Artificial Intelligence-Enhanced Flexible Biocompatible Triboelectric Nanogenerator

Triboelectric nanogenerators (TENG) have emerged as a promising technology for energy harvesting in wearable devices and self-powered sensors. However, the effective utilization of TENGs in practical scenarios faces several challenges including limited power output, complexity of material combinations, wear and tear issues, complex integration into systems, sensitivity to frequency ranges, and miniaturization. A lot of work is being done addressing these challenges, however, the use of artificial intelligence in design optimization can pave new ways to reduce experimentation and can identify the optimal design parameters to obtain the best possible output from a range of variables. To unlock the full potential of TENGs, we propose a novel approach that integrates artificial intelligence (AI) techniques with the design of experiment (DOE) method to optimize the design and performance of TENGs. A microcavity-nanoparticle assembled microstructure with one triboelectric layer composed of porous polydimethylsiloxane (PDMS) sandwiched between a second triboelectric layer has been chosen as the initial prototype design. The use of porous PDMS in the structure with nanofillers allows the increase in surface area without the need for a spacer structure between the triboelectric films. This design structure can be well integrated with the human body motion in day-to-day life and provide improved energy conversion. A novel biogel material has been used as electrodes to ensure the biocompatibility of the TENG with the human skin along with flexibility. This study includes variation of key parameters or factors such as material combinations, porosity in PDMS, surface area, thickness, frequency of applied force or pressure, and applied force to prepare different sample prototypes and identify the optimal combinations for TENG performance. The frequency of the applied force mimics the human motion frequency range, which is from 0 to 20 Hz. The data generated from the prototype design with a range for each independent data point is analyzed by statistical tools and software to identify significant factors or parameters, and interactions between these factors are evaluated. After that, the optimization algorithms are applied to find the optimal levels of significant factors. Additional experiments are thereby conducted to validate the predicted performance and find the accuracy of the optimization. Finally, by refining the design parameters and conducting iterative experiments, a final optimized contact separation mode TENG design has been generated that is flexible, biocompatible, and efficient to power wearable biosensors for health monitoring. By harnessing energy from human motion, this AI-enhanced TENG offers a sustainable, reliable, and biocompatible power source for wearable healthcare applications.