Recently, vibration energy harvesting from surrounding environments to power wearable devices and wireless sensors in structure health monitoring has received considerable interest. Piezoelectric conversion mechanism has been employed to develop many successful energy harvesting devices due to its simple structure, long life span, high harvesting efficiency and so on. However, there are many difficulties of microscale cantilever configurations in energy harvesting from low frequency ambient. In order to improve the adaptability of energy harvesting from ambient vibrations, a two degrees of freedom (2-DOF) magnetic-coupled piezoelectric energy harvester is proposed in this paper. The electromechanical governing models of the cantilever and clamped hybrid energy harvester are derived to describe the dynamic characteristics for 2-DOF magnetic-coupled piezoelectric vibration energy harvester. Numerical simulations based on Matlab and ANSYS software show that the proposed magnetically coupled energy harvester can enhance the effective operating frequency bandwidth and increase the energy density. The experimental voltage responses of 2-DOF harvester under different structure parameters are acquired to demonstrate the effectiveness of the lumped parameter model for low frequency excitations. Moreover, the proposed energy harvester can enhance the energy harvesting performance over a wider bandwidth of low frequencies and has a great potential for broadband vibration energy harvesting.
There are numerous instruments and an abundance of complex information in the traditional cockpit display-control system, and pilots require a long time to familiarize themselves with the cockpit interface. This can cause accidents when they cope with emergency events, suggesting that it is necessary to evaluate pilot cognitive workload. In order to establish a simplified method to evaluate cognitive workload under a multitask condition. We designed a series of experiments involving different instrument panels and collected electroencephalograms (EEG) from 10 healthy volunteers. The data were classified and analyzed with an approximate entropy (ApEn) signal processing. ApEn increased with increasing experiment difficulty, suggesting that it can be used to evaluate cognitive workload. Our results demonstrate that ApEn can be used as an evaluation criteria of cognitive workload and has good specificity and sensitivity. Moreover, we determined an empirical formula to assess the cognitive workload interval, which can simplify cognitive workload evaluation under multitask conditions.