Apr 10, 2025
5:00pm - 7:00pm
Summit, Level 2, Flex Hall C
Jin Pyo Lee1,Xinran Zhou2,Yangyang Xin1,Dace Gao3,Peiwen Huang1,Pooi See Lee1
Nanyang Technological University1,Donghua University2,Linköping University3
Jin Pyo Lee1,Xinran Zhou2,Yangyang Xin1,Dace Gao3,Peiwen Huang1,Pooi See Lee1
Nanyang Technological University1,Donghua University2,Linköping University3
The fast development of Internet of Things (IoT) devices requires a compact, sustainable, and ubiquitous power source as alternative solution to traditional chemical batteries. For this, electromagnetic generator (EMG) based on electromagnetic induction is considered as a promising candidate due to its distinct advantages such as high reliability, high power density, long lifetime, and wide applicability to extreme environments. Recently, the EMG have been developed with magnetic levitation structure and magnetic flux concentrator (MFC) methods to highly increase the electrical output performance by introducing magnetic repulsive force and concentrating magnetic field for boosting electromagnetic induction, respectively. However, as the EMG device is smaller in dimension, the undesirable interaction between magnets and insufficient area severely limit the application of the MFC. Here, we devised high-performance millimeter scale electromagnetic generator (mmEMG) via highly concentrating magnetic flux with the MFC films applied on both ends of coils. We investigated the effect of the MFC depending on various position and structural parameters by simulation, confirmed with experimental results. Also, we studied influence of magnetic properties on the MFC by characterization with different soft magnetic materials. With this process, we optimized the mmEMG with MFC device showing 1.78 mW, 5.6-fold improvement in electrical output compared to without MFC. We successfully demonstrated charging process of a commercial electronics as portable power source and constructed self-powered force feedback-based gripper control system by integrating our devices for gripping force sensing with soft gripper connected to air pressure controllable system assisted by machine learning, which enabled us to intelligently control the gripping force without need of physical testing.