42 results
Showing 1–20 · 0.24 seconds · RSS
Machine Learning. Engineering
ISSN 3049-4761 · IOP Publishing · 42 articles in the index
-
Journal Article
Adaptive unknown fault detection and few-shot continual learning for condition monitoring in ultrasonic metal welding
Ultrasonic metal welding (UMW) is widely used in industrial applications but is sensitive to tool wear, surface contamination, and material variability, which can lead to unexpected process faults and unsatisfactory weld quality. Conventional monitoring systems typically rely on supervised learning…
-
Journal Article
In-process large language model requesting through knowledge graphs for interfacing with manufacturing digital twins
Digital twins (DTs) have potential to improve manufacturing processes by providing in-process monitoring and reducing trial-and-error while improving process prediction and performance. However, identifying anomalies or defects in a DT environment remains a very time intensive process that requires…
-
Journal Article
Unsupervised structural health monitoring using physics-based vibration features derived from an instrumented bridge mockup
Structural health monitoring (SHM) of civil infrastructure is often limited by the scarcity of labeled damage data required for supervised machine learning approaches. This study presents an unsupervised SHM framework that combines physics-based vibration features with novelty detection algorithms…
-
Journal Article
Hidden electrode subgroups revealed by independent analysis of charge and discharge formulas using Shapley behavioural spaces
Battery electrode materials exist in distinct charged and discharged states described by different chemical formulas, yet most machine learning studies use a single formula as the compositional descriptor for a material. We apply Shapley behavioural spaces to the Mendeleev-encoded charge and…
-
Journal Article
Hybrid quantum-classical classification of waste using the TrashNet dataset
This study presents a hybrid quantum-classical model for classifying waste images from the TrashNet dataset. TrashNet is a publicly accessible dataset of waste classifications. A classical convolutional neural network (CNN) is used to extract and reduce image features. These features are then…
-
Journal Article
Uncertainty quantification in machine learning models on synchrotron diffraction patterns
Machine learning (ML) models have been widely used to study functional relationships between attributes in physical science, engineering, medical science, social science, and other areas. The primary focus of such ML models is a deterministic prediction of the response variable with greater…
-
Journal Article
A CFD dataset for supporting machine-learning-based models of hydrogen–ammonia combustion
We present a computational fluid dynamics (CFD) dataset comprising 175 Reynolds-averaged Navier–Stokes simulations of staged ammonia–hydrogen combustion in the modified stagnation-point reverse-flow combustor developed at the Université Libre de Bruxelles. The dataset spans a three-dimensional…
-
Journal Article
Computation of waveguide eigenmodes by physics-informed neural networks
Physics-informed neural networks (PINNs) have emerged as powerful deep-learning frameworks for solving partial differential equations by directly embedding physical laws into the learning process. Leveraging this, we propose a PINN-based approach to solve the Helmholtz eigenvalue problem that…
-
Journal Article
Deducing window size and weights of Parzen classifier, using variogram function
The Parzen window classifier assigns a class label to an unlabeled observation based on the similarity to known observations (majority voting). This method requires the optimization of two parameters: the Parzen window radius and kernel function. These two parameters are often optimized…
-
Journal Article
A data-efficient approach to cyclic plasticity modeling via physics-augmented neural networks
Formulating constitutive models for plasticity presents significant challenges. In particular, cyclic inelastic analysis is computationally intensive, making machine learning a promising tool for capturing complex material behavior with reduced computational cost and enabling near real-time…
-
Journal Article
Algebraic adversarial attacks on explainability models
Classical adversarial attacks are phrased as a constrained optimisation problem. Despite the efficacy of a constrained optimisation approach to adversarial attacks, one cannot trace how an adversarial point was generated. In this work, we propose an algebraic approach to adversarial attacks and…
-
Journal Article
New algorithm for battery knee-onset identification and early prediction of end of life
Lithium-ion battery degradation is characterised by an initial quasi-linear capacity fade, transitioning to an accelerated, non-linear degradation phase at the knee onset point, which intensifies until the knee point, followed by rapid degradation to end-of-life (EOL). Accurate identification of…
-
Journal Article
Pretrained LLMs as real-time controllers for robot operated serial production line
The manufacturing industry is undergoing a transformative shift, driven by cutting-edge technologies like 5G, AI, and cloud computing. Despite these advancements, effective system control, which is crucial for optimizing production efficiency, remains a complex challenge due to the intricate…
-
Journal Article
Deep learning-based reconstruction of irregular laser ablation crater geometry from single plasma projection
During ultrafast laser ablation, a plasma plume forms as the laser interacts with the material, while the resulting crater geometry reflects how energy is deposited. We present a machine learning framework that reconstructs the 2D ablation pattern using only a single side-view image of the…
-
Journal Article
Simultaneous estimation of electrical and thermal properties of aluminum alloys from tone-burst eddy current thermography time-temperature data using machine learning
In this study, a machine learning (ML)-based inversion technique is presented to estimate the electrical and thermal properties of aluminum alloys using tone burst eddy current thermography (TBET). To obtain the necessary training data for the ML algorithm, a three-dimensional finite element…
-
Journal Article
Fredholm integral equations neural operator (FIE-NO) for data-driven boundary value problems
In this paper, we present a novel neural operator method, termed FIE-NO, inspired by Fredholm integral equations (FIE) and random Fourier features (RFF), tailored for solving data-driven boundary value problems (BVPs) with irregular boundaries. Unlike traditional computational approaches that…
-
Journal Article
Heterogeneous graph neural networks for short-term state forecasting in power systems across domains and time scales: a hydroelectric power plant case study
Accurate short-term state forecasting is crucial for the efficient and stable operation of modern power systems, especially as renewable and distributed energy resources introduce greater variability in power generation and load dynamics. Power systems today span multiple physical domains such as…
-
Journal Article
Efficient two-stage sequential machine learning model for progressive damage analysis in fiber-reinforced composites
This study presents a two-stage integrated machine learning surrogate model for microscale analysis of the nonlinear behavior of fiber-reinforced composites, with a particular focus on progressive damage analysis. The framework couples a recurrent neural network, used to capture the temporal…
-
Journal Article
Patch-based DWconv-ECA-transformer network for incipient fault diagnosis of analog circuits
To address the challenges in incipient fault diagnosis of analog circuits, including the difficulty of extracting fault features, the high computational complexity of existing models, and the limited diagnostic accuracy, this paper proposes a fault diagnosis method based on patch partitioning and…
-
Journal Article
A failure-efficient prediction of remaining useful life
Predicting the remaining useful life (RUL) is crucial for effective condition-based maintenance, enhancing cost-effectiveness, productivity, and safety by enabling timely maintenance interventions. A fundamental challenge in data-driven RUL prediction is the inherent need for historical failure…