arXiv subjects

Basuraj Bhowmik

Publications and source records attributed to Basuraj Bhowmik.

6 recordsLinked to original sources

arXiv2026

SPECTRA: A Physics-informed Digital Twin for Real-time Structural Anomaly Inference under Operational Variability

Structural health monitoring is moving from damage detection alone towards real-time decision support for ageing and safety-critical infrastructure. This shift requires monitoring methods that can separate true structural change from benign environmental and operational variability, while remaining interpretable to engineers. This paper presents SPECTRA, a physics-informed eigen-compressed digital twin framework for real-time structural anomaly inference. The framework combines a healthy structural twin, eigen-compressed dynamic representation, full-order twin innovation, residual-augmented spectral features, kernel principal component analysis, and a persistent decision rule. The central idea is that structural anomalies are inferred not from statistical features alone, but from disagreement between the measured response and a physics-informed healthy twin. The method is assessed through seven numerical benchmarks: smooth Duffing-type nonlinear drift, sudden stiffness loss, gradual stiffness degradation, bilinear breathing stiffness, environmental and operational variability-confounded local damage, local damping loss, and an operational-only negative-control case. The results show that SPECTRA detects abrupt, gradual, nonlinear and damping-related damage mechanisms, while avoiding persistent false alarms under operational variability alone. Across the accepted benchmark suite, the persistent decision rule gives zero pre-damage persistent false alarms, finite detection delay in damage cases, and zero persistent alarms in the no-damage negative-control case. The framework provides a reproducible route for testing physics-informed anomaly inference before deployment in infrastructure digital twins.

cs.CE↗

arXiv2026

A Physics-Aware Variational Graph Autoencoder for Joint Modal Identification with Uncertainty Quantification

Reliable modal identification from output-only vibration data remains a challenging problem under measurement noise, sparse sensing, and structural variability. These challenges intensify when global modal quantities and spatially distributed mode shapes must be estimated jointly from frequency-domain data. This work presents a physics-aware variational graph autoencoder, termed UResVGAE, for joint modal identification with uncertainty quantification from power spectral density (PSD) representations of truss structures. The framework represents each structure as a graph in which node attributes encode PSD and geometric information, while edges capture structural connectivity. A residual GraphSAGE-based encoder, attention-driven graph pooling, and a variational latent representation are combined to learn both graph-level and node-level modal information within a single, unified formulation. Natural frequencies and damping ratios are predicted through evidential regression, and full-field mode shapes are reconstructed through a dedicated node-level decoder that fuses global latent information with local graph features. Physical consistency is promoted via mode-shape reconstruction and orthogonality regularisation. The framework is assessed on numerically generated truss populations under varying signal-to-noise ratios and sensor availability. Results demonstrate accurate prediction of natural frequencies, damping ratios, and mode shapes, with high modal assurance criterion values and stable performance under noisy and sparse sensing conditions. Reliability analysis indicates that the predictive uncertainty is broadly consistent with empirical coverage. The proposed framework offers a coherent and physically grounded graph-based route for joint modal identification with calibrated uncertainty from frequency-domain structural response data.

cs.CE↗

arXiv2025

Investigating dimensionally-reduced highly-damped systems with multivariate variational mode decomposition: An experimental approach

Structural health monitoring (SHM) is an essential engineering field aimed at ensuring the safety and reliability of civil infrastructures. This study proposes a methodology using multivariate variational mode decomposition (MVMD) for damage detection and modal identification. MVMD decomposes multi-sensor vibration responses into intrinsic modal components, facilitating the extraction of natural frequencies and damping ratios by analyzing amplitude decay in the identified modes. Mode shapes are determined through peak-normalization of Fourier spectra corresponding to each mode. The methodology is further applied to detect damage by identifying changes in the extracted modal parameters and spatial features of the structure. The proposed approach enables damage detection by tracking variations in modal parameters and spatial structural characteristics. To validate its efficacy, the methodology is applied to a benchmark eight-degree-of-freedom (8-DOF) system from Los Alamos National Laboratory (LANL), demonstrating its robustness in identifying structural damage under non-stationary excitation and narrowband frequency content. The results confirm that MVMD provides a reliable and adaptable framework for modal analysis and damage assessment in complex infrastructure systems, addressing key challenges such as environmental variability and practical scenarios.

stat.AP↗

arXiv2024

When Fire Attacks: How does Concrete Stand up to Heat ?

Fire is a process that generates both light and heat, posing a significant threat to life and infrastructure. Buildings and structures are neither inherently susceptible to fire nor completely fire-resistant; their vulnerability largely depends on the specific causes of the fire, which can stem from natural events or human-induced hazards. High temperatures in structures can lead to severe health risks for those directly affected, discomfort due to smoke, and compromised safety if the structure fails to meet safety standards. Elevated temperatures can also cause significant structural damage, becoming the primary cause of casualties, economic losses, and material damage. This study aims to investigate the thermal and structural behavior of concrete beams when exposed to extreme fire conditions. It examines the effects of different temperatures on plain and reinforced concrete (PCC and RCC, respectively) using finite element method (FEM) simulations. Additionally, the study explores the performance of various concrete grades under severe conditions. The analysis reveals that higher-grade concrete exhibits greater displacement, crack width, stress, and strain but has lower thermal conductivity compared to lower-grade concrete. These elevated temperatures can induce severe stresses in the concrete, leading to expansion, spalling, and the potential failure of the structure. Reinforced concrete, on the other hand, shows lower stress concentrations and minimal strain up to 250{\deg}C. These findings contribute to the existing knowledge and support the development of improved fire safety regulations and performance-based design methodologies.

cs.CE↗

arXiv2023

CNN-Based Structural Damage Detection using Time-Series Sensor Data

Structural Health Monitoring (SHM) is vital for evaluating structural condition, aiming to detect damage through sensor data analysis. It aligns with predictive maintenance in modern industry, minimizing downtime and costs by addressing potential structural issues. Various machine learning techniques have been used to extract valuable information from vibration data, often relying on prior structural knowledge. This research introduces an innovative approach to structural damage detection, utilizing a new Convolutional Neural Network (CNN) algorithm. In order to extract deep spatial features from time series data, CNNs are taught to recognize long-term temporal connections. This methodology combines spatial and temporal features, enhancing discrimination capabilities when compared to methods solely reliant on deep spatial features. Time series data are divided into two categories using the proposed neural network: undamaged and damaged. To validate its efficacy, the method's accuracy was tested using a benchmark dataset derived from a three-floor structure at Los Alamos National Laboratory (LANL). The outcomes show that the new CNN algorithm is very accurate in spotting structural degradation in the examined structure.

cs.LG↗

arXiv2023

Mastering Complex Modes: A New Method for Real-Time Modal Identification of Vibrating Systems

A novel algorithm for real-time modal identification in linear vibrating systems with complex modes is introduced, utilizing a combination of first order eigen-perturbation and second order separation techniques. In practical settings, structures with complex modes are frequently encountered and their presence often poses a challenge in accurately estimating the source signal in real-time. The proposed methodology addresses this issue by incorporating the right angle phase shift of the response in the sensor output and updating the second order statistics of the complex response through first order eigen-perturbation. Empirical evidence of the efficacy of the technique is demonstrated through numerical case studies and validation using various numerically modeled systems, as well as a standard ASCE-SHM benchmark problem with complex modes, highlighting the capability of the proposed method to achieve precise real-time modal property identification and online source separation with a minimal number of initially required batch data.

eess.SY↗