Active Learning over DNN: Automated Engineering Design Optimization for Fluid Dynamics Based on Self-Simulated Dataset

TL;DR

This paper presents an active learning approach combined with deep neural networks to optimize fluid dynamic shapes efficiently, reducing data requirements and enabling user-guided design without extensive domain expertise.

Contribution

It introduces a novel active learning framework for fluid shape optimization using DNNs, significantly decreasing data needs and integrating user input for practical engineering design.

Findings

Reduced data samples from ~8000 to 625 using active learning.

Successfully optimized shapes with minimal human domain knowledge.

Achieved low-drag shapes through an interactive user interface.

Abstract

Optimizing fluid-dynamic performance is an important engineering task. Traditionally, experts design shapes based on empirical estimations and verify them through expensive experiments. This costly process, both in terms of time and space, may only explore a limited number of shapes and lead to sub-optimal designs. In this research, a test-proven deep learning architecture is applied to predict the performance under various restrictions and search for better shapes by optimizing the learned prediction function. The major challenge is the vast amount of data points Deep Neural Network (DNN) demands, which is improvident to simulate. To remedy this drawback, a Frequentist active learning is used to explore regions of the output space that DNN predicts promising. This operation reduces the number of data samples demanded from ~8000 to 625. The final stage, a user interface, made the model capable of optimizing with given user input of minimum area and viscosity. Flood fill is used to define a boundary area function so that the optimal shape does not bypass the minimum area. Stochastic Gradient Langevin Dynamics (SGLD) is employed to make sure the ultimate shape is optimized while circumventing the required area. Jointly, shapes with extremely low drags are found explored by a practical user interface with no human domain knowledge and modest computation overhead.