A ship maintenance management system

Abstract

The application belongs to the technical field of maintenance management systems, and particularly relates to a ship maintenance management system, which comprises an edge layer, a transmission layer, a platform layer and an application layer which are sequentially connected in communication, the edge layer is used for global real-time collection and preprocessing of ship related data, the transmission layer realizes stable and safe data transmission, the platform layer performs data deep analysis, fault prediction and digital twin mapping, and the application layer provides multi-scenario intelligent maintenance management services which are presented through multiple terminals.

Description

Technical Field

[0001] This invention belongs to the technical field of maintenance management systems, and particularly relates to a ship maintenance management system. Background Technology

[0002] Ships navigate in complex and ever-changing environments, encompassing nearshore, offshore, and port scenarios. The operational status of critical equipment (such as main engines, auxiliary engines, and navigation equipment) and the hull structure directly impacts navigation safety. Traditional ship maintenance management relies heavily on planned or reactive maintenance models, lacking the ability to collect real-time data on equipment operation, environmental parameters, and the health status of the hull structure. It primarily depends on manual inspections to monitor equipment status, which is not only inefficient and prone to errors but also fails to detect potential hidden faults, often leading to delayed fault warnings, unplanned downtime, and even safety accidents. Furthermore, it results in excessively high maintenance costs and wasted spare parts resources.

[0003] Existing ship maintenance systems have significant technical shortcomings: In terms of data transmission, single communication links are insufficient to meet the transmission needs of different navigation scenarios, bandwidth is limited in the open sea, data transmission efficiency is insufficient in near-shore and port areas, and data security protection is weak; In terms of data processing and application, there is a lack of deep integration of core technologies such as big data analysis, AI algorithms, and digital twins, making it impossible to achieve accurate fault prediction and intelligent diagnosis, maintenance plan formulation is highly subjective and lacks data support, and functions such as spare parts management and remote collaborative maintenance are incomplete, making it difficult to meet the intelligent and efficient maintenance management needs of modern ships and unable to adapt to the maintenance management scenarios of the entire ship life cycle. Summary of the Invention

[0004] The purpose of this invention is to provide a ship repair and management system to address the aforementioned technical problems.

[0005] In view of this, the present invention provides a ship maintenance management system, comprising an edge layer, a transmission layer, a platform layer and an application layer connected in sequence. The edge layer is used for real-time acquisition and preprocessing of ship-related data across the entire domain. The transmission layer enables stable and secure data transmission. The platform layer performs in-depth data analysis, fault prediction and digital twin mapping. The application layer provides intelligent maintenance management services for multiple scenarios, presented through multiple terminals.

[0006] Preferably, the edge layer deploys multiple types of intelligent terminals, including a multi-dimensional sensor network, an intelligent edge gateway, and a visual acquisition terminal. The multi-dimensional sensor network deploys corresponding types of sensors in key ship equipment, key hull structures, and engine and cargo hold areas to collect equipment operating parameters, hull structure data, and environmental safety parameters. The intelligent edge gateway aggregates and preprocesses data by region, supports multiple communication protocols, provides local alarms, and encrypts data transmission. The visual acquisition terminal is deployed in key areas to collect visual images of equipment operation and preliminarily identify anomalies.

[0007] Preferably, the transmission layer adopts a multi-link redundancy design of satellite communication, maritime broadband, and local area network. In the far sea area, maritime satellite is used as the core for data transmission, and LZ77 and Huffman compression algorithms are used to prioritize the transmission of key data. In the near sea and port areas, the transmission layer switches to maritime broadband or 5G network. The transmission layer ensures data transmission security through VPN encrypted tunnel, AES-256 encryption algorithm and device identity authentication system.

[0008] Preferably, the big data processing engine of the platform layer uses Hadoop and Spark architecture to build a distributed data storage cluster to store massive amounts of data such as the entire life cycle operation data and maintenance records of ship equipment, and forms a standardized dataset through data cleaning, fusion and labeling.

[0009] Preferably, the multimodal AI algorithm engine of the platform layer integrates fault prediction algorithm, anomaly diagnosis algorithm and maintenance decision algorithm; the fault prediction algorithm is based on LSTM and GRU deep learning algorithms, the anomaly diagnosis algorithm adopts random forest and SVM machine learning algorithms combined with a rule engine, and the maintenance decision algorithm combines multi-dimensional factors to generate the optimal maintenance plan.

