Well log data inversion using higher order neural networks

Kou-Yuan Huang; Liang Chi Shen; Chun Yu Chen

doi:10.1109/IGARSS.2008.4779548

Well log data inversion using higher order neural networks

Kou-Yuan Huang^*, Liang Chi Shen, Chun Yu Chen

^*Corresponding author for this work

Department of Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Scopus citations

Abstract

We use the multilayer perceptron for well log data inversion. The gradient descent method is used in the back propagation learning rule. The input of the network is the apparent conductivity (Ca) and the output of the network is the true formation conductivity (Ct). The original and the higher order features are used for the training process. According to our experimental results, the expanding higher order input features can get a fast training and a smaller error between the desired output and the actual output. The network with 10 input nodes and expanding the input features to third order, 8 hidden nodes, 10 output nodes, can get the smallest average mean absolute error on simulated well log data. Then, we apply the network to the real field data.

Original language	English
Title of host publication	2008 IEEE International Geoscience and Remote Sensing Symposium - Proceedings
Edition	1
DOIs	https://doi.org/10.1109/IGARSS.2008.4779548
State	Published - 1 Dec 2008
Event	2008 IEEE International Geoscience and Remote Sensing Symposium - Proceedings - Boston, MA, United States Duration: 6 Jul 2008 → 11 Jul 2008

Publication series

Name	International Geoscience and Remote Sensing Symposium (IGARSS)
Number	1
Volume	3

Conference

Conference	2008 IEEE International Geoscience and Remote Sensing Symposium - Proceedings
Country	United States
City	Boston, MA
Period	6/07/08 → 11/07/08

Keywords

Higher order
Multilayer perceptron
Well log inversion

Access to Document

10.1109/IGARSS.2008.4779548

Link to publication in Scopus

Fingerprint Dive into the research topics of 'Well log data inversion using higher order neural networks'. Together they form a unique fingerprint.

Cite this

Huang, K-Y., Shen, L. C., & Chen, C. Y. (2008). Well log data inversion using higher order neural networks. In 2008 IEEE International Geoscience and Remote Sensing Symposium - Proceedings (1 ed.). [4779548] (International Geoscience and Remote Sensing Symposium (IGARSS); Vol. 3, No. 1). https://doi.org/10.1109/IGARSS.2008.4779548

@inproceedings{f8e5c817d9d04c5983e88469a7344e3e,

title = "Well log data inversion using higher order neural networks",

abstract = "We use the multilayer perceptron for well log data inversion. The gradient descent method is used in the back propagation learning rule. The input of the network is the apparent conductivity (Ca) and the output of the network is the true formation conductivity (Ct). The original and the higher order features are used for the training process. According to our experimental results, the expanding higher order input features can get a fast training and a smaller error between the desired output and the actual output. The network with 10 input nodes and expanding the input features to third order, 8 hidden nodes, 10 output nodes, can get the smallest average mean absolute error on simulated well log data. Then, we apply the network to the real field data.",

keywords = "Higher order, Multilayer perceptron, Well log inversion",

author = "Kou-Yuan Huang and Shen, {Liang Chi} and Chen, {Chun Yu}",

year = "2008",

month = dec,

day = "1",

doi = "10.1109/IGARSS.2008.4779548",

language = "English",

isbn = "9781424428083",

series = "International Geoscience and Remote Sensing Symposium (IGARSS)",

number = "1",

booktitle = "2008 IEEE International Geoscience and Remote Sensing Symposium - Proceedings",

edition = "1",

note = "null ; Conference date: 06-07-2008 Through 11-07-2008",

}

Huang, K-Y, Shen, LC & Chen, CY 2008, Well log data inversion using higher order neural networks. in 2008 IEEE International Geoscience and Remote Sensing Symposium - Proceedings. 1 edn, 4779548, International Geoscience and Remote Sensing Symposium (IGARSS), no. 1, vol. 3, 2008 IEEE International Geoscience and Remote Sensing Symposium - Proceedings, Boston, MA, United States, 6/07/08. https://doi.org/10.1109/IGARSS.2008.4779548

Well log data inversion using higher order neural networks. / Huang, Kou-Yuan; Shen, Liang Chi; Chen, Chun Yu.

2008 IEEE International Geoscience and Remote Sensing Symposium - Proceedings. 1. ed. 2008. 4779548 (International Geoscience and Remote Sensing Symposium (IGARSS); Vol. 3, No. 1).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Well log data inversion using higher order neural networks

AU - Huang, Kou-Yuan

AU - Shen, Liang Chi

AU - Chen, Chun Yu

PY - 2008/12/1

Y1 - 2008/12/1

N2 - We use the multilayer perceptron for well log data inversion. The gradient descent method is used in the back propagation learning rule. The input of the network is the apparent conductivity (Ca) and the output of the network is the true formation conductivity (Ct). The original and the higher order features are used for the training process. According to our experimental results, the expanding higher order input features can get a fast training and a smaller error between the desired output and the actual output. The network with 10 input nodes and expanding the input features to third order, 8 hidden nodes, 10 output nodes, can get the smallest average mean absolute error on simulated well log data. Then, we apply the network to the real field data.

AB - We use the multilayer perceptron for well log data inversion. The gradient descent method is used in the back propagation learning rule. The input of the network is the apparent conductivity (Ca) and the output of the network is the true formation conductivity (Ct). The original and the higher order features are used for the training process. According to our experimental results, the expanding higher order input features can get a fast training and a smaller error between the desired output and the actual output. The network with 10 input nodes and expanding the input features to third order, 8 hidden nodes, 10 output nodes, can get the smallest average mean absolute error on simulated well log data. Then, we apply the network to the real field data.

KW - Higher order

KW - Multilayer perceptron

KW - Well log inversion

UR - http://www.scopus.com/inward/record.url?scp=67649800339&partnerID=8YFLogxK

U2 - 10.1109/IGARSS.2008.4779548

DO - 10.1109/IGARSS.2008.4779548

M3 - Conference contribution

AN - SCOPUS:67649800339

SN - 9781424428083

T3 - International Geoscience and Remote Sensing Symposium (IGARSS)

BT - 2008 IEEE International Geoscience and Remote Sensing Symposium - Proceedings

Y2 - 6 July 2008 through 11 July 2008

ER -