Surrogate-Assisted Evolutionary Generative Design Of Breakwaters Using Deep Convolutional Networks

TL;DR

This paper introduces a surrogate-assisted evolutionary method using deep convolutional networks to efficiently design coastal breakwaters, significantly reducing computational costs while improving protective effectiveness.

Contribution

It presents a novel multi-objective optimization approach employing deep CNNs as surrogates and confidence estimators for breakwater design, enhancing efficiency and accuracy.

Findings

More effective solutions with lower costs were obtained.

The surrogate model reduced computational time significantly.

The approach outperformed non-surrogate methods in effectiveness.

Abstract

In the paper, a multi-objective evolutionary surrogate-assisted approach for the fast and effective generative design of coastal breakwaters is proposed. To approximate the computationally expensive objective functions, the deep convolutional neural network is used as a surrogate model. This model allows optimizing a configuration of breakwaters with a different number of structures and segments. In addition to the surrogate, an assistant model was developed to estimate the confidence of predictions. The proposed approach was tested on the synthetic water area, the SWAN model was used to calculate the wave heights. The experimental results confirm that the proposed approach allows obtaining more effective (less expensive with better protective properties) solutions than non-surrogate approaches for the same time.