Development of probabilistic and statistical models for evaluation of the effectiveness of proppant hydraulic fracturing (on example of the Tl-Bb reservoir of the Batyrbayskoe field)

Abstract

The main factors affecting the efficiency of proppant hydraulic fracturing of Tl-Bb clastic reservoir depending on various parameters are statistically studied. There are 36 hydraulic fracturing treatments pumped from 2008 to 2016 with an average increase in oil production rate of 8.6 tons per day. Probabilistic statistical models were built to determine the parameters that influence the effectiveness of hydraulic fracturing. The average annual increase in oil production is used as a dependent variable; geological, technological and technical parameters are used as independent variables. To determine the degree of impact of parameters wells are divided into two classes of effectiveness: Qo > 8 tons/day (class 1), Qo < 8 tons/day (class 2). For the first class of each parameter individual statistical models are built for prediction and its probability is calculated. For the combined use of individual models complex probability is calculated separately for geological technological and technical indicators. As a result, regression models are built using step-by-step regression analysis. The standard error of the model for geological and technological parameters is 2.0 tons/day and 2.2 tons/day for technical. The joint consideration of geological technological and technical parameters in the regression model reduces the standard error to 1.5 tons/day. It is concluded that a separate description of the processes of hydraulic fracturing allows evaluating the efficiency of hydraulic fracturing in specific geological and technological conditions at the design stage, based on the use of developed individual models. After the hydraulic fracturing performed using the developed models considering technical conditions it is possible to preliminary estimate the effectiveness of hydraulic fracturing. The developed methodology for predicting the efficiency of hydraulic fracturing, performed on the Tl-Bb reservoir, is recommended to be used at other reservoirs of the field after correction of the built models. On the other fields where information for building the probabilistic and statistical models is available, it is necessary to develop new probabilistic and statistical models.

Keywords

• hydraulic fracturing of formations
• prediction
• average annual increase in oil production rate
• clastic reservoir
• Batyrbayskoe field
• geological and technological parameters
• technical parameters
• probabilistic model
• statistical model
• complex probability
• combination of probabilities
• Pearson's criterion