Enhancement of AVR system performance by using hybrid harmony search and dwarf mongoose optimization algorithms

Innovations in control algorithms, integration of smart grid technologies, and advancements in materials and manufacturing techniques all push the boundaries of AVR performance. As the demand for power systems progresses with the complexity and variety of loads, conventional AVR designs may struggle to handle these ever-changing circumstances efficiently. Therefore, the need for new optimization methods is crucial to bolstering the efficiency, reliability, and adaptability of AVRs. Thus, this work aims to improve the performance of the AVR system controller by using a novel hybrid technique between the Harmony Search (HS) and Dwarf Mongoose Optimization (DMO) algorithms to tune the proportional-integral-derivative (PID) and proportional-integral-derivative acceleration (PIDA) parameters. The suggested hybrid approach ensures an accurate solution with balanced exploration and exploitation rates. The reliability of the proposed HS-DMOA is verified through comparison with different optimization techniques carried out on time and frequency performance indicators, disturbances in the form of changes to time constants, and dynamic input signals. The proposed hybrid HS-DMOA PID-based has better overshoot than PID-based HS, LUS, TLBO, SMA, RSA, and L-RSAM by 20.37%, 18.5%, 18.5%, 2.77%, 5.55%, and 2.77%, respectively. Regarding the phase margin, the proposed hybrid HS-DMOA PID-based is better than PID-based HS, LUS, and TLBO by 39%, 37%, and 38%, respectively. While the proposed hybrid HS-DMOA PIDA-based has a better overshoot than PIDA-based HS, LUS, and PID HS-DMOA-based by 14%, 17%, and 20%, respectively. Moreover, the robustness under dynamic disturbance proved the reliability of the proposed HS-DMOA PID and PIDA based through enhancement of overshoot around 0.3%~20% for different cases. Finally, the main contribution of the paper is to propose a relatively new hybrid optimization method to enhance the AVR PID and PIDA-based performance with detailed analysis in time and frequency domains under normal and dynamic disturbances.