Methods, devices, equipment and storage media for monitoring and diagnosing the operational status of ship equipment
By acquiring real-time operating data and service life of ship equipment, determining the life cycle stage, selecting and updating the target fault diagnosis model, the problem of reduced accuracy in equipment condition monitoring and fault diagnosis in existing technologies is solved. Dynamic matching and self-updating of equipment status are achieved, improving the intelligence and reliability of ship equipment operation and maintenance.
Patent Information
- Application Number
- CN202611514918.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-28
- Publication Date
- 2026-11-17
AI Technical Summary
Existing technologies for ship equipment condition monitoring and fault diagnosis do not consider the performance degradation differences of equipment at different stages of its life cycle, resulting in a decrease in diagnostic accuracy as the equipment ages. Furthermore, they lack a self-updating mechanism based on maintenance feedback, making it difficult to adapt to dynamic changes in equipment condition and lacking the ability to perceive the material-level aging of key components, leading to delayed fault warnings.
By acquiring real-time operating data and service life of ship equipment, determining the life cycle stage, selecting the corresponding target fault diagnosis model, processing and analyzing the data, and feeding back the feedback data after maintenance execution to the pre-set database to update the model, the fault diagnosis model and the equipment aging status are dynamically matched and self-updated.
It improves the accuracy of status identification at different life cycle stages, ensures that the monitoring system adapts to the dynamic changes in equipment status, and enhances the intelligence level and long-term reliability of ship equipment operation and maintenance.
Smart Images

Figure CN123104883A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of ship equipment monitoring technology, and in particular to a method, apparatus, equipment and storage medium for monitoring and diagnosing the operating status of ship equipment. Background Technology
[0002] In the current era of rapid development in the shipping industry, ships, as the core transportation carriers, are crucial for ensuring the safe and stable operation of cargo transportation, personnel safety, and shipping efficiency.
[0003] Ship equipment is complex, encompassing multiple key systems such as power, electricity, and navigation. The normal operation of these systems directly determines the overall performance of the ship.
[0004] Currently, the monitoring and maintenance of ship equipment mainly relies on traditional periodic inspections and human experience-based judgment.
[0005] While regular maintenance can uncover some potential problems, it has significant drawbacks. It cannot monitor the equipment status in real time, and during the intervals between maintenance, potential equipment malfunctions may arise and escalate, ultimately affecting ship operations or even causing safety accidents.
[0006] Moreover, regular maintenance often requires complete disassembly of the equipment, which not only consumes a lot of manpower, material resources and time, increasing maintenance costs, but frequent disassembly may also damage the equipment and shorten its service life.
[0007] Human experience-based judgment relies heavily on the professional knowledge and practical experience of maintenance personnel, is greatly influenced by personal subjective factors, and has different judgment standards among different maintenance personnel, making it difficult to guarantee an accurate assessment of the equipment's condition.
[0008] As ship equipment technology becomes increasingly complex, human experience and judgment are gradually proving inadequate in dealing with new types of malfunctions and complex problems.
[0009] To achieve real-time and accurate monitoring of ship equipment, sensor technology has begun to be applied in the shipbuilding industry.
[0010] Sensors can collect equipment operating parameters in real time, such as temperature, voltage, current, and pressure, as well as environmental parameters, such as wind speed, wind direction, air temperature, and humidity, providing data support for equipment condition assessment.
[0011] The rapid development of artificial intelligence technology has enabled machine learning to demonstrate enormous potential in the field of fault diagnosis.
[0012] In ship equipment fault diagnosis, by establishing machine learning models, a large amount of historical data can be analyzed and learned to uncover potential patterns and rules in the data, thereby accurately judging the operating status of the equipment and intelligently diagnosing faults.
[0013] Machine learning models can quickly identify abnormal equipment conditions by comparing and analyzing real-time data collected by sensors with baseline data under normal operating conditions, and output possible fault types and causes.
[0014] Compared to traditional fault diagnosis methods, machine learning is more accurate and efficient, and can detect potential equipment faults in a timely manner, providing maintenance personnel with a scientific basis for decision-making.
[0015] While existing ship equipment monitoring and maintenance technologies can ensure ship operation safety to a certain extent, they still have many shortcomings.
[0016] The sensor deployment lacks systematicness and comprehensiveness, failing to meet the needs of all-round monitoring.
[0017] In terms of data processing and analysis, there is a lack of effective methods for outlier detection and handling, resulting in low data quality and affecting the accuracy of fault diagnosis.
[0018] In fault diagnosis, traditional methods rely on human experience and simple rule judgments, which are difficult to deal with complex equipment faults. Meanwhile, the application of machine learning technology is still immature and lacks a complete and effective fault diagnosis process and method.
[0019] More importantly, existing monitoring methods are mostly focused on macroscopic physical quantities such as voltage, current, and temperature, lacking microscopic monitoring of the aging degree of materials inside key equipment modules.
[0020] Every device has modules that are prone to failure, and within those modules are the boards that are most prone to failure, and within those boards are the components that are particularly prone to failure.
[0021] For example, the most easily damaged component in power electronic equipment is the frequency converter, and the most easily damaged component in the frequency converter is the IGBT.
[0022] An IGBT with an A current rating of around 3000A costs over 100,000 RMB. If the IGBT fails, it will also damage the load downstream.
[0023] Changes in temperature and elemental composition of key boards and components can also reflect the health status of modules and equipment.
[0024] However, existing technologies lack the means to reflect the aging degree of modules by utilizing changes in elemental composition or molecular information, resulting in the inability to provide early warning of material-level failures before physical parameters become abnormal.
[0025] Therefore, it is essential to develop a ship equipment monitoring and maintenance system based on multi-sensor fusion, spectral material analysis, intelligent data processing, and machine learning for fault diagnosis.
[0026] This system can rationally deploy sensors to comprehensively collect equipment operating status and environmental parameters; utilize advanced data processing technology to preprocess and analyze the collected data; employ machine learning models to achieve intelligent diagnosis and early warning of equipment faults; and formulate scientific and reasonable maintenance plans based on the diagnostic results, thereby improving the reliability and safety of ship equipment, reducing maintenance costs, and enhancing shipping efficiency. Summary of the Invention
[0027] The main objective of this invention is to provide a method, apparatus, device, and storage medium for monitoring and diagnosing the operational status of ship equipment. This invention aims to address the technical problems in the prior art where ship equipment status monitoring and fault diagnosis typically employ fixed models that do not consider the performance degradation differences of equipment at different life stages, resulting in decreased diagnostic accuracy as equipment ages. Furthermore, the models lack a self-updating mechanism based on maintenance feedback, making it difficult to adapt to dynamic changes in equipment status and lacking the ability to perceive the material-level aging of key components, leading to delayed fault warnings.
[0028] In a first aspect, the present invention provides a method for monitoring and diagnosing the operating status of ship equipment, the method comprising the following steps: Acquire real-time operating data and service life of ship equipment, and determine the current life cycle stage of the ship equipment based on the service life of the equipment; Based on the life cycle stage, a corresponding target fault diagnosis model is selected from a pre-set database, and the real-time operating data is input into the target fault diagnosis model to parse and obtain the current working status of the ship equipment; A maintenance strategy is formulated based on the current working status, and the feedback data after maintenance execution is fed back to the preset database to update the target fault diagnosis model.
[0029] Optionally, the step of acquiring real-time operating data and service life of the ship's equipment, and determining the current life cycle stage of the ship's equipment based on the service life, includes: Data is collected by a set of sensors deployed at key locations on the ship's equipment. The average value of the sensor data collected multiple times is taken as the real-time operating data. The sensor set includes conventional physical quantity sensors and spectral analysis sensors. The spectral analysis sensors are used to collect data on changes in elemental composition or molecular information on the surface of key modules. The current service life of the ship's equipment is obtained, and the current service life is compared with the preset total service life. Based on the comparison result, the current life cycle stage of the ship's equipment is determined.
