Dynamic Congestion Management With Chance-Constrained MPC in Networked Microgrids Under Consumers-Related Uncertainties

Abstract

Congestion management involves controlling and optimizing energy flow to ensure efficient and reliable operation in energy systems. The objective of this paper is to effectively manage congestion in networked microgrids by balancing energy supply and demand, thereby preventing overloads and ensuring a stable and resilient energy system. One approach to achieve this is through Model Predictive Control (MPC), which can regulate power flow while considering realistic constraints and system dynamics. The results of applying linear MPC for congestion management in networked microgrids are promising when all parameters are deterministic. However, introducing uncertainty into the model poses challenges that linear MPC cannot address. This paper introduces a Chance-Constrained Model Predictive Control (CC-MPC) method to tackle this problem through mathematical reformulation and stochastic optimization. The decision to use CC-MPC over robust MPC or tube MPC was based on the unknown uncertainty of the problem, making stochastic MPC a more suitable option. The results from provided numerical examples demonstrate that the proposed method not only accurately predicts demand response but also effectively manages congestion in both small-scale and large-scale networked microgrids, even while accounting for the effects of uncertainty.