Engine early warning system and method

Abstract

The present invention relates to an engine early warning system and method, comprising a data acquisition module, a data transmission module, a data storage and processing module, a prediction model module, a decision support module, and a user interaction module; the data acquisition module is used to construct a sensor network, and various sensors such as temperature sensors, pressure sensors, vibration sensors, and fuel consumption sensors are deployed at key positions of a ship engine to collect various parameters during the operation of the engine in real time. The present invention monitors the operation data of the ship engine in real time, uses advanced data analysis technology and algorithms, and realizes early warning and predictive maintenance of potential engine failures, which is of great significance for improving the safety of ship operation, reducing maintenance costs, and improving economic benefits.

Description

Technical Field

[0001] The present invention relates to the field of engine early warning, and in particular to an engine early warning system and method. Background Art

[0002] Against the backdrop of the rapid development of the global shipping industry, ship engines, as the core power equipment of ships, have a direct impact on the navigation safety, operational efficiency and economic benefits of ships. However, traditional maintenance methods have many disadvantages, such as high cost, low efficiency and great safety hazards.

[0003] 1. High cost: Regular maintenance requires a lot of manpower, material and financial resources, and in many cases potential faults cannot be discovered in time.

[0004] 2. Low efficiency: Post-fault maintenance is often carried out after a fault has occurred, resulting in long downtime and affecting the normal operation of the ship.

[0005] 3. Great safety hazard: If potential faults are not discovered and handled in time, they may cause more serious faults and even lead to ship accidents.

[0006] Therefore, we propose an engine early warning system and method, which aims to achieve early warning and predictive maintenance of potential engine failures by real-time monitoring of the ship engine's operating data and using advanced data analysis techniques and algorithms. This is of great significance for improving the safety of ship operations, reducing maintenance costs, and improving economic benefits. Summary of the invention

[0007] The purpose of the present invention is to overcome the shortcomings of the prior art, adapt to actual needs, and provide an engine early warning system and method to solve the many disadvantages of the current traditional maintenance methods, such as high cost, low efficiency and great safety hazards. 1. High cost: Regular maintenance requires a lot of manpower, material and financial resources, and in many cases, potential faults cannot be discovered in time. 2. Low efficiency: Post-maintenance is often carried out after the fault has occurred, resulting in long downtime and affecting the normal operation of the ship. 3. Great safety hazards: If potential faults are not discovered and handled in time, they may cause more serious faults and even lead to technical problems such as ship accidents.

[0008] In order to achieve the purpose of the present invention, the technical solution adopted by the present invention is: design an engine warning system, including a data acquisition module, a data transmission module, a data storage and processing module, a prediction model module, a decision support module, and a user interaction module;

[0009] The data acquisition module is used to build a sensor network, deploying various sensors at key parts of the ship engine, including temperature sensors, pressure sensors, vibration sensors, and fuel consumption sensors, to collect various parameters during the operation of the engine in real time;

[0010] The data transmission module is used to provide a communication interface between the sensor network and the data storage and processing module, and supports wired or wireless communication;

[0011] The data storage and processing module has a built-in data storage unit, which is used to store the collected engine operation data by using a high-performance storage device, such as a hard disk, a solid-state drive or a cloud storage;

[0012] The prediction model module is used for model construction. The model construction is based on historical data and domain knowledge to build a prediction model suitable for ship engines. The model is based on statistical methods, machine learning algorithms or deep learning technology;

[0013] The decision support module is used for fault warning and diagnosis. According to the output results of the prediction model, it warns and diagnoses possible engine faults and provides detailed fault information, fault severity and prediction time.

[0014] The user interaction module is used to provide a visual interface, a friendly user interface, display the real-time operating status of the engine, prediction results and maintenance suggestion information, and supports multiple display methods such as charts, curves, and alarm prompts.

[0015] Preferably, a data acquisition unit is provided inside the data acquisition module, and the data acquisition unit is responsible for receiving data from the sensor network and performing preliminary processing and conversion, such as filtering, denoising, and amplification, to ensure the accuracy and reliability of the data.

[0016] Preferably, the data transmission module is provided with a data transmission protocol, and the real-time and security of the data are ensured by formulating standards and protocols for data transmission.

[0017] Preferably, the data storage unit built into the data storage and processing module uses big data analysis and machine learning algorithms to preprocess, extract features, and recognize patterns on the stored data to detect anomalies and potential faults in the data.

[0018] Preferably, the prediction model module is also used for model training and optimization, using a training data set to train the model, and optimizing model performance through cross-validation and parameter tuning methods, while continuously updating and optimizing the model based on real-time data.

