Oil Well Productivity Computation Based on a Brain-Inspired Cognitive Architecture

Abstract

This paper aims to mitigate the negative effect of production data errors on the curve of inflow performance relationship (IPR). To this end, the author proposed a brain-inspired productivity computation method for oil wells based on Shannhan's brain-inspired cognitive architecture. The architecture consists of two interacting sensorimotor loops. In the proposed method, the IPR parameters were fitted in the inner loop, and the fitting results were considered in the productivity computing in the outer loop. Then, the proposed model was applied to compute the oil productivity of a real oil well, compared with other common methods, and verified through numerical simulation. The results show that the new method can predict well productivity more accurately than the contrast methods. Suffice it to say that this research puts forward a simple and reliable method for IPR curve drawing.