Neuro-fuzzy controller based adaptive control for enhancing the frequency response of two-area power system

Islanded microgrids (IMGs) utilizing converters-based renewable energy resources face a major challenge in ensuring frequency stability due to their low inertia. Hence, IMGs need to have an efficient control strategy to ensure a proper equilibrium between power generation and load demand and maintain the system's frequency and voltage inside the permitted bounds during various disturbances. This paper introduces a streamlined control approach for two-area power systems, consisting of three tiers of control: local control (LC) at each source, secondary control (CC) for each area, and tertiary control (MCC) for the entire system. This approach simplifies control and allows for easier communication between the two-area power systems. The suggested control strategy successfully maintains the system's frequency and voltage under various disturbances, enhances dynamic performance, and ensures a good balance between power generation and load demand. Adaptive Neuro-Fuzzy Inference System (ANFIS) controller-model reference adaptive control (MR-ANFIS), a non-linear adaptive controller that can adapt to the various operational settings is suggested in this research for the frequency control loop of LC at each source to improve the transient response of the system frequency across all operating conditions. This paper highlights the effectiveness of the MR-ANFIS controller in regulating dual-area power system frequency and presents a comparative performance analysis with other control techniques like Fuzzy PI (FPI) and PI controllers. The suggested MR-ANFIS controller's efficiency is evaluated by analyzing the dynamic response of a two-area power system with minor disruptions (e.g., small step changes in weather conditions and load) and large disturbances (e.g., symmetrical short circuit faults occurrence and rejection of large amounts of load). FPI and PI controllers' parameters are optimally tuned using a recent optimization technique known as the Coati Optimization Algorithm (COA). The results obtained from various operational scenarios indicate that the proposed MR-ANFIS controller exhibits superior dynamic performance in regulating system frequency compared to traditional FPI and PI controllers. in scenarios involving minor disturbances, the MR-ANFIS controller demonstrates remarkable agility, achieving a reduction in overshoot by 95 %, undershoot by 96.97 %, settling time by 85.1 %, and ITAE by 73.22 %. Conversely, in scenarios involving larger disturbances, the MR-ANFIS controller still showcases enhanced performance, with reductions in overshoot by 27.79 %, undershoot by 38.23 %, settling time by 59.38 %, and ITAE by 40.73 %.