An ice layer monitoring, prediction, and adaptive control device for a liquid hydrogen ambient temperature vaporizer

Abstract

This invention relates to the field of cryogenic vaporization and thermal management technology, and discloses an ice layer monitoring, prediction, and adaptive control device for a liquid hydrogen ambient temperature vaporizer. The device's data acquisition unit collects multi-source sensor data on the vaporizer's operating status and the external environment; a meteorological prediction unit acquires meteorological change information for future periods and calculates the comprehensive influence coefficient of meteorology on the vaporizer's operating status; a data fusion and processing unit calculates the ice layer thickness correction value for each monitoring node based on the multi-source sensor data; a threshold judgment unit determines the icing state of the nodes, and after determining that a node is in an icing state, calculates judgment indicators characterizing the icing risk and ice layer development trend; a control response unit switches the de-icing mode and executes the corresponding de-icing control strategy according to the changes in the judgment indicators, switching between high-energy-consumption de-icing mode, low-energy-consumption de-icing mode, and observation mode. This invention reduces the de-icing energy consumption of the vaporizer and improves operational stability and thermal efficiency.

Description

Technical Field

[0001] This invention relates to the field of cryogenic vaporization and thermal management technology, specifically to an ice layer monitoring, prediction and adaptive control device for a liquid hydrogen ambient temperature vaporizer. Background Technology

[0002] Liquid hydrogen ambient temperature vaporizers are key switching devices between liquid hydrogen storage and transportation systems and gaseous hydrogen supply systems. Their vaporization process is highly dependent on external environmental conditions. Due to the extremely low temperature of liquid hydrogen (approximately -253°C), the outer wall of the vaporizer easily exchanges heat with water vapor in the air during operation, forming an ice layer. The continuous accumulation of ice not only significantly reduces heat exchange efficiency but may also lead to increased structural stress, uneven surface temperature distribution, and operational instability, severely impacting the safety and continuity of the vaporization process.

[0003] Most existing liquid hydrogen vaporizers employ a timed overall de-icing method, meaning de-icing is performed at fixed intervals, failing to dynamically adjust based on the actual icing state. While some systems incorporate temperature or ice thickness sensors for monitoring, these mostly only achieve single-node or localized monitoring, making it difficult to achieve precise de-icing control by region or node. This results in high de-icing energy consumption, slow response, and low overall de-icing efficiency.

[0004] Current ice thickness monitoring mainly relies on ultrasonic or capacitive ice thickness sensors. However, due to the uneven distribution of ice along the circumference of the shell, single-point measurements often cannot accurately reflect the overall icing condition, resulting in deviations in the formulation of de-icing strategies and making it difficult to balance energy consumption optimization and de-icing effectiveness.

[0005] Furthermore, vaporizers are often installed outdoors, and their operation is significantly affected by meteorological conditions (such as precipitation, wind speed, and changes in ambient temperature). However, current technologies lack quantitative analysis methods for the impact of meteorological factors, as well as model mechanisms for dynamically correcting meteorological parameters in de-icing control. Summary of the Invention

[0006] To address the technical problems existing in the prior art, this invention provides an ice layer monitoring, prediction, and adaptive control device for liquid hydrogen ambient temperature vaporizers. This device enables real-time monitoring of the icing state on the outer wall of the vaporizer, prediction of the development trend of icing risk, and intelligent switching of de-icing modes. While ensuring vaporization safety and heat exchange performance, it effectively reduces de-icing energy consumption and improves operational stability and thermal efficiency.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] This invention discloses an ice layer monitoring, prediction, and adaptive control device for a liquid hydrogen ambient temperature vaporizer, comprising:

[0009] The data acquisition unit is used to collect multi-source sensor data on the operating status of the vaporizer and the external environment; the multi-source sensor data includes ambient temperature, relative humidity, ambient wind speed, as well as the inner wall temperature, outer surface temperature and ice thickness of each monitoring node of the vaporizer.

[0010] The meteorological forecasting unit is used to acquire meteorological change information for the future cycle and calculate the comprehensive impact coefficient of meteorology on the gasifier's operating status.

[0011] The data fusion and processing unit is used to calculate the ice thickness correction value for each monitoring node based on the multi-source sensor data.

[0012] The threshold judgment unit is used to determine the icing status of each monitoring node of the gasifier. After the judgment node is in an icing state, it calculates the judgment index to characterize the icing risk and ice development trend based on the multi-source sensor data, the ice thickness correction value and the comprehensive influence coefficient. The judgment index includes the icing risk amount, the ice growth rate and the time to reach the thickness threshold.

[0013] The control response unit is used to switch the de-icing mode and execute the corresponding de-icing control strategy according to the change status of the judgment index; the de-icing mode includes high-energy de-icing mode, low-energy de-icing mode and observation mode.

[0014] As a further improvement to the above scheme, the meteorological forecasting unit periodically acquires the predicted precipitation for the next cycle. Predicted wind speed And predicting ambient temperature Based on this, the comprehensive influence coefficient of meteorology on the operating status of the gasifier is calculated. :

[0015] ;

[0016] In the formula, This indicates the impact of precipitation over a future cycle on the operating status of the gasifier. To determine the design baseline precipitation intensity, when This indicates increased precipitation and a higher risk of freezing. This indicates the impact of wind speed on the operating status of the equipment over a future period. For the design reference wind speed, when This indicates that wind speed has decreased, convective heat transfer capacity has declined, and the risk of icing has increased. This indicates the impact of ambient temperature on the operating status of the equipment over a future period. For the design reference temperature, when This indicates that the temperature is decreasing and the risk of icing is increasing. , , These are the weighting coefficients for each meteorological factor.

