Widespread Use of Numerical Modelling
Reservoir simulation is a maturing market. During the past decades the speed and capacity of computer hardware has increased. Concurrently, the real price of computers has dropped. Large scale simulations are now economic. With Alberta’s oil reservoirs in advanced stages of depletion, production optimisation will become necessary. Widespread use of numerical modelling has occurred. To ARE’s perception this is declining in Western Canada due to the very advanced stages of depletion in the conventional business. Thermal simulation is rapidly increasing.
Dangers in Reservoir Simulation
There are many dangers in reservoir simulation. The numerical procedures within the models have improved, and models are more stable and have become much easier to run. Models will run numerically in spite of basic errors in reservoir description or choice of grid. Reservoir simulation requires specialized knowledge of numerical considerations. For these reasons it can be desirable to have a specialty consultant perform these studies.
Critical Issues in Reservoir Simulation Studies
Reservoir simulation studies vary greatly in quality. The most important differences are reservoir description and comprehensive checking of output data. Most predictions should be checked against offset analogies. The emphasis on ARE’s reservoir simulation is therefore as follows:
- Comprehensively prepared reservoir descriptions. The first step is a good understanding of the depositional environment. This requires a basic knowledge of geology. It is critical for a simulation engineer to be able to read logs, since this is one’s “window” into the reservoir.
- Geology and reservoir engineering need to be connected. Detailed geological descriptions are required. Historically reservoir description was about layering. The use of geostatistics has created some very sophisticated reservoir descriptions. However, the age old problem persists. Geologists used to make maps that the engineers digitized. Now, geo-modelling specialists create very sophisticated models that are probablistic and which don’t necessarily represent reality. One thing has not changed, there needs to be a good connnection betweent the geological description and the reservoir model.
- Checking offset information. This aids in identifying critical modelling parameters. Frequently, unrepresentative models can be weeded out by comparison to offsetting production.
- Providing a quality report. Simulation involves an enormous amount of input data. A report should include a copy of the basic data input deck. There is no other method of reproducing the results in the future without this information. A run by run history match log should be supplied. The changes should be described so that the client will know what decisions and assumptions were made. Results should be displayed graphically. This allows rapid assimilation of the results and provides a vital quality control check.
ARE has prepared a considerable amount of material on preparing simulations comprehensively, which include numerous courses, a textbook, as well as a 2 part JCPT Distinguished Author Series What You Should Know About Assessing Simulation Results For Economic Evaluations.