- Faculty
-
Pengpeng Sun
Author： Source： Date：2024-03-14
Pengpeng Sun
(1987.03-)
Sex: Male
Member of the Communist Party of China
Birthplace: Dezhou, Shandong Province
Phone: (+86)13572484015
Email: jiea489756@163.com
Education Background
Ph.D. in Engineering, Computer Science and Technology, Northwestern Polytechnical University. Disciple of Professor Dong Yunwei, Doctoral Supervisor of Northwestern Polytechnic University, Distinguished Member of the Chinese Computer Society and Distinguished Member of IEEE (2013.09-2021.03)
Master of Power Electronics and Power Transmission, Northwestern Polytechnic University（2010.09-2013.03）
In 2010, he was admitted to the School of Automation of Northwestern Polytechnical University with a score of 428 and ranked first in the School of Automation.
Bachelor of Electrification Engineering and Automation, Qingdao University (2006-2010)
Areas of Research Interests/ Research Projects
Cyber-Physical Systems
Trusted Computing
Doctoral Work
Ph. D. Dissertation: Research on Key Technologies of CPS Preventive Control in Active Distribution Network
His Main Work:
1) The theory of cross-domain chain fault propagation and preventive control model for active distribution network CPS systems is proposed.
In order to characterize the normal composition and operational state behavior of the active distribution network CPS system, a model of the active distribution network CPS system based on the theory of hybrid automation is given.
In order to reduce the system state space, the continuous state of the active distribution network CPS system model is divided and correlated according to the fault time attributes, to obtain a graph model containing the physical and information variable attributes of the nodes and their connecting edges of the active distribution network CPS system as a cross-domain interlocking fault analysis model, which is used as the theoretical basis for cross-domain interlocking fault critical node identification and network robustness analysis.
In order to ensure the safe and stable operation of the system, a preventive control model based on finite automation is formulated on the basis of the cross-domain chain fault model to describe the impact of control measure interventions on the propagation of cross-domain chain faults.
2) A critical node identification method and evaluation method for preventing the occurrence of cross-domain interlocking faults in active distribution network CPS systems is proposed.
The method is based on a cross-domain interlocking fault model, using the assumption of interdependent functional subgraphs and the shortest path functional loop across the information network and the power network to measure the function of any node pair in the power network, and gives an evaluation formula to measure the function of the active distribution network CPS system, and then assesses the loss of function of the active distribution network CPS system affected by node faults.
The experiments are compared with 11 commonly used node importance measures through grid arithmetic examples, and the experiments show that the node importance assessment method proposed in the paper has good accuracy and effectiveness.
3) A robustness analysis method for cross-domain interlocking faults in active distribution network CPS systems is proposed.
The method is based on a cross-domain interlocking fault model, which takes into account factors such as interdependencies and overload faults in active distribution network CPS systems, and quantifies the behavior of the fault propagation process when it reaches steady state using robustness indicators.
The theoretical analysis shows that the robustness towards cross-domain interlocking faults is negatively related to the random initial fault rate of the nodes; it is positively related to the number of distributed power sources contained in the power network, to the average control access of each node in the power network, to the average power supply access of each node in the information network and to the values of the nodes' tolerance parameters.
The paper builds a cross-domain interlocking fault simulation platform for active distribution network CPS systems and verifies the above relationships numerically.
4) A reinforcement learning-based interruption of cross-domain chain fault prevention control framework and control strategy generation method is proposed.
Based on the preventive control model, the method aims at reducing the cross-domain interlocking fault propagation breadth and depth respectively, controlling the active distribution network CPS system from the initial steady state to the most desired steady state according to the principle of the shortest possible cross-domain interlocking fault propagation sequence, improving the robustness towards cross-domain interlocking faults as well as reducing the control cost, giving the compatibility theorem of the preventive control policy through the deterministic control policy gradient, the approximate action value function Q and the preventive control policy The compatibility theorem and the convergence of the preventive control policy to a locally optimal solution theorem are given to ensure the correctness and suboptimality of the generated preventive control policy.