[0017] As a further improvement to the above scheme, the process by which the data fusion and processing unit calculates the ice thickness correction value for the monitoring node includes:

[0018] The convective heat transfer coefficient and radiative heat transfer coefficient of each monitoring node are calculated using the following formulas:

[0019] ;

[0020] ;

[0021] In the formula, For monitoring nodes The convective heat transfer coefficient, , This represents the total number of monitoring nodes for the vaporizer. For monitoring nodes The radiative heat transfer coefficient; The current ambient wind speed; The baseline natural convection coefficient; This is the wind speed gain coefficient; It is the Stefan-Boltzmann constant; The emissivity is determined by the output of the ice thickness sensor in the data acquisition unit to determine whether there is ice on the surface of the monitoring node. When the surface of the monitoring node is not frozen, the emissivity of the composite material layer is taken; when it is frozen, the emissivity of the ice layer is taken. For monitoring nodes The external surface temperature; The ambient temperature;

[0022] By combining the energy balance formula and the thermal resistance formula, the estimated ice thickness at each monitoring node of the gasifier is obtained analytically. The calculation formula is as follows:

[0023] ;

[0024] In the formula, For monitoring nodes The temperature of the inner wall surface; The finning ratio of the finned tube; For monitoring nodes The thermal resistance of the composite wall layer; For composite materials Layer thickness; For composite materials The thermal conductivity of the layer; The thermal conductivity of the ice layer; For monitoring nodes Estimated ice thickness;

[0025] By weighted and fused with the estimated ice thickness and the ice thickness measured by the data acquisition unit, the corrected ice thickness value for the monitoring node is obtained. The calculation formula is as follows:

[0026] ;

[0027] In the formula, For monitoring nodes The ice thickness correction value; These are the weighting coefficients.

[0028] As a further improvement to the above scheme, the control response unit is also used to calculate the percentage change in the ice thickness correction value after de-icing at the monitoring node:

[0029] ;

[0030] In the formula, For monitoring nodes The percentage change in the ice thickness correction value; For monitoring nodes The corrected ice thickness value obtained in the most recent monitoring cycle before the start of this de-icing operation. For monitoring nodes The ice thickness correction value obtained in the first monitoring cycle after the completion of this de-icing operation;

[0031] like If the change value is less than the preset threshold, the de-icing effect of the monitoring node is determined to be insufficient, indicating a prediction bias. In this case, a machine learning algorithm is used to adjust the weighting coefficients. , , as well as Make corrections.

[0032] As a further improvement to the above solution, the threshold determination unit is based on the ambient temperature measured by the data acquisition unit. and relative humidity Calculate dew point temperature The icing status of each monitoring node of the vaporizer is determined, and the determination relationship is as follows:

[0033] ;

[0034] In the formula, For monitoring nodes external surface temperature, , This represents the total number of monitoring nodes for the vaporizer. Indicates monitoring node The freezing state; when When the node is frozen, it is determined that the node is in an iced state; when When this occurs, the node is determined to be in a non-icing state.

[0035] As a further improvement to the above scheme, the formula for calculating the icing risk is as follows:

[0036] ;

[0037] In the formula, For monitoring nodes The risk of icing , This represents the total number of monitoring nodes for the vaporizer. This is the comprehensive influence coefficient of the meteorological conditions on the operating status of the vaporizer; For reference temperature; For monitoring nodes The external surface temperature;

[0038] The formula for calculating the ice layer growth rate is:

[0039] ;

[0040] In the formula, For monitoring nodes The rate of ice layer growth; The monitoring time within one cycle;

[0041] The formula for calculating the time required to reach the thickness threshold is:

[0042] ;

[0043] In the formula, The preset ice thickness threshold is used; For monitoring nodes The amount of time required to reach the thickness threshold.

[0044] As a further improvement to the above solution, the triggering conditions for the control response unit to switch the de-icing mode include:

[0045] When a monitoring node is determined to be in an icing state and meets any of the following conditions, the monitoring node will be switched to high-energy de-icing mode:

[0046] (i) The risk of freezing within a continuously preset time period is higher than the set first temperature risk judgment threshold.

[0047] (ii) The rate of ice growth exceeds the set rate threshold;

[0048] (iii) The time taken to reach the thickness threshold did not exceed the set first duration threshold;

[0049] When a monitoring node is determined to be in an icing state and meets any of the following conditions, the monitoring node will be switched to low-energy de-icing mode:

[0050] (i) The freezing risk level within a continuous preset time period shall not be higher than the first temperature risk judgment threshold and not lower than the set second temperature risk judgment threshold.

[0051] (ii) The ice layer growth rate is higher than 0 and not higher than the aforementioned rate threshold;

[0052] (iii) The time taken to reach the thickness threshold exceeds the first duration threshold but does not exceed the set second duration threshold;

[0053] When the triggering conditions for high-energy and low-energy de-icing modes are not met, switch to observation mode, do not perform active de-icing operations, and only maintain ice layer monitoring and data recording.

[0054] As a further improvement to the above solution, the device further includes:

[0055] Control valve, used to control the flow rate of liquid hydrogen into the vaporizer;

[0056] The warm and dry gas purging and de-icing mechanism includes multiple warm and dry gas purging ports corresponding to the number of monitoring nodes;

[0057] The jet impact de-icing mechanism includes multiple jet nozzles corresponding to the number of monitoring nodes; the power of the jet impact de-icing mechanism is greater than that of the warm dry gas purging de-icing mechanism.

[0058] As a further improvement to the above scheme, the control response unit is also used to activate the corresponding jet nozzle in the jet impact de-icing mechanism when the monitoring node is determined to be in a high-energy-consumption de-icing mode, and reduce the liquid hydrogen flow rate through the control valve to reduce the heat exchange load; the control response unit is also used to activate the corresponding warm dry gas purging port in the warm dry gas purging de-icing mechanism when the monitoring node is determined to be in a low-energy-consumption de-icing mode.

[0059] As a further improvement to the above solution, the data acquisition unit includes:

[0060] Temperature sensor 1, used to measure ambient temperature ;

[0061] Humidity sensor, used to measure relative humidity ;

[0062] Anemometers are used to measure ambient wind speed. ;

[0063] Vaporizers are evenly arranged axially There are 1 monitoring node, and the following sensors are installed at each monitoring node:

[0064] Temperature sensor two is used to measure the inner wall temperature of the gasifier monitoring node. ;

[0065] Temperature sensor three is used to measure the outer surface temperature of the gasifier monitoring node. ;

[0066] Ultrasonic ice thickness sensor used to measure ice thickness at gasifier monitoring nodes. .

