold, it indicates that the tube segment is generally facing downward, in line with the gravity debonding direction, and is therefore considered to have the advantage of natural defrosting.

[0134] During each judgment cycle, the control logic traverses all heat exchange tube sections, selecting those with angles less than a threshold. The sections with the lowest temperature or thickest frost are prioritized for initial heating. If multiple sections meet these criteria, they are further prioritized by pre-set priorities (e.g., front and rear, energy load balance, etc.).

[0135] For example, a floating platform has ten heat exchange tube segments arranged along its front-to-back heat exchange surface, numbered one through ten, arranged parallel to the platform's X-axis. The current pitch angle is positive 25 degrees (front end up), the roll angle is zero, and the platform is tilted backward. In this attitude, calculations show that the first five heat exchange tube segments are at an angle of more than 50 degrees with the direction of gravity, while the last five are at an angle of approximately 25 degrees. The control logic compares the last half of the segments and finds that they meet the angle requirement of less than 30 degrees, thus determining that they possess an advantage in natural defrosting. Therefore, the control logic selects the ninth segment in the rear row, with the thickest frost, as the current heating target, initiating a latent heat heating cycle. If the platform's attitude continues to change during flight, the system will periodically refresh the angle determination to ensure that the next round of defrosting control continues to select the most favorable direction.

[0136] The thermal modeling algorithm used to predict the latent heat transfer rate includes the linked calculation of the heat medium flow rate, initial temperature, ambient temperature-humidity ratio and current frost layer thickness. It adopts one of a group of classic heat transfer models for modeling, including a one-dimensional steady-state heat conduction model, a non-steady-state phase change model or a thermal resistance network model. By calculating the deviation between the latent heat release value per unit time and the total latent melting demand of the frost layer, the predicted defrost time period is dynamically adjusted.

[0137] The input variables of the thermal modeling algorithm include four key parameters: heat medium flow rate, initial temperature, ambient temperature-humidity ratio, and current frost layer thickness. The heat medium flow rate is adjusted by a micropump and fed back in real time by a flow sensor, which is the basis for determining the mass flow of heat medium per unit time. The initial temperature is set by the heat medium heating device and collected by the sensor, which is used to calculate the release potential of the heat medium's sensible heat and latent heat. The ambient temperature-humidity ratio is used to determine the heat transfer impedance and evaporation effect during the external heat exchange process. The current frost layer thickness is obtained by fitting the thermal map data and is the key basis for calculating the target heat demand for defrosting.

[0138] The one-dimensional steady-state heat conduction model is suitable for modeling the heating behavior of a heat medium under conditions of constant flow rate and constant temperature difference. The model assumes that the heat flow is conducted in a single radial direction along the pipe wall, ignoring the effects of time changes and phase changes. The main calculation ideas are as follows: the temperature difference between the contact surface of the heat medium and the frost layer is used as the driving force; combined with the thermal conductivity constant (obtained from the table based on the heat medium and wall material); assuming that the frost layer is a uniform ice layer of equal thickness, Fourier's heat conduction law is used for estimation. In the calculation of the heat transfer rate per unit length, it can be expressed as:

[0139] Heat flux density = thermal conductivity × temperature difference / heat transfer path length; multiplying heat flux density by contact area and time gives the heat transferred by the heat medium over a specific period of time. This calculation is applicable during stable platform flight, the initial heating phase, and when the frost layer has high thermal conductivity.

[0140] Unsteady-State Phase Change Model: When the heat medium undergoes a gas-liquid phase transition during heat transfer, or when the frost layer partially melts after heating, resulting in unstable temperature changes, an unsteady-state heat transfer model can be used. This model considers temperature variations over time and combines the specific heat capacity and latent heat of fusion of ice with ambient temperature fluctuations to calculate the change in heat demand as the frost layer warms from its initial temperature to melting. The model logic includes dividing the frost layer into several layers, with each layer gradually increasing in temperature; superimposing the required heat on each layer, accounting for the time-varying heat input from the heat medium; and incorporating the latent heat released during the phase change to form a nonlinear energy transfer model. This method is suitable for scenarios with thick frost layers, long heating times, and phase changes in the heat medium (such as carbon dioxide condensation).

