Automated Annulus Pressure Monitoring and Analysis Tool for Sustained Casing Pressure Detection

Abstract Sustained casing pressure (SCP) management remains one of the most significant well integrity challenges in the oil and gas industry. SCP impacts well barrier reliability and contributes to cumulative risk at platform level. Generally, SCP risks are managed and remediation efforts executed during sidetrack or well intervention activities. Effective monitoring and management of annuli pressure is critical in the early detection of abnormal pressure behavior and the timely identification of potential barrier element degradation. Loss of well barrier integrity can lead to uncontrolled fluid migration to the surface or between formation zones. In line with industry-wide digital transformation initiatives, BP has prioritized the automation of annuli pressure surveillance through the integration of advanced programming logic and machine learning (ML) techniques. This paper presents the development and deployment of a novel SCP detection and annuli pressure monitoring methodology, implemented through a customized software tool - Proxima. The system utilizes raw sensor data to automatically identify abnormal annuli behavior using a multi-step process that includes data pre-processing, signal analysis, and post-processing. The approach combines classical signal processing with advanced ML algorithms to enhance SCP detection accuracy and enable predictive capabilities. The Proxima tool is designed to continuously monitor annuli pressures, distinguish between normal and anomalous behavior, and alert responsible engineers in real-time via automated email notifications. It reduces manual engineering effort and enables timely, data-driven decisions for SCP risk mitigation. This paper highlights the operational logic behind Proxima, demonstrates its performance through real SCP cases from the Azerbaijan-Georgia-Türkiye (AGT) region, and discusses the added value of ML-driven diagnostics in well integrity management.