Mechanism, Quantitative Diagnosis, and Classification-Based Remediation of Wellbore Blockage in Ultradeep Sour Gas Reservoirs

Summary Wellbore blockage poses a significant challenge to the efficient development of ultradeep sour gas reservoirs. In this study, we investigate the Yuanba Gas Field in the Sichuan Basin, China, utilizing an integrated suite of analytical techniques that includes calcination experiments, X-ray diffraction (XRD), scanning electron microscopy (SEM) with energy-dispersive spectroscopy (EDS), Fourier-transform infrared spectroscopy (FTIR), and gas chromatography-mass spectrometry (GC-MS) to systematically characterize the multisource composition and spatial distribution of blockage materials. The results reveal that the blockages originate from both formation-derived substances (e.g., iron-sulfur compounds and asphaltenes) and engineering-introduced contaminants (e.g., corrosion inhibitors and drilling fluid additives). Based on composition, they are categorized into inorganic, organic, and hybrid types, with their spatial distribution governed by the interplay between reservoir heterogeneity and operational intensity. To overcome the limitations of conventional experience-based remediation, we introduce a data-driven framework that integrates machine learning (ML) with wellbore mechanical analysis. The framework enables high-precision prediction of wellbore flowing pressure and quantitative diagnosis of blockage severity, successfully distinguishing between wellbore blockage (WA) and near-wellbore resistance (WB) while forecasting dynamic blockage trends. Building on this diagnostic capability, a “diagnosis-classification-dynamic optimization” remediation strategy is proposed. Field applications demonstrate its effectiveness: In 2025, 13 well interventions guided by this approach achieved an average tubing-pressure recovery of 3.4 MPa and a daily gas-production recovery of 3.5×104 m3 per well. In frequently blocked wells such as Y15 and Y2-2, the effective remediation period was extended from several tens of days to more than 160 days. This research provides a comprehensive, data-guided workflow—from accurate mechanistic characterization to adaptive remediation—for ensuring flow assurance in ultradeep high-sulfur gas reservoirs.