Particle-Laden Droplet-Driven Triboelectric Nanogenerator for Real-Time Sediment Monitoring Using a Deep Learning Method

Droplet Energy Harvesting System Based on Total-Current Nanogenerator

The droplet-based nanogenerator (DNG) is a highly promising technology for harvesting high-entropy water energy in the era of the Internet of Things. Yet, despite the exciting progress made in recent years, challenges have emerged unexpectedly for the AC-type DNG-based energy system as it transitions from laboratory demonstrations to real-world applications. In this work, we propose a high-performance DNG system based on the total-current nanogenerator concept to address these challenges. This system utilizes the water-charge-shuttle architecture for easy scale-up, employs the field effect to boost charge density of the triboelectric layer, adopts an on-solar-panel design to improve compatibility with solar energy, and is equipped with a novel DC-DC buck converter as power management circuit. These features allow the proposed system to overcome the existing bottlenecks of DNG and empower the system with superior performances compared with previous ones. Notably, with the core architecture measuring only 15 cm × 12.5 cm × 0.3 cm in physical dimensions, this system reaches a record-high open-circuit voltage of 4200 V, capable of illuminating 1440 LEDs, and can charge a 4.7 mF capacitor to 4.5 V in less than 24 min. In addition, the practical potential of the proposed DNG system is further demonstrated through a self-powered, smart greenhouse application scenario. These demonstrations include the continuous operation of a thermohygrometer, the operation of a Bluetooth plant monitor, and the all-weather energy harvesting capability. This work will provide valuable inspiration and guidance for the systematic design of next-generation DNG to unlock the sustainable potential of distributed water energy for real-world applications.

Synergizing Machine Learning Algorithm with Triboelectric Nanogenerators for Advanced Self-Powered Sensing Systems

The advancement of the Internet of Things (IoT) has increased the demand for large-scale intelligent sensing systems. The periodic replacement of power sources for ubiquitous sensing systems leads to significant resource waste and environmental pollution. Human staffing costs associated with replacement also increase the economic burden. The triboelectric nanogenerators (TENGs) provide both an energy harvesting scheme and the possibility of self-powered sensing. Based on contact electrification from different materials, TENGs provide a rich material selection to collect complex and diverse data. As the data collected by TENGs become increasingly numerous and complex, different approaches to machine learning (ML) and deep learning (DL) algorithms have been proposed to efficiently process output signals. In this paper, the latest advances in ML algorithms assisting solid-solid TENG and liquid-solid TENG sensors are reviewed based on the sample size and complexity of the data. The pros and cons of various algorithms are analyzed and application scenarios of various TENG sensing systems are presented. The prospects of synergizing hardware (TENG sensors) with software (ML algorithms) in a complex environment and their main challenges for future developments are discussed.

Detection of Microplastics Based on a Liquid-Solid Triboelectric Nanogenerator and a Deep Learning Method

Microplastics are sub-millimeter-sized fragments of plastics, which have been found in environments to a great extent. They are relatively new pollutants that are difficult to be degraded. They not only cause irreversible adverse effects on microorganisms, animals, and plants but also enter the human body through the food chain and affect human health. However, due to their small size, variety, and differences in physical and chemical properties of microplastics, traditional detection and identification still face challenges. This work provides a method for detecting and classifying microplastics in liquids using a liquid-solid triboelectric nanogenerator (LS-TENG) in combination with a deep learning model. The experiment showed that the type and content of microplastics in the liquid had a great effect on the contact electrification between the liquid and the perfluoroethylene-propylene copolymer. After adding polyethylene, polypropylene, polyvinyl chloride, polyethylene terephthalate, and polystyrene microplastics to the liquids, it was found that the type and content of different microplastics have a significant impact on the output voltage signal of the LS-TENG sensor. When the mass fraction of microplastics ranged from 0.025 to 0.25 wt %, the voltage output of the LS-TENG sensor had a linear relationship with the mass fraction of microplastics. Therefore, a method for quantitatively detecting the content of microplastics using the LS-TENG sensor has been established. Based on the LS-TENG output voltage signal, a convolutional neural network deep learning model was used to identify different types of labels, and high recognition accuracy was achieved. These are of great significance for expanding the application prospect of LS-TENG and realizing the detection of microplastics in liquids.

