Ocena charakteru nasycenia skał zbiornikowych miocenu Przedgórza Karpat z wykorzystaniem analizy regresyjnej wielokrotnej

Authors

  • Stanisław Rychlicki
  • Kazimierz Twardowski

Abstract

W artykule przedstawiono, opracowany dla osadów autochtonicznego miocenu Przedgórza Karpat statystyczny model oceny charakteru nasycenia skał zbiornikowych z wykorzystaniem wielokrotne analizy regresji. Podano sposób przygotowania danych geologiczno-geofizycznych, metodykę obliczeń oraz oceniono pewność uzyskanych reguł rozwiązujących.ESTIMATION OF CHARACTER OF SATURATION OF MIOCENE RESERVOIR ROCKS FROM THE CARPATHIAN FORELAND WITH THE USE OF MULTIVARIATE REGRESSION ANALYSISThe paper presents the technique of estimation of character of saturation of reservoir rocks with the use of linear multivariate regression analysis, elaborated for the Miocene of the Carpathian forefield. On the basis of results of earlier studies carried out 'by the authors the Miocene was subdivided into 5 depth zones and a special technique was elaborated for every of these zones. The technique was elaborated using summation, of results of attempts involving a whole array of physical parameters characterizing sampled horizon. The parameter P prod. was accepted as a measure of affiliation of a given horizon to class of gas-bearing rocks. The value of that parameter was calculated depending on results of sampling performed according to a given scheme. Linear multivariate regression equations, making possible prognosis of the value P prod., were calculated for particular depth zones. The equations comprised physical parameters characterizing rock which were derived from geological-geophysical and drilling works. Assignation of a given interval to the gas- or water-bearing class is made by comparison of forecasted values of P prod. parameter with limit values of that parameter, determined for particular depth zones (Tab. 5). Such technique makes possible determining the probability of gas-bearing capability for a given interval. The reliability of the technique changes from 37 to 80%, depending on the significance level accepted.

Downloads

Issue

Section

Articles