Rotor angle and frequency stability improvement of power system using artificial neural network controlled HVDC device

This paper presents an artificial neural network (ANN) based control strategy for improving rotor angle and frequency stability in power transmission networks. This approach combined an ANN controller within a high-voltage direct current (HVDC) device to provide adaptive modulation of the control signals and firing angle, thereby enhancing the damping characteristics of the system under disturbance conditions. The ANN model is designed to consider nonlinear system behavior and trained to optimize dynamic performance during transient events. The method is implemented and evaluated on a 40-bus, 330 kV Nigerian transmission network developed in MATLAB/PSAT (R2024a). Steady-state operating conditions are established using the Newton–Raphson power flow algorithm, while eigenvalue analysis is employed to determine critical bus locations. The Makurdi bus is identified as the most unstable point, exhibiting a positive eigenvalue of 3.3641 ± j5.2608 and a low damping ratio of 0.0564. A balanced three-phase fault is applied at this bus to assess system resilience. Comparative analyses between the proposed ANN-HVDC scheme and a conventional HVDC controller show that the ANN approach significantly enhances stability performance. The ANN-based control achieves synchronism recovery at a reduced critical clearing time of 2 ms, compared to 3 ms for the conventional scheme, thus preventing generator loss. Additionally, the ANN controller improves post-fault frequency recovery, increasing generator frequency from 46.5 Hz to 49.8 Hz and maintaining it within the acceptable 50 ± 0.5 Hz operational band. The results demonstrate that the proposed controller provides higher damping, improved transient response, and enhanced dynamic stability in weak power transmission networks.