[0030] Optionally, obtaining the current service life of the ship's equipment by comparing the current service life with a preset total service life and determining the current life cycle stage of the ship's equipment based on the comparison result includes: Obtain the current service life of the ship's equipment, compare the current service life with the preset total service life L, and generate a comparison result; When the comparison result shows that the current service life is between 0 and the preset total service life L / 3 years, the current life cycle stage of the ship equipment is determined to be the initial stage. When the comparison result indicates that the current service life is between L / 3 and 2L / 3 years, the current life cycle stage of the ship equipment is determined to be mid-term. When the comparison result indicates that the current service life is between 2L / 3 and L years, the current life cycle stage of the ship equipment is determined to be the late stage. When the comparison result indicates that the current service life is greater than L years, the current life cycle stage of the ship equipment is determined to be the scrapping period.
[0031] Optionally, the step of selecting a corresponding target fault diagnosis model from a pre-set database based on the lifecycle stage, and inputting the real-time operating data into the target fault diagnosis model to parse and obtain the current operating status of the ship's equipment, includes: Based on the life cycle stage, a target sub-database and a target fault diagnosis model matching the life cycle stage are selected from a preset database. The target sub-database stores the standard operating parameter range and historical fault data corresponding to the life cycle stage. After the real-time operating data is input into the target fault diagnosis model, the target fault diagnosis model parses the real-time operating data to obtain the current operating status of the ship's equipment.
[0032] Optionally, the step of inputting the real-time operating data into the target fault diagnosis model and then parsing the real-time operating data through the target fault diagnosis model to obtain the current operating status of the ship's equipment includes: After the real-time operation data is input into the target fault diagnosis model, the target fault diagnosis model sequentially removes outliers, eliminates dimensional differences, and extracts state features from the real-time operation data. The current working status of the ship's equipment is obtained by comparing and analyzing the processed feature data with the benchmark data in the target sub-database.
[0033] Optionally, the step of formulating a maintenance strategy based on the current working state and feeding back the feedback data after maintenance execution to the preset database to update the target fault diagnosis model includes: When the current working state is at the first preset abnormal level, remote monitoring is performed and the data collection frequency is increased; When the current working status is at the second preset abnormal level, an on-site inspection and maintenance instruction is generated. When the current working state is at the third preset abnormality level, the emergency maintenance procedure is initiated; wherein, the deviation between the real-time operating data and the benchmark data corresponding to the first preset abnormality level, the second preset abnormality level and the third preset abnormality level increases sequentially. Obtain feedback data after maintenance execution, and feed the feedback data back to the preset database to update the target fault diagnosis model.
[0034] Optionally, the step of obtaining feedback data after maintenance execution and feeding the feedback data back to the preset database to update the target fault diagnosis model includes: The actual fault record data after maintenance is performed is obtained as the feedback data; The feedback data is compared with the diagnostic results output by the target fault diagnosis model; When the feedback data matches the diagnostic result, the feedback data is added to the preset database, and the target fault diagnosis model is retrained based on the updated preset database to obtain the updated target fault diagnosis model.
[0035] Secondly, to achieve the above objectives, the present invention also proposes a ship equipment operating status monitoring and diagnostic device, the ship equipment operating status monitoring and diagnostic device comprising: The life cycle determination module is used to acquire real-time operating data and service life of ship equipment, and determine the current life cycle stage of the ship equipment based on the service life of the equipment. The fault diagnosis module is used to select the corresponding target fault diagnosis model from the preset database according to the life cycle stage, and input the real-time operation data into the target fault diagnosis model to parse and obtain the current working status of the ship equipment; The model update module is used to formulate a maintenance strategy based on the current working status and to feed back the feedback data after maintenance execution to the preset database to update the target fault diagnosis model.
[0036] Thirdly, to achieve the above objectives, the present invention also proposes a ship equipment operation status monitoring and diagnostic device, the ship equipment operation status monitoring and diagnostic device comprising: a memory, a processor, and a ship equipment operation status monitoring and diagnostic program stored in the memory and executable on the processor, the ship equipment operation status monitoring and diagnostic program being configured to implement the steps of the ship equipment operation status monitoring and diagnostic method as described above.
[0037] Fourthly, to achieve the above objectives, the present invention also proposes a storage medium storing a ship equipment operation status monitoring and diagnostic program, wherein when the ship equipment operation status monitoring and diagnostic program is executed by a processor, it implements the steps of the ship equipment operation status monitoring and diagnostic method described above.
[0038] The proposed method for monitoring and diagnosing the operational status of ship equipment involves acquiring real-time operational data and the equipment's service life, and determining the current lifecycle stage of the equipment based on its service life. A corresponding target fault diagnosis model is selected from a pre-set database based on this lifecycle stage, and the real-time operational data is input into the target fault diagnosis model to parse and obtain the current operating status of the ship equipment. A maintenance strategy is formulated based on the current operating status, and feedback data after maintenance execution is fed back to the pre-set database to update the target fault diagnosis model. By introducing the equipment lifecycle stage as a model selection criterion, dynamic matching between the fault diagnosis model and the equipment's aging state is achieved, eliminating diagnostic biases caused by equipment performance degradation, thereby improving the accuracy of status identification at different lifecycle stages. Simultaneously, the feedback data feedback mechanism after maintenance execution enables self-updating and iterative optimization of the diagnostic model, ensuring that the monitoring system can adapt to dynamic changes in equipment status and improving the intelligence level and long-term reliability of ship equipment operation and maintenance. Attached Figure Description
[0039] Figure 1 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of the present invention; Figure 2 This is a flowchart illustrating the first embodiment of the ship equipment operation status monitoring and diagnosis method of the present invention; Figure 3 This is a flowchart illustrating the second embodiment of the ship equipment operation status monitoring and diagnosis method of the present invention; Figure 4 This is a flowchart illustrating the third embodiment of the ship equipment operation status monitoring and diagnosis method of the present invention; Figure 5 This is a flowchart illustrating the fourth embodiment of the ship equipment operation status monitoring and diagnosis method of the present invention; Figure 6This is a schematic diagram of the overall process of the ship equipment operation status monitoring and diagnosis method of the present invention; Figure 7 This is a functional block diagram of the first embodiment of the ship equipment operation status monitoring and diagnostic device of the present invention.
[0040] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation
[0041] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
[0042] The solution of this invention mainly involves: acquiring real-time operating data and the service life of ship equipment, and determining the current life cycle stage of the ship equipment based on the service life; selecting a corresponding target fault diagnosis model from a pre-set database based on the life cycle stage, and inputting the real-time operating data into the target fault diagnosis model to parse and obtain the current working state of the ship equipment; formulating a maintenance strategy based on the current working state, and feeding back the feedback data after maintenance execution to the pre-set database to update the target fault diagnosis model. This approach, by introducing the equipment life cycle stage as a model selection criterion, enables the fault diagnosis model to adapt to equipment aging. Dynamic matching of status eliminates diagnostic biases caused by equipment performance degradation, thereby improving the accuracy of status identification at different lifecycle stages. Simultaneously, through a feedback data feedback mechanism after maintenance execution, the diagnostic model can be self-updated and iteratively optimized, ensuring the monitoring system can adapt to dynamic changes in equipment status. This enhances the intelligence level and long-term reliability of ship equipment operation and maintenance. It solves the technical problems of existing ship equipment status monitoring and fault diagnosis, which typically use fixed models, fail to consider performance degradation differences at different lifecycle stages, leading to decreased diagnostic accuracy with equipment aging, and lack of a self-updating mechanism based on maintenance feedback, making it difficult to adapt to dynamic changes in equipment status.
[0043] Reference Figure 1 , Figure 1 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of the present invention.
[0044] like Figure 1As shown, the device may include: a processor 1001, such as a CPU; a communication bus 1002; a user interface 1003; a network interface 1004; and a memory 1005. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen or an input unit such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be high-speed RAM or non-volatile memory, such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.
[0045] Those skilled in the art will understand that Figure 1 The device structure shown does not constitute a limitation on the device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.
[0046] like Figure 1 As shown, the memory 1005, which serves as a storage medium, may include an operating system, a network communication module, a user interface module, and a ship equipment operation status monitoring and diagnostic program.
