{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,7,30]],"date-time":"2025-07-30T12:56:54Z","timestamp":1753880214358,"version":"3.41.2"},"reference-count":47,"publisher":"ASME International","issue":"8","license":[{"start":{"date-parts":[[2024,3,7]],"date-time":"2024-03-07T00:00:00Z","timestamp":1709769600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.asme.org\/publications-submissions\/publishing-information\/legal-policies"}],"funder":[{"DOI":"10.13039\/100000185","name":"Defense Advanced Research Projects Agency","doi-asserted-by":"publisher","award":["HR0011-18-9-0009"],"award-info":[{"award-number":["HR0011-18-9-0009"]}],"id":[{"id":"10.13039\/100000185","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["1943699"],"award-info":[{"award-number":["1943699"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["asmedigitalcollection.asme.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2024,8,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>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\u2014e.g., optimizing aerodynamic lift. In contrast, this article 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 (e.g., fluid velocity fields) without directly specifying a set of objectives to optimize. This article explores this approach via two related parts. First, we study clustering algorithms that can detect novel device families from simulation data. Specifically, we contribute a modification to the hierarchical density-based spatial clustering of applications with noise 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 the context of clustering device behavior. This leads to an insight regarding the important role that a design\u2019s representation has in compactly encoding its behavior. We test our contributions via the task of discovering simple fluidic devices and show that our proposed clustering algorithm outperforms other density-based algorithms, but that K-means clustering outperforms density-based algorithms, as measured by adjusted Rand score. 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.<\/jats:p>","DOI":"10.1115\/1.4065017","type":"journal-article","created":{"date-parts":[[2024,3,7]],"date-time":"2024-03-07T15:24:07Z","timestamp":1709825047000},"update-policy":"https:\/\/doi.org\/10.1115\/crossmarkpolicy-asme","source":"Crossref","is-referenced-by-count":0,"title":["Automatically Discovering Mechanical Functions From Physical Behaviors via Clustering"],"prefix":"10.1115","volume":"24","author":[{"given":"Kevin N.","family":"Chiu","sequence":"first","affiliation":[{"name":"University of Maryland, College Park Department of Mechanical Engineering, , MD 20742"}]},{"given":"Mark D.","family":"Fuge","sequence":"additional","affiliation":[{"name":"University of Maryland, College Park Department of Mechanical Engineering, , MD 20742"}]}],"member":"33","published-online":{"date-parts":[[2024,4,16]]},"reference":[{"issue":"4","key":"2025041520023468400_CIT0001","first-page":"26","article-title":"Design Prototypes: A Knowledge Representation Schema for Design","volume":"11","author":"Gero","year":"1990","journal-title":"AI Magazine"},{"issue":"25","key":"2025041520023468400_CIT0002","doi-asserted-by":"publisher","first-page":"5074","DOI":"10.1063\/1.1764592","article-title":"Flow Resistance for Microfluidic Logic Operations","volume":"84","author":"Vestad","year":"2004","journal-title":"Appl. Phys. Lett."},{"issue":"1","key":"2025041520023468400_CIT0003","doi-asserted-by":"publisher","first-page":"218","DOI":"10.1038\/nphys1513","article-title":"Static Control Logic for Microfluidic Devices Using Pressure-Gain Valves","volume":"6","author":"Weaver","year":"2010","journal-title":"Nat. Phys."