System collapse and power supply interruptions caused by extreme weather, has largely affected the resiliency of power distribution systems. Currently, forming microgrids with distributed generators have been employed to resolve these challenges. They are highly valued in terms of fast response and restoration of critical loads. Technically, the load recovery process comprises three key steps, including failure isolation, initial load recovery, and finally, the optimal microgrid formation process. In most cases, the two initial steps are often easy to implement because they do not normally require any optimization, as in the case with the third step. However, optimization of the microgrid formation processes has remained a challenge to date despite the development of different resilience-oriented microgrid formation methods based on heuristic search and mathematical programming approaches.
To address the above challenges, researchers from Hohai University: Professor Yue Yuan and led by Dr. Junpeng Zhu, together with Professor Weisheng Wang from China Electric Power Research Institute, proposed an exact microgrid formation model for effective optimization of the microgrid operation step of the load recovery process, employing mathematical programming methods. Their work is currently published in the International Journal of Electrical Power & Energy Systems.
The research team evaluated the power balance and operational feasibility of the distribution systems by considering power loss and voltage constraints. It is worth mentioning that additional considerations, including the scale of binary variables, linear and nonlinear characteristics, and the load controllability assumptions were employed to improve the modeling and computational accuracy. In their work, binary variables and non-linear equations were used to represent the reconfigured network and exact power flow, respectively, making the microgrid formation model a mixed-integer non-linear program.
Cases studies were conducted to validate the feasibility of the model by comparing the obtained results to the numerical results based on IEEE 33 and PE&G 69 standard distribution systems. The authors observed that the proposed method could effectively and preciously obtain optimal power solutions by avoiding the under-voltage and power shortage problems commonly associated with microgrids formatted by existing linearized DistFlow-based methods. Unlike the traditional methods, the proposed method can recover more load since it can achieve microgrids with lower power loss. Additionally, exactness, computational efficiency, and convergence analysis revealed that the model has a high performance and solving process. The time solving of the proposed approach depends on the iteration times and scale of integer variables.
In summary, the study presented a novel microgrid formation model for the purposes of load restoration in resilient distribution systems. From the findings, it can be concluded that the proposed model can recover more load efficiently and, therefore, more advantageous than traditional and existing linearized DistFlow based methods. In a statement to Advances in Engineering, Dr. Junpeng Zhu highlighted that their findings would give a better concept of load recovery and optimization of microgrid formations to minimize the collapse and power interruption in the distribution systems.
Zhu, J., Yuan, Y., & Wang, W. (2020). An exact microgrid formation model for load restoration in resilient distribution system. International Journal of Electrical Power & Energy Systems, 116, 105568.