Adaptive and Robust Operation with Active Fuzzy Harvester under Nonstationary and Random Disturbance Conditions.

The objective of this paper is to amplify the output voltage magnitude from a piezoelectric vibration energy harvester under nonstationary and broadband vibration conditions. Improving the transferred energy, which is converted from mechanical energy to electrical energy through a piezoelectric transducer, achieved a high output voltage and effective harvesting. A threshold-based switching strategy is used to improve the total transferred energy with consideration of the signs and amplitudes of the electromechanical conditions of the harvester. A time-invariant threshold cannot accomplish effective harvesting under nonstationary vibration conditions because the assessment criterion for desirable control changes in accordance with the disturbance scale. To solve this problem, we developed a switching strategy for the active harvester, namely, adaptive switching considering vibration suppression-threshold strategy. The strategy adopts a tuning algorithm for the time-varying threshold and implements appropriate intermittent switching without pre-tuning by means of the fuzzy control theory. We evaluated the proposed strategy under three realistic vibration conditions: a frequency sweep, a change in the number of dominant frequencies, and wideband frequency vibration. Experimental comparisons were conducted with existing strategies, which consider only the signs of the harvester electromechanical conditions. The results confirm that the presented strategy achieves a greater output voltage than the existing strategies under all nonstationary vibration conditions. The average amplification rate of output voltage for the proposed strategy is 203% compared with the output voltage by noncontrolled harvesting.