With the rapid development of wireless communication and micro-power technologies, smart wearable devices with various functionalities appear more and more in our daily lives. Nevertheless, they normally possess short battery life and need to be recharged with external power sources with a long charging time, which seriously affects the user experience. To help extend the battery life or even replace it, a non-resonant piezoelectric–electromagnetic–triboelectric hybrid energy harvester is presented to effectively harvest energy from low-frequency human motions. In the designed structure, a moving magnet is used to simultaneously excite the three integrated energy collection units (i.e., piezoelectric, electromagnetic, and triboelectric) with a synergistic effect, such that the overall output power and energy-harvesting efficiency of the hybrid device can be greatly improved under various excitations. The experimental results show that with a vibration frequency of 4 Hz and a displacement of 200 mm, the hybrid energy harvester obtains a maximum output power of 26.17 mW at 70 kΩ for one piezoelectric generator (PEG) unit, 87.1 mW at 500 Ω for one electromagnetic generator (EMG) unit, and 63 μW at 140 MΩ for one triboelectric nanogenerator (TENG) unit, respectively. Then, the generated outputs are adopted for capacitor charging, which reveals that the performance of the three-unit integration is remarkably stronger than that of individual units. Finally, the practical energy-harvesting experiments conducted on various body parts such as wrist, calf, hand, and waist indicate that the proposed hybrid energy harvester has promising application potential in constructing a self-powered wearable system as the sustainable power source.

An innovative coupling numerical algorithm is proposed in the current paper, the front-tracking method–lattice Boltzmann method–machine learning (FTM-LBM-ML) method, to precisely capture fluid flow phase interfaces at the mesoscale and accurately simulate dynamic processes. This method combines the distinctive abilities of the FTM to accurately capture phase interfaces and the advantages of the LBM for easy handling of mesoscopic multi-component flow fields. Taking a single vacuole rising as an example, the input and output sets of the machine learning model are constructed using the FTM’s flow field, such as the velocity and position data from phase interface markers. Such datasets are used to train the Bayesian-Regularized Back Propagation Neural Network (BRBPNN) machine learning model to establish the corresponding relationship between the phase interface velocity and the position. Finally, the trained BRBPNN neural network is utilized within the multi-relaxation LBM pseudo potential model flow field to predict the phase interface position, which is compared with the FTM simulation. It was observed that the BRBPNN-predicted interface within the LBM exhibits a high degree of consistency with the FTM-predicted interface position, showing that the BRBPNN model is feasible and satisfies the accuracy requirements of the FT-LB coupling model.