IoT Integration With CMPA-PINN for Islanding Detection Through Microgrid Hierarchical Control

Abstract

Internet of Things (IoT) and cloud computing are becoming increasingly important in the solution of many industrial problems. Effective management of microgrid (MG) requires a strong and scalable information and communication technology (ICT) infrastructure. IoT devices with effective measurement and control capabilities have the potential to be very important in the MG environment. MG was run in both grid-connected and island mode. This paper proposes to improve the MG hierarchical control with IoT using CMPA-PINN techniques for islanding detection. The proposed hybrid method is the joint execution of both the Coronavirus Mask Protection Algorithm (CMPA) and physics-informed neural networks (PINNs). Hence, it is named as CMPA-PINN approach. The major goal of this proposed method is to reduce the deviation of voltage, frequency, and total harmonic distortion (THD). The proposed CMPA is used to optimize the traffic flow over a communication network, and the PINNs are used to predict the optimized traffic flow. By then, the MATLAB platform has adopted the proposed method, and the current process is used to compute its execution. The proposed technique outperforms all current systems, including maximum power point tracking (MPPT), multi-agent reinforcement learning (MARL), and deep reinforcement learning (DRL). The proposed approach shows the THD is 2%, which is lower than other existing systems.