A ship LNG fuel tank thermal intrusion anomaly identification and disposal method and system based on BOG generation rate residual inversion

Abstract

The application discloses a kind of based on BOG generation rate residual inversion ship LNG fuel tank thermal intrusion anomaly identification and disposal method and system, belong to ship fuel tank monitoring control technical field.The method and system are by collecting fuel tank pressure, temperature, liquid level, BOG flow, compressor state, absorption equipment state, valve state and navigation environment data, calculate measured BOG generation rate and benchmark BOG generation rate;According to the residual of both, equivalent heat intrusion is converted, and in combination with residual amplitude, slope, duration, pressure recovery characteristics, valve correlation and navigation state correlation identify adiabatic degradation, local thermal bridge, valve leakage, abnormal sloshing heat exchange or sensor drift;Again output pressure window adjustment, compressor adjustment, absorption flow regulation, valve isolation and alarm instruction, and verify disposal effect by residual convergence condition.The application can identify abnormal thermal intrusion before tank pressure overrun, improve operation safety and BOG absorption stability.

Description

Technical Field

[0001] This invention relates to the field of ship fuel tank monitoring and control technology, specifically to a method and system for identifying and handling abnormal thermal intrusion in ship LNG fuel tanks based on BOG generation rate residual inversion. Background Technology

[0002] With the increasing number of LNG-fueled and dual-fuel vessels, BOG (Boil-Off Gas) is generated in the LNG fuel tanks during berthing, navigation, refueling, and load fluctuations. If BOG is not absorbed in a timely and stable manner, it can easily lead to increased fuel tank pressure and may cause frequent compressor start-ups and shutdowns, boiler combustion disturbances, gas pipeline pressure fluctuations, or safety venting.

[0003] Existing BOG (Boiled Air Gathering) handling solutions typically focus on BOG quantity calculation, compressor capacity configuration, heat exchanger configuration, boiler or main unit absorption capacity calculation, and piping layout. These solutions can achieve BOG handling under normal operating conditions, but their technical focus is mostly on equipment configuration or pressure threshold control, making it difficult to promptly identify the source of heat intrusion behind abnormal BOG increases.

[0004] In actual ship operation, an abnormally high BOG generation rate may be caused by factors such as decreased fuel tank insulation performance, localized thermal bridging in the hull, internal valve leakage, abnormal sloshing heat exchange, backflow or purging operations, and sensor drift. If the emergency response is only initiated when the fuel tank pressure exceeds the alarm threshold, the anomaly identification will be significantly delayed; if the BOG anomaly is judged solely based on the number of compressor start-stop cycles, it is easy to confuse normal sea state disturbances with continuous heat intrusion anomalies.

[0005] Therefore, there is a need to provide a monitoring and control method for ship LNG fuel tanks, which uses the BOG generation rate as a calculable state variable, the baseline BOG generation rate under normal operating conditions as a dynamic reference, and identifies the abnormal category of thermal intrusion through residual inversion. The identification results are then linked to the pressure setting window, BOG compressor, BOG disposal equipment, valves and alarm system, thereby achieving early identification, classification judgment and closed-loop handling. Summary of the Invention

[0006] The technical problem to be solved by this invention is: addressing the issue that existing ship LNG fuel tank BOG handling technology relies solely on pressure thresholds or equipment start-up / shutdown status as control criteria, resulting in delayed identification of thermal intrusion anomalies, difficulty in distinguishing anomaly sources, and insufficient targeting of handling instructions. This invention provides a method and system that can identify thermal intrusion anomaly categories in advance and generate closed-loop handling instructions based on BOG generation rate residuals.

[0007] To achieve the above objectives, the present invention provides the following technical solution: The present invention provides a method for identifying and handling abnormal thermal intrusion in ship LNG fuel tanks based on BOG generation rate residual inversion, comprising the following steps:

[0008] S1, collect operational data from the ship's LNG fuel tank. The operational data includes at least the fuel tank pressure, fuel tank gas phase temperature, fuel tank liquid phase temperature, liquid level, BOG venting flow rate, BOG compressor operating status, BOG disposal equipment operating status, fuel tank valve status, and ship navigation environment data.

