Abstract of the talk
Well test analysis is used to assess well condition and obtain reservoir parameters. Major improvements have occurred approximately 13 to 19 years apart, driven by the availability of both new types of data and new mathematical tools. While early interpretation methods (straight lines in the early 1950’s or log-log pressure plots in the late 1960’s) were limited to the estimation of well performance, well test analysis has become a very powerful tool for reservoir characterisation, following the formulation of an integrated methodology in the early 1980s; the introduction of pressure-derivative analysis in 1983; the development of complex interpretation models able to account for detailed geological features; the development of a stable deconvolution algorithm in the early 2000’s; and its successful extension to multiple interfering wells (2008-2018).
Although not strategically driven, progress in well test analysis techniques has essentially resulted in a significant increased capability for (1) identifying an applicable interpretation model and (2) verifying the consistency of that interpretation model. As a result, the amount of information that one can extract from well test data and more importantly, the confidence in that information, has increased significantly.
Any new improvement will have to further enhance the ability to identify and verify the well test interpretation model, and provide additional information. How to achieve this has always been difficult to predict, but, following the development of multiwell deconvolution, using richer signals and mining the huge amount of well data available is a most likely direction.