No abstract

No abstract

Early water breakthrough and a rapid increase in water cut are always observed in highpermeability completion intervals when perforations are uniformly distributed in the wellbore in heterogeneous reservoirs. Optimization of perforating parameters in partitioned sections in horizontal intervals helps homogenize the infl ow from the reservoir and thus is critically important for enhanced oil recovery. This paper derives a coupled reservoir-wellbore flow model based on inflow controlling theory. Genetic algorithms are applied to solving the model as they excel in obtaining the global optimum of discrete functions. The optimized perforating strategy applies a low perforation density in highpermeability intervals and a high perforation density in low-permeability intervals. As a result, the infl ow profi le is homogenized and idealized.

Graded blended cement made of graded Portland cement (PC), blast furnace slag (BFS) and fly ash (FA) is attractive for cement production. For manufacturing graded blended cement, a suitable mathematical expression should be introduced to describe the particle size distribution (PSD) of its components and control the quality of graded blended cement. This study aims to evaluate Rosin-Rammler-Sperling-Bennet (RRSB) distribution and lognormal distribution for describing the PSD of the components of graded blended cement. RRSB distribution and lognormal distribution are used to fit the PSD of ungraded and graded PC, BFS and FA. It is found that lognormal distribution exhibits smaller fitting errors for describing the PSDs of graded PC, BFS, FA and ungraded FA. What is more, lognormal distribution exhibits good simplicity and popularity. Hence, it is recommended to use lognormal distribution to control the PSD of graded blended cement in manufacturing process.