[0019] Preferably, the decision support module is also used for maintenance suggestions and optimization, and provides management personnel with targeted maintenance suggestions and optimization plans based on fault warning and diagnosis results. The suggestions include adjusting operating parameters, replacing parts, and shutting down for maintenance.

[0020] Preferably, the user interaction module is also used for interaction, allowing the user to perform operations and settings through the interface, such as adjusting monitoring parameters, viewing historical data, and exporting reports. At the same time, it supports user feedback and opinion collection to continuously improve and optimize the system.

[0021] Preferably, the present invention provides an early warning method for an engine early warning system, the early warning method comprising a research and development stage and an operation stage:

[0022] The development phase includes the following steps:

[0023] S1. Requirements analysis and system design: clarify project requirements, conduct market research and technical analysis; design system architecture and functional modules;

[0024] A. Clarify requirements: First, the specific requirements of the ship engine predictive maintenance system need to be clarified, including the parameters that need to be monitored, the predicted fault types, and the real-time requirements of the system;

[0025] B. System planning: According to the needs, plan the overall architecture and functional modules of the system, including data collection, data transmission, data processing, prediction model construction, and decision support.

[0026] S2. Technology research and development and algorithm optimization: Develop data collection and transmission systems; research and optimize prediction algorithms; establish prediction models and verify them.

[0027] A. Data acquisition and transmission processing system:

[0028] Sensor selection and deployment: According to system requirements, select appropriate sensors to monitor key parameters of ship engines, such as temperature, pressure, vibration, and fuel consumption. Sensors should be deployed in locations that can accurately reflect the operating status of the engine.

[0029] Data collection and transmission: Use IoT technology to collect sensor data in real time and transmit it to the cloud or local server for storage.

[0030] Data preprocessing: Clean, denoise, and filter the collected raw data to improve the quality and accuracy of the data.

[0031] B. Feature extraction and model building

[0032] Feature extraction: Use data analysis and mining techniques to extract useful feature information from preprocessed data. These features should be able to reflect the engine's operating status and potential faults.

[0033] Model construction: Based on the extracted feature information, a prediction model is constructed. The prediction model can use machine learning and deep learning algorithms, such as neural networks, support vector machines, and random forests. Parameter tuning and model verification are required during the model construction process to ensure the accuracy and reliability of the model.

[0034] C. Model training and optimization

[0035] Data partitioning: Divide the preprocessed data into training sets and test sets for model training and verification.

[0036] Model training: Use the training set data to train the model, and adjust the model parameters through an iterative optimization algorithm so that the model can better fit the training data.

[0037] Model validation: Use the test set data to validate the trained model and evaluate the prediction accuracy and generalization ability of the model. Further optimize the model based on the validation results.

[0038] S3. System integration and testing: Integrate various functional modules into a unified platform; conduct system testing to ensure stable operation.

[0039] S4. Pilot application and feedback adjustment: select typical ships for pilot application; collect user feedback to optimize and improve the system.

[0040] System deployment: Deploy the constructed predictive maintenance system to the actual ship and connect and integrate it with the ship's engine.

[0041] System testing: Before actual application, the system is fully tested to ensure that the system can run stably and meet actual needs.

[0042] System maintenance: Regularly maintain and update the system, including data backup, software upgrades, and hardware inspections. At the same time, continuously optimize and improve the system based on user feedback and actual application conditions.

[0043] S5. Promotion, application and subsequent support: Promote the application of predictive maintenance systems in the industry; provide continuous technical support and system upgrade services.

[0044] Preferably, the operation phase comprises the following steps:

[0045] S1. Data collection and preprocessing

[0046] A. Data collection: Using sensors and the Internet of Things to collect real-time operating data of ship engines, including key parameters such as vibration, temperature, pressure, and speed;

[0047] B. Data preprocessing: Clean, denoise and normalize the collected data to improve data quality and provide a reliable data basis for subsequent analysis;

[0048] S2. Data Analysis and Modeling

[0049] A. Data analysis: Use machine learning, statistical modeling, and pattern recognition data analysis methods to conduct in-depth analysis of pre-processed data to identify abnormal patterns and fault characteristics in engine operation;

[0050] B. Model prediction: Based on the data analysis results, an engine fault prediction model is established to predict the future performance and fault possibility of the engine through time series analysis, trend analysis, and anomaly detection technology;

[0051] S3. Fault warning and diagnosis

[0052] A. Fault warning: Set reasonable thresholds and warning mechanisms. When the engine operation data is abnormal, the system can send out warning signals in real time to remind managers to take timely measures;

[0053] B. Fault diagnosis: Use fault diagnosis algorithms to further analyze early warning signals, determine the specific location and cause of the fault, and provide guidance for maintenance work;

[0054] S4. Predictive maintenance plan formulation

[0055] A. Maintenance plan formulation: Based on the failure prediction results and risk assessment, formulate a scientific predictive maintenance plan, including maintenance time, maintenance content, and required spare parts;

[0056] B. Optimize maintenance resources: Through the implementation of predictive maintenance plans, optimize the allocation of maintenance resources for ship engines, reduce maintenance costs and improve maintenance efficiency.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] 1. Ship operation safety: The predictive maintenance system can detect and handle potential engine failures in a timely manner, prevent the failures from expanding and causing ship accidents, and improve the operational safety of ships.

