Bo Hu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Bo Hu has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
- Bram Noble
University of Saskatchewan
2 shared publications - Chi Yung Chung
The Hong Kong Polytechnic University
2 shared publications Greg Poelzer
School of Environment and Sustainability, University of Saskatchewan, Saskatoon, SK, Canada
2 shared publications- Yuzhong Gong
College of Control Science and Engineering, Zhejiang University, Hangzhou 310027, China
2 shared publications Chuanzi Xu
State Grid Hangzhou Power Supply Company,Hangzhou,China
1 shared publicationHong Li
State Grid Hangzhou Power Supply Company,Hangzhou,China
1 shared publication
Latest Publications
- Open Access
- 19 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 193581-193593
Safe Deep Reinforcement Learning-Based Real-Time Multi-Energy Management in Combined Heat and Power Microgrids
Combined heat and power microgrids (CHPMGs) have become increasingly popular recently due to their ability to offer cost-effective and resilient solutions to support critical infrastructures. As the centerpiece of a CHPMG, an autonomous real-time multi-energy management system can leverage advanced metering infrastructure to monitor and dispatch various distributed energy resources (DERs), minimize operational costs, and improve the overall system reliability. In this paper, a novel real-time energy management system (EMS) for CHPMGs is proposed using a data-driven model-free safe deep reinforcement learning method. The energy management problem in CHPMGs is first formulated into a constrained Markov decision process (CMDP). A safe deep deterministic policy gradient (SDDPG) method is then applied to solve the developed CMDP. SDDPG features an actor-critic structure, in which the actor network learns the optimal control policy, and the critic network learns to evaluate the state-action value. To satisfy CHPMGs operational constraints during the training process, two sets of neural networks are proposed to approximate the constrained system parameters. A mathematical optimization-based safety layer is also constructed upon the actor network to analytically correct the agent’s actions into safe actions satisfying specific constraints. The proposed method is validated by case studies using real-world data; it is also compared with conventional deep reinforcement learning (DRL) and optimization-based approaches with and without accurate uncertainty data.
- 19 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 193581-193593
Open AccessSafe Deep Reinforcement Learning-Based Real-Time Multi-Energy Management in Combined Heat and Power Microgrids
Combined heat and power microgrids (CHPMGs) have become increasingly popular recently due to their ability to offer cost-effective and resilient solutions to support critical infrastructures. As the centerpiece of a CHPMG, an autonomous real-time multi-energy management system can leverage advanced metering infrastructure to monitor and dispatch various distributed energy resources (DERs), minimize operational costs, and improve the overall system reliability. In this paper, a novel real-time energy management system (EMS) for CHPMGs is proposed using a data-driven model-free safe deep reinforcement learning method. The energy management problem in CHPMGs is first formulated into a constrained Markov decision process (CMDP). A safe deep deterministic policy gradient (SDDPG) method is then applied to solve the developed CMDP. SDDPG features an actor-critic structure, in which the actor network learns the optimal control policy, and the critic network learns to evaluate the state-action value. To satisfy CHPMGs operational constraints during the training process, two sets of neural networks are proposed to approximate the constrained system parameters. A mathematical optimization-based safety layer is also constructed upon the actor network to analytically correct the agent’s actions into safe actions satisfying specific constraints. The proposed method is validated by case studies using real-world data; it is also compared with conventional deep reinforcement learning (DRL) and optimization-based approaches with and without accurate uncertainty data.
