DEVELOPMENT OF STATISTICAL MODELS FOR PREDICTING OIL-AND-GAS CONTENT (ON THE EXAMPLE OF TERRIGENOUS DEVONIAN SEDIMENTS OF NORTH TATAR ARCH)

The economic importance of creation more sophisticated models for the prediction of oil and gas from produced and iden- tified local structures in increasingunder the conditions of more complex search for oil and gas in Udmurtia. Some factual mate- rial about local structures containing hydrocarbon deposits, and on those structures which conducted exploratory drilling, but deposits of hydrocarbons are not open on the North-Tatar territory are gained. A feature of this method is that, as indicators to be used by those who are always available to producers. It should be noted that these figures will be an integrated, which guarantees high reliability built probabilistic and statistical forecast models of oil and gas. This technique of forecasting petroleum can be implemented in a specific area of study, that is, when the analysis can be used for a number of lifts, some of which contain hydrocarbons, others into these deposits do not contain, ie are empty struc- tures. It should be noted that all of these uplifts can be characterized by the same parameters, which can be determined before placing them deep exploratory drilling. This fact allows us to construct probabilistic and statistical models that can be practically used in assessing local undeveloped oil and gas structures. With this technique, you can plan the sequence of drilling on local uplifts that will stabilize the oil production in the Republic of Udmurtia. Identification of the most promising raised regarding the proposed oil and gas produce by constructing mathematical models of geological forecast. By analyzing the characteristics of uplifts are quantified ones that actually form the oil and gas structures. A distinctive feature of this work is that the construction of models in the first phase will be used not the indicators with different dimensions, and the probabilities calculated for them. This will be constructed regression equations by which to calculate probabilities. By the values of the probabilities using stepwise linear discriminant and multivariate regression analyzes will be developed a comprehensive probabilistic criterion. This criterion will be used in the future to build a multidimensional model is directly on the very characteristics of structures. Based on multivariate statistical model developed is impossi- ble to determine the priority structure, recommended for deep exploratory drilling.