Fishing boat refrigeration cabin refrigerant leakage detection device and method

Abstract

The invention provides a device and a method for detecting refrigerant leakage in a refrigeration cabin of a fishing boat. The device provided by the invention comprises a detection module, a transmission module and a processing module which are connected in sequence, the detection module is used for determining the arrangement positions of the sensors based on the arrangement of the fishing boat refrigeration cabin and converting the refrigerant concentrations, detected by the sensors, of the multiple positions in the fishing boat refrigeration cabin into electric signals. The transmission module is used for transmitting the electric signal detected by the detection module to the processing module; and the processing module is used for fusing the refrigerant concentrations of the multiple positions based on the neural network model to obtain a fused comprehensive refrigerant concentration, and determining a corresponding leakage mode and a leakage position based on the comprehensive refrigerant concentration. The device and the method for detecting the leakage of the refrigerating fluid in the refrigerating cabin of the fishing boat are used for accurately and comprehensively monitoring and detecting the leakage condition of the refrigerating fluid in the refrigerating cabin of the fishing boat in real time, finding out the leakage of the refrigerating fluid as early as possible, and preventing the economic, personal and environmental hazards caused by the leakage of the refrigerating fluid.

Description

Technical Field

[0001] The present application relates to the technical field of ship refrigeration equipment, and in particular to a refrigerant leakage detection device and method for a refrigerated compartment of a fishing vessel. Background Art

[0002] At present, refrigerant leak detection mostly relies on manual inspection or fixed sensors. Although manual inspection can detect leaks to a certain extent, it is often limited by human subjectivity and the frequency of inspections, which may lead to detection delays and failure to detect problems in time. Although fixed sensors can provide continuous monitoring, they are often unable to cover every corner of the ship's refrigerated compartments due to the limitations of their installation locations, which makes it difficult to detect refrigerant leaks early.

[0003] Therefore, there is an urgent need for a device to monitor and detect the leakage of refrigerant in the refrigerated compartment of a fishing vessel in real time, accurately and comprehensively, to detect the refrigerant leakage as early as possible, and to prevent the economic, personal and environmental damage caused by the refrigerant leakage. Summary of the invention

[0004] In view of this, the present application provides a refrigerant leakage detection device and method for a refrigerated compartment of a fishing vessel, which is used to monitor and detect the leakage of refrigerant in the refrigerated compartment of a fishing vessel in real time, accurately and comprehensively, discover the refrigerant leakage as early as possible, and prevent the economic, personal and environmental damage caused by the refrigerant leakage.

[0005] Specifically, the present application is implemented through the following technical solutions:

[0006] The first aspect of the present application provides a refrigerant leakage detection device for a refrigerated compartment of a fishing vessel, the device comprising a detection module, a transmission module, and a processing module connected in sequence; wherein: The detection module is used to determine the locations of multiple sensors based on the layout of the refrigerated compartment of the fishing boat, detect the refrigerant concentration at multiple locations in the refrigerated compartment of the fishing boat based on the sensors, and convert the refrigerant concentration into an electrical signal; wherein the multiple sensors are densely distributed in the refrigerated compartment of the fishing boat, and the refrigerant concentration at the corresponding location detected by each sensor is different;

[0007] The transmission module is used to transmit the electrical signal detected by the detection module to the processing module;

[0008] The processing module is used to fuse the refrigerant concentrations at multiple locations based on the neural network model to obtain a fused comprehensive refrigerant concentration, and determine the corresponding leakage mode and leakage location based on the comprehensive refrigerant concentration;

[0009] Among them, a refrigerant leakage distribution map of the refrigerated compartment of the fishing vessel is constructed based on the electrical signal values ​​collected by each sensor, the refrigerant leakage distribution map is input into the neural network model, a refrigerant leakage estimation map is predicted based on the refrigerant leakage constraints between each sensor, the refrigerant leakage distribution map is corrected based on the refrigerant leakage estimation map, and the comprehensive refrigerant concentration is determined based on the corrected refrigerant leakage distribution map.

[0010] The second aspect of the present application provides a method for detecting refrigerant leakage in a refrigerated compartment of a fishing vessel, the method being applied to any one of the refrigerant leakage detection devices for refrigerated compartments of a fishing vessel provided in the first aspect of the present application, the method comprising:

[0011] Determine the locations of multiple sensors based on the layout of the refrigerated compartment of the fishing boat, and detect the refrigerant concentration at multiple locations in the refrigerated compartment of the fishing boat based on the sensors; wherein the multiple sensors are densely distributed in the refrigerated compartment of the fishing boat, and the refrigerant concentration at the corresponding location detected by each sensor is different;

[0012] Based on the neural network model, refrigerant concentrations at multiple locations are fused to obtain a fused comprehensive refrigerant concentration, and a corresponding leakage mode and leakage location are determined based on the comprehensive refrigerant concentration;

[0013] Among them, a refrigerant leakage distribution map of the refrigerated compartment of the fishing vessel is constructed based on the electrical signal values ​​collected by each sensor, the refrigerant leakage distribution map is input into the neural network model, a refrigerant leakage estimation map is predicted based on the refrigerant leakage constraints between each sensor, the refrigerant leakage distribution map is corrected based on the refrigerant leakage estimation map, and the comprehensive refrigerant concentration is determined based on the corrected refrigerant leakage distribution map.

[0014] The present application provides a refrigerant leakage detection device and method for the refrigerated compartment of a fishing vessel. On the one hand, the layout positions of multiple sensors in the detection module are determined based on the actual layout of the refrigerated compartment of the fishing vessel, ensuring that the refrigerant concentration at each position in the refrigerated compartment of the fishing vessel can be monitored in real time, which avoids the monitoring blind spot problem that may exist in fixed sensors. The traditional sensor layout is often limited by the installation position and may not cover every corner of the refrigerated compartment of the fishing vessel, resulting in the inability to monitor the leakage in real time in some areas. However, through reasonable layout and spacing optimization, the density distribution of multiple sensors ensures that all areas in the cabin can be effectively covered by the sensor, enhancing the comprehensiveness and accuracy of refrigerant concentration monitoring. On the other hand, by combining the data collected by the sensor, the prediction ability of the neural network model and the correction mechanism, the accuracy and reliability of refrigerant leakage detection and concentration estimation can be effectively improved. First, the electrical signal value collected by the sensor directly reflects the local gas concentration, but due to the error of the sensor itself, environmental interference or uneven layout, the initially constructed refrigerant leakage distribution map may have blind spots or errors. By inputting these data into the neural network model, the neural network model's ability to model spatial relationships can be used to automatically learn the propagation law of gas leakage in space and the correlation between different sensor data, thereby generating a more accurate refrigerant leakage prediction map. The neural network model can identify and remove outliers caused by sensor errors, noise or other environmental factors, thereby improving the robustness of the prediction. Furthermore, by comparing and correcting the initially generated refrigerant leakage distribution map, the accuracy of the refrigerant leakage distribution map can be improved overall, making the gas concentration distribution more consistent with the actual leakage scenario. This method of generating a comprehensive refrigerant concentration based on the corrected refrigerant leakage distribution map can provide a global and reliable comprehensive refrigerant concentration assessment by combining the refrigerant concentrations of each point through weighted average or other mathematical methods, accurately assessing the leakage risk of the refrigerated compartment of the fishing vessel and taking timely measures. The setting of this system not only improves the measurement accuracy, but also can cope with a variety of interference factors in complex environments, thereby ensuring the efficiency and accuracy of real-time monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic diagram of the structure of a refrigerant leakage detection device for a refrigerated compartment of a fishing vessel provided in Example 1 of the present application;

[0016] Figure 2 A schematic diagram of the structure of a refrigerant leakage detection device for a refrigerated compartment of a fishing vessel provided in this application;

[0017] Figure 3 This is a flow chart of a method for detecting refrigerant leakage in a refrigerated compartment of a fishing vessel provided in Example 2 of the present application. DETAILED DESCRIPTION

[0018] Here, exemplary embodiments will be described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application.

