Intelligent control method and system for latent heat defrosting

Abstract

The invention discloses an intelligent control method and system for latent heat defrosting, and particularly relates to the technical field of latent heat defrosting control. The method comprises the following steps that the defrosting starting time is accurately judged by establishing a multi-parameter environment sensing model, and carbon dioxide or Freon or air is selected as a heating medium; latent heat is input into the heat exchange area through the independent heating circulation path; the gravity direction of the heat exchange pipe is dynamically judged according to the platform posture, and the pipe section in the natural frost stripping direction is preferentially heated; an intermittent heating strategy based on thermal modeling is adopted to control energy consumption; when the energy is insufficient, heating is suspended, the state is recorded, and the energy is recovered for subsequent defrosting; according to the method, accurate judgment of defrosting starting, directional and efficient heating control and whole-process closed-loop management are achieved, the energy utilization efficiency, the frost layer stripping effect and the control reliability are remarkably improved, intelligence, self-adaption and restorability are achieved, and the method is suitable for meeting the requirement for stable defrosting in energy-limited and complex-environment scenes such as a high-altitude platform.

Description

Technical Field

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

[0002] In operating environments such as high-altitude floating platforms, external condensation structures of aircraft, and unattended heat transfer equipment, due to the extremely low external environmental temperature and frequent humidity changes, a frost layer is extremely likely to form on the heat exchange surface. Once the frost layer accumulates, it will significantly reduce the heat exchange efficiency, increase energy consumption, and may affect the stability of the overall attitude control of the platform and the normal operation of mission equipment. In severe cases, it may even cause equipment failure. Therefore, ensuring that the heat exchange structure maintains a clean and unobstructed heat exchange state during operation is one of the keys to ensuring the long-term stable operation of high-altitude equipment.

[0003] In the prior art, methods such as heating defrosting, electrothermal film heating, compression heat reflux, and mechanical scraping are often used to treat the frosting area. However, these traditional methods generally have the following deficiencies: high energy consumption, the thermal control process cannot be dynamically adjusted, which easily causes overheating or ineffective heating; lack of adaptability to environmental changes, the defrosting trigger logic depends on manually set thresholds or fixed cycles, and cannot accurately respond to the dynamic changes of the complex high-altitude environment; during periods of limited energy or high-priority mission occupancy, it is impossible to effectively coordinate the resource allocation between the defrosting task and other tasks of the platform, the defrosting process is easily interrupted, and there is no recovery mechanism.

[0004] In addition, most traditional defrosting methods adopt strategies such as "overall heating" or "local periodic activation", and cannot dynamically optimize the heating area according to factors such as the direction of gravity and the difference in frost layer thickness, resulting in uneven heat distribution and low defrosting efficiency. At the same time, the judgment of the completion of defrosting also lacks accuracy, usually based on time measurement or empirical judgment, there is a risk of insufficient defrosting or redundant heating, affecting the operation safety of the equipment. Therefore, the present invention proposes an intelligent control method and system for latent heat defrosting in order to solve the above problems. Summary of the Invention

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

[0006] An intelligent control method for latent heat defrosting, comprising the following steps:

[0007] Establish a multi-parameter environmental perception model including air pressure, temperature, humidity, heat exchange surface state, and platform attitude angle, and judge whether the critical condition for starting defrosting is reached based on this model;

[0008] Select at least one of carbon dioxide, freon, or air as the heat medium, and guide the latent heat of the heat medium to the predetermined heat exchange area through an independent heating cycle path isolated from the heat flow of the refrigeration system;

[0009] The heat exchange area is composed of heat exchange tubes arranged independently in terms of structure, and the direction of gravity is dynamically judged according to the current platform attitude angle, and the pipe section in the advantageous position in the natural defrosting direction is preferentially started for heating;

[0010] During the heating process, the target defrosting time period is determined according to the predicted phase change latent heat transfer rate, and intermittent heating is performed on the selected heat exchange pipe section, and the average power within each heating cycle is controlled not to exceed the current platform redundant energy capacity;

[0011] When it is detected that the energy is insufficient, the heating process is paused and the defrosting duration and area information that have been completed are recorded, and the remaining defrosting process is continued after the energy is restored until the closed-loop defrosting control is completed.

[0012] In a preferred embodiment, the construction process of the multi-parameter environment perception model includes dynamic coupling calculation of the high-altitude environmental air pressure and the relative motion state of the platform. The air pressure information is obtained by continuously collecting the current altitude, latitude and atmospheric temperature of the platform, and is derived using one of the empirical pressure-height formula and the moist air state equation to obtain the instantaneous air pressure prediction value;

[0013] The platform attitude angle information is realized through the average filtered values of multiple inertial measurement units, and a corrected angle model is formed after being corrected in combination with the actual load offset of the platform;

[0014] The heat exchange surface state information is continuously scanned and obtained by a micro infrared temperature difference sensor, and the surface frost thickness is estimated according to the gray-scale fitting function for the heat map data;

[0015] This multi-parameter environment perception model fits the curve based on the environmental change trend in the historical operation cycle, and combines the current acquisition results for preset multi-objective prediction to judge whether the critical point temperature, pressure and energy state of the first defrost trigger has been reached.

[0016] In a preferred embodiment, the structural layout method of the independent heating cycle path includes setting a closed pipe path that is physically completely isolated from the refrigeration and heat cycle, and the heat transfer medium filled inside is selected from one of carbon dioxide, freon or air, and controllable variable-speed flow is realized through a built-in multi-stage micro pump;

[0017] This heating path is preset with multiple heat exchange branch sections, and each branch section 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, by calculating the average rate of latent heat released by the heat transfer medium per unit length in the current environment and matching it with the current estimated frost thickness on the heat exchange surface, the flow rate and heating duration are dynamically selected;

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

[0020] In a preferred embodiment, the gravity direction recognition strategy for judging whether the heating target area is in the advantageous position of natural defrosting includes vector projection of the layout direction of the heat exchange tubes under the three-dimensional attitude angles of the platform, namely pitch, yaw and roll, and using the gravity vector as a reference. By calculating the angle between the axis of the heat exchange tube and the gravity vector, the most favorable potential defrosting direction is determined;

[0021] When the angle is less than the set first angle threshold, it is considered that the heat exchange section has the ability of natural defrosting; if all areas do not meet the requirements, it enters the waiting state until the next cycle of environmental parameter update and then re-evaluates whether to activate.

[0022] In a preferred embodiment, the thermal modeling algorithm for predicting the latent heat transfer rate includes a combined calculation of the heat medium flow rate, initial temperature, environmental temperature-humidity ratio and the current frost layer thickness, and uses one of a set of classical heat transfer models for modeling, including one-dimensional steady-state heat conduction model, unsteady-state phase change model or 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 defrosting time period is dynamically adjusted.

[0023] The intermittent heating strategy adopts a dynamic duty cycle control algorithm, and sets the opening and closing ratio within the heating cycle according to the current redundant energy capacity, the peak and valley of the task load 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: the activation period and the cooling period. The activation period is carried out after the platform task management layer allocates the available energy, and the cooling period is adjusted according to the external air temperature change trend to prevent condensation and refreezing during the defrosting process.

[0025] In a preferred embodiment, the defrosting state recording strategy for suspending 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 time stamp of the current heating time point. The recorded information forms a defrosting breakpoint state structure and is written into the platform task data storage area;

[0026] When the energy is restored, first read the previous breakpoint information, and estimate the current required compensation heat according to the unfinished defrosting area, the heated energy consumption and the remaining thickness of the frost layer, and compare it with the latent heat release rate of the heat medium to judge whether it has the ability to execute;

[0027] If the condition for continuing to execute is met, directly resume the heating process of the corresponding heat exchange tube section, otherwise delay entering the next cycle.

