Navigation big data intelligent interaction system and method based on model context protocol

Abstract

The invention relates to the technical field of shipping big data and artificial intelligence, in particular to a shipping big data intelligent interaction system and method based on a model context protocol. The system comprises a shipping big data platform, a ship monitoring system, a model context protocol module, an AI model module, a user interaction terminal and a safety management module. The method comprises the steps of data acquisition and preprocessing, request triggering and context packaging, AI model processing and reasoning, result packaging and returning, result distribution and interactive presentation, and component updating and hot plugging. Aiming at the problems of difficulty in docking between a ship monitoring system and an artificial intelligence model, relatively high data security risk, insufficient system expansibility and the like in the prior art, the invention proposes that a data interaction format and a communication process are standardized through a unified model context protocol, and safe and efficient docking between a shipping big data platform and an AI model is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of shipping big data and artificial intelligence, and particularly to a shipping big data intelligent interaction system and method based on the Model Context Protocol (MCP). Background Art

[0002] As the shipping industry enters the digital age, modern ships are equipped with a large number of sensors and monitoring systems, and data such as the ship's operating status, equipment performance, and environmental parameters are collected in real time through a shipping big data platform. Although the existing technologies have the capabilities of real-time ship monitoring and data analysis, there are still problems such as data security risks, lack of unity in communication standards, high integration complexity, and difficulty in component upgrading in the deep integration of ship monitoring systems and artificial intelligence (AI) algorithms (especially large language models LLM).

[0003] In the prior art, although there are public ship monitoring systems that can obtain the ship's operating data in real time, and there are also public artificial intelligence algorithms for data analysis and decision support, the deep integration of the two still faces many challenges. For example, in traditional solutions, if an advanced AI model such as a large language model (LLM) is to be introduced into shipping data analysis, a large amount of ship data needs to be transmitted to the AI model for processing. However, this direct transmission method may lead to data security risks - sensitive navigation data and business secrets are at risk of leakage during transmission. In addition, the interfaces of ship equipment and AI models provided by different suppliers are different, lacking a unified communication standard, resulting in complex system integration and high costs. When upgrading or replacing components, it often requires significant modification of the overall system architecture. Moreover, most current shipping data analysis platforms adopt a fixed architecture design, which cannot support the plug-and-play and dynamic expansion of system components. If a new functional module (such as a new AI model algorithm or a third-party data source) is to be added to the shipping big data platform, it usually requires downtime maintenance or re-development of interfaces, seriously affecting the continuous operation and scalability of the system. Especially in scenarios involving multi-vendor equipment, incompatible communication protocols and data formats create data silos, hindering the full exploration of the value of shipping big data.

[0004] Based on the above background, there is an urgent need for a new technical solution that can achieve the efficient docking of a shipping big data platform and an AI model (such as LLM) while ensuring data security, support the standardized access of different manufacturers' equipment and models, and allow the hot-pluggable upgrade of system components, so as to provide an intelligent human-computer interaction function and improve the timeliness and intelligence level of shipping data analysis. Summary of the Invention

[0005] The present invention solves the problems of data security risks, lack of unity in communication standards, high integration complexity, and difficulty in component upgrading in the deep integration of existing ship monitoring systems and artificial intelligence (AI) algorithms (especially large language models LLM). It provides a shipping big data intelligent interaction system and method based on the model context protocol (MCP) to achieve intelligent interaction, component hot plugging, and multi-vendor compatibility under the premise of data security.

[0006] The technical solutions claimed by the present invention are as follows:

[0007] A shipping big data intelligent interaction system based on the model context protocol, comprising: a ship monitoring system for collecting ship status data and transmitting it in the form of a wired or wireless network, a shipping big data platform connected to the ship monitoring system and receiving the transmitted ship status data and further preprocessing the ship status data, a data storage and management module connected to the shipping big data platform and storing and managing the preprocessed data, a model context protocol module connected to the data storage and management module, encapsulating and parsing the data stored in the data storage and management module based on the model context protocol to obtain context data and transmitting the context data, an AI model module connected to the model context protocol module and receiving the transmitted context data and used for processing the context data to generate a result data packet in the model context protocol format, a user interaction terminal connected to the data storage and management module, and a security management module connected to the model context protocol module and the AI model module for ensuring the security and compliance of data interaction;

[0008] The model context protocol module is also used for data interaction and conversion between the shipping big data platform and the AI model module; the AI model module is connected to the model context protocol module and returns the generated result data packet to the model context protocol module. The model context protocol module is connected to the data storage and management module, receives the result data packet, unpacks and extracts the result content, and then transmits it back to the data storage and management module for storage.

