Transfer Learning in Bayesian Optimization for Aircraft Design

TL;DR

This paper introduces a transfer learning approach within Bayesian optimization to improve aircraft design optimization, addressing cold start issues and heterogeneity in design variables and constraints.

Contribution

It presents a novel ensemble surrogate model with transfer learning, combined with dimension reduction and meta data surrogate selection for aircraft design optimization.

Findings

Significant improvement in early convergence compared to standard Bayesian optimization.

Enhanced prediction accuracy for objective and constraint surrogate models.

Effective handling of heterogeneous design variables and constraints.

Abstract

The use of transfer learning within Bayesian optimization addresses the disadvantages of the so-called \textit{cold start} problem by using source data to aid in the optimization of a target problem. We present a method that leverages an ensemble of surrogate models using transfer learning and integrates it in a constrained Bayesian optimization framework. We identify challenges particular to aircraft design optimization related to heterogeneous design variables and constraints. We propose the use of a partial-least-squares dimension reduction algorithm to address design space heterogeneity, and a \textit{meta} data surrogate selection method to address constraint heterogeneity. Numerical benchmark problems and an aircraft conceptual design optimization problem are used to demonstrate the proposed methods. Results show significant improvement in convergence in early optimization iterations compared to standard Bayesian optimization, with improved prediction accuracy for both objective and constraint surrogate models.