Infrared identification adaptive anti-freezing control system and method for sea water vaporizer

Abstract

The application discloses an infrared identification adaptive anti-freezing control system and method for a seawater vaporizer, and belongs to the technical field of anti-freezing control and temperature monitoring. The system comprises an infrared thermal imaging unit, which is used for collecting infrared thermal images of key parts of the seawater vaporizer; a temperature sensing unit, which is used for monitoring ambient temperature and seawater inlet temperature; a control unit, which is configured to perform low-temperature abnormal point analysis on an area prone to icing, judge a current icing risk level, and generate a corresponding anti-freezing control signal; an anti-freezing execution unit, which is used for receiving the anti-freezing control signal and executing anti-freezing measures on the seawater vaporizer; a data storage and communication unit, which is used for storing temperature data, anti-freezing control signals and a reference temperature field model, and realizing remote state monitoring and alarming; and an alarm unit, which is used for triggering sound-light alarming or sending an alarm signal to a remote monitoring center when the icing risk exceeds a safety threshold or the system itself fails.

Description

Technical Field

[0001] This application relates to the field of antifreeze control and temperature monitoring, and more specifically, to an infrared recognition adaptive antifreeze control system and method for seawater vaporizers. Background Technology

[0002] Seawater vaporizers are key heat exchange devices in ship LNG fuel supply systems. They are typically installed on the ship's deck or in open areas directly exposed to seawater and the surrounding atmosphere, utilizing seawater heat to complete the vaporization process of liquefied natural gas. During navigation in high-latitude seas, especially under harsh sea conditions such as low temperatures, cold waves, and strong winds in winter, both ambient and seawater temperatures drop significantly. Strong winds further enhance convective heat transfer on the equipment surface, causing the surface temperature of localized areas of the seawater vaporizer to rapidly drop below freezing. Due to the complex structure of the seawater vaporizer, low-temperature cold spots are more likely to form at locations with uneven heat transfer, such as pipe connections, fin roots, and areas where support structures intersect, thus preferentially leading to icing.

[0003] When ice forms on the surface of the seawater vaporizer, the resulting ice layer covers the heat exchange surface, reducing the heat exchange capacity between the seawater and the heat exchange fluid and significantly decreasing vaporization efficiency. Simultaneously, the ice layer and ice bridge structure reduce the cross-sectional area of ​​the flow channels, increasing flow resistance and causing system operating parameters to deviate from design conditions. With continued icing, the temperature difference between the inside and outside of the ice layer and the freezing expansion effect can also lead to pipe cracking, structural damage, and even heat exchange system failure, severely impacting the safety and stability of the ship's LNG fuel system.

[0004] Currently, antifreeze measures for seawater vaporizers mainly employ contact temperature sensors, electric heating devices, and fixed insulation structures. For example, a small number of temperature sensors are deployed in localized areas to control the heating device's start and stop based on single-point or small-scale temperature measurements; or electric heating and hot fluid circulation are initiated at preset time intervals; or fixed insulation covers and windbreaks are used to reduce direct cold air impact. However, existing technologies still have significant limitations: on the one hand, point-based temperature measurement methods cannot reflect the overall surface temperature field distribution of the seawater vaporizer, failing to effectively identify localized cold spots and potential icing areas, leading to icing risks often only being detected at a more severe stage; on the other hand, timed or fixed threshold-based control methods lack dynamic adaptability to changes in ambient temperature, seawater temperature, and operating conditions, easily resulting in insufficient or excessive antifreeze protection. Furthermore, existing systems lack coordinated and refined control strategies for actuators such as heating devices, regulating valves, and insulation mechanisms, easily causing frequent start-stop operations or prolonged high-load operation of the actuators, increasing system energy consumption and equipment wear, and leading to decreased control stability.

[0005] In summary, how to achieve non-contact, global, real-time monitoring of the surface temperature field of seawater vaporizers, how to accurately identify local icing risk areas and establish adaptive anti-freezing control strategies that match actual operating conditions, thereby ensuring the safe and stable operation of the system while reducing overall energy consumption and equipment wear, have become urgent technical problems to be solved. Summary of the Invention

[0006] To overcome a series of shortcomings in the existing technology, the purpose of this application is to provide an infrared recognition adaptive antifreeze control system for seawater vaporizers, comprising: The infrared thermal imaging unit is used to acquire infrared thermal images of key parts of the seawater vaporizer to achieve non-contact global monitoring of the surface temperature field of the seawater vaporizer. Temperature sensing unit is used to monitor ambient temperature and seawater inlet temperature; The control unit is configured to: perform low-temperature anomaly analysis on areas prone to freezing; determine the current freezing risk level; and generate corresponding antifreeze control signals. The antifreeze actuator is used to receive the antifreeze control signal and perform antifreeze measures on the seawater vaporizer; The data storage and communication unit is used to store temperature data, antifreeze control signals and reference temperature field models, and to realize remote status monitoring and alarm. The alarm unit is used to trigger an audible and visual alarm or send an alarm signal to a remote monitoring center when the risk of icing exceeds a safety threshold or when the system itself malfunctions.

[0007] The purpose of this application is also to provide an infrared identification adaptive antifreeze control method for seawater vaporizers, applied to the infrared identification adaptive antifreeze control system described above, comprising the following steps: Infrared thermal images of key parts of the seawater vaporizer were acquired and preprocessed; at the same time, the current ambient temperature, seawater inlet temperature, and seawater outlet temperature were also acquired. Real-time temperature field distribution data of the seawater vaporizer surface was identified based on preprocessed infrared thermal images; The real-time temperature field distribution data is compared and analyzed with the preset benchmark temperature field model under the no-icing-risk state. The temperature deviation index and temperature gradient anomaly index are calculated, and the current icing risk level is determined accordingly. When the icing risk level reaches the preset conditions or the temperature of the icing-prone area is lower than the critical icing temperature threshold, a corresponding antifreeze control strategy is generated. Antifreeze control signals are generated based on the antifreeze control strategy, and the antifreeze execution unit is controlled accordingly to perform antifreeze actions.

[0008] In some embodiments, the method for acquiring infrared thermal images is as follows: Configure the operating parameters of the infrared thermal imaging camera and install the infrared thermal imaging camera at the preset monitoring position of the seawater vaporizer; Adjust the lens parameters and monitoring field of view according to the structural characteristics of the seawater vaporizer so that the area prone to icing meets the requirements for temperature anomaly identification. An environmental protection and vibration reduction structure is set up for the infrared thermal imaging camera, and the image acquisition frequency is dynamically adjusted according to the operating status of the seawater vaporizer and the icing risk level. Infrared thermal images were obtained by continuously acquiring infrared thermal images of the seawater vaporizer using an infrared thermal imaging camera.

[0009] In some embodiments, the method for preprocessing infrared thermal images is as follows: Non-uniformity correction processing is performed on the infrared thermal image to correct the response differences between pixels of the infrared detector, and the corrected infrared thermal image is obtained. The corrected infrared thermal image is subjected to noise filtering to reduce image temperature noise and improve the image signal-to-noise ratio, resulting in a filtered infrared thermal image. Temperature calibration mapping is performed on the filtered infrared thermal image to convert the image grayscale values ​​into corresponding temperature field data, thus obtaining the initial temperature field data. Based on the emissivity parameters corresponding to the surface materials of different parts of the seawater vaporizer, the initial temperature field data is corrected for emissivity to improve the temperature measurement accuracy and obtain the corrected temperature field data. Based on the corrected temperature field data, a corresponding preprocessed infrared thermal image is generated.