[0059] 2. Reduce maintenance costs: Through predictive maintenance, maintenance can be performed before a failure occurs, avoiding the high cost of subsequent repairs. At the same time, it optimizes the allocation of maintenance resources and reduces unnecessary maintenance investment.

[0060] 3. Improve economic benefits: reduce downtime caused by failures, improve ship operation efficiency, and increase economic benefits. In addition, the application of predictive maintenance systems can also improve the reliability and service life of ship engines, further reducing long-term operating costs.

[0061] 4. Promote the development of intelligent ships: Predictive maintenance system is an important part of the development of intelligent ships. By introducing advanced sensors, Internet of Things, big data and artificial intelligence technologies, intelligent monitoring and maintenance of ship engines can be realized, providing strong support for the development of intelligent ships.

[0062] In summary, the ship engine predictive maintenance system project has important practical significance and broad application prospects. Through the implementation of this invention, the operational safety of ships can be significantly improved, maintenance costs can be reduced, economic benefits can be improved, and the development of ship intelligence can be promoted. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is a block diagram of the experimental system of the present invention;

[0064] Figure 2 It is a schematic diagram of the vibration signal power spectrum of the present invention. DETAILED DESCRIPTION

[0065] The present invention is further described below in conjunction with the accompanying drawings and embodiments:

[0066] An engine early warning system and method, including a data acquisition module, a data transmission module, a data storage and processing module, a prediction model module, a decision support module, and a user interaction module;

[0067] The data acquisition module is used to build a sensor network and deploy various sensors at key parts of the ship engine, including temperature sensors, pressure sensors, vibration sensors, and fuel consumption sensors, to collect various parameters during the operation of the engine in real time;

[0068] The data transmission module is used to provide a communication interface between the sensor network and the data storage and processing module, and supports wired or wireless communication;

[0069] The data storage and processing module has a built-in data storage unit, which uses high-performance storage devices such as hard disks, solid-state drives or cloud storage to store the collected engine operation data;

[0070] The prediction model module is used for model construction. The model construction is based on historical data and domain knowledge to build a prediction model suitable for ship engines. The model is based on statistical methods, machine learning algorithms or deep learning technology;

[0071] The decision support module is used for fault warning and diagnosis. Based on the output results of the prediction model, it warns and diagnoses possible engine failures and provides detailed fault information, fault severity and prediction time.

[0072] The user interaction module is used to provide a visual interface and a friendly user interface to display the real-time operating status of the engine, prediction results and maintenance recommendation information, and supports multiple display methods such as charts, curves, and alarm prompts.

[0073] Specifically, a data acquisition unit is provided inside the data acquisition module, which is responsible for receiving data from the sensor network and performing preliminary processing and conversion, such as filtering, denoising, and amplification, to ensure the accuracy and reliability of the data.

[0074] More specifically, the data transmission module is provided with a data transmission protocol, and by formulating data transmission standards and protocols, the real-time and security of the data are ensured.

[0075] Furthermore, the data storage unit built into the data storage and processing module uses big data analysis and machine learning algorithms to pre-process, extract features, and recognize patterns on the stored data to detect anomalies and potential faults in the data.

[0076] Furthermore, the prediction model module is also used to train and optimize the model. The model is trained using training data sets, and the model performance is optimized through cross-validation and parameter tuning methods. At the same time, the model is continuously updated and optimized based on real-time data.

[0077] It is worth mentioning that the decision support module is also used for maintenance recommendations and optimization. Based on fault warning and diagnosis results, it provides managers with targeted maintenance recommendations and optimization plans, including adjusting operating parameters, replacing parts, and shutting down for maintenance.

[0078] It is worth noting that the user interaction module is also used for interaction, allowing users to perform operations and settings through the interface, such as adjusting monitoring parameters, viewing historical data, and exporting reports. At the same time, it supports user feedback and opinion collection to continuously improve and optimize the system.

