Grey wolf optimizer-assisted R-method-based weighted minimization for automatic generation control in two-area interconnected power system

P. J. Krishna,V. P. Meena,N. Patnana,A. Mathur,V. P. Singh
DOI: https://doi.org/10.1007/s40435-022-01070-9
2022-10-30
International Journal of Dynamics and Control
Abstract:In this article, design and performance analysis of grey wolf optimization algorithm (GWO)-assisted R-method-based proportional–integral–derivative controller with derivative filter is proposed for automatic generation control problem of two-area interconnected power system. The filter with derivative gain is used to nullify the impact of noise in the input signal. The objective function formulated for tuning of controller parameters is based on sub-objectives utilizing integral of time multiplied square error (ITSE) of frequency deviations, tie-line power deviation, and area control errors (ACEs). These sub-objectives are combined to form a single overall objective function. The overall objective function is weighted summation of ITSEs of two areas, ITSE of tie-line power deviation and ITSEs of ACEs of two areas. The weights in overall objective function are considered to evaluate each sub-objective with its relative importance in design of control. The weights in this article are derived using R-method systematically in contrast with other methods where the weights are either assumed equal by neglecting the relative importance of sub-objectives or taken arbitrarily. Therefore, the overall developed objective function is minimized utilizing the GWO algorithm. The effectiveness of the proposed GWO optimizer-assisted R-method-based controller is tested for six distinct scenarios involving different load disturbances in the interconnected areas. Further, the performance of GWO-tuned controller is compared with other controllers tuned using differential evolution, elephant herding optimization, Nelder–Mead simplex, Jaya, Luus–Jaakola, covariance matrix adaptation evolution strategy, and success history-based parameter adaptation for differential evolution) algorithms. For all the six scenarios, time domain specifications are tabulated. The obtained results are also plotted to show the variations in frequencies and tie-line power. Additionally, one statistical analysis is presented to evaluate the overall effectiveness of the proposed controller.
What problem does this paper attempt to address?