Finally, it is shown that the reinforcement learning-based method for generating preventive control policies to block cross-domain chain fault propagation is better than the existing methods using multiple intelligences in reducing the number of fault nodes and avoiding state space explosion, as verified by the CEPR-CPS algorithm for a practical application system disclosed by the Chinese Academy of Electrical Sciences.
Scientific Achievement
Ÿ Published Papers
[1] Pengpeng Sun, Yunwei Dong, Chong Wang, et al. Cyber–Physical Active Distribution Networks Robustness Evaluation against Cross-Domain Cascading Failures. Appl. Sci., 2019, 9, 5021. (CAS Zone3, SCI)
[2] Xiaomin Wei, Yunwei Dong, Pengpeng Sun, and Mingrui Xiao. Safety analysis of AADL models for grid cyber-physical systems via model checking of stochastic games. Electronics, 8(2):212, 2019. (CAS Zone3, SCI)
[3] Pengpeng Sun, Yunwei Dong, et al. Preventive Control Policy Construction in Active Distribution Network of Cyber-physical System with Reinforcement Learning. Applied Sciences. 2020,12.(published) (CAS Zone3, SCI)
[4] SUN Peng-peng, DONG Yun-wei. Modeling and Optimizing Method of Two-Dimensional Omni-directional Wireless Charging Cyber-Physical Systems. Acta Electronica Sinica.Vol.47, Issue (3), p568-575, 2019. (CCF A-Class Scientific Journal in Chinese, EI)
[5] PengpengSun, Yunwei Dong, et al. A Wireless Power Transfer CPS Based on 2D Omni-directional Rotating Magnetic Field[C]//2016 IEEE International Conference on Software Quality, Reliability and Security Companion (QRS-C). IEEE, 2016: 252-258. (CCF C-Class Conference)
[6] Pengpeng Sun, Yunwei Dong, et al. On Vulnerability Analysis of Nodes against Cross-domain Cascading Failures Propagation in Active Distribution Networks. International Conference on Dependable Systems and Their Applications (DSA 2020). 2020.11. (EI Conference publication)
[7] Chong Wang, Yunwei Dong, PengpengSun. Cascading Failure Path Prediction based on Association Rules in Cyber-Physical Active Distribution Networks. SSCPS.2020,12. (EI Conference publication)
[8] Pengpeng Sun, Yunfeng Li, BO Liang. Solar-Powered Temperature Acquisition Based on Zigbee[J]. MECHATRONICS, No.3, 2013, p70-73.
Granted Patents
[1] A System Based on Self-Reconfiguration of Battery Pack and Its Intelligent Control Method, Patent No. ZL 2014 1 0654747.9 (Invention Patent, with the student as the first inventor)
[2] A Task Scheduling Method of the Embedded Operating System for Energy Harvesting System, Patent No. ZL 2014 1 0133089.9 (Invention Patent)
[3] Dong Yunwei, Sun Pengpeng, Yuan Sen, Wang Chong. Preventive Control Strategy and Fault Mode Intelligent Matching Method and Process for Active Distribution Network CPS System, Grant No.: 2021109302670.
[4] Sun Pengpeng, Dong Yunwei, Wang Chong, Yuan Sen. A Method for Discriminating Cross-Domain Interlocking Fault Source Nodes in Active Distribution Grid CPS, Grant No. 202110516668.
Honors and Awards
Winner of the China OI Underwater Robotics Competition, Project Director (2013.9)
Third Prize in the Air-to-Air Missile Cup, Project Director (2013.12)
Second Class Scholarship (2011 - 2012,2014 - 2016)
2010 and 2013 First Class Scholarship
Joined Scientific Projects
Ÿ Vertical Projects
Project of National Key R&D Program of China, "Fundamental Theories and Methods for Analysis and Control of Information-Physical Systems in Power Grids" (Basic Research Category, Project No. 2017YFB0903000), Key Student Member (2017.07-)
Mainly responsible for: Research on Key Technologies for Multi-Objective Prevention and Control of Interlocking Faults across Space in Active Distribution Networks CPS.
Ÿ Transversal Projects
He has participated in the design and development of a robotic arm control box, aircraft generator control (GCU) test bench development, underwater ROV, quadcopter design and development, Labview-based wind turbine monitoring upper computer design and development as project leader and key R&D staff.