},{"issue":"5813","key":"2025041520023468400_CIT0004","doi-asserted-by":"publisher","first-page":"832","DOI":"10.1126\/science.1136907","article-title":"Microfluidic Bubble Logic","volume":"315","author":"Prakash","year":"2007","journal-title":"Science"},{"issue":"16","key":"2025041520023468400_CIT0005","doi-asserted-by":"publisher","first-page":"19845","DOI":"10.1364\/OE.26.019845","article-title":"Realization of True All-Optical AND Logic Gate Based on Nonlinear Coupled Air-Hole Type Photonic Crystal Waveguides","volume":"26","author":"Jandieri","year":"2018","journal-title":"Opt. Express"},{"issue":"8090","key":"2025041520023468400_CIT0006","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/srep08090","article-title":"SynBioLGDB: A Resource for Experimentally Validated Logic Gates in Synthetic Biology","volume":"5","author":"Wang","year":"2015","journal-title":"Sci. Rep."},{"issue":"4","key":"2025041520023468400_CIT0007","doi-asserted-by":"publisher","first-page":"040501","DOI":"10.1103\/PhysRevLett.102.040501","article-title":"Realization of the Quantum Toffoli Gate With Trapped Ions","volume":"102","author":"Monz","year":"2009","journal-title":"Phys. Rev. Lett."},{"issue":"5","key":"2025041520023468400_CIT0008","doi-asserted-by":"publisher","first-page":"051001","DOI":"10.1103\/PhysRevApplied.9.051001","article-title":"Deutsch, Toffoli, and CNOT Gates Via Rydberg Blockade of Neutral Atoms","volume":"9","author":"Shi","year":"2018","journal-title":"Phys. Rev. Appl."},{"issue":"12","key":"2025041520023468400_CIT0009","doi-asserted-by":"publisher","first-page":"125119","DOI":"10.1063\/1.4973429","article-title":"Prolonged Silicon Carbide Integrated Circuit Operation in Venus Surface Atmospheric Conditions","volume":"6","author":"Neudeck","year":"2016","journal-title":"AIP Adv."},{"issue":"1","key":"2025041520023468400_CIT0010","doi-asserted-by":"publisher","first-page":"53","DOI":"10.1016\/0377-0427(87)90125-7","article-title":"Silhouettes: A Graphical Aid to the Interpretation and Validation of Cluster Analysis","volume":"20","author":"Rousseeuw","year":"1987","journal-title":"J. Comput. Appl. Math."},{"issue":"4","key":"2025041520023468400_CIT0011","doi-asserted-by":"publisher","first-page":"745","DOI":"10.1090\/S0025-5718-1968-0242392-2","article-title":"Numerical Solution of the Navier-Stokes Equations","volume":"22","author":"Chorin","year":"1968","journal-title":"Math. Comput."},{"issue":"2","key":"2025041520023468400_CIT0012","doi-asserted-by":"publisher","first-page":"021003","DOI":"10.1115\/1.3593409","article-title":"Computer-Based Design Synthesis Research: An Overview","volume":"11","author":"Chakrabarti","year":"2011","journal-title":"ASME J. Comput. Inf. Sci. Eng."},{"issue":"1","key":"2025041520023468400_CIT0013","doi-asserted-by":"publisher","first-page":"23","DOI":"10.1017\/S0890060409000080","article-title":"Structure, Behavior, and Function of Complex Systems: The Structure, Behavior, and Function Modeling Language","volume":"23","author":"Goel","year":"2009","journal-title":"Artif. Intell. Eng. Design, Anal. Manufact."},{"issue":"4","key":"2025041520023468400_CIT0014","doi-asserted-by":"publisher","first-page":"476","DOI":"10.1017\/S0890060417000403","article-title":"Function Modeling Combined With Physics-Based Reasoning for Assessing Design Options and Supporting Innovative Ideation","volume":"31","author":"Mokhtarian","year":"2017","journal-title":"Artif. Intell. Eng. Design, Anal. Manufact."},{"article-title":"Efficient Estimation of Word Representations in Vector Space","year":"2013","author":"Mikolov","key":"2025041520023468400_CIT0015"},{"key":"2025041520023468400_CIT0016","first-page":"3111","volume-title":"Distributed Representations of Words and Phrases and Their Compositionality","author":"Mikolov","year":"2013"},{"issue":"11","key":"2025041520023468400_CIT0017","doi-asserted-by":"publisher","first-page":"111414","DOI":"10.1115\/1.4037478","article-title":"Concept Clustering in Design Teams: A Comparison of Human and Machine Clustering","volume":"139","author":"Zhang","year":"2017","journal-title":"ASME J. Mech. Des."},{"issue":"5","key":"2025041520023468400_CIT0018","doi-asserted-by":"publisher","first-page":"054501","DOI":"10.1115\/1.4052904","article-title":"Phrase Embedding and Clustering for Sub-Feature Extraction From Online Data","volume":"144","author":"Park","year":"2022","journal-title":"ASME J. Mech. Des."