[0067] Compared with the prior art, the beneficial effects of the present invention are:

[0068] 1. This invention realizes a multi-node, multi-parameter fusion monitoring mechanism, which can comprehensively reflect the operating status and icing characteristics of the vaporizer. By deploying various sensors such as temperature, humidity, wind speed, and ice thickness on the vaporizer exterior and combining them with internal operating parameters, multi-dimensional real-time data acquisition and information fusion are achieved. The device can accurately assess the temperature field distribution, heat transfer characteristics, and ice growth trends of different axial or circumferential nodes, providing accurate status judgment basis for de-icing control. Based on this, a meteorological influence correction coefficient is constructed to realize real-time correction of icing risk caused by meteorological changes. The icing judgment index can be dynamically adjusted according to real-time meteorological conditions, thereby improving prediction accuracy and the foresight of the de-icing strategy. Based on the prediction results, high-energy de-icing, low-energy de-icing, or observation modes are adaptively selected to provide differentiated responses to the icing status of different nodes, accurately de-icing and optimizing energy allocation, significantly reducing de-icing energy consumption and ensuring the safety and continuity of vaporizer operation.

[0069] 2. This invention effectively improves the accuracy and stability of ice thickness monitoring and prediction by fusing analytical calculation results with sensor measurement data. The device calculates the analytical ice thickness based on an energy balance and thermal resistance model, and then weights and fuses this calculation with the ultrasonic measurement values ​​to form a comprehensive ice thickness result. This method can significantly reduce measurement errors caused by uneven ice distribution and improve the reliability of overall ice thickness identification.

[0070] 3. This invention, through a comprehensive technical approach of "multi-node monitoring + meteorological correction + dynamic control + parameter optimization," achieves intelligent de-icing and optimal energy consumption control of liquid hydrogen ambient air vaporizers in complex environments. It boasts advantages such as high monitoring accuracy, fast response speed, significant energy efficiency optimization, and strong system safety, providing crucial technical support for the efficient and safe operation of cryogenic fuel systems. Furthermore, the device of this invention exhibits excellent versatility and scalability, adaptable to liquid hydrogen ambient air vaporizers of different sizes, structures, and brands, without relying on specific equipment structures or operating platforms. Through multi-node deployment of monitoring sensors and modular control design, it can be flexibly applied to various scenarios such as liquid hydrogen refueling, storage and transportation, and aerospace propulsion, demonstrating strong engineering adaptability and widespread application value. Attached Figure Description

[0071] Figure 1 This is a framework diagram of the ice layer monitoring, prediction, and adaptive control device for a liquid hydrogen ambient temperature vaporizer in an embodiment of the present invention.

[0072] Figure 2 This is a schematic diagram of the ice layer monitoring, prediction and adaptive control device for the liquid hydrogen air-temperature vaporizer in Embodiment 1 of the present invention.

[0073] In the diagram, 1. Vaporizer; 2. Temperature sensor one; 3. Humidity sensor; 4. Anemometer; 5. Temperature sensor two; 6. Temperature sensor three; 7. Ultrasonic ice thickness sensor; 8. Warm and dry gas purging de-icing mechanism; 9. Jet impact de-icing mechanism; 10. Liquid hydrogen storage tank; 11. Control valve; 12. Gas hydrogen storage tank; 13. Central server. Detailed Implementation

[0074] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0075] Example 1

[0076] Please see Figure 1 and Figure 2 This embodiment provides an ice layer monitoring, prediction, and adaptive control device for a liquid hydrogen ambient air vaporizer, including: a data acquisition unit, a meteorological prediction unit, a data fusion and processing unit, a threshold judgment unit, and a control response unit. It may also include a control valve 11, a warm dry gas purging de-icing mechanism 8, and a jet impact de-icing mechanism 9. This device is used for ice layer monitoring and de-icing control of the liquid hydrogen ambient air vaporizer 1.

[0077] Figure 2 The vaporizer 1 in the diagram illustrates two heat exchange pipes. One end of each heat exchange pipe is connected to a liquid hydrogen storage tank 10 via a pipeline, and a control valve 11 for controlling the liquid hydrogen flow rate is installed on this section of the pipeline. The other end of the heat exchange pipe is connected to a gaseous hydrogen storage tank 12 via a pipeline. Taking one of the heat exchange pipes (on the right) as an example, four monitoring nodes are equidistantly arranged along the axial direction (i.e., the vertical direction in the diagram).

[0078] Among them, the meteorological forecasting unit, data fusion and processing unit, threshold judgment unit, and control response unit can all be implemented by the central server 13 to perform data processing, calculation, and discrimination functions.

[0079] The warm-dry air purging de-icing mechanism 8 includes four parallel warm-dry air purging ports, which selectively provide low-velocity dry air at a set temperature (20-40°C) to four monitoring nodes according to control commands. This airflow is mild and stable, and is mainly used to dry, soften, and slowly melt the nascent or medium-thickness ice layer at the nodes, so as to achieve de-icing under low-energy conditions.

[0080] The jet impact de-icing mechanism 9 comprises four parallel jet nozzles; its output power is significantly higher than that of the warm dry air purging de-icing mechanism. Each jet nozzle can provide a high-velocity (50–120 m / s), high-kinetic-energy air jet to the corresponding monitoring node. This jet primarily relies on strong mechanical impact to rapidly peel off and break up thick or firmly attached ice layers, making it suitable for high-risk or emergency icing conditions. Its de-icing mechanism is mainly based on kinetic energy impact, rather than relying on air heating for melting.

[0081] In this embodiment, to achieve differentiated de-icing control across multiple regions, the de-icing execution ports (including high-power jet nozzles and warm-dry air purging ports) corresponding to each monitoring node are arranged in parallel, and are independently scheduled node by node by the control unit of the central server 13. Therefore, the de-icing operations of each node do not interfere with each other, and can enter high-energy-consumption or low-energy-consumption de-icing modes according to their individual ice layer determination results, thereby achieving zoned, precise, and efficient de-icing control.

