Automatically Discovering Mechanical Functions from Physical Behaviors Via Clustering

Kevin Chiu,Mark Fuge

DOI: https://doi.org/10.1115/1.4065017

IF: 2.3

2024-03-07

Journal of Computing and Information Science in Engineering

Abstract:Abstract Computational design methods provide opportunities to discover novel and diverse designs that traditional optimization approaches cannot find or that use physical phenomena in ways that engineers have overlooked. However, existing methods require supervised objectives to search or optimize for explicit behaviors or functions. In contrast, this paper unpacks what it means to discover interesting behaviors or functions we do not know about a priori using data from experiments or simulation in a fully unsupervised way. Doing so enables computers to invent or re-invent new or existing mechanical functions given only measurements of physical fields without directly specifying a set of objectives to optimize. First, we study clustering algorithms that can detect novel device families from simulation data. Specifically, we contribute a modification to a well-established clustering algorithm via the use of the silhouette score to reduce excessively granular clusters. Second, we study multiple ways by which we preprocess simulation data to increase its discriminatory power in clustering device behavior. This leads to insight regarding the important role that a design's representation has in compactly encoding its behavior. We test our contributions via the task of discovering fluidic devices and show that our proposed clustering algorithm outperforms other density-based algorithms, but that KMeans clustering outperforms density-based algorithms. However, the device types may have an even stronger impact on the clustering. This opens up new avenues of research wherein computers can automatically derive new device functions, behaviors, and structures without the need for human labels or guidance.

engineering, manufacturing,computer science, interdisciplinary applications