- 22 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 15-21
Blockchain-Based and Secure Multi-Party Computation-Based Virtual Power Plant Management System
For the security transmission and information controllability and confidentiality when trading with virtual power plants and distributed nodes, ensuring the security of data communication between users and virtual power plants, as well as the difficulty of reliably correlating off-chain data and on-chain information and ensuring the credibility of data sources, this article proposes a virtual power plant management system and device based on blockchain and secure multi-party computation. Firstly, a hardware/software integrated terminal based on blockchain is developed to achieve terminal security certification and management, data collection and trusted computing, ensuring the authenticity and reliability of source data being uploaded to the blockchain.Secondly, using a secure multi-party computation protocol consisting of “secret sharing + homomorphic encryption + Pedersen commitment” to ensure that all parties participating in aggregated transactions in the virtual power plant maintain data privacy and security when data is uploaded onto the blockchain. In order to encourage protocol participants to remain loyal, an incentive mechanism is designed. Based on the secure multi-party computation encryption algorithm, this effectively addresses information security and data reliability issues that may arise when virtual power plant participants engage in aggregated transactions, ensuring the safe and stable transmission of transaction data.
- 22 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 15-21
Blockchain-Based and Secure Multi-Party Computation-Based Virtual Power Plant Management System
For the security transmission and information controllability and confidentiality when trading with virtual power plants and distributed nodes, ensuring the security of data communication between users and virtual power plants, as well as the difficulty of reliably correlating off-chain data and on-chain information and ensuring the credibility of data sources, this article proposes a virtual power plant management system and device based on blockchain and secure multi-party computation. Firstly, a hardware/software integrated terminal based on blockchain is developed to achieve terminal security certification and management, data collection and trusted computing, ensuring the authenticity and reliability of source data being uploaded to the blockchain.Secondly, using a secure multi-party computation protocol consisting of “secret sharing + homomorphic encryption + Pedersen commitment” to ensure that all parties participating in aggregated transactions in the virtual power plant maintain data privacy and security when data is uploaded onto the blockchain. In order to encourage protocol participants to remain loyal, an incentive mechanism is designed. Based on the secure multi-party computation encryption algorithm, this effectively addresses information security and data reliability issues that may arise when virtual power plant participants engage in aggregated transactions, ensuring the safe and stable transmission of transaction data.
- 31 August 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Smart Grid
- Vol. 12 (6) , 5201-5211
Price-Maker Bidding and Offering Strategies for Networked Microgrids in Day-Ahead Electricity Markets
This paper proposes a price-maker bidding and offering model for networked microgrids (NMG) in a pool-based day-ahead electricity market. The objective of this model is to maximize the net revenue of NMG by coordinating the joined individual microgrids to submit aggregated offers/bids to the market operator. A hybrid stochastic-robust optimization framework is developed to offset multiple associated uncertainties. The bidding and offering model is first formulated as a hard-to-solve mixed-integer nonlinear programming (MINLP) problem, which is later converted to its easy-to-solve mixed-integer linear programming (MILP) counterpart. To resolve privacy concerns of each microgrid and improve the scalability of the proposed bidding and offering model, a coordinated scheduling framework for NMG based on the Dantzig-Wolfe decomposition (DWD) method is proposed to obtain the global optimum. Numerical simulations with real-world measured data validate the effectiveness of the proposed price-maker bidding model, which is shown to outperform existing price-taker models.
- 31 August 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Smart Grid
- Vol. 12 (6) , 5201-5211
Price-Maker Bidding and Offering Strategies for Networked Microgrids in Day-Ahead Electricity Markets
This paper proposes a price-maker bidding and offering model for networked microgrids (NMG) in a pool-based day-ahead electricity market. The objective of this model is to maximize the net revenue of NMG by coordinating the joined individual microgrids to submit aggregated offers/bids to the market operator. A hybrid stochastic-robust optimization framework is developed to offset multiple associated uncertainties. The bidding and offering model is first formulated as a hard-to-solve mixed-integer nonlinear programming (MINLP) problem, which is later converted to its easy-to-solve mixed-integer linear programming (MILP) counterpart. To resolve privacy concerns of each microgrid and improve the scalability of the proposed bidding and offering model, a coordinated scheduling framework for NMG based on the Dantzig-Wolfe decomposition (DWD) method is proposed to obtain the global optimum. Numerical simulations with real-world measured data validate the effectiveness of the proposed price-maker bidding model, which is shown to outperform existing price-taker models.