Search

A deep learning method is proposed to identify damages in jacket offshore platforms, which integrates the random decrement technique (RDT) and Transformer neural network. RDT is utilized to process the noisy strain response signals from various measurement points on the platform under random excitation, and the Transformer network is used to localize the damages and assess their severities. Numerical simulations and experimental studies are used to demonstrate that the proposed method is effective. The numerical model applied in this study is a jacket platform, including single and multiple damage cases in the entire section and partial section (two-fifths of the entire section) with minor severities (<5%), and different noise levels are considered. Additionally, the effect of measurement positions on identification accuracy of minor and major damage is discussed. Furthermore, the proposed method is validated by experimental studies using a jacket platform model fixed on a shaking table, where different severities of single and multiple damages are considered. Results of the numerical and experimental studies demonstrate that the proposed RDT-Transformer approach can accurately determine the location of structural damage and identify its severity. • Combine the RDT with Transformer for jacket offshore platforms damage identification under random excitation. • RDT can significantly improve the effectiveness of damage identification of Transformer under noise condition. • Validate the RDT-Transformer method through the numerical and experimental studies of jacket offshore platform. [ABSTRACT FROM AUTHOR]

Machine learning methods for damage detection of thermoplastic composite pipes (TCPs) under noise conditions are presented, which combine the random decrement technique (RDT) with random forest (RF) and long and short-term memory (LSTM) networks. RDT is applied first to process the measured noisy strain response data of the TCP under random excitation. Then the RF or LSTM method is used to conduct the damage localization and severity estimation of the pipe. The applicability of the proposed methods is verified by means of numerical and experimental studies. The numerical example consists of a TCP subjected to internal pressure and random wave excitation considering several noise levels. The damages are simulated as circular holes on different layers of the pipe and varying severity, characterized by their radii and depths. The damage detection is carried out using RDT-RF and RDT-LSTM methods. The experimental studies consist of laboratory tests of a TCP model using Fiber Bragg Grating sensors. The damage cases, simulated as cracks with different lengths and depths on the reinforcement layer, are discussed. Both the numerical simulation and experimental tests show that the proposed RDT-RF and RDT-LSTM methods have an excellent performance in damage detection of TCPs. • Propose machine learning methods, RF and LSTM combined with RDT for damage detection of thermoplastic composite pipes. • RDT can remarkably improve the capability of damage detection of RF and LSTM methods under noise conditions. • Validate RDT-RF and RDT-LSTM through the numerical simulation and experimental test of thermoplastic composite pipes. [ABSTRACT FROM AUTHOR]