[0009] Preferably, the ship monitoring system includes an on-board sensor network and a control system; the on-board sensor network is responsible for collecting various status data of the ship, monitoring the operating conditions of the ship, the performance of mechanical equipment, and environmental condition information through sensors and monitoring devices, and transmitting this information to the shipping big data platform for processing and analysis; the control system is responsible for processing and managing various operations of the ship and ensuring the coordinated operation of the ship system. The control system receives data from the on-board sensor network, and after necessary processing, executes corresponding control commands; the ship status data includes: ship navigation, mechanical equipment status, and environmental data; the preprocessing includes: data cleaning, format standardization, outlier filtering, and time synchronization; the data storage and management module stores data according to time series or subject categories.

[0010] Preferably, the AI model module is an artificial intelligence model component deployed in the shipping big data platform environment, specifically: a large language model trained specifically or a machine learning model for specific tasks; the AI model module is deployed remotely or in the form of cloud services and communicates with the model context protocol module through a predefined model context protocol interface.

[0011] Preferably, the model context protocol module supports the hot plugging of new functional modules or data sources, enabling new components to access the shipping big data platform by providing standard interfaces.

[0012] Preferably, the security management module also includes measures for data security protection: during data encapsulation and transmission, sensitive ship data is encrypted, desensitized, etc., and access control policies are used to ensure that only authorized models and modules can obtain specific context data.

[0013] Preferably, the security management module adopts a multi-level security strategy, including data encryption transmission, identity authentication and permission control, sensitive information desensitization, and operation log recording; during the stage when the model context protocol module parses and obtains context data, the security management module is used to perform desensitization processing on highly sensitive raw data; the security management module is also used to check and authenticate the result data returned by the AI model module.

[0014] Preferably, the user interaction terminal is used for the human-machine interaction interface, specifically the display console in the ship control room, the computer terminal of the operation and maintenance personnel, or the application program on the mobile device.

[0015] Preferably, the user uses the terminal to issue queries, instructions or receive alarm notifications to the shipping big data platform. Specifically, when the user asks a natural language question or selects a predefined analysis requirement, the shipping big data platform sends the request to the model context protocol module for processing. Finally, the analysis result of the AI model module is transmitted to the user interaction terminal via the shipping big data platform; the user interaction terminal displays it in an easy-to-understand form, including: natural language answers, visual charts or operation suggestions.

[0016] The present invention also provides a shipping big data intelligent interaction method based on a model context protocol, comprising the following steps:

[0017] S1: Data collection and preprocessing: The ship monitoring system continuously collects ship status data and transmits the data to the shipping big data platform;

[0018] S2: Trigger request and context encapsulation: When a user initiates a query request through the user interaction terminal or when the security management module detects an anomaly that requires AI analysis, the shipping big data platform extracts the relevant historical and real-time data it has collected and submits it to the model context protocol module. The model context protocol module selects the data most relevant to the request from the data storage and management module according to the request type, and constructs a context data packet in accordance with the model context protocol format;

[0019] S3: AI model processing and reasoning: The model context protocol module sends the context data packet to the AI model module through a standard interface and calls the corresponding model algorithm for processing. The AI model module parses the received context packet, extracts the query and data contained therein, and then uses its trained knowledge and reasoning ability to generate answers or decision suggestions;

[0020] S4: Result packaging and return: The AI model module packages the inference results into a response data packet according to the model context protocol requirements and returns it to the model context protocol module. After receiving the response data packet, the model context protocol module first verifies the integrity and source legitimacy of the data packet, then parses and extracts the result data, and stores it in the data storage and management module;

[0021] S5: Result distribution and interactive presentation: The data storage and management module distributes the stored data results to the corresponding terminals or modules; the system can also actively integrate AI analysis conclusions into the operation process, improving the level of intelligence of shipping management.

