Constrained Multi-Objective Bayesian Optimization with Application to Aircraft Design

Grapin, Robin; Diouane, Youssef; Morlier, Joseph; Bartoli, Nathalie; Lefèbvre, Thierry; Saves, Paul; Bussemaker, Jasper

doi:10.2514/6.2022-4053

Cited by 10 publications

(12 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [91], R. Concerning the extensions of GP to mixed variables, in [206], we developed a new Gaussian process model for categorical and discrete inputs to extend the Bayesian optimization methods aforementioned to mixed integer variables. In particular, we proposed a new exponential correlation kernel that unifies both distance-based approaches and matrix-based approaches through a unified formulation.…”

Section: Constrained Bayesian Optimization In High Dimensionmentioning

confidence: 99%

“…The new GP models and the new Bayesian optimization capabilities have been combined and applied to various aircraft design problems. First, in [91,204], we optimized a reference configuration based on a A320 named CeRAS (Central Reference Aircraft System) [191]. Then, in [209], we optimized an hybrid electric aircraft for long range missions [212] and we developed an adaptive criterion to select automatically the number of effective dimensions when building the models during the optimization process.…”

Section: Constrained Bayesian Optimization In High Dimensionmentioning

confidence: 99%

See 1 more Smart Citation

High-dimensional multidisciplinary design optimization for aircraft eco-design

SAVES

2024

Preprint

View full text Add to dashboard Cite

Nowadays, there is a significant and growing interest in improving the efficiency of vehicle design processes through the development of tools and techniques in the field of MDO. Specifically, in aerostructure design, aerodynamic and structural variables influence each other and have a joint effect on quantities of interest like weight or fuel consumption and, as such, MDO arises as a powerful tool for automatically making interdisciplinary trade-offs. In the aircraft design context, the process generally involves mixed continuous and categorical design variables. For instance, the size of an aircraft’s structural parts can be described using continuous variables, while discrete variables may include either integer variables, like the number of panels, or categorical variables, like cross-sections or material choices. The objective of this Philosophiae Doctor (Ph.D) thesis is to propose an efficient approach for optimizing a multidisciplinary black-box model when the optimization problem is constrained and involves a large number of mixed integer design variables (typically 100 variables). The targeted optimization approach, called EGO, is based on a sequential enrichment of an adaptive surrogate model and, in this context, GP surrogate models are one of the most widely used in engineering problems to approximate time-consuming high fidelity models. EGO is a heuristic BO method that performs well in terms of solution quality. However, like any other global optimization method, EGO suffers from the curse of dimensionality, meaning that its performance is satisfactory on lower dimensional problems, but deteriorates as the dimensionality of the optimization search space increases. For realistic aircraft design problems, the typical size of the design variables can even exceed 100 and, thus, trying to solve directly the problems using EGO is ruled out. The latter is especially true when the problems involve both continuous and categorical variables increasing even more the size of the search space. In this Ph.D thesis, effective parameterization tools are investigated, including techniques like partial least squares regression, to significantly reduce the number of design variables. Additionally, Bayesian optimization is adapted to handle discrete variables and high-dimensional spaces in order to reduce the number of evaluations when optimizing innovative aircraft concepts such as the “DRAGON” hybrid airplane to reduce their climate impact.

show abstract

Section: Constrained Bayesian Optimization In High Dimensionmentioning

confidence: 99%

Section: Constrained Bayesian Optimization In High Dimensionmentioning

confidence: 99%

High-dimensional multidisciplinary design optimization for aircraft eco-design

SAVES

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…13, showing how architecture elements and associated design parameters are mapped to disciplinary analysis tools. The optimization problem has been executed using SEGOMOE (Bartoli et al 2016), a surrogatebased optimization algorithm which has recently been extended to also support multi-objective (Grapin et al 2022) and mixed-integer (Saves et al 2022) problems. Access to the optimization algorithm is provided by WhatsOpt (Lafage, Defoort & Lefebvre 2019) through an ask-tell interface.…”