[0047] The device of this invention calls the ship equipment operation status monitoring and diagnostic program stored in the memory 1005 through the processor 1001, and performs the following operations: Acquire real-time operating data and service life of ship equipment, and determine the current life cycle stage of the ship equipment based on the service life of the equipment; Based on the life cycle stage, a corresponding target fault diagnosis model is selected from a pre-set database, and the real-time operating data is input into the target fault diagnosis model to parse and obtain the current working status of the ship equipment; A maintenance strategy is formulated based on the current working status, and the feedback data after maintenance execution is fed back to the preset database to update the target fault diagnosis model.
[0048] The device of the present invention, through processor 1001 calling the ship equipment operation status monitoring and diagnostic program stored in memory 1005, also performs the following operations: Data is collected by a set of sensors deployed at key locations on the ship's equipment. The average value of the sensor data collected multiple times is taken as the real-time operating data. The sensor set includes conventional physical quantity sensors and spectral analysis sensors. The spectral analysis sensors are used to collect data on changes in elemental composition or molecular information on the surface of key modules. The current service life of the ship's equipment is obtained, and the current service life is compared with the preset total service life. Based on the comparison result, the current life cycle stage of the ship's equipment is determined.
[0049] The device of the present invention, through processor 1001 calling the ship equipment operation status monitoring and diagnostic program stored in memory 1005, also performs the following operations: Obtain the current service life of the ship's equipment, compare the current service life with the preset total service life L, and generate a comparison result; When the comparison result shows that the current service life is between 0 and the preset total service life L / 3 years, the current life cycle stage of the ship equipment is determined to be the initial stage. When the comparison result indicates that the current service life is between L / 3 and 2L / 3 years, the current life cycle stage of the ship equipment is determined to be mid-term. When the comparison result indicates that the current service life is between 2L / 3 and L years, the current life cycle stage of the ship equipment is determined to be the late stage. When the comparison result indicates that the current service life is greater than L years, the current life cycle stage of the ship equipment is determined to be the scrapping period.
[0050] The device of the present invention, through processor 1001 calling the ship equipment operation status monitoring and diagnostic program stored in memory 1005, also performs the following operations: Based on the life cycle stage, a target sub-database and a target fault diagnosis model matching the life cycle stage are selected from a preset database. The target sub-database stores the standard operating parameter range and historical fault data corresponding to the life cycle stage. After the real-time operating data is input into the target fault diagnosis model, the target fault diagnosis model parses the real-time operating data to obtain the current operating status of the ship's equipment.
[0051] The device of the present invention, through processor 1001 calling the ship equipment operation status monitoring and diagnostic program stored in memory 1005, also performs the following operations: After the real-time operation data is input into the target fault diagnosis model, the target fault diagnosis model sequentially removes outliers, eliminates dimensional differences, and extracts state features from the real-time operation data. The current working status of the ship's equipment is obtained by comparing and analyzing the processed feature data with the benchmark data in the target sub-database.
[0052] The device of the present invention, through processor 1001 calling the ship equipment operation status monitoring and diagnostic program stored in memory 1005, also performs the following operations: When the current working state is at the first preset abnormal level, remote monitoring is performed and the data collection frequency is increased; When the current working status is at the second preset abnormal level, an on-site inspection and maintenance instruction is generated. When the current working state is at the third preset abnormality level, the emergency maintenance procedure is initiated; wherein, the deviation between the real-time operating data and the benchmark data corresponding to the first preset abnormality level, the second preset abnormality level and the third preset abnormality level increases sequentially. Obtain feedback data after maintenance execution, and feed the feedback data back to the preset database to update the target fault diagnosis model.
[0053] The device of the present invention, through processor 1001 calling the ship equipment operation status monitoring and diagnostic program stored in memory 1005, also performs the following operations: The actual fault record data after maintenance is performed is obtained as the feedback data; The feedback data is compared with the diagnostic results output by the target fault diagnosis model; When the feedback data matches the diagnostic result, the feedback data is added to the preset database, and the target fault diagnosis model is retrained based on the updated preset database to obtain the updated target fault diagnosis model.
[0054] This embodiment, through the above-described scheme, acquires real-time operating data and the service life of the ship's equipment, and determines the current life cycle stage of the equipment based on the service life. Based on the life cycle stage, a corresponding target fault diagnosis model is selected from a pre-set database, and the real-time operating data is input into the target fault diagnosis model to parse and obtain the current working state of the ship's equipment. A maintenance strategy is formulated based on the current working state, and the feedback data after maintenance execution is fed back to the pre-set database to update the target fault diagnosis model. By introducing the equipment life cycle stage as the model selection criterion, dynamic matching between the fault diagnosis model and the aging state of the equipment can be achieved, eliminating diagnostic biases caused by equipment performance degradation, thereby improving the state identification accuracy at different life cycle stages. Simultaneously, the feedback data feedback mechanism after maintenance execution enables self-updating and iterative optimization of the diagnostic model, ensuring that the monitoring system can adapt to dynamic changes in equipment status, and improving the intelligence level and long-term reliability of ship equipment operation and maintenance.
[0055] Based on the above hardware structure, an embodiment of the ship equipment operation status monitoring and diagnosis method of the present invention is proposed.
[0056] Reference Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of the ship equipment operation status monitoring and diagnosis method of the present invention.
[0057] In the first embodiment, the method for monitoring and diagnosing the operating status of ship equipment includes the following steps: Step S10: Obtain real-time operating data and service life of the ship's equipment, and determine the current life cycle stage of the ship's equipment based on the service life of the equipment.
[0058] It should be noted that after obtaining the real-time operating data and service life of the ship's equipment, the service life can be used as a basis for stage division to define the aging process of the ship's equipment, determine the current life cycle stage of the ship's equipment, and make the selection of subsequent fault diagnosis models constrained by the life cycle stage.
[0059] Step S20: Select the corresponding target fault diagnosis model from the preset database according to the life cycle stage, and input the real-time operation data into the target fault diagnosis model to parse and obtain the current working status of the ship equipment.
[0060] It should be understood that, based on the life cycle stage, the life cycle stage can be used as the basis for model selection. The corresponding target fault diagnosis model can be selected from the preset database, and the real-time operation data can be input into the target fault diagnosis model for parsing to obtain the current working status of the ship equipment.
[0061] Step S30: Formulate a maintenance strategy based on the current working status, and feed back the feedback data after maintenance execution to the preset database to update the target fault diagnosis model.
[0062] It is understandable that a corresponding maintenance strategy is formulated based on the current working status, and the feedback data after maintenance execution is fed back to the preset database. The target fault diagnosis model is updated using the feedback data, so that the model optimization process incorporates actual maintenance verification information, and realizes the continuous iteration and updating of the fault diagnosis model based on maintenance feedback.
[0063] This embodiment, through the above-described scheme, acquires real-time operating data and the service life of the ship's equipment, and determines the current life cycle stage of the equipment based on the service life. Based on the life cycle stage, a corresponding target fault diagnosis model is selected from a pre-set database, and the real-time operating data is input into the target fault diagnosis model to parse and obtain the current working state of the ship's equipment. A maintenance strategy is formulated based on the current working state, and the feedback data after maintenance execution is fed back to the pre-set database to update the target fault diagnosis model. By introducing the equipment life cycle stage as the model selection criterion, dynamic matching between the fault diagnosis model and the aging state of the equipment can be achieved, eliminating diagnostic biases caused by equipment performance degradation, thereby improving the state identification accuracy at different life cycle stages. Simultaneously, the feedback data feedback mechanism after maintenance execution enables self-updating and iterative optimization of the diagnostic model, ensuring that the monitoring system can adapt to dynamic changes in equipment status, and improving the intelligence level and long-term reliability of ship equipment operation and maintenance.
[0064] Furthermore, Figure 3 This is a flowchart illustrating the second embodiment of the ship equipment operation status monitoring and diagnosis method of the present invention, as shown below. Figure 3 As shown, based on the first embodiment, a second embodiment of the ship equipment operation status monitoring and diagnosis method of the present invention is proposed. In this embodiment, step S10 specifically includes the following steps: Step S11: Data is collected by a set of sensors deployed at key locations of the ship's equipment. The average value of the sensor data collected multiple times is taken as the real-time operating data. The set of sensors includes conventional physical quantity sensors and spectral analysis sensors. The spectral analysis sensors are used to collect data on changes in elemental composition or molecular information on the surface of key modules.