[0019] The terms used in this application are only for the purpose of describing specific embodiments and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in this article refers to and includes any or all possible combinations of one or more associated listed items.

[0020] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0021] Specific embodiments are given below to introduce the technical solution of the present application in detail.

[0022] Figure 1 This is a schematic diagram of the structure of the refrigerant leakage detection device for the refrigerated compartment of a fishing vessel provided in Example 1 of the present application. Figure 1 , the device provided in this embodiment includes a detection module, a transmission module, and a processing module connected in sequence; wherein, The detection module is used to determine the locations of multiple sensors based on the layout of the refrigerated compartment of the fishing boat, detect the refrigerant concentration at multiple locations in the refrigerated compartment of the fishing boat based on the sensors, and convert the refrigerant concentration into an electrical signal; wherein the multiple sensors are densely distributed in the refrigerated compartment of the fishing boat, and the refrigerant concentration at the corresponding location detected by each sensor is different;

[0023] The transmission module is used to transmit the electrical signal detected by the detection module to the processing module;

[0024] The processing module is used to fuse the refrigerant concentrations at multiple locations based on the neural network model to obtain a fused comprehensive refrigerant concentration, and determine the corresponding leakage mode and leakage location based on the comprehensive refrigerant concentration;

[0025] Among them, a refrigerant leakage distribution map of the refrigerated compartment of the fishing vessel is constructed based on the electrical signal values ​​collected by each sensor, the refrigerant leakage distribution map is input into the neural network model, a refrigerant leakage estimation map is predicted based on the refrigerant leakage constraints between each sensor, the refrigerant leakage distribution map is corrected based on the refrigerant leakage estimation map, and the comprehensive refrigerant concentration is determined based on the corrected refrigerant leakage distribution map.

[0026] For details, please refer to Figure 1 The refrigerant leakage detection device for the refrigerated compartment of a fishing vessel comprises a detection module, a transmission module and a processing module. The detection module is connected to the transmission module, and the transmission module is connected to the processing module.

[0027] Furthermore, the detection module is used to determine the locations of multiple sensors based on the layout of the refrigerated compartment of the fishing vessel, detect the refrigerant concentration at multiple locations in the refrigerated compartment of the fishing vessel based on the multiple sensors, and convert the refrigerant concentration into an electrical signal. It should be noted that each sensor is arranged at a different location in the refrigerated compartment of the fishing vessel, and each sensor detects a different refrigerant concentration.

[0028] Optionally, the detection module includes multiple gas sensors, and the gas sensors are corrosion-resistant; the detection module is used to determine the layout positions of the multiple sensors based on the layout of the refrigerated compartment of the fishing boat, including: calculating the coverage area of ​​a single gas sensor based on the detection sensitivity and effective detection radius of the gas sensor; determining the optimal layout spacing between adjacent gas sensors based on the coverage area of ​​the gas sensor; wherein each gas sensor is interconnected, and the spacing between adjacent gas sensors is the same; based on the coverage area and the optimal layout spacing, the gas sensors are densely distributed in the refrigerated compartment of the fishing boat; based on the cabin shape, air flow path and cargo stacking area of ​​the refrigerated compartment of the fishing boat, the layout position of the gas sensor is adjusted.

[0029] Specifically, the sensor in the detection module is a gas sensor. Due to the particularity of the fishing environment where fishing boats are located, they often face high humidity, salt spray and other factors that may damage fishing boat equipment. Therefore, the sensor selected for the detection module must have corrosion resistance. The gas sensor is encapsulated by high-performance corrosion-resistant materials (such as stainless steel, polytetrafluoroethylene (PTFE) and other materials). These high-performance corrosion-resistant materials can effectively resist the erosion of harsh environments and ensure the long-term stability and accuracy of the gas sensor.

[0030] Furthermore, when determining the specific layout position of the gas sensor, it is necessary to comprehensively consider the characteristics of the gas sensor and the layout characteristics of the refrigerated compartment of the fishing vessel. In specific implementation, according to the technical parameters of the gas sensor, its detection sensitivity and effective detection radius are obtained, and the coverage area of ​​each gas sensor is calculated based on the obtained detection sensitivity and effective detection radius. Further, based on the coverage area of ​​a single gas sensor, the appropriate overlapping area size of the gas sensor detection range is calculated to ensure that there is no detection blind area. According to the size of the refrigerated compartment of the fishing vessel and the detection coverage characteristics of the gas sensor, the spacing formula is used to calculate the optimal layout spacing between the gas sensors. According to the calculated optimal layout spacing, combined with the size of the refrigerated compartment of the fishing vessel, the preliminary distribution points of the gas sensors are determined. The gas sensors are evenly distributed in various areas of the refrigerated compartment of the fishing vessel to ensure the overall symmetry of the distribution and the integrity of the detection range. Finally, the layout position of the gas sensor is adjusted based on the characteristics of the refrigerated compartment of the fishing vessel. The cabin shape of the fishing vessel's refrigerated compartment is divided into detailed areas, and gas sensors are installed in corners or special-shaped areas to ensure that the detection range of the gas sensors covers all areas; the air flow path in the refrigerated compartment of the fishing vessel is simulated or measured, the area with weak airflow is identified, and the arrangement of gas sensors in this area is increased or optimized; according to the position and height of the cargo stacking in the refrigerated compartment of the fishing vessel, the installation height or angle of the gas sensor is adjusted to cover the area above and around the stacked cargo to avoid the gas sensor being blocked by the cargo.

[0031] It should be noted that each gas sensor is distributed in each corner of the refrigerated compartment of the fishing boat at the same spacing density. Each gas sensor is interconnected to form a complete monitoring network. The detection range of the monitoring network covers the entire refrigerated compartment of the fishing boat to ensure that the refrigerant concentration in the entire compartment can be effectively monitored. In addition, by setting a detection range with a radius of 1 to 2 meters around the gas sensor, it can be ensured that the monitoring network has no blind spots.

[0032] Furthermore, after the gas sensor detects the refrigerant concentration at multiple locations in the refrigerated compartment of the fishing vessel, it converts it into an electrical signal or other form of measurable signal to facilitate subsequent data transmission and data processing. These signals represent the refrigerant concentration at various locations in the refrigerated compartment of the fishing vessel.

[0033] Specifically, the transmission module is used to transmit the electrical signal detected by the detection module to the processing module.

[0034] In specific implementation, in order to ensure that the electrical signals detected by the detection module can be transmitted to the processing module in a timely and accurate manner, the detection module is equipped with an efficient transmission module. The transmission module adopts a low-power design, such as low-power wireless technologies such as LoRa, Zigbee or NB-IoT. These low-power wireless technologies have achieved a good balance between distance and power consumption, which is suitable for the application of gas sensors. They can transmit data wirelessly over a long distance, and the battery life can be extended to months or even years. Furthermore, the gas sensor regularly sends an "active" signal to the processing module through the transmission module based on the heartbeat mechanism. At the same time, a backup mechanism can be configured. When the main communication channel fails, the system can automatically switch to the backup channel to ensure the reliability of data transmission. When the electrical signals obtained by the detection module through the transmission module form data packets, they will be sent to the processing module, which is responsible for centralized processing, analysis and storage of the received data. Through such a data transmission mechanism, the entire device can realize real-time monitoring of environmental changes, and also provide important support for subsequent data analysis and decision-making.

[0035] Specifically, the processing module is used to fuse the refrigerant concentrations of multiple locations detected by the detection module based on the neural network model to obtain a fused comprehensive refrigerant concentration, and determine the corresponding leakage mode and leakage location based on the comprehensive refrigerant concentration.

[0036] Among them, a refrigerant leakage distribution map of the refrigerated compartment of the fishing vessel is constructed based on the electrical signal values ​​collected by each sensor, the refrigerant leakage distribution map is input into the neural network model, a refrigerant leakage estimation map is predicted based on the refrigerant leakage constraints between each sensor, the refrigerant leakage distribution map is corrected based on the refrigerant leakage estimation map, and the comprehensive refrigerant concentration is determined based on the corrected refrigerant leakage distribution map.

[0037] Specifically, the refrigerant leakage distribution map represents the relationship between the electrical signal values ​​collected by each sensor, that is, the preliminary refrigerant leakage concentration distribution generated based on the sensor raw data. The refrigerant leakage prediction map represents the refrigerant leakage concentration distribution predicted based on the neural network model. Neural network models include DeeplabV3+, variational mode decomposition, convolutional neural network, recursive neural network, etc. The input of the neural network model is the refrigerant leakage distribution map, and the output of the neural network model is the predicted refrigerant leakage prediction map. The neural network model determines the refrigerant leakage constraints based on the analysis of the refrigerant leakage distribution map, and predicts the refrigerant leakage prediction map based on the refrigerant leakage constraints. The refrigerant leakage constraints represent the correlation and propagation law between different electrical signal values.

[0038] In the specific implementation, the electrical signal values ​​collected by each sensor in the refrigerated compartment of the fishing boat are collected. The spatial position of the sensor is calibrated, and the point data measured by the sensor is mapped to the corresponding spatial position in the entire refrigerated compartment using an interpolation algorithm or a distribution modeling method. The electrical signal value of each sensor is related to its position in space, and the electrical signal value of the unmeasured area is filled to generate a preliminary refrigerant leakage distribution map. The generated refrigerant leakage distribution map is used as input data, combined with the spatial constraint relationship between each sensor (such as the propagation law of leakage concentration, the association of adjacent points, etc.), and input into the pre-trained neural network model. The neural network model analyzes the input refrigerant leakage distribution map, based on the constraint rules and the leakage law learned during the training process, and outputs a refrigerant leakage prediction map. Furthermore, the refrigerant leakage prediction map is compared with the initially generated refrigerant leakage distribution map, and the deviation between the two is calculated. A fusion algorithm (such as weighted average or error feedback) is used to correct the abnormal values ​​or blind area data in the refrigerant leakage distribution map according to the deviation information to generate a corrected refrigerant leakage distribution map. Based on the corrected refrigerant leakage distribution map, the comprehensive refrigerant concentration is calculated according to the weighted rule of the refrigerant concentration at each point and its distance from the center position: ; Among them, the is the comprehensive refrigerant concentration; is the distance weighting coefficient; is the first Refrigerant concentration at point.

[0039] The neural network model includes an input layer, a hidden layer, and an output layer. The number of neurons in the input layer matches the data size of the refrigerant leakage distribution map. For example, if the input refrigerant leakage distribution map is a 10x10 two-dimensional grid, then the input layer has 100 nodes, each corresponding to a refrigerant concentration value in the refrigerant leakage distribution map. The hidden layer contains several fully connected layers, convolutional layers, or other network structures suitable for processing spatial data, which learn the spatial patterns and constraints in the input data. The number of nodes in the output layer is consistent with the resolution of the predicted leakage area, and a predicted refrigerant leakage estimation map is output. The neural network model is trained on the training data set through the back propagation algorithm to optimize the weight parameters. During the training process, the neural network model predicts a refrigerant leakage estimation map based on the constraints between each sensor by learning the spatial relationship between the sensor data and the physical propagation law of the refrigerant leakage. The neural network model takes into account the relative position between the sensors and the distribution law of the refrigerant concentration, identifies the hot spots where the leakage may occur, and infers how the refrigerant concentration propagates and diffuses in space. During the training process, the neural network model is optimized using an appropriate loss function (such as mean square error, MSE).

[0040] The method provided in this embodiment can effectively improve the accuracy and reliability of refrigerant leakage detection and concentration estimation by combining the data collected by the sensor, the prediction ability and correction mechanism of the neural network model. First, the electrical signal value collected by the sensor directly reflects the local gas concentration, but due to the error of the sensor itself, environmental interference or uneven layout, the initially constructed refrigerant leakage distribution map may have blind spots or errors. By inputting these data into the neural network model, the neural network model's modeling ability for spatial relationships can be used to automatically learn the propagation law of gas leakage in space and the correlation between different sensor data, thereby generating a more accurate refrigerant leakage prediction map. The neural network model can identify and remove outliers caused by sensor errors, noise or other environmental factors, thereby improving the robustness of the prediction. Furthermore, by comparing and correcting the initially generated refrigerant leakage distribution map, the accuracy of the refrigerant leakage distribution map can be improved as a whole, so that the gas concentration distribution is more in line with the actual leakage scenario. This method of generating comprehensive refrigerant concentration based on the corrected refrigerant leakage distribution map can provide a global and reliable comprehensive refrigerant concentration assessment by weighted average or other mathematical methods to accurately assess the leakage risk of the refrigerated compartments of fishing vessels and take timely measures. This system setting not only improves the measurement accuracy, but also can cope with a variety of interference factors in complex environments, thereby ensuring the efficiency and accuracy of real-time monitoring.