[0079] The present invention provides an early warning method for an engine early warning system, which includes a research and development stage and an operation stage:

[0080] The development phase includes the following steps:

[0081] S1. Requirements analysis and system design: clarify project requirements, conduct market research and technical analysis; design system architecture and functional modules;

[0082] A. Clarify requirements: First, the specific requirements of the ship engine predictive maintenance system need to be clarified, including the parameters that need to be monitored, the predicted fault types, and the real-time requirements of the system;

[0083] B. System planning: According to the needs, plan the overall architecture and functional modules of the system, including data collection, data transmission, data processing, prediction model construction, and decision support.

[0084] S2. Technology research and development and algorithm optimization: Develop data collection and transmission systems; research and optimize prediction algorithms; establish prediction models and verify them.

[0085] A. Data acquisition and transmission processing system:

[0086] Sensor selection and deployment: According to system requirements, select appropriate sensors to monitor key parameters of ship engines, such as temperature, pressure, vibration, and fuel consumption. Sensors should be deployed in locations that can accurately reflect the operating status of the engine.

[0087] Data collection and transmission: Use IoT technology to collect sensor data in real time and transmit it to the cloud or local server for storage.

[0088] Data preprocessing: Clean, denoise, and filter the collected raw data to improve the quality and accuracy of the data.

[0089] B. Feature extraction and model building

[0090] Feature extraction: Use data analysis and mining techniques to extract useful feature information from preprocessed data. These features should be able to reflect the engine's operating status and potential faults.

[0091] Model construction: Based on the extracted feature information, a prediction model is constructed. The prediction model can use machine learning and deep learning algorithms, such as neural networks, support vector machines, and random forests. Parameter tuning and model verification are required during the model construction process to ensure the accuracy and reliability of the model.

[0092] C. Model training and optimization

[0093] Data partitioning: Divide the preprocessed data into training sets and test sets for model training and verification.

[0094] Model training: Use the training set data to train the model, and adjust the model parameters through an iterative optimization algorithm so that the model can better fit the training data.

[0095] Model validation: Use the test set data to validate the trained model and evaluate the prediction accuracy and generalization ability of the model. Further optimize the model based on the validation results.

[0096] S3. System integration and testing: Integrate various functional modules into a unified platform; conduct system testing to ensure stable operation.

[0097] S4. Pilot application and feedback adjustment: select typical ships for pilot application; collect user feedback to optimize and improve the system.

[0098] System deployment: Deploy the constructed predictive maintenance system to the actual ship and connect and integrate it with the ship's engine.

[0099] System testing: Before actual application, the system is fully tested to ensure that the system can run stably and meet actual needs.

[0100] System maintenance: Regularly maintain and update the system, including data backup, software upgrades, and hardware inspections. At the same time, continuously optimize and improve the system based on user feedback and actual application conditions.

[0101] S5. Promotion, application and subsequent support: Promote the application of predictive maintenance systems in the industry; provide continuous technical support and system upgrade services.

[0102] The operation phase includes the following steps:

[0103] S1. Data collection and preprocessing

[0104] A. Data collection: Using sensors and the Internet of Things to collect real-time operating data of ship engines, including key parameters such as vibration, temperature, pressure, and speed;

[0105] B. Data preprocessing: Clean, denoise and normalize the collected data to improve data quality and provide a reliable data basis for subsequent analysis;

[0106] S2. Data Analysis and Modeling

[0107] A. Data analysis: Use machine learning, statistical modeling, and pattern recognition data analysis methods to conduct in-depth analysis of pre-processed data to identify abnormal patterns and fault characteristics in engine operation;

[0108] B. Model prediction: Based on the data analysis results, an engine fault prediction model is established to predict the future performance and fault possibility of the engine through time series analysis, trend analysis, and anomaly detection technology;

[0109] S3. Fault warning and diagnosis

[0110] A. Fault warning: Set reasonable thresholds and warning mechanisms. When the engine operation data is abnormal, the system can send out warning signals in real time to remind managers to take timely measures;

[0111] B. Fault diagnosis: Use fault diagnosis algorithms to further analyze early warning signals, determine the specific location and cause of the fault, and provide guidance for maintenance work;

[0112] S4. Predictive maintenance plan formulation

[0113] A. Maintenance plan formulation: Based on the failure prediction results and risk assessment, formulate a scientific predictive maintenance plan, including maintenance time, maintenance content, and required spare parts;

[0114] B. Optimize maintenance resources: Through the implementation of predictive maintenance plans, optimize the allocation of maintenance resources for ship engines, reduce maintenance costs and improve maintenance efficiency.