},{"key":"2025041520023468400_CIT0019","doi-asserted-by":"crossref","DOI":"10.1115\/DETC2016-59926","article-title":"Discovering Diverse, High Quality Design Ideas From a Large Corpus","author":"Ahmed","year":"2016"},{"key":"2025041520023468400_CIT0020","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9780511809071","volume-title":"Introduction to Information Retrieval","author":"Manning","year":"2008"},{"key":"2025041520023468400_CIT0021","doi-asserted-by":"crossref","DOI":"10.1115\/DETC2020-22714","article-title":"Learning to Abstract and Compose Mechanical Device Function and Behavior","author":"Wang","year":"2020"},{"key":"2025041520023468400_CIT0022","first-page":"1","article-title":"Machine Learning Computational Fluid Dynamics","author":"Usman","year":"2021"},{"issue":"12","key":"2025041520023468400_CIT0023","doi-asserted-by":"publisher","first-page":"6289","DOI":"10.1029\/2018GL078510","article-title":"Prognostic Validation of a Neural Network Unified Physics Parameterization","volume":"45","author":"Brenowitz","year":"2018","journal-title":"Geophys. Res. Lett."},{"issue":"C","key":"2025041520023468400_CIT0024","doi-asserted-by":"publisher","first-page":"103225","DOI":"10.1016\/j.cad.2022.103225","article-title":"Generative Design by Reinforcement Learning: Enhancing the Diversity of Topology Optimization Designs","volume":"146","author":"Jang","year":"2022","journal-title":"Comput. Aided Design"},{"key":"2025041520023468400_CIT0025","doi-asserted-by":"crossref","DOI":"10.1115\/DETC2021-69544","article-title":"Design Concept Generation With Variational Deep Embedding Over Comprehensive Optimization","author":"Fujita","year":"2021"},{"key":"2025041520023468400_CIT0026","doi-asserted-by":"crossref","DOI":"10.1115\/DETC2021-68103","article-title":"CreativeGAN: Editing Generative Adversarial Networks for Creative Design Synthesis","author":"Heyrani","year":"2021"},{"key":"2025041520023468400_CIT0027","doi-asserted-by":"crossref","DOI":"10.1115\/DETC2021-69328","article-title":"Automatically Discovering Mechanical Functions From Physical Behaviors via Clustering","author":"Chiu","year":"2021"},{"issue":"2","key":"2025041520023468400_CIT0028","doi-asserted-by":"publisher","first-page":"129","DOI":"10.1109\/TIT.1982.1056489","article-title":"Least Squares Quantization in PCM","volume":"28","author":"Lloyd","year":"1982","journal-title":"IEEE Trans. Inform. Theory"},{"key":"2025041520023468400_CIT0029","first-page":"226","article-title":"A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases With Noise","author":"Ester","year":"1996"},{"key":"2025041520023468400_CIT0030","doi-asserted-by":"crossref","first-page":"160","DOI":"10.1007\/978-3-642-37456-2_14","article-title":"Density-Based Clustering Based on Hierarchical Density Estimates","volume-title":"Advances in Knowledge Discovery and Data Mining","author":"Campello","year":"2013"},{"key":"2025041520023468400_CIT0031","first-page":"1133","article-title":"Antenna Design Using Genetic Algorithms","author":"Linden","year":"2002"},{"issue":"4","key":"2025041520023468400_CIT0032","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/2897824.2925870","article-title":"Learning How Objects Function Via Co-Analysis of Interactions","volume":"35","author":"Hu","year":"2016","journal-title":"ACM Trans. Graph."},{"key":"2025041520023468400_CIT0033","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR52688.2022.00147","article-title":"Fixing Malfunctional Objects With Learned Physical Simulation and Functional Prediction","author":"Hong","year":"2022"},{"issue":"2","key":"2025041520023468400_CIT0034","doi-asserted-by":"publisher","first-page":"367","DOI":"10.1111\/cgf.13644","article-title":"Learning a Generative Model for Multi-Step Human-Object Interactions From Videos","volume":"38","author":"Wang","year":"2019","journal-title":"Computer Graphics Forum."