Assessment of machine learning techniques for monthly flow prediction

Zahra Alizadeh, Jafar Yazdi, Joong Hoon Kim, Abobakr Khalil Al-Shamiri

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Monthly flow predictions provide an essential basis for efficient decision-making regarding water resource allocation. In this paper, the performance of different popular data-driven models for monthly flow prediction is assessed to detect the appropriate model. The considered methods include feedforward neural networks (FFNNs), time delay neural networks (TDNNs), radial basis neural networks (RBFNNs), recurrent neural network (RNN), a grasshopper optimization algorithm (GOA)-based support vector machine (SVM) and K-nearest neighbors (KNN) model. For this purpose, the performance of each model is evaluated in terms of several residual metrics using a monthly flow time series for two real case studies with different flow regimes. The results show that the KNN outperforms the different neural network configurations for the first case study, whereas RBFNN model has better performance for the second case study in terms of the correlation coefficient. According to the accuracy of the results, in the first case study with more input features, the KNN model is recommended for short-term predictions and for the second case with a smaller number of input features, but more training observations, the RBFNN model is suitable.

Original languageEnglish
Article number1676
JournalWater (Switzerland)
Volume10
Issue number11
DOIs
Publication statusPublished - 2018 Nov 17

Fingerprint

artificial intelligence
Learning systems
neural network
neural networks
Water Resources
Grasshoppers
prediction
Resource Allocation
learning
Decision Making
case studies
methodology
Neural networks
performance
Recurrent neural networks
grasshopper
Feedforward neural networks
resource allocation
grasshoppers
Machine Learning

Keywords

  • Gaussian process regression
  • Grasshopper optimization algorithm
  • K-nearest neighbor regression
  • Neural network
  • Support vector machine

ASJC Scopus subject areas

  • Biochemistry
  • Geography, Planning and Development
  • Aquatic Science
  • Water Science and Technology

Cite this

Assessment of machine learning techniques for monthly flow prediction. / Alizadeh, Zahra; Yazdi, Jafar; Kim, Joong Hoon; Al-Shamiri, Abobakr Khalil.

In: Water (Switzerland), Vol. 10, No. 11, 1676, 17.11.2018.

Research output: Contribution to journalArticle

Alizadeh, Zahra ; Yazdi, Jafar ; Kim, Joong Hoon ; Al-Shamiri, Abobakr Khalil. / Assessment of machine learning techniques for monthly flow prediction. In: Water (Switzerland). 2018 ; Vol. 10, No. 11.
@article{a770d725e8f54234b2beb11ffd4adae1,
title = "Assessment of machine learning techniques for monthly flow prediction",
abstract = "Monthly flow predictions provide an essential basis for efficient decision-making regarding water resource allocation. In this paper, the performance of different popular data-driven models for monthly flow prediction is assessed to detect the appropriate model. The considered methods include feedforward neural networks (FFNNs), time delay neural networks (TDNNs), radial basis neural networks (RBFNNs), recurrent neural network (RNN), a grasshopper optimization algorithm (GOA)-based support vector machine (SVM) and K-nearest neighbors (KNN) model. For this purpose, the performance of each model is evaluated in terms of several residual metrics using a monthly flow time series for two real case studies with different flow regimes. The results show that the KNN outperforms the different neural network configurations for the first case study, whereas RBFNN model has better performance for the second case study in terms of the correlation coefficient. According to the accuracy of the results, in the first case study with more input features, the KNN model is recommended for short-term predictions and for the second case with a smaller number of input features, but more training observations, the RBFNN model is suitable.",
keywords = "Gaussian process regression, Grasshopper optimization algorithm, K-nearest neighbor regression, Neural network, Support vector machine",
author = "Zahra Alizadeh and Jafar Yazdi and Kim, {Joong Hoon} and Al-Shamiri, {Abobakr Khalil}",
year = "2018",
month = "11",
day = "17",
doi = "10.3390/w10111676",
language = "English",
volume = "10",
journal = "Water (Switzerland)",
issn = "2073-4441",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "11",

}

TY - JOUR

T1 - Assessment of machine learning techniques for monthly flow prediction

AU - Alizadeh, Zahra

AU - Yazdi, Jafar

AU - Kim, Joong Hoon

AU - Al-Shamiri, Abobakr Khalil

PY - 2018/11/17

Y1 - 2018/11/17

N2 - Monthly flow predictions provide an essential basis for efficient decision-making regarding water resource allocation. In this paper, the performance of different popular data-driven models for monthly flow prediction is assessed to detect the appropriate model. The considered methods include feedforward neural networks (FFNNs), time delay neural networks (TDNNs), radial basis neural networks (RBFNNs), recurrent neural network (RNN), a grasshopper optimization algorithm (GOA)-based support vector machine (SVM) and K-nearest neighbors (KNN) model. For this purpose, the performance of each model is evaluated in terms of several residual metrics using a monthly flow time series for two real case studies with different flow regimes. The results show that the KNN outperforms the different neural network configurations for the first case study, whereas RBFNN model has better performance for the second case study in terms of the correlation coefficient. According to the accuracy of the results, in the first case study with more input features, the KNN model is recommended for short-term predictions and for the second case with a smaller number of input features, but more training observations, the RBFNN model is suitable.

AB - Monthly flow predictions provide an essential basis for efficient decision-making regarding water resource allocation. In this paper, the performance of different popular data-driven models for monthly flow prediction is assessed to detect the appropriate model. The considered methods include feedforward neural networks (FFNNs), time delay neural networks (TDNNs), radial basis neural networks (RBFNNs), recurrent neural network (RNN), a grasshopper optimization algorithm (GOA)-based support vector machine (SVM) and K-nearest neighbors (KNN) model. For this purpose, the performance of each model is evaluated in terms of several residual metrics using a monthly flow time series for two real case studies with different flow regimes. The results show that the KNN outperforms the different neural network configurations for the first case study, whereas RBFNN model has better performance for the second case study in terms of the correlation coefficient. According to the accuracy of the results, in the first case study with more input features, the KNN model is recommended for short-term predictions and for the second case with a smaller number of input features, but more training observations, the RBFNN model is suitable.

KW - Gaussian process regression

KW - Grasshopper optimization algorithm

KW - K-nearest neighbor regression

KW - Neural network

KW - Support vector machine

UR - http://www.scopus.com/inward/record.url?scp=85056799267&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85056799267&partnerID=8YFLogxK

U2 - 10.3390/w10111676

DO - 10.3390/w10111676

M3 - Article

AN - SCOPUS:85056799267

VL - 10

JO - Water (Switzerland)

JF - Water (Switzerland)

SN - 2073-4441

IS - 11

M1 - 1676

ER -