A Bayesian experimental autonomous researcher for mechanical design.
TL;DR: A Bayesian experimental autonomous researcher (BEAR) that combines Bayesian optimization and high-throughput automated experimentation that explores the toughness of a parametric family of structures and observes an almost 60-fold reduction in the number of experiments needed to identify high-performing structures relative to a grid-based search.
Abstract: While additive manufacturing (AM) has facilitated the production of complex structures, it has also highlighted the immense challenge inherent in identifying the optimum AM structure for a given application. Numerical methods are important tools for optimization, but experiment remains the gold standard for studying nonlinear, but critical, mechanical properties such as toughness. To address the vastness of AM design space and the need for experiment, we develop a Bayesian experimental autonomous researcher (BEAR) that combines Bayesian optimization and high-throughput automated experimentation. In addition to rapidly performing experiments, the BEAR leverages iterative experimentation by selecting experiments based on all available results. Using the BEAR, we explore the toughness of a parametric family of structures and observe an almost 60-fold reduction in the number of experiments needed to identify high-performing structures relative to a grid-based search. These results show the value of machine learning in experimental fields where data are sparse.
Citations
More filters
TL;DR: The applications in mechanical property prediction, materials design and computational methods using ML-based approaches are summarized, followed by perspectives on opportunities and open challenges in this emerging and exciting field.
Abstract: Artificial intelligence, especially machine learning (ML) and deep learning (DL) algorithms, is becoming an important tool in the fields of materials and mechanical engineering, attributed to its power to predict materials properties, design de novo materials and discover new mechanisms beyond intuitions. As the structural complexity of novel materials soars, the material design problem to optimize mechanical behaviors can involve massive design spaces that are intractable for conventional methods. Addressing this challenge, ML models trained from large material datasets that relate structure, properties and function at multiple hierarchical levels have offered new avenues for fast exploration of the design spaces. The performance of a ML-based materials design approach relies on the collection or generation of a large dataset that is properly preprocessed using the domain knowledge of materials science underlying chemical and physical concepts, and a suitable selection of the applied ML model. Recent breakthroughs in ML techniques have created vast opportunities for not only overcoming long-standing mechanics problems but also for developing unprecedented materials design strategies. In this review, we first present a brief introduction of state-of-the-art ML models, algorithms and structures. Then, we discuss the importance of data collection, generation and preprocessing. The applications in mechanical property prediction, materials design and computational methods using ML-based approaches are summarized, followed by perspectives on opportunities and open challenges in this emerging and exciting field.
169 citations
University of Pennsylvania1, National Institute of Standards and Technology2, University of Maryland, College Park3, Boston University4, University at Buffalo5, Fordham University6, Brookhaven National Laboratory7, Columbia University8, Massachusetts Institute of Technology9, Argonne National Laboratory10, University of Chicago11, Cornell University12, California Institute of Technology13, SLAC National Accelerator Laboratory14, Toyota15, Florida State University16, Lawrence Berkeley National Laboratory17, Wright-Patterson Air Force Base18
TL;DR: In this paper, the authors discuss the specific challenges and opportunities related to materials discovery and development that will emerge from this new paradigm and outline the current status, barriers and needed investments, culminating with a vision for the path forward.
Abstract: Summary Solutions to many of the world's problems depend upon materials research and development. However, advanced materials can take decades to discover and decades more to fully deploy. Humans and robots have begun to partner to advance science and technology orders of magnitude faster than humans do today through the development and exploitation of closed-loop, autonomous experimentation systems. This review discusses the specific challenges and opportunities related to materials discovery and development that will emerge from this new paradigm. Our perspective incorporates input from stakeholders in academia, industry, government laboratories, and funding agencies. We outline the current status, barriers, and needed investments, culminating with a vision for the path forward. We intend the article to spark interest in this emerging research area and to motivate potential practitioners by illustrating early successes. We also aspire to encourage a creative reimagining of the next generation of materials science infrastructure. To this end, we frame future investments in materials science and technology, hardware and software infrastructure, artificial intelligence and autonomy methods, and critical workforce development for autonomous research.
116 citations
04 Nov 2020
TL;DR: Several algorithms, with a focus on machine learning methods, are reviewed and explored to systematically tackle the three main stages of the additive manufacturing process: geometrical design, process parameter configuration, and in situ anomaly detection.
Abstract: Summary Increasing demand for the fabrication of components with complex designs has spurred a revolution in manufacturing methods. Additive manufacturing stands out as a promising technology when it comes to prototyping multi-functional and multi-material designs. However, challenges still exist in the additive manufacturing process, such as mismatched material properties, lack of build consistency, and pervasive imperfections in the printed part. These inherent challenges can be avoided by implementing algorithms to detect imperfections and modulate printing parameters in real time. In this paper, several algorithms, with a focus on machine learning methods, are reviewed and explored to systematically tackle the three main stages of the additive manufacturing process: geometrical design, process parameter configuration, and in situ anomaly detection. Current challenges and future opportunities for algorithmically driven additive manufacturing processes, as well as potential applications to other manufacturing methods, are also discussed.
