An unmanned ship oil spill monitoring system and method

Abstract

The invention discloses an unmanned ship oil spill monitoring system and method, comprising an information acquisition module for collecting parameter data of various sensors, an oil storage module for storing and transporting fuel oil of the ship, the oil storage module comprising an oil bunker and an oil pipe, the oil bunker and the oil pipe being connected to the information acquisition module through a liquid level sensor and a flow sensor, so as to monitor the storage and flow conditions of the fuel oil in real time; a ventilation module for controlling the gas pressure and gas flow in the oil bunker, the ventilation module being connected to the information acquisition module through a pressure sensor, so as to monitor the gas pressure change in the oil bunker in real time; a communication module for performing data communication with a data processing center; the data processing center being used to receive the data collected by the information acquisition module and the signal transmitted by the communication module, and performing fault analysis and oil spill monitoring on the oil storage module and the ventilation module in real time through an onboard algorithm program; and an alarm module for sending an alarm signal when an oil spill is detected.

Description

Technical Field

[0001] The present invention relates to the technical field of light oil spill monitoring for unmanned ships, and particularly to an unmanned ship oil spill monitoring system and method. Background Art

[0002] At present, the construction of the oil spill emergency system is mainly to deal with large-area oil spill incidents caused by ship accidents, and it cannot cope with the phenomenon of oil leakage during ship fuel supply operations (the amount of oil spilled due to oil leakage is small and the operating unit can often clean it up in time). Due to the characteristics of high frequency, unfixed location (ports or anchorages), and small amount of oil spill in ship fuel supply operations, some operating units do not pay attention to the prevention and cleaning of oil leakage, resulting in the existence of small amounts of oil pollution belts of unknown origin in ports or on the sea surface.

[0003] Ship fuel supply operation is a key operation of the ship's engine department, which is related to the safety of ship operation and the safety of the marine environment. During the fuel supply process, the joints of oil pipes, the ventilation holes of fuel tanks, and the pipe sealing plates of the fuel filling stations on the opposite side are key parts prone to oil spill. At present, in addition to measuring the oil level in the receiving tank, it mainly relies on arranging crew members to inspect each oil pipe joint, ventilation hole, and fuel filling station on the opposite side during the fuel supply process to check whether there is oil spill. This method is not applicable to unmanned ships, and the manual inspection method is prone to errors and omissions, resulting in the inability to detect oil spill incidents in time, thus endangering the marine ecological environment.

[0004] CN112229584A discloses a method and detection device for monitoring oil spill during ship fuel supply operation. It judges whether the light in the monitoring environment is sufficient through an illuminance sensor, and judges whether to collect monitoring data through an image acquisition device or an ultraviolet fluorescence oil pollution monitoring sensor according to the light condition, processes and analyzes the monitoring data, and judges whether there are oil flowers on the water surface to monitor whether there is oil spill during ship fuel supply operation; this method uses the illuminance sensor and light condition for image recognition, and discriminates based on the spectral bands of oil concentration and water concentration, but it is only applicable to ships using heavy oil. If light oil is used, it is not easy to distinguish.

[0005] CN106154263A discloses a radar signal processing device and monitoring method for ship oil spill detection. It simultaneously emits radar waves through two transceivers with symmetrical structures, and through preset parameters, designates any one of them to detect ship targets and the other to detect oil spills to achieve ship monitoring; this method monitors oil spills during ship traffic through radar signals and is applicable to large-scale, ocean-going heavy ships. Existing unmanned ships are smaller in size and are usually used in ports or coastal waters, and are not suitable for carrying this transceiver. Summary of the Invention

[0006] The purpose of this section is to outline some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Simplifications or omissions may be made in this section, as well as in the abstract and title of the present application, to avoid obscuring the purpose of this section, the abstract, and the title. However, such simplifications or omissions shall not be used to limit the scope of the present invention.

[0007] In view of the above existing problems, the present invention is proposed.

[0008] Therefore, the technical problem solved by the present invention is: how to quickly, accurately, effectively, and timely monitor oil spill events during the fuel supply operation of an unmanned ship to avoid polluting the marine environment.