[0082] The data acquisition unit is used to collect multi-source sensor data on the operating status of the vaporizer and the external environment. The multi-source sensor data includes ambient temperature, relative humidity, ambient wind speed, as well as the inner wall temperature, outer surface temperature and ice thickness of each monitoring node of the vaporizer.

[0083] The data acquisition unit includes:

[0084] Temperature sensor 2 is used to measure ambient temperature. ;

[0085] Humidity sensor 3 is used to measure relative humidity. ;

[0086] Anemometer 4 is used to measure ambient wind speed. ;

[0087] Vaporizers are evenly arranged axially There are 1 monitoring node, and the following sensors are installed at each monitoring node:

[0088] Temperature sensor 25 is used to measure the inner wall temperature of the gasifier monitoring node. Temperature sensor 25 can be a surface-mount temperature sensor.

[0089] Temperature sensor 36 is used to measure the outer surface temperature of the gasifier monitoring node. Temperature sensor 36 can be an infrared temperature sensor.

[0090] Ultrasonic ice thickness sensor 7 is used to measure the ice thickness at the gasifier monitoring node. .

[0091] The meteorological forecasting unit is used to obtain meteorological change information in the future cycle and calculate the comprehensive influence coefficient of meteorology on the operating status of gasifier 1.

[0092] The meteorological forecasting unit periodically acquires the predicted precipitation for the next period (e.g., 10 minutes). Predicted wind speed And predicting ambient temperature It should be noted that the data acquisition unit obtains the real-time wind speed at the current moment. This reflects the actual environmental conditions experienced by vaporizer 1 during the current monitoring cycle; while the meteorological forecasting unit provides the estimated wind speed for the next monitoring cycle, which can be understood as a prediction of the wind speed change trend in the next cycle. This is used to correct the icing trend calculation and de-icing strategy in advance, and to predict the ambient temperature. Similarly, the comprehensive influence coefficient of meteorological conditions on the operating status of gasifier 1 is calculated accordingly. :

[0093] ;

[0094] In the formula, This indicates the impact of precipitation over a future cycle on the operating status of gasifier 1. To determine the design baseline precipitation intensity, when This indicates increased precipitation and a higher risk of freezing. This indicates the impact of wind speed on the operating status of the equipment over a future period. For the design reference wind speed, when This indicates that wind speed has decreased, convective heat transfer capacity has declined, and the risk of icing has increased. This indicates the impact of ambient temperature on the operating status of the equipment over a future period. For the design reference temperature, when This indicates that the temperature is decreasing and the risk of icing is increasing. , , The weighting coefficients for each meteorological factor are preferably normalized and limited to non-negative constants, used to adjust the overall influence intensity of different meteorological factors on the gasifier's operating status.

[0095] The data fusion and processing unit is used to calculate the ice thickness correction value for each monitoring node based on multi-source sensor data.

[0096] The process by which the data fusion and processing unit calculates the ice thickness correction value for the monitoring nodes includes:

[0097] The convective heat transfer coefficient and radiative heat transfer coefficient of each monitoring node are calculated using the following formulas:

[0098] ;

[0099] ;

[0100] In the formula, For monitoring nodes The convective heat transfer coefficient, , This represents the total number of monitoring nodes for the vaporizer. For monitoring nodes The radiative heat transfer coefficient; The current ambient wind speed; The baseline natural convection coefficient; This is the wind speed gain coefficient; It is the Stefan-Boltzmann constant; The emissivity is determined by the output of the ultrasonic ice thickness sensor 7 in the data acquisition unit to determine whether there is an ice layer on the surface of the monitoring node. When the surface of the monitoring node is not frozen, the emissivity of the composite material layer is taken; when it is frozen, the emissivity of the ice layer is taken. For monitoring nodes The external surface temperature; The ambient temperature.

[0101] It should be noted that here, the presence of an ice layer on the node surface is determined based on the output of the ultrasonic ice thickness sensor 7, and the emissivity value is obtained accordingly; subsequently... The value of 1 or 0 is determined based on the relationship between the node surface temperature and the dew point temperature, and is used to identify whether the current conditions are icing and whether an icing risk assessment is required.

[0102] By combining the energy balance formula and the thermal resistance formula, the estimated ice thickness at each monitoring node of gasifier 1 is obtained analytically. The calculation formula is as follows:

[0103] ;

[0104] In the formula, For monitoring nodes The temperature of the inner wall surface; The finning ratio of the finned tube; For monitoring nodes The thermal resistance of the composite wall layer; For composite materials Layer thickness; For composite materials The thermal conductivity of the layer; The thermal conductivity of the ice layer; For monitoring nodes The estimated value of the ice layer thickness.

[0105] It should be noted that the parameters involved in the above formula are all defined based on the heat conduction model of the local pipe wall section of the vaporizer corresponding to monitoring node j. Monitoring node j does not refer to a section of pipe, but rather to a fixed measuring point location (e.g., 1 to N locations evenly distributed along the axial direction) on the outer wall of the vaporizer. Each monitoring node corresponds to a typical wall thickness section of the outer wall of the vaporizer, where temperature measurements and ice layer thickness calculations are performed. The composite wall thermal resistance in the formula... This represents the total thermal resistance of all wall material layers that heat must pass through to transfer from the inner wall to the outer wall (or the outer surface of the ice layer) at the cross-section corresponding to monitoring node j. Since the vaporizer pipe wall has a consistent axial structure, the thickness of each material layer remains the same at each monitoring node; therefore, the formula uses... Indicates the thickness of the k-th layer of material, without using Distinguish between nodes.

[0106] By weighted and fused with the estimated ice thickness and the ice thickness measured by the data acquisition unit, the corrected ice thickness value for the monitoring node is obtained. The calculation formula is as follows:

[0107] ;

[0108] In the formula, For monitoring nodes The ice thickness correction value; These are the weighting coefficients. This fusion calculation enables dynamic correction of theoretical predictions, improving the accuracy and reliability of ice thickness monitoring results.

[0109] The threshold judgment unit is used to judge the icing status of each monitoring node of the vaporizer 1. After the judgment node is in the icing state, the judgment index is calculated based on multi-source sensor data, ice thickness correction value and comprehensive influence coefficient to characterize the icing risk and ice development trend. The judgment index includes icing risk, ice growth rate and time to reach the thickness threshold.