93 citations
Cites background or methods from "A Bayesian experimental autonomous ..."
...TO also enables the fabrication of complex lattice structures with the same time efficiency as bulk structures in AM....
[...]
...With the GA as a basis, multiple studies have been conducted on applying GA to optimize the process parameters in AM.71–73 The first example that will be discussed 1546 Matter 3, 1541–1556, November 4, 2020 is a GAmodel with the design of experiments (DOE) to find the optimal combination of process parameters that can minimize surface roughness and porosity characteristics of the printed part....
[...]
...Moreover, theMLmodel actively learns from training data generated around the current design point to reduce the bias within the local convex hull.68 These studies show that ML can be used as a promising approach to accelerate the materials design process, which can be translated into a physical part with advances in AM....
[...]
...Studies have shown promising results on acquiring a satisfying amount of training data using high-throughput methods in AM.(78,88) Additionally, the computational cost for training will also increase dramatically when the setting variables are augmented....
[...]
..., Bayesian optimization) can be used to save the computational cost and time needed for ML.(85,88)...
[...]
TL;DR: A closed-loop system capable of carrying out parallel autonomous process optimization experiments in batch with significantly reduced cycle times is developed, and it is found that the definition of a set of meaningful, broad, and unbiased process parameters was the most critical aspect of a successful optimization.
Abstract: Autonomous process optimization involves the human intervention-free exploration of a range process parameters to improve responses such as product yield and selectivity. Utilizing off-the-shelf components, we develop a closed-loop system for carrying out parallel autonomous process optimization experiments in batch. Upon implementation of our system in the optimization of a stereoselective Suzuki-Miyaura coupling, we find that the definition of a set of meaningful, broad, and unbiased process parameters is the most critical aspect of successful optimization. Importantly, we discern that phosphine ligand, a categorical parameter, is vital to determination of the reaction outcome. To date, categorical parameter selection has relied on chemical intuition, potentially introducing bias into the experimental design. In seeking a systematic method for selecting a diverse set of phosphine ligands, we develop a strategy that leverages computed molecular feature clustering. The resulting optimization uncovers conditions to selectively access the desired product isomer in high yield. An automated closed-loop system optimizes a stereoselective Suzuki-Miyaura reaction using a machine learning algorithm that incorporates unbiased and categorical process parameters.
80 citations
TL;DR: A state-of-the-art account on engineered 2D nanomaterials and their applications in emerging water technologies involving separation, adsorption, photocatalysis, and pollutant detection and to guide the design of next-generation 2DM systems for the development of selective, multifunctional, programmable, and even intelligent water technologies.
Abstract: Water scarcity has become an increasingly complex challenge with the growth of the global population, economic expansion, and climate change, highlighting the demand for advanced water treatment technologies that can provide clean water in a scalable, reliable, affordable, and sustainable manner. Recent advancements on 2D nanomaterials (2DM) open a new pathway for addressing the grand challenge of water treatment owing to their unique structures and superior properties. Emerging 2D nanostructures such as graphene, MoS2, MXene, h-BN, g-C3N4, and black phosphorus have demonstrated an unprecedented surface-to-volume ratio, which promises ultralow material use, ultrafast processing time, and ultrahigh treatment efficiency for water cleaning/monitoring. In this review, we provide a state-of-the-art account on engineered 2D nanomaterials and their applications in emerging water technologies, involving separation, adsorption, photocatalysis, and pollutant detection. The fundamental design strategies of 2DM are discussed with emphasis on their physicochemical properties, underlying mechanism and targeted applications in different scenarios. This review concludes with a perspective on the pressing challenges and emerging opportunities in 2DM-enabled wastewater treatment and water-quality monitoring. This review can help to elaborate the structure–processing–property relationship of 2DM, and aims to guide the design of next-generation 2DM systems for the development of selective, multifunctional, programmable, and even intelligent water technologies. The global significance of clean water for future generations sheds new light and much inspiration in this rising field to enhance the efficiency and affordability of water treatment and secure a global water supply in a growing portion of the world.
77 citations
References
More filters
Proceedings Article•
03 Dec 2012TL;DR: This work describes new algorithms that take into account the variable cost of learning algorithm experiments and that can leverage the presence of multiple cores for parallel experimentation and shows that these proposed algorithms improve on previous automatic procedures and can reach or surpass human expert-level optimization for many algorithms.