Section: Showcase: Designing a Family Of Business Jetsmentioning

confidence: 99%

The AGILE 4.0 Project: MBSE to Support Cyber‐Physical Collaborative Aircraft Development

Bussemaker,

Boggero,

Nagel

2023

INCOSE International Symp

View full text Add to dashboard Cite

This paper presents the main output of the EU‐funded H2020 AGILE 4.0 project: the AGILE 4.0 MBSE‐MDAO Development Framework. The framework aims at connecting upstream Model‐Based Systems Engineering (MBSE) activities to downstream Multidisciplinary Design Analysis and Optimization (MDAO) activities, applied to the development of complex aeronautical systems. Upstream activities include the definition of stakeholders, needs, semantic requirements, and modeling the system architecture design space. Downstream activities include defining, implementing, and executing cross‐organizational and collaborative MDAO workflows to support system synthesis. The development framework is implemented in an Operational Collaborative Environment (OCE), a web‐based platform for project data and modeling tool access. A business jet family design application case demonstrates the use of the development framework and OCE.

show abstract

“…SEGOMOE fully described in [8] has been successful applied for different applications: aerodynamic shape optimization [10], nacelle optimization [9], overall aircraft configurations [11], as some industrial test case in collaboration with Bombardier [10]. Some recent developments have been made to consider highly non-linear constraints [11] or mixed integer variables [12]. The newest capability concerns multi-objective described in [12]From few initial points, an adaptive process is used to add some Pareto-optimal points chosen from an acquisition function that assigns a maximum value to the points having a high probability of improving our knowledge of the objectives and respecting the constraints in the probably optimal areas.…”

Section: Value-costmentioning

confidence: 99%

“…Some recent developments have been made to consider highly non-linear constraints [11] or mixed integer variables [12]. The newest capability concerns multi-objective described in [12]From few initial points, an adaptive process is used to add some Pareto-optimal points chosen from an acquisition function that assigns a maximum value to the points having a high probability of improving our knowledge of the objectives and respecting the constraints in the probably optimal areas. Then, after some iterations, final surrogate models are trained on the enriched database and an evolutionary algorithm such as NSGA-II [13] is applied to obtain the Pareto front.…”

Section: Value-costmentioning

confidence: 99%

Value-driven Model-Based Optimization coupling Design-Manufacturing-Supply Chain in the Early Stages of Aircraft Development: Strategy and Preliminary Results

Donelli

Ciampa

Lefèbvre

et al. 2022

AIAA AVIATION 2022 Forum

Self Cite

View full text Add to dashboard Cite

A value-driven model-based approach concurrently coupling design, manufacturing and supply chain in the early development stage of aircraft design has been developed within the European project AGILE4.0. The benefits of using this methodology have been highlighted by the aeronautical application case focused on the design, manufacturing and supply chain of an horizontal tail plane. Finding a Pareto-front simultaneously optimizing the design, manufacturing and supply chain domains is the next challenge to face. The research activity proposed in this paper represents the first step of this ambitious goal. The objective is to identify the optimization strategy to use for the global optimization campaign by exploring, first, simple and representative Multidisciplinary Design and Optimization (MDO) problems related to the supply chain domain. In the first MDO problem, a 4-objective optimization is executed and then the optimized attributes are aggregated in a single measure named value. In the second MDO problem instead, attributes are first aggregated in a value and then a biobjective value-cost optimization is executed. Thus, two optimization strategies are investigated, but both lead to the value-cost Pareto-front investigation. The application case addressed in this research activity provides interesting insights for the value-driven

show abstract

Constrained Multi-Objective Bayesian Optimization with Application to Aircraft Design

Cited by 10 publications

References 30 publications

High-dimensional multidisciplinary design optimization for aircraft eco-design

High-dimensional multidisciplinary design optimization for aircraft eco-design

The AGILE 4.0 Project: MBSE to Support Cyber‐Physical Collaborative Aircraft Development

Value-driven Model-Based Optimization coupling Design-Manufacturing-Supply Chain in the Early Stages of Aircraft Development: Strategy and Preliminary Results

Contact Info

Product

Resources

About