Waseda University

About: Waseda University is a education organization based out in Tokyo, Japan. It is known for research contribution in the topics: Catalysis & Large Hadron Collider. The organization has 24220 authors who have published 46859 publications receiving 837855 citations. The organization is also known as: Waseda daigaku & Sōdai.

One-Step Synthetic Strategy of Hybrid Materials from Bimetallic Metal–Organic Frameworks for Supercapacitor Applications

Abstract: This work reports a facile one-step method for the synthesis of new hybrid porous materials using bimetallic NiCo-MOF-74 as the starting precursor. By controlling the calcination atmosphere and temperature, the bimetallic NiCo-MOF-74 particles can be converted into a series of hybrid materials consisting of carbon, metal, and metal oxides. The direct carbonization of the bimetallic NiCo-MOF-74 particles at 800 °C under N2 atmosphere results in the formation of graphitic carbon/NixCo1−x composites (termed NC-800). In contrast, the heat treatment of NiCo-MOF-74 in air at 350 °C (termed NC-350) yields NixCo1−x/NixCo1−xO composites (with a small trace of carbon) as the product. When evaluated as electrode materials for supercapacitors, NC-800 and NC-350 exhibit high specific capacitances of 715 and 513 F g–1, respectively, at a high current density of 1 A g–1. Furthermore, these hybrid materials also show good cycling stability with no visible degradation in their specific capacitance after 2,500 cycles. The ...

Effective gravitational equations on a brane world with induced gravity

Abstract: We present the effective equations to describe the four-dimensional gravity of a brane world, assuming that a five-dimensional bulk spacetime satisfies the Einstein equations and gravity is confined on the ${Z}_{2}$ symmetric brane. Applying this formalism, we study the induced-gravity brane model first proposed by Dvali, Gabadadze, and Porrati. In a generalization of their model, we show that an effective cosmological constant on the brane can be extremely reduced in contrast with the case of the Randall-Sundrum model even if a bulk cosmological constant and a brane tension are not fine tuned.

Gender differences in knee kinematics and muscle activity during single limb drop landing

Abstract: The likelihood of sustaining an ACL injury in a noncontact situation is two to eight times greater for females than for males. However, the mechanism and risk factors of ACL injury are still unknown. We compared knee kinematics as well as electromyographic activity during landing between male and female athletes. Eighteen male athletes and nineteen female athletes participated in the experiment. The angular displacements of flexion/extension, valgus/varus, and internal/external tibial rotation, as well as the translational displacements of anterior/posterior tibial translation during single limb drop landing were calculated. Simultaneous electromyographical activity of the rectus femoris (RF) and hamstrings (Ham) was taken. During landing, internal tibial rotation of the females was significantly larger than that of the males, while differences were not observed in flexion, varus, valgus, and anterior tibial translation. Hamstrings/quadriceps ratio (HQR) for the 50 ms time period before foot contact was greater in males than in females. The mechanism of noncontact ACL injury during a single limb drop landing would be internal tibial rotation combined with valgus rotation of the knee. Increased internal tibial rotation combined with greater quadriceps activity and a low HQR could be one reason female athletes have a higher incidence of noncontact ACL injuries.

Siamese Neural Network Based Few-Shot Learning for Anomaly Detection in Industrial Cyber-Physical Systems

Abstract: With the increasing population of Industry 4.0, both AI and smart techniques have been applied and become hotly discussed topics in industrial cyber-physical systems (CPS). Intelligent anomaly detection for identifying cyber-physical attacks to guarantee the work efficiency and safety is still a challenging issue, especially when dealing with few labeled data for cyber-physical security protection. In this article, we propose a few-shot learning model with Siamese convolutional neural network (FSL-SCNN), to alleviate the over-fitting issue and enhance the accuracy for intelligent anomaly detection in industrial CPS. A Siamese CNN encoding network is constructed to measure distances of input samples based on their optimized feature representations. A robust cost function design including three specific losses is then proposed to enhance the efficiency of training process. An intelligent anomaly detection algorithm is developed finally. Experiment results based on a fully labeled public dataset and a few labeled dataset demonstrate that our proposed FSL-SCNN can significantly improve false alarm rate (FAR) and F1 scores when detecting intrusion signals for industrial CPS security protection.