Synergistic integration of piezoelectric−electromagnetic−triboelectric generators : system design, performance evaluation, and optimization analysis

Abstract

The swift advancement of the Internet of Things (IoT) and Industry 5.0, along with the surging demand for self-powered real-time environmental monitoring and AI-driven automation systems, has increased interest in vibration-based energy harvesting technologies. Conventional power sources like batteries suffer from limited lifespans, high maintenance requirements, and environmental issues, highlighting the need for sustainable and maintenance-free energy solutions. Among various energy harvesting methods, piezoelectric generators (PEG), electromagnetic generators (EMG), and triboelectric nanogenerators (TENG) have become prominent due to their capacity to transform ambient mechanical energy into electrical power. However, each of these methods has its own limitations, such as restricted frequency adaptability, low power density, and durability concerns. To tackle these technical challenges, a triple hybridization of energy harvesting systems that integrate multiple transduction mechanisms can be created, enabling a broader operational bandwidth, improved efficiency, and superior energy conversion performance. This research explores nonlinear multi-stable tri-hybrid energy harvesting systems that integrate PEG, EMG, and TENG units, with an emphasis on their system design, performance evaluation, and optimization strategies. In contrast to single-mode and dual-mode energy harvesters, tri-hybrid energy harvesting systems leverage the complementary advantages of each mechanism, resulting in higher power output, greater frequency adaptability, and improved durability. Specifically, PEG units effectively convert high-frequency mechanical vibrations into electrical energy, EMG units boost power generation in response to large-amplitude and low-frequency vibrations, and TENG units facilitate energy harvesting from time-varying, low-amplitude, and low-frequency mechanical excitations. By interacting these three mechanisms, the system achieves a multi-stability-based frequency up-conversion approach, enabling continuous and stable power generation across diverse environmental conditions. Despite the advantages of multi-stable hybridization, several significant challenges persist. Key concerns include optimizing the interaction among the three energy harvesting units, reducing negative interference, and ensuring that changes to one unit do not adversely affect the performance of the others. Furthermore, the complexity of the design escalates considerably when integrating multiple transduction mechanisms, thereby requiring careful attention to geometric configurations, material properties, and coupling effects. In addition, the spatial limitations of compact electronic devices require efficient structural designs to maximize power output while minimizing the overall footprint. To address these issues, this study employs a combined experimental and simulation methodology to systematically evaluate the performance of tri-hybrid vibration-driven energy harvesters. First, the role of triboelectric components (friction) within the overall system is analyzed, focusing on their effects on power generation, durability, and frequency response. Next, the auxetic piezoelectric structural unit is examined, highlighting its improved mechanical flexibility, strain distribution, and output efficiency. We also explore the effect of magneto-multi-stable oscillators into tri-hybrid design. Finally, strategies to enhance power density and optimize the integration of the hybrid nonlinear system are explored. The results offer valuable insights into the interactions between different energy harvesting mechanisms and offer guidelines for designing next-generation hybrid energy harvesters with improved efficiency and robustness. Experimental validations involve the fabrication and testing of tri-hybrid energy harvester prototypes, followed by controlled vibration and electromechanical characterization. The performance of tri-hybrid energy harvesters is evaluated under periodic and chaotic behaviors by analyzing voltage and current output at different vibration frequencies, the efficiency of multi-stable energy conversion mechanisms, the effects of material properties on charge generation, and the stability and durability of power output. Going beyond structural design, optimization techniques are essential for enhancing the performance of tri-hybrid vibration-driven energy harvesters. While conventional optimization techniques (e.g., finite element method) are commonly used, they tend to be computationally intensive and require experimental validation. On the other hand, advanced machine learning methods (e.g., deep reinforcement learning), have been investigated for energy harvesting applications, but they mainly emphasize adaptive control instead of direct optimization. In contrast, Physics-informed neural networks (PINNs) offer an efficient data-driven optimization framework by directly integrating governing physical laws into a neural network architecture. Unlike conventional machine learning models that depend solely on labeled datasets, PINNs allow for accurate predictions with minimal or scarce experimental data. This work demonstrates that the integration of an advanced time-marching PINN-based (AT-PINN) method with a genetic algorithm, forming "AT-PINN-GA approach", can enhance and optimize critical design parameters. The arrangement of magnets in EMG is optimized by modifying the distance between them, which affects the restoring force and multi-stability. This adjustment enables transitions into tri-stable or quad-stable nonlinear system states, resulting in increased displacement and improved energy conversion efficiency. For the piezoelectric plate configuration, the method is used to predict optimal structural designs that maximize strain-induced charge generation while minimizing material degradation and mechanical losses. Furthermore, the performance of TENG units can also be enhanced. Compared to traditional numerical simulations and experimental tuning, the PINN-based approach significantly lowers computational costs, allowing for real-time optimization through dynamic adjustments of system parameters. This ensures that the tri-hybrid energy harvester operates at peak efficiency under varying environmental conditions, making it a highly effective and adaptable solution for energy harvesting applications. In addition, traditional methods typically require determining a single parameter first and then roughly setting the ranges for four additional parameters. After an initial calculation, each parameter must be adjusted sequentially until an optimal configuration is achieved for effective energy harvesting at a specific frequency. This process involves multiple iterations, making it time-consuming and inefficient. The integration of PINN-based optimization greatly enhances calculation efficiency, resulting in a substantial reduction in computational time compared to conventional tuning methods. However, the increased complexity of tri-hybrid vibration-based energy harvesting structures leads to manufacturing and scalability issues, resulting in advanced microfabrication and materials engineering solutions. Power management and circuit integration present difficulties, as each transduction mechanism produces electrical output with varying voltage, current, and impedance characteristics. This requires sophisticated rectifiers, impedance matching circuits, and DC-DC converters for effective energy storage and utilization. Furthermore, nonlinear behaviors caused by electric hysteresis, charge leakage, and unpredicted distortion must be carefully managed to minimize energy loss and enhance long-term stability. Addressing these problems is essential for further optimizing the tri-hybrid energy harvester for practical applications. Future research should concentrate on developing advanced materials and structural configurations to improve durability, flexibility, and efficiency, ensuring the long-term stability and adaptability of tri-hybrid energy harvesters. Besides, incorporating self-adaptive power management circuits will facilitate real-time adjustments of energy harvesting parameters based on environmental conditions, optimizing performance across varying operational scenarios. Finally, applying tri-hybrid vibration-based energy harvesters in wearable IoT-connected electronics and remote sensor networks will enable them to utilize their self-sustaining capabilities, making them suitable for a wide array of real-world engineering applications. In conclusion, the present work offers a thorough analysis of tri-hybrid energy harvesting systems, showcasing their superior performance compared to conventional single and dual-mode harvesters. By integrating experimental, simulation and optimization approaches, this research contributes to the advancement of self-powered energy harvesting technologies, paving an efficient way for highly efficient, adaptive, and sustainable energy harvesting solutions.