Ship comprehensive energy efficiency improving platform and data management and application method thereof

Abstract

The invention provides a ship comprehensive energy efficiency improvement platform and a data management and application method thereof, and belongs to the technical field of ship comprehensive energy efficiency optimization and big data. The ship comprehensive energy efficiency improvement platform comprises a ship end, a cloud end and a shore end, wherein the ship end realizes equipment data acquisition, edge processing and scheduling through a data acquisition module, an edge calculation module and a data scheduling module; the cloud end completes data aggregation, isolation and distributed storage through the data aggregation module, the data isolation module and the distributed storage module; and the shore end realizes data standard management, algorithm version control and micro-service business application through the data center layer, the algorithm center layer and the business center layer. The ship energy efficiency comprehensive improvement platform provides comprehensive management and optimization for various energy-saving devices and various ship energy efficiency improvement software, and optimization and improvement of ship comprehensive energy efficiency are achieved.

Description

Technical Field

[0001] The present application relates to the field of ship energy efficiency optimization and big data technology, and in particular to a ship comprehensive energy efficiency improvement platform and its data management and application methods. Background Art

[0002] The International Maritime Organization (IMO)'s 2023 Ship Greenhouse Gas Emissions Reduction Strategy calls for the maritime industry to achieve peak emissions as soon as possible and net zero emissions by 2050. Currently, ships are required to meet stringent energy efficiency regulations, including EEDI (for newbuildings), EEXI, and CII (for existing ships), yet over 30% of the global fleet still fails to meet these standards. Despite the widespread adoption of energy-saving technologies such as shaft-driven power generation and wind-aided navigation, their effectiveness is limited by the vessel and navigation conditions, resulting in limited energy efficiency gains. Furthermore, ship energy efficiency data is fragmented and isolated, and its management is not standardized, making it difficult to effectively integrate and analyze, hindering comprehensive energy efficiency optimization. Summary of the Invention

[0003] This application provides a ship comprehensive energy efficiency improvement platform and its data management and application method, aiming to solve the problems of limited ship energy efficiency improvement, data silos and insufficient energy efficiency optimization of energy-saving equipment.

[0004] First, this application provides a comprehensive ship energy efficiency improvement platform, including ship-side, cloud-side, and shore-side:

[0005] The ship side includes a data acquisition module, an edge computing module, and a data scheduling module; wherein the data acquisition module is used to realize multi-source acquisition of equipment data using multiple industrial protocols and communication protocols; the edge computing module is used to perform data processing based on edge computing; the data scheduling module is used to complete the scheduling and forwarding of edge data through a data bus based on data communication middleware and edge database construction;

[0006] The cloud includes a data aggregation module, a data isolation module, and a distributed storage module; wherein the data aggregation module is used to aggregate ship-side data through load balancing, resource management, and security protection mechanisms; the data isolation module is used to isolate and process the aggregated data; and the distributed storage module is used to use a distributed time series database to complete data storage and active-active backup.

[0007] The shore end includes a data center layer, an algorithm center layer and a business center layer; wherein, the data center layer is used to perform standardized management of data; the algorithm center layer is used to complete the creation, version control and testing experiments of algorithms and data flows, and realize version control of algorithms; the business center layer is used to realize microservice applications of businesses based on a microservice architecture.

[0008] In a second aspect, the present application further provides a data management and application method for a ship comprehensive energy efficiency improvement platform, which is applied to the ship comprehensive energy efficiency improvement platform described in any one of the first aspects, comprising:

[0009] Onboard steps: Utilize multiple industrial and communication protocols to complete multi-source acquisition of equipment data; perform data processing based on edge computing; and send the processed data through a data bus, based on data communication middleware and edge database construction, to complete edge data scheduling and forwarding.

[0010] Cloud-side steps: Aggregate ship-side data through load balancing, resource management, and security protection mechanisms; isolate and process aggregated data; and use a distributed time-series database for data storage and active-active backup.

[0011] Onshore steps: Build a data center layer to manage data in a standardized manner; build an algorithm center layer to complete the creation, version control and testing experiments of algorithms and data flows, and implement algorithm version control; build a business center layer to implement microservice applications of the business based on the microservice architecture.

[0012] This application effectively solves the problems of limited ship energy efficiency improvement, data silos, and insufficient energy efficiency optimization of energy-saving equipment by constructing a three-in-one ship-side, cloud-side, and shore-side comprehensive energy efficiency improvement platform and its data management system. Specifically, the ship-side realizes multi-source collection, edge processing, and scheduling forwarding of equipment data through the data acquisition module, edge computing module, and data scheduling module; the cloud-side completes data aggregation, isolation, and distributed storage through the data aggregation module, data isolation module, and distributed storage module; the shore-side realizes standardized data management, algorithm creation and optimization, and microservice application of business through the data center layer, algorithm center layer, and business center layer.

[0013] Therefore, this application realizes the integrated management of the entire process from data collection to processing, storage and business application, and has the characteristics of high efficiency, flexibility and scalability, which significantly improves the intelligence and coordination level of ship comprehensive energy efficiency business. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments or traditional solutions of the present application, a brief introduction will be given below to the drawings required for use in the description of the embodiments or traditional solutions. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0015] Figure 1 This is a structural diagram of the ship comprehensive energy efficiency improvement platform for this application;

[0016] Figure 2This is the overall structure diagram of the ship comprehensive energy efficiency improvement platform according to the embodiment of the present application;

[0017] Figure 3 This is a technical architecture diagram of the ship comprehensive energy efficiency improvement platform according to an embodiment of the present application;

[0018] Figure 4 This is a data architecture diagram of the ship comprehensive energy efficiency improvement platform according to an embodiment of the present application;

[0019] Figure 5 This is a diagram of the structured data architecture of an embodiment of the present application;

[0020] Figure 6 This is an unstructured data architecture diagram of an embodiment of the present application;

[0021] Figure 7 This is a diagram of the ship comprehensive energy efficiency business architecture of an embodiment of the present application;

[0022] Figure 8 This is a diagram of the comprehensive data warehouse of the ship comprehensive energy efficiency platform according to an embodiment of the present application;

[0023] Figure 9 This is a diagram of the business function modules of the ship comprehensive energy efficiency platform according to an embodiment of the present application;

[0024] Figure 10 A flowchart of the data management and application method of the ship comprehensive energy efficiency improvement platform of this application;

[0025] Figure 11 This is a flowchart of ship comprehensive energy efficiency data mining in an embodiment of the present application;

[0026] Figure 12 This is a flowchart of the coordinated optimization of route speed according to an embodiment of the present application;

[0027] Figure 13 This is a flowchart of energy efficiency matching optimization for multiple energy-saving technologies according to an embodiment of the present application;

[0028] Figure 14 This is a flow chart of the multi-source and multi-modal ship data processing and application architecture according to an embodiment of the present application. DETAILED DESCRIPTION

[0029] In the embodiments of this application, the term "and / or" describes the relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally indicates that the related objects are in an "or" relationship. In the embodiments of this application, the term "plurality" refers to two or more, and other quantifiers are similar.

[0030] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0031] This application comprehensively solves the problems of limited ship energy efficiency improvement, data silos, and insufficient energy efficiency optimization of energy-saving equipment by constructing a three-in-one ship-side, cloud-side, and shore-side comprehensive energy efficiency improvement platform. Specifically, the ship-side uses a variety of industrial protocols and communication protocols through the data acquisition module to realize multi-source collection of equipment data, ensuring the comprehensiveness and real-time nature of the data; the edge computing module processes the data locally based on edge computing technology, reduces data transmission delays and improves processing efficiency; the data scheduling module uses the data bus and data communication middleware, combined with the edge database, to realize efficient scheduling and forwarding of edge data, ensuring the smooth flow of data. The cloud uses load balancing, resource management and security protection mechanisms through the data aggregation module to complete the centralized aggregation of ship-side data and solve the problem of data silos; the data isolation module isolates and processes the aggregated data to ensure the security and independence of the data; the distributed storage module uses a distributed time series database to realize efficient data storage and active-active backup, ensuring the reliability and accessibility of the data. The shore side manages data in a standardized manner through the data center layer to ensure data standardization and consistency; the algorithm center layer completes the creation, version control and test experiments of the algorithm, supports the development and application of multi-objective optimization algorithms, and realizes the precise optimization of the energy efficiency of energy-saving equipment; the business center layer is based on the microservice architecture to realize the microservice application of the business and improve the flexibility and scalability of business deployment. Therefore, this application has the advantages of efficient data integration, precise algorithm optimization, and flexible business application. It can give full play to the value of data, improve the comprehensive energy efficiency of ships, and support the coordinated optimization and intelligent management of multiple energy-saving equipment, providing comprehensive technical support for improving ship energy efficiency.

[0032] This application realizes the integrated management of the entire process of data from collection, transmission, storage, processing and business application, and has the characteristics of high efficiency, flexibility and scalability. Based on existing ship energy-saving equipment and energy-saving solutions, it uses artificial intelligence, big data analysis and Internet technologies to break data silos, standardize data management, strengthen data analysis, enhance data application, give full play to the value of data, enable the comprehensive energy efficiency of ships, and form a comprehensive solution for improving ship energy efficiency that combines software and hardware.

[0033] The following is combined with Figures 1 to 14 This application is described in detail.

[0034] Please refer to Figure 1, Figure 1 This is a block diagram of the ship comprehensive energy efficiency improvement platform proposed in this application. The platform comprises three components: ship-side, cloud-side, and shore-side. The ship-side is responsible for data collection, edge computing, storage, and scheduling, serving as the platform's data source. The cloud-side is responsible for data aggregation, isolation, storage, management, and processing, serving as the platform's core data processing center. The shore-side is responsible for standardized data management, algorithm optimization, and business applications, serving as the platform's business decision-making center.

[0035] The platform's data transmission mechanism utilizes multiple protocols to achieve efficient data transmission between the ship, cloud, and shore. The platform architecture utilizes a distributed time-series database, microservices, and multimodal data processing mechanisms to ensure efficient operation and scalability. It aims to enable full lifecycle management of ship data, optimize energy efficiency, and support decision-making. The Ship Comprehensive Energy Efficiency Improvement Platform utilizes multiple industrial and communication protocols for data collection, transmission, aggregation, storage, and processing. It supports the mining of ship energy efficiency data, the coordinated optimization of route speeds, and the energy efficiency evaluation and optimization of multiple energy-saving technologies.

