Fire-flooding, a mainstream thermal recovery technology for heavy oil reservoirs, requires large-scale continuous air injection and inevitably generates substantial fire-flooding tail gas with a high greenhouse gas content, posing a severe challenge to the environmental sustainability of offshore oil exploitation. To address this critical issue and realize carbon-neutral development, this study proposes an integrated near-zero emission framework for offshore heavy oil fire-flooding tail gas based on multipath storage combined with intelligent optimization algorithms, addressing the limitations of single storage methods and insufficient intelligent integration in existing research. The framework integrates fire-flooding tail gas capture, transportation, and reinjection into adjacent light oil and hydrate-bearing reservoirs to synergistically achieve carbon storage and enhanced oil recovery (EOR), with a multiobjective optimization system established via BP neural networks and optimization algorithms (genetic algorithm and particle swarm optimization) to efficiently optimize injection-production parameters. Key results show that all involved reservoirs exhibit three distinct production stages, and the optimal tail gas reinjection timing is within the first 1000 days of fire-flooding. The proposed emission mode and optimization method achieve the dual goals of near-zero carbon emission and production enhancement, verifying the feasibility and superiority of the multipath utilization scheme and providing important theoretical guidance and technical support for the efficient, low-carbon, and sustainable development of offshore heavy oil resources.
Yuan et al. (Thu,) studied this question.