[0010] Preferably, the digital twin modeling engine of the platform layer constructs a full-dimensional digital twin model covering the hull structure, equipment layout, pipeline routing and electrical system, realizing real-time mapping between the physical ship and the digital model, and supporting fault visualization and location and maintenance process simulation.

[0011] Preferably, the device health management module of the application layer combines edge layer data collection with AI fault prediction model to generate a device health score of 0-100 in real time. If the score is lower than the threshold, an early warning is automatically triggered. Furthermore, for key devices, a health report is generated that includes operating parameter trends, potential fault risks, and remaining service life.

[0012] Preferably, the intelligent fault diagnosis module of the application layer adopts a multi-dimensional diagnosis method of data, vision, and rules, automatically calls the abnormal diagnosis algorithm in combination with the equipment fault knowledge base, analyzes the cause of the fault and locates the location, generates a diagnostic report containing fault description and emergency handling suggestions, and supports manual uploading of fault phenomena and provides intelligent diagnostic suggestions.

[0013] Preferably, the predictive maintenance planning module of the application layer intelligently generates a personalized maintenance plan based on the equipment health status, fault prediction results, and navigation plan, combined with the fault prediction time, ship navigation route, and maintenance resources. The plan can be dynamically adjusted and supports reminder functions.

[0014] Preferably, the application layer includes a spare parts intelligent management module and a remote collaborative maintenance module; the spare parts intelligent management module uses RFID technology to identify spare parts, enabling warehousing, inventory monitoring, intelligent early warning, and demand forecasting; the remote collaborative maintenance module supports crew members initiating remote maintenance requests, and shore-based experts provide remote guidance for maintenance through voice / video communication combined with a digital twin model, and supports screen recording and archiving of the maintenance process.

[0015] The beneficial effects of this invention are: Through the full-domain deployment of multiple types of intelligent terminals at the edge layer and edge computing capabilities, comprehensive perception and preprocessing of ship equipment, hull structure, and environmental parameters are achieved. Combined with multi-link redundancy design and security encryption mechanisms at the transmission layer, the problems of unstable data transmission and insufficient security in different navigation scenarios are completely solved. At the same time, the big data processing, AI algorithms, and digital twin core technologies integrated at the platform layer break through the limitations of data dispersion and insufficient analysis depth in traditional maintenance management, allowing data to form a closed loop from collection to processing, providing accurate and comprehensive technical support for maintenance management.

[0016] Through the collaborative operation of multiple modules at the application layer, the system promotes the transformation of ship maintenance management from traditional planned and reactive maintenance to predictive maintenance, effectively avoiding potential failure risks and improving navigation safety. At the same time, it standardizes maintenance processes, optimizes spare parts management, and enhances remote collaboration efficiency, significantly reducing maintenance costs and resource waste. It meets the diverse needs of different users such as crew members and shore-based management personnel, realizes intelligent and efficient management of the entire life cycle of ship maintenance, and significantly improves the overall operation and management level. Attached Figure Description

[0017] Figure 1 This is a system diagram of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0019] The edge layer is the core of the system's data source. By deploying multiple types of intelligent terminals, it enables the full-domain, real-time collection of data such as the operating status of ship equipment, environmental parameters, and the health of the ship's structure. At the same time, it has edge computing capabilities to reduce transmission pressure.

[0020] Multi-dimensional sensor network deployment: High-precision sensors such as vibration sensors, temperature sensors, pressure sensors, oil sensors, and current and voltage sensors are deployed on key equipment such as ship main engines, auxiliary engines, propulsion systems, navigation equipment, and power generation equipment to collect equipment operating parameters in real time (such as vibration frequency, temperature changes, oil pressure, oil impurity content, current fluctuations, etc.); stress sensors and strain sensors are deployed on key hull structures (such as decks, hull welds, bulkheads, etc.) to monitor hull structural deformation and stress conditions; temperature and humidity sensors, smoke sensors, and gas sensors (such as combustible gases and toxic gases) are deployed in areas such as engine rooms and cargo holds to monitor environmental safety parameters.

[0021] Intelligent Edge Gateway: Deploy an intelligent edge gateway in each area to aggregate sensor data from that area. Use edge computing technology to preprocess the data (such as data cleaning, outlier filtering, and data format conversion), prioritize the identification of serious abnormal data and issue local alarms, and encrypt and transmit the preprocessed data to the platform layer. The gateway supports multiple communication protocols (such as Modbus, CAN, TCP / IP) to adapt to different types of devices and sensors, ensuring compatibility.