[0036] For example, the ship side is responsible for data collection, edge computing, storage and scheduling, and specifically includes a data collection module, an edge computing module, an edge storage module and a data scheduling module. The details are as follows:

[0037] The data acquisition module uses the Industrial Bus Protocol, Message Queue Telemetry Transport Protocol (MQTT) and communication protocols to achieve multi-source collection of equipment data, ensuring the comprehensiveness and real-time nature of data collection.

[0038] The edge computing module is used for data processing based on edge computing, specifically for processing collected data. It includes a digital filtering module, a data cleaning module, a data deduplication module, a data normalization module, a data discretization module, a comprehensive energy efficiency data feature extraction module, and a fault alarm module. The digital filtering module removes noise from the data; the data cleaning module cleans invalid or erroneous data; the data deduplication module eliminates duplicate data; the data normalization module converts data into a unified format; the data discretization module converts continuous data into discrete data; the comprehensive energy efficiency data feature extraction module extracts key features reflecting the ship's energy efficiency; and the fault alarm module monitors data anomalies in real time and triggers alarms.

[0039] The edge storage module uses a standalone time-series database and a relational database (such as MySQL) to store real-time and computational data. It supports efficient data storage and query, ensuring data integrity and availability.

[0040] The data scheduling module adopts a subscription-based publishing model, using a data bus, built on data communication middleware and edge databases to complete the scheduling and forwarding of edge data. The data scheduling module enables efficient data transmission and distribution, ensuring real-time and consistency.

[0041] For example, the cloud is responsible for data aggregation, isolation, storage, management, and processing, specifically including a data aggregation module, a data isolation module, a distributed storage module, an edge management module, a data management module, a security protection module, a data source expansion module, a data processing algorithm expansion module, an operation resource management expansion module, a multimodal data processing module, a batch-stream integrated data processing module, and a time series data processing module. Specific details are as follows:

[0042] The data aggregation module aggregates ship-side data through Message Queuing Telemetry Transport Protocol (MQTT) and database synchronization capabilities. The data aggregation module supports centralized management and integration of multi-source data.

[0043] The data isolation module is used to isolate the aggregated data to ensure data security and privacy to prevent data leakage and illegal access.

[0044] The distributed storage module uses a distributed time-series database for data storage and active-active backup. Active-active backup refers to two or more data centers being simultaneously active, processing business requests simultaneously, and synchronizing data with each other in real time. If one data center fails, the other can immediately take over, ensuring uninterrupted service. The distributed storage module supports efficient storage and disaster recovery of large amounts of data. Disaster recovery means that in the event of a catastrophic failure in the primary data center, business operations can be quickly restored by rapidly switching to a backup data center. Disaster recovery typically includes data backup, system backup, and a business recovery plan.

[0045] The edge management module manages edge devices, connections, and applications. It includes a device management module, a connection management module, and an application management module. The device management module manages the registration, monitoring, and maintenance of edge devices; the connection management module manages the connection status between devices and the cloud; and the application management module manages the deployment and updates of edge applications.

[0046] The data management module is used to manage data in a standardized manner. It includes a metadata management module, a data cache module, a data model management module, a rules engine module, and a time alignment module. The metadata management module manages data metadata; the data cache module improves data access efficiency; the data model management module defines and manages data models; the rules engine executes data processing rules; and the time alignment module ensures data temporal consistency.

[0047] The security protection module is used to comprehensively safeguard data and network security through access control, data encryption, backup and recovery, security auditing, and anomaly detection to prevent data leakage, tampering, and illegal access.

[0048] The Data Source Extension Module is used to expand data sources through various industrial protocols, multimodal data communication protocols, database synchronization, file access, and API (Application Programming Interface) integration. The Data Source Extension Module supports the access and integration of multi-source data.

[0049] The data processing algorithm expansion module is used to expand data processing algorithms and flexibly deploy and optimize algorithms through the creation, version control and testing of algorithms and data flows.

[0050] The multimodal data processing module is used to complete the collection, transmission, storage and processing of unstructured data such as videos, images and files; the multimodal data processing module uses the Real-time Transport Protocol (RTP) and Session Initiation Protocol (SIP) to alleviate the pressure of MQTT on video data transmission; and uses document-based databases and streaming media processing technology to make up for the shortcomings of time series databases in the compression, transmission and storage of unstructured data.

[0051] The batch-stream integrated data processing module integrates batch and stream processing capabilities. It utilizes a big data processing framework, message queue system, distributed database, and time series database technology components to simplify data processing.

[0052] The time series data processing module is used to support the write performance and query throughput of tens of millions of points per second on a single server, and complete the millisecond-level aggregation of tens of billions of data points to achieve efficient processing and analysis of large-scale time series data.

[0053] For example, the shore side is responsible for standardized data management, algorithm optimization, and business applications, specifically including the data center layer, algorithm center layer, and business center layer. The details are as follows:

[0054] The data center layer is responsible for standardized data management and includes a metadata management module, a data compression and storage module, a data caching module, a time series creation module, a complex time processing module, a sequence data matching and discovery module, anomaly detection module, and a ship energy efficiency data mining module. The metadata management module manages data metadata. The data compression and storage module optimizes data storage space. The data caching module improves data access efficiency. The time series creation module generates time series data. The complex time processing module processes complex time data. The sequence data matching and discovery module discovers correlation patterns in the data. The anomaly detection module monitors data anomalies. The ship energy efficiency data mining module extracts key features reflecting ship energy efficiency and explores the correlation between energy efficiency features and different types of data.

[0055] The algorithm center layer is used to complete the creation, version control and test experiments of algorithms and data flows, and realize the version control of algorithms. It includes an algorithm and data flow creation module, a version control module, a test experiment module, a route speed collaborative optimization module, and a multi-energy-saving technology energy efficiency evaluation module. Among them, the algorithm and data flow creation module is used to create algorithms and data flows. The version control module is used to manage the version of the algorithm. The test experiment module is used to test the performance of the algorithm. The route speed collaborative optimization module is used to integrate meteorological data and provide ships with optimal or near-optimal route speed auxiliary decision-making recommendations based on electronic chart raster modeling, ship energy consumption model and collaborative optimization gray box model. The multi-energy-saving technology energy efficiency evaluation module is used to provide ships with optimal or near-optimal route speed auxiliary decision-making recommendations based on independent energy-saving equipment energy efficiency modeling, multi-energy-saving equipment combination energy efficiency modeling, multi-energy-saving scheme energy efficiency evaluation and multi-energy-saving technology energy efficiency matching optimization.

[0056] The business center layer is based on a microservice architecture and is used to implement microservice applications for the business. It includes a data sampling service module, a full data service module, an indicator maintenance service module, a rule maintenance service module, an anomaly alarm service module, and a visualization module. Among them, the full data service module is used to provide full data services. The indicator maintenance service module is used to maintain business indicators. The rule maintenance service module is used to maintain business rules. The anomaly alarm service module is used to provide anomaly alarm services. The visualization module is based on the output data of the ship energy efficiency data mining module of the data center layer, the multi-energy-saving technology energy efficiency matching evaluation module of the algorithm center layer, and the route speed collaborative optimization module, providing ship energy efficiency data mining, multi-energy-saving technology energy efficiency matching evaluation, route speed collaborative optimization and visualization business applications.

[0057] In addition, the data transmission mechanism of the ship comprehensive energy efficiency improvement platform is as follows: a two-way data transmission mechanism is used between the ship side, the cloud side, and the shore side, using one or more data transmission protocols such as Message Queuing Telemetry Transport Protocol (MQTT), Transmission Control Protocol (TCP), User Datagram Protocol (UDP), Session Initiation Protocol (SIP), Real-time Transport Protocol (RTP), and File Transfer Protocol (FTP) to achieve the transmission of high-frequency or low-frequency structured data, unstructured data, and algorithm model update packages. Its platform architecture adopts a distributed time series database, microservices architecture, and multimodal data processing mechanism to achieve balanced development of data management, computing power, and business needs.

[0058] In summary, the ship comprehensive energy efficiency improvement platform of this application realizes the full life cycle management, energy efficiency optimization and decision support of ship data through the collaborative work of the ship side, cloud side and shore side.

[0059] Please refer to Figure 2 , Figure 2 This is the overall structural diagram of the ship comprehensive energy efficiency improvement platform according to the embodiment of the present application. Figure 2 It shows a multi-layered architecture system including ship-side, cloud-side and shore-side, where each layer interacts through two-way data transmission and a business center exists.

[0060] (1) Ship side:

[0061] Sensor data acquisition is at the lowest level of the ship's infrastructure and serves as the source of data acquisition, collecting various essential data during the ship's operation. Edge data storage is responsible for storing data generated or collected at the ship's edge. Edge computing and analysis performs real-time computation and analysis on data collected onboard to support rapid decision-making. Edge data scheduling is responsible for the rational scheduling and management of edge data onboard, ensuring timely data transmission and processing. Edge management is responsible for managing edge devices and related operations onboard. Edge data aggregation aggregates and integrates edge data from various sources onboard, preparing it for further transmission and processing.

[0062] (2) Cloud:

[0063] Data isolation ensures data of different types or security levels is isolated from each other to ensure data security. Data backup backs up important data to prevent data loss. Data management coordinates the management of cloud data resources, including storage and maintenance. Cloud-based general capabilities provide common services and functional support. Edge data aggregation further aggregates and processes edge data from the ship.

[0064] (3) Shore:

[0065] Data aggregation reaggregates data transmitted from the cloud, integrating data from different channels. Distributed data storage utilizes a distributed storage approach, storing data across multiple nodes to improve storage reliability and scalability. Data analysis and mining utilizes various data analysis and mining techniques to extract valuable information from large amounts of data. Data specification management develops and manages data-related specifications and standards to ensure data consistency and standardization. Workflow management manages business processes and workflows to ensure the orderly execution of various operations. Algorithm management manages and maintains various algorithms, including their development, updates, and application. Service governance governs and oversees the provision of various services to ensure service quality and stability. Resource scheduling rationally allocates shore-based resources to improve resource utilization efficiency. Data services provide data-related services and support for upper-layer applications. Other microservice-based business applications are business applications that exist in the form of microservices in addition to the explicitly listed functions. Route and speed coordination and optimization coordinates and optimizes routes and speeds based on ship energy efficiency data to improve ship operating efficiency. Ship energy efficiency data mining is a process that conducts in-depth research on ship energy efficiency-related data to discover information such as energy-saving potential. Multi-energy-saving technology energy efficiency matching assessment evaluates the energy efficiency matching of multiple energy-saving technologies to identify the optimal combination of energy-saving technologies. Visualization presents relevant data and analysis results in a visual format for intuitive understanding and decision-making. The business center is located on the right side of the chart and interacts with the ship, cloud, and shore terminals through two-way data transmission. This means that the business center plays a role in business coordination and coordination within the entire system. It relies on the data and services provided by the ship, cloud, and shore terminals, and also guides and manages their work.