[0009] S2. Based on the fuel tank pressure, fuel tank gas phase temperature, liquid level, and fuel tank geometric volume, establish the mass balance of the fuel tank gas phase space and calculate the measured BOG formation rate.

[0010] S3 calculates the baseline BOG generation rate based on normal fuel tank operation samples, environmental heat exchange parameters, liquid level, navigation environment data, and equipment operating status.

[0011] S4. Generate the BOG generation rate residual based on the difference between the measured BOG generation rate and the baseline BOG generation rate, and convert the BOG generation rate residual into the equivalent heat intrusion amount.

[0012] S5. Based on the amplitude, slope, duration, pressure recovery time constant, valve status correlation, and navigation status correlation of the BOG generation rate residual, the fuel tank thermal intrusion anomaly category is identified by inversion, and the anomaly confidence level is determined. The thermal intrusion anomaly category includes at least one of the following: adiabatic degradation, local thermal bridge, valve internal leakage, abnormal sloshing heat transfer, and sensor drift.

[0013] S6, based on the type of heat intrusion anomaly, the anomaly confidence level, and the equivalent heat intrusion amount, generates at least one of the following as a handling instruction: fuel tank pressure setting window adjustment instruction, BOG compressor start / stop or frequency converter regulation instruction, BOG absorption flow regulation instruction, valve isolation instruction, and alarm instruction.

[0014] S7. The disposal command is sent to the ship's fuel tank monitoring controller or integrated automation system, and the operation data continues to be collected after the disposal command is executed. The disposal effect is verified based on the changes in BOG generation rate residual, equivalent heat intrusion amount and pressure recovery time constant, so that the fuel tank pressure, BOG absorption amount and alarm status form a closed loop control.

[0015] The present invention also provides a system for identifying and handling abnormal thermal intrusion in ship LNG fuel tanks based on BOG generation rate residual inversion, including a data acquisition module, a BOG generation rate calculation module, a baseline generation rate prediction module, a residual inversion module, and a handling control module; wherein, the system is configured to execute the above method.

[0016] In a preferred embodiment, the measured BOG generation rate is obtained by balancing the change in the gas phase mass of the fuel tank with the amount of gas phase discharge and replenishment from the fuel tank, and compensation is made for the change in the gas phase space caused by the change in the fuel tank liquid level.

[0017] In a preferred embodiment, the baseline BOG generation rate is obtained from a normal thermal intrusion model, which includes a statistical model based on normal samples from ship trials or operations, a physical model based on the heat transfer of the fuel tank enclosure structure, or a hybrid model obtained by fusing the statistical model and the physical model.

[0018] In a preferred embodiment, the equivalent heat intrusion amount is obtained by multiplying the BOG generation rate residual by the latent heat of LNG vaporization, and the equivalent heat intrusion amount is accumulated through a sliding time window to distinguish between transient disturbances and persistent heat intrusion anomalies.

[0019] In a preferred embodiment, the handling instructions are output according to the level of abnormality, and are executed according to the type of abnormality, such as increasing the sampling frequency, starting the BOG compressor in advance, reconstructing the fuel tank pressure setting window, increasing the receiving flow of the BOG disposal equipment, restricting valve action, triggering valve isolation, outputting an alarm, or requiring manual confirmation.

[0020] Compared with existing technologies, this invention provides a method and system for identifying and handling abnormal thermal intrusion in ship LNG fuel tanks based on BOG generation rate residual inversion, which has the following beneficial effects:

[0021] 1. This invention further converts the BOG generation rate residual into an equivalent heat intrusion amount, giving the anomaly judgment a physical meaning and avoiding reliance solely on empirical pressure thresholds.

[0022] 2. This invention can distinguish between adiabatic degradation, local thermal bridges, valve internal leakage, abnormal sloshing heat transfer, and sensor drift by using residual amplitude, residual slope, duration, pressure recovery time constant, valve state correlation, and navigation state correlation.

[0023] 3. This invention transforms the abnormal classification results into control commands for the compressor, BOG disposal equipment, pressure setting window, valves, and alarm system, forming a closed-loop control process of monitoring, judgment, handling, and feedback, which is more in line with the actual implementation logic of industrial control systems.

