Research

Autonomous coordination and control of smart converters for sustainable power systems (Link)

(Postdoctoral research is funded by the program)

To achieve the national target of 100% renewables by 2040, renewables are increasingly integrated into electric power systems. Unfortunately, intermittent renewables increase the risk of grid instability with voltage fluctuations, frequency deviations, and inertia issues, which limit the further integration of renewables. The renewable interface converters offer new promising methods to provide various functions for grid support. To efficiently and successfully utilize the grid support capabilities of the converters requires optimized coordination of a large number of converters. However, optimal coordination is a significant challenge due to limited communication support, multi-timescale operation, various real-time control actions, and computational complexity.

This project aims at developing smart autonomous power converters for optimal support of electric power systems. The project will build a new control framework based on digitalization and AI that will provide optimal grid support functions and enable further integration of renewable energy sources. This is achieved by developing a novel combined optimization and control algorithm for coordination and an AI-based scheme for autonomous control of smart converters.

Reference:

1. B. Li and Q. Xu, "A Machine Learning-Assisted Distributed Optimization Method for Inverter-Based Volt-VAR Control in Active Distribution Networks," in IEEE Transactions on Power Systems (Early Access).

2. Jan Kronqvist, Boda Li, and Rolfes Jan. "A mixed-integer approximation of robust optimization problems with mixed-integer adjustments." arXiv preprint arXiv:2302.13962 (2023).

3. Jan Kronqvist, Boda Li, and Rolfes Jan. "Alternating mixed-integer programming and neural network training for approximating stochastic two-stage problems." arXiv preprint arXiv:2305.06785 (2023).

Background - Resilience crisis in industrial systems

Integrating cyber systems, power grids and other urban energy systems makes the future power system (FPS) more complex. Also, wide applications of information and communications technology have also introduced extra vulnerabilities to the system.  As the complexity of the FPS increases, the problem of resilient operation becomes more severe. Some less serious failures may induce wide-scale power outages; failures in one system may propagate to other tightly coupled systems. Besides, the FPS involves great risks induced by external disturbances, such as malicious attacks and extreme meteorological disasters. All these high-impact-low-probability events urgently require us to enhance the resilience of FPS.

Resilience enhancement of cyber-physical-power-system

Nowadays, by strengthening the fusions of information and physical facilities, the power system is evolving into a cyber-physical system. Aiming at creating catastrophic damages by inducing measurement and control errors, sophisticated attack schemes become threats to the security operation of the whole cyber-physical power system and weaken the resilience of the system. For this reason, we design a lot of methods to eliminate the adverse effects raising by the attacks:

• We noticed that the false data injection attacks have the feature of sparsity and low ranks. Thus, we designed a detection algorithm with the Fast Go-Decomposition method to identify the injected false data [1]. Then, a real-time detection method is developed by exploring the spatial-temporal relationships among measurements [2].

• Furthermore, we apply the game theory to the process of attack and defense in the system. We use the graphical evolutionary game to describe the propagation of anomaly events [3]. Multi-stage incomplete information non-cooperative game between attackers and defenders is modeled with partially observable Markov decision process to determine the best defense strategy [4]. Diversified software deployment is also considered and solved by the cooperative game to further enhance the system resilience [5].

Reference:

1. B. Li, T. Ding, C. Huang, J. Zhao, Y. Yang, and Y. Chen, "Detecting false data injection attacks against power system state estimation with fast go-decomposition approach," IEEE Transactions on Industrial Informatics, vol. 15, no. 5, pp. 2892-2904.

2. B. Li, Y. Chen, S. Huang, S. Mei, Z. Wang, and J. Li, "Real-time detecting false data injection attacks based on spatial and temporal correlations," in 2019 IEEE Power & Energy Society General Meeting (PESGM), pp. 1-5.

3. B. Li, Y. Chen, S. Huang, R. Yao, Y. Xia, and S. Mei, "Graphical evolutionary game model of virus-based intrusion to power system for long-term cyber-security risk evaluation," IEEE Access, vol. 7, pp. 178605-178617.

4. Z. Zhang, S. Huang, Y. Chen, B. Li and S. Mei, "Cyber-Physical Coordinated Risk Mitigation in Smart Grids Based on Attack-Defense Game," IEEE Transactions on Power System (Early Access).

5. Z. Zhang, S. Huang, R. Yao, Y. Chen, B. Li and S. Mei, "Diversified Software Deployment for Long-Term Risk Mitigation in Cyber-Physical Power Systems," IEEE Transactions on Power System (Early Access).

Resilience enhancement of urban energy system

The future power system becomes more and more complicated. Gas networks, heat networks and electrified transportation networks are integrated to form the new urban energy system. Noticing that extreme meteorological disasters occur frequently, it is important to enhance the resilience to defend against natural disasters. Our research on resilience can be divided into two parts, one is the risk analysis about disaster-induced damages, and the other is resilience enhancement measures. The works are introduced as follows:

• We studied the impact of different meteorological disasters on distribution systems and used the Bayesian network model to calculate the damage risks of system outages [1]. On this basis, the post-disaster damage inference method was then proposed [2].  Then, we analyze how the availability of distribution terminal functions influences system resilience [3]. 

• We noticed that mobile energy carriers had great potential for disaster-related resilience enhancement. Thus, we used idle electrical buses to provide power supply for outage areas [4][5]. During the bus dispatch, the power supply and the public transition services are both considered, achieving a joint restoration.

Reference:

1. B. Li, Y. Chen, S. Huang, H. Guan, Y. Xiong, and S. Mei, "A bayesian network model for predicting outages of distribution system caused by hurricanes," in 2020 IEEE Power & Energy Society General Meeting (PESGM), pp. 1-5.

2.  Y. Xiong, G. Zhou, Y. Chen, L. Chen, M. Zhang, and B. Li, "A quick disaster inference in power distribution network based on tree-like Bayesian network (In Chinese)," Power System Technology, vol. 44, no. 6, pp. 2222-2230.

3. B. Li, Y. Xiong, Z. Ren, Y. Chen, and L. Zhang, "Distribution network reconfiguration method considering availability of distribution terminal functions during typhoon disaster (In Chinese)," Automation of Electric Power Systems, vol. 45, no. 4, pp. 38-44.

4. B. Li, Y. Chen, W. Wei, S. Huang, Y. Xiong, S. Mei, and Y. Hou, "Routing and scheduling of electric buses for resilient restoration of distribution system," IEEE Transactions on Transportation Electrification (Early Access).

5. B. Li, Y. Chen, W. Wei, S. Huang, and S. Mei, "Resilient restoration of distribution systems in coordination with electric bus scheduling," IEEE Transactions on Smart Grid, vol. 12, no. 4, pp. 3314-3325.