Extracting local nucleation fields in permanent magnets using machine learning

Abstract

Microstructural features play an important role in the quality of permanent magnets. The coercivity is greatly influenced by crystallographic defects, like twin boundaries, as is well known for MnAl-C. It would be very useful to be able to predict the macroscopic coercivity from microstructure imaging. Although this is not possible now, in the present work we examine a related question, namely the prediction of simulated nucleation fields of a quasi-three-dimensional (rescaled and extruded) system constructed from a two-dimensional image. We extract features of the image and analyze them via machine learning. A large number of extruded systems are constructed from 10 × 10 pixel sub-images of an Electron Backscatter Diffraction (EBSD) image using an automated meshing procedure. A local nucleation field is calculated by micromagnetic simulation of each quasi-three-dimensional system. Decision trees, trained with the simulation results, can predict nucleation fields of these quasi-three-dimensional systems from new images within seconds. As for now we cannot quantitatively predict the macroscopic coercivity, nevertheless we can identify weak spots in the magnet and see trends in the nucleation field distribution.

Artificial Intelligence—Engineering Magnetic Materials: Current Status and a Brief Perspective

“…Consequently, the recognition of these sites allows determining specifically where H c should be enhanced via, for instance, diffusion processes (already industrially implemented). In a similar scope, AI has also enabled the assessment of the local magnetic nucleation fields, now implementing electron backscatter diffraction data responsible for training supervised learning algorithms, to generate hysteresis curves of Mn-Al-C magnets, as illustrated in Figure 6 [22]. By using micromagnetic simulation of quasi-3D systems based on 2D images, the influence of microstructural features such as crystallographic orientation and size of grains have been assessed to identify "weak" regions of the magnet and identify partial dependences on the relations of such features to predict magnetic results.…”

Self-supervised learning and prediction of microstructure evolution with convolutional recurrent neural networks

“… 32 , 33 , 34 , 35 They have seen diverse applications ranging from the discovery of new materials 36 , 37 , 38 , 39 , 40 to the predictions of materials’ properties, 41 , 42 , 43 , 44 , 45 the development of accurate and efficient potentials for atomistic simulations, 46 , 47 , 48 , 49 microscopic and spectroscopic data analysis and processing, 50 , 51 , 52 , 53 , 54 , 55 , 56 , 57 , 58 , 59 , 60 , 61 , 62 and effective inference of a material’s properties from a limited experimental dataset. 63 , 64 A large number of these works are devoted to material microstructure, with encouraging results, including microstructure classification and quantification, 50 , 51 , 52 , 53 , 54 , 65 , 66 , 67 image segmentation, 55 , 56 predictions of microstructure-property relations, 57 , 68 , 69 , 70 mapping processing-microstructure relations, 71 , 72 , 73 , 74 microstructure optimization, 75 , 76 , 77 and equilibrium configuration prediction. 78 Datasets in these works are mainly in the form of static microstructure images.…”

“…[32][33][34][35] They have seen diverse applications ranging from the discovery of new materials [36][37][38][39][40] to the predictions of materials' properties, [41][42][43][44][45] the development of accurate and efficient potentials for atomistic simulations, [46][47][48][49] microscopic and spectroscopic data analysis and processing, [50][51][52][53][54][55][56][57][58][59][60][61][62] and effective inference of a material's properties from a limited experimental dataset. 63,64 A large number of these works are devoted to material microstructure, with encouraging results, including microstructure classification and quantification, [50][51][52][53][54][65][66][67] image segmentation, 55,56 predictions of microstructure-property relations, 57,[68][69]…”

Towards End-to-End Deep Learning Performance Analysis of Electric Motors

“…In the aforementioned papers, DL CNNs were primarily used to assist the analysis and the final results were still obtained using FEA. However, there is an enormous potential in developing end-to-end DL methods [9,[25][26][27]. In this paper, DL CNNs are used for high-accuracy IPMSM output prediction.…”