[0110] The threshold judgment unit is based on the ambient temperature measured by the data acquisition unit. and relative humidity Calculate dew point temperature The icing status of each monitoring node of vaporizer 1 is determined, and the determination relationship is as follows:

[0111] ;

[0112] In the formula, For monitoring nodes external surface temperature, , This represents the total number of monitoring nodes for vaporizer 1; Indicates monitoring node The freezing state; when When the node is frozen, it is determined that the node is in an iced state; when When this occurs, the node is determined to be in a non-icing state.

[0113] The formula for calculating the risk of icing is:

[0114] ;

[0115] In the formula, For monitoring nodes The risk of icing , This represents the total number of monitoring nodes for vaporizer 1; The comprehensive influence coefficient of meteorological conditions on the operating status of vaporizer 1; The reference temperature is the baseline temperature that characterizes the icing tendency of the outer wall of the vaporizer. Its value is determined based on the condensation and icing experimental data of the liquid hydrogen ambient air vaporizer under typical conditions, combined with engineering operation experience. This temperature is usually set at around -5℃ to reflect the critical condition when water vapor in the air begins to sublimate and ic. For monitoring nodes The external surface temperature;

[0116] The formula for calculating the rate of ice layer growth is:

[0117] ;

[0118] In the formula, For monitoring nodes The rate of ice layer growth; The monitoring time within one cycle;

[0119] The formula for calculating the time required to reach the thickness threshold is:

[0120] ;

[0121] In the formula, The preset ice thickness threshold is used; For monitoring nodes The amount of time required to reach the thickness threshold.

[0122] The control response unit is used to switch the de-icing mode and execute the corresponding de-icing control strategy according to the change of the judgment index; the de-icing modes include high energy consumption de-icing mode, low energy consumption de-icing mode and observation mode.

[0123] The trigger conditions for the control response unit to switch the de-icing mode include:

[0124] When a monitoring node is determined to be in an icing state and meets any of the following conditions, the monitoring node will be switched to high-energy de-icing mode:

[0125] (i) The risk of freezing within a continuously preset time period is higher than the set first temperature risk judgment threshold.

[0126] (ii) The rate of ice growth exceeds the set rate threshold;

[0127] (iii) The time taken to reach the thickness threshold did not exceed the set first duration threshold.

[0128] In this mode, the device activates a high-power jet impingement de-icing mechanism and gradually reduces the flow of liquid hydrogen through a control valve to lower the heat exchange load; when the above conditions are no longer met, the device automatically exits the high-energy-consumption de-icing mode.

[0129] When a monitoring node is determined to be in an icing state and meets any of the following conditions, the monitoring node will be switched to low-energy de-icing mode:

[0130] (i) The freezing risk level within a continuous preset time period shall not be higher than the first temperature risk judgment threshold and not lower than the set second temperature risk judgment threshold.

[0131] (ii) The ice layer growth rate is higher than 0 and not higher than the rate threshold;

[0132] (iii) The time taken to reach the thickness threshold exceeds the first duration threshold but does not exceed the set second duration threshold.

[0133] In this mode, the device activates a low-power warm dry gas purging and de-icing mechanism to achieve energy-optimized icing suppression; when the above conditions are no longer met, the device automatically exits the low-power de-icing mode.

[0134] When the triggering conditions for high-energy and low-energy de-icing modes are not met, switch to observation mode, do not perform active de-icing operations, and only maintain ice layer monitoring and data recording.

[0135] The control response unit is also used to activate the corresponding jet nozzle in the jet impact de-icing mechanism 9 when the monitoring node is determined to be in the high-energy-consumption de-icing mode, and reduce the liquid hydrogen flow rate through the control valve 11 to reduce the heat exchange load; the control response unit is also used to activate the corresponding warm dry gas purging port in the warm dry gas purging de-icing mechanism 8 when the monitoring node is determined to be in the low-energy-consumption de-icing mode.

[0136] The control response unit is also used to calculate the percentage change in the ice thickness correction value after de-icing at the monitoring node:

[0137] ;

[0138] In the formula, For monitoring nodes The percentage change in the ice thickness correction value; For monitoring nodes The corrected ice thickness value obtained in the most recent monitoring cycle before the start of this de-icing operation. For monitoring nodes The ice thickness correction value obtained in the first monitoring cycle after the completion of this de-icing operation;

[0139] like If the change value is less than the preset threshold, the de-icing effect of the monitoring node is determined to be insufficient, indicating a prediction bias. In this case, a machine learning algorithm is used to adjust the weighting coefficients. , , as well as Make corrections.

[0140] This embodiment also verifies the working principle of the ice layer monitoring, prediction and adaptive control device for the above-mentioned liquid hydrogen ambient temperature vaporizer 1, as well as its real-time monitoring, risk prediction and de-icing response functions. It realizes the entire process of monitoring the ice layer thickness on the outer wall of vaporizer 1, predicting the icing risk trend and dynamically determining and responding to the de-icing mode under multi-node conditions.

[0141] like Figure 2 As shown, the test object is an ambient temperature vaporizer 1 used in a liquid hydrogen refueling system. This vaporizer 1 adopts a composite wall structure, composed of aluminum layers and stainless steel layers arranged sequentially. The thickness of the aluminum layer is... Thermal conductivity is Stainless steel layer thickness Thermal conductivity is The vaporizer 1 has an outer fin ratio of [missing information]. =5.

[0142] Regarding surface radiation characteristics, when the vaporizer 1 is in an ice-free state, the emissivity of the outer wall surface is taken as... Once the ice layer forms, the surface emissivity is taken as... This will more accurately reflect the actual characteristics of radiative heat transfer.

[0143] like Figure 2 As shown, four monitoring nodes are evenly arranged along the axial direction on the outer surface of the vaporizer 1. Each node is equipped with an inner wall temperature sensor 5, an outer surface infrared temperature sensor 6, and an ultrasonic ice thickness sensor 7, for real-time acquisition of temperature field distribution and icing state parameters during the operation of the vaporizer 1. The main input parameters of the experiment are shown in Table 1.

