Real-time Diagnostics of Oil Production Equipment using Data Mining

摘要

In this paper, we consider the problem of increasing the efficiency of heavy oil production by a qualitative enhanced oil recovery (EOR) application. A novel approach, based on the analysis of data related to successful events is proposed. In particular, a synthesised model, targeted to automated search of wells for EOR application, is developed. For data processing, we present a novel approach based on a hybrid implementation of neural network analysis techniques and evolutionary algorithms. The presented approach enables the selection of EOR in fuzzy, difficult to formalise, oilfield conditions and consequently reduces the dependency on human factors.