[0041] Optionally, the method of determining the corresponding leakage mode and leakage position based on the comprehensive refrigerant concentration includes: acquiring structural information of the refrigerated compartment of the fishing vessel, establishing a refrigerant propagation curve based on the structural information, and determining a first preset threshold and a second preset threshold based on the refrigerant propagation curve; the first preset threshold is less than the second preset threshold, and the first preset threshold and the second preset threshold correspond to different leakage levels; based on the relationship between the comprehensive refrigerant concentration and the first preset threshold and the second preset threshold, determining a target threshold reached by the comprehensive refrigerant concentration; determining the corresponding leakage mode based on the target threshold; calculating the difference between the refrigerant concentration at each position and the comprehensive refrigerant concentration, and determining the leakage position based on the sensor corresponding to the refrigerant concentration with the largest difference.

[0042] Specifically, the first preset threshold and the second preset threshold are determined based on the structural information of the refrigerated compartment of the fishing vessel, the characteristics of the refrigerant, and the concentration that is harmful to the human body. The first preset threshold is less than the second preset threshold. The first preset threshold is the warning concentration, which indicates a preliminary abnormality in the refrigerant concentration and can trigger an early warning. The second preset threshold is the dangerous concentration, which indicates that the refrigerant concentration has reached a level that causes obvious harm to the human body and requires emergency response measures to be triggered. The leakage mode (minor leakage) corresponding to the first preset threshold is different from the leakage mode (serious leakage) corresponding to the second preset threshold.

[0043] In the specific implementation, the detailed structural information of the refrigerated compartment of the fishing vessel is obtained, including the size, layout, location of walls, doors and windows, air flow channels, and cargo stacking areas of the compartment, and a three-dimensional spatial model of the refrigerated compartment of the fishing vessel is established based on the detailed structural information. According to the three-dimensional spatial model of the refrigerated compartment of the fishing vessel, combined with the fluid dynamics model and the gas leakage propagation theory, the influence of factors such as air flow, temperature change, and compartment partition on gas propagation is considered, and the refrigerant propagation curve of the refrigerant in the compartment is established. The refrigerant propagation curve characterizes the change of the refrigerant concentration in different areas of the compartment over time. By analyzing the change trend of the refrigerant propagation curve, the chemical properties of the refrigerant, and the safety of the refrigerant to the human body, the concentration range of the refrigerant that is harmful to the human body is identified, and the first preset threshold and the second preset threshold are set based on the analysis results. Further, the size relationship between the integrated refrigerant concentration obtained by fusion and the first preset threshold and the second preset threshold is compared. If the integrated refrigerant concentration is lower than the first preset threshold, it is determined that there is no abnormality and no leakage occurs. If the integrated refrigerant concentration is between the first preset threshold and the second preset threshold, it is determined that the first preset threshold is reached, and the leakage mode is determined to be a slight leakage, and emergency treatment may not be required. If the comprehensive refrigerant concentration is higher than the second preset threshold, it is determined that the second preset threshold is reached, and the leakage mode is determined to be a serious leakage. It is necessary to immediately initiate an emergency response, including evacuating personnel, closing the source of leakage, etc. When it is determined that a leak has occurred, that is, the leakage mode is a minor leak or a serious leak, it is necessary to further determine the specific leak location. By performing a difference calculation on the refrigerant concentration of each position detected by the detection module, the difference between it and the comprehensive refrigerant concentration is calculated, and the sensor position with the largest concentration difference is identified, and the position is determined as the possible source of leakage, that is, the leakage position. If the concentration difference of multiple sensor positions is similar, it is necessary to further confirm the leakage location in combination with the airflow path and the layout of the refrigerated compartment of the fishing vessel.

[0044] Optionally, determining the first preset threshold and the second preset threshold based on the refrigerant propagation curve includes: determining the first preset threshold based on the refrigerant propagation curve and a first exposure concentration limit; the first exposure concentration limit represents an average concentration allowed for the human body in the refrigerated compartment of the fishing boat within a first preset time; determining the second preset threshold based on the refrigerant propagation curve and the second exposure concentration limit; the second exposure concentration limit represents the maximum concentration that the human body can withstand within a second preset time in the refrigerated compartment of the fishing boat, the first preset time being greater than the second preset time.

[0045] Specifically, the first preset time and the second preset time are set according to relevant standards, and this is not limited in the present embodiment. It should be noted that the first preset time is greater than the second preset time, the first preset time is usually a longer period of time, and the second preset time is usually a short period of time. For example, the first preset time is 8 hours, and the second preset time is 15 minutes. The first exposure concentration limit represents the average concentration that the human body can safely contact over a long period of time, and the second exposure concentration limit represents the maximum concentration that the human body can withstand in a short period of time. The first preset threshold is used to assess the risk of leakage under long-term exposure conditions, and the second preset threshold is used to assess the emergency leakage state in a short period of time.

[0046] In specific implementation, based on the refrigerant propagation curve, the position of the point where the refrigerant concentration reaches or remains at the first exposure concentration limit over time is calculated as the first preset threshold. That is, the area where the refrigerant concentration can maintain the first exposure concentration limit within the first preset time is determined based on the refrigerant propagation curve. Based on the refrigerant propagation curve, the position of the point where the refrigerant concentration can reach or exceed the concentration limit under the second exposure concentration limit is calculated as the second preset threshold. That is, the area where the refrigerant concentration reaches the second exposure concentration limit within the second preset time is determined based on the refrigerant propagation curve.

[0047] Optionally, before fusing the refrigerant concentrations at multiple locations based on the neural network model to obtain the fused comprehensive refrigerant concentration, the processing module is also used to perform abnormal value detection on the refrigerant concentrations detected by each sensor; if the refrigerant concentration difference in adjacent detection cycles exceeds a threshold, determine the corresponding sensor as a target sensor; determine a refrigerant leakage distribution map based on the refrigerant concentrations of multiple adjacent sensors of the target sensor, determine a leakage curve based on the refrigerant concentration of any adjacent sensor in the previous cycle, predict the predicted refrigerant concentration of any adjacent sensor in the current cycle based on the leakage curve, and correct the leakage curve based on the predicted refrigerant concentration and the refrigerant leakage distribution map; predict the predicted leakage value of the target sensor based on the corrected leakage curve, compare the predicted leakage value with the actual refrigerant concentration of the target sensor in the current cycle, and correct the abnormal detection result of the target sensor based on the comparison result; if the refrigerant concentration of the target sensor is still abnormal after correction, determine that the target sensor is a faulty sensor, and mark the refrigerant concentration detected by the faulty sensor as abnormal data and eliminate it.

[0048] Specifically, when the sensor is detecting, there may be times when the sensor is working abnormally or is damaged. If the sensor measurement value (refrigerant concentration) rises sharply in a short period of time and is higher than the normal value, but the actual refrigerant concentration has not reached this value, then the measurement value is an abnormal value, and the abnormal value will not be subsequently transmitted to the processing module for fusion. Therefore, before the processing module fuses the refrigerant concentrations of multiple locations, it is necessary to perform abnormal value detection on the refrigerant concentrations of each location detected by the sensor, eliminate abnormal data, and only fuse normal data.