[0144] Table 1: Main Input Parameters of the Experiment

[0145] .

[0146] I. Weather Correction and Icing Determination

[0147] Taking node 1 as an example, the meteorological correction coefficient is calculated by combining the meteorological forecast data for the next 10 minutes (changes in precipitation, wind speed, and ambient temperature). .

[0148] According to the formula:

[0149] ;

[0150] Among them, the impact of precipitation in the next cycle Wind speed influence Temperature effect Take weight parameters , , Substituting into the calculation, we get:

[0151] ;

[0152] This indicates that weather conditions increase the risk of icing by approximately 8%.

[0153] Based on the ambient temperature and relative humidity, the dew point temperature is calculated using the dew point formula. Node outer wall temperature ,satisfy and Therefore, the node is determined to be in an icing state, and the icing determination quantity is:

[0154] .

[0155] II. Calculation of Heat Transfer Parameters and Ice Thickness

[0156] The calculated heat transfer parameters for node 1 are as follows. (Based on the external wind speed of vaporizer 1) The convective heat transfer coefficient is calculated using the empirical formula as follows:

[0157] ;

[0158] The radiative heat transfer coefficient is determined according to the Stefan-Boltzmann law:

[0159] ;

[0160] in, It is the Stefan-Boltzmann constant. The surface emissivity of the ice layer.

[0161] Based on the energy balance equation and the thermal resistance model, the analytical ice layer thickness at node 1 can be obtained by the following formula:

[0162] ;

[0163] Substituting the parameters, the analytical ice thickness at node 1 is obtained as follows:

[0164] ;

[0165] Combining sensor measurements with analytical calculations, a weighted fusion algorithm is used to obtain the overall ice thickness at node 1:

[0166] ;

[0167] in, The ice thickness was measured using an ultrasonic sensor. These are weighting coefficients used to balance the weights of measured values ​​and model predictions in the overall result.

[0168] This fusion algorithm can effectively correct local measurement errors and improve the accuracy and stability of ice thickness monitoring results.

[0169] III. Threshold Judgment Unit Calculation

[0170] The threshold judgment unit of node 1 is used to calculate the icing risk, ice growth rate and time required to reach the preset thickness threshold, so as to comprehensively evaluate the icing development trend and de-icing response priority of the node, and provide a decision basis for subsequent control mode switching.

[0171] After determining that node 1 is in an icy state (Freeze1=1), the meteorological correction factor is first applied. and node outer surface temperature Calculate the icing risk of node 1:

[0172] ;

[0173] Among them, reference temperature The calculation results show that node 1 is currently in a significantly high-risk icing state, with a clear icing trend.

[0174] Ice growth rate Determined by the ratio of ice thickness change to monitoring cycle time:

[0175] ;

[0176] in, mm of the total ice thickness for this period The ice thickness in the previous period, monitoring period The results show that the ice layer at node 1 exhibits a continuous growth trend.

[0177] Time required to reach the preset ice thickness threshold Calculated based on current ice thickness, growth rate, and meteorological correction factors:

[0178] ;

[0179] Among them, the preset ice thickness threshold This result indicates that without de-icing measures, the ice layer at node 1 will reach its critical thickness in approximately 1.5 hours.

[0180] Comprehensive analysis reveals a significant temperature difference between the outer surface of node 1 and the reference temperature, indicating a substantial risk of icing. Furthermore, the ice layer is growing rapidly (approximately 4.17 mm / h), and is expected to reach the preset critical thickness within a short period. These results demonstrate that the node is in a rapid icing development phase, exhibiting clear high-risk characteristics, and providing sufficient basis for the control response unit to trigger a high-energy-consumption de-icing mode.

[0181] IV. De-icing Mode Determination and Response

[0182] Based on the threshold judgment logic, the device combines the real-time monitoring results of node 1 to dynamically identify and respond to the de-icing mode. The specific judgment criteria are as follows:

[0183] a. Frozen state: Node 1 is in a frozen state (Freeze1=1);

[0184] b. Risk of icing: ,in This indicates that the node is in a high-risk freezing state;

[0185] c. Growth rate: This indicates that the ice layer is growing at a relatively fast rate;

[0186] d. Critical time This indicates that the ice layer will approach its critical thickness in a short period of time.

[0187] When node 1 is in an icing state (condition a) and meets any one of the high-risk judgment conditions (conditions b, c, or d), the system control unit automatically enters the high-energy-consumption de-icing mode. In this mode, the high-power jet impact de-icing mechanism 9 is immediately activated to perform rapid de-icing on the outer wall of the vaporizer 1; at the same time, the liquid hydrogen flow rate is moderately adjusted by the control valve 11 to reduce the local heat exchange load and prevent secondary icing and structural thermal shock caused by excessive temperature difference, thereby ensuring the safety and stability of the de-icing process.

[0188] When the monitored parameters return to a safe range, meaning that the aforementioned high-risk conditions (b, c, d) are no longer met, the device automatically exits the high-power de-icing mode and switches to low-energy de-icing or observation mode. This adaptive switching mechanism achieves closed-loop control of the de-icing process, ensuring a dynamic balance between efficient heat transfer and safe operation in the liquid hydrogen vaporization process.

[0189] V. Statistics of all nodes

[0190] Based on the calculation steps and judgment logic of node 1, nodes 2, 3, and 4 are calculated synchronously to obtain the icing state, icing risk, ice growth rate, and time required to reach the preset thickness threshold for each node. The comprehensive calculation and de-icing mode judgment results for each node are summarized in Table 2.

[0191] Table 2: Calculation results and de-icing mode determination for each monitoring node

[0192] ;

[0193] As shown in Table 2, the icing risk of both node 1 and node 2 is significantly higher than the temperature risk assessment threshold. The temperature was +10℃, and the ice growth rate exceeded 3 mm / h, with the predicted time to reach the thickness threshold being less than 2 hours. Based on the high-risk judgment logic, the device automatically determined that the two nodes were in a high-risk state and entered a high-energy-consumption de-icing mode, activating the high-power jet impact de-icing mechanism 9 to quickly reduce the ice thickness and restore normal heat transfer capacity.