[0115] Embodiment 1:

[0116] Abnormal noise and vibration analysis

[0117] For mechanical systems, vibration and noise are the most important diagnostic information. They are a property that is inevitably generated during the operation of the machine. Even the most sophisticated machines will inevitably generate vibration and noise. Mechanical vibration is a special form of object movement, usually manifested as the reciprocating motion of a machine or system near its static equilibrium position, that is, the object changes from an equilibrium state to a reciprocating motion. As long as it is a running machine, especially marine machinery, it will vibrate to a greater or lesser extent. Therefore, vibration signal analysis and diagnosis accounts for the largest proportion of various diagnostic methods, is widely used, and has relatively mature theories and measurement technologies. It is a very important diagnostic method.

[0118] The normal working state of a ship diesel engine is accompanied by regular mechanical operation and vibration. During operation, the abnormal state of the engine can be monitored in real time through sensors. When the diesel engine vibrates abnormally or makes abnormal noise, information can be collected through sensors, and then analyzed and judged through the processor. Based on the most basic dynamics, kinematics, mechanical structure, and working principle technical parameters of the diesel engine, signal analysis and data processing technology are used to process the various data information obtained by the sensor, and the abnormal characteristic parameters are judged from them. The combination of scientific measurement methods and sensor data not only ensures the accuracy and timeliness of information, but also improves the efficiency of fault detection. Combined with the design, manufacturing, maintenance, and working state parameters of ship diesel engines, possible mechanical structure failure mechanisms and component abnormalities are analyzed to warn of abnormal parts of the obtained vibration data and indicate possible faults.

[0119] The vibration measurement system is mainly composed of three parts: the first is the sensor part, that is, the vibration signal measurement part, which is mainly composed of various vibration sensors; the second is data processing, which converts the acquired diesel engine operating status signal or parameter into a standard voltage or current signal for display and analysis through a piezoresistive amplifier.

[0120] Test case:

[0121] By changing the gap size between the piston and the cylinder liner, different cylinder vibration signals generated during the operation of the diesel engine can be obtained. This experiment uses vibration signals of three different gap sizes between the piston and the cylinder liner, 0.07-0.1mm, 0.2-0.25mm and 0.7-0.85mm, to measure, analyze and compare. The power spectrum of the vibration signal after being processed by the DSAP system is shown in the figure below. Figure 2 As shown in Figures (1), (2), and (3), it is obvious that when the gap between the piston and the cylinder liner is small, the vibration energy is mainly concentrated in the first frequency band. As the gap continues to increase, the energy of the vibration signal will move toward the high frequency direction and be mainly concentrated in the second frequency band. Therefore, when monitoring the operation of the diesel engine in real time, when the high-frequency components in the power spectrum increase significantly, it can be judged as a precursor to cylinder scuffing. During the experiment, the test system displayed Figure 2 The phenomenon shown in (4) was discovered through disassembly and inspection of the diesel engine. It was found that the diesel engine piston and cylinder liner showed obvious wear and tear, namely cylinder pulling, which accurately verified the experimental results.

[0122] DSAP System Composition

[0123] Data acquisition module:

[0124] Sensor network: Vibration and noise sensors deployed at key locations of ship diesel engines to capture vibration and noise signals during mechanical operation in real time;

[0125] Data acquisition unit: responsible for converting the signals collected by the sensor into digital format and transmitting them to the central processing unit via wired or wireless means;

[0126] Data processing and analysis module:

[0127] Signal preprocessing: Perform preprocessing operations such as filtering and denoising on the collected original signals to improve signal quality;

[0128] Feature extraction: Use time domain analysis, frequency domain analysis, time-frequency analysis and other methods to extract characteristic parameters that can reflect the mechanical state from the preprocessed signal;

[0129] Intelligent diagnosis: Based on machine learning or deep learning algorithms, pattern recognition is performed on the extracted characteristic parameters to determine whether the machine is in an abnormal state or fault;

[0130] User Interface and Report Generation Module:

[0131] Real-time monitoring interface: provides an intuitive graphical interface to display the real-time vibration and noise status of the mechanical system, as well as potential fault warning information;

[0132] Report Generator: Automatically generates detailed diagnostic reports including fault type, severity, and recommended repair actions.

[0133] The working process of the DSAP system in the first embodiment:

[0134] S1. Deploy sensor network:

[0135] Install vibration and noise sensors at key parts of marine diesel engines (such as cylinder heads, crankcases, and bearing seats);

[0136] Ensure that the connection between the sensor and the data acquisition unit is stable and reliable;

[0137] S2. Data collection and transmission:

[0138] Start the ship's diesel engine, and the sensor begins to capture vibration and noise signals in real time;

[0139] The data acquisition unit converts the collected signals into digital format and transmits them to the central processing unit of the DSAP system through the network;

[0140] S3. Data processing and analysis:

[0141] The DSAP system pre-processes the received signal to remove noise and interference components;

[0142] Extract characteristic parameters of the signal, such as amplitude, frequency, and phase;