[0009] To solve the above technical problem, the present invention provides the following technical solutions: An information acquisition module for collecting parameter data of each sensor. The information acquisition module includes a pressure sensor, a temperature sensor, a flow sensor, and a liquid level sensor, where at least two or more of the pressure sensor, the temperature sensor, the flow sensor, and the liquid level sensor are provided;

[0010] A fuel storage module for storing and transporting the fuel of the ship. The fuel storage module includes an oil tank and an oil pipe. The oil tank and the oil pipe are connected to the information acquisition module through the liquid level sensor and the flow sensor to monitor the storage and flow conditions of the fuel in real time;

[0011] A ventilation module for controlling the gas pressure and gas flow in the oil tank. The ventilation module is connected to the information acquisition module through the pressure sensor to monitor the change of gas pressure in the oil tank in real time;

[0012] A communication module for data communication with a data processing center;

[0013] The data processing center is configured to receive the data collected by the information acquisition module and the signals transmitted by the communication module, and perform fault analysis and oil spill monitoring on the fuel storage module and the ventilation module in real time through an algorithm program carried thereon;

[0014] An alarm module for emitting an alarm signal when an oil spill is detected.

[0015] As a preferred solution of the oil spill monitoring system for an unmanned ship according to the present invention, the communication module adopts wireless communication technology to achieve real-time data transmission and remote control.

[0016] As a preferred solution of the oil spill monitoring system for an unmanned ship according to the present invention, the alarm module adopts an audible and visual alarm and information notification method to timely remind relevant personnel to take emergency measures.

[0017] As a preferred solution of the unmanned ship oil spill monitoring system described in the present invention, the pressure sensor, the liquid level sensor and the flow sensor are arranged in the fuel tank to monitor whether the pressure value of the fuel in the fuel tank is abnormal and whether the fuel flow is abnormal.

[0018] As a preferred solution of the unmanned ship oil spill monitoring system described in the present invention, the temperature sensor and the flow sensor are arranged at each node of the oil pipe to monitor in real time whether an oil spill event occurs at each node of the oil pipe or whether the fuel flow in the oil pipe is abnormal.

[0019] As a preferred solution of the unmanned ship oil spill monitoring method described in the present invention, the method includes the following steps:

[0020] Collect the historical operation status data in the ship oil spill monitoring system, extract the parameter information containing oil spill faults therein and perform data preprocessing;

[0021] Establish a fault discrimination model based on the neural network algorithm, and substitute the preprocessed parameter information into the fault discrimination model for learning and training;

[0022] Deploy the trained fault discrimination model to the ship oil spill monitoring system, start running, and determine the oil spill fault according to the sensor data collected in real time:

[0023] If the vector value output by the fault discrimination model ≥ the fault threshold set by the monitoring system, it is determined that an oil spill fault has occurred, and the oil spill fault signal is pushed to the alarm module for sound and light alarm and information notification to timely remind relevant personnel to take emergency measures;

[0024] If the vector value output by the fault discrimination model < the fault threshold set by the monitoring system, it is determined that no oil spill fault has occurred, and the signal of no oil spill fault is pushed to the data processing center for real-time monitoring of the data of each sensor.

[0025] As a preferred solution of the unmanned ship oil spill monitoring method described in the present invention, the collected historical operation status data includes ship type, ship line drawing, ship scale, ship main scale ratio, ship coefficient, and ship unit status parameters;

[0026] Extract the parameter information containing the oil spill fault, including engine speed, generator coil temperature, front and rear bearing temperatures, hydraulic system oil temperature, oil pressure, oil level, and the pressure, temperature, liquid level, flow rate of the oil pipe and fuel tank, and the pressure and temperature of the ventilation device.

[0027] As a preferred solution of the oil spill monitoring method for unmanned ships described in the present invention, establishing the fault discrimination model and conducting learning and training includes:

[0028] Based on the input layer of the neural network and several classes, establish a combined network for network classification;

[0029] Define the class number for the oil spill fault as 0, and the class number for the non - oil spill fault as 1;

[0030] Establish the fault discrimination model with the multi - classification logarithmic loss function as the objective function, and its mathematical expression formula is as follows:

[0031]

[0032] Among them, μ is the weight coefficient obtained from the average variance of sensor parameters, n is the number of input data for acquisition, m is the number of classes to be classified, and y ij is whether the input x i is accurately classified, that is, whether j is the true class of x i and p ij is the output of the network classifier for the input x i That is, the probability value of belonging to class j.

