Abdellah Riyahi

Abdellah Riyahi

Auto-generated profile, based on publicly available information

Abdellah Riyahi

University of Hassan II Casablanca

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 March 2026 in Journal of the Civil Engineering Forum
Verified
Journal Article
Open access
ARAbdellah Riyahi
Abdellah Riyahi
MMMohammed Mestari
Mohammed Mestari
BBBouchra BOUIHI
Bouchra BOUIHI
Structural Health Monitoring (SHM) is crucial for maintaining the sustainability and safety of civil infrastructure. The Z24 Bridge in Switzerland remains one of the benchmark datasets used to validate vibration-based damage detection methods. Traditional approaches based exclusively on modal parameters are frequently limited by data scarcity and environmental variability. Recent advances in artificial intelligence have enabled data-driven neural networks to learn discriminative features directly from raw measurements. Meanwhile, hybrid methods such as Physics-Informed Neural Networks (PINNs) incorporate governing physical laws into the learning process. This study presents a comparative analysis of three successive artificial Neural Network models (NN V1–V3) and One Physics-Informed Neural Network (PINN V1), all applied to the Z24 Bridge dataset. The NN models progressively improve in depth, optimization strategy, and regularization, achieving ≈97.7% validation accuracy and a macro AUC ≈1.00 with NN V3. However, they remain completely dependent on the quality and quantity of training data. In contrast, the PINN incorporates the differential equation of a damped oscillator into its loss function, balancing a data-driven term with a physics-based residual. This approach enables more stable learning with limited labeled data and ensures consistency with structural dynamics. Experimental results highlight the trade-off between accuracy and robustness: while NN V3 yields the highest predictive performance (≈97.7% validation accuracy, macro AUC ≈1.00),PINN V1 achieves slightly lower accuracy (≈92%) but offers improved stability and interpretability. This dual perspective demonstrates that hybrid physics-informed models provide a more reliable basis for decision-making in SHM. The findings underscore the potentialof combining machine learning with physical knowledge, paving the way for future developments such as hybrid PINNs (HPINNs), multi-sensor integration, and high-performance computing deployment.

5 May 2025 in International Journal of Innovation and Modern Applied Science
Verified
Journal Article
Closed access
ARAbdellah Riyahi
Abdellah Riyahi
Although early damage detection in civil engineering structures is difficult, because damage is not directly observed, yet it is key for structural health monitoring and predictive maintenance. It must be inferred from partial, noisy, and indirect vibration measurements. In this paper, a theoretical framework in which early damage detection is formulated as an improperly posed inverse problem has been developed. Uniqueness, stability, and practical recoverability are limited by weak damage, sparse sensing, environmental variability, and poor conditioning. Addressing these challenges requires recasting damage inference as a regularized problem, balancing data consistency and prior structural knowledge. The interpretation of smoothness, sparsity, bounded variation, physics-based, and probabilistic regularization principles is conducted in relation to different damage morphologies. Also, the paper distinguishes detectable, identifiable, estimable, and practically recoverable damage, as well as discussing the relevance of this framework for data-driven, physics-informed, and maintenance-oriented Structural Health Monitoring.