A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Optimal Control of Irrigation Canals Using Recurrent Dynamic Neural Network (RDNN)
A recurrent dynamic neural network (RDNN) based on Hopfield model was applied to linear quadratic regulator controller of an irrigation canal. The Saint-Venant equations of open-channel flow were linearized using the Taylor series and a finite difference approximation of the original nonlinear, partial differential equations. Using the linear optimal control theory, a Linear Quadratic Regulator (LQR) controller was developed for an irrigation canal with a single pool and results were observed. And then LQR controller was redesigned with applying Hopfield recurrent dynamic neural network (RDNN) to solution of Riccati equation for calculating the LQR control gains. The main advantage of Hopfield neural network model is its robustness and flexibility when dealing with unknown perturbations and parameters. The results of this study shows that an RDNN based LQR controller (RDNN/LQR) for irrigation canals offers an alternative to traditional LQR controller when dealing with the disturbances in the irrigation canals.
Optimal Control of Irrigation Canals Using Recurrent Dynamic Neural Network (RDNN)
A recurrent dynamic neural network (RDNN) based on Hopfield model was applied to linear quadratic regulator controller of an irrigation canal. The Saint-Venant equations of open-channel flow were linearized using the Taylor series and a finite difference approximation of the original nonlinear, partial differential equations. Using the linear optimal control theory, a Linear Quadratic Regulator (LQR) controller was developed for an irrigation canal with a single pool and results were observed. And then LQR controller was redesigned with applying Hopfield recurrent dynamic neural network (RDNN) to solution of Riccati equation for calculating the LQR control gains. The main advantage of Hopfield neural network model is its robustness and flexibility when dealing with unknown perturbations and parameters. The results of this study shows that an RDNN based LQR controller (RDNN/LQR) for irrigation canals offers an alternative to traditional LQR controller when dealing with the disturbances in the irrigation canals.
Optimal Control of Irrigation Canals Using Recurrent Dynamic Neural Network (RDNN)
Durdu, Omer F. (author)
World Water and Environmental Resources Congress 2004 ; 2004 ; Salt Lake City, Utah, United States
2004-06-25
Conference paper
Electronic Resource
English
"Pilote" Optimal Control of Irrigation Canals
| British Library Conference Proceedings | 1995
Stochastic Optimal and Suboptimal Control of Irrigation Canals
| British Library Online Contents | 1999
TECHNICAL PAPERS - Stochastic Optimal and Suboptimal Control of Irrigation Canals
| Online Contents | 1999
| Engineering Index Backfile | 1952
Pilote: Linear Quadratic Optimal Controller for Irrigation Canals
| British Library Online Contents | 1998