[0143] Use intelligent diagnostic algorithms to analyze characteristic parameters and determine whether the machinery is in an abnormal state;

[0144] S4. Fault warning and report generation:

[0145] When abnormal vibration or noise signals are detected, the DSAP system automatically triggers the fault warning mechanism;

[0146] Generates diagnostic reports including fault type, location, severity and recommended repair actions;

[0147] Display warning information and diagnostic reports to operators through the user interface;

[0148] S5. Maintenance and repair:

[0149] Operators arrange corresponding maintenance and repair work based on the warning information and diagnostic reports of the DSAP system;

[0150] During the maintenance process, the historical data and trend analysis functions provided by the DSAP system can be used to assist in determining maintenance plans and optimizing maintenance strategies.

[0151] In the test case of Example 1, the DSAP system was used to analyze the impact of different piston-cylinder liner gaps on the vibration signal of the diesel engine. By changing the gap size and measuring the corresponding vibration signal, the DSAP system successfully identified the trend of the vibration signal moving toward the high-frequency direction as the gap increased. When the system detected a significant increase in the high-frequency component, it accurately warned of the precursor of cylinder scuffing failure.

[0152] Through the abnormal noise and vibration analysis of Example 1, the ship engine predictive maintenance system can accurately capture abnormal vibration and noise during engine operation, and timely warn of potential mechanical failures, such as cylinder pulling and bearing damage, thereby improving the operational safety of the ship. At the same time, the system reduces downtime caused by failures, reduces maintenance costs, and improves the economic benefits of the ship.

[0153] Embodiment 2:

[0154] Oil composition and status analysis

[0155] During the use of ships, the long-term operation of diesel engines will inevitably cause wear of transmission parts. Different degrees of wear of metal parts will form different amounts of particles in the lubricating oil. When the power system of the diesel engine is abnormal, analyzing the oil is helpful to clarify the wear of the parts. Commonly used oil analysis methods include iron spectrum analysis and spectral analysis. In practical applications, the appropriate detection method should be selected according to the different functions of iron spectrum and spectrum: iron spectrum analysis can obtain and analyze the composition, size, and shape of metal particles, but it is difficult to distinguish non-ferrous metals with high sensitivity; spectral analysis is mainly used to detect the content of wear elements in lubricating oil, and cannot accurately obtain the particle shape and wear type parameters. The research on fault prediction and health management technology of marine diesel engines based on oil monitoring can refer to the following papers in detail: Yao Xiaoshan, Zhang Weidong, Zhou Ping, Zhu Ziliang. Research on fault prediction and health management technology of marine diesel engines based on oil monitoring. Journal of Wuhan University of Technology (Transportation Science and Engineering Edition), 2014:8;

[0156] The oil composition and state analysis of Example 2 can detect the wear inside the engine in advance by monitoring the content and type of metal particles in the lubricating oil, and timely warn of potential faults, such as bearing wear and gear damage. This not only improves the operational safety of the ship, but also reduces maintenance costs through preventive maintenance, extends the service life of the engine, and further improves the economic benefits of the ship.

[0157] Embodiment three:

[0158] Temperature and pressure state detection is a common method for diesel engine fault analysis, mainly including cooling water discharge temperature detection, exhaust temperature monitoring, lubricating oil temperature detection, and cylinder pressure monitoring. According to the proportion of fault occurrence, cylinder pressure can display various engine status information. Through detection, it can not only calculate the compression pressure and pressure rise rate parameters, but also realize the status analysis of the power balance and combustion quality of each cylinder. In the process of temperature and pressure detection, many important analysis tasks can be completed through the indicator diagram. The indicator diagram is divided into two categories: direct measurement and recording method or indirect measurement and recording method: the direct measurement and recording method can realize the measurement of the state parameters of the cylinder pressure changing with the crankshaft angle, and realize the intuitive display of the diesel engine working process after sorting and calculation; the indirect measurement and recording method is mainly based on other physical quantities in the operation of the diesel engine to analyze and judge the gas pressure in the cylinder.

[0160] The temperature and pressure state detection of the third embodiment can timely detect abnormal conditions inside the engine, such as overheating and overpressure, by real-time monitoring of the key temperature and pressure parameters of the engine, and effectively prevent failures caused by excessive temperature or abnormal pressure. This not only improves the operational safety of the ship, but also optimizes the operating efficiency of the engine by accurately controlling the temperature and pressure, further improving the economic benefits of the ship.