[0066] Please refer to Figure 3 , Figure 3 This is a technical architecture diagram of the ship comprehensive energy efficiency improvement platform according to an embodiment of the present application. Figure 3 The data transmission and processing architecture involving the ship side (or test bench side), the cloud side, and the shore side is shown. Each side contains different components, databases, and protocols, and exchanges data in a specific way, as shown below:

[0067] (1) Ship end or test bench end:

[0068] The communication protocol layer includes MQTT (Message Queuing Telemetry Transport Protocol, used for communication between IoT devices), RTP / SIP (Real-time Transport Protocol / Session Initiation Protocol, used for real-time data transmission such as audio and video), TCP / UDP (Transmission Control Protocol / User Datagram Protocol, basic protocols for network transmission), and TsFile (a time series data file format). These protocols and formats are used for data transmission between devices.

[0069] The edge device layer includes the NeuronEX edge gateway and the IoTDB edge time series library. The edge gateway is responsible for device connection and data forwarding, while the IoTDB is used to store time series data.

[0070] The data source layer includes industrial protocols such as Modbus (used for communication between industrial devices), DB Source (database data source that can connect to local databases to obtain data), and RESTful API (application programming interface for data exchange with other systems). These provide different types of data sources.

[0071] The bus serves as a data transmission channel, connecting components at each layer to realize data transmission within the ship or test bench.

[0072] (2) Cloud:

[0073] The data storage layer includes the cloud-based real-time database IoTDB (used to store real-time data) and the unstructured data storage area (for storing unstructured data such as documents, pictures, etc.).

[0074] The foundational component layer includes YARN / HDFS / Hadoop (YARN is a resource manager, HDFS is a distributed file system, and Hadoop is a big data processing framework that provides distributed computing and storage capabilities) and EMQX (an open-source IoT message server for device connectivity and message processing). These components provide support for cloud-based data processing and storage.

[0075] The data transmission protocol adopts FTP / MQTT / JDBC (File Transfer Protocol / Message Queuing Telemetry Transport Protocol / Java Database Connectivity, used for data transmission, device communication and database connection, etc.) to realize data interaction with the ship or shore end.

[0076] (3) Shore:

[0077] The database layer includes the full database IoTDB (storing all relevant time series full data), the exception database IoTDB (storing exception-related data), the sampling database IoTDB (storing sampled time series data), the application database IoTDB (storing application-related time series data), MySQL (such as a relational database, which can store structured data), MongoDB (such as a non-relational database, suitable for storing unstructured or semi-structured data), the platform model library Dir (which can store platform-related model data) and the distribution model library (used to distribute model-related data).

[0078] The middleware layer includes Kafka (a distributed stream processing platform for processing real-time data streams), YARN / HDFS / Hadoop (similar to the cloud, providing big data processing capabilities), EMQX (for device connection and message processing), and ZLMediaKit (for audio and video related processing and transmission).

[0079] The application layer includes Spark ML (a machine learning library based on Spark, used for machine learning-related applications), BigDL (a distributed deep learning library, used for deep learning applications), application side (specific business applications), visualization (presenting data in a visual form) and others (such as other related applications or functions).

[0080] The data transmission relationship of the ship comprehensive energy efficiency improvement platform is as follows: data upload and synchronization between the ship or test bench and the cloud is carried out through protocols such as FTP, MQTT, and JDBC. Data exchange between the cloud and the shore is achieved through data transmission components and protocols, supporting shore-side operations such as data storage, processing, and application. The entire architecture, through the coordinated operation of different components and protocols at each end, realizes a complete process from data collection on the ship or test bench, transmission to cloud storage, and shore-side processing and application.

[0081] Please refer to Figure 4 , Figure 4 This is a data architecture diagram of the ship comprehensive energy efficiency improvement platform according to an embodiment of the present application. Figure 4 The data flow and processing architecture covering edge, cloud, and shore is presented. Each end has different components and functions, as follows:

[0082] (1) Edge end (i.e. ship end):

[0083] IoTDB is a time series database (abbreviated as time series library or time series database) used to store and manage time series data generated by IoT devices, and is responsible for local data storage at the edge.

[0084] The edge gateway NeuronEX, whose main function is to connect various devices and collect, process and forward device data, is a key component for data interaction between the edge and external devices and the upper cloud.

[0085] (2) Cloud:

[0086] IoTDB (active-active) is an IoTDB database that utilizes an active-active architecture, ensuring high database availability and data reliability. This architecture means two databases are simultaneously operational, backing up and supporting each other. If one fails, the other seamlessly takes over. In addition to the active-active IoTDB, IoTDB also includes a standard IoTDB database for various storage or processing needs.

[0087] The interface service provides an interface for data interaction, which is used for data communication with edge devices and shore ends, and is the channel for data to enter and exit the cloud.

[0088] The client serves as the entry point for users or other systems to access cloud services and is used for operations such as submitting requests and obtaining data.

[0089] EMQX, a message middleware, is an open-source IoT message server that handles functions such as device connection, message publishing and subscription, and implements message communication between devices and between devices and other components in the cloud.

[0090] (3) Shore:

[0091] The data cache is used to temporarily store data to reduce pressure on the backend database and improve data access speed. It typically caches frequently accessed data. It includes three data queues for temporary data storage, enabling asynchronous data processing, ensuring stable and reliable data transmission, and avoiding data loss or congestion during the data transmission process.

[0092] The database cluster contains multiple databases and provides powerful data storage and management capabilities.

[0093] The IoTDB cluster, consisting of multiple IoTDB databases, is used to store large-scale time series data, enabling distributed data storage and processing, and improving storage and query performance. The model library stores various business models and algorithmic models for use in data analysis, forecasting, decision support, and other application scenarios. MySQL, as a relational database, is commonly used to store structured data, such as user information and order data in business systems. The non-relational database MongoDB is suitable for storing unstructured or semi-structured data, such as log data and document data.

[0094] Interface services provide various types of interfaces for interacting with external systems.

[0095] RESTful is an interface based on the REST architectural style, used for data retrieval and submission via the HTTP protocol. It offers advantages such as being lightweight and cross-platform. Web APIs are used for data exchange between web applications, allowing other web systems to call services provided by the client. C++ provides an interface based on the C++ language, facilitating integration and data exchange between applications developed in C++ and client systems. Python provides an interface based on the Python language, meeting the data access and function call needs of applications developed in Python.

[0096] The data flow of the ship's comprehensive energy efficiency improvement platform is as follows: After IoTDB and NeuronEX on the edge collect device data, they transmit the data to the cloud via interface services. After the cloud receives the data, part of it is stored in IoTDB (active-active and standard IoTDB), and part is cached and asynchronously processed through data queues. The processed data is then transmitted to the shore-side database cluster through data queues for storage. The shore-side provides various types of access interfaces through interface services based on different business needs. The entire architecture, through the collaborative work of components at all ends, realizes a complete process from device data collection, cloud storage and processing, to shore-side data management and service provision.

[0097] Please refer to Figure 5 , Figure 5 This is a diagram of the structured data architecture of an embodiment of the present application. Figure 5 The data processing and application architecture covering the ship side, cloud side, shore side, and application side is demonstrated, and the complete process from data collection to data application is described as follows:

[0098] (1) Ship side:

[0099] Data acquisition equipment includes flow meters, energy meters, odometers, inclinometers, depth sounders, anemometers, edge systems and other devices. These devices are responsible for collecting various data during the operation of the ship, such as flow, power consumption, mileage, tilt angle, water depth, wind speed and other information.

[0100] The edge database consists of a main database, a functional database (synchronization database), and a backup database. It uses IoTDB (Time Series Database) as the edge database to store the full amount of normal data collected by the ship. Normal full data refers to all data collected by the equipment under normal conditions. This data is fully stored in the edge database, providing a foundation for subsequent data transmission and processing.

[0101] (2) Cloud and shore:

[0102] The data lake includes a master database, a functional database (fault signature database, AI training database, etc.), and a backup database. IoTDB is also used as the data lake's storage database. In this phase, data is stored and processed using different strategies:

[0103] Normal data automatic frequency reduction storage is for normal data (routine data in daily operation). In order to save storage space, it is stored in a way that automatically reduces the storage frequency. That is, there is no need to store all data according to the original frequency of collection.

[0104] Automatic original frequency storage of fault-related data means that when a fault or abnormality occurs on a ship, the fault-related data will be stored according to the original acquisition frequency for subsequent fault analysis and troubleshooting, ensuring that detailed information related to the fault is not lost.

[0105] The customized data storage extracted from the issued task is based on the specific task requirements issued by the shore or other upper-level systems, and extracts relevant customized data for storage to meet the needs of specific business or analysis.

[0106] Among them, in terms of shore-side data centers:

[0107] Kafka, as a message queue, is used for asynchronous data transmission and buffering. It receives data from the data lake and distributes it to subsequent computing components according to specific rules, ensuring the stability and reliability of data transmission and avoiding data loss or congestion during the data transmission process.

[0108] Flink and Spark are big data computing frameworks. Flink is primarily used for streaming data processing, enabling real-time processing of continuous data streams. Spark is suitable for a variety of scenarios, including batch processing and interactive data analysis. They can perform various computational and analytical operations on data retrieved from Kafka, such as data cleansing, aggregation, and modeling. Furthermore, other computing frameworks can be incorporated to meet diverse computing needs.

[0109] The storage database uses IoTDB, which is again used to store calculated and processed data as part of the data storage of the shore data center, providing support for subsequent data applications.

[0110] (3) Application end

[0111] Intelligent applications utilize collected, stored, and processed data to develop various intelligent applications, such as intelligent decision support systems that provide optimization recommendations based on ship operation data. Comprehensive analysis involves a comprehensive analysis of data to uncover the potential value and patterns underlying it, such as analyzing the relationship between ship energy consumption and navigation conditions.