Abstract: The use of machine learning algorithms frequently involves careful tuning of learning parameters and model hyperparameters. Unfortunately, this tuning is often a "black art" requiring expert experience, rules of thumb, or sometimes brute-force search. There is therefore great appeal for automatic approaches that can optimize the performance of any given learning algorithm to the problem at hand. In this work, we consider this problem through the framework of Bayesian optimization, in which a learning algorithm's generalization performance is modeled as a sample from a Gaussian process (GP). We show that certain choices for the nature of the GP, such as the type of kernel and the treatment of its hyperparameters, can play a crucial role in obtaining a good optimizer that can achieve expertlevel performance. We describe new algorithms that take into account the variable cost (duration) of learning algorithm experiments and that can leverage the presence of multiple cores for parallel experimentation. We show that these proposed algorithms improve on previous automatic procedures and can reach or surpass human expert-level optimization for many algorithms including latent Dirichlet allocation, structured SVMs and convolutional neural networks.
5,654 citations
01 Jan 2016
TL;DR: This review paper introduces Bayesian optimization, highlights some of its methodological aspects, and showcases a wide range of applications.
Abstract: Big Data applications are typically associated with systems involving large numbers of users, massive complex software systems, and large-scale heterogeneous computing and storage architectures. The construction of such systems involves many distributed design choices. The end products (e.g., recommendation systems, medical analysis tools, real-time game engines, speech recognizers) thus involve many tunable configuration parameters. These parameters are often specified and hard-coded into the software by various developers or teams. If optimized jointly, these parameters can result in significant improvements. Bayesian optimization is a powerful tool for the joint optimization of design choices that is gaining great popularity in recent years. It promises greater automation so as to increase both product quality and human productivity. This review paper introduces Bayesian optimization, highlights some of its methodological aspects, and showcases a wide range of applications.
3,703 citations
TL;DR: The common design motifs of a range of natural structural materials are reviewed, and the difficulties associated with the design and fabrication of synthetic structures that mimic the structural and mechanical characteristics of their natural counterparts are discussed.
Abstract: Natural structural materials are built at ambient temperature from a fairly limited selection of components. They usually comprise hard and soft phases arranged in complex hierarchical architectures, with characteristic dimensions spanning from the nanoscale to the macroscale. The resulting materials are lightweight and often display unique combinations of strength and toughness, but have proven difficult to mimic synthetically. Here, we review the common design motifs of a range of natural structural materials, and discuss the difficulties associated with the design and fabrication of synthetic structures that mimic the structural and mechanical characteristics of their natural counterparts.
3,083 citations
TL;DR: This work focuses on the interplay between the mechanisms that individually contribute to strength and toughness, noting that these phenomena can originate from very different lengthscales in a material's structural architecture.
Abstract: The attainment of both strength and toughness is a vital requirement for most structural materials; unfortunately these properties are generally mutually exclusive. Although the quest continues for stronger and harder materials, these have little to no use as bulk structural materials without appropriate fracture resistance. It is the lower-strength, and hence higher-toughness, materials that find use for most safety-critical applications where premature or, worse still, catastrophic fracture is unacceptable. For these reasons, the development of strong and tough (damage-tolerant) materials has traditionally been an exercise in compromise between hardness versus ductility. Drawing examples from metallic glasses, natural and biological materials, and structural and biomimetic ceramics, we examine some of the newer strategies in dealing with this conflict. Specifically, we focus on the interplay between the mechanisms that individually contribute to strength and toughness, noting that these phenomena can originate from very different lengthscales in a material's structural architecture. We show how these new and natural materials can defeat the conflict of strength versus toughness and achieve unprecedented levels of damage tolerance within their respective material classes.
2,350 citations
01 Jan 2015
TL;DR: A conceptual overview of rapid prototyping and layered manufacturing is given, beginning with the fundamentals so that readers can get up to speed quickly as mentioned in this paper, with a broad range of technical questions to ensure comprehensive understanding of the concepts covered.
Abstract: This book covers in detail the various aspects of joining materials to form parts. A conceptual overview of rapid prototyping and layered manufacturing is given, beginning with the fundamentals so that readers can get up to speed quickly. Unusual and emerging applications such as micro-scale manufacturing, medical applications, aerospace, and rapid manufacturing are also discussed. This book provides a comprehensive overview of rapid prototyping technologies as well as support technologies such as software systems, vacuum casting, investment casting, plating, infiltration and other systems. This book also: Reflects recent developments and trends and adheres to the ASTM, SI, and other standards Includes chapters on automotive technology, aerospace technology and low-cost AM technologies Provides a broad range of technical questions to ensure comprehensive understanding of the concepts covered.
1,878 citations