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

[0029] In a preferred embodiment, the completion criterion of the closed-loop defrost control is jointly determined based on the heat transfer surface temperature recovery rate, the frost layer shedding detection result, and the statistical result of the remaining untreated area, and the determination is based on meeting the termination threshold;

[0030] The surface temperature is detected in real time by multiple conformable temperature-sensitive sensors and compared with the initial unfrosted temperature to calculate whether the heating rate is stably tending to zero;

[0031] The completion of the physical peeling of the frost layer is judged by the feedback of the micro-vibration acceleration;

[0032] The untreated area is distinguished according to the temperature difference before and after the comparison of the thermal images before and after the treatment;

[0033] When all three indicators reach the preset threshold, it is automatically determined that the defrosting is 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 for establishing a multi-parameter environmental perception model including air pressure, temperature, humidity, heat transfer surface state, and platform attitude angle, and determining whether the critical condition for starting defrosting is reached based on this model;

[0036] A heat medium guiding unit for selecting at least one of carbon dioxide, Freon, or air as the heat medium, and guiding the latent heat of the heat medium to the predetermined heat transfer area through an independent heating cycle path isolated from the heat flow of the refrigeration system;

[0037] A directional heating unit for dynamically determining the gravity direction according to the current platform attitude angle for the heat transfer area composed of heat exchange tubes independently arranged in structure, and preferentially starting heating for the tube sections in the advantageous position of the natural defrosting direction;

[0038] A power control unit for determining the target defrosting time period according to the predicted phase change latent heat transfer rate during the heating process, and performing intermittent heating on the selected heat exchange tube sections, and controlling the average power within each heating cycle not to exceed the current platform redundant energy capacity;

[0039] A status continuation control unit, which is used to pause the heating process when detecting insufficient energy, record the completed defrosting duration and area information, and continue to execute the remaining defrosting process after the energy is restored until the closed-loop defrosting control is completed.

[0040] The technical effects and advantages of the present invention:

[0041] By establishing a multi-parameter environmental perception model and combining air pressure, temperature and humidity, heat exchange surface status and platform attitude information, the present invention realizes accurate judgment of defrosting triggering, significantly improving the scientificity and timeliness of defrosting start. Traditional timed heating or single-parameter triggering mechanisms are prone to mis-start or response lag in high-altitude dynamic environments, resulting in energy waste or difficult-to-control frost layer accumulation. By integrating historical trends and real-time collection, the present invention judges whether the temperature-pressure-energy critical point is reached, avoiding the common problems of "early start, late start, mis-start" in the defrosting process from the root cause, ensuring that the heating control is only executed when necessary, and improving the energy scheduling efficiency and task response ability of the entire platform.

[0042] By combining an independent heating cycle path with a gravity direction recognition strategy, the present invention constructs a directional defrosting and area-priority heating mechanism, significantly improving the thermal efficiency utilization rate and reliability of the defrosting process. Compared with the global heat equalization mode of traditional defrosting methods, the present invention guides the heat medium to preferentially flow to the heat exchange pipe sections in the direction of natural frost peeling advantage, and dynamically adjusts the flow rate and heating time, so that the heat acts concentratedly on the areas more prone to defrosting, reducing ineffective heating and heat loss. At the same time, through a multi-stage micro pump and a one-way heat conduction structure, heat backflow is avoided, effectively improving the frost layer peeling rate per unit time, reducing the defrosting duration, and ensuring the long-term operation stability of the equipment.

[0043] By constructing a defrosting breakpoint status structure and a multi-index closed-loop completion judgment mechanism, the present invention ensures that the defrosting task has the full-process closed-loop control ability of interruptible recovery, process traceability and execution termination. In the case of insufficient energy or task conflict, the present invention can record core variables such as heating status, heat input, and temperature rise trend, and accurately resume the control process according to the remaining frost layer state after the energy is restored, avoiding repeated heating and resource waste. At the same time, when the defrosting is completed, the termination timing is jointly determined by three indicators: temperature recovery rate, micro-vibration response and heat map recognition, effectively preventing overheating and incomplete defrosting, and realizing a truly intelligent, efficient and controllable defrosting closed-loop control process. Description of the Drawings

[0044] For the convenience of those skilled in the art to understand, the present invention will be further described below in conjunction with the drawings;

[0045] Figure 1 It is a schematic diagram 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. Specific embodiments

[0047] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0048] Refer to Figure 1 - Figure 2 The following embodiments are obtained:

[0049] Embodiment 1:

[0050] The present invention relates to a latent heat defrosting intelligent control method for the high-altitude floating platform environment, aiming to achieve an efficient, controllable and reliable defrosting process under the conditions of extremely low air pressure, frequent attitude changes and limited energy. The method dynamically determines whether the defrosting start condition is met by constructing a multi-parameter environmental perception model based on air pressure, temperature, humidity, heat transfer surface state and platform attitude angle, and realizes the 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 transfer medium, drives the heat transfer medium into the heat exchange tubes independently arranged in structure through a multi-stage controllable micro-pump, and dynamically selects the heating area in the direction of natural frost peeling advantage according to the platform attitude, effectively improving the frost layer peeling efficiency.

[0051] During the defrosting execution process, the matching relationship between the latent heat transfer rate and the frost layer melting demand is predicted through a heat transfer model, the defrosting time range is set, and an intermittent heating is carried out according to the current redundant energy capacity by adopting a dynamic duty cycle control strategy. When the energy is insufficient, the system records the executed defrosting state based on the time stamp and heat input, and writes it into the platform task data storage area; after the energy is restored, the platform task management layer schedules and judges whether it has the condition to continue the execution, and restores the heating process according to the breakpoint information until the closed loop is completed. The whole method runs through the perception, judgment, execution and feedback links with a highly adaptable control logic, and combines the gravity direction recognition, local energy consumption adjustment and termination state multi-factor determination mechanism, significantly improving the defrosting stability and resource utilization efficiency of the floating platform in a complex operating environment.

[0052] Specifically, it includes the following steps:

[0053] Build a multi-parameter environmental perception model that includes air pressure, temperature, humidity, the state of the heat exchange surface, and the platform attitude angle, and determine whether the critical condition for starting defrosting is reached based on this model; the significance of this step is to construct a basic model for comprehensively perceiving the current operating environment of the platform. Due to the high uncertainty of the environment where the floating platform is located, only through multi-dimensional perception of meteorological conditions, frost layer state, and equipment attitude can it be accurately judged whether defrosting is necessary and the timing, avoiding false triggering or delayed start, and ensuring the pertinence and real-time nature of the judgment of the control system.

[0054] Select at least one of carbon dioxide, freon, or air as the heat medium, and through an independent heating cycle path isolated from the heat flow of the refrigeration system, guide the heat medium to input latent heat to the predetermined heat exchange area; the core significance of this step is to construct an independent heating channel that does not interfere with the refrigeration process, so as to achieve precise control of heat flow. Through the selectable types of heat medium, the system has the adaptability to different environmental requirements, and at the same time enhances the thermal response efficiency and energy scheduling flexibility of the system.

[0055] The heat exchange area is composed of heat exchange tubes independently arranged in terms of structure, and the direction of gravity is dynamically judged according to the current platform attitude angle, and the pipe section in the advantageous position of the natural defrosting direction is preferentially selected to start heating; the design significance here is to optimize the directionality of the frost layer peeling process. By perceiving the relationship between the platform attitude and the direction of gravity, the heating is more concentrated on the position where the frost layer is most likely to fall off naturally, making the best use of the gravity effect, improving the defrosting efficiency and reducing energy consumption waste.

[0056] During the heating process, determine the target defrosting time period according to the predicted phase change latent heat transfer rate, and perform intermittent heating on the selected heat exchange pipe section, controlling the average power within each heating cycle not to exceed the current platform redundant energy capacity; this step emphasizes the precise matching of energy and heat, avoiding overheating and energy waste. Through thermal model prediction, reasonably arrange the heating time, and adopt intermittent control means to reduce the interference to other tasks of the platform while ensuring the defrosting effect, and achieve the balance between the thermal control process and energy sustainability.

