Zhishang Wang

Zhishang Wang, Ph.D.

Division of Computer Engineering
Department of Computer Science and Engineering
The University of Aizu

Email: zwang@u-aizu.ac.jp
Tel.: +81 0242-37-2500 (Int. 3204)

Employment

Postdoctoral Researcher, The University of Aizu, April 2023 – Present.
AI Engineer, Aizu Computer Science Laboratories, January 2020 – March 2021.

Education

Research

AEBiS

A Virtual Power Plant (VPP) is a network of distributed power generating units, flexible power consumers, and storage systems. A VPP balances the load on the grid by allocating the power generated by different linked units during periods of peak load. Demand-side energy equipment, such as Electric Vehicles (EVs) and mobile robots, can also balance the energy supply-demand when effectively deployed. However, fluctuation of the power generated by the various power units makes the supply power balance a challenging goal. Moreover, the communication security between a VPP aggregator and end facilities is critical and has not been carefully investigated. An AI-enabled, blockchain-based electric vehicle integration system is developed for power management in a smart grid platform based on EV and solar carport. We have developed a low-power AI-chip and various software tools for EV charge prediction, in which the EV fleet is employed as a consumer and as a supplier of electrical energy.

Trustworthy Multi-Blockchain-Based Collaborative Learning

Collaborative edge learning has emerged in various domains like vehicular networks and medical care, allowing local model training on edge devices while preserving privacy. A hybrid clustered blockchain method (HCB) is proposed for collaborative edge learning, where model transmission is performed in an on-chain-merge-off-chain manner. Each cluster performs an off-chain transmission of local model updates and an on-chain distribution of global model updates, and elects a delegate node to serve as a model aggregator. The delegate nodes form a main blockchain in which the global model updates of each cluster are exchanged. A delegate-based adaptive model aggregation for robust collaborative learning called DAMA-RCL ensures high-quality model selection and aggregation during collaborative learning. A disassembling-reassembling method is also introduced to enable practical model transmission on the blockchain network.

Patent

Selected Publications

Invited Talk

Oral Presentation

Services

Awards

  • Outstanding Research Assistant (AY2021), The University of Aizu.
  • Best Presentation Award. Postgraduate Forum of ACM International Conference on Research in Adaptive and Convergent Systems 2022, Aizuwakamatsu, Japan. October 3-6, 2022.