Research on the Improved Learning Algorithm of FNNC for the Control of Hydraulic Turbine Generating Units

The tradition control strategy does not have strong learning ability in control, so it is difficult to gain better control performance to control the complicated non-linear hydraulic turbine generating units system. The fuzzy neural network controller can make reasoning on-line based on fuzzy reasoning system and its fuzzy reasoning rules, member function and parameters can be trained on-line through neural network. So it is a good control strategy to control hydraulic turbine generating units. However, the learning speed of fuzzy neural network affects the control performance of hydraulic turbine generating units system in control. A quick improved learning method is presented to train fuzzy neural network. In training, learning parameters can change according to the control error, which solves the existent problem of the fuzzy neural network in on-line learning. Simulation experiment results show that the designed fuzzy neural network controller can control hydraulic turbine generating units efficaciously and has quick controlling speed and less controlling max-error. So it provides a good control and learning strategy for the control of hydraulic turbine generating units system.