[0057] When it is detected that the energy is insufficient, pause the heating process and record the completed defrosting duration and area information, and continue to execute the remaining defrosting process after the energy is restored until the closed-loop defrosting control is completed. The significance of this step is to endow the system with the ability to interrupt and resume, making the defrosting task continuous and fault-tolerant. When the energy is insufficient, do not forcibly execute defrosting, but retain the current progress data and continue to execute after the conditions are restored, which not only ensures the safe and stable operation of the system, but also improves the rationality of resource use.

[0058] The construction process of the multi-parameter environmental perception model includes dynamic coupling calculation of the high-altitude environmental air pressure and the relative motion state of the platform. The air pressure information is obtained by continuously collecting the current altitude, latitude, and atmospheric temperature of the platform, and is derived using one of the empirical pressure-altitude formula and the moist air state equation to obtain the instantaneous air pressure prediction value;

[0059] The platform attitude angle information is realized through the average filtered values of multiple inertial measurement units, and is combined with the actual load offset of the platform to form a corrected angle model after correction;

[0060] The heat transfer surface state information is obtained by continuously scanning with a micro infrared temperature difference sensor, and the surface frost thickness is estimated from the heat map data according to the gray scale fitting function;

[0061] The multi-parameter environmental perception model fits the curve based on the environmental change trend in the historical operation cycle, combines the current acquisition results to perform preset multi-objective prediction, and judges whether the critical temperature-pressure energy state for the first defrost trigger has been reached.

[0062] In the present invention, in order to accurately identify the defrost control trigger conditions during the operation of the high-altitude floating platform, a multi-parameter environmental perception model is constructed. The construction process of this model first includes dynamic coupling calculation between the high-altitude environmental air pressure and the relative motion state of the platform, which is used to reflect the comprehensive influence of environmental pressure changes on the platform attitude and external heat exchange conditions.

[0063] Specifically, the platform obtains the altitude information of the current operation through the equipped altitude measurement device, and at the same time combines the geographical latitude data provided by the positioning module and the current atmospheric temperature value provided by the environmental temperature sensor to form a basic input variable set. These variables are highly correlated and jointly determine the thickness and density distribution of the air column where the platform is located, thus affecting the air pressure value.

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

[0065] First, set the standard air pressure value at sea level, usually taken as 101325 Pa; then, taking the currently measured altitude as a variable, calculate the atmospheric temperature corresponding to the altitude according to the temperature decrease law in the troposphere section of the standard atmosphere (generally, the temperature drops about 6.5 degrees Celsius for every 1000 meters increase). Subsequently, substitute these parameters into the exponential function relationship to obtain the corresponding static air pressure value.

[0066] If higher accuracy is required, especially in areas with relatively high humidity or frequent platform changes, it is possible to 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. By calculating the partial pressures of the two separately and then combining them, the current mixed air pressure is obtained. In the calculation, the partial pressure of water vapor can be indirectly obtained from the air temperature and relative humidity through the Clausius-Clapeyron relationship. Based on the theoretical logarithmic relationship between the saturated water vapor pressure and temperature in the Clausius-Clapeyron relationship, an exponential empirical function is used to fit and express it in practical applications to improve the computational simplicity and engineering adaptability. Specifically:

[0067] The saturated water vapor pressure is calculated using an exponential empirical function with temperature as the variable. Its expression is: the saturated water vapor pressure is equal to the constant 6.112 multiplied by the exponential function, and the exponent part is 17.67 multiplied by the Celsius temperature divided by the temperature plus 243.5. The actual water vapor pressure is equal to the saturated water vapor pressure multiplied by the relative humidity (the percentage is converted to a decimal between 0 and 1). Subsequently, the partial pressure of dry air is obtained by subtracting the actual water vapor pressure from the total pressure, forming the partial pressure input parameters required for the moist air state equation, and finally obtaining the instantaneous air pressure prediction value.

[0068] Ultimately, the platform will select one of the above two methods based on whether the current environment has drastic humidity changes, i.e., the data standard deviation reaches the preset requirement, or whether it enters the sensitive altitude range. The empirical formula is used under ordinary stable flight conditions, and the moist air state equation is automatically enabled in the case of sudden humidity changes, altitude jumps, or when approaching the defrost boundary judgment.

[0069] The prediction result finally forms a real-time updated instantaneous air pressure value, which serves as one of the core bases for the subsequent model to judge whether defrosting is initiated. In this way, it is ensured that the platform can achieve stable and high-precision air pressure modeling under different altitudes and atmospheric structures, providing scientific support for the latent heat defrosting timing.

[0070] During the operation of the platform, attitude measurement is usually affected by factors such as asymmetric platform load and motion inertia interference, resulting in deviations in the original angle data collected by the IMU. To improve the accuracy of judging the heat exchange tube and the gravity direction, it is necessary to introduce a centroid offset correction function.

[0071] The centroid offset correction function is a function model used to compensate for the impact of the current structural load distribution of the platform on the attitude angle. The input parameters of this function include: the standard centroid position of the platform structure, the current load mass distribution (such as batteries, communication devices, thermal management components, etc.), and the corresponding installation position coordinates. According to the principle of static moment conservation, the three-dimensional space offset vector between the current true centroid and the standard centroid is calculated. This offset vector is then superimposed and corrected with the angles collected by the IMU through the direction cosine matrix or quaternion attitude transformation to obtain the true pitch, yaw, and roll angles, thereby improving the accuracy of gravity direction judgment in defrosting area positioning.

[0072] For the acquisition of the heat exchange surface state, continuous surface scanning is carried out using a micro infrared temperature difference sensor installed directly opposite the heat exchange tube array. The sensor outputs an infrared thermal distribution map in the form of a frame image, and the gray value of this image corresponds to different surface temperatures. Using the gray-temperature calibration function, the gray value can be converted into the actual temperature distribution, and then the frost layer thickness of each area can be estimated through the spatial fitting algorithm of the heat distribution gradient.

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

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

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

[0076] Next, select the target area (such as the entire row or column) for fitting modeling, and linear functions, polynomial functions, piecewise fitting, or spline interpolation can be selected;

[0077] For the area where the temperature gradient suddenly changes or forms a "temperature platform" structure, it is judged as being covered by the frost layer;

[0078] The preliminary estimation of the frost thickness is established based on an empirical function. For example: when the gradient amplitude reaches a specific threshold and the continuous length exceeds a certain number of pixel distances, combined with the condition that the average surface temperature of this area is lower than the dew point temperature, it can be judged that there is a frost layer in this area; use the calibration experiment to obtain the corresponding relationship function between the temperature difference and the frost thickness, such as: frost thickness (unit: millimeter) = a×ΔT + b (a and b are experimental fitting coefficients), ΔT is the temperature difference, to complete the thickness numerical calculation.

[0079] In the model comprehensive judgment stage, the changing trends of environmental parameters in the historical operation cycle are introduced. The temperature, humidity, and air pressure change curves are constructed through polynomial regression or spline interpolation methods and compared with the currently real-time collected parameters. A comprehensive judgment mechanism for temperature and pressure energy is set in combination with threshold rules, that is, when the current air pressure prediction value of the platform, the combined state of temperature and humidity, and the estimated value of the heat exchange cream thickness jointly meet the requirements of the preset start critical point, it is judged that the current round of operation reaches the trigger condition for the first defrosting.

[0080] The specific operation process is as follows:

[0081] Historical data collection and storage: The platform records environmental parameters such as the outside temperature, relative humidity, air pressure, and the temperature of the heat exchange surface and the estimated value of the frost layer in each operation cycle, forming a time series data set.

[0082] Trend fitting modeling method: The system performs regression modeling on the above multi-dimensional environmental variables respectively. The optional methods include: polynomial regression (usually second-order or third-order) for fitting the fluctuating change trend;

[0083] Spline interpolation (such as cubic spline) for smoothing the continuous change curve;

[0084] Moving average or weighted moving average for suppressing abnormal fluctuations.

