Integration of Machine Learning and Finite Element Analysis for Assessing the Conditions of Cure-in-Place Pipe Liner Materials for Damaged Pipe Rehabilitation

Underground pipeline systems are vital in modern infrastructure, serving as essential components for water, gas, and other utilities. Nevertheless, pipelines frequently face corrosive environments due to the transport of aggressive substances containing water, oxygen, and sulfides, as well as external factors like soil and moisture. Corrosion holes form when the protective layer degrades, leading to localized reactions that create and deepen pits over time. The structural integrity of pipelines can be compromised in this situation, leading to significant challenges to their long-term reliability and safety. Cure-in-place pipe (CIPP) lining is a widely adopted rehabilitation technique that restores pipeline integrity without extensive excavation by inserting and curing a flexible liner within the damaged pipe. Despite its advantages, there is limited research on the condition assessment of the pipeline associated with CIPP liners. It is important to evaluate the high-risk area and the rehabilitation effect of the CIPP liners on pipeline integrity. This study develops a machine learning (ML)-based condition assessment model using finite element analysis (FEA) data of CIPP liner materials to evaluate the structural integrity of pipelines affected by corrosion holes. The FEA data provides detailed insights into stress distribution and mechanical properties under various loading conditions. The machine learning model has the capability to predict high-risk areas and potential material failure with sufficient data provided by the FEA simulation data. The results revealed that the debonding risk regions are typically formed around the edges of the corrosion hole, and the risk level of the area closer to the length direction is higher, while the areas prone to material failure are found to coincide almost entirely with the corrosion hole itself, with higher risk near the center of the hole. These findings contribute to optimizing CIPP installations and ensuring the long-term reliability of rehabilitated pipelines.