Seminar Series - Dr. Soobum Lee
Piezoelectric Energy Harvester and Sensor Network Design
Piezoelectric Energy Harvester and Sensor Network Design using Advanced Optimization Methodologies
Abstract
This talk presents how design optimization methodologies can contribute to energy-sustainable and reliable engineering systems, with the special focus on energy harvesters and sensor network. Energy harvesting technologies are attracting great attention because they can realize self-powered wireless sensors and small electronics that do not require battery replacement or recharging, by utilizing ambient energy sources such as wind, vibration, sunlight, and heat. This presentation focuses on ambient vibration energy converted into electrical energy using piezoelectricity. The design challenges from the piezoelectric energy harvester and sensor systems include: random/stochastic nature of ambient vibration; insufficient energy conversion efficiency; and durability. In order to address them, this talk will introduce advanced design methods including: (i) reliability-based design optimization for vibration energy harvester in tire pressure monitoring sensors, (ii) highly efficient motion-driven vibration harvester for wind turbine blade monitoring, and (iii) a new piezoelectric sensor network design method for structural health monitoring.
Wasted energy in a vehicle’s rotating tire has a great potential to enable self-powered tire pressure monitoring sensors (TPMS) by collecting tire vibration energy. Due to the presence of harsh acceleration in a rotating tire, a design tradeoff needs to be studied to prolong the harvester’s fatigue life and ensure sufficient power generation. In this study, a new design formulation is addressed to consider the uncertainty in car speed, harvester dimension, and material properties, and solve this design problem using reliability-based design optimization (RBDO). The RBDO design result shows higher structural reliability than the case without consideration of uncertainty, and its performance is experimentally verified.
The second topic is about the development of a practical motion-driven vibration energy harvester named ActiveCharge™” that can effectively harvest wasted kinematic energy. The target application of this motion-driven harvester is, but is not limited to, powering the wireless sensors for wind turbine blade monitoring. The energy sustainability of wireless sensors for the wind turbine blades has been an important issue – disposable batteries are not an option because they have a limited lifetime (less than 5 years), and it is almost impossible to access the inside of a wind turbine blade to replace them. The study builds a working prototype of ActiveCharge™ with an integrated power management circuit and a wireless sensor for monitoring wind turbine blades.
The third topic introduces design optimization for piezoelectric sensor network considering the detectability of multiple failure modes. Sensor network design without considering uncertainty from system and environment may lead to inaccurate failure detection and untimely maintenance. The proposed method diagnoses failure using a probabilistic sensing model and the Mahalanobis Distance measure, while considering the uncertainty from structure properties and operation condition. The optimal piezoelectric sensor network design is prototyped and its failure detection capability is experimentally verified.
Pioneering use of advanced design optimization methods in piezoelectric energy harvester and sensor network design will enable intelligent engineering systems that autonomously generate necessary power and monitor their state, and promise far broader impact on structural and bio health monitoring technologies.
About the Speaker
Dr. Soobum Lee obtained his Ph.D. in Mechanical Engineering from Korea Advanced Institute of Science and Technology (KAIST) in 2007. He has worked at Korea Atomic Energy Research Institute (KAERI) and the University of Maryland at College Park (UMCP) as a postdoc researcher, the University of Notre Dame as a research assistant professor, and currently he is an assistant professor at the University of Maryland, Baltimore County (UMBC). His main research interests include energy harvester design, topology optimization, robust design, and reliability based design optimization. He is the author of about 90 international publications. He is currently the principal investigator of Energy Harvesting & Design Optimization Lab (EDLab) in UMBC.
Light refreshments will be served.
->-> Please RSVP below by clicking on GOING: