Structural Damage Early Warning Method of Quayside Container Crane Based on Fuzzy Entropy Ratio Variation Deviation

Area of Science:

• Structural Engineering
• Signal Processing
• Machine Learning

Background:

• Real-time structural health monitoring (SHM) and early warning systems are critical for safety and maintenance.
• Quayside container cranes (QCCs) require robust methods for detecting abnormal states, especially with incomplete damage data.

Purpose of the Study:

• To develop a timely early warning method for structural abnormal states in QCCs using fuzzy entropy ratio variation deviation (FERVD).
• To improve structural health assessment and safety maintenance for QCCs.

Main Methods:

• Applying dual-tree complex wavelet transform (DTCWT) to monitoring data for adaptive frequency band decomposition.
• Extracting response signal features using fuzzy entropy (FE) to construct the FERVD indicator.
• Establishing dynamic thresholds for early warning based on the FERVD indicator.
• Developing a finite element model (FEM) for QCCs to simulate various damage scenarios.

Main Results:

• The FERVD indicator effectively differentiates between healthy structural states and various damage conditions.
• Numerical experiments and case studies confirmed the method's ability to provide timely warnings for different damage states.
• FERVD values remained within threshold ranges for healthy structures, validating its applicability in QCC SHM.

Conclusions:

• The proposed FERVD-based method offers a timely and effective solution for structural abnormal state early warning in QCCs.
• The method demonstrates robustness and applicability in real-world structural health monitoring scenarios.
• This approach enhances safety and maintenance strategies for critical infrastructure like QCCs.