Forecasting hydrological parameters for reservoir system utilizing artificial intelligent models and exploring their influence on operation performance
Obtaining successful operation rules for dam and reservoir systems is crucial for improving water management to meet the increase in agricultural, domestic and industrial activities. Several research efforts have been developed to generate optimal operation rules for dam and reservoir systems utiliz...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Published: |
Elsevier
2019
|
| Subjects: | |
| Online Access: | http://eprints.um.edu.my/20126/ https://doi.org/10.1016/j.knosys.2018.10.013 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Obtaining successful operation rules for dam and reservoir systems is crucial for improving water management to meet the increase in agricultural, domestic and industrial activities. Several research efforts have been developed to generate optimal operation rules for dam and reservoir systems utilizing different optimization algorithms. The main purpose of an operation rule is to minimize the gap between water supply and water demand patterns. To examine the optimized model performance, the simulation of a dam and reservoir system is usually carried out for a particular period utilizing the generated operation rule. During the simulation procedure, although reservoir inflow and evaporation are stochastic variables that are required to be forecasted during simulation, they are considered deterministic variables. This study attempts to integrate a forecasting model for reservoir inflow and evaporation with the operation rules generated from optimization models during the simulation procedure. The present study employs several optimization models to generate an optimal operation rule and two different forecasting models for reservoir inflow and reservoir evaporation. The three different optimization algorithms used in this study are the genetic algorithm (GA), particle swarm optimization (PSO) algorithm and shark machine learning algorithm (SMLA). Two different forecasting models have been developed for reservoir inflow and evaporation using the radial basis function neural network (RBF-NN) and support vector regression (SVR). It is necessary to analyze the proposed simulation procedure for examining the operation rule to comprehend the analysis under different optimal operation rules and levels of accuracy for both hydrological variables. The suggested models have been applied to generate optimal operation policies and reservoir inflow and evaporation forecasts for the Timah Tasoh dam (TTD) located in Malaysia. The results show that the major findings regarding the model performance during the simulation period indicate the necessity to pay attention to evaluating the optimized model performance by considering the results of the forecasting model for both the hydrological variables of reservoir inflow and reservoir evaporation rather than the deterministic values. |
|---|