10 September 2024 in EUREKA: Physics and Engineering
The use of Exploratory Data Analysis (EDA) and machine learning in material science has rapidly advanced in recent years. EDA enables researchers to thoroughly explore and analyze material datasets, while machine learning allows for the development of predictive models capable of understanding complex patterns within the data. This study aims to develop an optimization tool to enhance the analysis of tensile strength in stainless steel by leveraging integrated data exploration and machine learning approaches within the Streamlit framework. The developed tool consists of four main features: data visualization, correlation analysis, 3D visualization, and machine learning. The developed machine learning model has 14 input variables, including chemical elements and heat treatment temperatures. In this research, the machine learning features comprise three models: Decision Tree, Random Forest, and Artificial Neural Network. The research findings indicate that the optimization tool can automatically display stainless steel tensile strength data using available pandas profiling in the visualization feature. The correlation feature can illustrate the relationship between chemical elements and heat treatment temperatures concerning stainless steel tensile strength. The 3D visualization feature can be utilized to identify optimal values of chemical elements and heat treatment temperatures according to desired tensile strength. Meanwhile, the machine learning feature can accurately predict stainless steel tensile strength based on chemical composition and heat treatment temperatures. This is evident from the performance evaluation metrics of the Random Forest model, which achieved MAE of 10.36, RMSE of 14.44, and R-squared of 0.97
29 May 2019 in SN Applied Sciences
1 July 2023 in Materials Research Express
The first-order shear deformation theory (FSDT) was used to explore the natural frequency response of functionally graded piezoelectric plates subjected to static electrical and mechanical strain in this present study. A monomorph model for a functionally graded piezoelectric plate with material properties that change according to sigmoid law with respect to plate thickness has been considered. A three-dimensional finite element model with a free tetrahedral element mesh was created using COMSOL 4.2 Multiphysics® software, with each node having three degrees of freedom. Variations in the FGPM plate’s free vibration boundary conditions, composition, and geometry have all been investigated. In free vibration analysis, non-dimensional natural frequency of FGPM plate initially decreasing considerably and then remaining almost constant with the increase in volume fraction index when material property graded by power law. When material properties are varied by sigmoid law, with an increase in volume fraction index, the non-dimensional natural frequency of FGPM plates remains virtually constant. FGPM plates have a lower non-dimensional natural frequency if the thickness to width ratio is greater. Non-dimensional natural frequencies of Clamped-Clamped FGPM plates (C–C–C–C) are greater compared to Clamped-Free FGPM (C–F–C–F) and Simply Supported Free FGPM (S–F–S–F).
12 August 2026 in The European Physical Journal Special Topics
In signal analysis, information content reflects the extent to which uncertainty is reduced and decision-making confidence is improved. Nevertheless, measuring information content from noisy time series with mathematical modelling remains challenging. In nonlinear time-series analysis, recurrence plots and their quantification analysis are increasingly used to study vibro-acoustic and oscillatory systems. Joint recurrence quantification analysis (JRQA) allows the computation of features based on the dynamics of multiple, potentially physically distinct, phase spaces. Here, a two-coupled model of analytical saline oscillators is used, which offers controllable coupling parameters and rich synchronisation behaviour. However, JRQA has not been used to analyse coupling direction and phase synchronisation in analytical saline oscillator models, nor has diagonal-wise joint recurrence-based power spectral entropy been explored for measuring common information content in signals. Our results show that the joint recurrence rate increases with the coupling strength parameter and synchronisation level, while the power spectral entropy, as a measure of information content, decreases in the noise-free case. Further, by progressively decreasing the signal-to-noise ratio from 20 dB to –20 dB through the addition of Gaussian white noise, the information content is increased. Our findings demonstrate that joint recurrence-based power spectral entropy can provide an effective measure of common information embedded in time-series data. By quantifying the shared recurrent dynamics between signals, joint recurrence quantification analysis may be useful for multi-sensor signal analysis, particularly when the objective is to identify weak dynamical signatures that are not clearly observable in individual sensors. This approach could improve decision-making confidence in the future by enabling the detection of micro-vibrations buried in noise using low-cost, off-the-shelf sensors.