Open research questions in Structural Health Monitoring Techniques

What the literature leaves open

• The difficulty in quantifying prestress loss in P-FRCM plates. The challenge of minimizing contact resistance effects in the measurement of resistance across the P-FRCM plates. The need to balance strength and ductility in the design of P-FRCM plates.

  Self-sensing based full-range prestress-loss quantification for P-FRCM plates · 2026 · DOI

• No explicit limitations stated in the text, - Small sample size of 12 test groups with 3 replicates, - Limited to uniaxial tension tests

  Self-sensing based full-range prestress-loss quantification for P-FRCM plates · 2026 · DOI

• Although various methods have been introduced in the literature, developing robust and reliable structural health monitoring (SHM) procedures remains an open research challenge.

  A Deep Learning Approach for Autonomous Compression Damage Identification in Fiber-Reinforced Concrete Using Piezoelectric Lead Zirconate Titanate Transducers · 2024 · DOI

• The objective of this study is to pinpoint areas where research is lacking in the existing literature on the environmental factors that impact the displacement of bridges, along with the techniques and technology used to monitor these structures.

  Structural Health Monitoring of Bridges under the Influence of Natural Environmental Factors and Geomatic Technologies: A Literature Review and Bibliometric Analysis · 2024 · DOI

• The approach relies on simulated data, which may not accurately represent real-world scenarios. The study focuses on a specific type of structure (plate-like). The number of damage cases studied is limited.

  Leveraging finite element model with transfer learning for data-driven-based structural damage identification · 2026 · DOI

• To apply the proposed framework to other types of structures. To investigate the use of other machine learning models and transfer learning methods. To study the effect of different damage types and severity levels on the damage identification performance.

  Leveraging finite element model with transfer learning for data-driven-based structural damage identification · 2026 · DOI

• The study identifies the challenge of variability in human judgment in manual inspection methods for post-earthquake damage assessment. The analysis highlights the challenge of developing effective AI-driven approaches for post-earthquake damage assessment. The study notes the challenge of balancing the need for rapid assessment with the need for accurate and reliable results.

  AI-Driven Post-Earthquake Rapid Structural Damage Assessment: A Scoping Review of Component-Level and Holistic Approaches · 2026 · DOI

• The study suggests that future research should focus on the development of more effective AI-driven approaches for post-earthquake damage assessment. The analysis highlights the need for more research on the application of AI and DL models for post-earthquake damage assessment. The study recommends that future research should examine the potential of AI-driven approaches for improving disaster response and recovery efforts.

  AI-Driven Post-Earthquake Rapid Structural Damage Assessment: A Scoping Review of Component-Level and Holistic Approaches · 2026 · DOI

• The study aims to address the lack of comprehensive analysis of the temperature-induced effects on double-ribbed plate girders in cable-stayed bridges.

  HEALTH‑MONITORING‑BASED INVESTIGATION OF THERMAL EFFECTS ON DOUBLE‑RIBBED PLATE GIRDERS OF CABLE‑STAYED BRIDGES · 2026 · DOI

• 1. Multi-site generalization. Extend deployments to multiple wharves and environmental regimes to test cross-asset transferability and robustness to distribution shift. 2. Compute-efficient twins. Develop reduced-order/ surrogate DTs with event-triggered updating and edge execution to retain low latency while cutting energy and cost. 3. Interpretable, uncertainty-aware ML. Integrate calibrated probabilities, conformal prediction, and DT-consistency checks to produce explanations and reliable risk bounds for operators and regulators. 4. Standardized benchmarks. Release harmonized protocols (EOV handling, chronological/LOP splits, Dev-only thresholding) and shared labeled datasets to enable reproducible, apples-to-apples comparisons across SHM frameworks. 5. Sensing durability and autonomy. Investigate energy harvesting, redundancy, and automated quality control (noise-floor/PSD drift/coherence tests) for long-term offshore operation. In summary, the proposed framework delivers high diagnostic accuracy (F1 0.95; AUC 0.98), rapid response (≈45 s latency), and tight DT–field agreement (MAC ≥ 0.92; frequency deviation within ±3%), offering a scalable path to next-generation, predictive condition monitoring for critical maritime infrastructure.

  Field-Validated Digital Twin–Enabled Structural Health Monitoring for Offshore High-Pile Wharves · 2026 · DOI

• Further validation of LiDAR under field and dynamic loading conditions is needed. Evaluation of LiDAR for measuring other structural health parameters, such as strain and acceleration.

