Yuze Song
Yuze Song has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Buyu Guo
Hainan Institute, Zhejiang University, Sanya 572025, China
1 shared publication - Chuan Lv
School of New Energy Engineering, Chengdu Technological University, Yibin 644000, China
1 shared publication - Haoxian Dong
Space Star Technology Co., Ltd Tianjin Branch, Tianjin 300000, China
1 shared publication - Lvjun Zhou
Key Laboratory of Radiation Physics and Technology of Ministry of Education, Institute of Nuclear Science and Technology, Sichuan University, Chengdu 610064, China
1 shared publication - Peiliang Li
State Key Laboratory of Ocean Sensing & Ocean College, Zhejiang University, Zhoushan 316021, China
1 shared publication - Xiaofei Xiong
Nansha Islands Coral Reef Ecosystem National Observation and Research Station, Guangzhou 510300, China
1 shared publication
Latest Publications
- Open Access
Multi-Source Fatigue Fracture of 2Cr13 Martensitic Stainless Steel Compressor Blades: The Critical Role of Surface Integrity in Marine Engineering Reliability
Compressor rotor blades in marine engineering applications are exposed to harsh, corrosive environments and complex aerodynamic loads, making them prone to premature failure. This study investigates the fracture of 12th-stage 2Cr13 martensitic stainless-steel blades following a maintenance overhaul. Despite the replacement of several cracked blades, five blades fractured shortly after restart, accompanied by abnormal vibration. A comprehensive failure analysis was conducted, including macroscopic inspection, fractographic observation, energy-dispersive spectroscopy, metallographic examination, and mechanical property testing. The results indicate that the fractures are multi-source high-cycle fatigue. Crack initiation in the new blade originated from pre-existing transverse mechanical damage, while in the old blades, it initiated from sharp pits and microcracks introduced by sandblasting, which compromised surface integrity. The material exhibited a normal tempered sorbite structure and adequate mechanical properties, with slight strengthening due to service-induced precipitation and dislocation accumulation. The failure followed a typical evolution of multi-source initiation, propagation, crack coalescence, and final overload ductile fracture. These findings highlight the critical role of surface integrity in blade reliability.
Open AccessMulti-Source Fatigue Fracture of 2Cr13 Martensitic Stainless Steel Compressor Blades: The Critical Role of Surface Integrity in Marine Engineering Reliability
Compressor rotor blades in marine engineering applications are exposed to harsh, corrosive environments and complex aerodynamic loads, making them prone to premature failure. This study investigates the fracture of 12th-stage 2Cr13 martensitic stainless-steel blades following a maintenance overhaul. Despite the replacement of several cracked blades, five blades fractured shortly after restart, accompanied by abnormal vibration. A comprehensive failure analysis was conducted, including macroscopic inspection, fractographic observation, energy-dispersive spectroscopy, metallographic examination, and mechanical property testing. The results indicate that the fractures are multi-source high-cycle fatigue. Crack initiation in the new blade originated from pre-existing transverse mechanical damage, while in the old blades, it initiated from sharp pits and microcracks introduced by sandblasting, which compromised surface integrity. The material exhibited a normal tempered sorbite structure and adequate mechanical properties, with slight strengthening due to service-induced precipitation and dislocation accumulation. The failure followed a typical evolution of multi-source initiation, propagation, crack coalescence, and final overload ductile fracture. These findings highlight the critical role of surface integrity in blade reliability.
- Open Access
MLP-MFF: Lightweight Pyramid Fusion MLP for Ultra-Efficient End-to-End Multi-Focus Image Fusion
Limited depth of field in modern optical imaging systems often results in partially focused images. Multi-focus image fusion (MFF) addresses this by synthesizing an all-in-focus image from multiple source images captured at different focal planes. While deep learning-based MFF methods have shown promising results, existing approaches face significant challenges. Convolutional Neural Networks (CNNs) often struggle to capture long-range dependencies effectively, while Transformer and Mamba-based architectures, despite their strengths, suffer from high computational costs and rigid input size constraints, frequently necessitating patch-wise fusion during inference—a compromise that undermines the realization of a true global receptive field. To overcome these limitations, we propose MLP-MFF, a novel lightweight, end-to-end MFF network built upon the Pyramid Fusion Multi-Layer Perceptron (PFMLP) architecture. MLP-MFF is specifically designed to handle flexible input scales, efficiently learn multi-scale feature representations, and capture critical long-range dependencies. Furthermore, we introduce a Dual-Path Adaptive Multi-scale Feature-Fusion Module based on Hybrid Attention (DAMFFM-HA), which adaptively integrates hybrid attention mechanisms and allocates weights to optimally fuse multi-scale features, thereby significantly enhancing fusion performance. Extensive experiments on public multi-focus image datasets demonstrate that our proposed MLP-MFF achieves competitive, and often superior, fusion quality compared to current state-of-the-art MFF methods, all while maintaining a lightweight and efficient architecture.
Open AccessMLP-MFF: Lightweight Pyramid Fusion MLP for Ultra-Efficient End-to-End Multi-Focus Image Fusion
Limited depth of field in modern optical imaging systems often results in partially focused images. Multi-focus image fusion (MFF) addresses this by synthesizing an all-in-focus image from multiple source images captured at different focal planes. While deep learning-based MFF methods have shown promising results, existing approaches face significant challenges. Convolutional Neural Networks (CNNs) often struggle to capture long-range dependencies effectively, while Transformer and Mamba-based architectures, despite their strengths, suffer from high computational costs and rigid input size constraints, frequently necessitating patch-wise fusion during inference—a compromise that undermines the realization of a true global receptive field. To overcome these limitations, we propose MLP-MFF, a novel lightweight, end-to-end MFF network built upon the Pyramid Fusion Multi-Layer Perceptron (PFMLP) architecture. MLP-MFF is specifically designed to handle flexible input scales, efficiently learn multi-scale feature representations, and capture critical long-range dependencies. Furthermore, we introduce a Dual-Path Adaptive Multi-scale Feature-Fusion Module based on Hybrid Attention (DAMFFM-HA), which adaptively integrates hybrid attention mechanisms and allocates weights to optimally fuse multi-scale features, thereby significantly enhancing fusion performance. Extensive experiments on public multi-focus image datasets demonstrate that our proposed MLP-MFF achieves competitive, and often superior, fusion quality compared to current state-of-the-art MFF methods, all while maintaining a lightweight and efficient architecture.