Study On Power Conversion Controller For Flywheel Energy Storage System Based On Artificial Intelligence Methods

In today’s society, electricity energy has become one of the indispensable secondary energy for the developments of our society and economy. With today’s growing fossil energy shortage, the storage and conversion of electricity energy has become the focus of the people’s growing attention and hot spots of research in a great many fields. Thereinto, the flywheel energy storage system shows a very good application prospects in many fields such as aerospace, transportation, electric power, etc, because of its higher energy storage density, shorter charging time, stronger anti-overcharge and anti-overdischarge ability, higher energy conversion efficiency, less environmental requirements and so on than others.The power conversion controlle system for the flywheel energy storage system is studied mainly in this dissertation. The main contents involved are dealed with including the transformation of the stationary coordinate system to the rotating coordinate system, the mathematical modeling of the permanent magnet synchronous motor in dq coordinate system, analysis and research of power converter, design of controller, etc. Firstly, the permanent magnet synchronous motor is selected to be used as charging motor, and then the mathematical model of the motor is established in dq coordinate with combination of the working principle of the permanent magnet synchronous motor and coordinate transformation theory. The topology, the basic working principles and the modulation strategy of the matrix converter used in the power conversion are analyzed in detail. Dual-space vector modulation method of matrix converter is introduced especially. The logic relationship between input and output is achieved by deducing and analyzing its switch states, therefore the simulation model of matrix converter is established. The simulation results show that the established matrix converter has good characteristics due to fine output voltage waveforms and less harmonic content of the input phase current waveform.On the basis of the studying and analyzing traditional double closed-loop speed control system, the fuzzy PID controler and BP neural network PID controller are designed respectively with integration of general designing methods for speed controller according to the basic theory of the fuzzy control and neural network control. The control for flywheel energy storage system is simulated using above two controllers under the toolbox environment of MATLAB/Simulink, and meanwhile all parameters of two PID controllers are setted effectively. Simulation results show that the application of Fuzzy Control or ANN Control can make the whole flywheel energy storage system obtain better waveforms of output valtage, current, torque and speed than traditional method. Aimming at the shortcomings of above two controllers, Fuzzy Neutral Network(FNN) controller is used as the speed-loop controller of the flywheel energy storage system with acquisition of great better stability, rapider response speed, higher steady-state precision, and it is verified by simulation that the FNN controller has great advantages over fuzzy PID controller or BP neural network PID controller, therefore the theory basis is founded for the application of the flywheel energy storage system in practice.