[0085] Comparison and analysis of real-time data and fitting curves: The currently collected values of temperature, humidity, air pressure, etc. will be compared with the trend curves, and the change rate, change direction, and deviation value will be extracted to identify whether there are abnormal environmental mutations or continuous deterioration.

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

[0087] The air pressure is lower than a certain absolute value (for example, five thousand pascals);

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

[0089] The estimated surface frost thickness exceeds one millimeter;

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

[0091] Any two of the above three items continuously meet for more than a certain duration (for example, twenty minutes).

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

[0093] For example, the platform operates at an altitude of 11,000 meters. The measured external temperature is -40 degrees Celsius, the humidity is 50%, and the platform attitude maintains a small-angle pitch. However, since the temperature of the condensation heat exchange surface has been below -5 degrees Celsius for 30 consecutive minutes and the estimated frost thickness is greater than 1 mm, and at the same time the predicted air pressure value is close to one-fourth of an atmosphere, the system determines that it has entered the high-risk frost formation interval according to the model and automatically judges that the defrosting critical point has been reached.

[0094] The structural layout of the independent heating cycle path includes setting a closed pipeline path that is physically completely isolated from the refrigeration and heating cycle. The heat transfer medium filled inside is selected from one of carbon dioxide, Freon, or air, and controlled variable-speed flow is achieved through a built-in multi-stage micro-pump.

[0095] The heating path is preset with multiple heat exchange branch segments, and each branch segment 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, by calculating the average rate of latent heat release per unit length of the heat transfer medium in the current environment and matching it with the currently estimated frost thickness on the heat exchange surface, the flow rate and heating duration are dynamically selected.

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

[0098] To ensure that the latent heat defrosting process has thermal isolation, directional heating ability, and energy efficiency control effect, a structural layout method of an independent heating cycle path is proposed. This path is physically completely isolated from the refrigeration and heating cycle in terms of structural design, avoiding heat backflow or energy interference during operation, thereby improving the thermal management stability of the entire platform.

[0099] The core of this structure includes a closed pipeline path, which serves as the main channel of the heating circuit and is filled with a pre-selected heat transfer medium. The heat transfer medium can be selected from at least one of carbon dioxide, Freon, or air. The selection of the heat transfer medium is determined according to the specific platform operating environment. For example, in a high-altitude and low-pressure environment, carbon dioxide is more suitable for short-time and high-efficiency heat release due to its high phase change latent heat and density regulation ability; while Freon is suitable for low-temperature and low-vibration environments, and air is suitable for use in high-redundancy environments.

[0100] The heat transfer medium flows in this closed path driven by a built-in multi-stage micro-pump, which has the ability to control variable speed regulation. Its adjustment method is based on the input signal of the platform control logic, allowing the heat transfer medium to achieve different flow rates at different heating stages to meet the refined heat control requirements. For example, a higher flow rate can be set in the initial defrosting stage for rapid heating, and the flow rate can be reduced in the later maintenance stage for energy-saving operation.

[0101] Multiple heat exchange branch segments are preset in this heating path. Each branch segment is an independent structure with unidirectional heat conduction performance, which can be achieved by integrating a heat flow rectifier in the branch channel or adopting a material heterojunction structure. For example, one end of the branch segment uses a high thermal conductivity material (such as copper or aluminum), and the other end uses a low thermal conductivity blocking layer (such as a ceramic coating or an aerogel lining) to form a unidirectional heat flow conduction effect, thus effectively preventing heat from flowing back to the main heat exchange condensation side during the defrosting process and ensuring heat exchange stability.

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

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

[0104] The first way: the total heat model composed of the sum of sensible heat and latent heat; this model believes that during the flow of the heat medium in the pipeline, heat will be released in two ways: one is to release sensible heat through a temperature drop, and the other is to release latent heat of phase change if a phase change occurs (such as from 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, obtained through a flow meter), and the unit can be kilograms per second; then measure the temperature difference between the inlet and outlet of the heat medium in this section of the pipeline (that is, the inlet temperature minus the outlet temperature); look up the thermophysical property table according to different heat media (such as carbon dioxide, freon or air) to obtain the specific heat capacity of this substance (the heat required per unit mass to increase by one degree); finally, if the heat medium has a gas-liquid phase change (such as supercritical carbon dioxide condensation), it is also necessary to look up the table to obtain the latent heat value per unit mass of this heat medium; add these two parts of heat to obtain the total heat (thermal power) released per unit time and per unit length of the pipeline. This method is applicable to the situation where the heat medium may experience coexistence of temperature change and phase change behavior in the heating section, and is particularly suitable for operating scenarios using carbon dioxide or heat media with condensation properties.

[0106] The second way: only consider the sensible heat model. If the heat medium only mainly changes in temperature and does not undergo a phase change in the control strategy (such as using air or sealed gaseous freon), the latent heat term can be omitted, and only the sensible heat transfer during the process of the heat medium changing from high temperature to low temperature needs to be calculated. This method is more suitable for application scenarios with limited resources, short pipeline space or limited heat medium flow rate, and is particularly suitable for lightweight platforms or temporary defrosting strategies.

[0107] When the heat medium has significant phase change potential (such as supercritical carbon dioxide), the first method is recommended; when the heat medium is gaseous and remains single-phase throughout the defrosting process, the second method is recommended. Both of the above methods are existing technical methods in the basic field of thermodynamics. The choice of which one depends on the type of heat medium and the thermodynamic objectives of the platform during implementation, and both have standardized data sources and ready-made calculation tables for support.

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

[0109] The specific steps are as follows: First, estimate the average thickness of the frost layer in different regions on the heat exchange surface through an infrared thermal imaging fitting algorithm, usually in millimeters or meters; then, according to the standard physical property parameters, take 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 fusion of ice, that is, the energy required for a unit mass of ice to change from solid to liquid state, which is usually about 334,000 joules per kilogram; multiply the frost layer thickness (unit volume) by the density to get the mass, and then multiply by the latent heat of fusion per unit mass to obtain the energy required for the entire defrosting area.

[0110] For example, if the average frost thickness in a certain area is 1 millimeter and the area is 1 square meter, the total volume is 1 liter; multiplying by the density of ice of about 0.917 kilograms, the theoretical minimum heat required to melt the frost layer in this area is about 300,000 joules.

[0111] The system control logic uses the corresponding relationship between the average rate of latent heat release per unit length of the heat medium in the current environment and the total melting energy required for the current frost layer on the heat exchange surface as the judgment basis, and dynamically adjusts the micro-pump flow rate and the heat medium heating time according to preset rules at different time periods to ensure that the heat supply not only meets the defrosting requirements but also does not cause energy waste.

[0112] For example, the micro-pump flow rate is selected by the control logic based on the balance between the current required heat and the allowable heat consumption:

[0113] If the current frost thickness is large, the ambient temperature continues to be low, and 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 enhance the heat transfer rate;

[0114] If the current energy state is sufficient and there are no high-priority tasks on the platform, a 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 only maintain low-intensity maintenance heating to prevent heat waste.

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

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

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

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

[0120] If the platform energy supply is at a medium level, segmented heating is adopted, and after each heating, it enters the "cooling - feedback - determination" stage;

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

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

[0123] For example, in a platform operation cycle: The current frost layer thickness is 2 mm; The ambient temperature is - 20 °C; The current redundant energy is in a normal state; The current flow rate of the micropump is 100 grams of carbon dioxide per second; According to the calculation, the heat released per second is about 5000 joules; The total heat required for defrosting is about 300,000 joules. Then the control logic will judge that heating needs to continue for about 60 seconds. However, to avoid thermal shock and energy waste, the system will execute two rounds of heating, each round for 30 seconds, with a 10 - second break in the middle, forming an intermittent thermal control strategy of "heat - stop - heat". If the surface reaches the preset requirements detected by the sensor at the end of the second round, it is considered that the defrosting task is completed and heating stops.

