Microstructural diagram for steel based on crystallography with machine learning

Tsutsui, Kazumasa; Terasaki, Hidenori; Maemura, Tatsuya; Hayashi, Kotaro; Moriguchi, Koji; Morito, Shigekazu

doi:10.1016/j.commatsci.2018.12.003

Cited by 33 publications

(13 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the same material system - dual phase steel - the damage mechanism has been detected using deep learning 52 . Finally, machine learning has been recently leveraged to construct the phase diagram of low carbon steel 53 .…”

Section: Resultsmentioning

confidence: 99%

Degradation Mechanism Detection in Photovoltaic Backsheets by Fully Convolutional Neural Network

Zhang

Grant

Bruckman

et al. 2019

Sci Rep

View full text Add to dashboard Cite

Materials and devices age with time. Material aging and degradation has important implications for lifetime performance of materials and systems. While consensus exists that materials should be studied and designed for degradation, materials inspection during operation is typically performed manually by technicians. The manual inspection makes studies prone to errors and uncertainties due to human subjectivity. In this work, we focus on automating the process of degradation mechanism detection through the use of a fully convolutional deep neural network architecture (F-CNN). We demonstrate that F-CNN architecture allows for automated inspection of cracks in polymer backsheets from photovoltaic (PV) modules. The developed F-CNN architecture enabled an end-to-end semantic inspection of the PV module backsheets by applying a contracting path of convolutional blocks (encoders) followed by an expansive path of decoding blocks (decoders). First, the hierarchy of contextual features is learned from the input images by encoders. Next, these features are reconstructed to the pixel-level prediction of the input by decoders. The structure of the encoder and the decoder networks are thoroughly investigated for the multi-class pixel-level degradation type prediction for PV module backsheets. The developed F-CNN framework is validated by reporting degradation type prediction accuracy for the pixel level prediction at the level of 92.8%.

show abstract

Section: Resultsmentioning

confidence: 99%

Degradation Mechanism Detection in Photovoltaic Backsheets by Fully Convolutional Neural Network

Zhang

Grant

Bruckman

et al. 2019

Sci Rep

View full text Add to dashboard Cite

show abstract

“…The labeling of each phase thus remains a challenging and tedious task. Few authors have therefore combined machine learning approaches with EBSD analyses for automatic phase characterization [9,12,22].…”

Section: Introductionmentioning

confidence: 99%

Leveraging EBSD data by deep learning for bainite, ferrite and martensite segmentation

Breumier

Ostormujof²,

Gey³

et al. 2022

Materials Characterization

View full text Add to dashboard Cite

“…[16][17][18][19][20][21][22][23] Given that, Shi et al [24][25][26] have reviewed the application of ML method in materials engineering as well as the functional field. In addition, we identify several common themes associated with the application of ML approaches, such as predictions of phase diagrams, [27] crystal structures, [28] damage identification, [29][30][31][32][33] and materials properties, [34][35][36][37] which significantly accelerates the discovery of new materials via a data-driven materials research approach. The ML approaches commonly exhibit excellent performance in dealing with the complex multivariate nonlinear relationship between input and output variables in view of the existing data.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning‐Aided Process Design: Modeling and Prediction of Transformation Temperature for Pearlitic Steel

Qiao

Zhu

Yuan

2021

steel research int.

View full text Add to dashboard Cite

In this article, different machine learning (ML) algorithms are provided to predict the transformation temperature of pearlite using relevant material descriptors, austenitizing temperature, and cooling rate. To search for an appropriate model, the predictive performance of ML model including artificial neural network (ANN), generalized regression neural network (GRNN), radial basis function neural network (RBFNN), and extreme learning machine (ELM) is evaluated and compared on testing dataset. To quickly find the appropriate parameters, artificial fish swarm algorithm (AFSA) is applied to further improve the prediction accuracy of ANN model. In view of the distinguished prediction performance, the regression analysis shows that the GRNN model performs favorably against the other learning models. To verify the superiority of proposed model, a type of pearlitic steel with hypereutectoid composition is prepared. The phase transition temperature and transformation products are determined using diametral dilatometric and microscopy technique through continuous cooling transformation with cooling rates of 1 and . As a result, the experimental results agree well with the predictive results. The work proposes a reliable model to predict the phase transition temperature which can facilitate the optimization of process parameters to achieve the best possible microstructure.

show abstract

Microstructural diagram for steel based on crystallography with machine learning

Cited by 33 publications

References 35 publications

Degradation Mechanism Detection in Photovoltaic Backsheets by Fully Convolutional Neural Network

Degradation Mechanism Detection in Photovoltaic Backsheets by Fully Convolutional Neural Network

Leveraging EBSD data by deep learning for bainite, ferrite and martensite segmentation

Machine Learning‐Aided Process Design: Modeling and Prediction of Transformation Temperature for Pearlitic Steel

Contact Info

Product

Resources

About