  Full field LiDAR-based measurement of deflection and support condition effects in a scaled bridge model: comparison with traditional point sensors · 2026 · DOI

• Traditional SHM methods may not capture the full extent of structural behavior. There is a need for noncontact techniques that can capture changes in deformation patterns across support-condition cases.

  Full field LiDAR-based measurement of deflection and support condition effects in a scaled bridge model: comparison with traditional point sensors · 2026 · DOI

• To apply the proposed methodology to real-world scenarios. To investigate the effect of noise contamination on the method. To develop a more robust approach that can handle complex structures and various types of damage.

  A Hybrid Experimental-Numerical Approach for Structural Health Monitoring Using Sem and Measured FRF · 2026 · DOI

• Traditional methods for damage detection have limitations, such as requiring access to all parts of the structure. There is a need for a new approach that can detect damage with high accuracy and prioritize maintenance.

  A Hybrid Experimental-Numerical Approach for Structural Health Monitoring Using Sem and Measured FRF · 2026 · DOI

• Post-reinforcement numerical analysis and follow-up monitoring to verify the effectiveness of the proposed measures, - Detailed design verification of prestressed anchoring with high-strength metal anchor points

  Mayfly–Deep Learning Fusion for High-Dimensional Parameter Identification and Reinforcement of Historical Buildings · 2026 · DOI

• High-dimensional parameter identification under sparse, noise-contaminated modal data can reduce robustness and lead to prohibitive computational cost - The need for a more efficient and accurate method for structural health assessment of historic buildings

  Mayfly–Deep Learning Fusion for High-Dimensional Parameter Identification and Reinforcement of Historical Buildings · 2026 · DOI

• To apply the proposed method to more complex systems. To investigate the effect of noise on the proposed method. To compare the proposed method with other methods.

  Nonlinear damping identification using an extended stabilized layers method · 2026 · DOI

• The traditional stabilized layers method does not account for time-varying stiffness. The method proposed by Lisitano and Bonisoli (2021) has limitations in identifying nonlinear damping.

  Nonlinear damping identification using an extended stabilized layers method · 2026 · DOI

• Further research could investigate the application of the proposed framework to other structural systems, - The development of more advanced optimization algorithms could potentially improve the efficiency of the method

  Multi-Criterion Mode Selection in Stochastic Subspace Identification (SSI): Enhancing Reliability in Noisy Environments · 2026 · DOI

• The classical Stochastic Subspace Identification method is often insufficient in high-noise environments. The method relies solely on frequency and damping stability, which may not be enough for reliable modal identification.

  Multi-Criterion Mode Selection in Stochastic Subspace Identification (SSI): Enhancing Reliability in Noisy Environments · 2026 · DOI

• Field validation results will be published in a subsequent version upon completion of Phase 5. Artificial excitation (impact hammer) will be integrated into the system. Multi-modal vibration analysis will be developed.

  EVM-Insight: A Low-Cost Eulerian Video Magnification System for Structural Health Monitoring · 2026 · DOI

• Traditional SHM approaches are expensive and impractical for ageing infrastructure. There is a need for a low-cost, non-invasive structural health monitoring system.

  EVM-Insight: A Low-Cost Eulerian Video Magnification System for Structural Health Monitoring · 2026 · DOI

• Conventional single-channel independent modeling strategies neglect spatial correlations and cross-modal causal associations. Existing methods fail to exploit the spatial correlations and causal relationships in multi-source heterogeneous monitoring data.

  Cross-Modal Graph Attention for Bridge SHM Data Imputation · 2026 · DOI

• The structural engineering of vertical towers presents a set of analytical challenges distinct from other infrastructure categories. There is a need for a comprehensive framework for structural integrity assessment and safety governance of vertical tower systems.

  TOWER-CORE: Structural Integrity Assessment and Safety Governance for Vertical Tower Systems · 2026 · DOI

