Search

Why quaternions in neural networks (NNs)? Are there quaternions in the human brain? “No” may be an ordinary answer. However, quaternion NNs (QNNs) are a powerful framework that strongly connects artificial intelligence (AI) and the real world. In this article, we deal with NNs based on quaternions and describe their basics and features. We also detail the underlying ideas in their engineering applications, especially when we adaptively process the polarization information of electromagnetic waves. We focus on their role in remote sensing, such as Earth observation radar mounted on artificial satellites or aircraft and underground radar, as well as mobile communication. There, QNNs are a class of NNs that know physics, especially polarization, composing a framework by fusing measurement physics with adaptive-processing mathematics. This fusion realizes a seamless integration of measurement and intelligence, contributing to the construction of a human society having harmony between AI and real human lives.

In metal metallurgy, precipitation strengthening is widely used to increase material strength by utilizing the impediment effect of the second-phase particles on dislocation movements. Inspired by this mechanism, in this paper, novel multiphase heterogeneous lattice materials are developed with enhanced mechanical properties utilizing a similar hindering effect of second-phase lattice cells on the shear band propagation. For this purpose, biphase and triphase lattice samples are fabricated using high-speed multi jet fusion (MJF) and digital light processing (DLP) additive manufacturing techniques, and a parametric study is carried out to investigate their mechanical properties. Different from the conventional random distribution, the second-phase and third-phase cells in this work are continuously distributed along the regular pattern of a larger-scale lattice to form internal hierarchical lattice structures. The results show that the triphase lattices possess balanced mechanical properties. Interestingly, this indicates that introducing a relatively weak phase also has the potential to improve the stiffness and plateau stress, which is distinct from the common mixed rule. This work is aimed at providing new references for the heterogeneous lattice design with outstanding mechanical properties through material microstructure inspiration.

Ground-penetrating radar (GPR)-based landmine detection has advantages such as high safety and high efficiency. There are various methods to process the data obtained from GPR systems. One of the commonly used methods is to visualize the quaternion-type polarization data by quaternion self-organizing map (QSOM). However, QSOM cannot take into account the geometric property of the polarization data. Then, degree-of-polarization (DoP) error is introduced in the self-organization process of QSOM, which leads to unsatisfactory visualization results. To overcome the limitation, in this article, we propose a novel processing, Riemannian QSOM (RQSOM), which takes into account the geometric property to eliminate the DoP error with the help of logarithmic and exponential maps in Riemannian geometry. We analyze its basic dynamics and compare it with that of QSOM. We conduct experiments to visualize a mock plastic landmine with QSOM and RQSOM. The experimental results show that RQSOM realizes expected self-organization dynamics, thereby achieving better visualization results compared with QSOM.

This article sheds light on an unexpected phenomenon of partial-cycling-induced feigned death in graphite electrodes (coin half cells). The partial cycling within specific voltage windows can trigger rapid but recoverable capacity fading of graphite electrodes. During partial cycling, the extremely rapid capacity fading (up to 80 %) appears in the initial dozens of cycles, and the voltage profile gradually becomes abnormal. Nevertheless, the capacity fading phenomenon can be erased by one full cycling cycle. By combining multiple characterizations (including scanning electron microscope, energy dispersive spectrometer, electrochemical impedance spectroscopy, and X-ray diffraction) with evaluating the influence of the fluoroethylene carbonate additive, it is further confirmed that the partial-cycling-induced feigned death in graphite electrodes is directly related to the composition evolution of solid electrolyte interphase. Meanwhile, a necessary condition for triggering the feigned death is that the lower cut-off voltage is close to the voltage plateaus for LiC18/LiC12transition and LiC12/LiC6transition. Based on these findings, this work provides an in-depth understanding of the cycling performance of graphite electrodes and also presents a challenge for the electrolyte design.