[0124] At the same time, the flow direction of the heat medium is not fixed, but is adjusted in real - time according to the platform attitude information. The platform attitude obtains its current angles with the gravity direction in the pitch and roll states through the corrected angle information. The control logic performs a dot - product calculation of the gravity direction vector and the direction vectors of each branch heat - exchange tube to calculate the included angle size, and preferentially activates the branch segment with the smallest included angle with the gravity direction. This approach can ensure that heat is preferentially input into the area in the favorable direction for natural frost shedding, improving the frost layer shedding efficiency.

[0125] Taking an embodiment as an example, assume that the platform is currently in a slightly forward-tilted state (pitch angle of five degrees), and the heat map shows that the heat exchange cream layer at the front is thicker. The control logic determines that the angle between the gravity vector and the direction of the front row of heat exchange tubes is only twenty degrees, while the angles in the remaining areas are above forty degrees. Therefore, the front row of branches is preferentially activated for heating. This dynamic control method not only improves the defrosting efficiency but also significantly reduces the peak energy consumption per unit time, ensuring the energy balance of the entire platform.

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

[0127] The gravity direction recognition strategy for determining whether the heating target area is in a natural defrosting advantageous position includes vector projection of the layout direction of the heat exchange tubes under the three-dimensional attitude angles of the platform, namely pitch, yaw, and roll, and using the gravity vector as a reference. By calculating the angle between the axis of the heat exchange tube and the gravity vector, the most favorable potential defrosting direction is determined;

[0128] When this angle is less than the set first angle threshold, it is considered that this heat exchange section has the ability of natural defrosting; if all areas do not meet the requirements, it enters the waiting state until the environmental parameters are updated in the next cycle and then it is evaluated again whether to activate.

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

[0130] When the heat exchange tubes are structurally laid out, they have their respective spatial orientations, which can be preset as a set of fixed direction vectors, usually defined in the platform's own coordinate system. For example: forward X-axis, leftward Y-axis, upward Z-axis. In the control logic, the axis direction of each heat exchange tube section is represented by a unit vector, denoted as "heat exchange axis vector", and after each update of the attitude angle, the spatial projection relationship of these axis vectors relative to the gravity direction under the current platform attitude is calculated.

[0131] The gravity direction always points towards the center of the earth in the earth coordinate system. When the platform attitude changes, its representation in the platform local coordinate system will change. According to the current pitch and roll angles, using the direction cosine transformation matrix or quaternion transformation, the global gravity vector (standard is vertically downward) is transformed into the gravity vector in the platform's own coordinate system.

[0132] At this time, the axial vector of each heat exchange tube section can be used to calculate the included angle with the transformed gravity vector. The vector included angle formula is used: the cosine value of the vector included angle = the dot product result of the two vectors ÷ the product of the moduli of the two vectors. Since both vectors are unit vectors, only the dot product result needs to be calculated to obtain the cosine value of the included angle, and then the actual angle can be obtained through the inverse cosine function. This angle is the "included angle between the direction of the heat exchange tube section and the gravity direction", and this included angle reflects whether the frost layer is likely to slide naturally under the action of gravity.

[0133] The control logic presets a first angle threshold, such as thirty degrees, as the boundary for judging whether natural defrosting ability is available. When the included angle between a certain heat exchange tube section and the gravity direction is less than this threshold, it indicates that the direction of this tube section is basically downward, conforming to the gravity peeling direction, and it is judged that it has the advantage of natural defrosting.

[0134] In each judgment cycle, the control logic will traverse all heat exchange tube sections, screen out the tube sections that meet the condition that the included angle is less than this threshold, and preferentially activate the part with the lowest temperature or the thickest frost layer among them as the first-round heating target. If there are multiple areas that meet the conditions, they can be further sorted and executed according to the preset priorities (such as front and rear parts, energy load balance, etc.).

[0135] Taking a certain floating platform as an example, ten heat exchange tube sections are arranged along the front and rear directions on its heat exchange surface, numbered from one to ten, and are all arranged parallel to the X-axis of the platform. The current pitch angle is positive twenty-five degrees (the front end is raised), the roll angle is zero, and the platform is tilted backward as a whole. In this posture, after calculation, the included angles between the first five heat exchange tube sections and the gravity direction are more than fifty degrees, and the included angles of the last five are about twenty-five degrees. After comparison, the control logic finds that the latter half meets the condition that the included angle is less than thirty degrees, and judges that it has the advantage of natural defrosting. Therefore, the control logic selects the ninth tube section with the thickest frost layer in the back row as the current heating target and starts the latent heat heating cycle. If the posture of the platform continues to change during flight, the system will periodically refresh the included angle judgment to ensure that the most favorable direction is selected for the next round of defrosting control.

[0136] The thermal modeling algorithm for predicting the latent heat transfer rate includes a coupled calculation of the heat medium flow rate, the initial temperature, the ambient temperature and humidity ratio, and the current frost layer thickness. One of a set of classical heat transfer models is used for modeling, including a one-dimensional steady-state heat conduction model, an unsteady-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 defrosting time period is dynamically adjusted.

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

[0138] The one-dimensional steady-state heat conduction model is applicable to the modeling of the heating behavior of the heating medium under the conditions of constant flow rate and constant temperature difference transfer. The modeling assumes that the heat flow conducts in a single direction along the radial direction of the pipe wall, ignoring the influence of time change and phase change. The main calculation idea is as follows: taking the temperature difference between the contact surface of the heating medium and the frost layer as the driving force; combining the thermal conductivity constant (obtained by looking up the table according to the heating medium and wall materials); assuming that the frost layer is an ice layer with uniform thickness, and using Fourier's law of heat conduction 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 the heat flux density by the contact area and time can obtain the heat transferred by the heating medium in a certain period. The applicable scenarios are the stable flight of the platform, the initial stage of heating, and the stage with strong frost layer thermal conductivity.

[0140] Unsteady-state phase change model: When the heating medium undergoes gas-liquid phase change during heat transfer, or the frost layer melts locally and the temperature changes unstably after being heated, an unsteady-state heat transfer model can be used for modeling. This model considers the change of temperature with time, combines the specific heat capacity of ice, latent heat of fusion, and ambient temperature fluctuation, and calculates the change of heat demand during the process of the frost layer heating up from the initial temperature to the melting process. The model logic includes: dividing the frost layer into several layers, and the temperature of each layer gradually increases; adding the required heat layer by layer, considering the change of the heat input by the heating medium with time; integrating the latent heat release during the phase change process to form a non-linear energy transfer model. This method is applicable to the scenarios where the frost layer is thick, the heating time is long, and the heating medium has phase change (such as carbon dioxide condensation).

[0141] The thermal resistance network model simulates the heat transfer path in the way of series-parallel combination of multiple thermal resistance structures, and can be used for the overall heat flow estimation under complex structures. Each layer of structure (such as heating medium - pipe wall - frost layer - external air) is regarded as a thermal resistance unit, and the modeling method is as follows: calculating the thermal resistance of each section of material according to the thermal conductivity, thickness, and area; combining the thermal resistances in series-parallel to obtain the equivalent total thermal resistance; using Ohm's analogy method to calculate the heat flow according to the total thermal resistance and the temperature difference at both ends. Heat flow = temperature difference ÷ total thermal resistance; this model is applicable to the situations of multi-layer materials, non-uniform frost layer, and heat flow affected by multiple interfaces, such as the pipeline has a fouling layer, thermal insulation coating and other structures.

[0142] After the modeling is completed, compare the latent heat value released by the heat medium per unit time with the heat required for the target defrosting area. The target heat is determined by the frost layer thickness, area, ice density, and latent heat of fusion. The control logic calculates the ratio between the calculated unit heat release rate and the total demand, determines the theoretical shortest defrosting duration, and sets aside an estimated safety margin time period.

