Adaptive-Width Generalized Correntropy Diffusion Algorithm for Robust Control Strategy of Microgrid Autonomous Operation

Abstract

This study introduces a new method for achieving the robust performance of an isolated microgrid (MG) using an adaptive-width generalized correntropy diffusion algorithm (AWGC-DA). In the approach, the width of the kernel is adjustable, allowing the program to reject misleading input from attackers; the technique may identify attackers using a simple identification rule. The Response Surface Methodology is utilized in combination with three optimization algorithms: the Coot bird metaheuristic optimizer (COOT), the sunflower optimization (SFO), and the particle swarm optimization (PSO) to improve the technique's effectiveness. To assess the effectiveness of the new technique, a standard microgrid is tested under three different scenarios: (1) disconnected from the utility grid (autonomous status); (2) variations in load while in autonomous status; and (3) a three-phase fault while in autonomous status. In addition, numerous simulations are performed using the PSCAD/EMTDC software to verify the efficacy of the proposed technique. The main contribution of this study is to enhance the system transients by minimizing the overshoots, steady state error, and settling time which increases the efficacy of the autonomous MG process. Finally, to demonstrate its validity, the performance of the suggested approach is compared to that of other methods, including the LMSRE-based adaptive control, EBS-ABA, COOT, SFO, and PSO algorithms. The results confirm that the AWGC-DA method outperforms the other methods regarding the system's transient response.

Hybrid Transient Search Algorithm With Levy Flight for Optimal PI Controllers of Islanded Microgrids

“…To establish the superiority and advantages of the LTSO, this section provides a comparative analysis. It compares the results of the LTSO approach to those obtained through alternative control methods, including LMSRE, EBS-ABA, AWGC-DA, SFO, COOT, and PSO techniques [26]. The MG system was tested under different operational scenarios: i) transitioning the system into autonomous mode by disconnecting from the primary grid, ii) adapting to varying load conditions while isolated, and iii) responding to a 3phase fault while operating in islanded mode.…”

“…Researchers have invented many advanced optimization techniques to address these complications and optimize the control of MGs. Such as Enhanced Transient Search Optimization [22], the Coot bird metaheuristic optimizer (COOT) [19], the Enhanced Bald Eagle Search Algorithm [23], genetic algorithms [24], modified virtual rotor-based derivative technique supported with Jaya optimizer based on balloon effect [25], the Adaptive-Width Generalized Correntropy Diffusion Algorithm (AWGC-DA) [26], the Sunflower (SFO) algorithm [27], Enhanced Block-Sparse Adaptive Bayesian algorithm (EBS-ABA) [28], the Cuttlefish optimization algorithm [29], the ant colony algorithm [30], Circle Search Algorithm [31], the particle swarm optimization (PSO) [32], [33], and The least mean (LM) and the square root of exponential (SRE) [34]. These optimization methods aim to enhance decentralized controllers in MG systems, fine-tune parameters, and improve performance.…”

“…Table 2 shows the weights used in this study. The results of the LMSRE, EBS-ABA, AWGC-DA, SFO, COOT, and PSO are extracted from [26] and compared with the results of the proposed technique explained in the next section.…”