Machine Learning Application in Reservoir Water Level Forecasting for Sustainable Hydropower Generation Strategy
The aim of this study is to accurately forecast the changes in water level of a reservoir located in Malaysia with two different scenarios; Scenario 1 (SC1) includes rainfall and water level as input and Scenario 2 (SC2) includes rainfall, water level, and sent out. Different time horizons (one day...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-07-01
|
Series: | Sustainability |
Subjects: | |
Online Access: | https://www.mdpi.com/2071-1050/12/15/6121 |
id |
doaj-b7cc4e8645084871a29bc05a5d6eafa4 |
---|---|
record_format |
Article |
spelling |
doaj-b7cc4e8645084871a29bc05a5d6eafa42020-11-25T04:03:24ZengMDPI AGSustainability2071-10502020-07-01126121612110.3390/su12156121Machine Learning Application in Reservoir Water Level Forecasting for Sustainable Hydropower Generation StrategyMichelle Sapitang0Wanie M. Ridwan1Khairul Faizal Kushiar2Ali Najah Ahmed3Ahmed El-Shafie4Uniten R&D Sdn Bhd, Universiti Tenaga Nasional (UNITEN), Kajang 43000, MalaysiaUniten R&D Sdn Bhd, Universiti Tenaga Nasional (UNITEN), Kajang 43000, MalaysiaAsset Management Department, Generation Division, Tenaga Nasional Berhad, Kuala Lumpur 59200, MalaysiaInstitute of Energy Infrastructure (IEI), Universiti Tenaga Nasional (UNITEN), Kajang 43000, MalaysiaDepartment of Civil Engineering, Faculty of Engineering, University of Malaya (UM), Kuala Lumpur 50603, MalaysiaThe aim of this study is to accurately forecast the changes in water level of a reservoir located in Malaysia with two different scenarios; Scenario 1 (SC1) includes rainfall and water level as input and Scenario 2 (SC2) includes rainfall, water level, and sent out. Different time horizons (one day ahead to seven days) will be investigated to check the accuracy of the proposed models. In this study, four supervised machine learning algorithms for both scenarios were proposed such as Boosted Decision Tree Regression (BDTR), Decision Forest Regression (DFR), Bayesian Linear Regression (BLR) and Neural Network Regression (NNR). Eighty percent of the total data were used for training the datasets while 20% for the dataset used for testing. The models’ performance is evaluated using five statistical indexes; the Correlation Coefficient (R<sup>2</sup>), Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Relative Absolute Error (RAE), and Relative Squared Error (RSE). The findings showed that among the four proposed models, the BLR model outperformed other models with R<sup>2</sup> 0.998952 (1-day ahead) for SC1 and BDTR for SC2 with R<sup>2</sup> 0.99992 (1-day ahead). With regards to the uncertainty analysis, 95PPU and d-factors were adopted to measure the uncertainties of the best models (BLR and BDTR). The results showed the value of 95PPU for both models in both scenarios (SC1 and SC2) fall into the range between 80% to 100%. As for the d-factor, all values in SC1 and SC2 fall below one.https://www.mdpi.com/2071-1050/12/15/6121forecasting water levelmachine learning algorithmsboosted decision tree regressiondecision forest regressionneural network regressionBayesian linear regression |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Michelle Sapitang Wanie M. Ridwan Khairul Faizal Kushiar Ali Najah Ahmed Ahmed El-Shafie |
spellingShingle |
Michelle Sapitang Wanie M. Ridwan Khairul Faizal Kushiar Ali Najah Ahmed Ahmed El-Shafie Machine Learning Application in Reservoir Water Level Forecasting for Sustainable Hydropower Generation Strategy Sustainability forecasting water level machine learning algorithms boosted decision tree regression decision forest regression neural network regression Bayesian linear regression |
author_facet |
Michelle Sapitang Wanie M. Ridwan Khairul Faizal Kushiar Ali Najah Ahmed Ahmed El-Shafie |
author_sort |
Michelle Sapitang |
title |
Machine Learning Application in Reservoir Water Level Forecasting for Sustainable Hydropower Generation Strategy |
title_short |
Machine Learning Application in Reservoir Water Level Forecasting for Sustainable Hydropower Generation Strategy |
title_full |
Machine Learning Application in Reservoir Water Level Forecasting for Sustainable Hydropower Generation Strategy |
title_fullStr |
Machine Learning Application in Reservoir Water Level Forecasting for Sustainable Hydropower Generation Strategy |
title_full_unstemmed |
Machine Learning Application in Reservoir Water Level Forecasting for Sustainable Hydropower Generation Strategy |
title_sort |
machine learning application in reservoir water level forecasting for sustainable hydropower generation strategy |
publisher |
MDPI AG |
series |
Sustainability |
issn |
2071-1050 |
publishDate |
2020-07-01 |
description |
The aim of this study is to accurately forecast the changes in water level of a reservoir located in Malaysia with two different scenarios; Scenario 1 (SC1) includes rainfall and water level as input and Scenario 2 (SC2) includes rainfall, water level, and sent out. Different time horizons (one day ahead to seven days) will be investigated to check the accuracy of the proposed models. In this study, four supervised machine learning algorithms for both scenarios were proposed such as Boosted Decision Tree Regression (BDTR), Decision Forest Regression (DFR), Bayesian Linear Regression (BLR) and Neural Network Regression (NNR). Eighty percent of the total data were used for training the datasets while 20% for the dataset used for testing. The models’ performance is evaluated using five statistical indexes; the Correlation Coefficient (R<sup>2</sup>), Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Relative Absolute Error (RAE), and Relative Squared Error (RSE). The findings showed that among the four proposed models, the BLR model outperformed other models with R<sup>2</sup> 0.998952 (1-day ahead) for SC1 and BDTR for SC2 with R<sup>2</sup> 0.99992 (1-day ahead). With regards to the uncertainty analysis, 95PPU and d-factors were adopted to measure the uncertainties of the best models (BLR and BDTR). The results showed the value of 95PPU for both models in both scenarios (SC1 and SC2) fall into the range between 80% to 100%. As for the d-factor, all values in SC1 and SC2 fall below one. |
topic |
forecasting water level machine learning algorithms boosted decision tree regression decision forest regression neural network regression Bayesian linear regression |
url |
https://www.mdpi.com/2071-1050/12/15/6121 |
work_keys_str_mv |
AT michellesapitang machinelearningapplicationinreservoirwaterlevelforecastingforsustainablehydropowergenerationstrategy AT waniemridwan machinelearningapplicationinreservoirwaterlevelforecastingforsustainablehydropowergenerationstrategy AT khairulfaizalkushiar machinelearningapplicationinreservoirwaterlevelforecastingforsustainablehydropowergenerationstrategy AT alinajahahmed machinelearningapplicationinreservoirwaterlevelforecastingforsustainablehydropowergenerationstrategy AT ahmedelshafie machinelearningapplicationinreservoirwaterlevelforecastingforsustainablehydropowergenerationstrategy |
_version_ |
1724440249777520640 |