Identification of flow cut-off pattern and prediction of cut-off time of gas–liquid two-phase flow in pipeline-riser system

For offshore pipeline-riser systems, cyclic flow cut-off in slugging flow pattern should be identified and necessarily mitigated to satisfy the requirement of normal operation of gas–liquid separator. Since conventional studies on flow pattern identification were not directly related to the harmfulness of slugging and the decision of flow control, this study develops a risk assessment method for offshore riser systems based on the characteristic parameters of riser outlet signals, achieving an identification accuracy exceeding 90% at different valve openings. Notably, the method maintains exceptional performance under high-pressure conditions, with identification accuracy reaching 96.74% at 10 MPa. On the other hand, this study focuses on the prediction of cut-off duration time in a cycle instead of the cycle period. We establish a predictive model for flow cut-off duration time under different choking conditions. Validation across diverse pipeline systems demonstrates strong reliability of the model, exhibiting maximum prediction errors of 30%. The model's coefficients can be fitted by a small dataset without significant loss of accuracy, and the coefficients are conservative, indicating strong applicability in the field operation. These findings offer significant theoretical insights and practical solutions for enhancing the operational safety of offshore oil and gas transportation systems.