| 1/2005 - 2 |
Reservoir Inflow Forecast Using Neural Networks: A Case Study of Wangchu River of BhutanJigme SINGYE, Katsumi MASUGATA, Murai TADAKUNI |
Extra paper information in    |
Click to see author's profile in  SCOPUS,  IEEE Xplore,  Web of Science |
| Not available online | Views: 1,552 |
Author keywords
reservoir inflow, neural network, hydro power
References keywords
References keywords will be displayed on the next page reload.
About this article
Date of Publication: 2005-04-02
Volume 5, Issue 1, Year 2005, On page(s): 10 - 16
ISSN: 1582-7445, e-ISSN: 1844-7600
Digital Object Identifier: Not assigned
Abstract
Efficient river inflow forecast is essential for various purposes such as for flood control, distribution of water for irrigation purposes, etc. It is also a highly useful tool for the efficient operation and maintenance of hydro power plants as the economics of the hydro electricity depend largely on the reservoir height as well as the inflow rate into the dam. Through effective inflow prediction mechanism, proper maintenance and operation schedule of the generating machines can be planned, thereby reducing the forced outages and increasing the generation output. In this paper, a neural network based approach is presented to forecast the daily river inflow for a Wangchu River of Bhutan since on this river basin lie the two biggest hydro plants, Chukha Hydropower Corporation (CHPC) and Tala Hydro Project Authority (THPA), with generating capacities of 336 MW and 1020 MW, respectively. For a small Himalayan kingdom where half the total national revenue comes from the sale of hydro power alone, with almost around 80 percent exported to India, it is essential to have a proper mechanism to accurately forecast the inflow as it would help implement the optimal utilization of water resources as well as plan efficient load scheduling. The latter is particularly important for power export as prior electric generation anticipation based on river inflow is critical in transacting the power sale in advance, thereby significantly improving the revenue earned. In this paper, two types of neural networks are designed and their performances compared in predicting the next day inflow. Both the networks are extensively tested using the inflow and other weather related data from the year 1999 to 2003. |
| References | | | Cited By |
On-line references are not available - see the PDF file if available.
Copyright ©2001-2025
Faculty of Electrical Engineering and Computer Science
Stefan cel Mare University of Suceava, Romania
All rights reserved: Advances in Electrical and Computer Engineering is a registered trademark of the Stefan cel Mare University of Suceava. No part of this publication may be reproduced, stored in a retrieval system, photocopied, recorded or archived, without the written permission from the Editor. When authors submit their papers for publication, they agree that the copyright for their article be transferred to the Faculty of Electrical Engineering and Computer Science, Stefan cel Mare University of Suceava, Romania, if and only if the articles are accepted for publication. The copyright covers the exclusive rights to reproduce and distribute the article, including reprints and translations.
Permission for other use: The copyright owner's consent does not extend to copying for general distribution, for promotion, for creating new works, or for resale. Specific written permission must be obtained from the Editor for such copying. Direct linking to files hosted on this website is strictly prohibited.
Disclaimer: Whilst every effort is made by the publishers and editorial board to see that no inaccurate or misleading data, opinions or statements appear in this journal, they wish to make it clear that all information and opinions formulated in the articles, as well as linguistic accuracy, are the sole responsibility of the author.
Faculty of Electrical Engineering and Computer Science
Stefan cel Mare University of Suceava, Romania
All rights reserved: Advances in Electrical and Computer Engineering is a registered trademark of the Stefan cel Mare University of Suceava. No part of this publication may be reproduced, stored in a retrieval system, photocopied, recorded or archived, without the written permission from the Editor. When authors submit their papers for publication, they agree that the copyright for their article be transferred to the Faculty of Electrical Engineering and Computer Science, Stefan cel Mare University of Suceava, Romania, if and only if the articles are accepted for publication. The copyright covers the exclusive rights to reproduce and distribute the article, including reprints and translations.
Permission for other use: The copyright owner's consent does not extend to copying for general distribution, for promotion, for creating new works, or for resale. Specific written permission must be obtained from the Editor for such copying. Direct linking to files hosted on this website is strictly prohibited.
Disclaimer: Whilst every effort is made by the publishers and editorial board to see that no inaccurate or misleading data, opinions or statements appear in this journal, they wish to make it clear that all information and opinions formulated in the articles, as well as linguistic accuracy, are the sole responsibility of the author.