[0022] S6: Component Update and Hot Plugging: When new functional modules need to be introduced, AI models replaced, or new data sources integrated, the new components first implement interface adaptation according to the requirements of the model context protocol. When the new components are connected to the shipping big data platform, they are registered through the model context protocol module, enabling the shipping big data platform to identify their functions and data interaction formats. After registration, the new components can work in real-time collaboration with the existing system.

[0023] The context data packet in the above method includes: raw ship status data, a description of the problem, an identifier of the data source, and access permission information attached by the security management module.

[0024] Beneficial Effects

[0025] The present invention provides a shipping big data intelligent interaction system and method based on the Model Context Protocol (MCP). First, by introducing the model context protocol, it standardizes the data exchange format and process between the shipping big data platform and the artificial intelligence model, thus establishing a unified communication bridge between different system components, achieving seamless docking between the shipping big data platform and different AI models and devices, greatly reducing the difficulty of system integration, enhancing the compatibility of the system in a multi-vendor environment, and solving the problems of the lack of unity in communication standards, high integration complexity, and difficulty in component upgrade in the deep integration of the existing ship monitoring system and artificial intelligence (AI) algorithms (especially large language models LLM). Second, the present invention allows for the hot plugging of AI model modules and the ship monitoring system, and the system can update or add functional components without downtime, improving the flexibility and expandability of the platform. Third, the use of context encapsulation and access control mechanisms ensures the security of data during the AI model processing, avoids the leakage of sensitive information, improves the reliability of data interaction, and solves the problem of data security risks in the deep integration of the existing ship monitoring system and artificial intelligence (AI) algorithms (especially large language models LLM). Finally, by introducing AI technologies such as large language models, the platform can provide more intelligent and user-friendly interaction methods, achieve in-depth analysis of massive shipping data and real-time decision support, and overall improve the intelligent level of shipping operations.

[0026] The model context protocol module supports the hot plugging of new functional modules or data sources, enabling new components to access the shipping big data platform by providing standard interfaces. The new components only need to implement the adaptation interface with the model context protocol and register, and then they can work in real-time collaboration without the need for downtime or system reconstruction.

[0027] In summary, the present invention closely combines the shipping big data platform with the AI model through the MCP protocol, which not only ensures the security and standardization of data interaction, but also endows the system with high flexibility and intelligent interaction capabilities. The system provided by the present invention enables the shipping big data platform to safely and efficiently analyze and interact with ship monitoring data using AI algorithms such as large language models, while ensuring that the system has good scalability and compatibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 FIG. is a schematic diagram of the architecture of the intelligent interaction system for shipping big data based on the model context protocol according to an embodiment of the present invention.

[0029] Figure 2 FIG. is a schematic diagram of the flow of the intelligent interaction method for shipping big data based on the model context protocol according to an embodiment of the present invention. DETAILED IMPLEMENTATION METHODS

[0031] The embodiments of the present invention will be further described below in conjunction with the accompanying drawings. It should be understood that these embodiments are used to explain the present invention, but do not constitute a limitation on the protection scope of the present invention.

[0032] The First Group of Embodiments: An Intelligent Interaction System for Shipping Big Data Based on the Model Context Protocol

[0033] This group of embodiments provides an intelligent interaction system for shipping big data based on the model context protocol (MCP). As Figure 1 shown, the system includes: a ship monitoring system for collecting ship status data and transmitting it in the form of a wired or wireless network, a shipping big data platform connected to the ship monitoring system to receive the transmitted ship status data and further preprocess the ship status data, a data storage and management module connected to the shipping big data platform to store and manage the preprocessed data, a model context protocol module connected to the data storage and management module to encapsulate and parse the data stored in the data storage and management module based on the model context protocol to obtain context data and transmit the context data, an AI model module connected to the model context protocol module to receive the transmitted context data and used to process the context data to generate a result data packet in the model context protocol format, a user interaction terminal connected to the data storage and management module, and a security management module connected to the model context protocol module and the AI model module to ensure the security and compliance of data interaction;

[0034] The model context protocol module is also used for data interaction and conversion between the shipping big data platform and the AI model module; the AI model module is connected to the model context protocol module and returns the generated result data packet to the model context protocol module. The model context protocol module is connected to the data storage and management module, and after receiving the result data packet, unpacks it to extract the result content and then passes it back to the data storage and management module for storage.

