Abstract:Multi-disciplinary design optimisation (MDO) is one of critical methodologies to the implementation of enterprise systems (ES). MDO requiring the analysis of fluid dynamics raises a special challenge due to its extremely intensive computation. The rapid development of computational fluid dynamic (CFD) technique has caused a rise of its applications in various fields. Especially for the exterior designs of vehicles, CFD has become one of the three main design tools comparable to analytical approaches and wind tunnel experiments. CFD-based design optimisation is an effective way to achieve the desired performance under the given constraints. However, due to the complexity of CFD, integrating with CFD analysis in an intelligent optimisation algorithm is not straightforward. It is a challenge to solve a CFD-based design problem, which is usually with high dimensions, and multiple objectives and constraints. It is desirable to have an integrated architecture for CFD-based design optimisation. However, our review on existing works has found that very few researchers have studied on the assistive tools to facilitate CFD-based design optimisation. In the paper, a multi-layer architecture and a general procedure are proposed to integrate different CFD toolsets with intelligent optimisation algorithms, parallel computing technique and other techniques for efficient computation. In the proposed architecture, the integration is performed either at the code level or data level to fully utilise the capabilities of different assistive tools. Two intelligent algorithms are developed and embedded with parallel computing. These algorithms, together with the supportive architecture, lay a solid foundation for various applications of CFD-based design optimisation. To illustrate the effectiveness of the proposed architecture and algorithms, the case studies on aerodynamic shape design of a hypersonic cruising vehicle are provided, and the result has shown that the proposed architecture and developed algorithms have performed successfully and efficiently in dealing with the design optimisation with over 200 design variables.

Optimisation Multidisciplinaire Avec Modèles Réduits Et Calcul Parallèle

Multidisciplinary Design Optimization on Conceptual Design of Aero-engine

A Bi-Level Meta-Modeling Approach for Structural Optimization Using Modified POD Bases and Diffuse Approximation

An Approach Based on Enhanced Collaborative Optimization and Kriging Approximation in Multidisciplinary Design Optimization

Surrogate-based optimization method applied to multidisciplinary design optimziation architectures

Nouveaux Algorithmes Numériques Pour L’utilisation Efficace Des Architectures Multi-Cœurs Et Hétérogènes

Méthodologie De Prototypage Rapide Pour Systèmes Embarqués Parallèles : Modélisation Des Systèmes Et Amélioration Des Heuristiques D'ordonnancement De Tâches

Multi-Disciplinary Optimisation For Front Auto Body Based On Multiple Optimisation Methods

A generalised collaborative optimisation method and its combination with kriging metamodels for engineering design

A generalized methodology for multidisciplinary design optimization using surrogate modelling and multifidelity analysis

Metamodeling techniques for CPU-intensive simulation-based design optimization: a survey

An efficient application of Bayesian optimization to an industrial MDO framework for aircraft design

Multidisciplinary and Multi-Objective Design Optimization Method Based on Physical Programming

A Supportive Architecture For Cfd-Based Design Optimisation

Multidisciplinary Optimization of Auto-Body Lightweight Design Using Hybrid Metamodeling Technique and Particle Swarm Optimizer

High-Fidelity Surrogate Based Multi-Objective Optimization Algorithm

Multidisciplinary Design Optimization Based On Physical Programming

Multi-fidelity multidisciplinary meta-model based optimization of a slender body with fins

A metamodel-based multidisciplinary design optimization process for automotive structures

An Improved Multidisciplinary Feasible Method Using DACE Approximation Approach

A multi-objective optimization strategy based on combined meta-models: Application to a wind turbine