[0049] In the specific implementation, refrigerant concentration data from multiple sensors are collected to form multiple time series data sets, each of which contains the refrigerant concentration value of each sensor in different time periods. For each time series data set, the difference in refrigerant concentration between sensors in two adjacent detection periods is calculated. If the difference in refrigerant concentration between adjacent detection periods exceeds the set threshold, the sensor is determined to be the target sensor and may be abnormal. Furthermore, according to the spatial layout of the sensors, several sensors adjacent to the target sensor are selected, and the refrigerant concentration data of the target sensor and its adjacent sensors are used to construct a refrigerant leakage distribution map using an interpolation algorithm to simulate the distribution of refrigerant concentration in space. The refrigerant concentration data of a target adjacent sensor is selected from the adjacent sensors of the previous cycle of the current detection cycle as a reference point. Based on the refrigerant concentration data of the target adjacent sensor, a leakage curve is generated in combination with a propagation model (such as a diffusion equation or a model based on a propagation curve). Based on the leakage curve and the refrigerant concentration data of the adjacent sensor of the previous cycle of the current cycle, the predicted refrigerant concentration of the adjacent sensor in the current cycle is predicted: ; Among them, the is the refrigerant concentration of the adjacent sensor in the current cycle; is the refrigerant concentration of the adjacent sensor in the previous cycle; is the leakage curve prediction function; is the leakage curve.

[0050] Furthermore, the predicted refrigerant concentration is compared with the actually measured refrigerant concentration of the adjacent sensor, and the parameters of the leakage curve, such as the concentration decay rate, are corrected according to the difference in refrigerant concentration. Based on the corrected leakage curve and the refrigerant concentration of the target sensor in the previous cycle, the refrigerant concentration of the target sensor in the current cycle is predicted, and the refrigerant concentration of the target sensor in the current cycle is compared with the actual refrigerant concentration of the target sensor in the current cycle. The abnormal detection result of the target sensor is corrected based on the difference in refrigerant concentration. If the difference in refrigerant concentration is within the preset range, it is determined that the concentration detection result of the target sensor is reasonable, and the abnormal detection result can be corrected to normal, and the detection capability of the target sensor is improved by adjusting the prediction curve or updating the preset range. If the difference in refrigerant concentration is large, it is determined that there may be a problem with the detection result of the target sensor, which may be due to factors such as sensor failure or external interference. The target sensor is marked as a faulty sensor, and its data is removed or regarded as invalid data to avoid it affecting the overall analysis results.

[0051] The device provided in this embodiment corrects and optimizes sensor data through multiple mechanisms, thereby enhancing the robustness, accuracy and reliability of the system. First, by using a neural network model to fuse the refrigerant concentrations at multiple locations, a more comprehensive and accurate estimation of the refrigerant concentration can be provided by integrating the data of different sensors. This multi-source information fusion method avoids local errors or errors that may occur in a single sensor and improves the overall monitoring accuracy. On this basis, the outlier detection mechanism ensures effective screening and control of the quality of sensor data. When the concentration difference of a sensor in adjacent cycles exceeds a predetermined threshold, the system automatically identifies it as a target sensor for further analysis, thereby promptly discovering possible faulty sensors. By correcting the leakage curve and predicting the concentration of adjacent sensors, the expected value of the model can be dynamically adjusted to correct the data of the target sensor. This process is based on the comparison between the actual concentration and the predicted value, which can effectively identify the accuracy of the sensor and further correct its data. If the concentration of the target sensor is still abnormal, the system will mark it as a faulty sensor and remove its data to avoid the abnormal value of the faulty sensor from misleading the overall analysis. This design not only ensures that the system detects and corrects data in real time, but also dynamically identifies and eliminates faulty sensors through intelligent self-correction, thereby improving the accuracy and stability of the entire system, avoiding monitoring blind spots caused by sensor failure or errors, and thus improving the reliability and safety of the gas leak monitoring system.

[0052] Optionally, after the processing module determines the corresponding leakage pattern and leakage location, the processing module is also used to store relevant data of refrigerant leakage in a non-volatile memory, and the non-volatile memory can retain the relevant data when the refrigerant leakage detection device for the refrigerated compartment of the fishing vessel is powered off or restarted, and the relevant data at least includes the electrical signal, the leakage pattern, and the leakage location.

[0053] Specifically, the non-volatile memory has the ability to retain data, and the recorded data will not be lost even when the power is turned off.

[0054] The device provided in this embodiment, by setting a non-volatile memory, immediately records the relevant data in the non-volatile memory after the processing module identifies the leakage pattern. This design makes subsequent data analysis more convenient and reliable, and users can review and analyze these relevant data in detail afterwards, so as to gain an in-depth understanding of the cause and frequency of the leakage and improve the maintenance efficiency of the equipment. In addition, this design not only improves the accuracy of refrigerant leakage monitoring, but also helps companies comply with environmental regulations and reduce pollution to the atmosphere. By regularly analyzing the stored data, companies can take preventive measures to avoid serious leakage incidents that may occur in the future, thereby protecting the ecological environment and reducing operating costs.

[0055] Figure 2 This is a schematic diagram of the structure of the refrigerant leakage detection device for the refrigerated compartment of a fishing vessel provided in this application. Figure 2 , the device also includes an alarm module, which is connected to the processing module; the alarm module includes a buzzer, an LED warning light and a wireless communication unit; wherein, The buzzer and the LED warning light are used to issue an audible warning and a visual warning respectively when a refrigerant leak is determined to exist, so as to remind the staff on the fishing vessel that there is a risk of leakage;

[0056] The wireless communication unit is used to send the leakage pattern and leakage position to the control system on the fishing boat, and the staff performs leakage inspection based on the leakage pattern and leakage position.

[0057] For details, please refer to Figure 2, the refrigerant leakage detection device for the refrigerated compartment of a fishing vessel also includes an alarm module, which is interconnected with the processing module. The alarm module includes a high-volume buzzer, a high-brightness LED warning light, and a wireless communication unit. When the processing module determines that there is a refrigerant leak (that is, when the concentration of the refrigerant is detected to be abnormal), the alarm module connected to the processing module will immediately take effect. First, the high-volume buzzer will emit a strong sound to attract the attention of the crew on board and remind the crew to pay attention to potential dangers. At the same time, the high-brightness LED warning light will flash quickly to provide visual warning information to ensure that it can effectively attract the attention of the crew even in a noisy environment. Further, in addition to the sound and light alarm, the alarm module will also send the alarm information (leakage mode and leakage location) to the control system on the fishing vessel in a timely manner through the wireless communication unit. In this way, the managers on the fishing vessel can obtain the alarm information in real time so as to take countermeasures quickly to ensure the safety of the vessel and its staff.