[0141] The thermal resistance network model simulates heat transfer paths by combining multiple thermal resistance structures in series and parallel. It can be used to estimate overall heat flow in complex structures. Each layer (e.g., heat medium-pipe wall-frost layer-external air) is considered a thermal resistance unit. The modeling method is as follows: the thermal resistance of each material segment is calculated based on its thermal conductivity, thickness, and area. The individual thermal resistances are combined in series and parallel to obtain an equivalent total thermal resistance. The heat flow is calculated using the Ohm analogy method based on the total thermal resistance and the temperature difference between the two ends. Heat flow = temperature difference / total thermal resistance. This model is suitable for situations with multiple layers of materials, non-uniform frost layers, and heat flow affected by multiple interfaces, such as pipes with scale layers or thermal insulation coatings.

[0142] Once the model is complete, the latent heat released by the heat medium per unit time is compared with the heat required for the target defrost area. The target heat is determined by the frost layer thickness, area, ice density, and latent heat of melting. The control logic uses the ratio of the calculated unit heat release rate to the total demand to determine the theoretical minimum defrost duration, while also allowing for an estimated safety margin.

[0143] For example: the heat medium can release 20,000 joules of heat per unit time; the current frost layer is estimated to require 100,000 joules of heat to melt; the control logic estimates that the theoretical heating time is 50 seconds, and the safety time is set to 60 seconds; if there are fluctuations in the heat medium, it can be set to two rounds of heating, each round of 30 seconds, with a cooling feedback interval of 10 seconds.

[0144] Assume the platform currently uses Freon as the heat medium, with an inlet temperature of zero degrees Celsius, an outlet temperature of -10 degrees Celsius, a flow rate of one hundred grams per second, a current frost layer thickness of one millimeter, and a coverage area of ​​0.5 square meters. A one-dimensional steady-state heat conduction model is selected for modeling. The thermal conductivity is approximately 0.2 watts per meter K, a temperature difference of ten degrees, and a path length of two millimeters. The calculated heat flux is one kilowatt per square meter, which, multiplied by the contact area, yields a heat release of five hundred watts per second. Based on ice density and latent heat, the total heat demand in the defrost area is approximately one hundred thousand joules. Therefore, the system determines that this defrost cycle will require approximately two hundred seconds of heating. The system will set four heating cycles of fifty seconds each, with a five-second cooling interval, to establish a stable defrost control rhythm.

[0145] The intermittent heating strategy uses a dynamic duty cycle control algorithm to set the on-off ratio within the heating cycle based on the current redundant energy capacity, mission load peaks and valleys, and the external air pressure fluctuation rate. The duty cycle is limited by the power control threshold and allows automatic adjustment at different stages.

[0146] The heating cycle is divided into two stages: activation period and cooling period. The activation period is carried out after the platform task management layer allocates available energy. The cooling period is adjusted according to the trend of external temperature changes to prevent condensation and refreezing during the defrosting process.

[0147] The so-called duty cycle refers to the ratio between the actual heating time and the total duration of a fixed heating cycle. In this strategy, the duty cycle is not set to a static value, but is calculated in real time based on three key dynamic variables:

[0148] Current redundant energy capacity: The platform's remaining energy resources currently available for defrosting. After task scheduling, the task management layer calculates and issues instructions to determine whether there is sufficient electricity or heat source for high-intensity heating in the current cycle.

[0149] Mission load peak and valley conditions: Determine the current mission energy consumption stage of the platform, such as whether communication, navigation, and attitude control are in a high energy consumption range. If in the mission load peak stage, the duty cycle should be reduced to reduce conflicts.

[0150] External air pressure fluctuation rate: When the air pressure in the external environment drops rapidly, it may be accompanied by changes in temperature and humidity. If frequent fluctuations or instability occur, a cooling period must be reserved to allow for the transition between heat release and condensation to avoid overheating due to sudden changes.