[0033] As a preferred solution of the oil spill monitoring method for unmanned ships described in the present invention, the learning and training includes:

[0034] Take the parameter data during historical oil spill faults as single - sample data, and mark the sample data with multi - class labels;

[0035] Randomly select 70% of the sample data from all the multi - class labels as the training set, and the remaining 30% of the sample data as the verification set;

[0036] Use the training set to train the fault discrimination model, and use the verification set to verify the model discrimination accuracy of the fault discrimination model until the output classification quantity meets the classification target quantity and then stop training.

[0037] Advantages of the present invention:

[0038] 1. The present invention realizes accurate monitoring and intelligent emergency measures by using the information acquisition module to monitor the parameter data of the ship's oil tank pipeline in real - time and using the data processing center for data analysis and processing, improving the response efficiency to oil spill accidents;

[0039] 2. The oil spill monitoring system for unmanned ships of the present invention has the advantages of high efficiency and accuracy, real - time communication, intelligent emergency measures, and automated operation, and is applicable to the field of marine environmental protection;

[0040] 3. By calculating the parameters of the oil tank pipeline sensor and combining with the fault classification model, the present invention determines whether there is an oil spill fault in the ship's oil storage module, and combines with the alarm module to give an alarm in time, improving the emergency handling efficiency of oil spill incidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings. Among them:

[0042] Figure 1 It is a schematic flow chart of a method for monitoring oil spills on an unmanned ship shown in the present invention;

[0043] Figure 2 It is a schematic diagram of the network topology structure distribution of a system for monitoring oil spills on an unmanned ship shown in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] In order to make the above objects, features and advantages of the present invention more obvious and understandable, the specific embodiments of the present invention will be described in detail below with reference to the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all embodiments.

[0045] Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0046] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0047] Embodiment 1

[0048] In maritime practice, due to human factors such as the negligence of operators, night fatigue operations, and judgment errors, the situation of inadequate inspection and implementation often occurs. During the ship refueling operations at home and abroad, fuel leakage accidents often occur due to the influence of the above various factors, causing huge marine ecological environmental pollution and property losses.

[0049] Common ship oil spills are divided into operational oil spills and accidental oil spills. Operational oil spills include cases where crew members do not comply with relevant regulations and illegally discharge bilge water, waste oil, used engine oil, etc., or due to work mistakes during oil loading and unloading, such as misopening valves or flange connections falling off, overfilling during refueling causing overflow outside the tank or oil pipelines bursting, etc. Accidental oil spills include sudden oil spill accidents caused by unexpected accidents such as ship collisions, groundings, reef strikes, fires, and explosions, resulting in a large amount of cargo oil or fuel leakage.

[0050] The harms caused by oil spill accidents to the ecological environment are as follows: First, due to the covering and asphyxiating effect of the oil film and its own oxygen consumption, in severe cases, a large number of organisms in the polluted water area will die due to lack of oxygen; Second, harmful substances in the oil will affect fish, shellfish, and humans through the food chain; Third, once various organic hydrocarbons in oil pollution are absorbed by organisms, their properties will become very unstable, cycle in the food chain, are not easily decomposed, and reach a toxic level through concentration and enrichment. Therefore, ecological monitoring must always be carried out during the monitoring of oil spill accidents, and monitoring cannot be stopped just because the floating oil or the oil concentration in the water returns to normal.

[0051] According to an embodiment of the present invention, in combination with Figure 1 the flowchart shown, a method for monitoring oil spills on unmanned ships specifically includes the following steps:

[0052] S1: Collect historical operation status data in the ship oil spill monitoring system, extract parameter information containing oil spill faults, and perform data preprocessing. Among them, it should be noted that:

[0053] The collected historical operation status data includes ship type, ship hull form diagram, ship dimensions, ship main dimension ratio, ship coefficient, and ship unit status parameters;

[0054] The extracted parameter information containing oil spill faults includes engine speed, generator coil temperature, front and rear bearing temperatures, hydraulic system oil temperature, oil pressure, oil level, and the pressure, temperature, liquid level, flow rate of oil pipelines and oil tanks, and the pressure and temperature of the ventilation module.

