arXiv subjects

Ling Lan

Publications and source records attributed to Ling Lan.

8 recordsLinked to original sources

arXiv2026

A welding penetration prediction model for laser welding process based on self-supervised learning using physics-informed neural networks

The laser welding full-penetration is of critical importance, as it constitutes one of the fundamental factors in achieving defect-free welded joints. Accurate prediction of the penetration state is therefore essential for ensuring weld quality. To this end, this paper introduces SimPhysNet, a novel algorithm that achieves high classification accuracy in laser welding penetration prediction using only a limited number of labelled images. This approach effectively overcomes the limitations of supervised learning classification algorithms, which are hindered in industrial applications by their dependence on extensive, high-quality labelled data. The core of SimPhysNet is a unique self-supervised learning paradigm that embeds physical priors into a contrastive learning framework. By incorporating a physics-informed neural network (PINN), the model is guided to extract physically meaningful features of the molten pool and keyhole from a large set of unlabelled data, while three image augmentation tasks further enhance its generalization capabilities. Subsequently, a few-shot learning strategy, based on prototypical networks, enables robust classification by constructing class representations from a minimal set of labelled images. Experimental results demonstrate that SimPhysNet achieves a classification accuracy of 96.06% using only 200 labelled images (approximately 5% of the total labelled dataset), which is comparable to the performance of conventional supervised learning algorithms that utilize the entire labelled dataset. This work presents a new, efficient, and highly accurate method, providing the way for the intelligent automation of laser welding.

cs.CV↗

arXiv2025

CbLDM: A Diffusion Model for recovering nanostructure from atomic pair distribution function

The nanostructure inverse problem is an attractive problem that helps researchers to understand the relationship between the properties and the structure of nanomaterials. This study focuses on the problem of recovering the model system of monometallic nanoparticles (MMNPs) from their pair distribution function (PDF) and regards it as a highly ill-posed conditional generation task. This study proposes a Condition-based Latent Diffusion Model (CbLDM) as a feasible solution to this problem. This model demonstrates an acceleration approach within the framework of a latent diffusion model by using conditional priors to estimate the conditional posterior distribution, which is an approximate distribution of p(z|x). In addition, this study uses Laplacian matrix instead of distance matrix to recover the nanostructure, which helps to improve stability. Our study demonstrates that a latent diffusion model with a conditional prior can generate nanostructures that are consistent with PDF observations and physically meaningful, thereby laying the groundwork for subsequent more complex inverse problems.

cs.LG↗

arXiv2024

A continuous symmetry breaking measure for finite clusters using Jensen-Shannon divergence

A quantitative measure of symmetry breaking is introduced that allows the quantification of which symmetries are most strongly broken due to the introduction of some kind of defect in a perfect structure. The method uses a statistical approach based on the Jensen-Shannon divergence. The measure is calculated by comparing the transformed atomic density function with its original. Software code is presented that carries the calculations out numerically using Monte Carlo methods. The behavior of this symmetry breaking measure is tested for various cases including finite size crystallites (where the surfaces break the crystallographic symmetry), atomic displacements from high symmetry positions, and collective motions of atoms due to rotations of rigid octahedra. The approach provides a powerful tool for assessing local symmetry breaking and offers new insights that can help researchers understand how different structural distortions affect different symmetry operations.

cond-mat.mtrl-sci↗

arXiv2023

Towards End-to-End Structure Solutions from Information-Compromised Diffraction Data via Generative Deep Learning

The revolution in materials in the past century was built on a knowledge of the atomic arrangements and the structure-property relationship. The sine qua non for obtaining quantitative structural information is single crystal crystallography. However, increasingly we need to solve structures in cases where the information content in our input signal is significantly degraded, for example, due to orientational averaging of grains, finite size effects due to nanostructure, and mixed signals due to sample heterogeneity. Understanding the structure property relationships in such situations is, if anything, more important and insightful, yet we do not have robust approaches for accomplishing it. In principle, machine learning (ML) and deep learning (DL) are promising approaches since they augment information in the degraded input signal with prior knowledge learned from large databases of already known structures. Here we present a novel ML approach, a variational query-based multi-branch deep neural network that has the promise to be a robust but general tool to address this problem end-to-end. We demonstrate the approach on computed powder x-ray diffraction (PXRD), along with partial chemical composition information, as input. We choose as a structural representation a modified electron density we call the Cartesian mapped electron density (CMED), that straightforwardly allows our ML model to learn material structures across different chemistries, symmetries and crystal systems. When evaluated on theoretically simulated data for the cubic and trigonal crystal systems, the system achieves up to $93.4\%$ average similarity with the ground truth on unseen materials, both with known and partially-known chemical composition information, showing great promise for successful structure solution even from degraded and incomplete input data.