[0022] Visual acquisition terminal: High-definition cameras and AI visual analysis modules are deployed in key areas such as the engine room, driver's cab, and cargo hold to collect real-time visual images of equipment operation (such as equipment surface wear, oil and water leaks, and instrument pointer positions). Anomalies (such as oil leaks and abnormal instrument readings) are initially identified through edge visual analysis.

[0023] The transport layer is responsible for data transmission between the edge layer and the platform layer. It adopts a multi-link redundancy design of "satellite communication, maritime broadband, and local area network" to ensure the stability and security of data transmission in different navigation scenarios (near sea, open sea, and port).

[0024] The long-range communication link is based on satellite communication and uses maritime satellites (such as Iridium and Inmarsat) to realize data transmission in the long-range area. To address the problem of limited satellite bandwidth, the transmitted data is compressed and encoded (using compression algorithms such as LZ77 and Huffman). Priority is given to transmitting critical abnormal data and core operational data, while non-critical data is cached to the edge gateway and retransmitted after entering the near-shore area.

[0025] Offshore and port communication links: Upon entering offshore or port areas, the system automatically switches to maritime broadband or 5G networks to enable high-speed transmission of large amounts of data (such as high-definition video and complete equipment operation logs), thereby improving data synchronization efficiency.

[0026] Security encryption mechanism: VPN encrypted tunnel and AES-256 encryption algorithm are used to encrypt the transmitted data throughout the process to prevent data theft or tampering; a device identity authentication system is established so that only authorized edge terminals can access the transmission network to prevent unauthorized access.

[0027] The platform layer is the core hub of the system, integrating core capabilities such as big data processing, AI algorithm engine, and digital twin modeling to achieve in-depth data analysis, fault prediction, maintenance decision-making, and digital twin mapping.

[0028] Big Data Processing Engine: Constructs a distributed data storage cluster (using Hadoop and Spark architecture) to store massive amounts of data such as the entire lifecycle operation data of ship equipment, maintenance records, spare parts data, and crew operation logs; through data cleaning, data fusion, and data labeling, it forms standardized datasets to provide data support for AI analysis and decision-making.

[0029] Multimodal AI algorithm engine: Integrates multiple AI algorithms such as fault prediction algorithm, anomaly diagnosis algorithm, and maintenance decision algorithm to achieve accurate prediction and intelligent diagnosis of equipment faults.

[0030] Fault prediction algorithm: Based on deep learning algorithms such as LSTM (Long Short-Term Memory Network) and GRU (Gated Recurrent Unit), combined with historical equipment operation data, maintenance records and environmental parameters, a fault prediction model is built to accurately predict the type, time and severity of possible equipment failures (such as predicting the remaining service life of the main bearing and the probability of generator failure). Anomaly diagnosis algorithm: It uses machine learning (such as random forest and SVM) combined with a rule engine to analyze real-time collected equipment operation data and visual data, quickly identify anomalies (such as identifying equipment imbalance faults through vibration data and identifying oil leakage anomalies through visual data), and locate the fault location and cause. Maintenance decision algorithm: Combining multiple factors such as fault type, equipment importance, voyage plan, spare parts inventory, and maintenance cost, it intelligently generates the optimal maintenance plan (such as prioritizing the maintenance of equipment that affects navigation safety, selecting the lowest cost maintenance method, and matching the most suitable maintenance personnel).

[0031] Digital Twin Modeling Engine: Constructs a full-dimensional digital twin model of the ship, covering all elements such as hull structure, equipment layout, pipeline routing, and electrical systems, achieving real-time mapping between the physical ship and the digital model. By integrating real-time data collected at the edge layer into the digital twin model, it can intuitively display the operating status of equipment and the health of the hull structure, supporting fault visualization and location (such as highlighting faulty equipment and its location in the digital model) and maintenance process simulation (such as simulating spare parts replacement procedures and optimizing maintenance steps).

[0032] At the application level, it provides intelligent application services for various users such as ship crew members, shore-based management personnel, maintenance teams, and spare parts suppliers, and presents them through multiple terminals such as Web, mobile (APP) and ship local terminal to meet the usage needs in different scenarios.

