科研之友
微信
新浪微博
Facebook
分享链接
摘要
This paper presents a new pressure-rate deconvolution algorithm to estimate reservoir properties and predict well performance from measured well responses. The algorithm combines the conveniences of deconvolution in Laplace domain with a new approach to transform sampled data from time domain to Laplace domain, which removes the necessity to extrapolate the data beyond the sampling interval imposed by Laplace transformation. This technique leads to a series form of the deconvolution expression in Laplace domain. Appropriate algorithms are presented to calculate the coefficients of the series to evaluate the results either in time domain or the Laplace space. The time domain computations, which are in the form of a closed solution, do not require evaluation or inversion of the Laplace transforms, can be implemented easily, and returns the equivalent constant-rate production data with minimum user interference. The Laplace domain calculations require the implementation of an inversion algorithm, but provide a wide range of applications in the solution of many pressure-or rate-transient analysis problems. For both cases, the computations are robust and accurate. An optimization algorithm in time domain is introduced in order to alleviate the noise effect on the constant-rate pressure behavior and its logarithmic derivative. The outcome of the process is the data converted to equivalent responses at constant-rate production, which is amenable to analysis by standard PTA/RTA models. Similar to the recent time-domain deconvolution algorithms, which are controlled by a user supplied regularization parameter, the proposed approach also requires minimal user interference.
- 出版日期2017-8