[0035] The shipping big data platform is used to receive various types of data from the ship monitoring system; the data storage and management module is used to store and manage the data received by the shipping big data platform.

[0036] In a specific embodiment of the present invention, the ship monitoring system may include an on-board sensor network and a control system; the on-board sensor network is responsible for collecting various state data of the ship, monitoring the operating conditions of the ship, the performance of mechanical equipment, environmental conditions, etc. through sensors and monitoring devices, and transmitting these data to the shipping big data platform for processing and analysis; the control system is responsible for processing and managing various operations of the ship and ensuring the coordinated operation of the ship system. The control system receives data from the on-board sensor network and executes corresponding control commands after necessary processing; for example: navigation equipment monitoring, engine performance sensors, cargo environment sensors, and communication and navigation systems correspond to the on-board sensor network; navigation equipment monitoring and communication and navigation systems are part of the control system. Therefore, navigation equipment monitoring and communication and navigation systems correspond to the control system.

[0037] The ship monitoring system sends the collected ship state data to the shipping big data platform through a wired or wireless network. The shipping big data platform preprocesses the received data (such as data cleaning, format standardization, outlier filtering, and time synchronization); the data storage and management module stores it according to time series or topic categories to prepare for subsequent data analysis and extraction.

[0038] The model context protocol module is used to realize data interaction and conversion between the shipping big data platform and the AI model module. The model context protocol module encapsulates and parses data according to the model context protocol (MCP); the model context protocol module supports hot plugging of new function modules or data sources, and enables new components to access the shipping big data platform by providing standard interfaces; new components only need to implement the adaptation interface with the model context protocol and register, and then they can work collaboratively in real time without shutting down or reconstructing the system.

[0039] In a specific embodiment of the present invention, on the one hand, when a user issues a query request through a user interaction terminal or the platform triggers an internal analysis task, the model context protocol module extracts relevant data from the shipping big data platform and integrates it to form context information. For example, for the user's question "What is the meteorological risk of the current shipping route", the model context protocol module will collect relevant information such as the current shipping route location, meteorological sensor data, and historical weather records, and package it into a context data packet in the MCP format. This context data packet contains the data content and the metadata describing this content (such as data type identifier, unit, timestamp, security level, etc.), thus providing a clear semantic context for the AI model module.

[0040] On the other hand, the model context protocol module is also responsible for parsing and converting the response results returned by the AI model module. After processing the received context data, the AI model module generates a result data packet (such as an answer to a question or an analysis conclusion of an abnormal situation), and the result data packet is returned in the MCP format. After receiving the result data packet, the model context protocol module unpacks it to extract the result content and the attached confidence level, source and other information, and passes it back to the data storage and management module for further processing or storage. Through the encapsulation and parsing functions of the model context protocol module, it is ensured that the information exchanged between different modules has a consistent format and semantics, eliminating data understanding ambiguity.

[0041] The AI model module can be an artificial intelligence model component deployed in the shipping big data platform environment. For example, it can be a large language model (LLM) that has been specially trained and has the ability to analyze data in the shipping field and perform natural language question and answer; or it can be a machine learning model for specific tasks (such as an algorithm model for shipping route optimization, a neural network model for equipment failure prediction, etc.). The AI model module communicates with the model context protocol module through a predefined MCP interface: when receiving a context data packet, it triggers the inference calculation process of the model. During this process, the AI model module only performs calculations based on the information provided in the context, thus avoiding dependence on irrelevant data. At the same time, the security management module cooperates with the AI model module when necessary to perform access control and permission verification on the context data, such as preventing the AI model from reading unauthorized sensitive fields, ensuring that the calculation process of the model is carried out within a controlled range.

[0042] The user interaction terminal is used for the human-machine interaction interface and can be a display console in the ship control room, a computer terminal of the operation and maintenance personnel, or an application on a mobile device. Through the user interaction terminal, the user can send queries, instructions or receive warning notifications to the shipping big data platform. In a specific embodiment of the present invention, when the user asks a natural language question or selects a predefined analysis requirement, the shipping big data platform sends the request to the model context protocol module for processing. Finally, the analysis result of the AI model module is transmitted to the user interaction terminal via the shipping big data platform and presented in an easy-to-understand form, such as a natural language answer, a visualization chart or an operation suggestion, etc., to realize the intelligent interaction function. For example, when it is detected that a certain device parameter is abnormal, the platform can automatically prompt the user with possible fault reasons and processing suggestions in a dialogue form through the user interaction terminal, and generate an intelligent feedback through the analysis of the real-time data by the AI model module.