[0112] Business applications include route and speed optimization, multi-energy-saving device matching, and a large-scale visualization screen. Route and speed optimization optimizes routes and speeds based on the ship's real-time and historical data, combined with external information such as weather and sea conditions, to improve navigation efficiency and reduce energy consumption. Multi-energy-saving device matching analyzes the operating data of various energy-saving devices on a ship to identify the optimal device matching solution for greater energy savings. The large-scale visualization screen displays data in a visual format, intuitively presenting information such as the ship's operating status and various indicators, facilitating monitoring and decision-making by managers.

[0113] Figure 5 The entire architecture shown in the figure gradually processes the ship-side data and applies it to various business scenarios through collection, caching, aggregation, storage, and calculation, realizing the complete process from data collection to value realization.

[0114] Please refer to Figure 6 , Figure 6 This is a diagram of the unstructured data architecture of an embodiment of the present application. Figure 6 The audio and video data processing and application architecture involving ship-side / test bench, cloud, shore-side data center and application side is demonstrated as follows:

[0115] (1) Ship side / test bench:

[0116] The communication protocols of the shipboard edge computing storage module use RTP (Real-time Transport Protocol) and SIP (Session Initiation Protocol). RTP is primarily used to transmit real-time data, such as audio and video streams, over the network. It provides packet sequence numbering and timestamps to ensure the correct transmission and playback of real-time data. SIP is an application layer control protocol used to create, modify, and release sessions between one or more participants. It is commonly used in scenarios such as VoIP (Voice over Internet Protocol) and video conferencing. Data storage is for full audio and video data, that is, all collected audio and video data is completely stored, preserving the original data for subsequent processing and analysis. Monitoring equipment includes cameras, cameras, and microphones. Cameras are used to capture video image information; cameras can also be used to capture still images or videos; microphones are responsible for capturing audio information. These devices are the source of audio and video data and are used to monitor relevant scenarios on board or on the test bench.

[0117] (2) Cloud:

[0118] The data storage strategy includes reduced-frequency storage of normal data, original-frequency storage of key data, and customized data storage extracted by dispatching tasks. Among them, reduced-frequency storage of normal data is to reduce the storage frequency of audio and video data under normal conditions to save storage space, because the changes in data under normal conditions are relatively stable and do not need to be stored at a high frequency. Original-frequency storage of key data is to store key audio and video data, such as data when abnormal situations or specific events occur, according to the original acquisition frequency to ensure the integrity and accuracy of key information and facilitate subsequent detailed analysis. Customized data storage extracted by dispatching tasks is to extract relevant customized audio and video data for storage based on specific task requirements issued by the shore or other upper-level systems to meet the data needs of specific business scenarios.

[0119] (3) Onshore data center:

[0120] ZLMediaKit, a video transmission middleware, is a high-performance streaming media service framework. Its functional modules include video recording, real-time viewing, data transcoding, and video stream management. Video recording allows users to record audio and video streams, saving real-time audio and video data as files for later viewing and analysis. Real-time viewing allows users to view audio and video streams in real time, enabling remote monitoring. Data transcoding converts collected audio and video data into different encoding formats to accommodate different playback devices and network environments. Video stream management manages audio and video streams, including stream creation, destruction, and distribution, to ensure stable transmission and playback of video streams.

[0121] The interface includes API and HOOK. The API provides an application program interface, allowing developers to easily call ZLMediaKit functions for secondary development and integration. The HOOK is used to perform callback operations when specific events occur, such as triggering corresponding processing logic when a video stream starts, ends, or an exception occurs.

[0122] Device management-related functions include device management, on-demand streaming, PTZ management, local recording, cloud recording, and device alarms. Device management involves managing surveillance equipment connected to the shore-side data center, including operations such as device registration, configuration, and status monitoring. On-demand streaming involves pulling corresponding audio and video streams from devices or storage based on user or application needs, avoiding unnecessary bandwidth usage and resource waste. PTZ management involves managing surveillance equipment with pan-tilt-zoom (PTZ) functionality, enabling remote control of camera operations such as horizontal rotation, vertical pitch, and zoom. Local recording involves storing audio and video data on a local server for backup or for quick local viewing. Cloud recording involves storing audio and video data in the cloud, enabling remote storage and sharing of data. Device alarms involve issuing timely alerts when abnormal conditions (such as device failure, network interruption, etc.) occur in surveillance equipment or when specific events (such as intrusion, fire, etc.) are detected.

[0123] (4) Application end

[0124] Protocol support refers to protocols such as FLV (Flash Video), HLS (HTTP Live Streaming), HTTP, and HTTPS. FLV is commonly used to play videos on web pages. HLS is an adaptive streaming protocol based on HTTP that automatically adjusts video quality based on network conditions. HTTP and HTTPS are protocols used to transmit data over the Internet. HTTPS adds encryption and authentication features to HTTP to ensure secure data transmission.

[0125] Application devices support access from multiple devices, including mobile phones and computers, allowing users to view audio and video data or perform related operations in real time through different terminals. Furthermore, support for VR devices is also available, providing users with a more immersive monitoring experience.

[0126] Figure 6 The architecture shown implements a complete process from audio and video data collection on the ship / test bench, to cloud storage and processing, to further management in the shore-side data center and diversified access on the application side.

[0127] Please refer to Figure 7 , Figure 7 This is a diagram of the ship comprehensive energy efficiency business architecture of an embodiment of the present application. Figure 7 The architecture of a ship energy efficiency management system, including the shipside or test bench, cloud, and shore-side, is presented, describing the complete process from device access and data interaction to data management and business applications, as follows:

[0128] (1) Ship end or test bench:

[0129] Terminal devices include various terminals such as sensors and software systems. These terminal devices are responsible for collecting various types of data during the operation of the ship, such as energy consumption data, navigation parameters, etc., and are the data source for the entire system.

[0130] The device access layer includes industrial protocols, MQTT protocol, FTP and JDBC. Among them, industrial protocols refer to industrial protocols such as Modbus and IEC104, which are used to connect different types of industrial equipment and realize the collection and transmission of equipment data. The Modbus protocol is an application layer communication protocol, commonly used in the field of industrial automation, with the characteristics of simplicity and openness; the IEC104 protocol is mainly used for remote communication of power systems. MQTT (Message Queuing Telemetry Transport) is a lightweight message transmission protocol suitable for IoT device communication in low-bandwidth and unstable network environments, and is often used for data transmission between devices and servers. The File Transfer Protocol (FTP) is used to upload and download files on the network and can be used to transmit large amounts of data files collected by devices. JDBC refers to Java Database Connectivity, which is used to connect and operate databases in Java programs and store collected data in local databases.

[0131] The basic service layer includes real-time data storage, real-time data computation, data model management, a service bus, and a message bus. Real-time data storage stores collected real-time data, ensuring data loss and providing a foundation for subsequent computation and analysis. Real-time data computation performs preliminary computations and processing on real-time data, such as filtering and aggregation, to reduce the burden of subsequent data processing. Data model management involves creating, modifying, and maintaining data models to ensure data is stored and processed in an appropriate structure. The service bus integrates and manages communication between different services, enabling service discovery, invocation, and routing. The message bus is responsible for message transmission and distribution, enabling asynchronous communication between different components.

[0132] The external service layer includes device management interaction, business data interaction, and security interaction. Device management interaction refers to interaction with cloud-based or onshore device management systems, such as device registration, configuration updates, and status reporting. Business data interaction is responsible for transmitting business data with the cloud-based or onshore system, sending collected business-related data to upper-layer systems and receiving instructions and data from these systems. Security interaction ensures the security of data interaction, including data encryption, identity authentication, access control, and other functions to prevent data leakage and unauthorized access.

[0133] (2) Cloud and shore

[0134] Edge management includes device management, connection management, application management, rule engines, and data processing. Device management refers to the unified management of devices connected to the system, including operations such as adding, removing, and monitoring device status. Connection management refers to managing the connection between devices and the system to ensure the stability and reliability of the connection, such as establishing, disconnecting, and reconnecting. Application management refers to the management of applications running on the edge, including deployment, updating, and monitoring. The rule engine processes and judges data based on preset rules, such as triggering alarms and executing specific operations. Data processing refers to the further processing of data collected from the ship or test bench, such as data cleaning and conversion.

[0135] Data management encompasses real-time, relational, and cached data management, as well as data computation. Real-time, relational, and cached data management includes metadata management, data model management, data migration, time series creation, high-compression storage, and data caching. Data computation encompasses frequency domain data processing, complex time processing, sequence data matching, sequence data discovery, sequence data mining, and anomaly detection.

[0136] Specifically, real-time, relational, and cached data management refers to the management of real-time, relational, and cached data to ensure efficient data storage and access. Real-time data reflects current equipment status and business operations; relational data stores structured business data; and cached data temporarily stores frequently accessed data to improve data access speed. Metadata management refers to the management of metadata, which is information about data, such as its definition, structure, and source, that facilitates better understanding and management of the data. Data model management refers to the unified management of data models across the entire system to ensure consistency and accuracy. Data migration refers to the migration of data from one storage location to another for operations such as data backup, archiving, or storage system upgrades. Time series creation refers to the creation of time series data. Time series data is crucial in ship energy efficiency management and can be used to analyze equipment operating status and energy consumption trends. High-compression storage uses high-compression algorithms to store data, saving storage space while maintaining data integrity and availability. Data caching refers to the establishment of a data caching mechanism to improve data access efficiency.

[0137] Specifically, in data calculations: Frequency domain data processing refers to the processing of frequency domain data. Frequency domain analysis can reveal the characteristics of data at different frequencies, which helps analyze the operating status of equipment and diagnose faults. Complex time processing refers to the processing of complex time-related problems, such as the alignment of time series and the calculation of time windows. Sequence data matching refers to matching sequence data, such as matching the operating mode and energy consumption mode of the equipment. Sequence data discovery refers to the discovery of regularities and patterns in sequence data to provide support for energy efficiency analysis and optimization. Sequence data mining refers to the use of data mining technology to conduct in-depth analysis of sequence data to mine potential information and knowledge. Anomaly detection refers to the detection of anomalies in data, such as equipment failures and abnormal energy consumption, and the issuance of timely alarms.