[0194] Although nodes 3 and 4 are in an icy state, their risk levels and growth rates are relatively low, with the predicted time to reach the critical thickness both falling within the range of 2 to 12 hours. Specifically, the risk level of node 4 is slightly below the temperature risk assessment threshold. Although the growth rate and time conditions are still met, the device determines both nodes to be in low-energy de-icing mode and automatically starts the warm dry gas purging device to maintain local de-icing and energy consumption optimization, thereby achieving refined de-icing control by region and node.

[0195] In this embodiment, the central server 13 periodically collects and updates the ice layer determination parameters of each monitoring node, including the icing state quantity. , risk of icing Ice layer growth rate and the amount of time to reach the thickness threshold The server uses comprehensive data from multiple nodes to determine whether the current de-icing effect has met expectations, and adjusts the de-icing control strategy for the next monitoring cycle accordingly.

[0196] To achieve a quantitative evaluation of the de-icing effect, this embodiment introduces the percentage change in ice thickness. As an indicator for judging the deviation of de-icing effect, it is defined as:

[0197] ;

[0198] in, This is the corrected ice thickness value obtained for this node in the most recent monitoring cycle before the start of this de-icing operation. This is the corrected ice thickness value obtained in the first monitoring cycle after the completion of this de-icing operation. If the de-icing effect of the node is below a preset threshold (e.g., 10% to 20%), it is determined that the node's de-icing effect is insufficient and there is a prediction bias.

[0199] When the device detects When the de-icing effect is inconsistent with the model prediction due to the value being below the threshold, a machine learning algorithm will be automatically invoked to adjust the meteorological impact coefficient. Weight parameters in and the weighting coefficient in the comprehensive ice thickness fusion calculation Adaptive correction is performed. The above algorithm uses measured data after de-icing as feedback, and achieves adaptive optimization of the model through deviation quantification, parameter updating and iterative learning, thereby improving the accuracy of ice layer monitoring, the reliability of icing trend prediction and the adaptability of de-icing strategy.

[0200] VI. Results Analysis

[0201] From the above calculations and experimental results, it can be seen that:

[0202] (1) By introducing meteorological correction coefficients and dynamic weighting algorithms, this invention can accurately identify the formation and growth trend of ice layer in the early stage of icing (ice layer thickness of about 2 to 5 mm), and effectively predict the risk of icing in advance;

[0203] (2) The device can complete the closed-loop operation of “monitoring-calculation-judgment-control” within every 10-minute monitoring cycle, realizing rapid response and real-time adjustment;

[0204] (3) The proposed de-icing mode has a clear hierarchical logic and can automatically select high-energy or low-energy de-icing strategies according to the risk level differences of each monitoring node, thereby achieving precise de-icing in different zones and avoiding unnecessary energy waste.

[0205] (4) Compared with the traditional timed overall de-icing method, this device can achieve node-level energy consumption optimization while ensuring the de-icing effect, significantly improve the energy utilization efficiency and intelligent level of de-icing management of the liquid hydrogen vaporizer 1, and effectively enhance the safety and stability of the device operation.

[0206] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An ice layer monitoring, predicting and self-adaptive control device for liquid hydrogen air temperature vaporizer, characterized in that, The method comprises the following steps: a data acquisition unit is configured to acquire multi-source sensor data of the running state of the gasifier and the external environment; the multi-source sensor data comprises environmental temperature, relative humidity, environmental wind speed, and inner wall temperature, outer surface temperature, and ice layer thickness of each monitoring node of the gasifier; The weather prediction unit is used to acquire weather change information in a future period and calculate a comprehensive influence coefficient of weather on the running state of the gasifier : wherein, , , are the predicted precipitation, the predicted wind speed and the predicted ambient temperature in the future one period, respectively; represents the influence of the precipitation in the future one period on the operating state of the gasifier, is the design reference precipitation intensity, and when , it indicates that the precipitation is enhanced and the icing risk is increased; represents the influence of the wind speed in the future one period on the operating state of the equipment, is the design reference wind speed, and when , it indicates that the wind speed is weakened, the convective heat exchange capacity is decreased, and the icing risk is increased; represents the influence of the ambient temperature in the future one period on the operating state of the equipment, is the design reference temperature, and when , it indicates that the temperature is decreased and the icing risk is increased; , , are the weight coefficients of the respective meteorological factors; a data fusion and processing unit is configured to calculate an ice layer thickness correction value of each monitoring node according to the multi-source sensor data, and the specific process comprises the following steps: calculating a convective heat transfer coefficient and a radiative heat transfer coefficient of each monitoring node; wherein is the amount of icing risk of the monitoring node , , is the total number of monitoring nodes of the vaporizer; is the reference temperature; is the outside surface temperature of the monitoring node . obtaining an ice layer thickness estimation value of each monitoring node of the gasifier through simultaneous calculation of an energy balance formula and a conduction thermal resistance formula; wherein to monitor the rate of ice layer growth of the node ; is the monitoring time within a cycle obtaining an ice layer thickness correction value of the monitoring node by weighted fusion of the ice layer thickness estimation value and the ice layer thickness measured by the data acquisition unit; In the formula, is a preset ice layer thickness threshold value; is a monitoring node is the amount of time to reach the thickness threshold value; is a monitoring node is the ice layer thickness correction value of the monitoring node a threshold judgment unit is configured to determine the icing state of each monitoring node of the gasifier, and after determining that the node is in the icing state, based on the multi-source sensor data, the ice layer thickness correction value, and the comprehensive influence coefficient, a determination index for representing the icing risk and the ice layer development trend is calculated; the determination index comprises an icing risk amount, an ice layer growth rate, and an amount of time to reach a thickness threshold; wherein the calculation formula of the icing risk amount is:

2. The ice layer monitoring, predicting and self-adaptive control device for liquid hydrogen air temperature vaporizer according to claim 1, characterized in that, the calculation formula of the ice layer growth rate is: the calculation formula of the amount of time to reach the thickness threshold is: wherein is the convective heat transfer coefficient of the monitoring node , , is the total number of monitoring nodes of the vaporizer; is the radiative heat transfer coefficient of the monitoring node ; is the current ambient wind speed; is the reference natural convection coefficient; is the wind speed gain coefficient; is the Stefan-Boltzmann constant; is the emissivity, determined according to the output of the ice layer thickness sensor in the data acquisition unit, whether or not there is an ice layer on the surface of the monitoring node, taking the emissivity of the surface of the composite layer when the monitoring node surface is not iced, and the emissivity of the ice layer when iced; is the outer surface temperature of the monitoring node ; is the ambient temperature; a control response unit is configured to switch the deicing mode and execute a corresponding deicing control strategy according to the change state of the determination index; the deicing mode comprises a high-energy-consumption deicing mode, a low-energy-consumption deicing mode, and an observation mode. wherein is the temperature of the inner wall surface of the monitoring node ; is the finned ratio of the finned tube; is the composite wall surface layer thermal resistance of the monitoring node ; is the thickness of the first layer of the composite material; ; is the thermal conductivity of the first layer of the composite material; ; is the thermal conductivity of the ice layer; is the estimated value of the thickness of the ice layer of the monitoring node ; The process of calculating the ice layer thickness correction value of the monitoring node by the data fusion and processing unit comprises the following steps: In the formula, is the ice layer thickness correction value of the monitoring node ; is the weight coefficient; is the ice layer thickness of the monitoring node acquired by the data acquisition unit.

3. The ice layer monitoring, predicting and self-adaptive control device for liquid hydrogen air temperature vaporizer according to claim 2, characterized in that, calculating a convective heat transfer coefficient and a radiative heat transfer coefficient of each monitoring node, and the calculation formula is: wherein is the ice layer thickness correction value of the monitoring node as a percentage of the change of the ice layer thickness correction value of the monitoring node is the ice layer thickness correction value of the monitoring node obtained in the last monitoring cycle before the start of the present de-icing operation, is the ice layer thickness correction value of the monitoring node obtained in the first monitoring cycle after the end of the present de-icing operation; If the change is less than the preset change threshold, it is determined that the deicing effect of the monitoring node is insufficient, and there is a prediction deviation; at this time, the machine learning algorithm is used to correct the weight coefficients , , and .

4. The ice layer monitoring, predicting and self-adaptive control device for liquid hydrogen air temperature vaporizer according to claim 1, characterized in that, The threshold determination unit is based on the ambient temperature measured by the data acquisition unit. and relative humidity Calculate dew point temperature The icing status of each monitoring node of the vaporizer is determined, and the determination relationship is as follows: wherein monitoring node of the outer surface temperature of the node, , total number of monitoring nodes of the vaporizer; indicates the icing state of the monitoring node ; when , it is determined that the node is in an icing state; when , it is determined that the node is in a non-icing state.

5. The ice layer monitoring, predicting and self-adaptive control device for liquid hydrogen air temperature vaporizer according to claim 1, characterized in that, obtaining an ice layer thickness estimation value of each monitoring node of the gasifier through simultaneous calculation of an energy balance formula and a conduction thermal resistance formula, and the calculation formula is: obtaining an ice layer thickness correction value of the monitoring node by weighted fusion of the ice layer thickness estimation value and the ice layer thickness measured by the data acquisition unit, and the calculation formula is: The control response unit is further configured to calculate the percentage change of the ice layer thickness correction value after deicing of the monitoring node: The triggering conditions for the control response unit to switch the deicing mode comprise: when it is determined that the monitoring node is in the icing state and any of the following conditions is met, the high-energy-consumption deicing mode is switched for the monitoring node: (1) the icing risk amount within a continuous preset time is higher than a set first temperature risk determination threshold; (2) the ice layer growth rate is higher than a set rate threshold; (3) the amount of time to reach the thickness threshold does not exceed a set first time length threshold; when it is determined that the monitoring node is in the icing state and any of the following conditions is met, the low-energy-consumption deicing mode is switched for the monitoring node: (1) the icing risk amount within a continuous preset time is not higher than the first temperature risk determination threshold and not lower than a set second temperature risk determination threshold; (2) the ice layer growth rate is higher than 0 and not higher than the rate threshold; (3) the amount of time to reach the thickness threshold exceeds the first time length threshold and does not exceed a set second time length threshold. When the trigger conditions of the high-energy and low-energy de-icing modes are not met, the observation mode is switched to, no active de-icing operation is performed, and only ice layer monitoring and data recording are maintained.

6. The ice layer monitoring, predicting and self-adaptive control device for liquid hydrogen air temperature vaporizer according to claim 5, characterized in that, Also included are: a control valve for controlling the flow of liquid hydrogen into the gasifier; a warm dry gas purge de-icing mechanism including a plurality of warm dry gas purge ports corresponding in number to the monitoring nodes; a jet impact de-icing mechanism including a plurality of jet injection ports corresponding in number to the monitoring nodes; the power of the jet impact de-icing mechanism is greater than that of the warm dry gas purge de-icing mechanism.

7. The ice layer monitoring, predicting and self-adaptive control device for liquid hydrogen air temperature vaporizer according to claim 6, characterized in that, The control response unit is also configured to, when determining that a monitoring node is in a high-energy de-icing mode, activate the corresponding jet injection port in the jet impact de-icing mechanism and reduce the flow of liquid hydrogen through the control valve to reduce the heat exchange load; and the control response unit is also configured to, when determining that a monitoring node is in a low-energy de-icing mode, activate the corresponding warm dry gas purge port in the warm dry gas purge de-icing mechanism.

8. The ice layer monitoring, predicting and self-adaptive control device for liquid hydrogen air temperature vaporizer according to claim 1, characterized in that, The data acquisition unit includes: a temperature sensor one for measuring the ambient temperature ; Humidity sensor for measuring relative humidity ; Anemometer for measuring ambient wind speed ; Gasifier axial uniform arrangement a plurality of monitoring nodes, each monitoring node being provided with the following sensors: temperature sensor two for measuring the inner wall surface temperature of the gasifier monitoring node ; a temperature sensor three for measuring the temperature of the outer surface of the gasifier monitoring node ; An ultrasonic ice thickness sensor for measuring the thickness of an ice layer of a gasifier monitoring node .

Images