An official website of the United States government
Rolling bearing is a critical component of machinery that has been widely applied in manufacturing, transportation, aerospace, and power and energy industries. The timely and accurate bearing fault detection thus is of vital importance. Computational data-driven deep learning has recently become a prevailing approach for bearing fault detection. Despite the progress of the deep learning approach, the deep learning performance is hinged upon the size of labeled data, the acquisition of which is expensive in actual implementation. Unlabeled data, on the other hand, are inexpensive. In this research, we develop a new semi-supervised learning method built upon the autoencoder to fully utilize a large amount of unlabeled data together with limited labeled data to enhance fault detection performance. Compared with the state-of-the-art semi-supervised learning methods, this proposed method can be more conveniently implemented with fewer hyperparameters to be tuned. In this method, a joint loss is established to account for the effects of labeled and unlabeled data, which is subsequently used to direct the backpropagation training. Systematic case studies using the Case Western Reserve University (CWRU) rolling bearing dataset are carried out, in which the effectiveness of this new method is verified by comparing it with other well-established baseline methods. Specifically, nearly all emulation runs using the proposed methodology can lead to around 2%–5% accuracy increase, indicating its robustness in performance enhancement.
more »
« less
Machine learning for human design: Sketch interface for structural morphology ideation using neural networks
Formal computational approaches in the realm of engineering and architecture, such as parametric modelling and optimization, are increasingly powerful, allowing for systematic and rigorous design processes. However, these methods often bring a steep learning curve, require previous expertise, or are unintuitive and unnatural to human design. On the other hand, analog design methods such as hand sketching are commonly used by architects and engineers alike, and constitute quick, easy, and almost primal modes of generating and transferring design concepts, which in turn facilitates the sharing of ideas and feedback. In the advent of increasing computational power and developments in data analysis, deep learning, and other emerging technologies, there is a potential to bridge the gap between these seemingly divergent processes to develop new hybrid approaches to design. Such methods can provide designers with new opportunities to harness the systematic and data-driven power of computation and performance analysis while maintaining a more creative and intuitive design interface. This paper presents a new method for interpreting human designs in sketch format and predicting their structural performance using recent advances in deep learning. The paper also demonstrates how this new technique can be used in design workflows including performance-based guidance and interpolations between concepts.
more »
« less
- Award ID(s):
- 1854833
- PAR ID:
- 10287795
- Date Published:
- Journal Name:
- Proceedings of the International Association for Shell and Spatial Structures (IASS) Symposium 2020/2021
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Generative artificial intelligence (GenAI) systems introduce new possibilities for enhancing professionals’ workflows, enabling novel forms of human–AI co-creation. However, professionals often strug- gle to learn to work with GenAI systems effectively. While research has begun to explore the design of interfaces that support users in learning to co-create with GenAI, we lack systematic approaches to investigate the effectiveness of these supports. In this paper, we present a systematic approach for studying how to support learn- ing to co-create with GenAI systems, informed by methods and concepts from the learning sciences. Through an experimental case study, we demonstrate how our approach can be used to study and compare the impacts of different types of learning supports in the context of text-to-image GenAI models. Reflecting on these results, we discuss directions for future work aimed at improving interfaces for human–AI co-creation.more » « less
-
Designing and generating new data under targeted properties has been attracting various critical applications such as molecule design, image editing and speech synthesis. Traditional hand-crafted approaches heavily rely on expertise experience and intensive human efforts, yet still suffer from the insufficiency of scientific knowledge and low throughput to support effective and efficient data generation. Recently, the advancement of deep learning has created the opportunity for expressive methods to learn the underlying representation and properties of data. Such capability provides new ways of determining the mutual relationship between the structural patterns and functional properties of the data and leveraging such relationships to generate structural data, given the desired properties. This article is a systematic review that explains this promising research area, commonly known as controllable deep data generation. First, the article raises the potential challenges and provides preliminaries. Then the article formally defines controllable deep data generation, proposes a taxonomy on various techniques and summarizes the evaluation metrics in this specific domain. After that, the article introduces exciting applications of controllable deep data generation, experimentally analyzes and compares existing works. Finally, this article highlights the promising future directions of controllable deep data generation and identifies five potential challenges.more » « less
-
From higher computational efficiency to enabling the discovery of novel and complex structures, deep learning has emerged as a powerful framework for the design and optimization of nanophotonic circuits and components. However, both data-driven and exploration-based machine learning strategies have limitations in their effectiveness for nanophotonic inverse design. Supervised machine learning approaches require large quantities of training data to produce high-performance models and have difficulty generalizing beyond training data given the complexity of the design space. Unsupervised and reinforcement learning-based approaches on the other hand can have very lengthy training or optimization times associated with them. Here we demonstrate a hybrid supervised learning and reinforcement learning approach to the inverse design of nanophotonic structures and show this approach can reduce training data dependence, improve the generalizability of model predictions, and significantly shorten exploratory training times. The presented strategy thus addresses several contemporary deep learning-based challenges, while opening the door for new design methodologies that leverage multiple classes of machine learning algorithms to produce more effective and practical solutions for photonic design.more » « less
-
Data-driven methods have attracted increasingly more attention in materials research since the advent of the material genome initiative. The combination of materials science with computer science, statistics, and data-driven methods aims to expediate materials research and applications and can utilize both new and archived research data. In this paper, we present a data driven and deep learning approach that builds a portion of the structure–property relationship for polymer nanocomposites. Analysis of archived experimental data motivates development of a computational model which allows demonstration of the approach and gives flexibility to sufficiently explore a wide range of structures. Taking advantage of microstructure reconstruction methods and finite element simulations, we first explore qualitative relationships between microstructure descriptors and mechanical properties, resulting in new findings regarding the interplay of interphase, volume fraction and dispersion. Then we present a novel deep learning approach that combines convolutional neural networks with multi-task learning for building quantitative correlations between microstructures and property values. The performance of the model is compared with other state-of-the-art strategies including two-point statistics and structure descriptor-based approaches. Lastly, the interpretation of the deep learning model is investigated to show that the model is able to capture physical understandings while learning.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