[0161] Embodiment 4:

[0162] Operation speed analysis

[0163] The speed fluctuation signal of a marine diesel engine can reflect its working quality and status to a certain extent. By analyzing the instantaneous speed fluctuation signal, it is helpful to clarify the possible fault information or abnormal status of the diesel engine. Although the instantaneous speed method can determine the cylinder position of the diesel engine that is not operating normally, it cannot accurately analyze and judge the cause of the cylinder failure.

[0164] The operating speed analysis of the fourth embodiment can accurately judge the operating status of the engine and promptly discover potential faults such as cylinder misfire and uneven fuel supply by real-time monitoring and analyzing the engine speed fluctuation signal. This not only improves the operational safety of the ship, but also reduces fuel consumption and operating costs by optimizing the engine speed control, further improving the economic benefits of the ship.

[0165] Embodiment five:

[0166] Fault prediction methods,

[0167] Taking prediction method as the research entry point, it is divided into empirical model-based, reliability model-based, physical model-based and data-driven methods.

[0168] The fault prediction method of Example 5 can achieve accurate prediction and diagnosis of ship engine faults by comprehensively using multiple prediction models and technologies. This not only improves the operational safety of the ship, but also reduces the probability of faults and maintenance costs through preventive maintenance, and extends the service life of the engine. At the same time, this method provides strong support for the intelligent maintenance of ship engines and promotes the development of ship intelligence.

[0169] In summary, through the comprehensive use of abnormal noise and vibration analysis, oil composition and state analysis, temperature and pressure state detection, and operating speed analysis, real-time monitoring and accurate prediction of the ship engine status are achieved. It can not only timely detect and deal with potential engine faults, prevent the faults from expanding and causing ship accidents, thereby improving the operational safety of the ship; and combined with predictive maintenance strategies, maintenance can be carried out before the fault occurs, avoiding the high cost of subsequent repairs, and optimizing the allocation of maintenance resources, reducing long-term operating costs. In addition, the system can also reduce downtime caused by faults, improve the operating efficiency of the ship, increase economic benefits, and extend the service life of the engine. More importantly, the predictive maintenance system is an important part of the development of ship intelligence and has promoted the intelligent process of the ship industry.

[0170] In addition, the components designed in the present invention are all universal standard parts or components known to technical personnel in this field. Their structures and principles can be known to technical personnel through technical manuals or through conventional experimental methods. They can be fully implemented by technical personnel in this field. Needless to say, the content protected by the present invention does not involve improvements to internal structures and methods.

[0171] The embodiments of the present invention disclose preferred embodiments, but are not limited thereto. A person skilled in the art can easily understand the spirit of the present invention based on the above embodiments and make different extensions and changes. However, as long as they do not deviate from the spirit of the present invention, they are all within the protection scope of the present invention.

Claims

1. An engine warning system, characterized in that: It includes data acquisition module, data transmission module, data storage and processing module, prediction model module, decision support module and user interaction module; The data acquisition module is used to build a sensor network, deploying various sensors at key parts of the ship engine, including temperature sensors, pressure sensors, vibration sensors, and fuel consumption sensors, to collect various parameters during the operation of the engine in real time; The data transmission module is used to provide a communication interface between the sensor network and the data storage and processing module, and supports wired or wireless communication; The data storage and processing module has a built-in data storage unit, which uses a high-performance storage device to store the collected engine operation data; The prediction model module is used for model construction. The model construction is based on historical data and domain knowledge to build a prediction model suitable for ship engines. The model is based on statistical methods, machine learning algorithms or deep learning technology; The decision support module is used for fault warning and diagnosis. According to the output results of the prediction model, it warns and diagnoses possible engine faults and provides detailed fault information, fault severity and prediction time. The user interaction module is used to provide a visual interface, a friendly user interface, display the real-time operating status of the engine, prediction results and maintenance suggestion information, and supports multiple display methods such as charts, curves, and alarm prompts.

2. An engine warning system as claimed in claim 1, characterized in that: The data acquisition module is internally provided with a data acquisition unit, which is responsible for receiving data from the sensor network and performing preliminary processing and conversion to ensure the accuracy and reliability of the data.

3. An engine warning system as claimed in claim 1, characterized in that: The data transmission module is provided with a data transmission protocol, and the real-time and security of the data are ensured by formulating the data transmission standards and protocols.

4. An engine warning system as claimed in claim 1, characterized in that: The data storage unit built into the data storage and processing module uses big data analysis and machine learning algorithms to preprocess, extract features, and recognize patterns on the stored data to discover anomalies and potential faults in the data.

5. The engine warning system according to claim 1, characterized in that: The prediction model module is also used to train and optimize the model. The model is trained using a training data set and the model performance is optimized through cross-validation and parameter tuning methods. At the same time, the model is continuously updated and optimized based on real-time data.

