The Statistical Simulation of Dataset in 3d Area With "Cubic" Соrrelation Function on Example Rivne NPP Geophysical Monitoring

Abstract

Background. The model and algorithm were developed by using optimal in the mean square sense "cubic" correlation function. An example of supplementing the results of geophysical studies of karst-suffuses phenomena with simulated data in the task of monitoring the density of the chalk stratum on the territory of the Rivne NPP is presented. The complex geophysical research was conducted on Rivne NPP area. The monitoring observations radioisotope study of soil density and humidity near the perimeter of buildings is of the greatest interest among these. In this case a problem was occurred to supplement simulated data that were received at the control of chalky strata density changes at the research industrial area with use of radioisotope methods on a grid that included 29 wells. This problem was solved in this work by statistical simulation method that provides the ability to display values (the random field of a research object in 3D area) in any point of the monitoring area. Methods. Based on the spectral decomposition of random fields in 3D space, a statistical model of the distribution of the average density of the chalk layer in the 3D observation area was built. Results. An algorithm for statistical simulation of random fields with a "cubic" correlation function is formulated. On the basis of the developed software, additional simulated realizations of the random component of the research subject on the grid of observations of the necessary detail and regularity were obtained. A statistical analysis of the results of the numerical simulation of the distribution of the average density of the chalk layer was carried out and their adequacy was tested. Conclusions. The method of statistical modeling of random fields with "cubic" correlation functions allows you to supplement data with a given accuracy.