arXiv subjects

Md Mesbah Uddin

Publications and source records attributed to Md Mesbah Uddin.

3 recordsLinked to original sources

arXiv2023

Addressing computational challenges in physical system simulations with machine learning

In this paper, we present a machine learning-based data generator framework tailored to aid researchers who utilize simulations to examine various physical systems or processes. High computational costs and the resulting limited data often pose significant challenges to gaining insights into these systems or processes. Our approach involves a two-step process: initially, we train a supervised predictive model using a limited simulated dataset to predict simulation outcomes. Subsequently, a reinforcement learning agent is trained to generate accurate, simulation-like data by leveraging the supervised model. With this framework, researchers can generate more accurate data and know the outcomes without running high computational simulations, which enables them to explore the parameter space more efficiently and gain deeper insights into physical systems or processes. We demonstrate the effectiveness of the proposed framework by applying it to two case studies, one focusing on earthquake rupture physics and the other on new material development.

cs.LG↗

arXiv2023

Study the effect of scratching depth and ceramic-metal ratio on the scratch behavior of NbC/Nb Ceramic/Metal nano-laminates using molecular dynamics simulation and machine learning

Developing a new class of coating materials is necessary to meet the increasing demands of energy and defense-related technologies, aerospace engineering, and harsh environmental conditions. Functional-based coatings, such as ceramic-metal nanolaminates, have gained popularity due to their ability to be customized according to specific requirements. To design and develop advanced coatings with the necessary functionalities, it is crucial to understand the effects of various parameters on the mechanical and tribological properties of these coatings. In this study, we investigate the impact of penetration depth, individual layer thickness, and ceramic-metal ratio on the mechanical and tribological properties of ceramic-metal nanolaminates, particularly NbC/Nb. Our findings reveal that the thickness of the individual metallic and ceramic layers significantly affects the coatings' properties. However, some models exhibited punctures on the top ceramic layer, which altered the scratching behavior and reduced the impact of layer thickness on it. This is because the top ceramic layer's thickness is too low, and the indenter can easily puncture it instead of pushing the ceramic atoms. The minimum thickness required to resist indentation is called the critical thickness, which depends on the indentation size and penetration depth. In the latter part of this paper, we employed machine learning to reduce computational costs, and the model predicts the friction coefficient with an R-squared value of 0.958.

cond-mat.mtrl-sci↗

arXiv2023

Estimating uncertainty of earthquake rupture using Bayesian neural network

Bayesian neural networks (BNN) are the probabilistic model that combines the strengths of both neural network (NN) and stochastic processes. As a result, BNN can combat overfitting and perform well in applications where data is limited. Earthquake rupture study is such a problem where data is insufficient, and scientists have to rely on many trial and error numerical or physical models. Lack of resources and computational expenses, often, it becomes hard to determine the reasons behind the earthquake rupture. In this work, a BNN has been used (1) to combat the small data problem and (2) to find out the parameter combinations responsible for earthquake rupture and (3) to estimate the uncertainty associated with earthquake rupture. Two thousand rupture simulations are used to train and test the model. A simple 2D rupture geometry is considered where the fault has a Gaussian geometric heterogeneity at the center, and eight parameters vary in each simulation. The test F1-score of BNN (0.8334), which is 2.34% higher than plain NN score. Results show that the parameters of rupture propagation have higher uncertainty than the rupture arrest. Normal stresses play a vital role in determining rupture propagation and are also the highest source of uncertainty, followed by the dynamic friction coefficient. Shear stress has a moderate role, whereas the geometric features such as the width and height of the fault are least significant and uncertain.

stat.ML↗