[0043] The security management module runs through the entire data flow process of the system and is used to ensure the security and compliance of data interaction. The security management module adopts multi-level security policies, including data encryption transmission, identity authentication and permission control, sensitive information desensitization, and operation log recording, etc. In the context data packet generation stage, the security management module will perform desensitization processing on the original data involving highly sensitive information, such as hiding part of the identification information or blurring the accuracy, and then encapsulate and send it to the AI model by the model context protocol module. For the results returned by the AI model, the security management module will also check to prevent unauthorized information from being leaked to the end user or other system parts. In addition, if the AI model module is deployed remotely or in the form of cloud services, the security management module can establish a secure channel (such as a VPN or a dedicated encryption channel) to transmit context data and result data, and authenticate the remote model service to ensure that the data is not intercepted by a third party during the transmission process. Through the above measures, the present invention maximally reduces the risk of data leakage and improper use while giving full play to the powerful analysis ability of the AI model.

[0044] The second group of embodiments: A shipping big data intelligent interaction method based on the model context protocol

[0045] This group of embodiments provides a shipping big data intelligent interaction method based on the model context protocol, as Figure 2 shown, including the following steps:

[0046] S1: Data collection and preprocessing: The ship monitoring system continuously collects ship status data and transmits the data to the shipping big data platform; the shipping big data platform preprocesses the original ship status data, such as data cleaning, format standardization, outlier filtering, and time synchronization, to ensure that the data quality meets the analysis requirements.

[0047] S2: Trigger Request and Context Encapsulation: When the user initiates a query request through the user interaction terminal (such as asking whether the ship's fuel consumption is normal) or when the safety management module detects an anomaly that requires AI analysis, the shipping big data platform extracts relevant historical and real-time data and submits it to the model context protocol module. The model context protocol module selects the data most relevant to the request (such as fuel consumption sensor readings, ship speed, historical average fuel consumption, and the status of related equipment) from the data storage and management module according to the request type, and constructs a context data packet in the MCP format. The context data packet not only contains the above raw data, but also includes a description of the problem, an identifier of the data source, and access permission information attached by the safety management module, etc.

[0048] S3: AI Model Processing and Inference: The model context protocol module sends the context data packet to the AI model module through a standard interface and invokes the corresponding model algorithm for processing. The AI model module parses the received context packet, extracts the questions and data contained therein, and then uses the knowledge and reasoning ability obtained through its training to generate answers or decision suggestions. In a specific embodiment of the present invention, if the AI model module is a large language model, then it will generate an answer in natural language form based on the context data and problem description, as well as an analysis and explanation of the relevant data. If the AI model module is a specific algorithm model, such as a model for detecting equipment failures, it will give an evaluation result of the equipment status and a prediction of the probability of failure based on the data provided in the context. During the entire processing process, the safety management module monitors the invocation of the AI model to ensure that the AI model can only access data fields within the authorized scope, and records the logs of model invocation and response for auditing.

[0049] S4: Result Encapsulation and Return: The AI model module packages the inference result into a response data packet according to the requirements of the MCP protocol and returns it to the model context protocol module. The response data packet includes the output result content of the model, as well as relevant meta-information, such as the confidence score of the result, model identifier, timestamp, etc. After receiving the response data packet, the model context protocol module first verifies the integrity and source legality of the data packet (with the help of the digital signature or verification mechanism provided by the safety management module), and then parses and extracts the result data. For an answer in natural language form, the model context protocol module refines it into structured data or maintains the original human-readable format, and stores it in the result database of the shipping big data platform (the data storage and management module connected to the shipping big data platform) or forwards it directly as needed.