• DIRECTIONS 10.1 Current Limitations Tower-Core v1.0.0 provides a validated framework for the continuous structural integrity assessment of vertical tower systems within several boundary conditions that define the current scope. First, the DFMM modal identification algorithm assumes stationary ambient vibration excitation from broadband wind loading. During severe storm events, the non-stationarity of the excitation may degrade the quality of frequency identification, and adaptive windowing techniques for nonstationary conditions are not yet implemented in the current module. Second, the Palmgren–Miner fatigue damage rule implemented in the SJFAM is a linear accumulation model that does not explicitly account for load sequence effects — the accelerated crack propagation that can occur when high-amplitude load cycles follow low-amplitude cycles and retardation effects that can occur in the reversed order. For tower structural details subject to rare high-amplitude loading events (extreme storms), these sequence effects may be relevant. Third, the spatial resolution of the damage TOWER-SAFETY-01 | Version 1.0.0 | MIT License | DOI: 10.5281/zenodo.20394041 github.com/gitdeeper12/TOWER-CORE TOWER-CORE — Structural Integrity Assessment and Safety Governance for Vertical Tower Systems Baladi, S. (2026) assessment is limited by the sensor spacing: the current sensor complement provides damage assessment at instrumented detail locations, with extrapolation to uninstrumented locations via the mode shape-based interpolation functions. Cracks or degradation at locations between sensors may not be detected until they have grown to a size that affects the measured modal parameters. 10.2 Future Research Directions Four priority extensions are planned for subsequent Tower-Core development. First, non-stationary modal identification: extension of the DFMM frequency tracking algorithm to non-stationary conditions through the implementation of the Short-Time SSI algorithm, which processes windowed segments of the acceleration time series and tracks the evolution of modal parameters through the loading event. This extension is particularly relevant for the identification of natural frequency shifts during severe storm events — precisely the conditions under which the overturning stability margin is most critical. Second, probabilistic fatigue life prediction: integration of a probabilistic framework for fatigue life prediction based on Monte Carlo simulation of the S-N curve scatter and Miner sum variability, enabling the TSII to incorporate calibrated uncertainty bounds that support risk-based maintenance decision making.

  TOWER-CORE: Structural Integrity Assessment and Safety Governance for Vertical Tower Systems · 2026 · DOI

Most-cited papers in Structural Health Monitoring Techniques

• Integrated structural health monitoring in bridge engineering · Automation in Construction · 2022 · 347 citations
• Structural health monitoring of civil engineering structures by using the internet of things: A review · Journal of Building Engineering · 2022 · 339 citations
• Three decades of statistical pattern recognition paradigm for SHM of bridges · Structural Health Monitoring · 2022 · 222 citations
• CNN and Convolutional Autoencoder (CAE) based real-time sensor fault detection, localization, and correction · Mechanical Systems and Signal Processing · 2022 · 213 citations
• Machine learning-based seismic response and performance assessment of reinforced concrete buildings · Archives of Civil and Mechanical Engineering · 2023 · 199 citations
• Vision-based real-time structural vibration measurement through deep-learning-based detection and tracking methods · Engineering Structures · 2023 · 199 citations
• Comparison of Visual Inspection and Structural-Health Monitoring As Bridge Condition Assessment Methods · Journal of Performance of Constructed Facilities · 2015 · 192 citations
• State-of-the-art non-destructive methods for diagnostic testing of building structures – anticipated development trends · Archives of Civil and Mechanical Engineering · 2010 · 172 citations
• Sensing Techniques for Structural Health Monitoring: A State-of-the-Art Review on Performance Criteria and New-Generation Technologies · Sensors · 2025 · 154 citations
• Optimization-based stacked machine-learning method for seismic probability and risk assessment of reinforced concrete shear walls · Expert Systems with Applications · 2024 · 146 citations

Most recent work

• Real-time Vehicle-Induced Response Identification via crowdsourced labeling for high-frequency unlabeled sensor data · Engineering Applications of Artificial Intelligence · 2026
• PhysScaleFormer: A multiscale physics-enhanced deep learning framework for real-time dynamic response prediction in vibratory truss structures · Engineering Applications of Artificial Intelligence · 2026
• An efficient reduced-order uncertainty quantification framework for high-confidence interval prediction of aircraft dynamic loads · Aerospace Science and Technology · 2026
• Dynamic monitoring of bridges subjected to freezing-induced variability by kernel-enhanced deep anomaly detection · Engineering Structures · 2026
• Accurate online reconstruction algorithm for UAV propeller blade deformation by FBG sensing · Aerospace Science and Technology · 2026
• Leveraging finite element model with transfer learning for data-driven-based structural damage identification · Journal of Vibration Engineering & Technologies · 2026
• Deep learning-based realtime multiload response prediction and inverse analysis of offshore bridges · Engineering Structures · 2026
• TOWER-CORE: Structural Integrity Assessment and Safety Governance for Vertical Tower Systems · Zenodo (CERN European Organization for Nuclear Research) · 2026
• Synthetic Data Generation for the Prototyping of Bridge Damage Detection Algorithms · Infrastructures · 2026
• Understanding the spectral acceleration amplification factors of reinforced concrete frames through explainable artificial intelligence · Computers & Structures · 2026

Related topics in Engineering