[0151] Based on these three parameters, the algorithm recalculates the recommended duty cycle each cycle and compares it to the preset power control threshold to ensure that heating does not cause platform energy overload. For example, if the platform's current redundant power is 200 watts, the mission load is low, and the air pressure fluctuation is mild, the control logic may set the duty cycle to 80, meaning that heating is active for 48 seconds per minute and cooling for 12 seconds.

[0152] In each heating cycle, time is divided into two phases:

[0153] Activation Period: During this phase, the platform's task management layer has completed energy scheduling and clearly allocated available energy to the defrost process. The control logic then activates the micropump and directs the heat medium into the target heat exchanger segment, performing the heating operation. The length of the activation period is determined by the duty cycle calculation results described above and can vary from cycle to cycle.

[0154] Cooling Period: After heating, the system enters a thermally controlled cooling phase, during which the heat transfer medium stops flowing and the heat exchange surfaces no longer receive new heat input. This cooling period serves not only as an energy conservation strategy but also as a way to prevent condensation and refreezing.

[0155] The cooling period is not fixed but adjusted flexibly based on external temperature trends. When the platform's external temperature rises rapidly, causing the frost to melt and evaporate again, the cooling period can be shortened. Conversely, if the temperature drops sharply, the cooling period should be extended to prevent residual heat from causing recondensation and forming "re-frost."

[0156] For example, during one operational cycle, the platform is currently located at high altitude. Task scheduling data indicates a surplus power of 300 watts, an ambient temperature of -10°C, and stable humidity. At this point, defrost control enters a single cycle: the control logic sets a total cycle length of 60 seconds. Based on calculations, the duty cycle is set to 70, resulting in an activation period of 42 seconds and a cooldown period of 18 seconds. With a heat transfer rate of 100 grams per second and a temperature difference of 10°C, the estimated heat release is approximately 5,000 joules per second. This total release of approximately 210,000 joules over 42 seconds covers the required heat flux to melt an area of ​​0.5 square meters with a 1mm thick frost layer. During the subsequent cooldown period, the heat exchange surface temperature is monitored by a thermal sensor to ensure no significant risk of refreezing. If residual heat is still detected at the cooling terminal, the duty cycle is reduced to 60 for the next heating cycle to prevent overheating.

[0157] The introduction of an intermittent heating strategy significantly improves energy flexibility and thermal control safety. Through dynamic duty cycle calculation and climate-adaptive adjustment of cooling cycles, the platform accurately balances defrost efficiency and energy consumption in highly variable high-altitude environments, effectively reducing the risk of residual frost and the probability of energy waste.

[0158] The defrost status recording strategy for pausing the heating process includes synchronously recording the currently heated area, the input heat energy, the average temperature rise of the heat exchange surface, and the timestamp of the current heating time point. This recorded information forms a defrost breakpoint status structure and is written into the platform task data storage area;

[0159] When energy is restored, the system first reads the previous breakpoint information and estimates the current required compensation heat based on the unfinished defrosting area, the heated energy consumption, and the remaining thickness of the frost layer. This is then compared with the latent heat release rate of the heat medium to determine whether the system has the capability to execute the function.

[0160] If the conditions for continuing execution are met, the heating process of the corresponding heat exchange tube section is directly resumed, otherwise it is delayed to enter the next cycle.

[0161] When energy shortage or priority scheduling conflicts occur during the execution of the heating task, causing the heating process to be suspended, the control logic automatically extracts and saves the following key status information to form a "defrost breakpoint status structure" and writes it to the platform task data storage area. Specifically, it includes:

[0162] Currently heated area: records the heat exchange pipe section number and spatial location that has completed the heating operation, so as to facilitate the determination of the specific areas that have been processed and unprocessed when continuing the subsequent tasks.

[0163] Thermal energy invested: The total amount of heat released is calculated based on the type of heat medium, flow rate, temperature difference and heating time. The unit is joules, which is used to compare the subsequent compensation heat required.