[0065] It should be noted that by using a sensor network deployed at key monitoring points to acquire raw signals, and by averaging multiple sampling values within the same time window, random fluctuations that may exist in a single measurement are eliminated, thereby generating stable monitoring values. By introducing spectral analysis technology to obtain changes in the elemental composition or molecular information of key modules, and incorporating material-level aging characteristics into the diagnostic model, the accuracy and foresight of fault early warning are further improved.
[0066] In a specific implementation, the sensor set includes not only conventional physical quantity sensors such as temperature, voltage, and current, but also spectral analysis sensors.
[0067] For critical modules (such as Insulated Gate Bipolar Transistor (IGBT) modules in frequency converters), small-scale laser-induced breakdown spectroscopy (LIBS) technology can be integrated into the equipment to obtain changes in elemental levels at locations unrelated to the key modules (the surface of the circuit board, without traces or solder joints) in order to understand the aging degree of the modules.
[0068] Alternatively, near-infrared spectroscopy can be used to obtain changes in molecular information, reflecting changes in elemental information on the sample surface.
[0069] This elemental or molecular information, as a new feature dimension, is input into the model along with conventional information such as wind speed, wind direction, temperature, humidity, seawater, and salinity.
[0070] Considering the size of LIBS equipment (approximately 20cm x 30cm at the minimum), in engineering implementation, for large ship equipment with sufficient space, the spectral analysis sensor can be fixedly integrated into the monitoring cabinet; for small equipment with limited space, a portable spectrometer can be used for testing during regular maintenance and the data can be entered into the system. This covers both fixed integration and regular testing implementation methods to ensure engineering feasibility.
[0071] Step S12: Obtain the current service life of the ship's equipment, compare the current service life with the preset total service life, and determine the current life cycle stage of the ship's equipment based on the comparison result.
[0072] Understandably, after obtaining the current service life of the ship's equipment, the length of time the equipment has been in operation can be compared with the preset total lifespan standard, and the aging period of the equipment can be determined based on the comparison results.
[0073] Furthermore, step S12 specifically includes the following steps: Obtain the current service life of the ship's equipment, compare the current service life with the preset total service life L, and generate a comparison result; When the comparison result shows that the current service life is between 0 and the preset total service life L / 3 years, the current life cycle stage of the ship equipment is determined to be the initial stage. When the comparison result indicates that the current service life is between L / 3 and 2L / 3 years, the current life cycle stage of the ship equipment is determined to be mid-term. When the comparison result indicates that the current service life is between 2L / 3 and L years, the current life cycle stage of the ship equipment is determined to be the late stage. When the comparison result indicates that the current service life is greater than L years, the current life cycle stage of the ship equipment is determined to be the scrapping period.
[0074] It should be noted that by comparing the current cumulative years of use with the preset total lifespan L, and setting the boundary threshold according to one-third or two-thirds of the total lifespan, the time interval in which the comparison result falls is mapped to the initial, middle, late or scrap period, so that the determination of the life cycle stage has a quantified time boundary, realizing the logical determination from continuous years of use to discrete life cycle attributes.
[0075] This embodiment utilizes the aforementioned scheme to collect data through a set of sensors deployed at key locations on the ship's equipment. The average value of multiple consecutive sensor data acquisitions is used as real-time operational data. The sensor set includes conventional physical quantity sensors and spectral analysis sensors. The spectral analysis sensors are used to collect data on changes in elemental composition or molecular information on the surface of key modules. The current service life of the ship's equipment is obtained and compared with a preset total service life. Based on the comparison result, the current life cycle stage of the ship's equipment is determined. Multi-sampling averaging eliminates random interference during sensor acquisition, ensuring the accuracy and stability of real-time operational data. Furthermore, the life cycle stages can be quantitatively divided based on the comparison between the equipment's service life and total lifespan, providing a clear aging status basis for the selection of subsequent diagnostic models. This achieves high-quality data input and accurate identification of the equipment's life cycle status. By introducing spectral analysis technology to obtain changes in the elemental composition or molecular information of key modules, material-level aging characteristics are incorporated into the diagnostic model, further improving the accuracy and foresight of fault warnings.
[0076] Furthermore, Figure 4 This is a flowchart illustrating the third embodiment of the ship equipment operation status monitoring and diagnosis method of the present invention, as shown below. Figure 4 As shown, based on the first embodiment, a third embodiment of the ship equipment operation status monitoring and diagnosis method of the present invention is proposed. In this embodiment, step S20 specifically includes the following steps: Step S21: Select a target sub-database and a target fault diagnosis model that match the life cycle stage from the preset database, wherein the target sub-database stores the standard operating parameter range and historical fault data corresponding to the life cycle stage.
[0077] It should be noted that by using the lifecycle stage as the selection criterion, matching sub-databases and diagnostic models can be retrieved from the pre-set database, and the standard parameter ranges and historical fault data of the corresponding stage stored in the sub-database can be used as diagnostic benchmarks.
[0078] Step S22: After inputting the real-time operating data into the target fault diagnosis model, the real-time operating data is parsed through the target fault diagnosis model to obtain the current working status of the ship equipment.
[0079] It is understandable that by using the target fault diagnosis model to perform feature analysis on the input real-time operating data, a current working state representing the equipment's operating condition is generated, making the state recognition process constrained by the model algorithm, thus realizing the feature mapping from the original operating data to the equipment's working state.
[0080] Furthermore, step S22 specifically includes the following steps: After the real-time operation data is input into the target fault diagnosis model, the target fault diagnosis model sequentially removes outliers, eliminates dimensional differences, and extracts state features from the real-time operation data. The current working status of the ship's equipment is obtained by comparing and analyzing the processed feature data with the benchmark data in the target sub-database.
[0081] It should be noted that by performing a standardized processing procedure on real-time operating data through the target fault diagnosis model, outlier removal, dimensional difference elimination, and state feature extraction are completed in sequence. The processed feature data is then compared and analyzed with the benchmark data in the target sub-database, so that the state identification process has a unified data benchmark, and accurate analysis from raw operating data to the current working state of the equipment is achieved.
[0082] This embodiment, through the above-described scheme, selects a target sub-database and a target fault diagnosis model from a pre-set database that match the lifecycle stage. The target sub-database stores the standard operating parameter range and historical fault data corresponding to the lifecycle stage. After inputting the real-time operating data into the target fault diagnosis model, the model parses the real-time operating data to obtain the current operating status of the ship's equipment. By matching the corresponding sub-database and diagnosis model to the lifecycle stage, it ensures that the diagnostic benchmark is adapted to the equipment's aging level, avoids errors caused by using a uniform model, and obtains the current operating status by parsing real-time data through the model, achieving accurate identification and adaptive diagnosis of equipment status under different lifecycle stages.
[0083] Furthermore, Figure 5 This is a flowchart illustrating the fourth embodiment of the ship equipment operation status monitoring and diagnosis method of the present invention, as shown below. Figure 5 As shown, based on the first embodiment, a fourth embodiment of the ship equipment operation status monitoring and diagnosis method of the present invention is proposed. In this embodiment, step S30 specifically includes the following steps: Step S31: When the current working state is at the first preset abnormal level, perform remote monitoring and increase the data acquisition frequency.
[0084] It should be noted that in response to the current working state reaching the first preset abnormality level, the remote monitoring mechanism is activated and the data sampling density is increased. By increasing the monitoring granularity, the trend of abnormal development is captured, so that the investment of monitoring resources is matched with the severity of the abnormality, and excessive maintenance measures are avoided for minor abnormalities.
[0085] Step S32: When the current working state is the second preset abnormal level, generate on-site inspection and maintenance instructions.
[0086] Understandably, in response to the current working state reaching the second preset abnormal level, an on-site inspection and maintenance instruction is generated, triggering the on-site intervention mechanism for technical personnel. This enables the substantive investigation and repair of moderate fault hazards, avoiding situations where remote monitoring cannot solve hardware problems or cause downtime losses due to emergency maintenance.