[0058] Optional, please continue to refer to Figure 2 The device further comprises a ventilation control module, the ventilation control module is connected to the processing module, and the ventilation control module is connected to the ventilation system of the fishing boat via a relay; wherein, The ventilation control module is used to send an opening command to the ventilation system of the fishing boat through the relay when it is determined that there is a refrigerant leakage, so as to cause outside air to flow into the refrigerated compartment and discharge the refrigerant in the refrigerated compartment until the refrigerant concentration in the refrigerated compartment is lower than the safety threshold, and then close the ventilation system; the opening time and operating frequency of the ventilation system are positively correlated with the refrigerant concentration.

[0059] For details, please refer to Figure 2 The refrigerant leakage detection device for the refrigerated compartment of the fishing boat also includes a ventilation control module. The ventilation control module is interconnected with the processing module. The ventilation control module is closely connected with the ventilation system in the refrigerated compartment of the fishing boat through a relay, and is used to adjust the environmental conditions inside the refrigerated compartment of the fishing boat.

[0060] In specific implementation, when the processing module detects that the comprehensive refrigerant concentration in the refrigerated compartment of the fishing boat reaches the safety threshold, that is, the refrigerant concentration exceeds the normal range, the control unit will send an opening command to the ventilation system through the relay, automatically start the ventilation equipment (ventilator and exhaust equipment), prompting fresh air to flow into the refrigerated compartment of the fishing boat quickly, and discharge the high-concentration refrigerant. During the ventilation process, the changes in the refrigerant concentration in the refrigerated compartment of the fishing boat are continuously monitored through the built-in sensor. Once the refrigerant concentration drops to a safe range, the ventilation control module automatically turns off the ventilation equipment to save energy and maintain a stable temperature in the refrigerated compartment of the fishing boat.

[0061] It should be noted that the opening time and operating frequency of the ventilation system are positively correlated with the refrigerant concentration in the refrigerated compartment of the fishing vessel. The higher the refrigerant concentration, the more it exceeds the safety threshold. Correspondingly, the longer the ventilation system is opened, the higher the operating frequency. The ventilation control module can also be set according to demand to adjust the operating frequency and duration of the ventilation equipment to ensure the best ventilation effect. Through such automated control, the operation of the refrigerated compartment of the fishing vessel has not only become more efficient, but also enhanced overall safety. The intelligent design of the system enables continuous monitoring and response under complex environmental conditions, providing a reliable solution for various demanding refrigeration needs.

[0062] Optional, please continue to refer to Figure 2 The device further comprises a power management module, which uses a lithium-ion battery and a thermocouple energy recovery system to provide power for the operation of the device; the thermocouple energy recovery system is installed on the inner and outer walls of the refrigerated compartment of the fishing boat to detect the internal temperature and the external temperature of the refrigerated compartment of the fishing boat respectively;

[0063] The thermocouple energy recovery system compares the internal temperature and the external temperature in real time, calculates the temperature difference between the inner and outer walls, utilizes the Seebeck effect of the thermocouple material, converts the detected temperature difference into an electric potential signal, generates current, and stores it in the lithium-ion battery.

[0064] For details, please refer to Figure 2 The fishing vessel refrigerated compartment refrigerant leakage detection device also includes a power management module, which is interconnected with other modules of the fishing vessel refrigerated compartment refrigerant leakage detection device (detection module, transmission module, processing module, alarm module, ventilation control module) to provide power for the operation of other modules. The power management module includes a lithium-ion battery and a thermocouple energy recovery system. Among them, the capacity of the lithium-ion battery is 3000mAh, which can support the entire device to work continuously for 48 hours.

[0065] Furthermore, the thermocouple energy recovery system is installed inside and outside the refrigerated compartment of the fishing boat to continuously monitor the internal and external temperatures of the refrigerated compartment of the fishing boat. The thermocouple energy recovery system uses the temperature difference between the inside and outside of the refrigerated compartment of the fishing boat, converts the temperature difference into electrical energy through the special characteristics of the thermocouple material, and stores the electrical energy in the lithium-ion battery, thereby providing additional power support for other modules. Specifically, the thermocouple energy recovery system is installed on the inner and outer walls of the refrigerated compartment of the fishing boat. It can efficiently sense the temperature difference and generate current through the semiconductor material behind it. The collected electrical energy can not only enrich the reserve of the lithium-ion battery, but also directly power the detection module and the transmission module under high demand conditions, thereby extending the overall operation time of the device.

[0066] Furthermore, the power management module is equipped with an advanced power monitoring system to monitor the power status of the lithium-ion battery in real time, ensuring that when the power is lower than the preset power, an alarm can be issued in time to remind the user to charge. In addition, the power management module is equipped with an intelligent power scheduling module, which can dynamically adjust the power distribution of each module according to the different working states of the leakage detection device. For example, when the detection module is collecting data, the power management module can give priority to providing more power support to make it work more stable and efficient; while in standby mode, the power management module can automatically reduce power consumption to maintain the lowest power consumption. This flexible power management strategy not only improves the operating efficiency of the system, but also effectively extends the battery life.