[0055] Furthermore, data preprocessing includes:

[0056] Data cleaning: cleaning of format content, non - required data cleaning, cleaning of missing values, and cleaning of logical errors;

[0057] Data transformation: performing feature construction, data grading, and data quantization on the data;

[0058] Data integration: performing data statistics on the transformed data and merging the data into a unified data storage;

[0059] As an example, the transformation of data can be processed in a standardized and normalized manner, which is not limited in the embodiments of the present invention.

[0060] It should be noted that the collected historical operation status data of the ship contains duplicate and useless interference information, which is screened by means of data preprocessing and transformed into unified numerical parameters to facilitate the efficiency and accuracy of later model calculations.

[0061] In an alternative embodiment, the extraction technology of key features is used to extract the parameter information containing oil spill faults to avoid incorrect parameter extraction.

[0062] As an alternative embodiment, the sensor parameters of the oil pipe, oil tank, ventilation module and hydraulic system are used as the main key features to streamline the calculation amount.

[0063] S2: Establish a fault discrimination model based on the neural network algorithm, and substitute the preprocessed parameter information into the fault discrimination model for learning and training. In this step, it should be noted that establishing a fault discrimination model and performing learning and training includes:

[0064] Establish a combined network for network classification based on the input layer and number of classes of the neural network;

[0065] Define the number of classes for the oil spill fault as 0 and the number of classes for the non-oil spill fault as 1;

[0066] Establish a fault discrimination model with the multi-class logarithmic loss function as the objective function, and its mathematical expression formula is as follows:

[0067]

[0068] Among them, μ is the weight coefficient obtained from the average variance of the sensor parameters, n is the number of input data collected, m is the number of classes to be classified, and y ij is whether the input x i is accurately classified, that is, whether j is the true class of x i , and p ij is the output of the network classifier for the input x i , that is, the probability value belonging to class j.

[0069] Furthermore, the learning and training include:

[0070] Take the parameter data during historical oil spill faults as single sample data and mark the sample data with multi-class labels;

[0071] Randomly select 70% of the sample data from all the multi-class labels as the training set, and the remaining 30% of the sample data as the validation set;

[0072] Train a fault discrimination model using the training set, and use the validation set to verify the model discrimination accuracy of the fault discrimination model until the training stops when the number of output classifications meets the classification target number.

[0073] Furthermore, use the validation set to verify the correct rate of the fault discrimination model, and its mathematical expression formula is as follows:

[0074]

[0075] where n true,j : the number of samples of the jth class that are accurately classified, τ: the number of classes, accuracy: the accuracy rate.

[0076] S3: Deploy the trained fault discrimination model to the ship oil spill monitoring system, start running, and determine the oil spill fault according to the sensor data collected in real time:

[0077] If the vector value output by the fault discrimination model ≥ the fault threshold set by the monitoring system, it is determined that an oil spill fault has occurred, and the oil spill fault signal is pushed to the alarm module for sound and light alarm and information notification to timely remind relevant personnel to take emergency measures;

[0078] If the vector value output by the fault discrimination model < the fault threshold set by the monitoring system, it is determined that no oil spill fault has occurred, and the signal of no oil spill fault is pushed to the data processing center for real-time monitoring of the data of each sensor.

[0079] It should be noted that the fault threshold set by the monitoring system can be set according to the actual situation of the unmanned ship.

[0080] The data preprocessing and fault identification calculation methods of the foregoing parameters can be carried out by means and methods in the prior art, and will not be elaborated in this example.

[0081] Referring to Figure 2 , the embodiment of the present invention also discloses an unmanned ship oil spill monitoring system, including an information collection module, an oil storage module, a ventilation module, a communication module, a data processing center and an alarm module, wherein:

[0082] The information collection module is used to collect the parameter data of each sensor. The information collection module includes a pressure sensor, a temperature sensor, a flow sensor and a liquid level sensor, wherein at least two or more of the pressure sensor, the temperature sensor, the flow sensor and the liquid level sensor.

[0083] The oil storage module is used to store and transport the fuel of the ship. The oil storage module includes an oil tank and an oil pipe. The oil tank and the oil pipe are connected to the information collection module through a liquid level sensor and a flow sensor to monitor the storage and flow conditions of the fuel in real time.

