The Quantity Analysis Method Research of Oil and Gas Geo-Anomaly Information Mining

Qing Li,Suhong Liu,Xiang Zhao,Peijuan Wang

DOI: https://doi.org/10.1109/igarss.2003.1295234

2003-01-01

Abstract:During oil-gas exploration, much information is collected concerning geo-physical exploration, geo-chemical exploration, remote-sensing and geology exploration. Depending on the traditional exploration methods, it is difficult to make further progress with the deep step in oil exploration, so new theories and methods are urgently needed. Geo-anomaly theory was first put forward by Zhao Pengda, a Chinese math geologist, in 1991. It has been widely applied in solid mineral exploration. In this research, it is promoted in oil and gas exploration. How to mine valuable oil and gas geo-anomaly information and knowledge from all collected data is a big problem. From this point of view, taking oil and gas exploration in one basin in east China as an example, research has been done on mining oil and gas geo-anomaly information using data mining technology. During the research, some new mining methods such as block convulsion filtering, BP neural network and grey prediction are applied besides entropy, and complex methods. Research shows that the block convulsion filtering method is beneficial in removing noise from the data, and grey prediction methods are better ways to extract geo-anomaly information from drilling and seismic data, the BP network method is used well in predicting reserve and cover layer parameters with high accuracy, complex and entropy methods make a great effect in predicting the oil and gas preferable area. Through the above research on oil and gas geo-anomaly information mining, good results have been achieved in quantity methods for oil and gas geo-anomaly information extraction. The research contributes a lot to get effective exploration knowledge and predict preferable objects for oil and gas exploration.