[0010] In some embodiments, the method for identifying real-time temperature field distribution data on the surface of a seawater vaporizer based on preprocessed infrared thermal images is as follows: Based on the structural features and heat transfer characteristics of the seawater vaporizer, the preprocessed infrared thermal image is divided into regions of interest to obtain a regionalized temperature field image. Based on the regional temperature field image, statistical features are extracted from the temperature distribution information of each region of interest to obtain the temperature statistical feature data corresponding to each region of interest; Based on the temperature change relationship between adjacent pixels in each region of interest, the temperature spatial gradient information of the corresponding region is calculated to obtain the temperature gradient distribution data corresponding to each region of interest. Based on temperature statistical characteristic data and temperature gradient distribution data, local low temperature areas and abnormal temperature gradient areas are identified to obtain information on potential icing cold spots on the surface of the seawater vaporizer. By associating and storing timestamps, region numbers, temperature statistical characteristic data, temperature gradient distribution data, and potential icing cold spot information, real-time temperature field distribution data of the seawater vaporizer surface can be obtained.

[0011] In some embodiments, the method for constructing the baseline temperature field model under the preset no-icing-risk condition is as follows: Determine the safe operating conditions of the seawater vaporizer under a no-icing-risk state, and obtain the corresponding safe operating condition reference parameter information. The safe operating condition reference parameter information is used to characterize the operating condition boundary conditions under a no-icing-risk state. Under safe operating conditions, infrared thermal images and real-time operating parameters of the seawater vaporizer are continuously collected at a preset sampling period to construct a time-series dataset containing temperature field data and operating parameter data. Based on the seawater flow, ambient temperature and wind speed in the operating condition parameter data, the time series dataset is classified according to the operating condition. Data that meet the preset operating condition parameter deviation range are divided into the same data group to obtain the classified operating condition dataset. Statistical analysis was performed on the temperature field data corresponding to each data group, and the temperature statistical characteristics corresponding to each region of interest were extracted to obtain typical temperature field distribution data under each working condition. Based on typical temperature field distribution data, a mapping relationship model between temperature field and operating condition parameters is established to obtain a reference temperature prediction model. Typical temperature field distribution data are fused with a baseline temperature prediction model to obtain a baseline temperature field model under conditions without freezing risk.

[0012] In some embodiments, the temperature deviation index is calculated as follows: The real-time average temperature of each region of interest in the real-time temperature field is obtained, and the reference average temperature of the corresponding region in the reference temperature field model under the current operating conditions is obtained, so as to obtain the average temperature difference data of each region of interest. Based on the structural characteristics and historical icing risk levels of each region of interest, the average temperature difference data are assigned corresponding weighting coefficients and then comprehensively calculated to obtain the temperature fusion feature quantity. The temperature fusion feature is subjected to time-domain filtering to reduce the impact of short-term temperature fluctuations and measurement noise on the calculation results, thus obtaining the temperature deviation index.

[0013] In some embodiments, the temperature gradient anomaly index is calculated as follows: The spatial temperature gradient corresponding to each pixel in the real-time temperature field is calculated based on the finite difference method to obtain the temperature gradient field data. The temperature gradient magnitude of each pixel is calculated based on the temperature gradient field data, and the temperature gradient magnitude of each region of interest is statistically analyzed to obtain the temperature gradient statistical feature data corresponding to each region of interest. By comparing the statistical characteristic data of the temperature gradient corresponding to each region of interest with the benchmark temperature field model, the gradient deviation data corresponding to each region of interest is obtained. Based on the temperature gradient change relationship between adjacent sampling times, time series analysis is performed on the temperature gradient corresponding to each region of interest to obtain gradient change trend data corresponding to each region of interest. Based on a comprehensive evaluation of gradient deviation data and gradient change trend data, the temperature gradient anomaly index corresponding to each region of interest is obtained.

[0014] In some embodiments, the method for determining the critical freezing temperature threshold is as follows: The basic freezing temperature parameters are determined by using the freezing point of pure water under standard atmospheric pressure as the benchmark reference value. Real-time seawater salinity measurements were obtained, and the basic freezing temperature parameters were corrected for salinity based on the freezing point depression law to obtain the theoretical freezing temperature. By combining the seawater flow state and the vaporizer operating pressure, the theoretical freezing temperature is corrected for operating conditions to obtain the corrected freezing temperature. The actual critical freezing temperature threshold is obtained by adding a preset engineering safety margin to the freezing temperature after the working condition correction. The actual freezing critical temperature threshold is dynamically adjusted based on real-time seawater salinity measurements to obtain the freezing critical temperature threshold used for antifreeze control determination.

[0015] In some embodiments, the infrared recognition adaptive antifreeze control method further includes the following steps: During the antifreeze action, real-time infrared thermal images and real-time temperature data are continuously collected to determine whether the current antifreeze action has reached the preset antifreeze standard. If the target is not met, the antifreeze action is maintained or enhanced; if the target is met, the workload of the antifreeze execution unit is gradually reduced based on the preset hysteresis control range to suppress frequent start-stop operations.

[0016] Compared with the prior art, this application has the following beneficial effects: This application employs infrared thermal imaging for non-contact global temperature field monitoring of seawater vaporizers on ship decks. Compared to traditional point-based temperature measurement methods, it can more comprehensively and in real-time identify local low temperatures and potential icing areas, improving the accuracy and timeliness of icing monitoring. By introducing a benchmark temperature field model and an icing risk level discrimination mechanism, intelligent antifreeze identification and graded control based on the actual temperature field distribution are achieved, enhancing the accuracy and reliability of antifreeze control. Simultaneously, by combining the coordinated adaptive adjustment of electric heating devices, hot fluid regulating valves, and zoned movable insulation curtains, the antifreeze strategy can be dynamically optimized according to different icing risks, reducing the overall energy consumption of the system. Furthermore, through a temperature-risk joint hysteresis control mechanism, frequent start-stop of actuators is effectively suppressed, improving system operational stability and equipment lifespan. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the infrared recognition adaptive antifreeze control system for a seawater vaporizer disclosed in an embodiment of this application.

[0018] Figure 2 This is a schematic flowchart of the infrared recognition adaptive antifreeze control method for seawater vaporizers disclosed in an embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of the embodiments of this invention will be described in more detail below with reference to the accompanying drawings. In the drawings, the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The described embodiments are some embodiments of this invention, but not all embodiments.

