Table of Contents Author Guidelines Submit a Manuscript
The Scientific World Journal
Volume 2013, Article ID 584516, 7 pages
Research Article

Prediction of Missing Flow Records Using Multilayer Perceptron and Coactive Neurofuzzy Inference System

Department of Civil Engineering, National Pingtung University of Science and Technology, Neipu Hsiang, Pingtung 91201, Taiwan

Received 25 August 2013; Accepted 2 October 2013

Academic Editors: R. Beale and R.-J. Dzeng

Copyright © 2013 Samkele S. Tfwala et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


Hydrological data are often missing due to natural disasters, improper operation, limited equipment life, and other factors, which limit hydrological analysis. Therefore, missing data recovery is an essential process in hydrology. This paper investigates the accuracy of artificial neural networks (ANN) in estimating missing flow records. The purpose is to develop and apply neural networks models to estimate missing flow records in a station when data from adjacent stations is available. Multilayer perceptron neural networks model (MLP) and coactive neurofuzzy inference system model (CANFISM) are used to estimate daily flow records for Li-Lin station using daily flow data for the period 1997 to 2009 from three adjacent stations (Nan-Feng, Lao-Nung and San-Lin) in southern Taiwan. The performance of MLP is slightly better than CANFISM, having of 0.98 and 0.97, respectively. We conclude that accurate estimations of missing flow records under the complex hydrological conditions of Taiwan could be attained by intelligent methods such as MLP and CANFISM.