English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Physics simulations are a computational bottleneck in computer-aided design(CAD) optimization processes. Hence, in order to make accurate (computationallyexpensive) simulations feasible for use in design optimization, one requireseither an optimization framework that is highly sample-efficient or fastdata-driven proxies (surrogate models) for long running simulations. In thiswork, we leverage recent advances in optimization and artificial intelligence(AI) to address both of these potential solutions, in the context of designingan optimal unmanned underwater vehicle (UUV). We first investigate and comparethe sample efficiency and convergence behavior of different optimizationtechniques with a standard computational fluid dynamics (CFD) solver in theoptimization loop. We then develop a deep neural network (DNN) based surrogatemodel to approximate drag forces that would otherwise be computed via directnumerical simulation with the CFD solver. The surrogate model is in turn usedin the optimization loop of the hull design. Our study finds that the BayesianOptimization Lower Condition Bound (BO LCB) algorithm is the mostsample-efficient optimization framework and has the best convergence behaviorof those considered. Subsequently, we show that our DNN-based surrogate modelpredicts drag force on test data in tight agreement with CFD simulations, witha mean absolute percentage error (MAPE) of 1.85%. Combining these results, wedemonstrate a two-orders-of-magnitude speedup (with comparable accuracy) forthe design optimization process when the surrogate model is used. To ourknowledge, this is the first study applying Bayesian optimization and DNN-basedsurrogate modeling to the problem of UUV design optimization, and we share ourdevelopments as open-source software.

Sample-Efficient and Surrogate-Based Design Optimization of Underwater Vehicle Hulls