A5072936331- Ferroelectric and Piezoelectric Materials
- Ultrasonics and Acoustic Wave Propagation
- Acoustic Wave Resonator Technologies
- Multiferroics and related materials
- Advanced ceramic materials synthesis
- Adversarial Robustness in Machine Learning
- Advanced Sensor and Energy Harvesting Materials
- Anomaly Detection Techniques and Applications
- Dielectric materials and actuators
- Advanced materials and composites
- Non-Destructive Testing Techniques
- Structural Health Monitoring Techniques
- Microwave Dielectric Ceramics Synthesis
- Metal and Thin Film Mechanics
- Innovative Energy Harvesting Technologies
- Smart Materials for Construction
- Shape Memory Alloy Transformations
- Domain Adaptation and Few-Shot Learning
- Bacillus and Francisella bacterial research
- Rock Mechanics and Modeling
- Electromagnetic Effects on Materials
- Pharmaceutical and Antibiotic Environmental Impacts
- Energy Harvesting in Wireless Networks
- Numerical methods in engineering
- Metaheuristic Optimization Algorithms Research
Wuhan University
2015-2024
Guangxi University
2024
Yanshan University
2021-2024
Ningbo University
2023-2024
East China University of Science and Technology
2022-2024
Technical University of Darmstadt
2019-2023
Southern University of Science and Technology
2021-2023
Wuchang University of Technology
2021-2022
Northwest A&F University
2022
Wuhan University of Technology
2020-2021
Abstract Most oxide ceramics are known to be brittle macroscopically at room temperature with little or no dislocation‐based plasticity prior crack propagation. Here, we demonstrate the size‐dependent ductile transition in SrTiO 3 using nanoindentation pop‐in events visible as a sudden increase displacement nominally constant load. We identify that indentation event temperature, below critical indenter tip radius, is dominated by dislocation‐mediated plasticity. When radius increases size,...
A one-dimensional pencil-like magnetoelectric (ME) sensor prototype is proposed which consists of a magnetostrictive cylinder, truncated conical spacer, and piezoelectric disk assembled in rigid frame. By adopting the displacement-transfer mode this sensor, not only strain loss at ME interface avoided but also volume fractions both phases can be adjusted broader range. Using nonlinear model linear model, coupling performance systematically analyzed using lead titanate zirconate (PZT) disks...
Shape and texture are two prominent complementary cues for recognizing objects. Nonetheless, Convolutional Neural Networks often biased towards either or shape, depending on the training dataset. Our ablation shows that such bias degenerates model performance. Motivated by this observation, we develop a simple algorithm shape-texture debiased learning. To prevent models from exclusively attending single cue in representation learning, augment data with images conflicting shape information...
This paper proposed a piezoelectric energy harvester based on spring-mass-spring oscillator, of which the piezoelectrics operate in d33 mode. Theoretical analysis reveals that oscillator can not only generate larger vibration than ambient system but also buffer force possible accidental impact applied stacks. By using lead zirconate titanate (PZT-4) ceramics as model materials, we systematically characterized performance harvester. Results show at resonance frequency, output satisfactory...
This paper focuses on learning transferable adversarial examples specifically against defense models (models to attacks). In particular, we show that a simple universal perturbation can fool series of state-of-the-art defenses. Adversarial generated by existing attacks are generally hard transfer models. We observe the property regional homogeneity in perturbations and suggest defenses less robust regionally homogeneous perturbations. Therefore, propose an effective transforming paradigm...
Superelastic materials capable of recovering large nonlinear strains are ideal for a variety applications in morphing structures, reconfigurable systems, and robots. However, making oxide superelastic has been long-standing challenge due to their intrinsic brittleness. Here, we fabricate ferroelectric BaTiO3 (BTO) micropillars that not only but also possess excellent fatigue resistance, lasting over 1 million cycles without accumulating residual or noticeable variation stress-strain curves....