Dr Dongfang Zhao | Electrical Properties of Materials | Best Innovator Award

Dr. Dongfang Zhao is a dedicated researcher and lecturer at Shanghai Polytechnic University, specializing in artificial intelligence and mechanical fault diagnosis. Since 2014, he has advanced the frontiers of intelligent systems in industrial diagnostics. After completing his Ph.D. at Shanghai University in 2021, he pursued postdoctoral research at the Control Science and Engineering Research Station. Recognized nationally for his outstanding doctoral dissertation, Dr. Zhao continues to drive innovation in AI-enhanced fault detection. He has published 20+ SCI/EI-indexed papers, earning ~500 citations, and contributes significantly to academic and practical advancements. 📚🤖🔧

Dr Dongfang Zhao, Shanghai Polytechnic University, China

Profile

SCOPUS

🎓 Education

Dr. Zhao earned his Ph.D. in Measurement, Control, and Instrumentation from Shanghai University in 2021, where his dissertation was nominated for a national excellence award. He built his foundation in engineering and research through a series of rigorous academic training steps. Following graduation, he immediately joined a prestigious postdoctoral program in Control Science and Engineering. His academic path reflects a commitment to merging theoretical depth with applied innovation in intelligent diagnostics and control systems. 📖🎓🔍

💼 Experience

Currently serving as a lecturer and master’s supervisor at Shanghai Polytechnic University, Dr. Zhao is actively engaged in research and teaching. He completed a postdoctoral fellowship at the Control Science and Engineering Research Station of Shanghai University. Professionally, he has collaborated on several government and industrial research projects, including two National Natural Science Foundation of China initiatives and AI-enabled fault analysis programs. His diverse academic and industry-facing experiences reflect a holistic career in intelligent systems and real-world fault mitigation. 🏫🧪⚙️

🛠️ Contributions

Dr. Dongfang Zhao has developed three advanced AI models tailored to tackle diverse industrial diagnostic challenges. DRANet enhances fault detection accuracy under strong noise 💥, while DBANet ensures precise diagnosis under variable rotational speeds 🔄 using extreme multi-scale entropy. Furthermore, SCANet excels in unlabeled diagnostic scenarios 🚫🏷️, reducing domain confusion and boosting generalization 🌐. These innovations significantly advance smart maintenance, ensuring reliability, adaptability, and intelligence in complex mechanical systems—paving the way for next-generation predictive diagnostics in Industry 4.0 🔧🌍.

📡 Research Projects 

Dr. Dongfang Zhao has led five significant research projects, including a prestigious Super Postdoctoral Incentive Program project in Shanghai, an AI-enabled initiative from the Shanghai Education Commission, and two provincial/ministerial key lab open funds. He also directed a school-level youth innovation project at Shanghai Polytechnic University. Additionally, as a technical backbone, he has contributed to two National Natural Science Foundation of China projects, a Shanghai Science & Agriculture program, and a classified aerospace project. These engagements demonstrate his leadership in AI-driven diagnostics and interdisciplinary collaboration across academia and high-impact industries. 🚀🧠📡🔍💼

🔬 Research Focus

Dr. Zhao’s research revolves around AI-based fault diagnosis for mechanical systems, particularly in complex and noisy environments. His work spans domain adaptation networks (DRANet, DBANet, SCANet) and advanced entropy-based methods for speed-invariant diagnostics. He develops frameworks that increase system robustness, generalization, and diagnostic precision under non-ideal conditions. This includes intelligent modeling for unlabeled or variable-speed environments—challenges typical in modern industrial systems. His research directly supports smart manufacturing, predictive maintenance, and Industry 4.0 frameworks. 🧠📡🔩

📘 Publications

Pearson Coefficient Enhanced Multi-Branch Joint Attention Network and Adaptive Decomposition Based Dual Adaptive Fault Diagnosis Scheme for Rolling Bearing

👨‍🔬 Authors: Dongfang Zhao, et al.
📘 Journal: Signal, Image and Video Processing
📅 Year: 2025
📌 Overview: This paper presents a novel dual adaptive fault diagnosis framework leveraging Pearson correlation coefficients and a joint attention mechanism, significantly improving the accuracy of bearing fault detection under complex conditions.
📊🧠🛠️

Multiscale Attention Feature Fusion Network for Rolling Bearing Fault Diagnosis Under Variable Speed Conditions

👨‍🔬 Authors: Dongfang Zhao, et al.
📘 Journal: Signal, Image and Video Processing

📅 Year: 2024
📌 Overview: This study introduces a multiscale attention-based deep learning model that enhances feature fusion and generalization in rolling bearing fault diagnosis, particularly effective under varying speed scenarios.
⚙️🔍📉

Prof Can Li | Electrical Properties of Materials | Best Researcher Award

Prof. Dr. Can Li is an innovative researcher and Assistant Professor at the University of Hong Kong, specializing in neuromorphic computing, AI hardware, and memristor-based systems 🧬💾. He earned his Ph.D. in Electrical and Computer Engineering from UMass Amherst, after completing his M.S. and B.S. in Microelectronics at Peking University 📘🔌. With over 90 publications, HK$45M+ in grants, and multiple patents, his work bridges advanced electronics, machine learning, and brain-inspired systems 🌍📊. He actively mentors students, collaborates internationally, and serves on major editorial boards, shaping the future of intelligent computing technology 🚀🔍.

Prof Can Li, The University of Hong Kong, Hong Kong

Profile

SCOPUS

GOOGLESCHOLAR

Education 🎓

Prof. Dr. Can Li earned his Ph.D. in Electrical and Computer Engineering from the University of Massachusetts Amherst in 2018, under the mentorship of Prof. Qiangfei Xia 🧠📘. His dissertation, “CMOS Compatible Memristor Networks for Brain-Inspired Computing”, laid the foundation for his cutting-edge work in neuromorphic hardware ⚙️🧬. He holds both M.S. (2012) and B.S. (2009) degrees in Microelectronics from Peking University, China, mentored by Prof. Wengang Wu 📡🔌. His academic path combines solid-state electronics, AI hardware, and advanced semiconductor design, empowering his innovations in brain-inspired and in-memory computing systems 🌐🧑‍🔧.

Experience 💼

Prof. Can Li is currently an Assistant Professor at The University of Hong Kong (2020–Present) 🇭🇰, serving in the Department of Electrical and Electronic Engineering ⚡. His academic role focuses on advanced system design, hardware acceleration, and energy-efficient computing 💻🔋. Before this, he worked as a Research Associate at Hewlett Packard Labs in Palo Alto, California (2018–2020) 🇺🇸, contributing to architectural innovation within the System Architecture Lab 🧠🛠️. His industry and academic experience reflect a deep commitment to cutting-edge research in computer architecture and system performance optimization across real-world and theoretical applications 🚀📊.

Achievements & Innovations 🏅

Prof. Can Li has received numerous prestigious honors, including being ranked in the Top 1% worldwide by citations (Clarivate, 2022–2024) 🌍📊, the Croucher Tak Wah Mak Innovation Award (2023) 🧠🏆, the RGC Early Career Award (2021) 🧑‍🔬, and the NSFC Excellent Young Scientists Fund (2021) 🌟. He holds 17 granted patents across the US and China, focused on analog content-addressable memory (CAM), fuzzy search, optical TCAMs, and neuromorphic systems 🔧💡. These contributions demonstrate his pioneering work at the interface of hardware acceleration, AI computing, and next-gen memory systems 🚀🖥️.

Book Contributions 📘

Prof. Can Li has significantly contributed to the field of neuromorphic and in-memory computing through key book chapters in high-impact scientific texts. He co-authored “In-Memory Computing with Non-volatile Memristor CAM Circuits” and “Ta/HfO₂ Arrays for In-Memory Memristor Computing” in Memristor Computing Systems (Springer, 2022) 📘⚡. Additionally, he authored “Silicon Based Memristor Devices and Arrays” in the Handbook of Memristor Networks (Springer, 2019) 🔍🧠. These works showcase his pioneering role in advancing memristor technology and its integration into next-generation computational architectures 🖥️🧩.

Research Focus 🔬

Prof. Can Li’s research centers on cutting-edge innovations in artificial intelligence hardware and emerging memory technologies. His work focuses on developing AI inference and training accelerators using advanced memristor-based architectures 🧠💾. He explores novel applications of memristor crossbar arrays, including image processing, hardware security, and pattern matching 🖼️🔐🔍. A core part of his research also involves the CMOS-compatible integration of memristor devices at the array level, enhancing scalability and manufacturability ⚙️🔧. By bridging nanotechnology with AI computing, Prof. Li is advancing the future of energy-efficient, high-performance computing systems for next-generation intelligent electronics 🚀🖥️.

Publications 📚

Efficient Nonlinear Function Approximation in Analog Resistive Crossbars for Recurrent Neural Networks

Authors: Junyi Yang, Ruibin Mao, Mingrui Jiang, Can Li, Arindam Basu
Journal: Nature Communications, 2025

Current Opinions on Memristor-Accelerated Machine Learning Hardware

Authors: Mingrui Jiang, Yichun Xu, Zefan Li, Can Li
Journal: Current Opinion in Solid State and Materials Science, 2025
Emojis: 💾🧠🧩📈

Efficient Coherent Polarization Beam Combining of 16-Channel Femtosecond Fiber Lasers

Authors: Jiayi Zhang, Bo Ren, Can Li, Wenxue Li, Pu Zhou
Journal: Guangxue Xuebao/Acta Optica Sinica, 2025

Research Progress of Ultrafast Fiber Laser Amplifier Based on Gain Managed Nonlinearity (Invited)

Authors: Can Li, Bo Ren, Kun Guo, Jingyong Leng, Pu Zhou
Journal: Hongwai yu Jiguang Gongcheng/Infrared and Laser Engineering, 2025

Event-Based Multi-Object Tracking With Sparse Motion Features

Authors: Song Wang, Zhu Wang, Can Li, Xiaojuan Qi, Hayden Kwok Hay So
Journal: IEEE Access, 2025

An InGaZnO Synaptic Transistor Using Titanium-Oxide Traps at Back Channel for Neuromorphic Computing

Authors: B. F. Yang, Chen Zhang, Z. H. Zhang, Can Li, Xiaodong Huang
Journal: IEEE Transactions on Electron Devices, 2025