[0138] Data services include data sampling service, full data service, indicator maintenance service, indicator maintenance service, rule maintenance service, and exception alarm service. Among them, data sampling service refers to providing data sampling function, which can sample raw data at a certain frequency to meet the data granularity requirements of different applications. Full data service refers to providing access to full data, making it convenient for users to obtain complete historical data. Indicator maintenance service refers to the maintenance of energy efficiency-related indicators, such as indicator definitions and updates to calculation methods. Rule maintenance service refers to the maintenance of rules in the system, including operations such as adding, modifying, and deleting rules. Exception alarm service means issuing alarm information in a timely manner when an abnormal situation is detected, notifying relevant personnel to handle it.

[0139] Business applications (i.e., comprehensive improvement of ship energy efficiency) include alarm events, energy efficiency analysis, energy efficiency assessment, multi-energy-saving and energy-efficiency matching assessment, and comprehensive improvement of ship energy efficiency. Alarm events refer to the management and processing of alarm events generated in the system, including functions such as alarm display, classification, and notification. Energy efficiency analysis analyzes a ship's energy efficiency, identifying energy consumption patterns and potential energy savings through analysis of energy consumption data and navigation parameters. Energy efficiency assessment assesses a ship's energy efficiency level, providing corresponding assessment indicators and grades to facilitate quantitative evaluation of the ship's energy efficiency performance. Multi-energy-saving and energy-efficiency matching assessment evaluates the synergistic effects of multiple energy-saving devices or measures to identify the optimal energy-saving combination. Route and speed collaborative optimization involves the coordinated optimization of routes and speeds based on factors such as the ship's energy consumption data and navigation conditions to reduce energy consumption and improve navigation efficiency. Comprehensive improvement of ship energy efficiency refers to achieving comprehensive improvements in ship energy efficiency through the above-mentioned business applications, achieving the goals of energy conservation, emission reduction, and lowering operating costs.

[0140] Please refer to Figure 8 , Figure 8 This is a diagram of the comprehensive data warehouse of the ship comprehensive energy efficiency platform in an embodiment of the present application. Figure 7The platform functional module architecture diagram is shown, and the specific functions of each functional module and its sub-modules of the platform are described as follows:

[0141] The edge adaptation module is primarily responsible for data collection and preliminary processing related to edge devices, as well as algorithm integration. This module includes industrial bus protocol data collection, collected data storage integration, edge data filtering and analysis, and AI / ML algorithm integration. Industrial bus protocol data collection refers to the collection of data generated by edge devices via the industrial bus protocol, which is a critical step in obtaining raw data. Collected data storage integration involves the integrated storage processing of collected data to ensure its proper preservation. Edge data filtering and analysis involves filtering data transmitted from edge devices, removing invalid or erroneous data, and performing preliminary analysis. AI / ML algorithm integration involves integrating artificial intelligence (AI) and machine learning (ML) algorithms to enable more advanced analysis and processing of collected data.

[0142] The data interaction module is primarily responsible for the transmission and aggregation of data between different devices and layers. It consists of two submodules: the data forwarding module and the data aggregation module. The data forwarding module forwards data between different locations, such as edge devices and the ship, the ship and the cloud, and the cloud and shore. It supports multiple data transmission methods and ensures smooth data transmission between different entities. The data aggregation module, together with the data forwarding module, forms the sending and receiving ends of the real-time data bus, realizing data aggregation and centralizing dispersed data for further processing.

[0143] The data pre-processing module is used to extract, clean, extract features, encode, and load raw data. This includes data extraction, data cleaning, feature extraction, data encoding, and data loading. Data extraction refers to extracting the required data from the raw data source. Data cleaning removes noise, erroneous values, duplicate values, and other data to improve data quality. Feature extraction extracts representative features from raw data for subsequent analysis and modeling. Data encoding converts data into a format suitable for computer processing and storage. Data loading transfers processed data into the appropriate data storage.

[0144] The comprehensive data warehouse is used to store and manage different types of data. It includes three sub-modules: the real-time data warehouse, the real-time data analysis library, and the analysis results library. The real-time data warehouse is used to store real-time data transmitted back from the edge, and includes a full database, a downsampled database, an anomaly database, and an external application database. The full database stores all real-time data; the downsampled database downsamples and stores data to save space; the anomaly database stores data with anomalies; and the external application database provides data for external applications. The real-time data analysis library is used to store indicator data, analytical data, and processed and summarized data generated by processing and summarizing real-time data in the real-time data warehouse for further data analysis and decision support. The analysis results library is used to store the needs of different business departments and analysis results data for different applications to meet the storage and use requirements of different users for data analysis results.

[0145] The data management module is primarily responsible for comprehensive data management and includes multiple management functions, namely metadata management, master data management, data standards management, data lifecycle management, data quality management and monitoring, data service monitoring and management, and data service security auditing. Metadata management refers to the management of metadata (data about data, such as data definitions and structures). Master data management refers to the management of master data (core business entity data that is consistent and shared across the entire organization). Data standards management refers to the development and management of data standards to ensure data consistency and standardization. Data lifecycle management refers to the management of data throughout its entire lifecycle, from generation to use to final destruction. Data quality management and monitoring refers to monitoring data quality to ensure that data meets predetermined quality standards. Data service monitoring and management refers to monitoring the operating status of data services and performing corresponding management operations. Data service security auditing refers to auditing the security of data services to ensure data security.

[0146] The data service interface provides different types of interfaces to facilitate external systems accessing and using the platform's data and services. It includes basic interfaces and encapsulated interfaces. Basic interfaces are used to interact with different databases and computing frameworks, including IoTDB-like SQL interfaces, relational database SQL interfaces, NoSQL interfaces, and Flink / Spark interfaces. Encapsulated interfaces encapsulate commonly used data service functions for easier invocation, including data forwarding interfaces, real-time data query interfaces, and resource directory service interfaces.

[0147] The algorithm management module is primarily responsible for algorithm management and scheduling, including built-in common algorithms, algorithm scheduling, and basic algorithm management. Built-in common algorithms refer to commonly used algorithms built into the platform for data processing and analysis. Algorithm scheduling involves the rational scheduling and execution of algorithms based on business needs and data processing workflows. Basic algorithm management involves managing basic algorithms, such as creating, modifying, and deleting them, to ensure their availability and effectiveness.

[0148] Please refer to Figure 9 , Figure 9 This is a business function module diagram of the ship comprehensive energy efficiency platform according to an embodiment of the present application. Figure 9 The diagram shows the structure of the integrated data warehouse. With the integrated data warehouse as the center, multiple databases with different functions are distributed in a circular pattern. These databases together form a relatively complete data storage and management system, as shown below:

[0149] The central component is the integrated data warehouse, located at the center of the entire diagram and the core of the entire data storage and management system. It integrates multiple types of data warehouses, plays a key role in aggregating and coordinating various types of data, and provides a unified data foundation for upper-level data applications and analysis.

[0150] The ring distribution refers to the ring surrounding the integrated data warehouse. Based on different functions and uses, it is divided into three major categories: real-time data warehouse, real-time data analysis library, and analysis result library. Each major category contains several specific databases:

[0151] Related to the real-time data warehouse are the full database, the downsampled database, the anomaly database, and the external application database. The full database stores all raw data, containing the most complete information and serving as the fundamental data source for comprehensive data analysis. For example, in a ship energy efficiency management system, it can store all energy consumption, navigation parameters, and other data from the beginning of a ship's operation. The downsampled database stores raw high-frequency data after downsampling. While preserving key data characteristics, this reduces data volume, saving storage space and computing resources. For example, data originally collected once per second can be sampled and stored once per minute. The anomaly database is specifically used to store records of anomalies detected during data collection and analysis. When abnormal equipment operating conditions or data indicators exceed normal ranges occur, the relevant data is stored in this database to facilitate subsequent troubleshooting and problem analysis. The external application database provides data support for external applications. This database stores filtered and processed data to meet the data needs of various external applications while ensuring the security of core data.

[0152] Related to the real-time data analysis library are processing and summary databases, indicator databases, and analytical databases. Among them, the processing and summary database is a data repository for processing and summarizing the raw data. For example, energy consumption data is stored here after being summed and averaged according to different time periods (hours, days, months, etc.), making it easy to quickly obtain statistical information. The indicator database is used to store various indicator data generated through calculations. These indicators are key data points extracted from the raw data based on business needs and data analysis purposes, such as the energy consumption per unit mileage indicator of ships and the operating efficiency indicator of equipment. The analytical database is mainly used to store data suitable for analysis. These data have undergone preprocessing, feature extraction and other operations, and can more conveniently support the application of various analysis algorithms and models to discover potential patterns and trends in the data.

[0153] Related to the analysis results database are fixed report databases, ad hoc query databases, online analysis databases, and data mining result databases. The fixed report database stores data used to generate fixed-format reports. These reports are typically generated according to established templates and rules to meet daily business reporting and regulatory requirements, such as monthly energy consumption reports and equipment operating status reports. The ad hoc query database supports users' immediate and flexible data query operations. Users can quickly obtain the required data from this database based on their temporary needs without going through complex data processing and report generation processes. The online analysis database provides data support for online analytical processing (OLAP). It supports multi-angle and multi-level data analysis, allowing users to slice, dice, and drill down on data from different dimensions (such as time, location, and equipment type) to deeply explore the underlying information. The data mining result database stores the results data generated by data mining algorithms. Data mining aims to discover hidden patterns, relationships, and knowledge from large amounts of data. This data can provide deeper support and reference for decision-making, such as predicting equipment failure probability and optimizing ship routes.

[0154] Please refer to Figure 10 , Figure 10 This is a flow chart of the data management and application method of the ship comprehensive energy efficiency improvement platform of this application. A data management and application method of the ship comprehensive energy efficiency improvement platform, which is applied to the ship comprehensive energy efficiency improvement platform described in any of the above embodiments, includes:

[0155] S1010, ship-side steps: use a variety of industrial protocols and communication protocols to complete multi-source collection of equipment data; process data based on edge computing; send the processed data through the data bus, based on data communication middleware and edge database construction, to complete the scheduling and forwarding of edge data.

[0156] Specifically, in the ship-side step, data processing based on edge computing includes:

[0157] S1011 performs real-time pre-processing on collected device data, including missing data processing, abnormal data processing, noise data processing and data standardization.

[0158] S1012, feature extraction is performed on the pre-processed data to extract key features reflecting the ship's energy efficiency, including fuel consumption rate, energy efficiency operation index, and carbon intensity;

[0159] S1013, forwarding the extracted feature data to the cloud via the data bus, and storing the original data in the edge database for local analysis and backup.