[0164] Average temperature rise of the heat exchange surface: The temperature difference between the initial temperature and the current temperature of the heating area is obtained through a bonded temperature sensor, and the weighted average value of the local area is taken to reflect the defrost progress.

[0165] Timestamp of the current heating point: A high-precision clock module records the time stamp of heating interruptions, facilitating correlation with task queue priorities and synchronization information, and supporting delay strategy execution judgment. This state structure is written to the platform task data storage area in real time to ensure data is not lost due to power outages.

[0166] When the platform energy condition is restored, the control logic will trigger the breakpoint re-evaluation mechanism, and first read the defrost breakpoint status structure written during the most recent interruption from the task data storage area, and perform the following operations based on the completed status and remaining defrost target recorded therein:

[0167] Determine the unfinished defrost area: Compare the "heated area" in the structure with the "planned defrost area" in the current heat exchange tube layout table to identify the pipe sections that have not been processed or have been insufficiently processed.

[0168] Estimate the current required heat compensation: Calculate the total energy required based on the remaining frost thickness (estimated using the infrared heat map) and the area of ​​the unheated area. Refer to the heat demand calculation model: Defrosting heat requirement = remaining area × frost thickness × ice density × latent heat of melting.

[0169] Compare the latent heat release rate of the heat medium: read the current allowed heating time, the current inlet temperature and flow rate of the heat medium from the heat medium control parameters, and combine them with the aforementioned thermal modeling algorithm to estimate the heat release capacity per unit time.

[0170] Determine whether the execution conditions are met: If the current platform energy can support the defrosting heat requirements of the remaining areas, that is, the execution conditions are met, the recovery heating process will be entered; if not, the logic will enter the delay mode and will be evaluated again when the next judgment cycle is triggered.

[0171] If the conditions for continuing execution are met, the control logic automatically uses the "heated area" in the breakpoint structure as the starting point, sequentially selecting the target pipe segment from the unfinished area to resume the heat medium injection and heating process. The control process recalculates the time allocation in the thermal model and compensates for the heat difference in the unfinished section, ensuring the complete defrost thermal closed loop.

[0172] For example: in the previous round, the heating of pipe sections 1 to 3 has been completed, with a total heat input of 400,000 joules; the remaining pipe sections 4 to 5 have not been heated, and the infrared thermal map determines that the frost thickness in this area is about one millimeter; the on-demand melting heat is about 300,000 joules; the current platform has a heat medium release capacity of about 100,000 joules per minute; calculation results show that the task can be completed within three minutes; the control logic resumes the heating task, setting three heating cycles, each of 60 seconds, with 15 seconds of cooling in between, to complete the defrost compensation task.

[0173] The platform's redundant energy capacity is obtained by comparing the platform's current task power demand with the energy supply capacity in real time, and determining whether to allow the activation of the heating operation based on the redundant capacity threshold. The heating control sub-process is only executed when the platform's task scheduling mechanism sends a non-interference signal.

[0174] The "heating control sub-process" refers to a specific heating task scheduled by the defrost control logic after trigger conditions are met (such as reaching the defrost threshold, sufficient energy, and appropriate posture and orientation). It is a conditionally driven sub-action unit within the entire defrost process. Assume that the entire "latent heat defrost intelligent control process" is divided into: environmental assessment → energy assessment → defrost initiation → heating control sub-process → status recording → resumption or termination. The heating control sub-process refers to the actual heating operation, including the actual steps of activating the heat transfer medium, controlling on / off, monitoring temperature, and calculating time and power.

[0175] The platform's energy supply capacity is provided by the Energy Core Module, which can be a solar panel, chemical battery pack, or other renewable energy storage unit. This capacity is calculated in real time based on the following parameters: current power output capacity (e.g., the product of voltage and current); remaining charge in the energy storage device; energy-to-load conversion efficiency (taking into account energy losses from the energy storage unit to the terminal execution structure); and the impact of temperature on battery performance (particularly critical on high-altitude platforms). In practice, the energy supply capacity is automatically converted to the platform's current maximum available power output, for example, 300 watts.