[0020] Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] The embodiments and directional terms described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0022] like Figure 1 As shown, the infrared recognition adaptive antifreeze control system for a seawater vaporizer includes: The infrared thermal imaging unit is used to acquire infrared thermal images of key parts of the seawater vaporizer to achieve non-contact global monitoring of the surface temperature field of the seawater vaporizer. Temperature sensing unit is used to monitor ambient temperature and seawater inlet temperature; The control unit, connected to the infrared thermal imaging unit and the temperature sensing unit, is configured to: identify the temperature field distribution on the vaporizer surface based on the infrared thermal image and perform low-temperature anomaly analysis on the easily icing area; compare the real-time infrared thermal image with the pre-stored benchmark temperature field model under no-icing-risk conditions to determine the current icing risk level; and generate a corresponding antifreeze control signal when an icing risk is identified or the surface temperature of the easily icing area is detected to be lower than the preset antifreeze threshold. An antifreeze execution unit, connected to the control unit, is used to receive antifreeze control signals and perform antifreeze measures on the seawater vaporizer; The antifreeze measures include at least one of the following: heating key parts of the seawater vaporizer using an electric heating device; regulating the flow rate of the hot fluid flowing into the seawater vaporizer using a regulating valve, wherein the hot fluid is engine cooling water; and adjusting the air intake area around the seawater vaporizer using a movable heat preservation curtain. The data storage and communication unit is used to store temperature data, antifreeze control signals and reference temperature field models, and to realize remote status monitoring and alarm. The alarm unit is used to trigger an audible and visual alarm or send an alarm signal to a remote monitoring center when the risk of icing exceeds a safety threshold or when the system itself malfunctions.

[0023] The infrared recognition adaptive antifreeze control system for seawater vaporizers described in this application achieves non-contact global perception of the surface temperature field of the seawater vaporizer and real-time identification of areas prone to icing through coordinated monitoring of infrared thermal imaging and ambient temperature. By dynamically comparing real-time infrared thermal images with a reference temperature field model, it accurately determines the icing risk level and provides early warning of abnormal low-temperature areas. Based on this, the system incorporates adaptive linkage control of various antifreeze measures, such as electric heating, hot fluid flow regulation, and movable insulation curtain airflow control, effectively mitigating the icing trend on the surface of the seawater vaporizer in low-temperature environments. Simultaneously, through data storage, remote communication, and alarm mechanisms, it achieves continuous monitoring of the system's operating status and timely response to abnormal risks, thereby significantly improving the antifreeze capability, operational stability, and safety reliability of the seawater vaporizer in cold sea conditions, meeting the long-term continuous operation requirements of ship LNG gas systems.

[0024] In some embodiments, the infrared thermal imaging unit includes: An infrared thermal imaging camera, operating in the 8μm–14μm band, is used to acquire infrared thermal images of the surface of a seawater vaporizer. The protective housing is used to protect the infrared thermal imaging camera from water, salt spray and vibration. The front end of the protective housing is provided with an infrared transmission window to ensure the normal transmission of infrared thermal radiation signals. The adjustable bracket is used to adjust the installation angle and field of view of the infrared thermal imaging camera relative to the seawater vaporizer, so that the area prone to icing is within the monitoring field of view of the infrared thermal imaging camera. The automatic calibration module is used to perform periodic temperature calibration of the infrared thermal imaging camera by using a preset blackbody or temperature reference block to ensure the normal transmission of infrared thermal radiation signals.

[0025] Specifically, the infrared thermal imaging camera selected is an uncooled long-wave infrared thermal imaging camera with a resolution of 640×512 pixels, a thermal sensitivity (NETD) of ≤50mK, a temperature measurement range of -40℃ to 150℃, and a frame rate of 10fps.

[0026] Furthermore, the infrared thermal imaging camera is mounted on an adjustable bracket at a distance of about 3 meters from the seawater vaporizer. It uses a 25mm focal length lens with a field of view of about 60°×48° and a spatial resolution of about 7.5mm per pixel, which is sufficient to identify local temperature anomalies in pipe joints (50mm~100mm in diameter) and fin roots (8mm~15mm spacing).

[0027] Specifically, the protective housing has an IP67 rating for waterproofing and dustproofing and is made of 316L stainless steel. An infrared transmission window made of germanium (Ge) is installed at the front, with a transmittance of ≥95% in the 8μm–14μm wavelength range. The housing is internally equipped with rubber damping pads and a spring-damped vibration reduction structure.

[0028] Specifically, to ensure temperature measurement accuracy, a surface-source blackbody calibration source with an emissivity ≥0.97 and a temperature control accuracy of ±0.1℃ is installed near the vaporizer, and it is automatically calibrated every 4 hours. During the calibration process, a linear radiation calibration model is used. ,in, , These are the calibration coefficients obtained from blackbody calibration.

[0029] For example, the blackbody temperature is set to -10℃ and 30℃ respectively, and the camera reads the corresponding grayscale values ​​as follows: and Therefore, the calculation is obtained. , .

[0030] In actual measurements, the raw radiance acquired by the infrared thermal imaging unit and the temperature of the measured surface satisfy an approximate Stefan-Boltzmann relationship: in, The emissivity of the coating on the outer surface of the seawater vaporizer. Stefan-Boltzmann constant ( = 5.67×10 -8 W / (m 2 ·K 4 )).

[0031] For example, the emissivity of the anti-corrosion coating is ε=0.92; the emissivity of the exposed stainless steel surface is ε=0.15~0.30. During the installation and commissioning phase, the effective emissivity of various surfaces is determined by comparing with reference temperatures and stored in the emissivity parameter table of the data storage and communication unit.

[0032] In some embodiments, the infrared thermal imaging unit is further configured to: Non-uniformity correction and noise filtering are performed on the acquired infrared thermal images; Based on the pre-defined region of interest, the average temperature, minimum temperature, and temperature gradient parameters of different structural regions of the seawater vaporizer are calculated, and local low-temperature cold spots and potential icing areas are identified accordingly.

[0033] In some embodiments, the temperature sensing unit includes: Environmental sensors are used to collect information on ambient air temperature and / or wind speed at sea. Seawater inlet temperature sensor is used to collect the temperature of the seawater entering the seawater vaporizer; A vaporizer outlet temperature sensor is used to monitor the temperature of the medium at the outlet of the seawater vaporizer in order to construct operating condition discrimination parameters for the seawater vaporizer.

[0034] Specifically, the environmental sensors include a PT100 platinum resistance ambient temperature sensor (-50℃~100℃, accuracy ±0.3℃) and an ultrasonic anemometer (0~60m / s, resolution 0.1m / s); the seawater inlet temperature sensor adopts a titanium alloy sheathed K-type thermocouple seawater inlet temperature sensor (-10℃~50℃, accuracy ±0.5℃).

[0035] In some embodiments, the control unit is further configured to: Based on the identified icing risk level, ambient temperature, and historical operating data, the intensity and duration of antifreeze measures are adaptively adjusted. Specifically: when the icing risk level is high, the electric heating device is activated and the hot fluid flow rate is increased; when the icing risk level is medium, the hot fluid flow rate is adjusted first, in conjunction with intermittent electric heating control; when the icing risk level is low, the opening of the insulation curtain is finely adjusted to reduce the scouring area of ​​the seawater vaporizer by cold air.

[0036] In some embodiments, the control unit presets start and stop conditions corresponding to the antifreeze measures, and dynamically adjusts the hysteresis threshold between the start and stop conditions based on real-time acquired temperature field data, ambient temperature and icing risk level, so as to avoid frequent start and stop of the antifreeze execution unit near critical operating conditions.

