A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Inflow Forecasting of Bhavanisagar Reservoir Using Artificial Neural Network (ANN): A Case Study
Hydrologic forecasting of inflows into a reservoir plays an important role in efficient reservoir management and control. Efficient reservoir operation and management rely on the proper forecast of the inflow into the reservoir and it leads to enhanced reservoir yields and better flood protection. But, most of the hydrological parameters are subjected to uncertainty. Hence, an appropriate forecasting method, a feedforward Artificial Neural Network (ANN) was used in this study to obtain reliable information of inflow into a reservoir. The ANN models were trained and simulated using MATLAB with raw and transformed data. Synthetic data and stochastic models are generated to obviate a lack of data and they are utilized to forecast inflow. A total of 24 years (1989–2013) of historical data in the form of average monthly inflow to Bhavanisagar reservoir was used to train, test and validate the model. Then, the results are compared with the observed values of the reservoir. Further, it was found that the Mean Square Error (MSE) obtained is within the range. Hence, this model is used to simulate the inflow for the period 2049–2064 (as per IPCC AR4 report). From the predicted values, appropriate storage and discharge from the reservoir can be decided to prevent the extreme crisis in the near future.
Inflow Forecasting of Bhavanisagar Reservoir Using Artificial Neural Network (ANN): A Case Study
Hydrologic forecasting of inflows into a reservoir plays an important role in efficient reservoir management and control. Efficient reservoir operation and management rely on the proper forecast of the inflow into the reservoir and it leads to enhanced reservoir yields and better flood protection. But, most of the hydrological parameters are subjected to uncertainty. Hence, an appropriate forecasting method, a feedforward Artificial Neural Network (ANN) was used in this study to obtain reliable information of inflow into a reservoir. The ANN models were trained and simulated using MATLAB with raw and transformed data. Synthetic data and stochastic models are generated to obviate a lack of data and they are utilized to forecast inflow. A total of 24 years (1989–2013) of historical data in the form of average monthly inflow to Bhavanisagar reservoir was used to train, test and validate the model. Then, the results are compared with the observed values of the reservoir. Further, it was found that the Mean Square Error (MSE) obtained is within the range. Hence, this model is used to simulate the inflow for the period 2049–2064 (as per IPCC AR4 report). From the predicted values, appropriate storage and discharge from the reservoir can be decided to prevent the extreme crisis in the near future.
Inflow Forecasting of Bhavanisagar Reservoir Using Artificial Neural Network (ANN): A Case Study
Lecture Notes in Civil Engineering
Ramanagopal, S. (editor) / Gali, Madhavi Latha (editor) / Venkataraman, Kartik (editor) / Suriya, S. (author) / Saran, K. (author) / Chris Anto, L. (author) / Anbalagan, C. (author) / Vinodh, K. R. (author)
2020-08-29
13 pages
Article/Chapter (Book)
Electronic Resource
English
Inflow forecasting , Bhavanisagar reservoir , ANN , Feedforward algorithm , MATLAB Engineering , Geoengineering, Foundations, Hydraulics , Sustainable Architecture/Green Buildings , Waste Water Technology / Water Pollution Control / Water Management / Aquatic Pollution , Building Construction and Design , Transportation Technology and Traffic Engineering
Reservoir Inflow Forecasting Using Neural Network Models
| British Library Conference Proceedings | 1998
Inflow Forecasting for Real-Time Reservoir Operation Using Artificial Neural Network
| British Library Conference Proceedings | 2009
Multivariate Reservoir Inflow Forecasting Using Temporal Neural Networks
| Online Contents | 2001
Multivariate Reservoir Inflow Forecasting Using Temporal Neural Networks
| British Library Online Contents | 2001