[0160] Step S1010 above describes the complete process of ship-side data processing, from multi-source acquisition of equipment data using a variety of industrial and communication protocols, to data preprocessing and feature extraction based on edge computing, and then forwarding the processed feature data to the cloud via the data bus. Meanwhile, the original data is stored in the edge database for local analysis and backup. This process ensures comprehensive data collection, real-time processing, and efficient transmission, providing high-quality data support for ship energy efficiency management. At the same time, the combination of edge computing and local storage reduces dependence on cloud resources, enhances data security and traceability, and lays a solid technical foundation for ship energy efficiency optimization and operational management.

[0161] S1020, cloud-side steps: complete the aggregation of ship-side data through load balancing, resource management and security protection mechanisms; isolate and process the aggregated data; use a distributed time series database to complete data storage and active-active backup.

[0162] Specifically, in the cloud step, the isolation processing of aggregated data includes:

[0163] S1021 uses a data isolation module to isolate and store data from different ship ends according to data type and source.

[0164] S1022, compress and encrypt the isolated data to achieve data security and storage efficiency.

[0165] S1023: Store the compressed and encrypted data in a distributed time series database and perform active-active backup simultaneously to achieve high data availability.

[0166] Step S1020 describes the cloud-based data processing process and operations. Its purpose is to efficiently aggregate data from different shipboard locations through load balancing, resource management, and security protection mechanisms. It also employs data isolation, compression, encryption, and distributed time-series database storage with active-active backup to ensure data security, storage efficiency, and high availability. Specifically, the data isolation module isolates and stores data by type and source. Compression and encryption enhance data security and storage efficiency. The distributed time-series database and active-active backup mechanism ensure data reliability and high availability, providing stable, secure, and efficient technical support for cloud-based data management and analysis.

[0167] S1030, shore-side steps: Build a data center layer to manage data in a standardized manner; build an algorithm center layer to complete the creation, version control and testing experiments of algorithms and data flows, and implement algorithm version control; build a business center layer to implement microservice applications of the business based on the microservice architecture.

[0168] Specifically, in the onshore step, the business center layer is built, and the microservice applications of the business based on the microservice architecture include:

[0169] S1031 modularizes the ship energy efficiency analysis, route optimization, speed optimization, and multi-energy-saving technology matching optimization functions to form independent microservices.

[0170] S1032, realizes communication and data interaction between microservices through API gateway, and realizes flexibility and scalability of business logic;

[0171] S1033 deploys microservices based on containerization technology, combined with load balancing and automatic scaling mechanisms to achieve performance and stability of the business center layer.

[0172] Step S1030 above describes the architecture and implementation of shore-side data and business processing. Its purpose is to achieve standardized data management, algorithm version control and testing, and modular business applications based on a microservices architecture by building a data center layer, an algorithm center layer, and a business center layer. Specifically, the business center layer modularizes functions such as ship energy efficiency analysis and route optimization into independent microservices, enabling communication and data exchange between microservices through an API gateway. Incorporating containerization technology, load balancing, and auto-scaling mechanisms, it ensures the flexibility and scalability of business logic, as well as system performance and stability, providing efficient, reliable, and easily scalable technical support for shore-side business applications.

[0173] Please refer to Figure 11 , Figure 11 This is a flowchart of ship comprehensive energy efficiency data mining in an embodiment of the present application. Figure 11The data processing and analysis process for ship energy efficiency analysis is demonstrated, including four main steps: data collection, feature extraction, data mining, and intelligent analysis. The details are as follows:

[0174] The data collection process includes meteorological data, ship operation data, and other data. Meteorological data, as one of the data sources, is an external environmental factor that affects ship energy efficiency. Ship operation data encompasses multiple aspects. Equipment detection environment data reflects the environmental conditions of the ship's equipment; installed sensor data is data collected by additional equipment; ship assembly system data contains information related to the overall ship assembly; ship communication and navigation equipment data is related to communication and navigation equipment; crew record operation data is operational information manually recorded by crew members; and ship-side energy efficiency analysis data is existing energy efficiency analysis data on board. These data comprehensively reflect the ship's operating status and are an important basis for analyzing ship energy efficiency. Other data refers to relevant data other than meteorological and ship operation data.

[0175] Collected data can present various issues, requiring processing for missing values, outliers, noise, duplicate data, and data transformation. Missing value processing addresses missing values and ensures data integrity. Outlier processing identifies and addresses data that deviates from the normal range to prevent interference with subsequent analysis. Noisy data processing removes noise to more accurately reflect the real world. Duplicate data processing eliminates duplicate records and reduces data redundancy. Data transformation transforms data to make it more suitable for subsequent analysis.

[0176] The feature extraction process includes energy efficiency structured data, other structured data, and unstructured data. Energy efficiency structured data includes fuel oil consumption rate (FCR), energy efficiency operation index (EEOI), carbon intensity (CII), fuel consumption per unit distance, and fuel consumption per unit transport work. These indicators are key structured data for measuring ship energy efficiency. Other structured data, such as time and frequency domain data features, describe data characteristics from the time and frequency dimensions, which facilitates a deeper understanding of ship operation-related data. Unstructured data includes text feature extraction, image and video feature extraction, and audio feature extraction. Features are extracted from unstructured data such as text, images, video, and audio to fully utilize various data types.

[0177] The data mining phase utilizes a variety of data mining methods for different types of data, including time series data mining, spatial data mining, spatiotemporal data mining, empirical data mining, and fusion data mining. Time series data mining processes data with time series characteristics and analyzes how data changes over time. Spatial data mining explores the spatial characteristics and relationships of data, which is helpful for analyzing spatial information such as the geographic location involved in ship operations. Spatiotemporal data mining integrates time and space dimensions to comprehensively analyze the spatiotemporal characteristics of ship operations. Empirical data mining utilizes existing empirical knowledge to conduct data mining and assist in analytical decision-making. Fusion data mining integrates multiple types of data and then performs mining to obtain more comprehensive and accurate information.

[0178] The intelligent analysis phase includes ship energy efficiency impact analysis, voyage energy efficiency improvement analysis, and ship energy efficiency improvement analysis. Ship energy efficiency impact analysis analyzes the impact of various factors on ship energy efficiency by analyzing the aforementioned data mining results. Voyage energy efficiency improvement analysis analyzes how to improve energy efficiency for each voyage. Ship energy efficiency improvement analysis comprehensively analyzes methods and strategies for improving ship energy efficiency.

[0179] above Figure 11 It demonstrates the entire process from data acquisition to the ultimate realization of intelligent analysis of ship energy efficiency. By processing and mining different types of data, it provides a data-driven analysis path for improving ship energy efficiency.

[0180] Please refer to Figure 12 , Figure 12 This is a flowchart of the collaborative optimization of route speed in an embodiment of the present application. Figure 12 The paper presents a process for joint optimization of wind-assisted ship routes and speeds. Starting from data acquisition, through meteorological data analysis, model construction, and multi-objective solution, it ultimately provides auxiliary decision-making recommendations, as follows:

[0181] (1) Data acquisition: The upper left corner of the figure shows the data sources, including flow meters, shaft power meters, speed meters, GPS, and meteorological data related equipment. These devices are used to obtain flow, shaft power, distance, position, and meteorological data during the operation of the ship. The acquired data is stored in the "Acquire Data" module and serves as the basis for subsequent analysis.

[0182] (2) Meteorological data analysis: Part of the data flowing out of the "Acquire Data" module enters the meteorological data analysis link. This link contains two sub-modules:

[0183] Meteorological data spatiotemporal interpolation: The acquired meteorological data is interpolated in the time and space dimensions to fill in the data gaps, making the meteorological data more continuous and complete, thereby more accurately reflecting the meteorological conditions in different time and space.

[0184] Analysis of the spatiotemporal distribution characteristics of wind speed, wind direction, current velocity, and wave height: This module analyzes the spatiotemporal distribution characteristics of key meteorological elements, such as wind speed, wind direction, current velocity, and wave height, to identify their changing patterns and characteristics. Through the processing of these two submodules, the results of the "Ship Energy Efficiency Spatiotemporal Distribution Analysis" are obtained, providing meteorological information support for subsequent model construction.

[0185] (3) Model construction: Based on the results of meteorological data analysis, the model construction phase is entered, which includes three sub-modules:

[0186] Fuel consumption model and ship force analysis: By building a model for the ship's fuel consumption and analyzing the force conditions of the ship during navigation, we can understand the relationship between the ship's energy consumption and the force, providing a physical basis for optimizing route speed.

[0187] Joint optimization model: Determine the decision variables, objective function and constraints, and build a joint optimization model that comprehensively considers multiple factors to guide the optimization decision of route and speed.

[0188] Joint optimization algorithm: Use algorithms such as A* to perform global search and find the optimal solution within the framework of the joint optimization model, thereby constructing a "joint optimization model for wind-assisted ship route speed."

[0189] (4) Multi-objective solution solving part: Using the constructed joint optimization model, enter the multi-objective solution solving stage:

[0190] Model training: Train the model to better adapt to actual conditions and improve the accuracy and reliability of the model.

[0191] Model solving: Solve the model through certain methods to obtain the optimization solution.

[0192] Through comparative analysis of navigation position, navigation speed and route fuel consumption, different plans are compared and analyzed from multiple dimensions such as navigation position, speed and fuel consumption, and the "optimal speed" and "optimal route" are determined through these analyses.

[0193] (5) Decision-making assistance part: Integrate the optimal speed and optimal route information obtained from solving the multi-objective solution, and finally give "decision-making assistance suggestions" to provide a reference for optimizing the navigation plan for the actual navigation of the ship, so as to achieve the improvement of ship energy efficiency and the rational use of resources.

[0194] Figure 12 The entire flow chart, from data acquisition to final decision recommendations, is closely connected with each link, reflecting the complete process from basic data to optimized decision-making.

[0195] In some embodiments, in the above-mentioned shore step of S1030, the method further includes:

[0196] S1041: Based on the ship's operating conditions, energy-saving equipment performance limitations, and environmental factors, define the constraints for energy efficiency matching of multiple energy-saving technologies, including power generation balance constraints, hull dynamics balance constraints, propulsion shaft power balance constraints, and navigation planning constraints.

[0197] S1042: Based on the demand for improving ship energy efficiency, set optimization goals for energy efficiency matching of multiple energy-saving technologies, including the highest fuel saving rate, the lowest initial investment cost, the longest compliance period and the best return on investment.