[0176] Subtracting the total power requirement of the current task from the platform's current maximum available power yields the platform's redundant energy capacity available for non-core tasks. This value serves as the key basis for determining whether the heating task is executable.

[0177] A preset redundant capacity threshold is defined as the minimum amount of energy required to execute the defrost task. The specific value is determined by the heat medium type, current flow rate, heating duration, and heat demand. For example, if the micropump and heating module require a combined 100 watts and must run for at least 30 seconds to achieve the initial defrost effect, the redundant capacity threshold is set to 100 watts. The judgment process is as follows: If the redundant energy capacity ≥ the preset threshold, the heating execution conditions are determined to be met; if the redundant energy capacity is less than the preset threshold, the heating task is paused or delayed. For example, if the current energy supply capacity is 300 watts, the total task power demand is 190 watts, and the redundant energy capacity is 110 watts, then if the threshold is set to 100 watts, the heating task is determined to be executable.

[0178] To prevent the heating process from interfering with high-priority tasks (such as maintaining communications and navigation updates), the activation of the heating control sub-process must also meet a "non-interference signal" condition. This signal, generated by the task scheduling mechanism, is issued based on the following criteria: whether a high-priority task is currently scheduled for execution; whether the platform is currently in a transitional state (such as maneuvering, trajectory change, or communication window); and whether there are energy consumption warnings or energy switching processes. Only if all judgments are clear will the heating control logic enter execution. This mechanism prevents the defrost task from being forcibly activated during task switching, intense platform motion, or energy fluctuations, thereby improving platform operational safety.

[0179] Assume the platform is currently equipped with solar panels with a total output power of 250 watts. Communications, attitude control, and navigation utilize a total of 180 watts, leaving 70 watts of redundant capacity. The current defrost task requires the heating module to consume at least 90 watts of power. The control logic determines that the redundant capacity threshold is not met and refuses to activate the heating process. Simultaneously, the task scheduler determines that communication will require high-power transmission within the next minute, marking it as a high-interference period. Even if redundant power is restored later, defrost will continue to be delayed due to the interference signal. When the solar panels provide 320 watts of output in the next cycle, the redundant power reaches 140 watts, and there are no high-priority tasks scheduled, the system's control logic receives an "actionable signal" and immediately activates the heating process, entering the first stage of the heating cycle.

[0180] The completion criteria for closed-loop defrost control are determined based on the heat exchange surface temperature recovery rate, frost shedding detection results, and statistical results of the remaining untreated area, and are determined based on meeting the termination threshold.

[0181] The surface temperature is detected in real time by multiple bonded temperature sensors and compared with the initial unfrosted temperature to calculate whether the heating rate is stable and tends to zero;

[0182] The frost layer is removed by micro-vibration acceleration feedback to determine whether physical removal is completed;

[0183] The distribution of untreated areas is distinguished by comparing the temperature difference before and after treatment based on thermal image recognition;

[0184] When all three indicators reach the preset threshold, the defrost is automatically determined to be completed and the task record is cleared.

[0185] During the heating process, the platform uses multiple bonded temperature sensors to collect real-time data on the surface temperature of the heat exchange tubes. Each sensor is attached to a key heating area to record the temperature curve.

[0186] The system's control logic compares the current temperature recorded by the sensor with the initial, unfrosted temperature and calculates the rate of temperature rise per unit time, known as the "temperature recovery rate." If this rate continues to decrease and approaches zero, it indicates that the frost layer has been largely cleared, heat conduction is unimpeded, and heat input cannot further increase the surface temperature. At this point, the surface is nearing "defrost completion." For example, if the initial unfrosted temperature is 0°C, the current temperature is -1°C, and the temperature rise has been less than 0.2°C over the past 30 seconds, and the recovery rate termination threshold is set at 0.1°C / minute, the temperature rise is considered "saturated." This determination is a continuous dynamic comparison, and generally requires a low rate of change for two consecutive observation periods (e.g., 15-second periods) to be considered achieved.