},{"issue":"6","key":"2025041520023468400_CIT0035","doi-asserted-by":"publisher","first-page":"1389","DOI":"10.1002\/j.1538-7305.1957.tb01515.x","article-title":"Shortest Connection Networks and Some Generalizations","volume":"36","author":"Prim","year":"1957","journal-title":"Bell Syst. Tech. J."},{"volume-title":"Dynamic Programming","year":"1957","author":"Bellman","key":"2025041520023468400_CIT0036"},{"issue":"7","key":"2025041520023468400_CIT0037","first-page":"793","article-title":"Sur La Sph\u00e9re Vide","volume":"6","author":"Delaunay","year":"1934","journal-title":"Bull. de l\u2019Acad\u00e9mie des Sci. de l\u2019URSS, Classe des Sci. Math. Nat."},{"article-title":"pygmsh: A Python Frontend for Gmsh (v7.1.17)","year":"2022","author":"Schl\u00f6mer","key":"2025041520023468400_CIT0038"},{"issue":"11","key":"2025041520023468400_CIT0039","doi-asserted-by":"publisher","first-page":"1309","DOI":"10.1002\/nme.2579","article-title":"Gmsh: a Three-Dimensional Finite Element Mesh Generator With Built-in Pre- and Post-processing Facilities","volume":"79","author":"Geuzaine","year":"2009","journal-title":"Int. J. Numer. Methods Eng."},{"article-title":"meshio: Tools for mesh files (v5.3.4)","year":"2022","author":"Schl\u00f6mer","key":"2025041520023468400_CIT0040"},{"issue":"100","key":"2025041520023468400_CIT0041","first-page":"9","article-title":"The Fenics Project Version 1.5","volume":"3","author":"Aln\u00e6s","year":"2015","journal-title":"Arch. Numer. Soft."},{"issue":"1","key":"2025041520023468400_CIT0042","doi-asserted-by":"publisher","first-page":"193","DOI":"10.1007\/BF01908075","article-title":"Comparing Partitions","volume":"2","author":"Hubert","year":"1985","journal-title":"J. Class."},{"issue":"1","key":"2025041520023468400_CIT0043","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1111\/j.2517-6161.1977.tb01600.x","article-title":"Maximum Likelihood From Incomplete Data via the Em Algorithm","volume":"39","author":"Dempster","year":"1977","journal-title":"J. R. Stat. Soc. Seri. B (Methodol.)"},{"issue":"85","key":"2025041520023468400_CIT0044","doi-asserted-by":"publisher","first-page":"2825","DOI":"10.3389\/fninf.2014.00014","article-title":"Scikit-Learn: Machine Learning in Python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J. Mach. Learn. Res."},{"issue":"11","key":"2025041520023468400_CIT0045","doi-asserted-by":"publisher","first-page":"2017","DOI":"10.21105\/joss.00205","article-title":"HDBSCAN: Hierarchical Density Based Clustering","volume":"2","author":"McInnes","year":"2017","journal-title":"J. Open Source Soft."},{"article-title":"Bayesian Optimization: Open Source Constrained Global Optimization Tool for Python","year":"2014","author":"Nogueira","key":"2025041520023468400_CIT0046"},{"issue":"5","key":"2025041520023468400_CIT0047","doi-asserted-by":"publisher","first-page":"1299","DOI":"10.1162\/089976698300017467","article-title":"Nonlinear Component Analysis as a Kernel Eigenvalue Problem","volume":"10","author":"Schlkopf","year":"1998","journal-title":"Neural Comput."}],"container-title":["Journal of Computing and Information Science in Engineering"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/asmedigitalcollection.asme.org\/computingengineering\/article-pdf\/24\/8\/081002\/7328615\/jcise_24_8_081002.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/asmedigitalcollection.asme.org\/computingengineering\/article-pdf\/24\/8\/081002\/7328615\/jcise_24_8_081002.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,4,16]],"date-time":"2025-04-16T00:02:39Z","timestamp":1744761759000},"score":1,"resource":{"primary":{"URL":"https:\/\/asmedigitalcollection.asme.org\/computingengineering\/article\/24\/8\/081002\/1198485\/Automatically-Discovering-Mechanical-Functions"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,4,16]]},"references-count":47,"journal-issue":{"issue":"8","published-print":{"date-parts":[[2024,8,1]]}},"URL":"https:\/\/doi.org\/10.1115\/1.4065017","relation":{},"ISSN":["1530-9827","1944-7078"],"issn-type":[{"type":"print","value":"1530-9827"},{"type":"electronic","value":"1944-7078"}],"subject":[],"published":{"date-parts":[[2024,4,16]]},"article-number":"081002"}}