[0033] The equipment health management module monitors equipment operating status in real time, enabling equipment health status assessment and fault prediction. By collecting equipment operating data from the edge layer and combining it with an AI fault prediction model, it generates a real-time equipment health score (0-100 points). Scores below a threshold automatically trigger warnings. For critical equipment, it generates health reports (including operating parameter trends, potential fault risks, and remaining service life), providing a basis for maintenance decisions. For example, when the main unit's vibration data exceeds the normal range, the system uses AI algorithms to analyze and determine that it indicates abnormal bearing wear, predicting a remaining service life of 200 hours, immediately triggering a warning and pushing it to crew and shore-based management personnel.

[0034] The intelligent fault diagnosis module employs a multi-dimensional diagnostic approach, combining data, vision, and rules to achieve rapid fault location and accurate diagnosis. When the system detects abnormal data or visual anomalies, it automatically invokes the anomaly diagnosis algorithm, combining it with the equipment fault knowledge base (including historical fault cases and maintenance manuals) to analyze the cause of the fault, locate its position, and generate a detailed diagnostic report (including fault description, cause analysis, scope of impact, and emergency handling suggestions). It also supports manual uploading of fault symptoms, providing intelligent diagnostic suggestions to assist crew members in quickly handling the fault.

[0035] The predictive maintenance planning module replaces traditional planned maintenance, intelligently generating personalized maintenance plans based on equipment health status, fault prediction results, and navigation plans. The system automatically plans maintenance time, content, required spare parts, and personnel by combining factors such as equipment fault prediction time, ship navigation route (e.g., proximity to port, availability of maintenance ports), and maintenance resources (maintenance personnel, spare parts). Maintenance plans can be dynamically adjusted; if equipment health improves or an emergency fault occurs, the maintenance plan is automatically optimized. Maintenance plan reminders (mobile app push notifications, local terminal alarms) are supported to ensure timely execution of maintenance work.

[0036] The digital twin visualization module, based on the ship's digital twin model, enables the visualization of equipment operating status, fault information, and maintenance progress. Crew members and shore-based management personnel can view the ship's 3D model through terminals, and click on equipment to view detailed operating parameters, health scores, and historical maintenance records. When a fault occurs, the corresponding equipment in the digital twin model is automatically highlighted, and the fault type and location are marked. It supports visualized monitoring of the maintenance process, transmitting maintenance footage in real time via camera, and combining the maintenance steps with the digital twin model to assist in remote maintenance guidance.

[0037] The intelligent spare parts management module enables full lifecycle management of spare parts, including warehousing, inventory monitoring, intelligent early warning, demand forecasting, and procurement management. The system uses RFID technology to identify spare parts, automatically recording information (model, quantity, expiration date, and storage location) upon receipt. It monitors spare parts inventory in real time, automatically triggering warnings when the quantity falls below a safety threshold or approaches its expiration date. Combining equipment failure prediction results and historical spare parts consumption data, it uses AI algorithms to predict future spare parts demand and generate procurement suggestions. It supports barcode scanning registration for spare parts entering and leaving the warehouse, automatically updating inventory data and avoiding errors from manual statistics.

[0038] The maintenance process management module standardizes maintenance procedures and enables full-process management of maintenance tasks, including task allocation, progress tracking, quality acceptance, and cost statistics. The system automatically assigns maintenance tasks to corresponding crew members or maintenance teams based on the maintenance plan, clearly defining the maintenance time, content, and requirements. Maintenance personnel provide real-time progress updates (e.g., pending maintenance, in progress, completed) via a mobile app. After maintenance is completed, maintenance records (including maintenance steps, replaced parts, and maintenance photos) are uploaded for quality acceptance by management personnel. The system automatically calculates maintenance costs (spare parts costs, labor costs) and generates a maintenance cost report.

[0039] The remote collaborative maintenance module addresses the issue of insufficient maintenance technology when ships are sailing at sea, enabling remote collaborative maintenance between shore-based experts and crew. Crew members can initiate remote maintenance requests via a mobile app or the ship's local terminal, uploading fault diagnosis reports, equipment operating data, and on-site videos. Shore-based experts can view relevant information through the system, communicate with crew members in real time via voice / video, and provide remote guidance on maintenance procedures using a digital twin model. The system also supports screen recording and archiving of the maintenance process, providing materials for subsequent maintenance recovery and personnel training.

[0040] The data analysis and reporting module performs multi-dimensional analysis of ship equipment operation data, maintenance data, and spare parts data, generating various statistical reports to provide data support for management decisions. It supports equipment failure rate analysis, maintenance cost analysis, spare parts consumption analysis, and crew maintenance efficiency analysis; reports can be customized (daily, weekly, monthly reports) and displayed intuitively in chart format (line graphs, bar charts, pie charts); data export is supported for convenient in-depth analysis by management personnel.

