Smart Cementing: IoT-Driven Real-Time Well Integrity Monitoring System

Thesis Abstract

The integrity of cement sheaths in oil and gas wells is increasingly challenged by complex subsurface conditions, corrosion, micro-annuli, and operational interruptions, necessitating real-time, connected monitoring to prevent catastrophic failures and non-productive time. This study develops and evaluates a comprehensive IoT-driven real-time well integrity monitoring system (WIIMS) that integrates embedded sensing, edge computing, and cloud analytics to provide continuous assessment of cement job longevity, zonal isolation, and casing integrity. The aim is to provide a robust, scalable framework that enables proactive maintenance and optimized cementing operations. Specific objectives include (1) designing a sensorization strategy and rugged communication network for downhole and surface deployment, (2) creating a data fusion architecture that assimilates cement-bond log proxies, temperature–pressure profiles, acoustic signals, and chemical tracer data, (3) developing predictive models for early failure detection using machine learning techniques, (4) implementing a decision-support module aligned with ISO 13526 and NORSOK standards for risk-based intervention, and (5) validating the system through a field-scale pilot in a controlled wellbore environment and a retrospective analysis of 24 cementing projects with heterogeneous geologies. The methodological framework adopts a mixed-methods approach anchored in the technology-organization-environment (TOE) framework and the Technology Acceptance Model (TAM) to assess technological feasibility and user adoption among operators. The population comprises 12 onshore and offshore rigs within a major operator’s portfolio, with a purposive sample of 60 intervals across 20 wells selected for instrumented deployment, complemented by 24 historical cementing case studies. Data collection instruments include (i) downhole sensors measuring temperature, pressure, vibration, and fluid conductivity; (ii) acoustic emission transducers for micro-annuli detection; (iii) cement-bond log proxies derived from gamma ray and neutron porosity sensors; (iv) tracer-based leakage indicators; (v) operator interviews and workflow diaries; and (vi) archival well integrity reports. Validity and reliability are ensured through calibration against standard cementing diagnostics, cross-validation with wireline cement-bond logs, and test-retest reliability of the edge devices. The data analysis plan comprises (a) time-series analyses and multivariate regression to identify correlations between cement integrity indicators and casing leakage events; (b) ensemble learning (random forest, gradient boosting) for predictive maintenance and anomaly detection; (c) survival analysis to estimate time-to-failure under varying operational conditions; (d) Bayesian updating for real-time probabilistic risk assessment; (e) thematic analysis of operator interviews to derive usability and adoption factors; and (f) model validation using hold-out wells and k-fold cross-validation with performance metrics including ROC-AUC, F1-score, and mean absolute error. The analytical framework also incorporates causal modeling to infer driving factors of cement failure and integrity degradation, complemented by a conceptual model illustrating data flow from sensing to decision support. Expected findings include (i) a validated sensor suite and communication protocol enabling reliable data transmission in challenging downhole environments, (ii) a high-performing predictive model achieving ROC-AUC ? 0.92 in identifying impending bonding failures within a 30-day horizon, (iii) a 15–20% improvement in decision-making speed for interventions due to the WIIMS dashboard, and (iv) demonstrable reductions in non-productive time and cementing rework by at least 12% in the pilot wells. The study provides evidence on the added value of integrating IoT with cementing operations and identifies thresholds for proactive mitigations such as cement slurry modification, stage cementing adjustments, and casing cement placement timing. The contribution to knowledge includes a novel, field-validated WIIMS architecture that bridges real-time sensing, edge analytics, and centralized governance within the framework of cementing reliability and well integrity. The abstract concludes with recommendations for standardization of data schemas, governance policies for data privacy and cyber security, and pathways for scale-up across diverse geology and operational contexts. The main conclusion posits that IoT-enabled real-time monitoring substantially enhances cementing reliability and well integrity when coupled with robust predictive analytics and operator-centric decision support. Recommendations emphasize continuing field trials across multiple basins, integration with existing drilling and completion software, enhancement of edge computing capabilities for latency-sensitive alerts, and development of industry-wide benchmarks for real-time cementing performance metrics.