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

[0144] Assume that the platform currently uses Freon as the heat medium, with an inlet temperature of 0 °C, an outlet temperature of -10 °C, a flow rate of 100 grams per second, a current frost layer thickness of 1 mm, and a coverage area of 0.5 square meters. Select a one-dimensional steady-state heat conduction model for modeling. After checking, the thermal conductivity is about 0.2 W / (m·K), the temperature difference is 10 °C, and the path length is 2 mm. Calculate the heat flux density to be 1000 W / m², and multiply by the contact area to get the heat released per second to be 500 W. Based on the ice density and latent heat calculations, the total heat demand for the defrosting area is about 100,000 joules. Therefore, the system determines that this round of defrosting requires heating for about 200 seconds. The system will set four rounds of heating, 50 seconds per round, with a 5-second cooling interval to form a stable defrosting 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, task load peak and valley, and external air pressure fluctuation rate. This 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: an activation period and a 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 air 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 length and the total duration of the cycle in a fixed heating cycle. In this strategy, the duty cycle is not set to a static value but is calculated and generated in real time based on three key dynamic variables:

[0148] Current redundant energy capacity: The remaining energy resources currently available for defrosting on the platform. After task scheduling, it is calculated by the task management layer and an instruction is issued to determine whether there is sufficient power or heat source for high-intensity heating in the current cycle.

[0149] Task load peak and valley situation: Judge the current task energy consumption stage of the platform. For example, whether communication, navigation, and attitude control are in a high energy consumption interval. If in the task 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, temperature and humidity changes may occur. If frequent fluctuations or instability occur, a cooling period needs to be reserved for heat release and condensation transition to avoid overheating caused by mutations.

[0151] Based on the above three parameters, the algorithm recalculates the recommended duty cycle every cycle and compares it with the preset power control threshold to ensure that the heating behavior does not cause platform energy overload. For example, if the current platform redundant power is 200 watts, the task load is at a low level, and the air pressure change is mild, the control logic can set the duty cycle of this cycle to 80, that is, the heating is activated for 48 seconds and cooled for 12 seconds every minute.

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

[0153] Activation period: In this phase, the platform task management layer has completed energy scheduling and clearly allocated the available energy for the defrosting process. The control logic accordingly starts the micro pump and makes the heat medium enter the target heat exchange pipe section to perform heating operations. The length of the activation period is determined according to the above duty cycle calculation results and is allowed to vary in different cycles.

[0154] Cooling period: After heating, the system enters the thermal control cooling stage, the heat medium stops flowing, and the heat exchange surface no longer receives new heat source input. The setting of the cooling period is not only an energy-saving strategy but also has the function of preventing condensation and refreezing.

[0155] The length of the cooling period is not fixed but is flexibly adjusted with reference to the external air temperature change trend. When the external temperature of the platform rises rapidly and there is a tendency for the frost layer to evaporate again after melting, the cooling time can be appropriately shortened; conversely, if the temperature drops suddenly, the cooling period should be extended to avoid re-condensation on the surface caused by heat residue and the formation of "return frost".

[0156] Taking one operating cycle as an example, the platform is currently in a high-altitude area. The task scheduling data shows that the current power surplus is 300 watts, the external environmental temperature is -10 degrees Celsius, and the humidity is medium and stable. At this time, the defrosting control enters a round of cycle execution: The control logic sets the total cycle length to 60 seconds; according to the calculation, the duty cycle is set to 70, that is, the activation period is 42 seconds and the cooling period is 18 seconds; the heat medium flow rate is 100 grams per second, the temperature difference is 10 degrees, and it is estimated that the heat released per second is about 5,000 joules; a total of about 210,000 joules are released within 42 seconds; this energy covers the heat of fusion required for a frost layer with a thickness of 1 mm and an area of 0.5 square meters; during the subsequent cooling period, the temperature of the heat exchange surface is monitored by a thermal sensor to ensure that there is no obvious risk of refreezing; if residual heat is still detected at the end of the cooling, the duty cycle of the next round of heating will be reduced to 60 to prevent overheating.

[0157] The introduction of the intermittent heating strategy significantly improves the flexibility of energy use and the safety of the thermal control strategy. Through the dynamic calculation of the duty cycle and the climate self-adaptive adjustment of the cooling cycle, the platform can accurately balance the "defrosting efficiency" and "energy consumption control" in the high-altitude variable environment, effectively reducing the risk of frost layer residue and the probability of energy waste.

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

[0159] When the energy is restored, first read the previous breakpoint information, and estimate the current required compensation heat based on the unfinished defrosting area, the heated energy consumption, and the remaining frost layer thickness. Compare it with the latent heat release rate of the heat medium to judge whether it has the ability to execute;

[0160] If the conditions for continuing to execute are met, directly resume the heating process of the corresponding heat exchange pipe section, otherwise delay entering the next cycle.

[0161] When the heating task is paused due to insufficient energy or priority scheduling conflict during the execution process, the control logic will automatically extract and save the following key status information to form a "defrosting breakpoint status structure" and write it into the platform task data storage area. Specifically, it includes:

[0162] Currently heated area: Record the numbers of the heat exchange pipe sections where the heating operations have been completed and their spatial positions, which is convenient for determining the specific processed and unprocessed areas when the subsequent tasks continue.

[0163] Input thermal energy: Calculate the total heat released currently based on the type of heat medium, flow rate, temperature difference, and heating time. The unit is joule, which is used to compare the subsequent required compensation heat.

[0164] Average temperature rise of the heat exchange surface: Obtain the temperature difference between the initial temperature and the current temperature of the heating area through a conformal temperature sensor, and take the weighted average of the local area to reflect the defrosting progress.

[0165] Time stamp of the current heating time point: Record the time identifier at the heating interruption with a high-precision clock module, which is convenient for associating the priorities and synchronization information in the task queue and supports the judgment of the execution of the delay strategy. This status structure is written into the platform task data storage area in real time to ensure that the data is not lost due to power interruption.

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

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

[0168] Estimate the current required compensation heat: Based on the remaining thickness of the frost layer (continuously estimated through the infrared thermal image) and the area of the unheated area, calculate the total required energy. Refer to the heat demand calculation model: Heat required for defrosting = Area of the remaining area × Frost layer thickness × Ice density × Latent heat of fusion.

[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 with the aforementioned thermal modeling algorithm to estimate the heat release capacity per unit time.

[0170] Judge whether the execution condition is established: If the current platform energy can support the defrosting heat demand of the remaining area, that is, the execution condition is met, then enter the recovery heating process; if not, the logic enters the delay mode and is re-evaluated when the next judgment cycle is triggered.

[0171] If the condition for continuing execution is met, the control logic will automatically call the "heated area" in the breakpoint structure as the starting point, sequentially select the target pipe segments from the uncompleted area, and resume the heat medium injection and heating process. The control process will recalculate the time allocation in the thermal modeling and compensate for the heat difference of the uncompleted part in the previous time to ensure the integrity of the overall defrosting heat closed-loop.

[0172] For example: In the previous round, the heating of pipe segments 1 to 3 has been completed, with a total input heat of 400,000 joules; the remaining pipe segments 4 to 5 have not been heated, and the infrared thermal image judges that the frost thickness in this area is about 1 mm; the required melting heat is about 300,000 joules; the current platform's heat medium release capacity per minute is about 100,000 joules; the calculation results show that the task can be completed within three minutes; the control logic resumes the heating task, sets a three-round heating cycle, 60 seconds for each round, and 15 seconds for intermediate cooling to complete the defrosting compensation task.

[0173] The method for obtaining the redundant energy capacity of the platform includes comparing the total current task power demand of the platform with the energy supply capacity in real time, judging whether to allow the activation of the heating operation according to the redundant capacity threshold, and the heating control sub-process is only executed under the condition that the platform task scheduling mechanism issues a non-interference signal.

[0174] The "heating control sub-process" refers to a specific heating task operation stage that is scheduled and executed by the defrost control logic after the trigger conditions are met (such as reaching the defrost critical point, sufficient energy, appropriate posture direction, etc.). It is a condition-driven sub-level action unit in the entire defrosting process. Assume that the entire "latent heat defrosting intelligent control process" is divided into: environmental judgment → energy evaluation → defrost start → heating control sub-process → state recording → recovery or termination. Among them, the heating control sub-process refers to the part of "actually executing heating", including actual operation steps such as starting the heat medium, controlling on and off, monitoring temperature, calculating time and power, etc.

[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. The supply capacity is calculated in real time based on the following parameters: current power output capacity (such as the product of voltage and current); remaining power of the energy storage device; energy load conversion efficiency (considering the loss of energy from the energy storage unit to the terminal execution structure); the impact of temperature on battery performance (especially critical on high-altitude platforms); in practice, the energy supply capacity can be automatically converted to the current maximum available power output upper limit of the platform, for example, the current maximum output power is 300 watts.

[0176] When the current maximum available power of the platform is subtracted from the total power demand of the current task, the result is the redundant energy capacity that the platform can allocate to non-core tasks. This value is the key basis for judging whether the heating task can be executed.

[0177] The redundant capacity threshold is preset, which is the minimum amount of energy required for the defrost task to run. The specific value is determined by the heat medium type, current flow rate, heating time and heat demand. For example, if the micropump and heating module require a total of one hundred watts and run for at least thirty seconds to achieve the initial defrost effect, the redundant capacity threshold is set to one hundred watts. The judgment process is as follows: If the redundant energy capacity ≥ the preset threshold → it is judged that the heating execution conditions are met; if the redundant energy capacity < the preset threshold → the heating task is suspended or delayed; for example: the current energy supply capacity is three hundred watts; the total power demand of the task is one hundred and ninety watts; the redundant energy capacity is one hundred and ten watts; if the threshold is set to one hundred watts → it is judged that the heating task can be executed.

[0178] In order to prevent the heating process from interfering with high-priority tasks (such as communication maintenance and navigation updates), the start of the heating control subprocess must also meet the "non-interference signal" condition. This signal is generated by the task scheduling mechanism and is issued according to the following rules: whether there is a high-priority task scheduled to start; whether the current platform is in a transition state (such as maneuvering, track change, communication window); whether there is an energy consumption warning or energy switching process; if all judgments are risk-free, the heating control logic can enter the execution state. This mechanism prevents the defrost task from being forcibly activated during task switching, violent platform movement, or energy fluctuations, thereby improving the safety of platform operation.

[0179] Assume that the platform is currently equipped with solar panels with a total output power of 250 watts. Communication, attitude control, and navigation together consume 180 watts, and the remaining redundant capacity is 70 watts. The current defrosting task requires that the heating module needs at least 90 watts of power. Therefore, the control logic determines that the redundant capacity threshold is not met and rejects activating the heating process. At the same time, the task scheduler determines that strong power transmission for communication is required within the next minute, marking it as a high-interference interval. Even if the redundant power recovers later, the defrosting is still delayed because the interference signal has not been cleared. When the solar panels provide an output of 320 watts in the next cycle, the redundant power reaches 140 watts, and there is no high-priority task plan, the control logic of the system receives an "executable signal" and immediately activates the heating process and enters the first-stage heating cycle.

[0180] The completion standard of the closed-loop defrosting control is jointly determined based on the heat transfer surface temperature recovery rate, the detection result of frost layer shedding, and the statistical result of the remaining untreated area, and the determination is based on meeting the termination threshold.

[0181] The surface temperature is detected in real time by multiple close-fitting temperature-sensitive sensors and compared with the initial non-frost temperature to calculate whether the heating rate is stably approaching zero.

[0182] Whether the frost layer has completed physical peeling is judged by the feedback of micro-vibration acceleration for frost layer shedding.

[0183] For the untreated area, the temperature difference before and after processing is distinguished and distributed according to the thermal image recognition comparison.

[0184] When all three indicators reach the preset threshold, it is automatically determined that the defrosting is completed, and the task record is cleared.

[0185] During the heating process, the platform collects the surface temperature of the heat exchange tube in real time through multiple close-fitting temperature-sensitive sensors. Each sensor is attached to a key heating area to record the temperature change curve.

[0186] The control logic of the system compares the current temperature collected by the sensor with the temperature value when there is no frost initially, and calculates the rate of temperature rise per unit time, that is, the "temperature recovery rate". If this rate continues to decrease and approaches zero, it indicates that the frost layer has been basically cleared, there is no obvious obstacle in the heat conduction process, and the heat input cannot further increase the surface temperature. At this time, the surface state is close to "defrosting completed". For example: the initial non-frost temperature is 0 °C; the current temperature is -1 °C; the temperature rise in the past 30 seconds is less than 0.2 °C; if the set termination threshold of the recovery rate is 0.1 °C / minute; then it is judged that "the temperature rise tends to be saturated". This judgment process is a continuous dynamic comparison. Generally, it is considered that this indicator is achieved only when the low-rate change is maintained for two consecutive observation periods (such as every 15 seconds as a period).

[0187] To determine whether physical peeling of the frost layer has occurred, the present invention adopts a micro-vibration acceleration feedback method. By integrating a micro-accelerometer element at the key part of the heat exchange structure, it is used to sense the tiny structural vibrations or pulses caused by the shedding of the frost layer. When heating causes the frost layer to change from solid state to liquid or gaseous state, the change in its mass and adhesion force will cause tiny release vibrations in the local structure. This signal is captured by the accelerometer and the background interference (such as platform attitude disturbance) is removed through filtering, and a pulse response with typical frequency characteristics is extracted. The control logic judges that if a micro-vibration signal within a specific peak frequency range is detected; and the matching degree of this signal with the known "peeling response template" exceeds a preset recognition threshold (for example, the similarity exceeds 80%); then it is determined that physical peeling of the frost layer in this area has been completed. This method can judge the actual defrosting effect without relying on image recognition, and is especially suitable for feedback control in the case of high altitude and low light or closed equipment structure.

[0188] The third criterion for defrosting completion judgment is the statistics of the remaining unprocessed areas, mainly relying on thermal image recognition means. By comparing the temperature distributions of the same heat exchange area before and after heating through an infrared thermal image. The control logic operates according to the following process: obtain the thermal image of the current heat exchange surface; perform pixel-level matching with the original thermal image taken before the start of heating; extract the area where the temperature difference is lower than a certain temperature rise threshold (for example, less than one degree Celsius), and regard it as the "unresponsive area"; count the ratio of its total area to the total area of the heating area;

[0189] If the ratio is less than the termination threshold (such as 5%), it is judged that defrosting is basically completed; if the ratio is higher than the threshold and there is no micro-vibration response or temperature rise sign in the corresponding area, there may be a situation of local non-heating or defrosting failure, and it will be marked as the priority processing area in the next round.

[0190] Among the above three judgment indicators, if all meet the preset termination threshold, the control logic automatically determines that the current closed-loop defrosting process has been completed. At this time, it will immediately: terminate the heating process; clear the current defrosting task scheduling record; release the energy scheduling lock; send a "defrosting completion signal" to the task management mechanism; ensure the recovery of platform energy resources, update the task scheduling status, and prepare for the next defrosting cycle.

[0191] During a certain defrosting process, the platform selected three middle heat exchange tubes for heating, and the total heating time was 120 seconds. The temperature-sensitive sensor recorded that the temperature rise rate continued to be lower than 0.1 degree Celsius per minute in the last two cycles; the accelerometer detected typical peeling pulse responses twice at the end position; the infrared thermal image comparison showed that the area of the unresponsive area was only 3% of the total area; based on this, the control logic judged that all three conditions were met and immediately closed the heat medium flow and entered the defrosting completion state.