[0198] S1043, using a multi-objective optimization algorithm to solve the optimal solution for energy efficiency matching of multiple energy-saving technologies, the algorithm includes one or more of the penalty function method, projected gradient method, multiplier method, sequential quadratic programming (SQP), back propagation algorithm (BP), particle swarm optimization (PSO), genetic algorithm (GA), and NSGA-II algorithm.

[0199] S1044 first performs energy efficiency matching of multiple energy-saving technologies under a single operating condition, and then combines multiple operating conditions such as economic operating conditions, port entry and exit conditions, and full-speed operating conditions. Through an energy efficiency optimization method based on the weighted application of multiple energy-saving equipment based on the time proportion of multiple operating conditions, energy efficiency matching of multiple energy-saving technologies under multiple operating conditions is achieved, and ultimately the comprehensive energy efficiency of multiple energy-saving technologies is optimized.

[0200] Please refer to Figure 13 , Figure 13 This is a flowchart of energy efficiency matching optimization for multiple energy-saving technologies in an embodiment of the present application. Figure 13 The paper presents an optimization analysis framework for energy efficiency matching of energy-saving technologies, covering multiple aspects such as constraints, optimization objectives, solution algorithms, operating conditions, and the final energy efficiency optimization method, as follows:

[0201] (1) Constraints part.

[0202] On the far left of the diagram, four constraints are listed:

[0203] Power generation balance constraint: Ensure that the power generation reaches a balanced state during the relevant operation process to maintain the stability and reasonable distribution of power supply.

[0204] Hull dynamic balance constraint: From the perspective of hull dynamics, it ensures the dynamic balance of the hull during navigation, involving multiple balance conditions such as hull force and motion state.

[0205] Propulsion shaft power balance constraint: Balance the power of the propulsion shaft to ensure the stable operation of the propulsion system and reasonable use of power.

[0206] Navigation planning constraints: Consider the planning factors of ship navigation, such as routes, sailing time, stopover points and other related planning restrictions.

[0207] (2) Optimize the target part.

[0208] Next to the constraints, four optimization objectives are set, namely:

[0209] Highest fuel saving rate: Aims to maximize the reduction of fuel consumption rate and improve fuel efficiency by optimizing the application of energy-saving technologies.

[0210] Lowest initial investment cost: strive to minimize the initial capital investment and control cost expenditure when introducing energy-saving technologies.

[0211] Maximum compliance period: Ensure that the energy-saving technologies and related equipment used can comply with relevant regulations and standards for the longest period of time, reducing the cost of updates or renovations due to non-compliance.

[0212] Optimal ROI (return on investment): With the goal of obtaining the best return on investment, the relationship between the input cost and output benefits of energy-saving technology is comprehensively considered.

[0213] (3) Solution algorithm part.

[0214] On the right side of the optimization objective, a variety of solution algorithms are listed, namely:

[0215] Penalty function method: By introducing the penalty function, the constraints are converted into part of the objective function so that the constraints can be satisfied during the solution process.

[0216] Projected Gradient Method: An iterative algorithm for solving optimization problems that finds the optimal solution by projecting the gradient on the feasible region.

[0217] Multiplier method: It is a method for solving constrained optimization problems by introducing multipliers to deal with constraints.

[0218] Sequential Quadratic Programming (SQP): is an algorithm for solving nonlinear programming problems that approaches the optimal solution by iteratively solving a series of quadratic programming subproblems.

[0219] Back propagation algorithm BP: Commonly used in neural networks, it adjusts network parameters by back propagating errors to optimize the objective function.

[0220] Particle Swarm Optimization (PSO): An optimization algorithm that simulates the foraging behavior of bird flocks and searches for the optimal solution by particles in the solution space.

[0221] Genetic Algorithm (GA): It draws on the mechanisms of inheritance, mutation, and selection in the process of biological evolution to optimize and solve problems.

[0222] NSGA-II algorithm: Non-dominated Sorting Genetic Algorithm II, is a multi-objective optimization algorithm used to solve multi-objective optimization problems and find a set of Pareto optimal solutions.

[0223] (4) Working conditions section.

[0224] Further to the right, three working conditions are shown:

[0225] Economic operating conditions: The operating conditions in which a ship pursues economy as the main goal, usually focusing on energy consumption and cost control.

[0226] Port entry and exit conditions: refers to the operating conditions when the ship enters and leaves the port. The navigation status and operating requirements of the ship under these conditions are different from those in open waters.

[0227] Full speed operating condition: The operating condition where the ship runs at maximum speed, which places higher requirements on the power system and energy consumption.

[0228] (5) Energy efficiency optimization method part.

[0229] The rightmost diagram illustrates an energy efficiency optimization method for combining multiple energy-saving devices based on weighted time proportions across multiple operating conditions. First, the energy efficiency of multiple energy-saving technologies is matched for a single operating condition. Then, considering multiple operating conditions, the time proportions under different operating conditions are weighted to further achieve energy efficiency matching for multiple energy-saving technologies across multiple operating conditions, ultimately achieving the goal of optimizing the overall energy efficiency of multiple energy-saving technologies.

[0230] Figure 13 It demonstrates the complete process and logical relationship from setting constraints and goals, to selecting solution algorithms, and then combining different working conditions to ultimately achieve energy efficiency optimization of energy-saving technologies.

[0231] Please refer to Figure 14 , Figure 14 This is a flow chart of the multi-source and multi-modal ship data processing and application architecture according to an embodiment of the present application.

[0232] (1) Ship side

[0233] On a technical level, the ship's end is equipped with an integrated edge-source, multi-source, and multi-modal data acquisition and calculation module capable of receiving a wide variety of data types, including report data (files), high-frequency signal data, low-frequency signal data, and image data. Functionally, the ship's end possesses powerful multi-source and multi-modal data processing capabilities, enabling rapid response to strong real-time edge data. By fully leveraging the characteristics of edge computing, it effectively reduces the pressure on data transmission to other levels, alleviating the burden on subsequent data transmission and processing.

[0234] (2) Cloud

[0235] On a technical level, the cloud utilizes adaptive data field encoding technology, enabling flexible data field processing. It employs lossless / lossy compression and encryption to ensure data security during storage and transmission. It utilizes a data layering and classified storage mechanism to rationally arrange data storage. Furthermore, it provides a fixed IP bridge connecting ship and shore, facilitating data exchange between the two. Functionally, the cloud enables distributed storage of multi-mode, multi-source data, reducing transmission pressure while improving transmission security. It also standardizes data fields, making them more standardized and organized, laying a solid foundation for subsequent shore-based data processing and application.

[0236] (3) Shore

[0237] On the technical level, the shore side has a data association retrieval and data set generation mechanism based on unified coding, which facilitates data retrieval and data set creation; it has built a multi-level tree-like data storage structure of ship-system-module-equipment-measurement point to form a hierarchical data storage system; it uses data and algorithm synchronous editing and version automatic control technology based on read-write protection to ensure the security and version management of data and algorithms; it uses route and speed collaborative optimization technology that integrates meteorological forecast data to optimize the route and speed of ships; it has multi-energy-saving equipment coupling energy efficiency evaluation and matching optimization technology to evaluate and match energy-saving equipment; it uses system construction technology based on low-code components to facilitate rapid system construction. On the functional level, the shore side has convenient data set construction and statistical analysis capabilities to support data management and data application; it ensures data consistency and integrity and supports multi-user parallel operation; it realizes applications in the field of microservices and ship energy efficiency, and has low-code and efficient expansion capabilities, which can flexibly expand system functions according to business needs.

[0238] It is understandable that Figure 14The illustrated ship data processing and application architecture covers the shipside, middle layer, cloud, and shoreside. The shipside collects reports, high-frequency and low-frequency data, and image data through an integrated edge-source, multi-modal data acquisition and computation module. It features multi-source, multi-modal data processing, strong real-time data response, and the use of edge computing to reduce transmission pressure. The middle layer utilizes a unified synchronization engine for multiple databases and multi-protocol real-time data transmission technology to ensure real-time data transmission and compatibility with multiple communication protocols. The cloud layer utilizes technologies such as adaptive data field encoding, lossless / lossy compression encryption, and hierarchical classification storage to bridge the gap between ship and shore, enabling distributed storage of multi-mode and multi-source data, reducing transmission pressure, improving transmission security, and standardizing data fields. The shoreside utilizes various technologies, such as data association retrieval based on unified encoding, enabling convenient data set construction and analysis, supporting data management applications, ensuring data consistency and integrity, enabling microservices and ship energy efficiency applications, and enabling efficient low-code expansion. The overall architecture is dedicated to achieving efficient collection, transmission, storage, and application of multi-source and multi-modal ship data to improve ship energy efficiency.

[0239] In summary, this application proposes a ship comprehensive energy efficiency improvement platform and its data management and application method. The ship comprehensive energy efficiency improvement platform is a ship comprehensive energy efficiency improvement data, algorithm, and business integrated platform based on a distributed time series database. Compared with the mainstream Hadoop and Doris ecological architectures, the platform has outstanding performance in balancing data management and computing capabilities, especially in time series data processing, with advantages such as small size and simple technical foundation. The platform supports ship energy efficiency data mining, route speed collaborative optimization, and energy efficiency matching evaluation of multiple energy-saving technologies, closely adapting to business needs and enabling ship energy efficiency improvement. By integrating components such as Spark, Kafka, HBase, and IoTDB, the platform integrates batch processing and stream processing, simplifies data processing complexity, and improves efficiency. At the same time, the platform supports the collection, transmission, storage, and processing of multimodal data, and uses technologies such as RTP, SIP, MongoDB, and ZLMediaKit to make up for the shortcomings of MQTT and IoTDB in video data and unstructured data processing, and has powerful multimodal data comprehensive processing capabilities. In addition, the platform is designed based on a microservice architecture, and has convenient data source, algorithm and operation resource management expansion capabilities, as well as flexible scalability of upper-level applications and business services. It has strong overall scalability and adaptability.

