Learning Robust Model Predictive Control for Voltage Control of Islanded Microgrid

This paper proposes a novel control design for voltage tracking of an islanded AC microgrid in the presence of nonlinear loads and parametric uncertainties at the primary level of control. The proposed method is based on the Tube-Based Robust Model Predictive Control (RMPC), an online optimization-based method which can handle the constraints and uncertainties as well. The challenge with this method is the conservativeness imposed by designing the tube based on the worst-case scenario of the uncertainties. This weakness is amended in this paper by employing a combination of a learning-based Gaussian Process (GP) regression and Tube-Based RMPC. The advantage of using GP is that both the mean and variance of the loads are predicted at each iteration based on the real data, and the resulted values of mean and the bound of confidence are utilized to design the tube in Tube-Based RMPC. The theoretical results are also provided to prove the recursive feasibility and stability of the proposed learning based Tube-Based RMPC. Finally, the simulation results are carried out on both single and multiple DG (Distributed Generation) units. Note to Practitioners —In this paper, we present a new way to control the voltage in an islanded microgrid to improve Power Quality (PQ). The method we propose is based on an online optimization technique called Tube-Based Robust Model Predictive Control. It can handle uncertainties and disturbances that occur when the microgrid operates independently, ensuring the voltage remains stable. However, there's a challenge with this method. It tends to be too cautious because it assumes the worst-case scenario for uncertainties. To make the control more efficient, we improve it by combining a learning-based technique called Gaussian Process regression with Tube-Based RMPC. The advantage of using GP is that it predicts the uncertainty of the electrical devices based on real data. We use these predictions to design the control in Tube-Based RMPC more accurately. We also provide theoretical results to show that our new learning-based control is reliable and stable. We tested our approach through computer simulations on different scenarios with one or multiple power sources in the microgrid. The results show the effectiveness of our control design in regulating the voltage even with uncertain and nonlinear loads. Overall, this paper suggests a practical and reliable way to control the voltage in an independent microgrid using a combination of online optimization and learning techniques.