[0037] In some embodiments, the electric heating device includes at least an externally attached electric heating strip and / or an inter-fin electric heating element; the electric heating device is arranged in a zoned manner, and the heating power of each heating zone can be adjusted independently to achieve zoned and refined heating control of areas with high icing risk.

[0038] In some embodiments, the movable heat-insulating curtain is made of a flexible composite material resistant to seawater corrosion, and multiple partitioned heat-insulating curtains are arranged around the seawater vaporizer; each heat-insulating curtain is independently driven to open and close by a corresponding electric actuator, so as to adaptively adjust the air intake area around the seawater vaporizer according to different wind directions and wind speed conditions.

[0039] like Figure 2 As shown, the purpose of this application is also to provide an infrared identification adaptive antifreeze control method for seawater vaporizers, comprising the following steps: Infrared thermal images of key parts of the seawater vaporizer were acquired and preprocessed; at the same time, the current ambient temperature, seawater inlet temperature, and seawater outlet temperature were also acquired. Real-time temperature field distribution data of the seawater vaporizer surface was identified based on preprocessed infrared thermal images; The real-time temperature field distribution data is compared and analyzed with the preset benchmark temperature field model under the no-icing-risk state. The temperature deviation index and temperature gradient anomaly index are calculated, and the current icing risk level is determined accordingly. When the icing risk level reaches the preset conditions or the temperature of the icing-prone area is lower than the critical icing temperature threshold, a corresponding antifreeze control strategy is generated. Anti-freeze control signals are generated based on the anti-freeze control strategy, and the anti-freeze action of the anti-freeze execution unit is controlled accordingly. During the antifreeze action, real-time infrared thermal images and real-time temperature data are continuously collected to determine whether the current antifreeze action has reached the preset antifreeze standard. If the target is not met, the antifreeze action is maintained or enhanced; if the target is met, the workload of the antifreeze execution unit is gradually reduced based on the preset hysteresis control range to suppress frequent start-stop operations.

[0040] The infrared identification adaptive antifreeze control method for seawater vaporizers described in this application achieves real-time, non-contact monitoring of the surface temperature field of the seawater vaporizer through infrared thermal image preprocessing and collaborative acquisition of multiple temperature parameters. By dynamically comparing the real-time temperature field distribution data with a reference temperature field model and combining it with temperature deviation index and temperature gradient anomaly index for comprehensive analysis, it achieves accurate identification of icing risk level and early warning of icing-prone areas. Based on this, it adaptively generates corresponding antifreeze control strategies by combining the icing risk level and the critical icing temperature threshold, and dynamically adjusts and controls the antifreeze execution unit. At the same time, by continuously monitoring the execution effect of antifreeze actions in a closed loop and gradually adjusting the execution intensity by combining the hysteresis control interval, it effectively reduces the frequent start-stop phenomenon of the antifreeze system, thereby significantly improving the antifreeze control accuracy, operational stability and energy utilization efficiency of the seawater vaporizer in the low-temperature marine environment, and meeting the requirements for long-term continuous and safe operation of the ship's LNG gas system.

[0041] In some embodiments, the method for acquiring infrared thermal images is as follows: Configure the operating parameters of the infrared thermal imaging camera and install the infrared thermal imaging camera at the preset monitoring position of the seawater vaporizer; Adjust the lens parameters and monitoring field of view according to the structural characteristics of the seawater vaporizer so that the area prone to icing meets the requirements for temperature anomaly identification. An environmental protection and vibration reduction structure is set up for the infrared thermal imaging camera, and the image acquisition frequency is dynamically adjusted according to the operating status of the seawater vaporizer and the icing risk level. Infrared thermal images were obtained by continuously acquiring infrared thermal images of the seawater vaporizer using an infrared thermal imaging camera.

[0042] The infrared thermal image acquisition method described in this application achieves accurate infrared monitoring of the icing-prone areas of the seawater vaporizer by adaptively configuring the operating parameters, monitoring position, and field of view of the infrared thermal imaging camera. By setting up environmental protection and vibration reduction structures and dynamically adjusting the image acquisition frequency according to the operating status and icing risk, the stability and continuity of infrared imaging in complex marine environments are improved. Based on this, reliable acquisition of the surface temperature field of the seawater vaporizer is achieved, thereby providing accurate data support for icing risk identification and antifreeze control.

[0043] In some embodiments, the method for preprocessing infrared thermal images is as follows: Non-uniformity correction processing is performed on the infrared thermal image to correct the response differences between pixels of the infrared detector, and the corrected infrared thermal image is obtained. The corrected infrared thermal image is subjected to noise filtering to reduce image temperature noise and improve the image signal-to-noise ratio, resulting in a filtered infrared thermal image. Temperature calibration mapping is performed on the filtered infrared thermal image to convert the image grayscale values ​​into corresponding temperature field data, thus obtaining the initial temperature field data. Based on the emissivity parameters corresponding to the surface materials of different parts of the seawater vaporizer, the initial temperature field data is corrected for emissivity to improve the temperature measurement accuracy and obtain the corrected temperature field data. Based on the corrected temperature field data, a corresponding preprocessed infrared thermal image is generated.

[0044] For example, the non-uniformity correction process employs a pixel-by-pixel gain-bias correction model: in, For the corrected first The radiance value of each pixel; For the first Gain correction coefficient for each pixel; For the first The offset correction coefficient for each pixel; For the first time collected The radiance value of each pixel.

[0045] Noise filtering uses a 5×5 two-dimensional Gaussian filter: in, These are the weight coefficients for a two-dimensional Gaussian filter, with a standard deviation of 1.0 pixel. These are the pixel values ​​of the filtered image; For a moment The original image values ​​of the next neighboring pixels; The target pixel coordinates; This represents the neighborhood offset within the filtering window.

[0046] The infrared thermal image preprocessing method described in this application eliminates differences in the response of infrared detector pixels through non-uniformity correction, thereby improving the consistency of infrared images; reduces temperature noise interference and improves the image signal-to-noise ratio through noise filtering; achieves accurate conversion of grayscale values ​​to temperature field data through temperature calibration mapping, and improves temperature measurement accuracy by combining emissivity parameters corresponding to different surface materials; on this basis, high-precision preprocessing of the temperature field on the surface of seawater vaporizer is achieved, providing a reliable data foundation for icing risk identification and antifreeze control.

[0047] In some embodiments, the method for identifying real-time temperature field distribution data on the surface of a seawater vaporizer based on preprocessed infrared thermal images is as follows: Based on the structural features and heat transfer characteristics of the seawater vaporizer, the preprocessed infrared thermal image is divided into regions of interest to obtain a regionalized temperature field image. Based on the regional temperature field image, statistical features are extracted from the temperature distribution information of each region of interest to obtain the temperature statistical feature data corresponding to each region of interest; Based on the temperature change relationship between adjacent pixels in each region of interest, the temperature spatial gradient information of the corresponding region is calculated to obtain the temperature gradient distribution data corresponding to each region of interest. Based on temperature statistical characteristic data and temperature gradient distribution data, local low temperature areas and abnormal temperature gradient areas are identified to obtain information on potential icing cold spots on the surface of the seawater vaporizer. By associating and storing timestamps, region numbers, temperature statistical characteristic data, temperature gradient distribution data, and potential icing cold spot information, real-time temperature field distribution data of the seawater vaporizer surface can be obtained.

