Transparent and Robust LiCl–Organohydrogel Triboelectric Nanogenerator With Deep Learning Assisted Sensing

ABSTRACT Hydrogels have emerged as promising materials for flexible and self‐powered electronic devices due to their high ionic conductivity and mechanical adaptability. We developed a polyacrylamide–polyethylenimine (PAAm–PEI) organohydrogel using a glycerol–water binary solvent system and systematically investigated the effects of various alkali and alkaline earth metal salts on triboelectric performance. Among them, LiCl‐incorporated organohydrogels demonstrated superior overall characteristics, achieving ∼400 V, and an exceptional strain of up to 3000%, high transparency (∼87%), and excellent antifreeze and anti‐evaporation properties. The LiCl ions formed robust hydrogen bonding and ionic interactions with water, effectively suppressing ice crystallization and minimizing water loss. The resulting organohydrogel‐based triboelectric nanogenerator (H‐TENG) maintained 91.4% and 80% of its initial output after exposure to 60°C and −20°C, respectively, and recovered 85% of voltage after self‐healing within 10 s. Moreover, by integrating a hybrid CNN–LSTM deep learning model, the TENG achieved a high classification accuracy of 95.64% in Morse code recognition, demonstrating reliable intelligent sensing capability. This work not only provides an effective strategy to balance ionic conductivity, mechanical flexibility, and optical transparency in organohydrogels, but also highlights the potential of deep learning‐assisted TENG systems for advanced human–machine interfaces and next‐generation wearable electronics.