[0087] Step S33: When the current working state is the third preset abnormality level, start the emergency maintenance procedure; wherein, the deviation between the real-time operating data and the benchmark data corresponding to the first preset abnormality level, the second preset abnormality level and the third preset abnormality level increases sequentially.
[0088] It should be understood that in response to the current working state reaching the third preset abnormality level, an emergency maintenance procedure is initiated. Based on the maximum deviation between the real-time operating data and the baseline data corresponding to the third preset abnormality level, the equipment is characterized as being in the highest risk state. This ensures that the urgency of the maintenance measures matches the severity of the abnormality, thus avoiding safety accidents or equipment damage caused by serious malfunctions.
[0089] Step S34: Obtain feedback data after maintenance execution, and feed the feedback data back to the preset database to update the target fault diagnosis model.
[0090] Understandably, the actual results information collected after maintenance is completed is used as feedback data and sent back to the pre-set database for retraining of the target fault diagnosis model, so that the model update process incorporates actual maintenance verification information.
[0091] Furthermore, step S34 specifically includes the following steps: The actual fault record data after maintenance is performed is obtained as the feedback data; The feedback data is compared with the diagnostic results output by the target fault diagnosis model; When the feedback data matches the diagnostic result, the feedback data is added to the preset database, and the target fault diagnosis model is retrained based on the updated preset database to obtain the updated target fault diagnosis model.
[0092] It should be noted that by collecting actual fault record data after maintenance is performed as feedback truth values, and comparing them with the diagnostic results output by the model, the data is only included in the database and the model is retrained when the verification is consistent to obtain the updated model. This makes the model update process constrained by the actual maintenance results, avoids the model performance degradation caused by erroneous data, and realizes closed-loop iteration and accuracy optimization of the diagnostic model based on real feedback verification.
[0093] In the specific implementation, see Figure 6 , Figure 6 This is a schematic diagram of the overall process of the ship equipment operation status monitoring and diagnosis method of the present invention, as shown below. Figure 6 As shown: Step 1. Deployment of key position sensors Various types of sensors are deployed at key locations on the ship's equipment, including temperature sensors, voltage sensors, current sensors, pressure sensors, power consumption sensors, battery status sensors, central processing unit (CPU) usage sensors, input / output (IO) usage sensors, memory usage sensors, and meteorological and environmental sensors (such as wind speed sensors, wind direction sensors, air temperature sensors, humidity sensors, seawater temperature sensors, salinity sensors, etc.); spectral analysis sensors (such as small LIBS sensors or near-infrared sensors) are also included to collect data on changes in elemental composition or molecular information on the surface of key modules.
[0094] By strategically deploying these sensors, comprehensive monitoring of the ship's equipment operating status, environmental parameters, and material aging status can be achieved.
[0095] Step 2. Determine the anomaly level The level of equipment anomaly is determined based on the degree of deviation between real-time data and baseline data.
[0096] The specific classification criteria are as follows: ① Mild anomaly: Data deviation is within 10% of the baseline range. The equipment can still operate normally, but there may be potential hidden dangers.
[0097] ② Moderate anomaly: Data deviation is between 10% and 30% of the baseline range, which affects equipment performance to some extent and may indicate local faults.
[0098] ③ Severe anomaly: Data deviation exceeds 30% of the baseline range, equipment operation shows obvious malfunction, which may endanger the safety of the ship and requires immediate measures.
[0099] Determining the anomaly level provides a basis for subsequent fault diagnosis and maintenance plan development.
[0100] Step 3. Establishing benchmark files and comparing data Establish a baseline document for the normal operation of ship equipment, and build an equipment condition monitoring database based on this document.
[0101] The benchmark documents are shown in Table 1: Table 1. Example of benchmark file information:
[0102] This benchmark file contains the standard operating parameter ranges of the equipment under various operating conditions, such as temperature range, voltage range, current range, pressure range, power consumption range, battery status indicators, upper limit of CPU utilization, upper limit of IO utilization, upper limit of memory utilization, etc., as well as the corresponding meteorological and environmental condition ranges, and also includes the elemental composition benchmark range or molecular information benchmark range of key modules.
[0103] The real-time collected data is compared and analyzed with the data in the benchmark file, and a machine learning model is established to determine whether the device's operating status is abnormal.
[0104] In particular, this basic document includes data on the normal operation of the device in different years (before its reliable lifespan).
[0105] Step 31: If the service life of a certain device is L years, then the data of the device is divided into four periods: initial stage (L / 3 years), middle stage (L / 3~2L / 3 years), late stage (2L / 3~L years), and scrapping stage (>L years).
[0106] The benchmark database contains four sub-databases: the initial database, the intermediate database, and the final database.
[0107] Step 32: The data in each sub-database is divided into four categories: normal data, slightly abnormal data, moderately abnormal data, and severely abnormal data, with data labels set as 1, 2, 3, and 4 respectively.
[0108] Step 4. Fault Diagnosis Model Establishment The proposed model is used to diagnose faults in abnormal equipment data.
[0109] Collect historical fault data, environmental data, spectral analysis data, and corresponding equipment operating parameters, and input them into the established model for modeling.
[0110] When a device malfunction is detected, the data is input into a trained machine learning model, which then outputs the possible fault type and cause.
[0111] Specifically, the model comprises five modules: Module 1: Data Filtering Module: Outlier Removal During the operation of ship equipment, the data collected by sensors may contain various outliers, which may be caused by factors such as sensor failure and environmental interference.
[0112] This module uses statistical methods or machine learning-based anomaly detection algorithms to comprehensively analyze the collected data, identify and remove outliers, and ensure the accuracy and reliability of the data.
[0113] Module 2: Data Scale Transformation Module: Eliminating the Impact of Data Differences in Different Dimensions. Ship equipment monitoring involves various parameters, such as temperature, voltage, current, and pressure, each with different dimensions and ranges. The data scale transformation module uses methods such as normalization and standardization to scale data of different dimensions to a uniform scale, eliminating the impact of data differences on the final results and providing a good data foundation for subsequent feature extraction and model training.
[0114] Module 3: Feature Extraction and Selection Module. Feature Extraction: Extracts features that reflect the equipment status from the preprocessed raw data. These features may include time-domain features (such as mean, variance, peak value, etc.), frequency-domain features (such as frequency components, power spectral density, etc.), and time-frequency-domain features (such as wavelet coefficients, etc.).
[0115] This also includes material characteristics (such as the rate of change in elemental content, changes in molecular bond strength, etc.). Through feature extraction, the raw data is transformed into more representative feature vectors, which facilitates subsequent analysis and processing.
[0116] Feature selection: Select the most relevant features for equipment status monitoring from the extracted features, and remove redundant and irrelevant features.
[0117] Feature selection algorithms, such as correlation-based feature selection and model-based feature selection, are employed to reduce feature dimensionality and improve model training efficiency and generalization ability.
[0118] Module 4: Model Selection and Training Module Model Selection: Select a suitable machine learning model based on the problem type (such as classification, regression, etc.) and data characteristics of the equipment status monitoring.
[0119] Common models include decision trees, support vector machines, and neural networks. Different models have different characteristics and applicable scenarios. This module comprehensively considers factors such as the model's accuracy, efficiency, and interpretability to select the optimal model.
[0120] Model training: The preprocessed data is divided into training set, validation set and test set.
[0121] The selected model is trained using a training set, and the model's parameters are adjusted so that the model can learn patterns and rules in the data.
[0122] During training, a validation set is used to monitor the model's performance and prevent overfitting.
[0123] Training is stopped when the model reaches its optimal performance on the validation set, and the model's generalization ability is evaluated using the test set.
[0124] Each of the above modules includes multiple methods, and the appropriate method will be selected based on the distribution characteristics of the data, the purpose of the analysis, and the overall effect (generalization ability, analysis accuracy, analysis time, etc.).
[0125] The methods within each module can be used individually or in combination.
[0126] For example, the data filtering module includes statistical methods (normal distribution, standard scores, interquartile range, etc.), distance methods (K-nearest neighbors, local outlier method), density methods (DBSCAN, etc.), and machine learning methods (random forest, support vector machine, etc.).
