Classification and Evaluation System of Oil and Gas Reservoir Based on Artificial Neural Network

In recent years, the dissemination of information technology is more and more, this industry in recent years to the unprecedented rapid development, the human elite, artificial neural network as the representative of artificial intelligence has achieved rapid development, computer vision what is good, technical college entrance examination, English language and other geosciences, especially in the field of paint address of new technologies emerge endlessly. In this paper, artificial neural network technology is applied to reservoir classification. Sedimentary microfacies, grain size, sandstone thickness, porosity, permeability, acoustic time difference and resistivity, which are closely related to reservoir classification, are taken as input variables to establish a BP neural network model. Through the actual data testing and verification, a very special and achieved ideal results, indicating that the artificial neural network technology has a good prospect in the field of oil and gas geology.

I. Introduction

It is well known that traditional oil and gas reservoir classification methods are mainly qualitative or semi-quantitative identification methods [1]. The disadvantage is that it is highly artificial, and secondly, it is difficult to fully consider many reservoir evaluation parameters in a scientific manner, which is often overlooked. In this paper, artificial neural network is applied to the study of reservoir classification, which avoids the influence of human factors and makes the reservoir evaluation more quantitative and scientific. Taking Yan 'an Group in TBC area as an example, the practicability and effectiveness of objective evaluation neural network are given.