Power quality improvement of hybrid renewable energy systems-based microgrid for statcom: hybrid-deep-learning model and mexican axoltl dingo optimizer (MADO)

Abstract

Abstract The power generation from HRES, such as PV, FC, and Battery, is heavily dependent on weather conditions, resulting in PQ problems like voltage fluctuations, swells, harmonics, and sags. Rapid adjustment is therefore increasingly necessary in energy transmission and distribution networks. To address these issues, this paper proposes the use of STATCOM for reactive power compensation to mitigate voltage sag, swell, fluctuations, and THD. The paper discusses the modeling of HRES (PV-FC-Battery) and evaluates the fixed operating limit of the system with the integration of STATCOM. The main objective is to optimize gain parameters of FOPID controllers-based STATCOM control circuit using a hybrid deep learning model that includes CNN and RNN, and a hybrid meta-heuristic optimization model to improve voltage stability and response given the irregular nature of HRES. The optimized FOPID controller is evaluated to mitigate PQ issues such as voltage swell, fluctuation, sag, and harmonics. The proposed meta-heuristic optimization model, MADO, is a combination of MAO and DOA. The proposed approach is evaluated using power flow and power quality analyses on the IEEE 9 bus system’s power.