APPLICATION OF REGRESSION ANALYSIS TO EVALUATE THE EFFICIENCY OF OIL WELL OPERATING WITH THE PARAFFIN OIL

Abstract

The relevance of the research is caused by the need for timely and rapid analysis of the state of oil wells which work is complicated by paraffinicity. The increase of the number of oil well indicators due to complicated production conditions leads to the difficulties in their interpretation by the geophysical services specialists, and as a result to incorrect decisions about the operating efficiency of oil wells. The existing models for evaluation of the oil wells operating efficiency include either a small number of well operation indicators which makes them insufficiently complex, or an excessive number of indicators which makes the models difficult and resource-intensive. The main aim of the research is to define the geological and physical indicators that have a significant impact on the oil paraffinicity to create a regression model for predicting oil well operating efficiency from a geological point of view. Objects of researches are oil wells and oil well operating efficiency with regard to oil paraffinicity. Methods: regression analysis, evaluation of the model adequacy using Student’s t-criterion, determination coefficient, Fisher’s F-criterion, checking unbiasedness and efficiency of estimates using Gauss–Markov conditions, estimating residual autocorrelation using Durbin–Watson statistics, non-parametric statistics, regression residual analysis. Results. The authors have determined the form and type of regression dependence between oil paraffinicity, reservoir temperature and oil viscosity in reservoir conditions. The adequacy of the obtained model is proved. The resulting mathematical model in conjunction with the model of the technological oil well efficiency can be the basis for decision support systems of specialists of geophysical services for complex analysis of the oil well operating efficiency.

Keywords

• regression analysis
• oil well operating efficiency
• paraffinicity
• regression residual
• model adequacy