Adaptive neuro-fuzzy interface system based performance monitoring technique for hydropower plants
Kumar, Krishna LU
and Saini, R. P.
(2023)
In ISH Journal of Hydraulic Engineering
29(5).
p.611-621
- Abstract
Energy has played a significant role in developing civilization, but the continuous use of fossil fuels has hampered the environment. Hydropower is the alternative to fossil fuels. But most of the hydropower plants in hilly areas suffer from silt erosion problems. Erosion of underwater parts creates vibration and noise and reduces machine efficiency. Therefore, online monitoring of turbines and other equipment is necessary to minimize losses due to erosion and part-load operation. Various studies are reported in the literature and found that correlation-based machine efficiency monitoring is one of the popular techniques. ANN method is useful for system modeling with a wide range of applications. However, despite the excellent... (More)
Energy has played a significant role in developing civilization, but the continuous use of fossil fuels has hampered the environment. Hydropower is the alternative to fossil fuels. But most of the hydropower plants in hilly areas suffer from silt erosion problems. Erosion of underwater parts creates vibration and noise and reduces machine efficiency. Therefore, online monitoring of turbines and other equipment is necessary to minimize losses due to erosion and part-load operation. Various studies are reported in the literature and found that correlation-based machine efficiency monitoring is one of the popular techniques. ANN method is useful for system modeling with a wide range of applications. However, despite the excellent classification capacities, its development can be time-consuming, computer-intensive, and prone to overfitting. In this paper, an Adaptive Neuro-Fuzzy Interface System (ANFIS) has been utilized to develop a correlation that removes the drawbacks of ANN and can predict the efficiency of the machine with an R2-value of 0. 99,976 having a Mean Absolute Percentage Error (MAPE) of 0.0108% at 0.06482% Root Mean Square Percentage Error (RMSPE).
(Less)
- author
- Kumar, Krishna
LU
and Saini, R. P.
- publishing date
- 2023
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- anfis, artificial intelligence, Hydropower, monitoring, turbine performance
- in
- ISH Journal of Hydraulic Engineering
- volume
- 29
- issue
- 5
- pages
- 11 pages
- publisher
- Taylor & Francis
- external identifiers
-
- scopus:85136866834
- ISSN
- 0971-5010
- DOI
- 10.1080/09715010.2022.2115320
- language
- English
- LU publication?
- no
- additional info
- Publisher Copyright: © 2022 Indian Society for Hydraulics.
- id
- 183666a0-dc9f-428b-8eec-3ff29e556099
- date added to LUP
- 2024-04-15 13:08:10
- date last changed
- 2024-05-16 14:46:48
@article{183666a0-dc9f-428b-8eec-3ff29e556099,
abstract = {{<p>Energy has played a significant role in developing civilization, but the continuous use of fossil fuels has hampered the environment. Hydropower is the alternative to fossil fuels. But most of the hydropower plants in hilly areas suffer from silt erosion problems. Erosion of underwater parts creates vibration and noise and reduces machine efficiency. Therefore, online monitoring of turbines and other equipment is necessary to minimize losses due to erosion and part-load operation. Various studies are reported in the literature and found that correlation-based machine efficiency monitoring is one of the popular techniques. ANN method is useful for system modeling with a wide range of applications. However, despite the excellent classification capacities, its development can be time-consuming, computer-intensive, and prone to overfitting. In this paper, an Adaptive Neuro-Fuzzy Interface System (ANFIS) has been utilized to develop a correlation that removes the drawbacks of ANN and can predict the efficiency of the machine with an R<sup>2</sup>-value of 0. 99,976 having a Mean Absolute Percentage Error (MAPE) of 0.0108% at 0.06482% Root Mean Square Percentage Error (RMSPE).</p>}},
author = {{Kumar, Krishna and Saini, R. P.}},
issn = {{0971-5010}},
keywords = {{anfis; artificial intelligence; Hydropower; monitoring; turbine performance}},
language = {{eng}},
number = {{5}},
pages = {{611--621}},
publisher = {{Taylor & Francis}},
series = {{ISH Journal of Hydraulic Engineering}},
title = {{Adaptive neuro-fuzzy interface system based performance monitoring technique for hydropower plants}},
url = {{http://dx.doi.org/10.1080/09715010.2022.2115320}},
doi = {{10.1080/09715010.2022.2115320}},
volume = {{29}},
year = {{2023}},
}