A Robust Method of Tuning a Decentralized Proportional-Integral Load Frequency Controller in a Deregulated Environment Using Genetic Algorithms

Abstract

In this article, load frequency control of an interconnected power system is achieved by two methods. The first is based on the classical H 1 control method and is subjected to linear matrix inequalities. The second uses a proportional-integral controller that is tuned by genetic algorithm optimization and is also subjected to the same linear matrix inequalities in order to obtain robustness against disturbances. Both controllers are tested on a two-area power system with three scenarios of load disturbances, as well as on power systems in a deregulated environment to demonstrate their robust performances.

Effect of electric vehicles on load frequency control in interconnected thermal and hydrothermal power systems utilising CF‐FOIDF controller

“…It is gathered from Δω 1 /ΔPtie 12 responses shown in Figs. 8a and b and Δω 1 /Δω 2 /ΔPtie 12 /J S results revealed in Table 3 that the suggested CF-FOIDF controller offers much superior control performances than the GA tuned PI [17], multi objective GA (MOGA) tuned PID [18], modified evolutionary particle swarm optimisation-time varying acceleration coefficient (MEPSO-TVAC) tuned PID [19], LCOA tuned PID [26], hPSO-LFA tuned PID [34] and ICA tuned FPI/FFOPI (this study) controllers regarding magnitude of oscillations, T S (Δω 1 = 0.50 s, Δω 2 = 1.11 s, ΔPtie 12 = 4.39 s), U S (Δω 1 = 0.0008 pu, Δω 2 = 0.00005 pu, ΔPtie 12 = 0.0008 pu), J S (ISE = 2.95 × 10 −6 , ITSE = 6.96 × 10 −6 , IAE = 0.0053, ITAE = 0.0170) and stabilised responses even when the degree of SLD is extended from 1 to 20%. However, more refined system responses are obtained with CF-FOIDF when the effect of EVs is considered in the control areas.…”

“…Via suitable control arrangements, LFC reinstate the system stability and preserve the frequency/power at anticipated values. Various optimal, robust and intelligent control methodologies as stated few above are utilised as potential solutions to get a robust performance and stability of real PSs [12][13][14][15][16][17][18][19][20][21][22][23][24][25][26]. These include hybrid human brain emotional learning PI [12], sine-cosine algorithm based on wavelet mutation (SCAWM) based model-free non-linear sliding mode controller (MFNSMC) [13], hybrid SCA-HS algorithm based FO-SMC [14], firefly algorithm-pattern search (hFA-PS) tuned PI/PID [15], hybrid invasive weed optimisation-PS (hIWO-PS) tuned PI/2-DOF-PID [16], multi-objective genetic algorithm (MOGA)/GA tuned PI/PID [17,18], modified evolutionary particle swarm optimisation-time varying acceleration coefficient (MEPSO-TVAC) tuned PID [19], dragonfly algorithm (DA) tuned PID/2DOF-PID [20], blended biogeography based optimisation (BBBO) tuned PID [21], grey wolf optimisation (GWO)/ensemble of mutation and crossover strategies and parameters in differential evolution (EPSDE) tuned PI/PID [22], hybrid gravitational search algorithm-PS (hGSA-PS) tuned PI/PID with filter (PIDF) [23], salp swarm algorithm (SSA) tuned PIDF/ tilt IDF (TIDF)/cascade control-TIDF (CC-TIDF) [24], differential evolution (DE) tuned PID/TIDF [25] and lozi map-based chaotic optimisation algorithm (LCOA) tuned PID [26] controllers applied on different PS configurations.…”