[0048] For example, a vaporizer with 120 heat exchange tubes, 24 pipe joints, and 4 sets of finned modules is divided into K=16 zones. The average temperature and minimum temperature of the k-th zone are defined as follows: The temperature gradient is approximated by finite difference: In the formula, The region of interest is numbered; For the first Average temperature of the region of interest; For the first The lowest temperature in each region; coordinates ( The temperature value corresponding to the pixel at position (). For the first The number of valid pixels contained in each region; ) indicates that it belongs to the first A set of pixel coordinates for each region; For the temperature field along Temperature gradient in the direction; For the temperature field along Temperature gradient in the direction; , These are the temperature values ​​of the left and right adjacent pixels of the target pixel, respectively. , These are the temperature values ​​of the pixels directly above and below the target pixel, respectively. For the image in Spatial sampling interval in the direction; For the image in Spatial sampling interval in the direction.

[0049] The real-time temperature field distribution data identification method described in this application achieves targeted temperature analysis of key heat exchange areas in a seawater vaporizer by dividing the preprocessed infrared thermal image into regions of interest. It also achieves accurate identification of local low-temperature regions and regions with abnormal temperature gradients by extracting the temperature statistical characteristics and spatial gradient information of each region. Furthermore, it combines timestamps, region numbers, and potential icing spot information for associated storage, enabling dynamic tracking and continuous analysis of the surface temperature field of the seawater vaporizer. Based on this, it effectively improves the accuracy and real-time performance of identifying potential icing areas, providing reliable data support for subsequent icing risk assessment and adaptive antifreeze control.

[0050] In some embodiments, the method for constructing the baseline temperature field model under the preset no-icing-risk condition is as follows: Determine the safe operating conditions of the seawater vaporizer under a no-icing-risk state, and obtain the corresponding safe operating condition reference parameter information. The safe operating condition reference parameter information is used to characterize the operating condition boundary conditions under a no-icing-risk state. Under safe operating conditions, infrared thermal images and real-time operating parameters of the seawater vaporizer are continuously collected at a preset sampling period to construct a time-series dataset containing temperature field data and operating parameter data. Based on the seawater flow, ambient temperature and wind speed in the operating condition parameter data, the time series dataset is classified according to the operating condition. Data that meet the preset operating condition parameter deviation range are divided into the same data group to obtain the classified operating condition dataset. Statistical analysis was performed on the temperature field data corresponding to each data group, and the temperature statistical characteristics corresponding to each region of interest were extracted to obtain typical temperature field distribution data under each working condition. Based on typical temperature field distribution data, a mapping relationship model between temperature field and operating condition parameters is established to obtain a reference temperature prediction model. Typical temperature field distribution data are fused with a baseline temperature prediction model to obtain a baseline temperature field model under conditions without freezing risk.

[0051] The method for constructing the aforementioned benchmark temperature field model in this application determines the safe operating conditions of the seawater vaporizer under no-icing risk and continuously acquires infrared thermal images and operating condition parameter data to achieve stable acquisition of the temperature field characteristics under safe operating conditions. It improves the relevance and accuracy of temperature field analysis under different operating conditions by classifying the time-series dataset according to parameters such as seawater flow rate, ambient temperature, and wind speed. Furthermore, it establishes a mapping relationship between the temperature field and operating condition parameters based on typical temperature field distribution data to achieve predictive modeling of the benchmark temperature field. On this basis, through the fusion processing of typical temperature field data and the predictive model, a benchmark temperature field model under no-icing risk conditions is constructed, thus providing a reliable reference for subsequent icing risk identification and adaptive anti-freezing control.

[0052] In some embodiments, the temperature deviation index is calculated as follows: The real-time average temperature of each region of interest in the real-time temperature field is obtained, and the reference average temperature of the corresponding region in the reference temperature field model under the current operating conditions is obtained, so as to obtain the average temperature difference data of each region of interest. Based on the structural characteristics and historical icing risk levels of each region of interest, the average temperature difference data are assigned corresponding weighting coefficients and then comprehensively calculated to obtain the temperature fusion feature quantity. The temperature fusion feature is subjected to time-domain filtering to reduce the impact of short-term temperature fluctuations and measurement noise on the calculation results, thus obtaining the temperature deviation index.

[0053] The method for calculating the temperature deviation index described in this application achieves a quantitative characterization of the degree of temperature anomaly in seawater vaporizers by comparing and analyzing the average temperature of the corresponding region between the real-time temperature field and the benchmark temperature field model; it improves the accuracy of temperature anomaly identification in key icing-prone areas by weighted fusion of the structural characteristics and historical icing risk levels of each region of interest; and it reduces the impact of short-term temperature fluctuations and measurement noise through time-domain filtering, thereby achieving stable calculation of the temperature deviation index and providing a reliable basis for icing risk assessment and antifreeze control.

[0054] In some embodiments, the temperature gradient anomaly index is calculated as follows: The spatial temperature gradient corresponding to each pixel in the real-time temperature field is calculated based on the finite difference method to obtain the temperature gradient field data. The temperature gradient magnitude of each pixel is calculated based on the temperature gradient field data, and the temperature gradient magnitude of each region of interest is statistically analyzed to obtain the temperature gradient statistical feature data corresponding to each region of interest. By comparing the statistical characteristic data of the temperature gradient corresponding to each region of interest with the benchmark temperature field model, the gradient deviation data corresponding to each region of interest is obtained. Based on the temperature gradient change relationship between adjacent sampling times, time series analysis is performed on the temperature gradient corresponding to each region of interest to obtain gradient change trend data corresponding to each region of interest. Based on a comprehensive evaluation of gradient deviation data and gradient change trend data, the temperature gradient anomaly index corresponding to each region of interest is obtained.

[0055] The method for calculating the temperature gradient anomaly index described in this application calculates the spatial temperature gradient of the real-time temperature field using the finite difference method, thereby accurately extracting the local temperature change characteristics of the seawater vaporizer surface. It also performs statistical analysis of the temperature gradient magnitude and compares it with a benchmark temperature field model to achieve a quantitative assessment of the degree of temperature gradient anomaly. Furthermore, it incorporates time-series analysis of the temperature gradient change relationship between adjacent sampling times to improve the continuity and stability of temperature gradient anomaly identification. Based on this, it comprehensively calculates the temperature gradient anomaly index for each region of interest, thus providing a reliable basis for identifying potential icing areas and implementing antifreeze control.

[0056] In some embodiments, temperature gradient magnitude Represented as: .

[0057] In some embodiments, the reference temperature field model is obtained under conditions with no risk of icing (seawater temperature ≥10℃, ambient temperature ≥5℃, wind speed ≤10m / s).