6. The engine warning system according to claim 1, characterized in that: The decision support module is also used for maintenance suggestions and optimization. According to the fault warning and diagnosis results, it provides management personnel with targeted maintenance suggestions and optimization plans, including adjusting operating parameters, replacing parts, and shutting down for maintenance.

7. An engine warning system as claimed in claim 1, characterized in that: The user interaction module is also used for interaction, allowing users to perform operations and settings through the interface, such as adjusting monitoring parameters, viewing historical data, and exporting reports. At the same time, it supports user feedback and opinion collection to continuously improve and optimize the system.

8. The early warning method of an engine early warning system according to any one of claims 1 to 7, characterized in that: The early warning method includes the research and development stage and the operation stage: The development phase includes the following steps: S1. Requirements analysis and system design: clarify project requirements, conduct market research and technical analysis; design system architecture and functional modules; A. Clarify requirements: First, the specific requirements of the ship engine predictive maintenance system need to be clarified, including the parameters that need to be monitored, the predicted fault types, and the real-time requirements of the system; B. System planning: According to the requirements, plan the overall architecture and functional modules of the system, including data collection, data transmission, data processing, prediction model construction, and decision support; S2. Technology R&D and algorithm optimization: Develop data collection and transmission systems; research and optimize prediction algorithms; establish prediction models and verify them; A. Data acquisition and transmission processing system: Sensor selection and deployment: According to system requirements, select appropriate sensors to monitor the key parameters of the ship engine. The sensors should be deployed in a position that can accurately reflect the engine operating status; Data collection and transmission: Using IoT technology, sensor data is collected in real time and transmitted to the cloud or local server for storage; Data preprocessing: Clean, denoise, and filter the collected raw data to improve the quality and accuracy of the data; B. Feature extraction and model building Feature extraction: Use data analysis and mining techniques to extract useful feature information from preprocessed data. These features should be able to reflect the engine's operating status and potential faults. Model construction: Based on the extracted feature information, a prediction model is constructed. The prediction model can use machine learning and deep learning algorithms. Parameter tuning and model verification are required during the model construction process to ensure the accuracy and reliability of the model. C. Model training and optimization Data partitioning: Divide the preprocessed data into training sets and test sets for model training and verification; Model training: Use the training set data to train the model, and adjust the model parameters through iterative optimization algorithms so that the model can better fit the training data; Model validation: Use the test set data to validate the trained model, evaluate the prediction accuracy and generalization ability of the model, and further optimize the model based on the validation results; S3, System integration and testing: Integrate various functional modules into a unified platform; perform system testing to ensure stable operation; S4. Pilot application and feedback adjustment: select typical ships for pilot application; collect user feedback to optimize and improve the system; System deployment: Deploy the built predictive maintenance system on the actual ship and connect and integrate it with the ship engine; System testing: Before actual application, conduct a comprehensive test on the system to ensure that the system can run stably and meet actual needs; System maintenance: Regularly maintain and update the system, including data backup, software upgrades, and hardware inspections. At the same time, continuously optimize and improve the system based on user feedback and actual application conditions; S5. Promotion, application and subsequent support: Promote the application of predictive maintenance systems in the industry; provide continuous technical support and system upgrade services.

9. The early warning method according to claim 8, characterized in that: The operation phase includes the following steps: S1. Data collection and preprocessing A. Data collection: Using sensors and the Internet of Things to collect real-time operating data of ship engines, including key parameters such as vibration, temperature, pressure, and speed; B. Data preprocessing: Clean, denoise and normalize the collected data to improve data quality and provide a reliable data basis for subsequent analysis; S2. Data Analysis and Modeling A. Data analysis: Use machine learning, statistical modeling, and pattern recognition data analysis methods to conduct in-depth analysis of pre-processed data to identify abnormal patterns and fault characteristics in engine operation; B. Model prediction: Based on the data analysis results, an engine fault prediction model is established to predict the future performance and fault possibility of the engine through time series analysis, trend analysis, and anomaly detection technology; S3. Fault warning and diagnosis A. Fault warning: Set reasonable thresholds and warning mechanisms. When the engine operation data is abnormal, the system can send out warning signals in real time to remind managers to take timely measures; B. Fault diagnosis: Use fault diagnosis algorithms to further analyze early warning signals, determine the specific location and cause of the fault, and provide guidance for maintenance work; S4. Predictive maintenance plan formulation A. Maintenance plan formulation: Based on the failure prediction results and risk assessment, formulate a scientific predictive maintenance plan, including maintenance time, maintenance content, and required spare parts; B. Optimize maintenance resources: Through the implementation of predictive maintenance plans, optimize the allocation of maintenance resources for ship engines, reduce maintenance costs and improve maintenance efficiency.

Images