[0127] The data scaling module includes standardization, normalization, decimal scaling normalization, logarithmic transformation, and power transformation.
[0128] The feature extraction and selection module includes statistical feature extraction methods (mean, variance, standard deviation, quantiles, etc.), transformation-based feature extraction methods (Fourier transform, wavelet transform, etc.), model-based feature extraction methods (principal component analysis, autoencoder, etc.), filtering-based feature selection methods (variance thresholding, correlation coefficient, etc.), and model-based feature selection methods (recursive feature elimination, random forest, principal component analysis, etc.).
[0129] The model selection and training module includes linear regression, logistic regression, decision tree, support vector machine, partial least squares, extreme learning machine, neural network, k-nearest neighbor, and naive Bayes methods.
[0130] Module 5: Results Visualization Module displays the results of equipment status monitoring in an intuitive chart format, making it easy for operators and managers to quickly understand the equipment status.
[0131] Visual charts include line charts, bar charts, scatter plots, heatmaps, etc.
[0132] For example, plotting the curves of key equipment parameters over time allows for a visual observation of the trend of parameter changes; plotting the probability distribution of model prediction results allows for an understanding of the likelihood of the equipment being in different states.
[0133] Step 5. Data Acquisition and Equipment Status Monitoring: Sensors collect real-time data on ship equipment, including temperature, voltage, current, pressure, power consumption, battery status, CPU usage, I / O usage, memory usage, meteorological and environmental data, equipment logs, and historical maintenance records, and transmit the data to the data processing center.
[0134] Based on the service life of equipment XX, select an equipment status analysis model, predict the unknown data, and output the current working status of the equipment.
[0135] Step 6. Determine the maintenance plan and carry out maintenance based on the early warning signal. Based on the fault diagnosis results and the level of abnormality, determine the corresponding maintenance plan based on the early warning signal.
[0136] For minor anomalies, remote monitoring and regular inspections can be implemented to increase the frequency of data collection and closely monitor changes in equipment status. For moderate anomalies, technical personnel should be arranged to conduct on-site inspections and simple repairs, such as replacing vulnerable parts and adjusting parameters. For severe anomalies, emergency maintenance procedures should be initiated immediately, and a professional maintenance team should be arranged to carry out comprehensive repairs, including equipment disassembly, inspection, and replacement of core components.
[0137] At the same time, the maintenance plan will be fed back to the relevant maintenance personnel and maintenance records will be kept so that the equipment operating status and maintenance effect can be evaluated and analyzed in the future.
[0138] Specifically, if the prediction of the equipment's operating status is correct, the data is added to the established database, and the model is rebuilt.
[0139] It should be noted that this embodiment has the following beneficial effects: comprehensive data acquisition: various sensors such as temperature, voltage, current, pressure, power consumption, battery status, CPU usage, I / O usage, and memory usage are deployed at key locations of the ship's equipment, as well as meteorological and environmental sensors and spectral analysis sensors, to achieve comprehensive monitoring of the ship's equipment operating status and environmental parameters, providing a rich data foundation for subsequent equipment status analysis and fault diagnosis.
[0140] Precise fault diagnosis: By establishing a benchmark file, real-time collected data is compared and analyzed with benchmark data, and machine learning models are used to diagnose faults in abnormal equipment data. This can accurately determine whether the equipment is operating abnormally and output possible fault types and causes, thus improving the accuracy and timeliness of fault diagnosis.
[0141] Maintenance Records and Evaluation: The maintenance plan is fed back to the relevant maintenance personnel and maintenance records are kept. This facilitates the subsequent evaluation and analysis of the equipment's operating status and maintenance effectiveness, and helps to continuously optimize equipment maintenance strategies, thereby improving equipment reliability and service life.
[0142] Highly adaptable and scalable: Each module includes multiple methods, allowing users to select the appropriate method based on the data distribution characteristics, analysis objectives, and overall results. The methods within each module can be used individually or in combination, demonstrating strong adaptability and flexibility.
[0143] Easy to expand and upgrade: It has good scalability. With the continuous development of sensor technology and the increasing demand for ship equipment monitoring, new sensor types and data processing methods can be easily added to further improve the system's performance and functions.
[0144] In its specific implementation, the method for monitoring the operating status and diagnosing faults of ship equipment provided in the first embodiment of the present invention mainly includes the following steps: Step 1. Deployment of key position sensors In this example, 15 types of sensors were deployed at key locations on the ship's equipment, including temperature sensors, voltage sensors, current sensors, pressure sensors, power consumption sensors, battery status sensors, CPU usage sensors, I / O usage sensors, memory usage sensors, as well as meteorological and environmental sensors (such as wind speed sensors, wind direction sensors, air temperature sensors, humidity sensors, seawater temperature sensors, salinity sensors, etc.). For key high-value modules (such as the inverter IGBT module), additional small LIBS sensors or near-infrared sensors were deployed to collect changes in elemental levels or molecular information at unrelated locations on the circuit board surface.
[0145] Step 2. Determining the Anomaly Level The level of equipment anomaly is determined based on the degree of deviation between real-time data and baseline data.
[0146] The specific classification criteria are as follows: ① Mild anomaly: Data deviation is within 10% of the baseline range. The equipment can still operate normally, but there may be potential hidden dangers.
[0147] ② Moderate anomaly: Data deviation is between 10% and 30% of the baseline range, which affects equipment performance to some extent and may indicate local faults.
[0148] ③ Severe Anomaly: Data deviation exceeds 30% of the baseline range, indicating a significant equipment malfunction that may endanger ship safety, requiring immediate action. Determining the anomaly level provides a basis for subsequent fault diagnosis and maintenance plan development.
[0149] Step 3. Establishing benchmark files and comparing data Establish a baseline document for ship equipment and build an equipment condition monitoring database based on this document. The baseline document contains the operating parameter ranges of the equipment under different cycles and various operating conditions (normal, mild, moderate and severe abnormalities), such as temperature, voltage, current, pressure, power consumption, battery status indicators, upper limit of CPU utilization, upper limit of I / O utilization, upper limit of memory utilization, etc., as well as the corresponding meteorological and environmental conditions, and also includes the elemental composition baseline data of key modules.
[0150] The real-time collected data is compared and analyzed with the data in the benchmark file, and a machine learning model is established to determine whether the device's operating status is abnormal.
[0151] Step 31: If the lifespan of a certain device is 20 years, then the device's data is divided into four periods: initial (7 years), mid-term (8-14 years), late-term (15-20 years), and end-of-life (>20 years). Therefore, the baseline data database contains four sub-databases: the initial database, the mid-term database, and the late-term database.
[0152] Step 32: The data in each sub-database is divided into four categories: normal data, slightly abnormal data, moderately abnormal data, and severely abnormal data, with data labels set as 1, 2, 3, and 4 respectively.
[0153] Step 4. Fault Diagnosis Model Establishment Based on the data in the database, four fault diagnosis models were established (operating condition model in the early stage of equipment use, operating condition model in the middle stage of equipment use, operating condition model in the later stage of equipment use, and operating condition model in the end stage of equipment scrapping).
[0154] The model comprises five modules: data filtering, data scaling, feature extraction and selection, model selection and training, and results visualization. The methods used in each module are determined based on the overall performance of the model.
[0155] To achieve status monitoring and fault diagnosis of XX equipment on ships, the data filtering module uses principal component analysis to remove outliers, and the data scaling module uses decimal scaling normalization to scale data of different dimensions to the same scale.
[0156] The feature extraction and selection module uses the random forest method to calculate the importance of the data, removes data that has a negative effect on the model or does not have any effect, and uses the remaining data as the training set for the model.
[0157] The model selection and training module uses the support vector machine method to model the training set data.
[0158] The ratio of training set, validation set, and test set is 6:3:1.
[0159] The data dimension of the training set samples is N600. 78. The data dimension of the validation set samples is N300. 78. The data dimension of the test set samples is N100. 78.
[0160] The prediction accuracy on the validation set was 93.3%, and on the test set it was 89.0%.
