Application of neural networks in synthetic log generation
by Pham Son Tung; Nguyen Ninh Giang; Nguyen Dac Nhat; Ta Quoc Dung
International Journal of Oil, Gas and Coal Technology (IJOGCT), Vol. 30, No. 2, 2022

Abstract: This paper aimed to study the application of artificial neural network in solving the problem of missing logs, a problem frequently encountered in petroleum industry. Firstly, fully connected neural network (feed-forward neural network) and long-short-term memory network (recurrent neural network) were used to generate synthetic logs in order to determine the most suitable approach to the missing log issue. Secondly, various methods of data pre-processing, such as normalisation, outlier, and principal component analysis, were also analysed so that their effect on the application of neural network in solving incomplete log problem can be evaluated. Finally, the influence of number of stratigraphy layers on the accuracy of using neural networks to solve missing logs problem was equally considered in this research. The results showed that the most accurate way to generate synthetic logs is to use feed-forward neural network for one stratigraphy layer while the data pre-processing method should be normalisation. [Received: August 17, 2020; Accepted: June 21, 2021]

Online publication date: Wed, 04-May-2022

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Oil, Gas and Coal Technology (IJOGCT):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com