[0024] 4. This invention mainly utilizes existing pressure, temperature, liquid level, flow rate, valves, and navigation status signals of the ship, without requiring a significant increase in mechanical equipment, making it easy to apply in new shipbuilding and the retrofitting of existing ships. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the system structure of the present invention;

[0026] Figure 2 This is a flowchart of the method of the present invention;

[0027] Figure 3This is a schematic diagram of the BOG generation rate residual inversion of the present invention;

[0028] Figure 4 This is a schematic diagram of the graded treatment control of the present invention. Detailed Implementation

[0029] The following is in conjunction with the reference appendix Figure 1 To the reference appendix Figure 4 The following examples illustrate specific embodiments of the present invention. These examples are intended to explain the implementation of the invention and are not intended to limit the scope of protection of the invention. Those skilled in the art can make adaptive adjustments to the data acquisition method, controller type, anomaly threshold, and model parameters without departing from the concept of the invention.

[0030] Example 1: System Architecture and Data Acquisition

[0031] Reference Appendix Figure 1 The ship's LNG fuel tank is equipped with pressure sensors, gas phase temperature sensors, liquid phase temperature sensors, and liquid level sensors; the BOG output pipeline is equipped with a mass flow meter; the BOG processing branch is equipped with a BOG compressor, heater, and BOG digestion equipment, which can be a dual-fuel boiler, dual-fuel generator, main engine gas supply unit, or combustion device. A valve status acquisition unit collects valve opening and closing status, valve opening degree, upstream pressure, downstream pressure, and downstream temperature. A ship navigation environment acquisition unit collects roll, pitch, acceleration, speed, ambient temperature, and seawater temperature.

[0032] The monitoring controller receives the aforementioned operational data and communicates with the ship's integrated automation system. The monitoring controller can be a standalone controller or integrated into the ship's fuel supply control system or integrated automation system. The monitoring controller's outputs include pressure setpoint adjustment commands, BOG compressor start / stop or frequency converter control commands, BOG absorption flow rate setting commands, valve isolation commands, and alarm commands.

[0033] Reference Appendix Figure 2 The monitoring controller collects data at sampling intervals and sets a sliding calculation window. The sampling interval is preferably 1 to 60 seconds, and the sliding calculation window is preferably 5 to 60 minutes. To avoid misjudgments caused directly by sensor malfunctions, the monitoring controller performs reliability verification on pressure, temperature, liquid level, and flow data. Reliability verification includes range verification, abrupt change verification, redundant measuring point deviation verification, valve status and differential pressure consistency verification, and compressor frequency and BOG flow consistency verification.

[0034] Example 2: Calculation of Measured BOG Generation Rate

[0035] Reference Appendix Figure 2The monitoring controller first calculates the measured BOG formation rate based on the mass balance of the fuel tank gas phase space. The fuel tank gas phase mass can be calculated using the following formula:

[0036] ;

[0037] in, For the first The BOG mass of the fuel tank at each sampling time. For the first Fuel tank pressure at each sampling time, For the first The volume of the fuel tank gas phase space at each sampling time. The average molar mass of BOG. For the first Gas compressibility factor at each sampling time. This is the universal gas constant. For the first The gas phase temperature of the fuel tank at each sampling time.

[0038] The volume of the fuel tank gas phase space can be calculated using the following formula:

[0039] ;

[0040] in, This refers to the total volume of the fuel tank. For the first The volume of LNG liquid phase at each sampling time. The meaning is the same as in the aforementioned formula.

[0041] After obtaining the fuel tank gas phase mass at adjacent sampling times, the measured BOG formation rate can be calculated using the following formula:

[0042] ;

[0043] in, For the first Measured BOG generation rate at each sampling time. For the first The BOG mass of the fuel tank at each sampling time. The sampling interval is... For the first BOG mass flow rate output from the fuel tank at each sampling time For the first The mass flow rate of gas entering the gas phase space of the fuel tank at each sampling time.

[0044] When the ship is undergoing refueling, reflux, purging, or inerting operations, the mass flow rate of gas entering or exiting the fuel tank vapor space is corrected according to the corresponding pipeline flow meter. When none of the above operations are performed, the mass flow rate of gas entering the fuel tank vapor space can be zero, or determined based on the measured flow rate of the return branch. To reduce measurement noise, the monitoring controller uses a sliding window to smooth the measured BOG formation rate, resulting in a smoothed measured BOG formation rate:

[0045] ;

[0046] in, For the first Smoothed measured BOG generation rate at each sampling time point To calculate the number of sampling points within the sliding calculation window, The sampling point number within the sliding calculation window. For the first The measured BOG generation rate at each sampling time.

[0047] Example 3: Benchmark BOG Generation Rate Model

[0048] Reference Appendix Figure 2 and reference appendix Figure 3 The baseline BOG generation rate represents the BOG generation rate that should occur during normal operation of the fuel tank under current liquid level, temperature, environmental, and equipment conditions. This baseline value is not a fixed threshold but a dynamic reference value that varies with operating conditions. The baseline BOG generation rate can be obtained using the following formula:

[0049]

[0050] in, For the first The baseline BOG generation rate at each sampling time. This is a statistical prediction model trained on normal samples. For statistical prediction model parameters, For the first The operating condition feature vector at each sampling time, To ensure proper heat transfer to the fuel tank enclosure structure, Explainable heat introduced by pumps, piping, or recirculation operations. This is the correction for normal sloshing heat transfer caused by ship motion. For the first The latent heat of LNG vaporization at each sampling time.

[0051] The preferred operating condition feature vector includes fuel tank level, gas phase temperature, liquid phase temperature, ambient temperature, seawater temperature, ship roll, pitch, acceleration, speed, compressor status, BOG disposal equipment status, and valve status. The statistical prediction model can employ multiple linear regression, recursive least squares model, state-space model, or lightweight neural network model. To facilitate implementation by the ship controller, a recursive least squares model with a forgetting factor is preferred, and the model parameters are updated only when the system is confirmed to be in normal operating condition.

[0052] Normal operating conditions can be confirmed by the following conditions: fuel tank pressure is within the safe pressure window; there are no safety valve activations, no venting actions, and no high-level abnormal alarms; the compressor and BOG disposal equipment are in a recordable stable state; valve status is consistent with planned operating conditions; and the reliability of key sensors is higher than the set reliability threshold. When the above conditions are not met, model parameter updates are paused, and only baseline BOG generation rate prediction is performed.

[0053] Example 4: Residual, Equivalent Heat Intrusion, and Anomaly Candidate Judgment

[0054] Reference Appendix Figure 3 The smoothed measured BOG generation rate is compared with the baseline BOG generation rate to obtain the BOG generation rate residual:

[0055] ;

[0056] in, For the first BOG generation rate residual at each sampling time and The meaning is the same as in the aforementioned formula.

[0057] To ensure comparability of residuals under different liquid levels and temperature conditions, the BOG generation rate residuals are standardized into residual exponents:

[0058] ;

[0059] in, For the first The residual exponent at each sampling time. This represents the standard deviation of the normal BOG generation rate fluctuation under current operating conditions. To prevent small corrections where the denominator is zero, The meaning is the same as in the aforementioned formula.

[0060] Convert the BOG generation rate residual into equivalent heat intrusion:

[0061] ;

[0062] in, For the first The equivalent heat intrusion amount obtained by residual inversion at each sampling time point For BOG generation rate residuals, This refers to the latent heat of vaporization of LNG.

[0063] Within the sliding calculation window, the equivalent cumulative amount of heat intrusion can be calculated using the following formula:

[0064] ;

[0065] in, For the first Each sampling time corresponds to the equivalent cumulative amount of heat intrusion within the sliding calculation window. For the first The equivalent heat intrusion amount obtained by residual inversion at each sampling time point , and The meaning is the same as in the aforementioned formula.

[0066] When the residual index is positive and exceeds the set time continuously, and the equivalent cumulative heat intrusion is higher than the reference threshold, the system enters the heat intrusion anomaly candidate state. If the residual index rises briefly but quickly falls back, the system treats it as a transient disturbance and does not directly trigger high-level handling.