[0161] Step 5. Data Acquisition and Equipment Status Monitoring The sensors collect real-time data on ship equipment, including temperature, voltage, current, pressure, power consumption, battery status, CPU usage, I / O usage, memory usage, meteorological and environmental data, equipment logs, and historical maintenance records.
[0162] Data is measured once every half hour, and 10 data points are collected consecutively. The average value is then output.
[0163] Data matrix Nxxy=N1 165.
[0164] Nxⅹy represents the structure of the constructed data matrix as N1 rows and 165 columns.
[0165] The service life of equipment XX is three years. Therefore, the model built by the primary sub-database of the database is selected to predict the equipment status. The current working status of the equipment is a slightly abnormal condition.
[0166] Step 6. Determine the maintenance plan based on the early warning signal and carry out maintenance. Based on the fault diagnosis results and the level of abnormality, a corresponding maintenance plan is determined based on the early warning signals. For minor abnormal operating conditions, remote monitoring and regular inspections can be adopted to increase the frequency of data collection and closely monitor changes in equipment status.
[0167] Upon verification, the equipment was indeed in a slightly abnormal working state, the prediction result was correct, the data was added to the established database, and the model was rebuilt.
[0168] This embodiment, through the above-described scheme, executes remote monitoring and increases data acquisition frequency when the current working state is at the first preset anomaly level; generates on-site inspection and maintenance instructions when the current working state is at the second preset anomaly level; and initiates an emergency maintenance procedure when the current working state is at the third preset anomaly level. The deviation between the real-time operating data and the baseline data corresponding to the first, second, and third preset anomaly levels increases sequentially. Feedback data after maintenance execution is acquired and fed back to the preset database to update the target fault diagnosis model. This allows for tiered responses based on the deviation between real-time operating data and baseline data, implementing differentiated strategies such as remote monitoring, on-site inspection, or emergency maintenance for different anomaly levels. This ensures that maintenance resource investment matches the severity of fault risk, avoiding over-maintenance for minor faults or delayed response to serious faults. Furthermore, the feedback data feedback mechanism after maintenance execution enables closed-loop iteration and self-updating of the diagnostic model, ensuring the system adapts to dynamic changes in equipment status and improving the intelligence level and long-term reliability of ship equipment operation and maintenance.
[0169] Accordingly, the present invention further provides a device for monitoring and diagnosing the operating status of ship equipment.
[0170] Reference Figure 7 , Figure 7 This is a functional block diagram of the first embodiment of the ship equipment operation status monitoring and diagnostic device of the present invention.
[0171] In the first embodiment of the ship equipment operation status monitoring and diagnostic device of the present invention, the ship equipment operation status monitoring and diagnostic device includes: The life cycle determination module 10 is used to acquire real-time operating data and service life of ship equipment, and determine the current life cycle stage of the ship equipment based on the service life of the equipment.
[0172] The fault diagnosis module 20 is used to select the corresponding target fault diagnosis model from the preset database according to the life cycle stage, and input the real-time operation data into the target fault diagnosis model to parse and obtain the current working status of the ship equipment.
[0173] The model update module 30 is used to formulate a maintenance strategy based on the current working status and to feed back the feedback data after maintenance execution to the preset database to update the target fault diagnosis model.
[0174] The life cycle determination module 10 is also used to collect data through a set of sensors deployed at key locations of the ship's equipment, and to take the average value of the sensor data collected multiple times as the real-time operating data. The sensor set includes conventional physical quantity sensors and spectral analysis sensors. The spectral analysis sensors are used to collect data on changes in elemental composition or molecular information on the surface of the key modules. The module also obtains the current service life of the ship's equipment, compares the current service life with the preset total service life, and determines the current life cycle stage of the ship's equipment based on the comparison result.
[0175] The lifecycle determination module 10 is further configured to obtain the current service life of the ship's equipment, compare the current service life with the preset total service life L, and generate a comparison result; when the comparison result shows that the current service life is between 0 and the preset total service life L / 3 years, the ship's equipment is determined to be in the initial stage of its lifecycle; when the comparison result shows that the current service life is between L / 3 and 2L / 3 years, the ship's equipment is determined to be in the middle stage of its lifecycle; when the comparison result shows that the current service life is between 2L / 3 and L years, the ship's equipment is determined to be in the later stage of its lifecycle; and when the comparison result shows that the current service life is greater than L years, the ship's equipment is determined to be in the scrapping stage.
[0176] The fault diagnosis module 20 is further configured to select a target sub-database and a target fault diagnosis model that match the life cycle stage from a preset database, wherein the target sub-database stores the standard operating parameter range and historical fault data corresponding to the life cycle stage; after inputting the real-time operating data into the target fault diagnosis model, the target fault diagnosis model parses the real-time operating data to obtain the current working status of the ship equipment.
[0177] The fault diagnosis module 20 is further configured to input the real-time operating data into the target fault diagnosis model, and then use the target fault diagnosis model to sequentially remove outliers, eliminate dimensional differences, and extract state features from the real-time operating data; and then compare and analyze the processed feature data with the benchmark data in the target sub-database to obtain the current working status of the ship equipment.
[0178] The model update module 30 is further configured to: execute remote monitoring and increase data acquisition frequency when the current working state is at a first preset anomaly level; generate on-site inspection and maintenance instructions when the current working state is at a second preset anomaly level; and initiate an emergency maintenance procedure when the current working state is at a third preset anomaly level. The deviation between the real-time operating data and the baseline data corresponding to the first, second, and third preset anomaly levels increases sequentially. The module also acquires feedback data after maintenance execution and feeds the feedback data back to the preset database to update the target fault diagnosis model.
[0179] The model update module 30 is further configured to acquire actual fault record data after maintenance execution as the feedback data; compare the feedback data with the diagnostic results output by the target fault diagnosis model; when the feedback data is consistent with the diagnostic results, add the feedback data to the preset database, and retrain the target fault diagnosis model based on the updated preset database to obtain the updated target fault diagnosis model.
[0180] The steps for implementing each functional module of the ship equipment operation status monitoring and diagnosis device can be referred to in the various embodiments of the ship equipment operation status monitoring and diagnosis method of the present invention, and will not be repeated here.
[0181] Furthermore, this embodiment of the invention also proposes a storage medium storing a ship equipment operating status monitoring and diagnostic program. When the ship equipment operating status monitoring and diagnostic program is executed by a processor, it performs the following operations: Acquire real-time operating data and service life of ship equipment, and determine the current life cycle stage of the ship equipment based on the service life of the equipment; Based on the life cycle stage, a corresponding target fault diagnosis model is selected from a pre-set database, and the real-time operating data is input into the target fault diagnosis model to parse and obtain the current working status of the ship equipment; A maintenance strategy is formulated based on the current working status, and the feedback data after maintenance execution is fed back to the preset database to update the target fault diagnosis model.
[0182] Furthermore, when the ship equipment operation status monitoring and diagnostic program is executed by the processor, it also performs the following operations: Data is collected by a set of sensors deployed at key locations on the ship's equipment. The average value of the sensor data collected multiple times is taken as the real-time operating data. The sensor set includes conventional physical quantity sensors and spectral analysis sensors. The spectral analysis sensors are used to collect data on changes in elemental composition or molecular information on the surface of key modules. The current service life of the ship's equipment is obtained, and the current service life is compared with the preset total service life. Based on the comparison result, the current life cycle stage of the ship's equipment is determined.
[0183] Furthermore, when the ship equipment operation status monitoring and diagnostic program is executed by the processor, it also performs the following operations: Obtain the current service life of the ship's equipment, compare the current service life with the preset total service life L, and generate a comparison result; When the comparison result shows that the current service life is between 0 and the preset total service life L / 3 years, the current life cycle stage of the ship equipment is determined to be the initial stage. When the comparison result indicates that the current service life is between L / 3 and 2L / 3 years, the current life cycle stage of the ship equipment is determined to be mid-term. When the comparison result indicates that the current service life is between 2L / 3 and L years, the current life cycle stage of the ship equipment is determined to be the late stage. When the comparison result indicates that the current service life is greater than L years, the current life cycle stage of the ship equipment is determined to be the scrapping period.