physics.comp-ph↗

arXiv2023

Stretched Non-negative Matrix Factorization

An algorithm is described and tested that carries out a non negative matrix factorization (NMF) ignoring any stretching of the signal along the axis of the independent variable. This extended NMF model is called StretchedNMF. Variability in a set of signals due to this stretching is then ignored in the decomposition. This can be used, for example, to study sets of powder diffraction data collected at different temperatures where the materials are undergoing thermal expansion. It gives a more meaningful decomposition in this case where the component signals resemble signals from chemical components in the sample. The StretchedNMF model introduces a new variable, the stretching factor, to describe any expansion of the signal. To solve StretchedNMF, we discretize it and employ Block Coordinate Descent framework algorithms. The initial experimental results indicate that StretchedNMF model outperforms the conventional NMF for sets of data with such an expansion. A further enhancement to StretchedNMF for the case of powder diffraction data from crystalline materials called Sparse-StretchedNMF, which makes use of the sparsity of the powder diffraction signals, allows correct extractions even for very small stretches where StretchedNMF struggles. As well as demonstrating the model performance on simulated PXRD patterns and atomic pair distribution functions (PDFs), it also proved successful when applied to real data taken from an in situ chemical reaction experiment.

cond-mat.mtrl-sci↗

arXiv2022

Towards a machine-readable literature: finding relevant papers based on an uploaded powder diffraction pattern

We investigate a prototype application for machine-readable literature. The program is called "pyDataRecognition" and serves as an example of a data-driven literature search, where the literature search query is an experimental data-set provided by the user. The user uploads a powder pattern together with the radiation wavelength. The program compares the user data to a database of existing powder patterns associated with published papers and produces a rank ordered according to their similarity score. The program returns the digital object identifier (doi) and full reference of top ranked papers together with a stack plot of the user data alongside the top five database entries. The paper describes the approach and explores successes and challenges.

cond-mat.mtrl-sci↗

arXiv2022

Robustness test of the spacegroupMining model for determining space groups from atomic pair distribution function data

Machine learning models based on convolutional neural networks have been used for predicting space groups of crystal structures from their atomic pair distribution function (PDF). However, the PDFs used to train the model are calculated using a fixed set of parameters that reflect specific experimental conditions, and the accuracy of the model when given PDFs generated with different choices of these parameters is unknown. In this paper, we report that the results of the top-1 accuracy and top-6 accuracy are robust when applied to PDFs of different choices of experimental parameters $r_\text{max}$, $Q_\text{max}$, $Q_\text{damp}$ and atomic displacement parameters.

cond-mat.mtrl-sci↗

arXiv2019

Self-Folding Metasheets: The Optimal Pattern of Strain of Miura-Ori Folded State

Self-folding origami has emerged as a tool to make functional objects in material science. The common idea is to pattern a sheet with creases and activate them to have the object fold spontaneously into a desired configuration. This article shows that collinear quadrilateral metasheets are able to fold into the Miura-Ori configuration, if we only impose strain on part of their creases. In this study, we define and determine the optimal pattern of strain (OPS) on a collinear quadrilateral metasheet, that is the pattern of minimum "functional" creases with which the self-folding metasheet can fold into Miura-Ori state stably. By comparing the energy evolution along the folding pathway of each possible folded state under OPS, we conclude that the energy predominance of the desired Miura-Ori pathway during the initial period of time accounts for why the OPS works. Furthermore, we measure the projected force of the OPS on the intial flat metasheet and give insights on how to determine the OPS using only local information of the initial flat state.

cond-mat.soft↗