Revolutionizing the Perforation Process Via an End-To-End Automated Perforation Toolbox in the Gulf of Thailand (PerfSight)

Abstract This paper presents an innovative solution to generate the composite log using the in-house auto depth-shifting program and addresses the challenge of determining the optimal perforation depth for reservoir engineers to achieve optimal hydrocarbon recovery and reservoir performance. The traditional manual process of depth shifting is not only time-consuming but also prone to human errors, requiring collaborative efforts from multiple stakeholders. To address these challenges, Perfsight was developed, an in-house programming tool. It incorporates three key modules: auto depth matching, perforating depth optimization, and cement verification. The first module utilizes grid search to align depths between open-hole and cased-hole logs, accounting for variations in weight tension. The second module employs linear programming to optimize perforation interval selection, considering practical constraints. By assessing key factors such as resistivity, gamma rays, density, and neutron, our automated method ensures optimal perforation intervals. Lastly, the cement verification module utilizes a cement score equation to automatically identify the "good cement zone," mitigating the risk of flow behind the casing. Our study demonstrates a perforation enhancement process, transitioning from depth matching to cement verification in the perforating process. Our depth matching module achieves a commendable similarity score of 0.93, facilitating automatic correlation between open-hole and cased-hole logs. Following this, our tool utilizes linear programming to suggest perforating depths for each sand, considering operational constraints such as interval length and patching area to prevent water-coning. The model selects optimum intervals by prioritizing prolific, porous, clean, and top sections of the sands, resulting in a 20% increase in prolific score compared to manual selection. Subsequently, the model evaluates cement quality using a developed equation, identifying intervals with a cement score exceeding 80 as "good cement zones" to prevent water-flowing behind the casing. Our tool significantly streamlines the perforating process, reducing processing time from 4 hours to just 20 minutes, while also automating error-prone tasks such as cement quality verification. When integrated, these components produce a comprehensive one-stop service, generating a composite log.