[0067] Example 5: Residual Inversion Feature Extraction

[0068] Reference Appendix Figure 3 Upon entering the candidate state of thermal intrusion anomaly, the monitoring controller extracts residual inversion features. These features include at least the residual amplitude, residual duration, residual slope, pressure recovery time constant, valve status correlation, navigation status correlation, and sensor consistency. The residual slope can be calculated using the following formula:

[0069] ;

[0070] in, For the first The residual slope at each sampling time point For the first The residual exponent at each sampling time. The residual exponent at the start of the sliding calculation window. and The meaning is the same as in the aforementioned formula.

[0071] The pressure recovery time constant is obtained through the pressure decay process after the BOG (Bottle-Off Gas) absorption device operates. The pressure decay process can be characterized by the following formula:

[0072] ;

[0073] in, Time elapsed after the BOG disposal equipment operates fuel tank pressure at that time This is the stable value for pressure decay. Pressure at the initial moment of the action. is the base of the natural exponential function. is the pressure recovery time constant.

[0074] When the pressure recovery time constant gradually increases under the same BOG absorption capacity, it indicates that the fuel tank pressure is falling more slowly, which may indicate additional heat input or continuous BOG anomaly generation. Valve status correlation can be obtained by the synchronicity between residual changes and valve opening / closing status, pressure difference before and after the valve, and temperature changes after the valve. Navigation status correlation can be obtained by the synchronicity between residual changes and changes in roll, pitch, acceleration, speed, or specific liquid level ranges.

[0075] Example 6: Classification of Thermal Intrusion Anomalies

[0076] Reference Appendix Figure 3 The monitoring controller inputs the residual inversion features into the anomaly classification rule base or anomaly classification model, and outputs the anomaly category and anomaly confidence level for heat intrusion. The set of anomaly categories can be represented as:

[0077] ;

[0078] in, For the set of anomaly categories, Abnormal insulation degradation This is due to localized thermal bridging anomaly. This indicates an abnormal internal leakage in the valve. This indicates abnormal shaking and abnormal heat transfer. This indicates sensor drift anomaly.

[0079] Anomaly confidence can be calculated using the following formula:

[0080] ;

[0081] in, For the first The anomaly confidence of the anomaly class. This is the anomaly category number. For the current residual inversion characteristics and the first Distance scoring between anomaly feature templates Sum the sequence numbers of the exception categories. For the first Distance score corresponding to the anomaly feature template.

[0082] When the residual is continuously positively skewed, the residual slope increases slowly, it has low correlation with valve action, low correlation with short-term sea state changes, and the pressure recovery time constant gradually increases, it is identified as an adiabatic deterioration anomaly. When the positive skew of the residual increases in a specific liquid level range or under specific external temperature difference conditions, and is associated with anomalies in local temperature measurement points of the hull, it is identified as a local thermal bridge anomaly.

[0083] When a sudden increase in residual error is synchronized with a specific valve differential pressure, downstream temperature, or valve opening / closing status, and the residual error decreases after valve isolation, it is identified as an abnormal valve internal leakage. When the residual error is synchronized with changes in roll, pitch, acceleration, or speed, and decreases after the navigation status stabilizes, it is identified as an abnormal sloshing heat transfer anomaly. When the deviation between a single pressure, temperature, level, or flow measurement point and redundant measurement points increases and cannot be explained by thermal balance, valve status, or navigation status, it is identified as a sensor drift anomaly.

[0084] If no anomaly category exceeding the confidence threshold is found, the monitoring controller outputs an undefined thermal intrusion anomaly and increases the sampling frequency and alarm level. The confidence threshold can be set according to the ship's safety level, preferably between 0.60 and 0.90.

[0085] Example 7: Tiered Disposal Control

[0086] Reference Appendix Figure 4 The monitoring controller generates handling instructions based on the anomaly type, anomaly confidence level, equivalent heat intrusion amount, and current BOG absorption capacity. To achieve tiered handling, the anomaly severity can be calculated:

[0087] ;

[0088] in, For the first The severity of the anomaly at each sampling time. For residual exponent weights, For duration weighting, As the confidence level weight, As the cumulative heat intrusion weight, This is the normalized value for the residual duration. For the first The maximum anomaly confidence at each sampling time point For reference, cumulative heat intrusion amount, The meaning is the same as in the aforementioned formula.

