Predictive Control Strategy of Hydraulic Turbine Turning System Based on BGNN Neural Network
A model predictive control (MPC) strategy based on a novel Bayesian-Gaussian neural network (BGNN) model was proposed for the controller design of hydraulic turbine in this paper. The BGNN was used to learn the nonlinear dynamic model of controlled hydraulic turbine on-line as the predictive model for the design of MPC controller. Experiments show that the proposed nonlinear MPC strategy based on BGNN performs much better than the conventional PID controller.
KeywordsHydraulic Turbine BGNN Predictive Control neural network
Unable to display preview. Download preview PDF.
- 1.Jiang, C.: Nonlinear Simulation of Hydro turbine Governing System Based on Neural Network. In: IEEE International Conference on System, Man and Cybernetics, pp. 784–787 (1996)Google Scholar
- 2.Xu, F., Li, Z.: Computer Simulation about Hydraulic Generator Set. Hydraulic and Electric Power Press, Beijing (1998)Google Scholar
- 8.Jiang, C., Xiao, Z., Wang, S.: Neural Network Predict Control for the Hydro Turbine Generator Set. In: The Second International Conference on Machine Learning and Cybernetics, pp. 2–5 (2003)Google Scholar