[0084] A breather module, which is used to control the gas pressure and gas flow in the oil tank. The breather module is connected to the information acquisition module through a pressure sensor to monitor the change of gas pressure in the oil tank in real time.

[0085] A communication module, which is used to conduct data communication with the data processing center.

[0086] A data processing center, which is used to receive the data collected by the information acquisition module and the signals transmitted by the communication module, and conduct fault analysis and oil spill monitoring on the oil storage module and the breather module in real time through the loaded algorithm program.

[0087] An alarm module, which is used to send an alarm signal when an oil spill is detected.

[0088] As an example, the communication module adopts wireless communication technology to achieve real-time data transmission and remote control.

[0089] As an example, the alarm module adopts an audible and visual alarm and an information notification method to timely remind relevant personnel to take emergency measures.

[0090] As an example, a pressure sensor, a liquid level sensor and a flow sensor are arranged in the oil tank to monitor whether the pressure value of the fuel in the oil tank is abnormal and whether the fuel flow is abnormal.

[0091] As an example, a temperature sensor and a flow sensor are arranged at each node of the oil pipe to monitor in real time whether an oil spill event occurs at each node of the oil pipe or whether the fuel flow in the oil pipe is abnormal.

[0092] It should be noted that the pressure sensor, temperature sensor, flow sensor and liquid level sensor are installed at key parts of the ship's oil tank oil pipe (such as the oil pipe node) to ensure that various parameter data can be accurately collected. The data collected by the information acquisition module is transmitted to the data processing center through the communication module to achieve real-time monitoring and data sharing. Then, the artificial intelligence algorithm of the data processing center (such as the fault discrimination model designed in the foregoing method) is used to process and analyze the collected data to determine whether there is an oil spill and determine emergency measures.

[0093] Furthermore, when an oil spill is detected, an alarm signal is sent through the alarm module to remind relevant personnel to take corresponding measures.

[0094] As an example, according to the detected oil spill situation, necessary emergency measures are taken, such as closing the oil tank valve and starting the pump to recover the oil spill.

[0095] It should be noted that the unmanned ship oil spill monitoring system disclosed in the embodiment of the present invention further includes one or more processors and a memory.

[0096] The memory is used to store instructions that can be operated. When these instructions are executed by the one or more processors, they cause the one or more processors to perform operations, including the processes of the unmanned ship oil spill monitoring method in the foregoing embodiments, especially Figure 1 the process of the method shown.

[0097] It should be recognized that the embodiments of the present invention can be implemented or carried out by computer hardware, a combination of hardware and software, or by computer instructions stored in a non-transitory computer-readable memory. The method can use standard programming techniques, which include being implemented in a non-transitory computer-readable storage medium configured with a computer program in the computer program, wherein the storage medium so configured causes the computer to operate in a specific and predefined manner. Each program can be implemented in a high-level procedural or object-oriented programming language to communicate with the computer system.

[0098] The processes described in the embodiments of the present invention can be executed under the control of one or more computer systems configured with executable instructions, and can be implemented by hardware or a combination thereof as code (e.g., executable instructions, one or more computer programs, or one or more applications) that are commonly executed on one or more processors.

[0099] The computer program includes a plurality of instructions executable by one or more processors.

[0100] Furthermore, the method can be implemented in any type of computing platform that is operably connected and suitable, including but not limited to personal computers, minicomputers, mainframes, workstations, network or distributed computing environments, separate or integrated computer platforms, or communicating with charged particle tools or other imaging devices.

[0101] Aspects of the present invention can be implemented in machine-readable code stored on a non-transitory storage medium or device, whether removable or integrated into the computing platform, such as a hard disk, optical read and / or write storage medium, RAM, ROM, etc., such that it can be read by a programmable computer. When the storage medium or device is read by the computer, it can be used to configure and operate the computer to execute the processes described herein.

