Adaptive Control for Virtual Synchronous Generator Parameters Based on Soft Actor Critic.

This paper introduces a model-free optimization method based on reinforcement learning (RL) aimed at resolving the issues of active power and frequency oscillations present in a traditional virtual synchronous generator (VSG). The RL agent utilizes the active power and frequency response of the VSG as state information inputs and generates actions to adjust the virtual inertia and damping coefficients for an optimal response. Distinctively, this study incorporates a setting-time term into the reward function design, alongside power and frequency deviations, to avoid prolonged system transients due to over-optimization. The soft actor critic (SAC) algorithm is utilized to determine the optimal strategy. SAC, being model-free with fast convergence, avoids policy overestimation bias, thus achieving superior convergence results. Finally, the proposed method is validated through MATLAB/Simulink simulation. Compared to other approaches, this method more effectively suppresses oscillations in active power and frequency and significantly reduces the setting time.

Application of an optimal tilt controller in a partial loading schedule of multi-area power system considering HVDC link and virtual inertia.

Maintaining the reliability of the modern power system is becoming increasingly challenging due to the continuous growth in its size and complexity. One of the most significant challenges PS faces is frequency regulation, which is essential for maintaining a stable and reliable operation of PS. This article investigates the automatic generation control (AGC) of a multi-area PS considering partial loading scheduled in a hydro and thermal generating unit. A bird swarm algorithm (BSA) based two-degree of freedom tilt integral derivative (2DOF-TID) controller has been suggested. The superiority of the optimal 2DOF-TID controller has been selected by comparing the system dynamics with other conventional controllers. Also, the ascendancy of suggested BSA has been shown over the grasshopper algorithm and particle swarm optimization techniques. The investigation on partial loading schedules for the hydro-thermal system showed that the system's dynamic response deteriorates when loading on hydro units exceeds their nominal loading. Also, system dynamics are enhanced when thermal units are loaded beyond their capacity. The investigations are also conducted to analyse the effect of high voltage direct current (HVDC) link and virtual inertia emulation with an energy storage system. The studies with alternating current (AC)/HVDC and virtual inertia revealed that dynamic responses are better with the latter when compared with AC and HVDC links alone.

Sine augmented scaled arithmetic optimization algorithm for frequency regulation of a virtual inertia control based microgrid.

The biggest challenge associated with a renewable energy source (RES) based microgrids is to maintain their frequency stability. In solving this challenge, virtual inertia control (VIC) could be viewed as an unavoidable part of alternating current (AC) microgrids. For VIC, a phase-locked loop (PLL) is essential for obtaining information regarding the changes in the frequency of the microgrid. However, implementation of PLL might cause greater oscillation in frequency because of its system dynamics. Such type of issues can be resolved by using a multistage proportional integral derivative (PID) controller which restricts the undesirable frequency measurement and thus helps in improving the stability of the microgrid. In this paper, a novel Sine augmented scaled arithmetic optimization algorithm is proposed in purpose to tune the parameters of the aforementioned controller. The effectiveness of the proposed methodology is validated through contrastive simulation results, and the impact of a few standard strategies such as a change in system boundaries and various steps of RESs penetration are also demonstrated.