Interference effect between wells is very common in mature reservoirs. This is particularly true for transient pressure data from permanent down-hole gauges (PDG). As soon as the recorded pressure interfered by the pressure disturbance from other wells nearby, the PDG pressure becomes the product of two components, i.e. self response due to the well production and interference from neighbouring wells. This combined pressure data recorded by the PDG has a nonlinear nature, which can not be analysed by traditional well testing principles such as superposition or deconvolution.

There are two approaches handling this type of data. One is numerical well testing designed for solving well testing problems in a nonlinear reservoir system. Another is to pre-process the data by decomposing the total pressure response into linear reservoir response and the component that causes the nonlinearity of the response, i.e. interference effect. Thereafter, the extracted interference data can be analysed as interference test, and the remaining data can be analysed by traditional well testing methods.

This paper presents a newly developed multi-well deconvolution algorithm specially designed for analysing long term transient pressure with interference. The procedures include firstly, the diagnostic of the reservoir system response for the nonlinearity. As soon as the reason causing this nonlinearity is found, in this presented case, is due to interference, the following step is to extract the interference from the total pressure response. Then, the analysis of the decomposed data can be made using the available traditional well testing methods. Numerical well testing synthetic studies were performed to demonstrate these procedures.

The study results proved that the new method appeared to work well in homogeneous reservoirs with wells flowing at single phase, single and multiple rates. The potential in further developing this technique for practical application is obvious and looks very promising.

You can access this article if you purchase or spend a download.