Traditional evaluation of battery charging protocols typically requires hundreds of electrochemical cycles and months of experimentation to select charging schemes that maximize the battery performance without compromising the cycle life. In this work, by nesting clustering and classification algorithms, a data-driven method using only data within a few tens of cycles is proposed to accurately classify constant-current charging protocols and rapidly identify the critical current, beyond which rapid degradation tends to occur within a specified lifetime. Specifically, by utilizing unsupervised clustering to process early-stage features and generate prediction labels, a model for early-stage prediction of the rapid degradation is established with an accuracy higher than 92.75%. Subsequently, the critical current is determined by intersecting the classification boundary with the physical distribution domain of the features. The reliability and generalizability of the proposed method is also discussed, which suggests that only ∼30 cycles and ∼40 samples are required to accomplish acceptable identification. The method is also proven to suitable for different battery systems. Therefore, the data-driven method proposed in this work provides a novel pathway to rapidly evaluate fast-charging batteries and charging protocols.

Application of silicon in high capacity electrodes of lithium ion battery suffers from stress effects and, in turn, affects voltage performance of battery. This paper established a reaction-diffusion-stress coupled model and investigated the stress induced voltage hysteresis with consideration of diffusion induced stress, surface effects and interparticle compression. It was found stress and stress induced voltage hysteresis depended on particle size. In big particles, diffusion induced stress is dominant and leads to significant hysteresis in both stress and voltage, indicating energy dissipation due to stress effects. In small particles, e.g., radius of dozens nanometers, diffusion induced stress was negligible while surface effects played dominant role, leading to nearly vanished voltage hysteresis shifting away from equilibrium potential. According to calculation, particle sizes around 100 nm are appropriate choice for electrode design as both diffusion induced stress and surface effects are insignificant. Finally, interparticle compression pushed the stress hysteresis to compressive side and led to early termination of lithiation at cut-off voltage. Denser electrode would enhance this effect. It indicated that there must be a limit of mixing ratio of silicon because higher interparticle compressive stress due to more introduced silicon would impede the battery from fully charged.

The stress evolution, total charging time and capacity utilization of pulse charging (PC) method are investigated in this paper. It is found that compared to the conventional constant current (CC) charging method, the PC method can accelerate the charging process but will inevitably cause an increase in stress and a decrease in capacity. The charging speed for PC method can be estimated by the mean current. By introducing stress control, a modified PC method called the PCCC method, which starts with a PC operation followed by a CC operation, is proposed. The PCCC method not only can accelerate charging process but also can avoid the stress raising and capacity loss occurring in the PC method. Furthermore, the optimal pulsed current density and switch time in the PCCC method is also discussed.

A macroscopic architecture design of lithium metal electrodes for solving the problem of extremely excessive lithium metal is proposed in this paper. By employing a simple mechanical processing method, macroscopic hollows within lithium foils are introduced, and consequently, the amount of lithium metal has economized significantly. The cyclability of lithium foils with millimeter-size hollows is evaluated jointly via modeling and experiments. The results suggest that the well-designed macroscopic hollow causes controllable sacrifices of battery cycling performances and considerably boosts the utilization of lithium metal. The relationship of economization, cyclability, and utilization of lithium metal is also discussed. The universality of the results is also verified in different battery systems. Meanwhile, the initial hollows are found to heal morphologically after a series of electrochemical cycles, and the existence of lithium metal in the healing product is also confirmed, indicating that hollows provide room for the in-plane lithium dendrite growth. Based on these findings, this work provides a new perspective on the architectural design of lithium metal electrodes.

In this article, we propose a unified geolocation framework to automatically determine where on the earth a web video was shot. We analyze different social, visual, and textual relationships from a real-world dataset and find four relationships with apparent geography clues that can be used for web video geolocation. Then, the geolocation process is formulated as an optimization problem that simultaneously takes the social, visual, and textual relationships into consideration. The optimization problem is solved by an iterative procedure, which can be interpreted as a propagation of the geography information among the web video social network. Extensive experiments on a real-world dataset clearly demonstrate the effectiveness of our proposed framework, with the geolocation accuracy higher than state-of-the-art approaches.

Analytical solutions of diffusion induced stress in a cylindrical multilayer electrode during lithiation are obtained. Design insights are provided by analyzing the diffusion induced stress as a function of geometrical parameters, material properties and charging procedure. It is found that the stress is induced not only by the Li-ion concentration inhomogeneity in the active layer materials, but also by the restriction of the lithiated active layer by the current collector. Curvature and thickness of electrode, as well as rigidity of current collector, have great impacts on the stresses. The maximum stress during charging can be minimized by reducing the thickness of electrode, employing soft material for the current collector and decreasing the thickness of current collector. It is also found that the ratio of maximum stress to minimum stress in the active layers can be reduced from 2.1 to 1.1 if the radius-to-thickness ratio of electrode is increased from 1 to 10. Employing this radius-to-thickness ratio provides that the maximum stresses are quite uniform in the two active layers attached on current collector, independently of lithiation or delithiation. Finally, an optimized charging procedure is suggested based on the diffusion induced stresses, i.e. galvanostatic first followed by potentiostatic.

