Indonesian Journal of Electrical Engineering and Computer Science Vol. 14, No. 1, April 2019, pp. 443~449 ISSN: 2502-4752, DOI: 10.11591/ijeecs.v14.i1.pp443-449  443 Journal homepage: http://iaescore.com/journals/index.php/ijeecs Reservoir water level forecasting using normalization and multiple regression Siti Rafidah M-Dawam 1 , Ku Ruhana Ku-Mahamud 2 1 Faculty of Computer and Mathematical Sciences, Universiti Teknologi MARA Kedah, Malaysia 2 School of Computing, College of Arts and Sciences, Universiti Utara Malaysia, Sintok, Kedah, Malaysia Article Info ABSTRACT Article history: Received Oct 6, 2018 Revised Nov 19, 2018 Accepted Dec 27, 2018 Many non-parametric techniques such as Neural Network (NN) are used to forecast current reservoir water level (RWLt). However, modelling using these techniques can be established without knowledge of the mathematical relationship between the inputs and the corresponding outputs. Another important issue to be considered which is related to forecasting is the preprocessing stage where most non-parametric techniques normalize data into discretized data. Data normalization can influence the the results of forecasting. This paper presents reservoir water level (RWL) forecasting using normalization and multiple regression. In this study, continuous data of rainfall (RF) and changes of reservoir water level (WC) are normalized using two different normalization methods, Min-Max and Z-Score techniques. Its comparative studies and forecasting process are carried out using multiple regression. Three input scenarios for multiple regression were designed which comprise of temporal patterns of WC and RF, in which the sliding window technique has been applied. The experimental results showed that the best input scenario for forecasting the RWLt employs both the RF and the WC, in which the best predictors are three day’s delay of WC and two days’ delay of RF. The findings also suggested that the performance of the RWL forecasting model using multiple regression was dependent on the normalization methods. Keywords: Forecasting model Reservoir modelling Reservoir water release Sliding window Temporal data mining Copyright © 2019 Institute of Advanced Engineering and Science. All rights reserved. Corresponding Author: Siti Rafidah M-Dawam, Faculty of Computer and Mathematical Sciences, Universiti Teknologi MARA Kedah, P.O. Box 187, 08400 Merbok, Kedah, Malaysia. Email: srafidah192@kedah.uitm.edu.my 1. INTRODUCTION Forecasting RWL is crucial for reservoir’s operator in making decision on the reservoir water release (RWR) of a particular reservoir. It is a challenging and complex task, especially during flood and drought occurances due to unpredictable inflow such as RF [1]. Thus, a few researches have focused on non- structural approaches predicting reservoir inflows [2]. However, during flood or drought, the decision on RWR is not only based on the availability of water inflows, but also on the previous release, demands, time, etc. Besides daily RF, several researches also considered changes in the RWL (WC) as an input in the multipurpose reservoir forecasting model [2]. RF (hydrological data) and reservoir WC are found to be correlated in the flood prediction model [3]. Many literature conducted on the RWR operation have utilized RF data and RWL as inputs [4], and have applied different methods and techniques of Artificial Intelligence and machine learning[5–8]. Only a small number of researches conducted on RWR decisions highlighted on the time delay between the RF and the increase of RWL.