Missing River Discharge Data Imputation Approach using Artificial Neural Network

• Published in: ARPN Journal of Engineering and Applied Sciences
• Main Author: Mispan M.R.; Rahman N.F.A.; Ali M.F.; Khalid K.; Bakar M.H.A.; Haron S.H.
• Format: Article
• Language: English
• Published: Asian Research Publishing Network 2015
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85101170010&partnerID=40&md5=eab1dcbfa5795d1840e563222fff4331

The issue with missing data in hydrological models are very common and it occurs when no data value was stored during observation. In modelling, the missing data can affect the conclusion that can be drawn from the dataset. This paper presents the study on Levenberg-Marquadt back propagation algorithm in predicting missing stream flow data in Langat River Basin. Data series from the upper part of Langat River Basin were applied to build the Artificial Neural Network model. The result indicated good performance of the model with Pearson Correlation, r = 0.97261 for training data and overall data shows r = 0.91925. The study reveals that Levenberg-Marquadt back propagation algorithm for ANN can simulate well in the daily missing stream flow prediction if the model customizes with good configuration. © 2006-2015. Asian Research Publishing Network (ARPN). All rights reserved.