To accelerate charging without rapid degradation, fast charging protocols that regulate stress are theoretically and experimentally investigated. In the stress-regulated protocol, the diffusion-induced stress within the active particle rapidly reaches a critical value in the early stage of charging and maintains that value throughout the charging process. To resolve the issue of the theoretically infinite current at the beginning of the stress-regulated protocol, a modified method that contains an initial constant-current stage followed by a stress-regulated stage is proposed. The key parameters are identified from the governing equation and can be easily obtained via constant-current charging tests. The modeling and experimental results collectively suggest that the proposed protocol considerably shortens the charging process (10%-20% compared with the referenced constant-current protocol) without sacrificing cycling stability and capacity utilization. The proposed protocol appears superior to other fast charging protocols (including high-rate constant-current charging, boost charging, and exponential current charging) in achieving a balance of charging speed and cycling performance.

Many promising candidate electrode materials suffer from severe voltage decay, which is detrimental to lithium ion battery performance. Recently, the relationship between stress-mediated chemical potential and voltage variation has received much attention. While, how the plastic deformation affects the voltage variation remains unclear. A stress coupled phase field reaction model is employed to reveal the mechanism for the stress-related voltage decay accompanied by plastic deformation, wherein different lithiation behaviors under electrochemical reactions are considered. It is demonstrated that hydrostatic compression on the surface is responsible for the stress-induced voltage decay, and the plastic deformation can mitigate the voltage decay via not only relaxing the hydrostatic compression, but also changing hydrostatic compression on the surface to be hydrostatic tension under the small yield stress. In addition, under the same yield stress, two-phase lithiation process is prone to trigger the plastic deformation compared to the single phase lithiation process, facilitating to elevate the output voltage. Further, compared to the circular particle, the interconnected particle can improve the output voltage through the favorable transition from hydrostatic compression to hydrostatic tension at the interconnected region and the surface with large curvature.

Volatile organic solids, such as cyclododecane or menthol, have been employed as temporary reinforcement material during archaeological excavations. They are usually applied as melts and reinforcement is achieved once the melts solidify. Such solidification process can induce internal stress on the artifacts, which can be a big concern, especially to those very precious and fragile ones. However, information about such stress is still extremely limited at present. This paper proposes an experimental method based on resistance strain gauge technique to monitor the deformation induced by solidification of menthol melt. Bending tests are performed on very thin glass slides. The solidification process of menthol melt is well characterized by the development of mechanical strains. Then, menthol melts are applied to three kinds of simulated samples, i.e. glass, sandstone and rice paper, to investigate the mechanical response of preserved bodies upon solidification. It is found that menthol melt will generate certain amount expansion or contraction of the objects upon solidification. The stresses induced, evaluated according to obtained strains, are generally quite small, indicating that application of menthol as reinforcement material is safe in mechanics for cultural relics.

Analytical solutions of diffusion induced stress in a cylindrical multilayer electrode during lithiation are obtained. Design insights are provided by analyzing the diffusion induced stress as a function of geometrical parameters, material properties and charging procedure. It is found that the stress is induced not only by the Li-ion concentration inhomogeneity in the active layer materials, but also by the restriction of the lithiated active layer by the current collector. Curvature and thickness of electrode, as well as rigidity of current collector, have great impacts on the stresses. The maximum stress during charging can be minimized by reducing the thickness of electrode, employing soft material for the current collector and decreasing the thickness of current collector. It is also found that the ratio of maximum stress to minimum stress in the active layers can be reduced from 2.1 to 1.1 if the radius-to-thickness ratio of electrode is increased from 1 to 10. Employing this radius-to-thickness ratio provides that the maximum stresses are quite uniform in the two active layers attached on current collector, independently of lithiation or delithiation. Finally, an optimized charging procedure is suggested based on the diffusion induced stresses, i.e. galvanostatic first followed by potentiostatic.