[0050] S5: Result Distribution and Interactive Presentation: The shipping big data platform obtains the results of the parsed AI model and distributes the results to the corresponding terminals or modules according to specific applications. On the one hand, the shipping big data platform feeds back the results to the user who initiated the request through the user interaction terminal. For example, for the fuel consumption query problem, the user interaction terminal can display: "The current fuel consumption is normal. The average fuel consumption is X tons per day, slightly lower than the historical average. It is recommended to maintain the current speed." In addition, the shipping big data platform can also trigger other automated operations or alarm notifications according to the results. For example, when the AI model determines that there is an abnormal risk for a certain parameter, the shipping big data platform immediately sends an alarm message to the relevant person in charge. Through the above methods, the present invention realizes the intelligent interaction between the shipping big data platform and the user: the user can obtain professional analysis results in natural language, and the system can also actively integrate the AI analysis conclusions into the operation process, improving the intelligent level of shipping management.

[0051] S6: Component Update and Hot Plugging: When new functional modules need to be introduced, AI models need to be replaced, or new data sources need to be integrated, first, the new components implement interface adaptation according to the MCP requirements. When the new components are connected to the shipping big data platform, they are registered through the model context protocol module, enabling the platform to identify their functions and data interaction formats. After registration, the new components can work in real-time coordination with the existing system without suspending the platform service. For example, when replacing the route weather prediction model with higher precision, only need to deploy the new model as an instance of the AI model module and register it through the model context protocol module; because a unified context protocol is used, the format of the context data packet sent by the shipping big data platform to the model remains unchanged, and the new model can directly parse and process it, which reflects the advantage of component-level hot plugging. Similarly, when adding sensor devices provided by other suppliers, as long as the device is supported to output data conforming to the MCP format through software or middleware, the shipping big data platform can seamlessly incorporate its data into the analysis context. It can be seen that the architecture of the present invention has good compatibility with multi-vendor hardware and software, facilitating the horizontal expansion and upgrade of the system according to requirements, and protecting the continuity of the user's investment in the existing system.

[0052] It should be noted that the model context protocol (MCP) described in the present invention can be customized and extended according to specific implementations. For example, MCP can adopt a self-describing data format based on JSON or XML to facilitate reading and writing and understanding by humans and machines; MCP can also define the timing specifications of interactions, such as the request-response mode or the publish-subscribe mode, to adapt to different types of AI tasks (synchronous question and answer or asynchronous monitoring). In addition, the form of the AI model module is not limited to a single model. It can represent a model cluster or a model service platform, and can dynamically select a suitable model to perform inference according to the context request. These changes and variations all fall within the protection scope of the present invention.

[0053] Those skilled in the art should understand that, without departing from the principle of the present invention, many equivalent deformations or modifications can be made to the technical solutions of the above embodiments, and these should all be regarded as falling within the protection scope of the present invention. The above description of the present invention is exemplary rather than restrictive, and the protection scope is defined by the appended claims.

Claims

1. A shipping big data intelligent interaction system based on a model context protocol, characterized in that, Including: A ship monitoring system for collecting ship status data and transmitting it in the form of a wired or wireless network, a shipping big data platform connected to the ship monitoring system to receive the transmitted ship status data and further preprocess the ship status data, a data storage and management module connected to the shipping big data platform to store and manage the preprocessed data, a model context protocol module connected to the data storage and management module to encapsulate and parse the data stored in the data storage and management module based on the model context protocol to obtain context data and transmit the context data, an AI model module connected to the model context protocol module to receive the transmitted context data and used to process the context data to generate a result data packet in the model context protocol format, a user interaction terminal connected to the data storage and management module, and a security management module connected to the model context protocol module and the AI model module to ensure the security and compliance of data interaction; The model context protocol module is also used for data interaction and conversion between the shipping big data platform and the AI model module; the AI model module is connected to the model context protocol module and returns the generated result data packet to the model context protocol module, the model context protocol module is connected to the data storage and management module, and after receiving the result data packet, unpacks and extracts the result content and then transmits it back to the data storage and management module for storage.

2. The intelligent interactive system for shipping big data based on the model context protocol according to claim 1, wherein The ship monitoring system includes an on-board sensor network and a control system; the on-board sensor network is responsible for collecting various status data of the ship, monitoring the operating conditions, mechanical equipment performance, and environmental condition information of the ship through sensors and monitoring devices, and transmitting this information to the shipping big data platform for processing and analysis; the control system is responsible for processing and managing various operations of the ship and ensuring the coordinated operation of each system of the ship, the control system receives data from the on-board sensor network, and after necessary processing, executes corresponding control commands; The ship status data includes: ship navigation, mechanical equipment status, and environmental data; the preprocessing includes: data cleaning, format standardization, outlier filtering, and time synchronization; the data storage and management module stores the data according to time series or subject categories.

