Hybrid artificial intelligence based abc-pso system for ground water level forecasting in Udupi region

B. S. Supreetha, K. Prabhakara Nayak, K. Narayan Shenoy

Research output: Contribution to journalArticle

Abstract

The groundwater level modelling and forecasting have wide application for effective groundwater resources management. The traditional numerical groundwater level forecasting requires various hydrogeological parameters. The alternative approach for groundwater level forecasting is data-driven models. The ANN hybrid models are found to be more effective for predicting ground-water levels at different time domains. Soft computing based model is developed by considering historical groundwater level and rainfall data. We developed an innovative hybrid ABC algorithm based on PSO searching mechanism to carry out forecasts of future groundwater levels with the aid of earlier recorded groundwater levels and rainfall. The evaluation metrics of parameters such as RMSE, Error Variation Regression coefficient, and MAE have been used. The results obtained prove that hybrid soft computing technique is able to forecast the groundwater level over several years effectively. The model predicted trend followed the observed data closely (RMSE = 0.3928, R2 = 0.90029). The Mean Absolute Error and relative error of predicted results are 0.574 and 2.11% respectively. The ABC-PSO technique has shown promising results in accurate monthly groundwater level prediction vis-à-vis ANN methods.

Original languageEnglish
Pages (from-to)797-809
Number of pages13
JournalJournal of Engineering Science and Technology
Volume14
Issue number2
Publication statusPublished - 01-04-2019
Externally publishedYes

Fingerprint

Water levels
Artificial intelligence
Groundwater
Soft computing
Particle swarm optimization (PSO)
Rain
Groundwater resources

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Cite this

Supreetha, B. S. ; Prabhakara Nayak, K. ; Narayan Shenoy, K. / Hybrid artificial intelligence based abc-pso system for ground water level forecasting in Udupi region. In: Journal of Engineering Science and Technology. 2019 ; Vol. 14, No. 2. pp. 797-809.
@article{f526001d50f74039850cbab9837c2ddf,
title = "Hybrid artificial intelligence based abc-pso system for ground water level forecasting in Udupi region",
abstract = "The groundwater level modelling and forecasting have wide application for effective groundwater resources management. The traditional numerical groundwater level forecasting requires various hydrogeological parameters. The alternative approach for groundwater level forecasting is data-driven models. The ANN hybrid models are found to be more effective for predicting ground-water levels at different time domains. Soft computing based model is developed by considering historical groundwater level and rainfall data. We developed an innovative hybrid ABC algorithm based on PSO searching mechanism to carry out forecasts of future groundwater levels with the aid of earlier recorded groundwater levels and rainfall. The evaluation metrics of parameters such as RMSE, Error Variation Regression coefficient, and MAE have been used. The results obtained prove that hybrid soft computing technique is able to forecast the groundwater level over several years effectively. The model predicted trend followed the observed data closely (RMSE = 0.3928, R2 = 0.90029). The Mean Absolute Error and relative error of predicted results are 0.574 and 2.11{\%} respectively. The ABC-PSO technique has shown promising results in accurate monthly groundwater level prediction vis-{\`a}-vis ANN methods.",
author = "Supreetha, {B. S.} and {Prabhakara Nayak}, K. and {Narayan Shenoy}, K.",
year = "2019",
month = "4",
day = "1",
language = "English",
volume = "14",
pages = "797--809",
journal = "Journal of Engineering Science and Technology",
issn = "1823-4690",
publisher = "Taylor's University College",
number = "2",

}

Hybrid artificial intelligence based abc-pso system for ground water level forecasting in Udupi region. / Supreetha, B. S.; Prabhakara Nayak, K.; Narayan Shenoy, K.

In: Journal of Engineering Science and Technology, Vol. 14, No. 2, 01.04.2019, p. 797-809.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Hybrid artificial intelligence based abc-pso system for ground water level forecasting in Udupi region

AU - Supreetha, B. S.

AU - Prabhakara Nayak, K.

AU - Narayan Shenoy, K.

PY - 2019/4/1

Y1 - 2019/4/1

N2 - The groundwater level modelling and forecasting have wide application for effective groundwater resources management. The traditional numerical groundwater level forecasting requires various hydrogeological parameters. The alternative approach for groundwater level forecasting is data-driven models. The ANN hybrid models are found to be more effective for predicting ground-water levels at different time domains. Soft computing based model is developed by considering historical groundwater level and rainfall data. We developed an innovative hybrid ABC algorithm based on PSO searching mechanism to carry out forecasts of future groundwater levels with the aid of earlier recorded groundwater levels and rainfall. The evaluation metrics of parameters such as RMSE, Error Variation Regression coefficient, and MAE have been used. The results obtained prove that hybrid soft computing technique is able to forecast the groundwater level over several years effectively. The model predicted trend followed the observed data closely (RMSE = 0.3928, R2 = 0.90029). The Mean Absolute Error and relative error of predicted results are 0.574 and 2.11% respectively. The ABC-PSO technique has shown promising results in accurate monthly groundwater level prediction vis-à-vis ANN methods.

AB - The groundwater level modelling and forecasting have wide application for effective groundwater resources management. The traditional numerical groundwater level forecasting requires various hydrogeological parameters. The alternative approach for groundwater level forecasting is data-driven models. The ANN hybrid models are found to be more effective for predicting ground-water levels at different time domains. Soft computing based model is developed by considering historical groundwater level and rainfall data. We developed an innovative hybrid ABC algorithm based on PSO searching mechanism to carry out forecasts of future groundwater levels with the aid of earlier recorded groundwater levels and rainfall. The evaluation metrics of parameters such as RMSE, Error Variation Regression coefficient, and MAE have been used. The results obtained prove that hybrid soft computing technique is able to forecast the groundwater level over several years effectively. The model predicted trend followed the observed data closely (RMSE = 0.3928, R2 = 0.90029). The Mean Absolute Error and relative error of predicted results are 0.574 and 2.11% respectively. The ABC-PSO technique has shown promising results in accurate monthly groundwater level prediction vis-à-vis ANN methods.

UR - http://www.scopus.com/inward/record.url?scp=85071639666&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071639666&partnerID=8YFLogxK

M3 - Article

VL - 14

SP - 797

EP - 809

JO - Journal of Engineering Science and Technology

JF - Journal of Engineering Science and Technology

SN - 1823-4690

IS - 2

ER -