[0240] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A ship comprehensive energy efficiency improvement platform, characterized by: Including ship side, cloud side and shore side: The ship side includes a data acquisition module, an edge computing module, and a data scheduling module; wherein the data acquisition module is used to realize multi-source acquisition of equipment data using multiple industrial protocols and communication protocols; the edge computing module is used to perform data processing based on edge computing; the data scheduling module is used to complete the scheduling and forwarding of edge data through a data bus based on data communication middleware and edge database construction; The cloud includes a data aggregation module, a data isolation module, and a distributed storage module; wherein the data aggregation module is used to aggregate ship-side data through the message queue telemetry transmission protocol and database synchronization capability; the data isolation module is used to isolate and process the aggregated data; and the distributed storage module is used to use a distributed time series database to complete data storage and active-active backup. The shore end includes a data center layer, an algorithm center layer and a business center layer; wherein, the data center layer is used to perform standardized management of data; the algorithm center layer is used to complete the creation, version control and testing experiments of algorithms and data flows, and realize version control of algorithms; the business center layer is used to realize microservice applications of businesses based on a microservice architecture.

2. The ship comprehensive energy efficiency improvement platform according to claim 1 is characterized in that: The ship, the cloud and the shore are connected via a two-way data transmission mechanism to achieve the transmission of high-frequency or low-frequency structured data, unstructured data and algorithm model update packages; The platform adopts a distributed time-series database, microservice architecture and multimodal data processing mechanism to achieve balanced development of data management, computing power and business needs.

3. The ship comprehensive energy efficiency improvement platform according to claim 1, characterized in that: The ship end also includes an edge storage module; wherein: The data acquisition module is used to complete multi-source acquisition of device data through industrial bus protocol, message queue telemetry transmission protocol and communication protocol; The edge computing module includes a digital filtering module, a data cleaning module, a data deduplication module, a data normalization module, a data discretization module, a comprehensive energy efficiency data feature extraction module and a fault alarm module, which are used to process the collected data; The edge storage module uses a stand-alone time series database and a relational database to store real-time data and calculation data; The data scheduling module adopts a subscription-publishing model to complete the scheduling of edge data.

4. The ship comprehensive energy efficiency improvement platform according to claim 1, characterized in that: The cloud also includes an edge management module, a data management module, and a security protection module; wherein: The data aggregation module completes the aggregation of ship-side data through the message queue telemetry transmission protocol and database synchronization capability; The edge management module includes a device management module, a connection management module and an application management module, which are used to manage edge devices, connections and applications; The data management module includes a metadata management module, a data cache module, a data model management module, a rule engine module and a time alignment module, which are used to perform standardized management of data; The security protection module is used to ensure data and network security through access control, data encryption, backup and recovery, security auditing and anomaly detection.

5. The ship comprehensive energy efficiency improvement platform according to claim 1, characterized in that: The data center layer includes a metadata management module, a data compression and storage module, a data cache module, and a time series creation module for implementing standardized data management. The data center layer also has a complex time processing module, a sequence data matching and discovery module, and anomaly detection module for completing complex time processing, sequence data matching and discovery, and anomaly detection. The algorithm center layer includes an algorithm and data flow creation module, a version control module and a test experiment module; The business center layer is based on a microservice architecture, including a data sampling service module, a full data service module, an indicator maintenance service module, a rule maintenance service module and an abnormal alarm service module, which is used to provide data sampling, full data, indicator maintenance, rule maintenance and abnormal alarm services.

6. The ship comprehensive energy efficiency improvement platform according to claim 1, characterized in that: The data bidirectional transmission module adopts one or more data transmission protocols among the message queue telemetry transmission protocol, transmission control protocol, user datagram protocol, session initiation protocol, real-time transport protocol, and file transfer protocol.

7. The ship comprehensive energy efficiency improvement platform according to claim 1 or 5, characterized in that: The data center layer on the shore also includes a ship energy efficiency data mining module, which is used to process the ship's original data and edge computing return data, extract key features reflecting the ship's energy efficiency, and mine the association between energy efficiency features and different types of data.

8. The ship comprehensive energy efficiency improvement platform according to claim 1 or 5, characterized in that: The algorithm center layer on the shore also includes a route speed collaborative optimization module, which is used to integrate meteorological data and provide ships with optimal or near-optimal route speed auxiliary decision-making recommendations based on electronic chart raster modeling, ship energy consumption model and collaborative optimization gray box model.

9. The ship comprehensive energy efficiency improvement platform according to claim 1 or 5, characterized in that: The algorithm center layer on the shore also includes a multi-energy-saving technology energy efficiency evaluation module, which is used to provide ships with optimal or near-optimal ship energy-saving equipment installation design plans based on independent energy-saving equipment energy efficiency modeling, multi-energy-saving equipment combination energy efficiency modeling, multi-energy-saving scheme energy efficiency evaluation and multi-energy-saving technology energy efficiency matching optimization.

10. The ship comprehensive energy efficiency improvement platform according to claim 1 or 5, characterized in that: The business center layer on the shore also includes a visualization module, which is used to provide ship energy efficiency data mining, multi-energy-saving technology energy efficiency matching evaluation, route speed collaborative optimization and visualization business applications based on the output data of the ship energy efficiency data mining module, the multi-energy-saving technology energy efficiency matching evaluation module and the route speed collaborative optimization module.

11. The ship comprehensive energy efficiency improvement platform according to claim 1 or 4, characterized in that: The cloud also includes a data source expansion module, a data processing algorithm expansion module and an operation resource management expansion module; wherein: The data source expansion module is used to expand the data source through multiple industrial protocols, multimodal data communication protocols, database synchronization, file reading and API integration; The data processing algorithm expansion module is used to expand the data processing algorithm through the creation, version control and testing of algorithms and data flows; The operation resource management expansion module is used to achieve the expansion of operation resources through load balancing, resource scheduling and security protection.

12. The ship comprehensive energy efficiency improvement platform according to claim 1 or 4, characterized in that: The cloud also includes a multimodal data processing module for completing the collection, transmission, storage and processing of unstructured data such as videos, images and files; The multimodal data processing module adopts the real-time transport protocol and the session initialization protocol to alleviate the pressure of the message queue telemetry transmission protocol on video data transmission, and makes up for the shortcomings of the time series database technology in the compression, transmission and storage of unstructured data through document database and streaming media processing technology.

13. The ship comprehensive energy efficiency improvement platform according to claim 1 or 4, characterized in that: The cloud also includes a batch-stream integrated data processing module for integrating batch processing and stream processing capabilities; The batch-stream integrated data processing module adopts a big data processing framework, a message queue system, a distributed database, and a time series database technology component to simplify the complexity of data processing.

14. The ship comprehensive energy efficiency improvement platform according to claim 1 or 4, characterized in that: The cloud also includes a time series data processing module to support the write performance and query throughput of tens of millions of points per second of a three-node server cluster, and complete millisecond-level aggregation of tens of billions of data points.

15. A data management and application method for a ship comprehensive energy efficiency improvement platform, characterized in that: The method is applied to the ship comprehensive energy efficiency improvement platform according to any one of claims 1 to 14, comprising: Onboard steps: Utilize multiple industrial and communication protocols to complete multi-source acquisition of equipment data; perform data processing based on edge computing; and send the processed data through a data bus, based on data communication middleware and edge database construction, to complete edge data scheduling and forwarding. Cloud-side steps: Aggregate ship-side data through load balancing, resource management, and security protection mechanisms; isolate and process aggregated data; and use a distributed time-series database for data storage and active-active backup. Onshore steps: Build a data center layer to manage data in a standardized manner; build an algorithm center layer to complete the creation, version control and testing experiments of algorithms and data flows, and implement algorithm version control; build a business center layer to implement microservice applications of the business based on the microservice architecture.

16. The data management and application method of the ship comprehensive energy efficiency improvement platform according to claim 15 is characterized in that: In the ship-side step, the data processing based on edge computing includes: Perform real-time pre-processing of collected equipment data, including missing data processing, abnormal data processing, noise data processing and data standardization; Perform feature extraction on the pre-processed data to extract key features reflecting ship energy efficiency, including fuel consumption rate, energy efficiency operation index and carbon intensity; The extracted feature data is forwarded to the cloud via the data bus, while the original data is stored in the edge database for local analysis and backup.

17. The data management and application method of the ship comprehensive energy efficiency improvement platform according to claim 15 is characterized in that: In the cloud step, isolating the aggregated data includes: Adopting data isolation module, data from different ship ends are isolated and stored according to data type and source; The isolated data is compressed and encrypted to achieve data security and storage efficiency; The compressed and encrypted data is stored in a distributed time series database and synchronized with active-active backup to achieve high data availability, integrity, and security.

18. The data management and application method of the ship comprehensive energy efficiency improvement platform according to claim 15 is characterized in that: In the onshore step, the construction of the business center layer and the implementation of microservice applications based on the microservice architecture include: Modularize the functions of ship energy efficiency analysis, route optimization, speed optimization, speed and route coordination optimization, and multi-energy-saving equipment energy efficiency evaluation and matching optimization to form independent microservices; The API gateway enables communication and data interaction between microservices, achieving flexibility and scalability of business logic; Deploy microservices based on containerization technology, combined with load balancing and automatic scaling mechanisms, to achieve performance and stability of the business center layer.

19. The data management and application method of the ship comprehensive energy efficiency improvement platform according to claim 15 is characterized in that: In the onshore step, the method further comprises: Based on the ship's operating conditions, energy-saving equipment performance limitations, and environmental factors, define the energy efficiency matching constraints for multiple energy-saving technologies, including power generation balance constraints, hull dynamics balance constraints, propulsion shaft power balance constraints, and navigation planning constraints. Based on the demand for improving ship energy efficiency, we set optimization goals for energy efficiency matching of multiple energy-saving technologies, including the highest fuel saving rate, the lowest initial investment cost, the longest compliance period, and the best return on investment; A multi-objective optimization algorithm is used to find the optimal solution for energy efficiency matching of multiple energy-saving technologies. The algorithm includes one or more of the following: penalty function method, projected gradient method, multiplier method, sequential quadratic programming (SQP), back propagation algorithm (BP), particle swarm optimization (PSO), genetic algorithm (GA), and NSGA-II algorithm. First, energy efficiency matching of multiple energy-saving technologies in a single operating condition is carried out, and then combined with multiple operating conditions such as economic conditions, port entry and exit conditions, and full-speed conditions, an energy efficiency optimization method based on the combined application of multiple energy-saving equipment with weighted time proportions of multiple operating conditions is used to achieve energy efficiency matching of multiple energy-saving technologies in multiple operating conditions, and ultimately achieve the optimal comprehensive energy efficiency of multiple energy-saving technologies.

Images