3. The intelligent interaction system for shipping big data based on the model context protocol according to claim 1, characterized in that, The AI model module is an artificial intelligence model component deployed in the shipping big data platform environment, specifically: a large language model trained specifically or a machine learning model for specific tasks; the AI model module is deployed remotely or in the form of cloud services and communicates with the model context protocol module through a predefined model context protocol interface.

4. The intelligent interaction system for shipping big data based on the model context protocol according to claim 1, characterized in that The model context protocol module supports the hot plugging of new functional modules or data sources, and enables new components to access the shipping big data platform by providing standard interfaces.

5. The intelligent interactive system for shipping big data based on the model context protocol according to any one of claims 1-4, characterized in that, The security management module also includes measures for data security protection: during the data encapsulation and transmission process, sensitive ship data is processed such as encryption and desensitization, and access control policies are used to ensure that only authorized models and modules can obtain specific context data.

6. The intelligent interactive system for shipping big data based on the model context protocol according to claim 5, characterized in that, The security management module adopts a multi-level security strategy, including data encryption transmission, identity authentication and authority control, sensitive information desensitization and operation log recording; when the model context protocol module parses the context data, the security management module is used to perform desensitization processing on the highly sensitive original data; The security management module is also used to check and authenticate the result data returned by the AI model module.

7. The intelligent interaction system for shipping big data based on the model context protocol according to claim 6, wherein, The user interaction terminal is used for a human-computer interaction interface, specifically a display console in a ship control room, a computer terminal of an operation and maintenance personnel, or an application on a mobile device.

8. The intelligent interactive system for shipping big data based on the model context protocol according to claim 7, characterized in that The user uses the terminal to issue queries, instructions or receive alarm notifications to the shipping big data platform. Specifically, when the user asks a natural language question or selects a predefined analysis requirement, the shipping big data platform sends the request to the model context protocol module for processing. Finally, the analysis result of the AI model module is transmitted to the user interaction terminal via the shipping big data platform; the user interaction terminal displays it in an easy-to-understand form, including: natural language answers, visual charts or operation suggestions.

9. An intelligent interaction method for shipping big data based on a model context protocol, characterized in that, The following steps are involved: S1: Data collection and preprocessing: The ship monitoring system continuously collects ship status data and transmits the data to the shipping big data platform; S2: Trigger request and context encapsulation: When a user initiates a query request through the user interaction terminal or when the security management module detects an anomaly that requires AI analysis, the shipping big data platform extracts the relevant historical and real-time data it has collected and submits it to the model context protocol module. The model context protocol module selects the data most relevant to the request from the data storage and management module according to the request type, and constructs a context data packet in accordance with the model context protocol format; S3: AI model processing and reasoning: The model context protocol module sends the context data packet to the AI model module through a standard interface and calls the corresponding model algorithm for processing. The AI model module parses the received context packet, extracts the query and data contained therein, and then uses its trained knowledge and reasoning ability to generate answers or decision suggestions; S4: Result packaging and return: The AI model module packages the inference results into a response data packet according to the model context protocol requirements and returns it to the model context protocol module. After receiving the response data packet, the model context protocol module first verifies the integrity and source legitimacy of the data packet, then parses and extracts the result data, and stores it in the data storage and management module; S5: Result distribution and interactive presentation: The data storage and management module distributes the stored data results to the corresponding terminals or modules; the system can also actively integrate AI analysis conclusions into the operation process, improving the intelligent level of shipping management; S6: Component Update and Hot Plugging: When new functional modules need to be introduced, AI models need to be replaced, or new data sources need to be integrated, first, the new component adapts the interface according to the requirements of the model context protocol. When the new component accesses the shipping big data platform, it is registered through the model context protocol module, enabling the shipping big data platform to recognize its functions and data interaction formats. After registration, the new component can work in real-time collaboration with the existing system.

10. The intelligent interaction method for shipping big data based on the model context protocol according to claim 9, characterized in that The context data packet includes: raw ship status data, a description of the problem, an identifier of the data source, and access permission information attached by the security management module.

Images