[0192] Embodiment 2: A latent heat defrosting intelligent control system, including:

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

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

[0195] A directional heating unit, which is used to dynamically judge the gravity direction according to the current platform attitude angle for the heat exchange area composed of heat exchange tubes independently arranged in structure, and select the pipe section in the advantageous position of the natural defrosting direction to start heating preferentially;

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

[0197] A state continuous control unit, which is used to pause the heating process and record the completed defrosting duration and area information when detecting insufficient energy, and continue to execute the remaining defrosting process after the energy is restored until the closed-loop defrosting control is completed.

[0198] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0199] It should be understood that in various embodiments of the present application, the magnitudes of the sequence numbers of the above processes do not mean the order of execution. The execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0200] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

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

[0202] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims described above.

Claims

1. An intelligent control method for latent heat defrosting, characterized in that, It includes the following steps: Establish a multi-parameter environmental perception model including air pressure, temperature, humidity, heat exchange surface state and platform attitude angle, and judge whether the critical condition for starting defrosting is reached based on this model; Select at least one of carbon dioxide, freon or air as the heat medium, and through an independent heating cycle path isolated from the heat flow of the refrigeration system, guide the heat medium to input latent heat to a predetermined heat exchange area; The heat exchange area is composed of heat exchange tubes independently arranged in terms of structure, and the direction of gravity is dynamically judged according to the current platform attitude angle, and the pipe section in the advantageous position of the natural defrosting direction is preferentially selected to start heating; During the heating process, determine the target defrosting time period according to the predicted phase change latent heat transfer rate, and perform intermittent heating on the selected heat exchange pipe section, and control the average power within each heating cycle not to exceed the current platform redundant energy capacity; When energy shortage is detected, pause the heating process and record the defrosting duration and area information that have been completed, and continue to execute the remaining defrosting process after the energy is restored until the closed-loop defrosting control is completed.

2. The latent heat defrosting intelligent control method according to claim 1, wherein The construction process of the multi-parameter environmental perception model includes dynamic coupling calculation of the high-altitude environmental air pressure and the relative motion state of the platform. The air pressure information is obtained by continuously collecting the current altitude, latitude and atmospheric temperature of the platform, and is derived using one of the empirical pressure-height formula and the moist air state equation to obtain the instantaneous air pressure prediction value; The platform attitude angle information is realized through the average filtered values of multiple inertial measurement units, and is combined with the actual load offset of the platform to form a corrected angle model after correction; The heat exchange surface state information is obtained by continuous scanning of a micro infrared temperature difference sensor, and the surface frost thickness is estimated according to the gray-scale fitting function of the heat 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 acquisition results for preset multi-target prediction to judge whether the critical temperature, pressure and energy state for the first defrost trigger has been reached.

3. The intelligent control method for latent heat defrosting according to claim 2, characterized in that The structural layout method of the independent heating cycle path includes setting a closed pipe path that is physically completely isolated from the refrigeration heat cycle, the heat medium filled inside is selected from one of carbon dioxide, freon or air, and a controllable variable-speed flow is realized through a built-in multi-stage micro pump; This heating path is preset with multiple heat exchange branch sections, and each branch section has a one-way heat conduction performance to prevent heat from flowing back into the main heat exchange condensation side; In the control strategy, by calculating the average rate of latent heat released by the heat medium per unit length in the current environment and matching it with the current estimated frost thickness of the heat exchange surface, the flow rate and heating duration are dynamically selected; The flow direction of the heat medium is adjusted in real time according to the platform attitude information, and is preferentially guided to the branch section in the advantageous direction of gravity defrosting to improve the frost layer 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 recognition strategy for judging whether the heating target area is in the advantageous position of natural defrosting includes vector projection of the layout direction of the heat exchange pipes of the platform in the three-dimensional attitude angles, namely pitch, yaw and roll, and using the gravity vector as a reference, and determining the most favorable potential defrosting direction by calculating the angle between the axis of the heat exchange pipe and the gravity vector; When the included angle is less than the set first angle threshold, it is considered that the heat exchange section has the ability of natural defrosting; if all areas do not meet the requirements, it enters the waiting state and is re-evaluated whether to be activated until the environmental parameters are updated in the next cycle.

5. The intelligent control method for latent heat defrosting according to claim 4, characterized in that The thermal modeling algorithm for predicting the latent heat transfer rate includes a coupled calculation of the heat medium flow rate, initial temperature, environmental temperature-humidity ratio, and current frost layer thickness, and uses one of a set of classical heat transfer models for modeling, including one-dimensional steady-state heat conduction model, unsteady-state phase change model, or 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 defrosting time period is dynamically adjusted.

6. The latent heat defrosting intelligent control method according to claim 5, characterized in that, The intermittent heating strategy adopts a dynamic duty cycle control algorithm, and sets the opening and closing ratio within the heating cycle according to the current redundant energy capacity, peak and valley of task load, and external air pressure fluctuation rate. This duty cycle is limited by the power control threshold and allows automatic adjustment at different stages; The heating cycle is divided into two stages: the activation period and the cooling period. The activation period is carried out after the platform task management layer allocates available energy, and the cooling period is adjusted according to the change trend of the external air temperature to prevent condensation and refreezing during the defrosting process.

7. A latent heat defrosting intelligent control method according to claim 6, characterized in that, The defrosting state 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 time stamp of the current heating time point. This recorded information forms a defrosting breakpoint state structure and is written into the platform task data storage area; When the energy is restored, first read the previous breakpoint information, and estimate the current required compensation heat according to the unfinished defrosting area, the consumed heating energy, and the remaining thickness of the frost layer, and compare it with the latent heat release rate of the heat medium to judge whether it has the ability to execute; If the condition for continuing to execute is met, directly resume the heating process of the corresponding heat exchange pipe section, otherwise delay entering the next cycle.

8. A latent heat defrosting intelligent control method according to claim 7, characterized in that The method for obtaining the platform redundant energy capacity includes comparing the total power demand of the current platform tasks with the energy supply capacity in real time, and judging whether to allow the activation of the heating operation according to the redundant capacity threshold, and the heating control sub-process is only executed under the condition that the platform task scheduling mechanism issues a non-interference signal.

9. The intelligent control method for latent heat defrosting according to claim 8, characterized in that, The completion standard of the closed-loop defrosting control is jointly determined based on the heat exchange surface temperature recovery rate, the frost layer shedding detection result, and the statistical result of the remaining unprocessed area, and the judgment is based on meeting the termination threshold; The surface temperature is detected in real time by multiple close-fitting temperature-sensitive sensors, and compared with the initial non-frosting temperature to calculate whether the heating rate is stably tending to zero; The shedding of the frost layer is judged whether the physical peeling is completed through the feedback of micro-vibration acceleration; The unprocessed area is distinguished according to the temperature difference distribution before and after the comparison of the thermal images; When all three indicators reach the preset threshold, it is automatically determined that the defrosting is completed, and the task record is cleared.

10. A latent heat defrosting intelligent control system for implementing a latent heat defrosting intelligent control method according to any one of claims 1-9, characterized in that, Including: The environmental perception unit is used to establish a multi-parameter environmental perception model including air pressure, temperature, humidity, heat exchange surface state, and platform attitude angle, and judge whether the critical condition for starting defrosting is reached based on this model; The heat medium guiding unit is used to select at least one of carbon dioxide, Freon, or air as the heat medium, and guide the heat medium 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. A directional heating unit, which is used to dynamically judge the gravity direction based on the current platform attitude angle for a heat exchange area formed by heat exchange tubes independently arranged structurally, and select the pipe sections in the advantageous positions in the natural defrosting direction to preferentially start heating; A power control unit, which is used during the heating process to determine the target defrosting time period according to the predicted phase change latent heat transfer rate, and perform intermittent heating on the selected heat exchange pipe sections, controlling the average power within each heating cycle not to exceed the current platform redundant energy capacity; A state continuous control unit, which is used to pause the heating process when detecting insufficient energy, record the completed defrosting duration and area information, and continue to execute the remaining defrosting process after the energy is restored until the closed-loop defrosting control is completed.

Images