笔记：An integrated approach for enhanced early-phase space system design and optimization

An integrated approach for enhanced early-phase space system design and optimization

Abstraction

The integration of Model-Based Systems Engineering (MBSE) and Multidisciplinary Design Analysis and Optimization (MDAO) presents a powerful opportunity to enhance early-stage system design, particularly for complex space systems. However, the lack of efficient integration between these methods results in limitations such as unclear boundary between domain models, reduced automation, and challenges in maintaining traceability of optimization results. Overcoming these barriers is essential for conducting high-quality trade studies in systems engineering. In this work, we propose a novel framework that integrates MDAO with MBSE to streamline system modeling, optimization, and verification. This approach enables the seamless exchange of knowledge between design and optimization models, while performing optimizations and managing results directly within the MBSE environment. By using MBSE as a central knowledge repository, the framework minimizes errors and improves the traceability of optimization processes. Case studies demonstrate that this framework enhances both efficiency and accuracy during the early design phases of space mission development. Our findings indicate that integrating MDAO with MBSE allows for comprehensive system evaluation and more informed decision-making, ultimately improving the quality and efficiency of the design process. This integrated framework offers a flexible, scalable solution for multidisciplinary optimization, making it a valuable tool for the design of future complex systems.

Introduction

Traditional systems engineering stores team knowledge in dispersed documents, while detailed design, manufacturing, and testing increasingly rely on computer-aided technology, which can lead to disconnects and introduce errors. By using this model as the single source of truth (SSOT), MBSE helps manage system complexity, offers an efficient interdisciplinary platform for collboration across various fields, and facilitates the sharing of design knowledge.

Trade study: A means of evaluating system designs by devising alternative means to meet functional requirements, evaluating these alternatives in terms of the measures of effectiveness and system cost, ranking the alternatives according to appropriate selection criteria, dropping less promising alternatives, and proceeding to the next level of resolution, if needed. (NASA Systems Engineering Handbook)

MDAO typically requires defining the system’s operation before constructing the MDAO framework and performing optimization with analytical models from each discipline. As a result, MDAO demands a substantial initial effort, consuming 60–80 % of project time in aerospace contexts (Habermehl et al, 2022).

Translating MBSE models into a form usable by MDAO is not straightforward, and integrating MDAO results back into MBSE is equally challenging.

The primary contribution of this approach is to clearly delineate the boundaries between models within the space mission domain and the optimization domain in MBSE. A consistent model not only enhances the level of automation in performing optimizations within MBSE but also reduces the complexity of associated manual operations, making the optimization process more responsive to system changes.

While integrating MBSE with MDAO has the potential to leverage the strengths of both approaches, this synergy has yet to be fully realized (Habermehl et al, 2022).

Although optimization algorithms and discipline-specific models may originate from various modeling languages and tools, surrogate models and tool interoperability facilitate the linking and execution of MDAO models.

Challenges also arise in integrating MBSE with MDAO, particularly in the transfer of information between design models, optimization algorithms, and optimization models. Indeed, automatically generating MDAO models from MBSE models remains a significant challenge, particularly for high-fidelity, heterogeneous discipline models, which typically require careful authoring by discipline experts using specialized tools.

Chaudemar et al proposed two integration processes—strongly coupled and weakly coupled—based on the degree of coupling between MBSE and MDAO (Chaudemar and de Saqui-Sannes, 2021). Several studies support the return of optimization results to MBSE for further analysis and extensive validation (Gao et al, 2021; Chen et al, 2022; C´ ardenas et al, 2022). As MDAO represents only one aspect of the model, returning these solutions to MBSE allows for comprehensive verification and initial screening of viable solutions.

In summary, although research has demonstrated the potential of integrating MBSE with MDAO, significant challenges remain in fully achieving this integration.

MBSE和MDAO的边界模糊
如何利用MBSE推动MDAO有待解决

Research objective and method

Integrating MBSE with MDAO from a methodological perspective and to establish MDAO as an integral activity with SE.
To clarify the modeling boundaries between MBSE and MDAO.

图1展示了一个面向MDAO设计的SysML扩展元模型，是根据CSRM改编的。
图2展示了一个叫XDSM的visualization tool for illustrating processes and data interfaces within an optimizarion architecture.

Case study

一种应急遥感卫星，1t重，三个任务指标：higher payload performance, longer payload operational hours, and the ability to undertake additional emergency missions throughout the satellite’s lifetime.（1）有效载荷应达到不超过0.3 m的分辨率，（2）在每日数据流量余额的条件下，有效载荷应每天运行不少于60分钟，并且（3）卫星应保留至少80 kg的自由推进剂质量。

之后开始介绍空间轨道计算，推进剂消耗，通信系统，绘制出一张参数图展示多学科分析的效果。然后开始进行MDAO过程，最后用Excel保存信息。

卫星系统参数图