Tuning of controller parameters for suppressing low frequency oscillations in electric railway traction networks using meta-heuristic algorithms

ABSTRACT

Due to the interaction of electric multiple units (EMUs), and the electric traction networks, low frequency oscillations (LFOs) appear leading to traction blockade and overall stability related issues. For suppressing LFOs, coronavirus herd immunity optimiser (CHIO), a recently developed meta-heuristic, has been applied for tuning controller parameters. Controller parameters are tuned to minimise the integral time absolute error (ITAE) that regulates DC-link capacitor voltage. Results obtained using CHIO are compared with those found using other well-established algorithms like symbiotic organisms search (SOS) and particle swarm optimisation (PSO). The supremacy of CHIO over other mentioned algorithms for mitigating LFOs was demonstrated for a diverse range of operating conditions. Results demonstrates that overshoot for the proposed algorithm-based traction unit is 1.0061% whereas those for SOS and PSO based algorithm are obtained as 6.4542 % and 20.6166%, respectively which are quite high. CHIO is more stable than SOS and PSO and requires settling time of 0.1934 s only to reach steady-state condition, which is 50.21% faster than SOS and 65.03% faster than PSO. Also, the total harmonic distortion (THD) for line currents of the secondary side of traction transformer (TT) are obtained as 0.88%, 2.17%, and 12.48% for CHIO, SOS, and PSO, respectively.