Abstract This study explores the application of artificial intelligence (AI) neural networks to predict the service life (crack initiation stage) of tensile armor wires in flexible pipes subjected to stress corrosion cracking under carbon dioxide-rich environments (SCC-CO2). Conventional methods for assessing the lifespan of tensile armor focus on experimental data and physical models, often resulting in time-consuming and costly evaluations. By integrating neural networks with our existing theoretical model, we aim to create a more efficient, data-driven approach that can predict the onset of critical cracking in tensile armor wires under specific environmental conditions, enhancing early detection and maintenance strategies for flexible pipeline systems. The neural network model was trained on a comprehensive dataset exported based on an existing theoretical model, encompassing variables like utilization factor (UF), CO2 fugacity, temperature, and material degradation patterns. Performance metrics demonstrate the AI model’s capability to capture complex relationships and dependencies, offering high efficiency in service life predictions at the crack threshold stage. This approach has potential benefits for risk assessment and maintenance planning in offshore and subsea engineering. The integration of neural networks thus represents a significant advancement in predicting structural integrity under harsh environmental conditions, ultimately contributing to safer, more sustainable operations in industries reliant on flexible pipes.
Ye et al. (Sun,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: