The goal of this work was to investigate using harvested energy to directly control the vibration response of flexible aerospace systems. Small, lightweight, flexible Micro Air Vehicles (MAVs) operate near flutter, providing both harvesting opportunities and vibration suppression requirements. The possibility that ambient energy might be harnessed and recycled to provide energy to mitigate the vibrations through various control laws was investigated. The goal was to integrate harvesting, storage, control, and computation into one multifunctional structure, and illustrate its benefits.
The first task was to discover ways to minimize control effort for vibration suppression. Basic control laws were tuned to achieve the same performance. The required amount of energy in each case was calculated and compared. A saturation function was instituted over the top of each controller to limit the amount of energy called for in the early part of the control law. These bang-bang, or saturation, controllers clearly used the least amount of energy to produce the same performance. As much as two-thirds of the required energy can be saved by using a saturation control. This reduction makes running a control law off of harvested energy possible.
In implementing these control laws, it was discovered that the high voltages commanded by the control laws result in the piezoelectric coupling coefficient being non-constant. An adaptive control law had to be implemented to account for the change in coupling coefficient as the control voltage demand increased.
The next task was to integrate harvesting and storage into the same package with a control actuator and a control law (i.e. the circuitry) embedded in a multifunctional composite structure. The goal was to integrate all of these components in order to provide a multifunctional system capable of the following functions:
- Energy harvesting
- Energy storage
- Vibration suppression using active control
- Embedded computing (providing energy management and control laws)
- Structural integrity
This was all fabricated, modeled, and tested. Before proceeding, the harvesting, sensing, and control authority of several different types of piezoelectric material were considered, in order to choose the best components for each task. Macro fiber composites form the best control actuation devices, and monolithic piezoceramic forms the best sensing and harvesting device. These results were validated with extensive experiments.
The concept of a multifunctional composite beam was applied to a problem prevalent in UAVs: they tend to be light and travel near their flutter speed, which means that they are susceptible to instabilities caused by gusts. While the UAV is in normal flight, its wing vibrates. The multifunctional wing spar would transfer the wing vibration into electrical energy and store it in the embedded battery. When the UAV hits a gust, the sensor function of the multifunctional spar would then see the increased strain and turn on the active control system embedded in the PCB part of the spar. The resulting feedback control law would then quiet the gust response and keep the vibration suppressed during the period of the gust. Laboratory results show great agreement with the theoretical models and numerical simulations.
Two different controllers are used. A positive position feedback controller (basically a second order filter) and the reduced energy controller illustrate that the settling time is about the same, while the energy consumed is much less.
With validation of the model, simulations were used to predict how the system would behave as a gust suppression system for a small UAV. The gust and clear sky condition (the condition of vibration induced during normal flight) were simulated using the Dryden PSD signal for both clear sky and gust. The simulations were fed into the model of the multifunctional wing spar. The response of the wing to a gust shows a large tip deflection. The response of the wing tip with the controller turned on and the gust as input shows substantial vibration reduction.
Other results that spun off of the proposed research include a MEMs-based energy-harvesting device, the use of nonlinearity to improve the amount of energy captured by improving the mechanical efficiency, and a look at harvesting impacts. The main contribution here is to show that closed-loop control can be accomplished with harvested energy.
This work was done by Daniel J. Inman and Pablo Tarazaga of Virginia Tech for the Air Force Office of Scientific Research. AFOSR-0008