Development of Advanced Machine Learning Models for Multiple Leak Detection in Offshore Gas Pipelines Under Single- and Multiphase Flow Conditions

This study investigates the application of advanced machine learning (ML) models for single and multiple leak detection in offshore gas pipelines under single- and multiphase flow conditions. The primary goal is to enhance the accuracy and robustness of leak detection systems, which are critical for maintaining the integrity and safety of pipeline networks. The novelty of this study lies in the development and evaluation of a comprehensive multiple leak detection model using advanced stacked ML techniques, addressing gaps in existing research by providing a robust solution for complex leak scenarios. Using representative data generated from OLGA software, we evaluated six individual ML algorithms for single leak detection and localization, including Artificial Neural Network (ANN), Support Vector Regression (SVR), k-Nearest Neighbors Regression (KNNR), Decision Tree Regression (DTR), Extreme Gradient Boosting (XGBoost), and Random Forest Regression (RFR). For multiple leak detection, we developed and compared three stacked models: Voting Regressor 1 (VR1), Voting Regressor 2 (VR2), and Stacking Regressor (SR). The base learners for these stacked models were KNNR and RFR for VR1, RFR, GB, XGB, and LGBM for VR2, and RFR, GB, XGB, LGBM, and MLP for SR. The results show that for the single leak detection model, the RFR model outperforms other models with an absolute relative error of less than 1.22% with noised data for single-, and multi-phase flow conditions. In addition, the SR model, which combines multiple base learners with a meta-learner (RidgeCV), significantly outperforms the other models. Specifically, the SR model achieved R2 scores higher than 0.96 compared to less than 0.92 for VR1 and VR2, demonstrating its superior ability to capture diverse patterns and relationships in the data. These findings highlight the potential of advanced ML techniques to provide reliable and accurate leak detection solutions in complex pipeline systems.