[0067] The refrigerant leakage detection device for the refrigerated compartment of a fishing vessel provided in this embodiment, on the one hand, the layout positions of multiple sensors in the detection module are determined based on the actual layout of the refrigerated compartment of the fishing vessel, ensuring that the refrigerant concentration at each position in the refrigerated compartment of the fishing vessel can be monitored in real time, which avoids the monitoring blind spot problem that may exist in fixed sensors. The traditional sensor layout is often limited by the installation position and may not cover every corner of the refrigerated compartment of the fishing vessel, resulting in the inability to monitor the leakage in real time in some areas. However, through reasonable layout and spacing optimization, this design ensures that all areas in the cabin can be effectively covered by sensors, enhancing the comprehensiveness and accuracy of refrigerant concentration monitoring. On the other hand, by combining the data collected by the sensor, the prediction ability and correction mechanism of the neural network model, the accuracy and reliability of refrigerant leakage detection and concentration estimation can be effectively improved. First, the electrical signal value collected by the sensor directly reflects the local gas concentration, but due to the error of the sensor itself, environmental interference or uneven layout, the initially constructed refrigerant leakage distribution map may have blind spots or errors. By inputting these data into the neural network model, the neural network model's modeling ability for spatial relationships can be used to automatically learn the propagation law of gas leakage in space and the correlation between different sensor data, thereby generating a more accurate refrigerant leakage prediction map. The neural network model can identify and remove outliers caused by sensor errors, noise or other environmental factors, thereby improving the robustness of the prediction. Furthermore, by comparing and correcting the initially generated refrigerant leakage distribution map, the accuracy of the refrigerant leakage distribution map can be improved as a whole, making the gas concentration distribution more consistent with the actual leakage scenario. This method of generating a comprehensive refrigerant concentration based on the corrected refrigerant leakage distribution map can provide a global and reliable comprehensive refrigerant concentration assessment by combining the refrigerant concentrations of each point through weighted average or other mathematical methods, accurately assess the leakage risk of the refrigerated compartment of the fishing vessel and take timely measures. This system setting not only improves the measurement accuracy, but also can cope with a variety of interference factors in a complex environment, thereby ensuring the efficiency and accuracy of real-time monitoring. Thirdly, before the data module performs data fusion, by performing outlier detection on the refrigerant concentration detected by the sensor, abnormal data caused by sensor failure or external factors can be discovered and eliminated in real time. A sudden change in refrigerant concentration may not represent a leak, but may be a fault in the sensor itself or an abnormality caused by environmental factors. By comparing the data with adjacent sensors, sensors with inconsistent trends can be identified, faulty sensors can be determined, and their inaccurate data can be excluded to avoid the error of a single sensor affecting the judgment of the entire refrigerant leak. In addition, outlier detection can reduce the occurrence of false alarms. If only relying on the detection of a single sensor, once a fault or error occurs, it may trigger a false alarm in the system, resulting in unnecessary emergency response, wasting resources and affecting the work efficiency of the crew.Through intelligent comparison and analysis, the system can automatically identify anomalies, avoid false alarms caused by errors, and ensure the reliability and stability of the system. Fourthly, based on the relationship between the integrated refrigerant concentration and the preset threshold, the corresponding leakage mode and leakage location are determined, and the corresponding alarm method is adopted. First, the graded alarm can clearly divide the severity of the leakage according to the different refrigerant concentrations, so that the crew can take corresponding countermeasures according to the alarm level. When the refrigerant concentration exceeds the first preset threshold, the crew can monitor and make preliminary adjustments, such as strengthening ventilation or adjusting refrigeration equipment. When the concentration exceeds the second preset threshold, a more urgent alarm will be triggered, requiring immediate and more stringent emergency measures, such as evacuation of personnel or finding the source of the leak. This grading mechanism enables the crew to respond flexibly, without unnecessary emergency responses due to low-concentration leaks, and avoids the problem of untimely response in high-concentration leaks. Secondly, by calculating the difference between the refrigerant concentration and the comprehensive concentration at each location and determining the leak location, it is not only possible to judge the severity of the leak, but also to accurately locate the leak location, helping the crew to repair or handle it more efficiently. Compared with the traditional fixed sensor layout, this can avoid missing the detected dead corners and improve the accuracy of leak detection. In addition, the hierarchical alarm reduces false alarms, avoids the waste of resources caused by system false alarms in low-concentration leaks, and ensures timely response in high-concentration leaks to maximize the safety of personnel. In general, this intelligent and hierarchical alarm mechanism effectively improves the response speed and accuracy of refrigerant leak detection through threshold setting, concentration difference calculation and precise leak location, while enhancing the crew's ability to respond to leaks of different levels, thereby providing more flexible and accurate protection for the safety management of fishing boat refrigerated compartments. Fifthly, the power management module is set as a lithium-ion battery and a thermocouple energy recovery system. The combination of lithium-ion batteries and thermocouple energy recovery systems enables the refrigerant leak monitoring device to have efficient energy utilization capabilities. Among them, the thermocouple energy recovery system detects the temperature difference between the inside and outside of the refrigerated compartment of the fishing boat and uses the Seebeck effect to convert heat energy into electrical energy to achieve self-powered. This method does not rely on the replacement and charging of traditional batteries, greatly reducing manual maintenance costs and ensuring the continuous operation of the device, especially when energy is limited or the ship is far away from the supply station. Secondly, by real-time monitoring of the temperature difference between the inside and outside, the thermocouple system can dynamically generate electrical energy, so that even when the ship is operating at sea, the system can continue to provide energy, avoiding the risks of external power interruptions or failures, and improving the reliability and independence of the system. At the same time, lithium-ion batteries, as energy storage devices, have high energy density, long life and strong durability, which can effectively store electrical energy from the thermocouple system, providing guarantee for the long-term stable operation of the device.In summary, this power management module based on temperature difference power generation can achieve efficient utilization of green energy. It not only reduces dependence on external power sources, but also improves the autonomy, stability and sustainability of the equipment, ensuring the long-term and stable operation of the refrigerant leakage detection device in the refrigerated compartment of the fishing vessel.

[0068] Corresponding to the aforementioned embodiment of a refrigerant leakage detection device for a refrigerated compartment of a fishing vessel, the present application also provides an embodiment of a refrigerant leakage detection method for a refrigerated compartment of a fishing vessel.

[0069] Figure 3 This is a flow chart of a method for detecting refrigerant leakage in a refrigerated compartment of a fishing vessel provided in Example 2 of the present application. Figure 3 The method provided in this embodiment is applied to the refrigerant leakage detection device for a refrigerated compartment of a fishing vessel according to any one of the first aspects of the present application, and the method comprises:

[0070] S301. Determine the locations of multiple sensors based on the layout of the refrigerated compartment of the fishing boat, and detect the refrigerant concentrations at multiple locations in the refrigerated compartment of the fishing boat based on the sensors; wherein the multiple sensors are densely distributed in the refrigerated compartment of the fishing boat, and the refrigerant concentrations at the corresponding locations detected by each sensor are different.

[0071] S302, fusing refrigerant concentrations at multiple locations based on a neural network model to obtain a fused comprehensive refrigerant concentration, and determining a corresponding leakage mode and leakage location based on the comprehensive refrigerant concentration;

[0072] Among them, a refrigerant leakage distribution map of the refrigerated compartment of the fishing vessel is constructed based on the electrical signal values ​​collected by each sensor, the refrigerant leakage distribution map is input into the neural network model, a refrigerant leakage estimation map is predicted based on the refrigerant leakage constraints between each sensor, the refrigerant leakage distribution map is corrected based on the refrigerant leakage estimation map, and the comprehensive refrigerant concentration is determined based on the corrected refrigerant leakage distribution map.

[0073] The method of this embodiment can be used to execute Figure 1 The steps, specific implementation principles and implementation processes of the device embodiment shown are similar and will not be repeated here.

[0074] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0075] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present application scheme. A person of ordinary skill in the art can understand and implement it without paying any creative work.

[0076] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A refrigerant leakage detection device for a refrigerated compartment of a fishing vessel, characterized in that: The device comprises a detection module, a transmission module, and a processing module connected in sequence; wherein, The detection module is used to determine the locations of multiple sensors based on the layout of the refrigerated compartment of the fishing boat, detect the refrigerant concentration at multiple locations in the refrigerated compartment of the fishing boat based on the sensors, and convert the refrigerant concentration into an electrical signal; wherein the multiple sensors are densely distributed in the refrigerated compartment of the fishing boat, and the refrigerant concentration at the corresponding location detected by each sensor is different; The transmission module is used to transmit the electrical signal detected by the detection module to the processing module; The processing module is used to fuse the refrigerant concentrations at multiple locations based on the neural network model to obtain a fused comprehensive refrigerant concentration, and determine the corresponding leakage mode and leakage location based on the comprehensive refrigerant concentration; Among them, a refrigerant leakage distribution map of the refrigerated compartment of the fishing vessel is constructed based on the electrical signal values ​​collected by each sensor, the refrigerant leakage distribution map is input into the neural network model, a refrigerant leakage estimation map is predicted based on the refrigerant leakage constraints between each sensor, the refrigerant leakage distribution map is corrected based on the refrigerant leakage estimation map, and the comprehensive refrigerant concentration is determined based on the corrected refrigerant leakage distribution map.

2. The device according to claim 1, characterized in that The determining the corresponding leakage mode and leakage position based on the comprehensive refrigerant concentration includes: Acquire structural information of a refrigerated compartment of a fishing vessel, establish a refrigerant propagation curve based on the structural information, and determine a first preset threshold and a second preset threshold based on the refrigerant propagation curve; the first preset threshold is less than the second preset threshold, and the first preset threshold and the second preset threshold correspond to different leakage levels; Determining a target threshold value reached by the comprehensive refrigerant concentration based on a relationship between the comprehensive refrigerant concentration and the first preset threshold value and the second preset threshold value; determining a corresponding leakage mode based on the target threshold; The difference between the refrigerant concentration at each position and the comprehensive refrigerant concentration is calculated, and the leakage position is determined based on the sensor corresponding to the refrigerant concentration with the largest difference.

3. The device according to claim 2, characterized in that The determining the first preset threshold and the second preset threshold based on the refrigerant propagation curve comprises: Determining a first preset threshold value based on the refrigerant propagation curve and a first exposure concentration limit value; the first exposure concentration limit value represents an average concentration allowed for a human body in the refrigerated compartment of the fishing vessel within a first preset time; A second preset threshold is determined based on the refrigerant propagation curve and a second exposure concentration limit; the second exposure concentration limit represents the maximum concentration that the human body can withstand within a second preset time in the refrigerated compartment of the fishing boat, and the first preset time is greater than the second preset time.

4. The device according to claim 1, characterized in that Before fusing the refrigerant concentrations at multiple locations based on the neural network model to obtain the fused comprehensive refrigerant concentration, the processing module is also used to perform abnormal value detection on the refrigerant concentrations detected by each sensor; If the refrigerant concentration difference between adjacent detection cycles exceeds a threshold, the corresponding sensor is determined to be a target sensor; Determine a refrigerant leakage distribution map according to the refrigerant concentration of a plurality of adjacent sensors of the target sensor, determine a leakage curve based on the refrigerant concentration of any adjacent sensor in a previous cycle, predict a predicted refrigerant concentration of any adjacent sensor in a current cycle based on the leakage curve, and correct the leakage curve based on the predicted refrigerant concentration and the refrigerant leakage distribution map; Predicting a predicted leakage value of the target sensor based on the corrected leakage curve, comparing the predicted leakage value with an actual refrigerant concentration of the target sensor in a current cycle, and correcting an abnormal detection result of the target sensor based on the comparison result; If the refrigerant concentration of the target sensor is still abnormal after correction, the target sensor is determined to be a faulty sensor, and the refrigerant concentration detected by the faulty sensor is marked as abnormal data and discarded.

5. The device according to claim 1, characterized in that The detection module includes a plurality of gas sensors, and the gas sensors are corrosion-resistant; the detection module is used to determine the layout positions of the plurality of sensors based on the layout of the refrigerated compartment of the fishing vessel, and includes: Calculating the coverage area of ​​a single gas sensor based on the detection sensitivity and effective detection radius of the gas sensor; Based on the coverage area of ​​the gas sensors, determining the optimal arrangement spacing between adjacent gas sensors; wherein each gas sensor is interconnected, and the spacing between adjacent gas sensors is the same; Based on the coverage area and the optimal arrangement spacing, the gas sensors are densely distributed in the refrigerated compartment of the fishing boat; The layout position of the gas sensor is adjusted based on the cabin shape, air flow path and cargo stacking area of ​​the refrigerated cabin of the fishing boat.

6. The device according to claim 1, characterized in that The device also includes an alarm module, which is connected to the processing module; the alarm module includes a buzzer, an LED warning light and a wireless communication unit; wherein, The buzzer and the LED warning light are used to issue an audible warning and a visual warning respectively when a refrigerant leak is determined to exist, so as to remind the staff on the fishing vessel that there is a risk of leakage; The wireless communication unit is used to send the leakage pattern and leakage position to the control system on the fishing boat, and the staff performs leakage inspection based on the leakage pattern and leakage position.

7. The device according to claim 1, characterized in that The device also includes a ventilation control module, which is connected to the processing module and is connected to the ventilation system of the fishing boat via a relay; wherein, The ventilation control module is used to send an opening command to the ventilation system of the fishing boat through the relay when it is determined that there is a refrigerant leakage, so as to cause outside air to flow into the refrigerated compartment and discharge the refrigerant in the refrigerated compartment until the refrigerant concentration in the refrigerated compartment is lower than the safety threshold, and then the ventilation control module is closed; the opening time and operating frequency of the ventilation system are positively correlated with the refrigerant concentration.

8. The device according to claim 1, characterized in that The device also includes a power management module, which uses a lithium-ion battery and a thermocouple energy recovery system to provide power for the operation of the device; the thermocouple energy recovery system is installed on the inner and outer walls of the refrigerated compartment of the fishing boat to detect the internal temperature and the external temperature of the refrigerated compartment of the fishing boat respectively; The thermocouple energy recovery system compares the internal temperature and the external temperature in real time, calculates the temperature difference between the inner and outer walls, utilizes the Seebeck effect of the thermocouple material, converts the detected temperature difference into an electric potential signal, generates current, and stores it in the lithium-ion battery.

9. The device according to claim 1, characterized in that After the processing module determines the corresponding leakage mode and leakage position, the processing module is also used to store relevant data of refrigerant leakage in a non-volatile memory. The non-volatile memory can retain the relevant data when the refrigerant leakage detection device for the refrigerated compartment of the fishing vessel is powered off or restarted. The relevant data at least includes the electrical signal, the leakage mode, and the leakage position.

10. A method for detecting refrigerant leakage in a refrigerated compartment of a fishing vessel, characterized in that: The method is applied to the refrigerant leakage detection device for a refrigerated compartment of a fishing vessel according to any one of claims 1 to 9, and the method comprises: Determine the locations of multiple sensors based on the layout of the refrigerated compartment of the fishing boat, and detect the refrigerant concentration at multiple locations in the refrigerated compartment of the fishing boat based on the sensors; wherein the multiple sensors are densely distributed in the refrigerated compartment of the fishing boat, and the refrigerant concentration at the corresponding location detected by each sensor is different; Based on the neural network model, refrigerant concentrations at multiple locations are fused to obtain a fused comprehensive refrigerant concentration, and a corresponding leakage mode and leakage location are determined based on the comprehensive refrigerant concentration; Among them, a refrigerant leakage distribution map of the refrigerated compartment of the fishing vessel is constructed based on the electrical signal values ​​collected by each sensor, the refrigerant leakage distribution map is input into the neural network model, a refrigerant leakage estimation map is predicted based on the refrigerant leakage constraints between each sensor, the refrigerant leakage distribution map is corrected based on the refrigerant leakage estimation map, and the comprehensive refrigerant concentration is determined based on the corrected refrigerant leakage distribution map.

Images