[0089] When the anomaly category is insulation deterioration or local thermal bridging, the monitoring controller lowers the BOG compressor start-up pressure threshold, raises the BOG compressor minimum operating frequency, increases the BOG receiving flow rate within the allowable range of the BOG absorption equipment, reconstructs the fuel tank pressure setting window, and outputs insulation layer check or local thermal bridging troubleshooting prompts.

[0090] When the anomaly is classified as valve internal leakage, the monitoring controller outputs a corresponding valve isolation command to restrict unnecessary valve actions, verifies changes in pressure difference and temperature before and after the valve, and switches the BOG absorption control to conservative mode. When the anomaly is classified as abnormal sloshing heat exchange, the monitoring controller starts the BOG compressor in advance or appropriately increases the receiving flow of the BOG absorption equipment and extends the minimum compressor operating time to avoid frequent compressor starts and stops due to short-term residual fluctuations.

[0091] When the anomaly category is sensor drift, the monitoring controller switches to redundant sensors, reduces the weight of the abnormal sensor, outputs a sensor verification prompt, and prohibits triggering high-level interlocks solely based on the abnormal sensor. The above handling instructions can be categorized into four levels according to the severity of the anomaly: Level 1 anomalies involve increasing the sampling frequency and outputting an alarm prompt; Level 2 anomalies involve prematurely starting the BOG compressor or lowering the BOG compressor shutdown threshold; Level 3 anomalies involve reconfiguring the fuel tank pressure setting window, restricting unnecessary valve actions, and increasing the BOG dissipation equipment's receiving flow rate; Level 4 anomalies involve triggering the fuel tank safety interlock, valve isolation, and manual confirmation.

[0092] Example 8: Safety Constraints and Post-Disposal Verification

[0093] Reference Appendix Figure 4 All handling commands must meet the constraints of fuel tank safety pressure, compressor capacity, BOG disposal equipment capacity, and valve actuation. The target control pressure of the fuel tank must meet the following constraints:

[0094] ;

[0095] in, To provide the minimum allowable control pressure for the fuel tank, For the first The target control pressure of the fuel tank at each sampling time. The maximum control pressure allowed for the fuel tank.

[0096] The target flow rate of the BOG compressor must meet the following constraints:

[0097] ;

[0098] in, For the first BOG compressor target flow rate at each sampling time The maximum allowable flow rate for the BOG compressor.

[0099] The target received flow of the BOG absorption device must meet the following constraints:

[0100] ;

[0101] in, For the first The target received flow rate of the BOG absorption device at each sampling time. For the first The maximum allowable received flow rate of the BOG absorption device at each sampling time.

[0102] When the BOG (Boiler Gas) absorbing unit is a dual-fuel boiler, the maximum allowable flow rate is determined by the boiler combustion stability, steam load, gas valve opening, and the boiler's minimum stable combustion requirements. When the BOG absorbing unit is a dual-fuel generator or main engine gas supply unit, the maximum allowable flow rate is determined by the load demand, gas supply pressure, and combustion stability.

[0103] To improve the reliability of closed-loop control, the monitoring controller continues to track the residual index, equivalent cumulative heat intrusion, and pressure recovery time constant after the disposal command is executed. If the residual index decreases, the growth rate of equivalent cumulative heat intrusion decreases, and the pressure recovery time constant returns to the normal range after disposal, the disposal is confirmed to be effective; if the residual continues to rise, the anomaly level is increased, and a manual confirmation or safety interlock command is output. Through this post-disposal verification process, monitoring, identification, control, and feedback form a complete closed loop.

[0104] Example 9: Engineering Application Case

[0105] In a vessel with a dual-fuel boiler for BOG (Boiled Boiler Gas), BOG generated in the fuel tank is burned in the dual-fuel boiler after passing through a low-pressure compressor and a heater. A monitoring controller collects data on fuel tank pressure, vapor temperature, liquid level, BOG flow rate, compressor frequency, boiler steam load, and valve status, and executes the method of this invention according to a 10-second sampling interval and a 30-minute sliding calculation window.

