Recently, the team of Zhang Kai has made new progress in the field of model transfer research in history matching, with related findings published in the journal Geoenergy Science and Engineering. The paper is titled "Rapid History Matching through Evolutionary Algorithms and Multi-stage Experience Transfer".
Innovation: Numerous studies focus on improving the performance of history matching algorithms, yet few leverage the valuable historical data amassed during reservoir development. Typically, these studies initiate history matching anew at the target stage, resulting in considerable time and resource inefficiencies. In this study, we propose an innovative historical experience transfer paradigm that effectively utilizes historical data to inform and optimize history matching at the target stage. Algorithms developed under this paradigm demonstrate superior performance compared to conventional methods, achieving enhanced convergence and more accurate parameter inversion. Additionally, they substantially reduce the sample requirements at the target stage, thereby minimizing computational costs and enabling rapid history matching. Moreover, the proposed transfer paradigm is highly adaptable, offering the potential to enhance the efficacy of a wide range of history matching algorithms.
Abstract: History matching in reservoir development is an iterative process of model calibration, where model parameters are continuously adjusted to align the model outputs with actual production data as closely as possible. During the history matching process, a significant amount of valuable data and associated correction errors accumulate. Notably, a majority of current automated history matching algorithms do not sufficiently address this accumulation and initiate each iteration of the history matching from the beginning, thereby failing to capitalize on potential enhancements in model accuracy. Addressing this gap, this study introduces a novel approach for rapid update in history matching, leveraging historical correction errors through a Historical Experience Transfer System (HETS) with Differential Evolution (DE) to map the relationship between historical and target stages, termed as HETS-DE. Guided by historical experience, it effectively avoids local optima, enhances convergence speed, and reduces the need for sample collection in the target stage. In tandem with the DE algorithm for parameter space exploration, it facilitates swift updates in history matching processes. The efficacy of HETS-DE was compared against the traditional surrogate-assisted evolutionary algorithm using three reservoir models characterized by distinct geological attributes. The findings indicate that HETS-DE demonstrates superior convergence in history matching with fewer samples than traditional strategies. Notably, in the model of regular reservoir, HETS-DE achieves faster convergence and enhanced matching outcomes with just a quarter of the data samples of other surrogate-based methods These results underscore the algorithm's effectiveness in history matching and its potential for rapid update implementations, representing a new paradigm for reservoir history matching.
The Journal of Petroleum Science and Engineering covers the fields of petroleum and natural gas exploration, production and flow in its broadest possible sense. Topics include: reservoir engineering; reservoir simulation; rock mechanics; petrophysics; pore-level phenomena; well logging, testing and evaluation; mathematical modeling; enhanced oil and gas recovery; petroleum geology; compaction/diagenesis; petroleum economics; drilling and drilling fluids; thermodynamics and phase behavior; fluid mechanics; multi-phase flow in porous media; production engineering; formation evaluation; exploration methods; CO2 Sequestration in geological formations/sub-surface, etc. The latest impact factor of the journal is 4.346, and the average impact factor IF in the past 3 years is 3.646. This journal currently appears in the JCR Q1 or the Chinese Academy of Sciences ranking 2.
Paper link:
https://doi.org/10.1016/j.geoen.2024.213553
Citation:
Zhang W, Liu P, Zhang K*, Zhang Liming, Yan Xia, Yang Yongfei, Sun Hai, Wang Jian, Yao Jun. Rapid history matching through evolutionary algorithms and multi-stage experience transfer[J]. Geoenergy Science and Engineering, 2025, 246: 213553.
