Please use this identifier to cite or link to this item: http://dspace.library.iitb.ac.in/xmlui/handle/123456789/19339
Title: Predicting CO2 permeability of bituminous coal using statistical and adaptive neuro-fuzzy analysis
Authors: SHARMA, LK
VISHAL, V
SINGH, TN
Keywords: Hole Stress Conditions
Inference System
Reservoir Characterization
Feature-Selection
Carbon-Dioxide
Network
Sequestration
Performance
Saturation
Simulation
Issue Date: 2017
Publisher: ELSEVIER SCI LTD
Citation: JOURNAL OF NATURAL GAS SCIENCE AND ENGINEERING,42,216-225
Abstract: The permeability of coal is considered as one of the critical parameters for coal bed methane (CBM) extraction and for evaluating the ability of coal seams to transport and store anthropogenic carbon dioxide. The heterogeneity in coal from basin to basin, inter and intra seams makes the estimation of permeability more critical and complex. In this work, permeability of Indian coal was evaluated at varied injection pressure and effective stresses using adaptive neuro-fuzzy inference system (ANFIS) technique and the outcomes were compared with the traditional statistical method of multiple regression analysis (MRA). The uniqueness of this technique lies in the fact that it integrates the learning capacity of neural networks and reasoning capabilities of fuzzy logic. For the development of prediction model, 48 datasets measured from laboratory analysis were used in such a way that the first set of 38 datasets were corresponding to the input parameters to train the models and 10 datasets were used to test the accuracy of the developed models. The performance capacity of the predictive models was evaluated based on coefficient of determination (R-2), the mean absolute percentage error (MAPE) and the root mean square error (RMSE). The ANFIS predictive model had the R-2, MAPE and RMSE equal to 0.9989, 1.8535% and 0.0002, respectively, supersede the performance of the MRA. Thus, the results from the present study are very encouraging and can be used to predict the permeability of coal. (C) 2017 Elsevier B.V. All rights reserved.
URI: http://dx.doi.org/10.1016/j.jngse.2017.02.037
http://localhost:8080/xmlui/handle/123456789/19339
ISSN: 1875-5100
2212-3865
Appears in Collections:Article

Files in This Item:
There are no files associated with this item.


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.