[0058] In some embodiments, the icing risk level is classified based on a regional icing risk index, which comprehensively represents the degree of deviation from the regional minimum temperature, the proportion of cold spot area, and the duration of low temperature. The calculation formula is as follows: ,in, For the first Ice risk index for each region; , , These are the weighting coefficients for the temperature factor, cold spot area factor, and duration factor, respectively. The freezing point safety margin threshold is used to characterize the warning temperature benchmark; in this embodiment, it is set to 0.5℃. For the first The lowest surface temperature in each region; This is the temperature normalization coefficient, used to normalize the temperature difference term. In this embodiment, it is set to 5℃. For the first Cold spot area in each region; For the first The total area of ​​each region; For the first The duration of cold spots in each area; This is the time normalization coefficient, used to normalize the duration term. In this embodiment, it is set to 300 s.

[0059] Specifically, cold spot area The calculation method is as follows: ,in, For the first The area is below the freezing point safety margin threshold. The set of pixels; This represents the number of pixels in the cold spot. The actual area corresponding to a single pixel is approximately 56.25 mm in this embodiment. 2 .

[0060] Specifically, the first Average temperature difference in each region Defined as: ,in, For the first The total number of pixels in each region; For pixels in the reference temperature field Reference temperature; For the pixels in the current infrared thermal image The actual measured temperature.

[0061] In some embodiments, the method for determining the critical freezing temperature threshold is as follows: The basic freezing temperature parameters are determined by using the freezing point of pure water under standard atmospheric pressure as the benchmark reference value. Real-time seawater salinity measurements were obtained, and the basic freezing temperature parameters were corrected for salinity based on the freezing point depression law to obtain the theoretical freezing temperature. By combining the seawater flow state and the vaporizer operating pressure, the theoretical freezing temperature is corrected for operating conditions to obtain the corrected freezing temperature. The actual critical freezing temperature threshold is obtained by adding a preset engineering safety margin to the freezing temperature after the working condition correction. The actual freezing critical temperature threshold is dynamically adjusted based on real-time seawater salinity measurements to obtain the freezing critical temperature threshold used for antifreeze control determination.

[0062] The method for determining the critical freezing temperature threshold described in this application uses the freezing point temperature of pure water as a benchmark and corrects the freezing point drop pattern by incorporating seawater salinity, thereby achieving accurate calculation of the theoretical freezing temperature. It also improves the adaptability of freezing temperature determination under complex operating environments by incorporating seawater flow conditions and vaporizer operating pressure for condition correction. Furthermore, it introduces an engineering safety margin based on the condition-corrected freezing temperature and dynamically adjusts it according to real-time changes in seawater salinity, achieving adaptive determination of the critical freezing temperature threshold. This improves the accuracy and reliability of freezing risk identification and antifreeze control.

[0063] In some embodiments, the real-time monitoring mechanism during the antifreeze action includes: after the antifreeze control signal is issued and the antifreeze execution unit starts to operate, an enhanced monitoring mode is entered. The first 2 to 3 minutes after the antifreeze action is initiated are the critical period for temperature response. During this stage, the infrared thermal image acquisition frequency is maintained at 10 frames per second to closely track the temperature field change process. For the key area where the antifreeze action is being performed, in addition to the conventional division of 16 regions of interest, the temperature monitoring grid can be further refined within the key area to extract more temperature feature point data, so as to more accurately evaluate the local temperature distribution and heating effect. The actual working status of the antifreeze execution unit is monitored simultaneously. For the electric heating subsystem, the actual working current of each heating zone is monitored by a current sensor. The current value is used to determine whether the heating element is working normally and whether the actual output power meets the control command. At the same time, the surface temperature of the heating element is monitored by a temperature sensor on the heating element to prevent overheating damage. For the hot fluid regulation subsystem, the regulation is monitored by a valve position feedback signal. The actual opening position of the valve is monitored by a flow meter to determine whether the valve action is in place and whether the pipeline is unobstructed. For the insulation curtain regulation subsystem, the actual opening of each insulation curtain slat is monitored by a position sensor or limit switch, and the working current of the actuator is monitored to determine whether there is a jamming fault. A temperature response evaluation model is established. Based on the current antifreeze measures and the heating power input, the theoretical temperature rise rate is estimated based on the principle of heat transfer. The measured temperature rise rate is compared with the theoretical prediction value. If the measured rate is significantly lower than 70% of the prediction value, it is determined that there may be an abnormality, such as an actuator failure or an inappropriate control strategy, which needs to be adjusted in time. The complete process data from the start of the antifreeze action to the temperature of the target area recovering to above the safe threshold is recorded, including saving a temperature field snapshot every 30 seconds, recording the measurement values ​​of each sensor and the control quantity of the actuator every second, and calculating the total energy consumption of this antifreeze process. This data is stored in the data storage and communication unit for post-event analysis and control optimization.

[0064] In some embodiments, the decision logic for maintaining or enhancing the antifreeze action specifically involves: if, after the antifreeze action has been continuously executed for a preset time threshold, the target temperature field is still not detected to meet the preset antifreeze standard, then an enhanced control decision process is initiated. Specifically: First, the consistency of the current antifreeze control strategy is verified to determine whether the level of the antifreeze strategy is matched with the current freezing risk level. If the strategy level is matched but the temperature regulation effect does not meet expectations, the cause of the abnormal temperature response is further diagnosed and analyzed. The cause includes at least: the heat exchange load increases due to the deterioration of external environmental conditions, the heat exchange conditions deviate from the design range due to abnormal fluctuations in seawater flow or inlet water temperature, and the performance degradation or failure of the antifreeze execution unit. For different diagnostic results, corresponding enhanced compensation measures are implemented: when it is determined that the external environment has deteriorated, the current electric heating power is increased by 10% to 20%, or the opening of the insulation curtain is further reduced by 5% to 10% to reduce the intensity of convective heat transfer; when it is determined that the flow rate is insufficient, the opening status of the regulating valve is checked and the valve opening is increased; when it is determined that a local electric heating unit is faulty, the antifreeze load of that area is redistributed to the adjacent heating unit, and thermal compensation is achieved by increasing the output power of the adjacent area. If the current antifreeze strategy has reached the maximum control intensity under the corresponding risk level but still cannot effectively suppress the temperature drop, then the antifreeze strategy will be upgraded, the strategy level will be raised from medium risk to high risk, full antifreeze measures will be activated and the output intensity of each execution unit will be increased. During the enhanced antifreeze process, safety constraint monitoring is performed on each actuator: the surface temperature of the electric heating device is monitored in real time to ensure that it does not exceed the preset safety limit; the operating current and operating status of the electric actuator are monitored to avoid long-term stalling or overload operation. Simultaneously, antifreeze failure judgment conditions are set: when the antifreeze action continues for more than 30 minutes, and the system has activated all antifreeze measures and run at maximum intensity, but the temperature of the easily icing area has not yet recovered to the safe threshold and shows a continuous downward trend, it is determined that the antifreeze capacity is insufficient under the current operating conditions, triggering a high-level alarm, and sending early warning information to the operator and monitoring center through audible and visual alarms and remote communication, prompting emergency measures such as load reduction or safe shutdown to avoid equipment damage caused by continuous icing.

[0065] In some embodiments, the setting principle and mechanism of the preset hysteresis control interval are as follows: after the temperature field recovers to a safe state and meets the preset antifreeze standard, the antifreeze execution unit is not immediately shut down, but enters the hysteresis control stage to avoid frequent start-stop problems caused by temperature fluctuations near the threshold.