Model-driven deep unrolling: Towards interpretable deep learning against noise attacks for intelligent fault diagnosis

Intelligent fault diagnosis (IFD) has experienced tremendous progress owing to a great deal to deep learning (DL)-based methods over the decades. However, the "black box" nature of DL-based methods still seriously hinders wide applications in industry, especially in aero-engine IFD, and how to interpret the learned features is still a challenging problem. Furthermore, IFD based on vibration signals is often affected by the heavy noise, leading to a big drop in accuracy. To address these two problems, we develop a model-driven deep unrolling method to achieve ante-hoc interpretability, whose core is to unroll a corresponding optimization algorithm of a predefined model into a neural network, which is naturally interpretable and robust to noise attacks. Motivated by the recent multi-layer sparse coding (ML-SC) model, we herein propose to solve a general sparse coding (GSC) problem across different layers and deduce the corresponding layered GSC (LGSC) algorithm. Based on the ideology of deep unrolling, the proposed algorithm is unfolded into LGSC-Net, whose relationship with the convolutional neural network (CNN) is also discussed in depth. The effectiveness of the proposed model is verified by an aero-engine bevel gear fault experiment and a helical gear fault experiment with three kinds of adversarial noise attacks. The interpretability is also discussed from the perspective of the core of model-driven deep unrolling and its inductive reconstruction property.

Deep learning: an efficient method for plasmonic design of geometric nanoparticles

Plasmons of noble metal nanoparticles have played an important role in energy transfer research, biomedical sensing, drug preparation and photocatalysis in recent years. The scattering spectra of nanoparticles are dramatically affected by numerous complex parameters, such as morphology, material, and volumes, making the parameters design a necessary step before the experiment. However, the plasmonic design is limited by several difficulties, such as the high degree of freedom, expensive trial and error costs. Herein, a plasmonic design method based on dual strategy deep learning is proposed, which can provide the design proposals according to the required spectrum. To make the model closer to the real experimental situation, the boundary element method was used to build a scattering spectra dataset of geometric nanoparticles (>1200 000 samples). Driven by the above data, the artificial intelligence (AI) model learns the relationship between spectral features and design parameters. Then, the performance statistics of the model were implemented from multiple dimensions, and a high design precision of over 90% was achieved in testing cases (testing samples >120 000). Moreover, to verify the realizability of the proposed model, the scattering spectra of nanoparticles designed by AI were constructed using a dark-field microscope system. The experimental results showed that the deviation between the target and the actual spectra was very small and within the acceptable range. This proves the realizability of the AI model proposed in this paper, and sheds light on the application of AI in plasmonic design.

Triboelectric Nanogenerator as a Probe for Measuring the Charge Transfer between Liquid and Solid Surfaces

The phenomenon of triboelectricity involves the flow of charged species across an interface, but conclusively establishing the nature of the charge transfer has proven extremely difficult, especially for the liquid-solid cases. Herein, we developed a self-powered droplet triboelectric nanogenerator (droplet-TENG) with spatially arranged electrodes as a probe for measuring the charge transfer process between liquid and solid interfaces. The information on the electric signal on spatially arranged electrodes shows that the charge transfer between droplets and the solid is an accumulation process during the dropping and that the electron is the dominant charge-transfer species. Such a droplet-TENG showed a high sensitivity to the ratio of solvents in the mixed organic solution, and we postulated this is due to the possibility of generation of a hydrogen bond, affecting the electric signal on the spatially arranged electrodes. This work demonstrated a chemical sensing application based on the self-powered droplet triboelectric nanogenerator.