Integrating AI in Structural Health Monitoring (SHM): A Systematic Review on Ad- vances, Challenges, and Future Directions
A Generalized Physically Based Zig-Zag Theory for Efficient Analysis of Multilayered Composite Structures
Graded Multiscale Topology Optimization with Periodic Boundary Conditions and Subdomain Coupling
Reinforced Concrete Hulls for Wave Energy Converters a Structural Feasibility and Economic Viability of ISWEC Designs
Satellite Based Estimation of Forest Biomass for Structural Resource Planning Using Gaussian Processes and Sentinel-2 Imagery
Study on Strength Properties of Lightweight Expanded Clay Aggregate Concrete
A Step By Step Illustrative Procedure to Perform Isogeometric Analysis and Find the Nodal Displacements for a Two Dimensional Plate Structure
Lateral - Torsional Buckling of Various Steel Trusses
Comparative Study on Methodology of Neo-Deterministic Seismic Hazard Analysis Over DSHA and PSHA
A Step by Step Procedure to Perform Isogeometric Analysis of Beam and Bar Problems in Civil Engineering Including Sizing Optimisation of a Beam
Investigation on the Properties of Non Conventional Bricks
Analysis on Strength and Fly Ash Effect of Roller Compacted Concrete Pavement using M-Sand
Investigation on Pozzolanic Effect of Mineral Admixtures in Roller Compacted Concrete Pavement
Effect of Symmetrical Floor Plan Shapes with Re-Entrant Corners on Seismic Behavior of RC Buildings
Effect of Relative Stiffness of Beam and Column on the Shear Lag Phenomenon in Tubular Buildings
Integration of Artificial Intelligence into SHM is a tectonic shift in infrastructure management, bringing unprecedented advances in safety, reliability, and operational efficiency. AI-driven SHM utilizes machine learning, deep learning, and data-driven analytics methodologies for new ways of detecting, assessing, and predicting the performance of structures and structural anomalies. Some recent developments are demonstrating how AI will process a large amount of sensor data in near real time, extracting and analyzing with unprecedented accuracy from damage detection to anomaly iden- tification through to predictive maintenance. Regarding SHM, machine-learning algorithms like support vector machines and random forests are commonly employed to perform feature extraction and classifica- tion tasks. Deep learning techniques, including CNN and RNN methods, are particularly effective when addressing complex high-dimensional data related to various types of structural systems, from bridges to high-rise buildings and even offshore platforms. Despite the promise, challenges persist. These include the availability of high-quality labeled data, since AI models depend hugely on training data for performance, and interpretability problems of these models, which may act as a barrier for their adoption in critical infrastructure monitoring. Besides, integrating the solutions with existing SHM frameworks also requires seamless communication among different hard- ware and software systems that often necessitates major technological upgrades. Standardized dataset development, interpretable AI algorithms, and robust integration protocols call for future research. In that way, it will be very useful if the collaboration of researchers, industry practition- ers, and policymakers intensifies to bring the full potential of AI. The continuous maturation of AI is sure to continue its application in SHM and thereby help develop a host of safe, resilient, and sustainable in- frastructure systems as one of the key future developments in structural engineering.
Multilayered and sandwich structures are extensively utilized in aerospace, automotive, and marine applications due to their superior stiffness-to-weight ratios. However, their accurate modelling remains computationally challenging. Classical theories such as FSDT and ESL lack the capability to capture layer wise effects, while discrete-layer and 3D finite element models are often computationally prohibitive. This paper introduces ZZA_GEN, a generalized physically based zig-zag theory that overcomes these challenges by allowing user defined through-thickness displacement representations and abandoning rigid coefficient roles. Derived from the Zig-Zag Adaptive (ZZA) theory, ZZA_GEN achieves high fidelity in stress and displacement prediction with minimal computational overhead. It retains the simplicity of a C⁰ finite element formulation while maintaining accuracy equivalent to higher-order theories. Validation against classical benchmarks, including elastostatic and dynamic problems, shows excellent agreement with 3D FEA results. The theory's adaptability and integration potential with commercial solvers make it highly suitable for industrial applications.
This study introduces a comprehensive computational framework for graded multiscale topology optimization (MTO) that integrates periodic boundary conditions (PBCs), localized homogenization, and inter-subdomain coupling to enable advanced structural design. The design domain is partitioned into multiple independent yet interactively coupled subregions, each optimized to host distinct, spatially-tailored microstructures characterized by homogenized elasticity tensors. A tunable coupling parameter wc is incorporated to control stiffness continuity across subdomain interfaces, ensuring mechanical coherence throughout the structure. To address practical fabrication constraints, a projection-based filtering scheme is implemented, enhancing manufacturability without compromising structural performance. Numerical results validate the framework’s superiority over conventional SIMP and de-homogenization techniques, exhibiting marked enhancements in compliance reduction, material utilization, and stress uniformity.
Wave energy presents a promising path for renewable electricity, yet commercial adoption remains limited due to high capital costs of wave energy converters (WECs). This study investigates the structural and economic feasibility of using reinforced concrete instead of steel for the hull of the Inertial Sea Wave Energy Converter (ISWEC). A concrete-based hull is proposed and evaluated using linear and nonlinear Finite Element Analysis (FEA), with environmental loads assessed via Computational Fluid Dynamics (CFD). Key performance indicators such as cracking, reinforcement behavior, and serviceability were analysed. Cost analysis indicates potential savings in capital expenditure and lifecycle maintenance. Results support the viability of concrete as an alternative material in offshore WEC structures, paving the way for more sustainable and cost-effective wave energy systems.
This study presents a replicable, cost-efficient method for estimating forest biomass critical for sustainable structural material sourcing using Sentinel-2 satellite imagery and Gaussian Process Regression. A simplified inventory method, coupled with spectral data in the visible to mid-infrared bands, enables accurate biomass quantification across diverse forest structures in Mediterranean climates. Compared to traditional LiDAR-based techniques, this approach offers faster, lower-cost deployment without significant trade-off in accuracy, making it suitable for applications in construction timber forecasting, infrastructure planning, and environmental assessments. The method has been validated across several Mediterranean forest types and is packaged in a freely accessible programming tool for direct integration into engineering planning workflows.