[0184] Furthermore, when the ship equipment operation status monitoring and diagnostic program is executed by the processor, it also performs the following operations: Based on the life cycle stage, a target sub-database and a target fault diagnosis model matching the life cycle stage are selected from a preset database. The target sub-database stores the standard operating parameter range and historical fault data corresponding to the life cycle stage. After the real-time operating data is input into the target fault diagnosis model, the target fault diagnosis model parses the real-time operating data to obtain the current operating status of the ship's equipment.
[0185] Furthermore, when the ship equipment operation status monitoring and diagnostic program is executed by the processor, it also performs the following operations: After the real-time operation data is input into the target fault diagnosis model, the target fault diagnosis model sequentially removes outliers, eliminates dimensional differences, and extracts state features from the real-time operation data. The current working status of the ship's equipment is obtained by comparing and analyzing the processed feature data with the benchmark data in the target sub-database.
[0186] Furthermore, when the ship equipment operation status monitoring and diagnostic program is executed by the processor, it also performs the following operations: When the current working state is at the first preset abnormal level, remote monitoring is performed and the data collection frequency is increased; When the current working status is at the second preset abnormal level, an on-site inspection and maintenance instruction is generated. When the current working state is at the third preset abnormality level, the emergency maintenance procedure is initiated; wherein, the deviation between the real-time operating data and the benchmark data corresponding to the first preset abnormality level, the second preset abnormality level and the third preset abnormality level increases sequentially. Obtain feedback data after maintenance execution, and feed the feedback data back to the preset database to update the target fault diagnosis model.
[0187] Furthermore, when the ship equipment operation status monitoring and diagnostic program is executed by the processor, it also performs the following operations: The actual fault record data after maintenance is performed is obtained as the feedback data; The feedback data is compared with the diagnostic results output by the target fault diagnosis model; When the feedback data matches the diagnostic result, the feedback data is added to the preset database, and the target fault diagnosis model is retrained based on the updated preset database to obtain the updated target fault diagnosis model.
[0188] Those skilled in the art will understand that all or part of the steps in the methods described above can be implemented by a program instructing related hardware. The program is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium is a computer-readable storage medium, including: USB flash drive, mobile hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, and other media that can store program code.
[0189] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.
[0190] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.
[0191] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.
Claims
1. A method for monitoring and diagnosing the operational status of ship equipment, characterized in that, The method for monitoring and diagnosing the operational status of ship equipment includes: Acquire real-time operating data and service life of ship equipment, and determine the current life cycle stage of the ship equipment based on the service life of the equipment; Based on the life cycle stage, a corresponding target fault diagnosis model is selected from a pre-set database, and the real-time operating data is input into the target fault diagnosis model to parse and obtain the current working status of the ship equipment; A maintenance strategy is formulated based on the current working status, and the feedback data after maintenance execution is fed back to the preset database to update the target fault diagnosis model.
2. The method for monitoring and diagnosing the operating status of ship equipment as described in claim 1, characterized in that, The acquisition of real-time operating data and service life of ship equipment, and the determination of the current life cycle stage of the ship equipment based on the service life, includes: Data is collected by a set of sensors deployed at key locations on the ship's equipment. The average value of the sensor data collected multiple times is taken as the real-time operating data. The sensor set includes conventional physical quantity sensors and spectral analysis sensors. The spectral analysis sensors are used to collect data on changes in elemental composition or molecular information on the surface of key modules. The current service life of the ship's equipment is obtained, and the current service life is compared with the preset total service life. Based on the comparison result, the current life cycle stage of the ship's equipment is determined.
3. The method for monitoring and diagnosing the operating status of ship equipment as described in claim 2, characterized in that, The process of obtaining the current service life of the ship's equipment involves comparing the current service life with a preset total service life, and determining the current life cycle stage of the ship's equipment based on the comparison result, including: Obtain the current service life of the ship's equipment, compare the current service life with the preset total service life L, and generate a comparison result; When the comparison result shows that the current service life is between 0 and the preset total service life L / 3 years, the current life cycle stage of the ship equipment is determined to be the initial stage. When the comparison result indicates that the current service life is between L / 3 and 2L / 3 years, the current life cycle stage of the ship equipment is determined to be mid-term. When the comparison result indicates that the current service life is between 2L / 3 and L years, the current life cycle stage of the ship equipment is determined to be the late stage. When the comparison result indicates that the current service life is greater than L years, the current life cycle stage of the ship equipment is determined to be the scrapping period.
4. The method for monitoring and diagnosing the operating status of ship equipment as described in claim 1, characterized in that, The step of selecting a corresponding target fault diagnosis model from a pre-set database based on the lifecycle stage, and inputting the real-time operating data into the target fault diagnosis model to parse and obtain the current operating status of the ship's equipment includes: Based on the life cycle stage, a target sub-database and a target fault diagnosis model matching the life cycle stage are selected from a preset database. The target sub-database stores the standard operating parameter range and historical fault data corresponding to the life cycle stage. After the real-time operating data is input into the target fault diagnosis model, the target fault diagnosis model parses the real-time operating data to obtain the current operating status of the ship's equipment.
5. The method for monitoring and diagnosing the operating status of ship equipment as described in claim 4, characterized in that, The step of inputting the real-time operating data into the target fault diagnosis model, and then parsing the real-time operating data through the target fault diagnosis model to obtain the current operating status of the ship's equipment includes: After the real-time operation data is input into the target fault diagnosis model, the target fault diagnosis model sequentially removes outliers, eliminates dimensional differences, and extracts state features from the real-time operation data. The current working status of the ship's equipment is obtained by comparing and analyzing the processed feature data with the benchmark data in the target sub-database.
6. The method for monitoring and diagnosing the operating status of ship equipment as described in claim 1, characterized in that, The step of formulating a maintenance strategy based on the current working status and feeding back the feedback data after maintenance execution to the preset database to update the target fault diagnosis model includes: When the current working state is at the first preset abnormal level, remote monitoring is performed and the data collection frequency is increased; When the current working status is at the second preset abnormal level, an on-site inspection and maintenance instruction is generated. When the current working state is at the third preset abnormality level, the emergency maintenance procedure is initiated; wherein, the deviation between the real-time operating data and the benchmark data corresponding to the first preset abnormality level, the second preset abnormality level and the third preset abnormality level increases sequentially. Obtain feedback data after maintenance execution, and feed the feedback data back to the preset database to update the target fault diagnosis model.
7. The method for monitoring and diagnosing the operating status of ship equipment as described in claim 6, characterized in that, The step of obtaining feedback data after maintenance execution and feeding the feedback data back to the preset database to update the target fault diagnosis model includes: The actual fault record data after maintenance is performed is obtained as the feedback data; The feedback data is compared with the diagnostic results output by the target fault diagnosis model; When the feedback data matches the diagnostic result, the feedback data is added to the preset database, and the target fault diagnosis model is retrained based on the updated preset database to obtain the updated target fault diagnosis model.
8. A device for monitoring and diagnosing the operating status of ship equipment, characterized in that, The ship equipment operation status monitoring and diagnostic device includes: The life cycle determination module is used to acquire real-time operating data and service life of ship equipment, and determine the current life cycle stage of the ship equipment based on the service life of the equipment. The fault diagnosis module is used to select the corresponding target fault diagnosis model from the preset database according to the life cycle stage, and input the real-time operation data into the target fault diagnosis model to parse and obtain the current working status of the ship equipment; The model update module is used to formulate a maintenance strategy based on the current working status and to feed back the feedback data after maintenance execution to the preset database to update the target fault diagnosis model.
9. A ship equipment operation status monitoring and diagnostic device, characterized in that, The ship equipment operation status monitoring and diagnostic device includes: a memory, a processor, and a ship equipment operation status monitoring and diagnostic program stored in the memory and executable on the processor. The ship equipment operation status monitoring and diagnostic program is configured to implement the steps of the ship equipment operation status monitoring and diagnostic method as described in any one of claims 1 to 7.
10. A storage medium, characterized in that, The storage medium stores a ship equipment operation status monitoring and diagnostic program, which, when executed by a processor, implements the steps of the ship equipment operation status monitoring and diagnostic method as described in any one of claims 1 to 7.