[0102] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. An unmanned ship oil spill monitoring system, characterized in that: include: An information acquisition module, used to collect parameter data of each sensor, the information acquisition module includes a pressure sensor, a temperature sensor, a flow sensor and a liquid level sensor, wherein there are at least two or more of the pressure sensor, the temperature sensor, the flow sensor and the liquid level sensor; The oil storage module is used to store and transport fuel for the ship. The oil storage module includes an oil tank and an oil pipe. The oil tank and the oil pipe are connected to the information acquisition module through the liquid level sensor and the flow sensor to monitor the storage and flow of the fuel in real time. A ventilation module, used to control the gas pressure and gas flow in the oil bunker, the ventilation module is connected to the information acquisition module through the pressure sensor to monitor the gas pressure change in the oil bunker in real time; A communication module, used for data communication with a data processing center; The data processing center is used to receive the data collected by the information collection module and the signal transmitted by the communication module, and to perform fault analysis and oil spill monitoring on the oil storage module and the ventilation module in real time through the algorithm program installed; An alarm module is used to send out an alarm signal when an oil spill is detected; The unmanned ship oil spill monitoring system also includes an unmanned ship oil spill monitoring method, the method comprising the following steps: Collect historical operating status data from the ship oil spill monitoring system, extract parameter information containing oil spill faults and perform data preprocessing; Establishing a fault discrimination model based on a neural network algorithm, substituting the pre-processed parameter information into the fault discrimination model for learning and training; The trained fault discrimination model is deployed to the ship oil spill monitoring system, and the system is started to run, and the oil spill fault is judged according to the real-time collected sensor data: If the vector value output by the fault discrimination model is ≥ the fault threshold set by the monitoring system, it is determined that an oil spill fault has occurred, and the oil spill fault signal is pushed to the alarm module for sound and light alarm and information notification, so as to promptly remind relevant personnel to take emergency measures; If the vector value output by the fault discrimination model is less than the fault threshold set by the monitoring system, it is determined that no oil spill fault has occurred, and a signal indicating that no oil spill fault has occurred is pushed to the data processing center for real-time monitoring of the data of each sensor; Establishing the fault discrimination model and conducting learning and training includes: Building a network classification network based on the input layer and number of neural networks; The number of categories defined as oil spill faults is 0, and the number of categories defined as no oil spill faults is 1; The fault discrimination model with multi-classification logarithmic loss function as the objective function is established, and its mathematical expression formula is as follows: in, is the weight coefficient obtained by the average variance of sensor data, n is the number of collected data inputs, m is the number of categories to be classified, For input Whether the classification is accurate, that is, whether j is The real category, For the network classifier to input The output of is the probability value of belonging to category j.

2. The unmanned ship oil spill monitoring system according to claim 1, characterized in that: The communication module adopts wireless communication technology to achieve real-time data transmission and remote control.

3. The unmanned ship oil spill monitoring system according to claim 1, characterized in that: The alarm module uses an audible and visual alarm and information notification method to promptly remind relevant personnel to take emergency measures.

4. The unmanned ship oil spill monitoring system according to claim 1, characterized in that: The pressure sensor, the liquid level sensor and the flow sensor are arranged in the oil bunker to monitor whether the pressure value of the fuel in the oil bunker is abnormal and whether the fuel flow is abnormal.

5. The unmanned ship oil spill monitoring system according to claim 1, characterized in that: The temperature sensor and the flow sensor are arranged at each node of the oil pipeline to monitor in real time whether an oil spill occurs at each node of the oil pipeline or whether the fuel flow in the oil pipeline is abnormal.

6. The unmanned ship oil spill monitoring system according to claim 1, characterized in that: The historical operating status data collected in the monitoring method include ship type, ship lines diagram, ship scale, ship main scale ratio, ship system coefficient, and ship unit status parameters; Extract parameter information containing the oil overflow fault, including engine speed, generator coil temperature, front and rear bearing temperature, hydraulic system oil temperature, oil pressure, oil level, and pressure, temperature, liquid level, flow rate of the oil pipe and oil tank, and pressure and temperature of the breather.

7. The unmanned ship oil spill monitoring system according to claim 1, characterized in that: The learning training in the monitoring method includes: Taking parameter data of historical oil spill failures as single sample data, and marking the sample data as multi-category labels; Randomly extract 70% of the sample data from all the multi-category labels as a training set, and use the remaining 30% of the sample data as a validation set; The fault discrimination model is trained using the training set, and the fault discrimination model is verified for model discrimination accuracy using the verification set, until the training is stopped when the number of output classifications meets the number of classification targets.

Images