Wavelet-Based Time-Frequency Control of a Flywheel Energy Storage System

The objective of this paper was to implement a novel controller called “wavelet-based time-frequency control” (WFXLMS) in a computer simulation of a FES system with six degrees-of-freedom and compare its dynamic stability and active power consumption with the following conventional controllers: PID and fuzzy-logic. Specifically, all three controllers were applied to a FES system operating at 100,000 rpm, and the amplitude of vibration, rate of convergence, and current draw were compared. As of writing this paper, the PID and fuzzy-logic controllers have converged but the WFXLMS controller has not. The parameter values for the WFXLMS controller need further tweaking for a more comprehensive analysis. Despite this setback, both the fuzzy-logic and PID controllers did demonstrate convergence at 100,000 rpm. The fuzzy-logic converged immediately and the PID converged around 0.8 seconds. The PID demonstrated a periodic motion about the z axis while the fuzzy-logic settled at a constant displacement. Finally, the PID controller had a smaller maximum and average current draw over the fuzzy-logic. In conclusion, the PID controller provided sufficient control of the system with the least amount of current draw, but the fuzzy-logic controller provided the steadiest response.