Applying Genetic Programming to Reservoir History Matching Problem

Abstract

History matching is the process of updating a petroleum reservoir model using production data. It is a required step before a reservoir model is accepted for forecasting production. The process is normally carried out by flow simulation, which is very time-consuming. As a result, only a small number of simulation runs are conducted and the history matching results are normally unsatisfactory.In this work, we introduce a methodology using genetic programming (GP) to construct a proxy for reservoir simulator. Acting as a surrogate for the computer simulator, the "cheap" GP proxy can evaluate a large number (millions) of reservoir models within a very short time frame. Collectively, the identified goodmatching reservoir models provide us with comprehensive information about the reservoir. Moreover, we can use these models to forecast future production, which is closer to the reality than the forecasts derived from a small number of computer simulation runs.We have applied the proposed technique to a West African oil field that has complex geology. The results show that GP is able to deliver high quality proxies. Meanwhile, important information about the reservoirs was revealed from the study. Overall, the project has successfully achieved the goal of improving the quality of history matuching results without increasing the number of reservoir simulation runs. This result suggests this novel history matching approach might be effective for other reservoirs with complex geology or a significant amount of production data.

Optimizing Waterflooding Process for Recovery Using Nonlinear Model Predictive Control: Case Yariguí - Cantagallo

“…In oil recovery maximization, the objective is to find optimal water injection rates to increase the oil field production [14]- [16]. For history matching, the optimizer is looking for proper field model parameters (permeability, porosity, saturations, among others) which match numerical simulations and field measurements [17]- [19]. Finally, the GA are also used to decide optimal location of new wells taking into account future predictions of possible patterns and history data [20]- [22].…”

Coevolution of simulator proxies and sampling strategies for petroleum reservoir modeling

“…Instead, we used the simulator proxy model generated from our previous study [18] as a surrogate. In this way, this proof-of-concept study can be completed faster, avoiding the time-consuming computer simulation.…”

“…Initially, we ran reservoir simulator on a small number of samples selected by uniform design [4] as we have observed that uniform-design samples are suitable for evolutionary systems to construct proxies [18]. These simulation data becomes the initial training data for the proxy population to train proxy models at the initial estimation phase.…”

“…Previously, we have introduced a framework applying genetic programming (GP) [11] to construct a reservoir Tina simulator proxy to enhance the history matching process [18]. Unlike the full reservoir simulator, which gives the flow of fluids of a reservoir, this proxy only labels a reservoir model as "good" or "bad", based on whether or not its flow outputs match well with the production data.…”

Estudo Sobre Caracterização De Reservatórios Por Programação Genética

“…Kaydani afirma que PG é uma poderosa ferramenta para reconhecer possíveis padrões entre conjuntos de entrada e saída, o qual pode ser aplicado para predizer parâmetros de reservatório. Outras pesquisas utilizam com sucesso PG nas áreas de: ajuste de histórico, criação de proxies de simuladores de reservatório e controle operacional de reservatório (YU, et al, 2007), (YU, et al, 2008), (FALLAH-MEHDIPOUR, et al, 2012 MAPE, que é uma medida de erro absoluto porcentual e o erro médio, que corresponde à média do valor absoluto dos erros.…”