[0041] The embodiments of this application have been described above with reference to the accompanying drawings. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. This application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A ship repair management system, characterized in that: It includes an edge layer, a transmission layer, a platform layer, and an application layer that are connected in sequence. The edge layer is used for real-time acquisition and preprocessing of ship-related data across the entire domain. The transmission layer enables stable and secure data transmission. The platform layer performs in-depth data analysis, fault prediction, and digital twin mapping. The application layer provides intelligent maintenance management services for multiple scenarios, presented through multiple terminals.

2. The ship repair management system according to claim 1, characterized in that: The edge layer deploys multiple types of intelligent terminals, including a multi-dimensional sensor network, an intelligent edge gateway, and a visual acquisition terminal. The multi-dimensional sensor network deploys corresponding types of sensors in key ship equipment, key hull structures, and engine and cargo hold areas to collect equipment operating parameters, hull structure data, and environmental safety parameters. The intelligent edge gateway aggregates and preprocesses data by region, supports multiple communication protocols, provides local alarms, and encrypts data transmission. The visual acquisition terminal is deployed in key areas to collect visual images of equipment operation and preliminarily identify anomalies.

3. The ship repair management system according to claim 2, characterized in that: The transmission layer adopts a multi-link redundancy design of satellite communication, maritime broadband, and local area network. In the far sea area, maritime satellite is used as the core for data transmission, and LZ77 and Huffman compression algorithms are used to prioritize the transmission of critical data. In the near sea and port areas, the transmission layer switches to maritime broadband or 5G network. The transmission layer ensures data transmission security through VPN encrypted tunnel, AES-256 encryption algorithm and device identity authentication system.

4. The ship repair management system according to claim 3, characterized in that: The big data processing engine of the platform layer uses Hadoop and Spark architecture to build a distributed data storage cluster, storing massive amounts of data such as the entire life cycle operation data and maintenance records of ship equipment, and forming standardized datasets through data cleaning, fusion and labeling.

5. A ship repair management system according to claim 4, characterized in that: The platform layer's multimodal AI algorithm engine integrates fault prediction algorithms, anomaly diagnosis algorithms, and maintenance decision-making algorithms. The fault prediction algorithm is based on LSTM and GRU deep learning algorithms, the anomaly diagnosis algorithm uses random forest and SVM machine learning algorithms combined with a rule engine, and the maintenance decision-making algorithm combines multi-dimensional factors to generate the optimal maintenance plan.

6. A ship repair management system according to claim 5, characterized in that: The platform layer's digital twin modeling engine constructs a full-dimensional digital twin model covering hull structure, equipment layout, pipeline routing, and electrical systems, enabling real-time mapping between the physical ship and the digital model, and supporting visualized fault location and maintenance process simulation.

7. A ship repair management system according to claim 6, characterized in that: The device health management module in the application layer combines edge layer data collection with AI fault prediction models to generate a device health score of 0-100 in real time. If the score is lower than the threshold, an early warning is automatically triggered. Furthermore, for critical equipment, a health report is generated that includes trends in operating parameters, potential fault risks, and remaining service life.

8. A ship repair management system according to claim 7, characterized in that: The intelligent fault diagnosis module in the application layer adopts a multi-dimensional diagnosis approach that combines data, vision, and rules. It automatically calls anomaly diagnosis algorithms in conjunction with the equipment fault knowledge base to analyze the cause of the fault and locate its position, generating a diagnostic report that includes a fault description and emergency handling suggestions. It also supports manual uploading of fault phenomena and provides intelligent diagnostic suggestions.

9. A ship repair management system according to claim 8, characterized in that: The predictive maintenance planning module of the application layer intelligently generates personalized maintenance plans based on equipment health status, fault prediction results, and navigation plans, combined with fault prediction time, ship navigation route, and maintenance resources. The plans can be dynamically adjusted and support reminder functions.

10. A ship repair management system according to claim 1, characterized in that: The application layer includes a spare parts intelligent management module and a remote collaborative maintenance module; the spare parts intelligent management module uses RFID technology to identify spare parts, realizing warehousing, inventory monitoring, intelligent early warning and demand forecasting. The remote collaborative maintenance module supports crew members initiating remote maintenance requests, and shore-based experts can remotely guide the maintenance through voice / video communication combined with a digital twin model, and the maintenance process can be recorded and archived.

Images