[0187] To determine whether the frost has physically detached, the present invention employs micro-vibration acceleration feedback. By integrating miniature accelerometers in key locations within the heat exchange structure, these sensors sense the minute structural vibrations or pulses caused by frost shedding. When heating causes the frost to transition from solid to liquid or gaseous state, changes in its mass and adhesion cause tiny release vibrations in the local structure. This signal is captured by the accelerometer and filtered to remove background interference (such as platform attitude disturbances), extracting an impulse response with typical frequency characteristics. The control logic determines that if a micro-vibration signal within a specific peak frequency range is detected and its match with a known "peeling response template" exceeds a preset recognition threshold (e.g., a similarity exceeding 80%), the region is deemed to have physically detached. This method can determine the actual defrosting effect without relying on image recognition, making it particularly suitable for feedback control in high-altitude, low-light environments or when equipment structures are enclosed.

[0188] The third criterion for defrosting completion is the counting of remaining untreated areas. This is primarily accomplished through thermal image recognition, where infrared thermal images are used to compare the temperature distribution of the same heat exchange area before and after heating. The control logic operates as follows: The current heat exchange surface thermal image is acquired; pixel-by-pixel matching is performed with the original thermal image taken before heating begins; areas with a temperature difference below a certain temperature rise threshold (e.g., less than one degree Celsius) are identified as "unresponsive areas"; and the ratio of their total area to the total area of ​​the heated area is calculated.

[0189] If the ratio is less than the termination threshold (such as five percent), it is judged that defrosting has been basically completed; if the ratio is higher than the threshold and there is no micro-vibration response or temperature rise in the corresponding area, there may be local heating failure or defrosting failure, which will be marked as the next round of priority processing area.

[0190] If all three of these criteria meet the preset termination thresholds, the control logic automatically determines that the current closed-loop defrost process is complete. This immediately terminates the heating process, clears the current defrost task schedule, releases the energy scheduling lock, sends a "defrost completion signal" to the task management mechanism, ensures platform energy resource recovery, updates task scheduling status, and prepares for the next defrost cycle.

[0191] During one defrost cycle, the platform selected three central heat exchange tubes for heating, with a total heating time of 120 seconds. The temperature sensor recorded a temperature rise rate consistently below 0.1°C / minute for the last two cycles; the accelerometer detected typical spalling pulse responses twice at the end positions; and infrared thermal image comparison revealed that the unresponsive area accounted for only 3 percent of the total area. Based on these results, the control logic determined that all three conditions had been met and immediately shut off the heat transfer, entering the defrost completion state.

[0192] Example 2: A latent heat defrosting intelligent control system, comprising:

[0193] An environmental perception unit, which is used to establish a multi-parameter environmental perception model including air pressure, temperature, humidity, heat exchange surface state, and platform attitude angle, and to determine whether the critical conditions for starting defrost have been met based on the model;

[0194] A heat medium guiding unit is used to select at least one of carbon dioxide, freon or air as a heat medium and guide the heat medium to input latent heat to a predetermined heat exchange area through an independent heating circulation path isolated from the heat flow of the refrigeration system;

[0195] The directional heating unit is used to dynamically determine the direction of gravity in the heat exchange area formed by independently arranged heat exchange tubes based on the current platform posture angle, and prioritize heating the tube sections that are in a favorable position for natural frost stripping.

[0196] A power control unit is used to determine a target defrost time period based on the predicted phase change latent heat transfer rate during the heating process, and to perform intermittent heating on the selected heat exchange tube section, controlling the average power in each heating cycle to not exceed the current platform redundant energy capacity;

[0197] The status continuation control unit is used to suspend the heating process and record the completed defrosting time and area information when insufficient energy is detected, and continue the remaining defrosting process after energy is restored until the closed-loop defrost control is completed.