[0066] Hysteresis control achieves control hysteresis by setting start-up and stop-loss thresholds, where the temperature range between the start-up and stop-loss thresholds constitutes the hysteresis control range. Taking an electric heating device as an example, the start-up threshold temperature is set... The threshold temperature is 2℃. The temperature is 4℃. Electric heating is activated when the temperature of the easily icing area is below 2℃; electric heating is turned off only when the temperature rises above 4℃, thus forming a 2℃ hysteresis control range.

[0067] When the temperature is within the hysteresis control range, the anti-freeze actuator does not completely shut down, but continues to operate in a load-reduction maintenance mode: the output power of the electric heating device is reduced from approximately 60% of the rated power during the anti-freeze phase to 20%–30% of the rated power; the hot fluid flow rate is reduced from the maximum flow rate of 11 m³ / s. 3 / h is adjusted back to an operating level slightly above the baseline flow rate; the opening of the insulation curtain is gradually restored from the minimum opening of 15% to the intermediate opening of 40% to 50% in order to maintain basic antifreeze capability and reduce energy consumption; The duration of the hysteresis control is dynamically determined based on the duration of the preceding antifreeze process and the severity of the environment, and is generally not less than 15 minutes; when the external environment is still unstable or in a high-risk state, it can be extended to more than 30 minutes to enhance stability.

[0068] The purpose of this hysteresis control mechanism is to suppress control oscillations near the critical temperature, avoiding mechanical wear and electrical shocks caused by frequent start-stop of the actuator. Comparative operating data shows that without hysteresis control, the electric heating device starts and stops approximately 15 times within 30 minutes; after introducing hysteresis control, the number of start-stops is reduced to approximately 2 times, thereby significantly reducing equipment wear and improving system operational stability.

[0069] Furthermore, the hysteresis interval width can be adaptively optimized based on historical operating data. By analyzing the temperature control stability, antifreeze reliability, and energy consumption level under different combinations of hysteresis parameters, dynamic optimization and parameter optimization of the hysteresis control strategy can be achieved.

[0070] Furthermore, the output power of the electric heating device is calculated using proportional-integral (PI) control: in, To prevent freezing at the target temperature; For the first Each region at time Temperature deviation; For the first Each region at time The lowest temperature; This is the proportional control coefficient; These are integral control coefficients; For the time variable in the integration process; For the first Control output power of the electric heating device in each area.

[0071] For example, when When, deviation The proportional term outputs 720W; if it continues for 60 seconds, the integral term outputs approximately 180W, and the total output is approximately 900W, accounting for 60% of the rated power of 1500W.

[0072] Furthermore, the flow rate of the hot fluid is controlled by adjusting the opening of the regulating valve: ,in, For a moment The flow rate of the hot fluid; The maximum allowable hot fluid flow rate of the system; To regulate the normalized opening degree of the control valve.

[0073] For example, baseline traffic Maximum flow During medium-risk periods, traffic volume increases to 6.2. (Increases by approximately 20%), reaching 11.0 during high-risk periods. .

[0074] Furthermore, the opening degree of the insulation curtain is adjusted according to wind speed and risk level: in This is the reference opening degree for the thermal insulation curtain; This refers to the wind speed influence coefficient. This is a risk level adjustment coefficient; For a moment The opening degree of the heat insulation curtain; For a moment Environmental wind speed; This is the current indicator of the greatest risk of icing.

[0075] For example, wind speed 12 m / s, At that time, the opening of the insulation curtain was adjusted to 45.4%, and the opening was limited to between 15% and 80%.

[0076] Below is a complete operational example. During winter navigation in the North Atlantic at 60°N, an LNG vessel experiences low ambient temperatures and strong sea winds. Antifreeze control is dynamically implemented based on real-time monitoring results.

[0077] 02:00 , , Under normal monitoring conditions, the insulation curtain is open to 80%, and the heat transfer fluid is 5m. 3 / h, electric heating not started.

[0078] 02:05, Zone 5 Area 9 .because The risk index has been raised by 0.1. (Medium risk), enter anti-freeze mode: adjust the insulation curtain to 38.8%, increase the heat transfer fluid to 6.2m. 3 / h.

[0079] At 02:20, the wind speed increased to 18 m / s. The electric heating in zone 5 is activated, with an initial power of 660W. The insulation curtain is adjusted to 28.6%, and the heat transfer fluid level is increased to 7.78m. 3 / h.

[0080] At 02:35, the lowest temperature in zone 9 dropped to 0.2℃, and electric heating was activated. The system continued to operate under a medium-risk strategy.

[0081] At 03:00, the thermal compensation effect became apparent. But still lower Under hysteresis control, electric heating continues to be maintained.

[0082] At 04:30, the wind speed dropped to 8 m / s, and the temperature in each area rose back to above 4.0℃, but the exit conditions were not yet met.

[0083] 05:15, and After 300 seconds, the exit condition is met. The control unit smoothly shuts off the electric heating at a rate of 10% reduction every 30 seconds, restoring each actuator to its normal state.

[0084] Throughout the process, the electric heating was started and stopped only twice, and the data storage unit continuously recorded and uploaded key parameters every 10 seconds. Compared with traditional solutions, the energy consumption of electric heating was reduced by about 40%, the heat fluid consumption was reduced by about 25%, the icing detection rate was increased from 60% to over 95%, and the response time was shortened from 15 minutes to less than 30 seconds.

[0085] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An infrared recognition adaptive antifreeze control system for seawater vaporizers, characterized in that, include: The infrared thermal imaging unit is used to acquire infrared thermal images of key parts of the seawater vaporizer to achieve non-contact global monitoring of the surface temperature field of the seawater vaporizer. Temperature sensing unit is used to monitor ambient temperature and seawater inlet temperature; The control unit is configured to: perform low-temperature anomaly analysis on areas prone to freezing; and determine the current freezing risk level. Generate the corresponding antifreeze control signal; The antifreeze actuator is used to receive the antifreeze control signal and perform antifreeze measures on the seawater vaporizer; The data storage and communication unit is used to store temperature data, antifreeze control signals and reference temperature field models, and to realize remote status monitoring and alarm. The alarm unit is used to trigger an audible and visual alarm or send an alarm signal to a remote monitoring center when the risk of icing exceeds a safety threshold or when the system itself malfunctions.

2. An infrared recognition adaptive antifreeze control method for seawater vaporizers, applied to the infrared recognition adaptive antifreeze control system as described in claim 1, characterized in that, Includes the following steps: Infrared thermal images of key parts of the seawater vaporizer were acquired and preprocessed; at the same time, the current ambient temperature, seawater inlet temperature, and seawater outlet temperature were also acquired. Real-time temperature field distribution data of the seawater vaporizer surface was identified based on preprocessed infrared thermal images; The real-time temperature field distribution data is compared and analyzed with the preset benchmark temperature field model under the no-icing-risk state. The temperature deviation index and temperature gradient anomaly index are calculated, and the current icing risk level is determined accordingly. When the icing risk level reaches the preset conditions or the temperature of the icing-prone area is lower than the critical icing temperature threshold, a corresponding antifreeze control strategy is generated. Antifreeze control signals are generated based on the antifreeze control strategy, and the antifreeze execution unit is controlled accordingly to perform antifreeze actions.

