| Hybrid Workflow for Cementing Optimization |
| Paper ID : 1024-UGESR |
| Authors |
|
Hamid Saberi * Department of Petroleum Engineering, Faculty of Petroleum Engineering, Petroleum University of Technology (PUT), Ahvaz, Iran |
| Abstract |
| Well cementing design involves complex interactions among hydraulic, mechanical, and operational parameters, while conventional simulation-based approaches often fail to incorporate knowledge gained from previous field operations. This study presents a hybrid framework that integrates deterministic hydraulic modeling with machine-learning-assisted evaluation for cementing design optimization and benchmarking. The proposed methodology combines digital-twin wellbore simulation with clustering-based retrieval of historically analogous cementing jobs using operational and geological features such as depth, temperature, mud weight, casing geometry, and job type. A Cement Dependability Index (CDI) and an overall Job Score are introduced to quantify design quality and compliance with recommended operational practices. The framework enables engineers to benchmark candidate designs against the performance distribution of similar historical wells and iteratively optimize parameters such as spacer volume, pump rate, and centralizer placement. Validation using several thousand field cementing operations demonstrates that the proposed scoring methodology correlates consistently with well complexity and operational risk, while also identifying recurrent weaknesses in spacer effectiveness and standoff quality. The results indicate that integrating machine learning with hydraulic simulation improves decision support, reduces reliance on subjective engineering judgment, and establishes a continuously evolving data-driven workflow for reliable zonal isolation and well integrity management. |
| Keywords |
| Well Cementing; Machine Learning; Digital Twin; Hydraulic Simulation; Cement Dependability Index (CDI); K-means Clustering; Zonal Isolation; Well Integrity; Data-Driven Optimization; Petroleum Engineering |
| Status: Accepted |