[0198] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0199] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0200] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0201] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0202] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A latent heat defrosting intelligent control method, characterized in that: The following steps are involved: In a high-altitude floating platform environment, a multi-parameter environmental perception model is established, including air pressure, temperature, humidity, heat exchange surface state, and platform attitude angle. Based on this model, it is determined whether the critical conditions for starting defrost have been met. At least one of carbon dioxide, freon or air is selected as the heat medium, and the heat medium is guided to input latent heat to the predetermined heat exchange area through an independent heating cycle path isolated from the heat flow of the refrigeration system; The heat exchange area is composed of independently arranged heat exchange tubes. The most favorable potential defrost direction is determined based on the angle between the heat exchange tube axis and the gravity vector. The tube section with the advantageous position in the natural defrost direction is selected to start heating first. During the heating process, the target defrost time period is determined based on the predicted phase change latent heat transfer rate, and intermittent heating is performed on the selected heat exchange tube sections, ensuring that the average power in each heating cycle does not exceed the current platform's redundant energy capacity. When insufficient energy is detected, the heating process is suspended and the completed defrost time and area information are recorded. After energy is restored, the remaining defrost process is continued until closed-loop defrost control is completed.

2. A latent heat defrosting intelligent control method according to claim 1, characterized in that: The construction process of the multi-parameter environmental perception model includes the dynamic coupling calculation of the high-altitude ambient air pressure and the relative motion state of the platform. The air pressure information is continuously collected from the platform's current altitude, latitude, and atmospheric temperature. The empirical pressure-altitude formula and one of the moist air state equations are used to derive the instantaneous air pressure prediction value. The platform attitude angle information is realized by averaging the filtered values ​​of multiple inertial measurement units and is corrected by combining the actual platform load offset to form a corrected angle model. The heat exchange surface status information is continuously scanned and acquired by a micro-infrared temperature difference sensor, and the surface frost thickness is estimated based on the grayscale fitting function of the thermal map data. This multi-parameter environmental perception model is based on the fitting curve of the environmental change trend in the historical operation cycle, and combines the current collection results to perform preset multi-target predictions to determine whether the temperature and pressure energy state of the critical point for triggering the first defrost has been reached.

3. The latent heat defrosting intelligent control method according to claim 2, characterized in that: The independent heating cycle is designed to be physically isolated from the cooling cycle by a closed pipe path. The heat transfer medium filled inside is selected from carbon dioxide, freon, or air, and a built-in multi-stage micro pump is used to achieve controllable variable speed flow. The heating cycle path is pre-set with multiple heat exchange branch sections, each of which has one-way heat conduction performance to prevent heat from flowing back into the main heat exchange condensation side; In the control strategy, the average rate of latent heat release per unit length of heat medium under the current environment is calculated, and the flow rate and heating time are dynamically selected by matching the current estimated frost thickness of the heat exchange surface. The direction of heat medium flow is adjusted in real time according to the platform posture information, and is preferentially guided to the branch section in the favorable direction of gravity defrosting, so as to improve the frost peeling efficiency and reduce the peak energy consumption per unit time.

4. The latent heat defrosting intelligent control method according to claim 3, characterized in that: The gravity direction identification strategy used to determine whether the heating target area is in a favorable position for natural frost stripping involves vector projection of the heat exchange tube layout direction under the platform's three-dimensional attitude angles (pitch, yaw, and roll). Using the gravity vector as a reference, the angle between the heat exchange tube axis and the gravity vector is calculated to determine the most favorable potential frost stripping direction. When the angle is less than the set first angle threshold, the corresponding heat exchange branch section is considered to have natural defrosting capability; if all areas do not meet the requirements, the system enters a waiting state until the next cycle environmental parameters are updated and re-evaluated for activation.

5. The latent heat defrosting intelligent control method according to claim 4, characterized in that: The thermal modeling algorithm used to predict the latent heat transfer rate includes the linked calculation of the heat medium flow rate, initial temperature, ambient temperature-humidity ratio and current frost layer thickness. It adopts one of a group of classic heat transfer models for modeling, including a one-dimensional steady-state heat conduction model, a non-steady-state phase change model or a thermal resistance network model. By calculating the deviation between the latent heat release value per unit time and the total latent melting demand of the frost layer, the predicted defrost time period is dynamically adjusted.

6. The latent heat defrosting intelligent control method according to claim 5, characterized in that: The intermittent heating strategy uses a dynamic duty cycle control algorithm to set the on-off ratio within the heating cycle based on the current redundant energy capacity, mission load peaks and valleys, and the external air pressure fluctuation rate. The duty cycle is limited by the power control threshold and allows automatic adjustment at different stages. The heating cycle is divided into two stages: activation period and cooling period. The activation period is carried out after the platform task management layer allocates available energy. The cooling period is adjusted according to the trend of external temperature changes to prevent condensation and refreezing during the defrosting process.

7. The latent heat defrosting intelligent control method according to claim 6, characterized in that: The defrost status recording strategy for pausing the heating process includes synchronously recording the currently heated area, the input heat energy, the average temperature rise of the heat exchange surface, and the timestamp of the current heating time point. This recorded information forms a defrost breakpoint status structure and is written into the platform task data storage area; When energy is restored, the system first reads the previous breakpoint information and estimates the current required compensation heat based on the unfinished defrosting area, the heated energy consumption, and the remaining thickness of the frost layer. This is then compared with the latent heat release rate of the heat medium to determine whether the system has the capability to execute the function. If the conditions for continuing execution are met, the heating process of the corresponding heat exchange tube section is directly resumed, otherwise it is delayed to enter the next cycle.

8. The latent heat defrosting intelligent control method according to claim 7, characterized in that: The platform's redundant energy capacity is obtained by comparing the platform's current task power demand with the energy supply capacity in real time, and determining whether to allow the activation of the heating operation based on the redundant capacity threshold. The heating control sub-process is only executed when the platform's task scheduling mechanism sends a non-interference signal.

9. The latent heat defrosting intelligent control method according to claim 8, characterized in that: The completion criteria for closed-loop defrost control are determined based on the heat exchange surface temperature recovery rate, frost shedding detection results, and statistical results of the remaining untreated area, and are determined based on meeting the termination threshold. The surface temperature is detected in real time by multiple bonded temperature sensors and compared with the initial unfrosted temperature to calculate whether the heating rate is stable and tends to zero; The frost layer is removed by micro-vibration acceleration feedback to determine whether physical removal is completed; The distribution of untreated areas is distinguished by comparing the temperature difference before and after treatment based on thermal image recognition; When all three indicators reach the preset threshold, the defrost is automatically determined to be completed and the task record is cleared.

10. A latent heat defrost intelligent control system, used to implement a latent heat defrost intelligent control method according to any one of claims 1 to 9, characterized in that: include: An environmental perception unit, which is used to establish a multi-parameter environmental perception model including air pressure, temperature, humidity, heat exchange surface state, and platform attitude angle, and to determine whether the critical conditions for starting defrost have been met based on the model; A heat medium guiding unit is used to select at least one of carbon dioxide, freon or air as a heat medium and guide the heat medium to input latent heat to a predetermined heat exchange area through an independent heating circulation path isolated from the heat flow of the refrigeration system; The directional heating unit is used to determine the most favorable potential defrosting direction for the heat exchange area formed by the independently arranged heat exchange tubes on the structure based on the angle between the heat exchange tube axis and the gravity vector, and select the tube section with the advantageous position in the natural defrosting direction to start heating first; A power control unit is used to determine a target defrost time period based on the predicted phase change latent heat transfer rate during the heating process, and to perform intermittent heating on the selected heat exchange tube section, controlling the average power in each heating cycle to not exceed the current platform redundant energy capacity; The status continuation control unit is used to suspend the heating process and record the completed defrosting time and area information when insufficient energy is detected, and continue the remaining defrosting process after energy is restored until the closed-loop defrost control is completed.

Images