3. The infrared recognition adaptive antifreeze control method according to claim 2, characterized in that, The method for acquiring infrared thermal images is as follows: Configure the operating parameters of the infrared thermal imaging camera and install the infrared thermal imaging camera at the preset monitoring position of the seawater vaporizer; Adjust the lens parameters and monitoring field of view according to the structural characteristics of the seawater vaporizer so that the area prone to icing meets the requirements for temperature anomaly identification. An environmental protection and vibration reduction structure is set up for the infrared thermal imaging camera, and the image acquisition frequency is dynamically adjusted according to the operating status of the seawater vaporizer and the icing risk level. Infrared thermal images were obtained by continuously acquiring infrared thermal images of the seawater vaporizer using an infrared thermal imaging camera.

4. The infrared recognition adaptive antifreeze control method according to claim 3, characterized in that, The method for preprocessing infrared thermal images is as follows: Non-uniformity correction processing is performed on the infrared thermal image to correct the response differences between pixels of the infrared detector, resulting in a corrected infrared thermal image. The corrected infrared thermal image is subjected to noise filtering to reduce image temperature noise and improve the image signal-to-noise ratio, resulting in a filtered infrared thermal image. Temperature calibration mapping is performed on the filtered infrared thermal image to convert the image grayscale values ​​into corresponding temperature field data, thus obtaining the initial temperature field data. Based on the emissivity parameters corresponding to the surface materials of different parts of the seawater vaporizer, the initial temperature field data is corrected for emissivity to improve the temperature measurement accuracy and obtain the corrected temperature field data. Based on the corrected temperature field data, a corresponding preprocessed infrared thermal image is generated.

5. The infrared recognition adaptive antifreeze control method according to claim 2, characterized in that, The method for identifying real-time temperature field distribution data on the surface of a seawater vaporizer based on preprocessed infrared thermal images is as follows: Based on the structural features and heat transfer characteristics of the seawater vaporizer, the preprocessed infrared thermal image is divided into regions of interest to obtain a regionalized temperature field image. Based on the regional temperature field image, statistical features are extracted from the temperature distribution information of each region of interest to obtain the temperature statistical feature data corresponding to each region of interest; Based on the temperature change relationship between adjacent pixels in each region of interest, the temperature spatial gradient information of the corresponding region is calculated to obtain the temperature gradient distribution data corresponding to each region of interest. Based on temperature statistical characteristic data and temperature gradient distribution data, local low temperature areas and abnormal temperature gradient areas are identified to obtain information on potential icing cold spots on the surface of the seawater vaporizer. By associating and storing timestamps, region numbers, temperature statistical characteristic data, temperature gradient distribution data, and potential icing cold spot information, real-time temperature field distribution data of the seawater vaporizer surface can be obtained.

6. The infrared recognition adaptive antifreeze control method according to claim 2, characterized in that, The method for constructing the baseline temperature field model under the pre-defined no-icing-risk condition is as follows: Determine the safe operating conditions of the seawater vaporizer under a no-icing-risk state, and obtain the corresponding safe operating condition reference parameter information. The safe operating condition reference parameter information is used to characterize the operating condition boundary conditions under a no-icing-risk state. Under safe operating conditions, infrared thermal images and real-time operating parameters of the seawater vaporizer are continuously collected at a preset sampling period to construct a time-series dataset containing temperature field data and operating parameter data. Based on the seawater flow, ambient temperature and wind speed in the operating condition parameter data, the time series dataset is classified according to the operating condition. Data that meet the preset operating condition parameter deviation range are divided into the same data group to obtain the classified operating condition dataset. Statistical analysis was performed on the temperature field data corresponding to each data group, and the temperature statistical characteristics corresponding to each region of interest were extracted to obtain typical temperature field distribution data under each working condition. Based on typical temperature field distribution data, a mapping relationship model between temperature field and operating condition parameters is established to obtain a reference temperature prediction model. Typical temperature field distribution data are fused with a baseline temperature prediction model to obtain a baseline temperature field model under conditions without freezing risk.

7. The infrared recognition adaptive antifreeze control method according to claim 2, characterized in that, The method for calculating the temperature deviation index is as follows: The real-time average temperature of each region of interest in the real-time temperature field is obtained, and the reference average temperature of the corresponding region in the reference temperature field model under the current operating conditions is obtained, so as to obtain the average temperature difference data of each region of interest. Based on the structural characteristics and historical icing risk levels of each region of interest, the average temperature difference data are assigned corresponding weighting coefficients and then comprehensively calculated to obtain the temperature fusion feature quantity. The temperature fusion feature is subjected to time-domain filtering to reduce the impact of short-term temperature fluctuations and measurement noise on the calculation results, thus obtaining the temperature deviation index.

8. The infrared recognition adaptive antifreeze control method according to claim 7, characterized in that, The method for calculating the temperature gradient anomaly index is as follows: The spatial temperature gradient corresponding to each pixel in the real-time temperature field is calculated based on the finite difference method to obtain the temperature gradient field data. The temperature gradient magnitude of each pixel is calculated based on the temperature gradient field data, and the temperature gradient magnitude of each region of interest is statistically analyzed to obtain the temperature gradient statistical feature data corresponding to each region of interest. By comparing the statistical characteristic data of the temperature gradient corresponding to each region of interest with the benchmark temperature field model, the gradient deviation data corresponding to each region of interest is obtained. Based on the temperature gradient change relationship between adjacent sampling times, time series analysis is performed on the temperature gradient corresponding to each region of interest to obtain gradient change trend data corresponding to each region of interest. Based on a comprehensive evaluation of gradient deviation data and gradient change trend data, the temperature gradient anomaly index corresponding to each region of interest is obtained.

9. The infrared recognition adaptive antifreeze control method according to claim 8, characterized in that, The method for determining the critical freezing temperature threshold is as follows: The basic freezing temperature parameters are determined by using the freezing point of pure water under standard atmospheric pressure as the benchmark reference value. Real-time seawater salinity measurements were obtained, and the basic freezing temperature parameters were corrected for salinity based on the freezing point depression law to obtain the theoretical freezing temperature. By combining the seawater flow state and the vaporizer operating pressure, the theoretical freezing temperature is corrected for operating conditions to obtain the corrected freezing temperature. The actual critical freezing temperature threshold is obtained by adding a preset engineering safety margin to the freezing temperature after the working condition correction. The actual freezing critical temperature threshold is dynamically adjusted based on real-time seawater salinity measurements to obtain the freezing critical temperature threshold used for antifreeze control determination.

10. The infrared recognition adaptive antifreeze control method according to any one of claims 2-9, characterized in that, The infrared recognition adaptive antifreeze control method further includes the following steps: During the antifreeze action, real-time infrared thermal images and real-time temperature data are continuously collected to determine whether the current antifreeze action has reached the preset antifreeze standard. If the condition is not met, maintain or enhance anti-freezing measures. If the target has been reached, the workload of the antifreeze execution unit will be gradually reduced based on the preset hysteresis control range to suppress frequent start-stop operations.

Images