Ultralow-Frequency Biomechanical Energy Scavenging and Human Activity Recognition at Different Positions Using a Multifunctional Wearable Energy Harvester

被引:1
|
作者
Fan, Shuyu [1 ]
Fu, Mengyao [1 ]
Zhou, Yushan [1 ]
Hou, Dibo [1 ]
Zhang, Guangxin [1 ]
Cao, Yunqi [1 ]
机构
[1] Zhejiang Univ, Coll Control Sci & Engn, State Key Lab Ind Control Technol, Hangzhou 310027, Peoples R China
基金
中国国家自然科学基金;
关键词
Vibrations; Coils; Human activity recognition; Biomechanics; Rotors; Energy harvesting; Magnets; Biomechanical energy transduction; human activity recognition (HAR); human-motion adaptability; multifunctional electromagnetic energy harvester; ultralow-frequency vibrations; wearable devices; HUMAN MOTION; SENSOR; VIBRATIONS; DESIGN;
D O I
10.1109/TIM.2024.3406823
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
As a variety of wearable electronics with different functions have emerged, multifunctional devices with self-powered nature are desired for solving both problems of the tighter on-body wearing space and energy budget. Herein, we propose an eccentric-rotor-based multifunctional wearable electromagnetic vibration energy harvester (EMVEH) to function as an ultralow-frequency biomechanical energy scavenging device and a self-powered motion sensor for human activity recognition (HAR) at different positions, respectively. Kinetic simulations based on Euler-Lagrange equations provide a detailed understanding of its vibration pickup performance for effective energy scavenging and human-motion sensing. Results of comprehensive bench tests further verify its adaptability to excitations with ultralow frequencies (<5 Hz) and different directions, and the capability of mechanical excitation perception. Kinetic energy in limb motions during walking and running has been scavenged by a fabricated prototype into electric energy up to 221.49 mu W on average, whereas variations in multiple-feature-based profiles of output voltages in response to these human activities reflect repeatable and human-activity-related voltage patterns. On this basis, walking at 2-4 km/h and running at 4-8 km/h on a treadmill are recognized by these features and trained random forest (RF) classification models when the prototype is worn at the wrist, elbow, and ankle, respectively, all achieving high accuracies over 90%. Mutual constraints between dual design purposes of this multifunctional EMVEH are also fully discussed and released. Thus, the wearing-position-independent HAR performance combined with the efficient energy scavenging capability of our EMVEH expands the application range of wearable EMVEHs to more flexible and practical uses.
引用
收藏
页码:1 / 14
页数:14
相关论文
共 30 条
  • [21] Human Activity Recognition by Using Different Deep Learning Approaches for Wearable Sensors
    Erdas, Cagatay Berke
    Guney, Selda
    NEURAL PROCESSING LETTERS, 2021, 53 (03) : 1795 - 1809
  • [22] FusedAR: Energy-Positive Human Activity Recognition Using Kinetic and Solar Signal Fusion
    Sandhu, Muhammad Moid
    Khalifa, Sara
    Geissdoerfer, Kai
    Jurdak, Raja
    Portmann, Marius
    Kusy, Brano
    IEEE SENSORS JOURNAL, 2023, 23 (11) : 12411 - 12426
  • [23] An Ultra-Low Energy Human Activity Recognition Accelerator for Wearable Health Applications
    Bhat, Ganapati
    Tuncel, Yigit
    An, Sizhe
    Lee, Hyung Gyu
    Ogras, Umit Y.
    ACM TRANSACTIONS ON EMBEDDED COMPUTING SYSTEMS, 2019, 18 (05)
  • [24] An electromagnetic wearable 3-DoF resonance human body motion energy harvester using ferrofluid as a lubricant
    Wu, Shuai
    Luk, P. C. K.
    Li, Chunfang
    Zhao, Xiangyu
    Jiao, Zongxia
    Shang, Yaoxing
    APPLIED ENERGY, 2017, 197 : 364 - 374
  • [25] Model-Based Design of Energy-Efficient Human Activity Recognition Systems with Wearable Sensors
    Gruetzmacher, Florian
    Hein, Albert
    Kirste, Thomas
    Haubelt, Christian
    TECHNOLOGIES, 2018, 6 (04):
  • [26] Gathering energy from ultra-low-frequency human walking using a double-frequency up-conversion harvester in public squares
    Tan, Yisong
    Lu, Guangpeng
    Cong, Moyue
    Wang, Xinyu
    Ren, Limin
    ENERGY CONVERSION AND MANAGEMENT, 2020, 217
  • [27] A 3D multidirectional piezoelectric energy harvester using rope-driven mechanism for low frequency and ultralow intensity vibration environment
    Zhang, Jinhui
    Lin, Maoyu
    Zhou, Wei
    Tang, Lihua
    Qin, Lifeng
    SMART MATERIALS AND STRUCTURES, 2022, 31 (02)
  • [28] Using Wearable Sensors for Human Activity Recognition in Logistics: A Comparison of Different Feature Sets and Machine Learning Algorithms
    Syed, Abbas Shah
    Syed, Zafi Sherhan
    Shah, Muhammad Shehram
    Saddar, Salahuddin
    INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND APPLICATIONS, 2020, 11 (09) : 644 - 649
  • [29] Origin: Enabling On-Device Intelligence for Human Activity Recognition Using Energy Harvesting Wireless Sensor Networks
    Mishra, Cyan Subhra
    Sampson, Jack
    Kandemir, Mahmut Taylan
    Narayanan, Vijaykrishnan
    PROCEEDINGS OF THE 2021 DESIGN, AUTOMATION & TEST IN EUROPE CONFERENCE & EXHIBITION (DATE 2021), 2021, : 1414 - 1419
  • [30] Robust human activity recognition using single accelerometer via wavelet energy spectrum features and ensemble feature selection
    Tian, Yiming
    Zhang, Jie
    Wang, Jie
    Geng, Yanli
    Wang, Xitai
    SYSTEMS SCIENCE & CONTROL ENGINEERING, 2020, 8 (01): : 83 - 96