[0106] When the residual index exceeds the preset threshold for 30 consecutive minutes, and the residuals show low correlation with changes in roll, pitch, and speed, as well as low correlation with valve action, and the pressure recovery time constant is greater than the historical normal value, the monitoring controller determines that there is a risk of adiabatic degradation or localized thermal bridging. Subsequently, the monitoring controller lowers the BOG compressor start-up pressure threshold, increases the BOG receiving flow rate within the allowable range of boiler steam load, and outputs a fuel tank adiabatic status check prompt.

[0107] In another valve internal leakage scenario, the monitoring controller detects a sudden increase in the BOG generation rate residual, which is synchronized with changes in the differential pressure before and after a certain isolation valve and a decrease in the downstream temperature. The monitoring controller outputs a valve isolation and verification command. If the residual index decreases after valve isolation, the anomaly is confirmed to be related to valve internal leakage; if the residual does not decrease, the anomaly classification is re-executed and the alarm is maintained.

[0108] In high sea state navigation scenarios, if the residual index changes synchronously with roll, pitch, and acceleration, and then falls back after the navigation state stabilizes, the monitoring controller identifies it as abnormal sloshing heat transfer. In this case, the monitoring controller starts the BOG compressor in advance or appropriately increases the BOG absorption flow rate and extends the minimum compressor operating time, without directly triggering high-level safety interlocks, thus balancing safety and equipment stability.

[0109] This invention can be applied to the monitoring and control system of ship fuel tanks using LNG as fuel, and can also be applied to LNG bunkering ships, LNG-powered conversion ships, LNG transport auxiliary fuel tanks, and marine engineering equipment with independent cryogenic fuel tanks.

[0110] The data required for this invention mainly comes from existing shipboard sensors, BOG pipeline instruments, compressor controllers, BOG disposal equipment controllers, and shipboard integrated automation systems, exhibiting good engineering adaptability. By linking the BOG generation rate residual inversion results with equipment control commands, this invention can improve the stability of fuel tank pressure control, reduce the risk of abnormal BOG venting, decrease frequent compressor start-ups and shutdowns, and enhance the early identification and handling capabilities for thermal intrusion anomalies.

Claims

1. A method for identifying and handling abnormal thermal intrusion in ship LNG fuel tanks based on BOG generation rate residual inversion, characterized in that, Includes the following steps: S1. Collect operational data from the ship's LNG fuel tank. The operational data includes at least the fuel tank pressure, fuel tank gas phase temperature, fuel tank liquid phase temperature, liquid level, BOG venting flow rate, BOG compressor operating status, BOG disposal equipment operating status, fuel tank valve status, and ship navigation environment data. S2. Based on the fuel tank pressure, fuel tank gas phase temperature, liquid level and fuel tank geometric volume, establish the mass balance of the fuel tank gas phase space and calculate the measured BOG formation rate. S3. Calculate the baseline BOG generation rate based on normal fuel tank operation samples, environmental heat exchange parameters, liquid level, navigation environment data, and equipment operating status; S4. Generate a BOG generation rate residual based on the difference between the measured BOG generation rate and the benchmark BOG generation rate, and convert the BOG generation rate residual into an equivalent heat intrusion amount. S5. Based on the amplitude, slope, duration, pressure recovery time constant, valve status correlation and navigation status correlation of the BOG generation rate residual, the fuel tank thermal intrusion anomaly category is identified by inversion, and the anomaly confidence level is determined. The thermal intrusion anomaly category includes at least one of the following: adiabatic degradation, local thermal bridge, valve internal leakage, abnormal sloshing heat transfer and sensor drift. S6. Based on the heat intrusion anomaly category, anomaly confidence level and equivalent heat intrusion amount, generate at least one of the following as a disposal instruction: fuel tank pressure setting window adjustment instruction, BOG compressor start / stop or frequency conversion adjustment instruction, BOG absorption flow adjustment instruction, valve isolation instruction and alarm instruction. S7. Send the disposal command to the ship's fuel tank monitoring controller or integrated automation system, and continue to collect operating data after the disposal command is executed. Verify the disposal effect based on the changes in BOG generation rate residual, equivalent heat intrusion amount and pressure recovery time constant, so that the fuel tank pressure, BOG absorption amount and alarm status form a closed loop control.

2. The method for identifying and handling abnormal thermal intrusion in ship LNG fuel tanks based on BOG generation rate residual inversion according to claim 1, characterized in that, In step S2, the measured BOG generation rate is obtained by balancing the change in gas phase mass of the fuel tank with the amount of gas phase discharge and replenishment from the fuel tank, and the change in gas phase space caused by the change in fuel tank liquid level is compensated.

3. The method for identifying and handling abnormal thermal intrusion in ship LNG fuel tanks based on BOG generation rate residual inversion according to claim 1, characterized in that, In step S3, the baseline BOG generation rate is obtained from the normal thermal intrusion model. The normal thermal intrusion model includes a statistical model based on normal samples of ship trials or operations, a physical model based on the heat transfer of the fuel tank enclosure structure, or a hybrid model obtained by fusing the statistical model and the physical model.

4. The method for identifying and handling abnormal thermal intrusion in ship LNG fuel tanks based on BOG generation rate residual inversion according to claim 1, characterized in that, In step S4, the equivalent heat intrusion amount is obtained by multiplying the BOG generation rate residual with the latent heat of LNG vaporization, and the equivalent heat intrusion amount is accumulated through a sliding time window to distinguish between transient disturbances and persistent heat intrusion anomalies.

5. The method for identifying and handling abnormal thermal intrusion in ship LNG fuel tanks based on BOG generation rate residual inversion according to claim 1, characterized in that, In step S5, when the BOG generation rate residual is continuously positively biased, has low correlation with navigation status, low correlation with valve opening and closing status, and the pressure recovery time constant gradually increases, the abnormal category is identified as adiabatic deterioration or local thermal bridging.

6. The method for identifying and handling abnormal thermal intrusion in ship LNG fuel tanks based on BOG generation rate residual inversion according to claim 1, characterized in that, In step S5, when the BOG generation rate residual has a synchronous correlation with the opening / closing state, differential pressure state, or downstream temperature change of the specified valve, the abnormality category is identified as internal leakage of the valve, and an isolation or verification command corresponding to the specified valve is output.

7. The method for identifying and handling abnormal thermal intrusion in ship LNG fuel tanks based on BOG generation rate residual inversion according to claim 1, characterized in that, In step S5, when the BOG generation rate residual has a synchronous correlation with the ship's roll, pitch, acceleration, speed changes or liquid level range, and the residual falls back after the navigation state stabilizes, the abnormal category is identified as abnormal sloshing heat transfer.

8. The method for identifying and handling abnormal thermal intrusion in ship LNG fuel tanks based on BOG generation rate residual inversion according to claim 1, characterized in that, In step S6, the handling instructions are output according to the level of abnormality: Level 1 abnormality is to increase the sampling frequency and output a prompt alarm; Level 2 abnormality is to start the BOG compressor in advance or lower the BOG compressor shutdown threshold; Level 3 abnormality is to reconstruct the fuel tank pressure setting window, restrict unnecessary valve actions, and increase the receiving flow of the BOG disposal equipment; Level 4 abnormality is to trigger the fuel tank safety interlock, valve isolation, and manual confirmation.

9. A system for identifying and handling abnormal thermal intrusion in ship LNG fuel tanks based on BOG generation rate residual inversion, characterized in that, include: The data acquisition module is used to collect data on fuel tank pressure, temperature, liquid level, BOG flow rate, equipment status, valve status, and navigation environment. The system includes a BOG generation rate calculation module for calculating the measured BOG generation rate based on the fuel tank gas phase mass balance; a baseline generation rate prediction module for calculating the baseline BOG generation rate based on a normal thermal intrusion model; a residual inversion module for inverting the thermal intrusion anomaly category and determining the anomaly confidence level based on the residual between the measured BOG generation rate and the baseline BOG generation rate; and a handling control module for outputting at least one control command among pressure setting window adjustment, BOG compressor adjustment, BOG absorption flow rate adjustment, valve isolation, and alarm based on the thermal intrusion anomaly category and anomaly confidence level. The system is configured to execute the method for identifying and handling thermal intrusion anomalies in ship LNG fuel tanks based on BOG generation rate residual inversion as described in any one of claims 1 to 8.

Images