History Matching and Forecasting
Throughout my career, I have witnessed some funny moments concerning reservoir modeling and history matching. Here, I will talk about two of them. The first one is an adaptation of something I read: “History matching is the process of torturing a model until it confesses.” Despite the obvious anecdotal nature, this sentence is actually an interesting analogy. Torture makes the victim say anything you want to hear, even if it is not the truth. Please, do not get me wrong. I am not suggesting that history matching is a despicable practice comparable with torture. The analogy is more subtle. What I mean is that there is a scrupulous limit for history matching; if it starts looking like you are “torturing” the model, then you went too far. Stop there and rethink the process. You are not going to get reliable information from this model anyway.
The second incident occurred in a workshop I attended last year. At some point, a geologist made a presentation about history matching a field problem. He started the talk with the message “History Matching: 30 Years of Failure.” At first, I found it somewhat funny, but at the same time a little troubling. During the coffee break, I overheard a colleague, probably upset with the talk, paraphrasing with “Geological Modeling: 30 Years of Failure.” The geologist was unhappy with history matching because, in his past experience, that was the excuse for the engineer to distort his model. The engineer, on the other hand, blames the geological models that are not able to reproduce the observed production. That brings me exactly to the point I want to make. History matching should not be seen as a sealed task. History matching is only one part of something more comprehensive—reservoir modeling, a process that involves several disciplines working together toward the same goal, answering questions about the reservoir to be able to make useful forecasts and aid decision making.
Last year, we had a nice collection of papers presented at SPE conferences about history matching and forecasting. There are three points I will mention. First, there is a clear trend in the field applications of moving away from history matching a single model and adopting forecasts on the basis of multiple realizations. Second, there are some interesting papers on integrated reservoir studies. I selected one for the additional-reading list. Finally, I noticed a significant number of interesting papers on proxy and data-driven models for production forecast. I am not sure that this is a current trend, but I selected two papers for this month’s feature. (One is summarized in the main selection, and the other is in the additional-reading list.) Of course, you can find many other interesting papers at onepetro.org. I hope you enjoy the reading.
This Month's Technical Papers
Recommended Additional Reading
SPE 172594 Integration of Reservoir-Performance and Geoscience Studies in the History Match of a Complex Carbonate Reservoir—A Case Study From the Magwa Marrat Reservoir, Kuwait by Menayer Al Jadi, Kuwait Oil Company, et al.
SPE 173206 Physics-Based and Data-Driven Surrogates for Production Forecasting by Hector Klie, ConocoPhillips
SPE 174310 Seismic Assisted History Matching Using Binary Image Matching by Dennis Obidegwu, Heriot-Watt University, et al.
History Matching and Forecasting
Alexandre Emerick, SPE, Reservoir Engineer, Petrobras Research Center
01 April 2016
Enhancing Model Consistency in Ensemble-Based History Matching
The aim of this work is to present the effectiveness of a fully integrated approach for ensemble-based history matching on a complex real-field application.
Ensemble-Based Assisted History Matching With 4D-Seismic Fluid-Front Parameterization
An ensemble-based 4D-seismic history-matching case is presented in the complete paper. Seismic data are reparameterized as distance to a 4D anomaly front and assimilated with production data.
Rapid S-Curve Update Using Ensemble Variance Analysis With Model Validation
In the complete paper, the authors propose a novel method to rapidly update the prediction S-curves given early production data without performing additional simulations or model updates after the data come in.
Don't miss out on the latest technology delivered to your email weekly. Sign up for the JPT newsletter. If you are not logged in, you will receive a confirmation email that you will need to click on to confirm you want to receive the newsletter.
07 May 2018
08 May 2018