| 摘 要： |
The transition to more highly integrated naval vessels with significant, high power electric power systems for advanced sensors, weapons, and propulsion boost introduces significant complexity in managing power distribution, thermal control, and cross-domain system interactions - necessitating a cohesive digital engineering approach. This paper presents a low-cost, rapidly deployable Digital Ecosystem architecture tailored to support model-based systems engineering (MBSE) and digital twin development in electric ship programs. Aligned with DoDI 5000.97, the study implements a "three-legged" integration strategy using CSV files, Python scripts, and MATLAB routines to connect a SysML system model in Cameo with external analytical tools. This architecture enables seamless validation of system behavior and performance against requirements, including integration with CAD-derived geometries and physics-based simulations such as antenna pattern and structural analysis. Generative AI (ChatGPT) was used to accelerate model development by transforming legacy documents into structured, import-ready datasets. The result is a scalable, interoperable framework for digital thread and digital twin workflows that supports the agility, traceability, and cross-discipline coordination demanded by modern electric ship design and sustainment.
向集成度更高的舰艇过渡，配备先进的传感器、武器和推进系统所需的高功率电力系统，显著增加了对电力分配、热控以及跨域系统交互的管理复杂性，因此需要采用统一的数字工程方法。本文提出了一种低成本、快速部署的数字生态系统架构，专为电动船舶项目中的基于模型的系统工程（MBSE）和数字孪生开发而设计。该架构符合美国国防部第5000.97号指令要求，采用“三足”集成策略，通过CSV文件、Python脚本和MATLAB程序，将Cameo中的SysML系统模型与外部分析工具连接起来。该架构可实现系统行为与性能的无缝验证，并满足各项需求，包括与CAD生成的几何数据及天线图样、结构分析等物理仿真技术的集成。同时，利用生成式人工智能（ChatGPT）加速了模型开发过程，将传统文档转换为结构化、可直接导入的数据集。其结果是一个可扩展且具备互操作性的数字线程与数字孪生工作流框架，能够支持现代电动船舶设计与维护所需的敏捷性、可追溯性以及跨学科协调。 |
京公网安备 11010202009788