| dc.description.abstract |
Low-frequency oscillations (LFOs) in power systems arise from disturbances in the power
supply network and can negatively affect system stability. If not properly mitigated, these
oscillations may grow and lead to instability. Power System Stabilizers (PSS) are widely
used as supplementary control devices to damp such oscillations. However, the Gilgel
Gibe-I hydroelectric power plant experiences low-frequency oscillations with an
oscillation frequency of 1.5066 Hz despite the presence of a conventional power system
stabilizer (CPSS) in its existing system.this thesis aims to improve the damping of LFOs
and enhance the small-signal stability of the plant using an Adaptive Neuro-Fuzzy
Inference System (ANFIS)-based PSS, which does not require an explicit mathematical
model of the system. To evaluate the effectiveness of the proposed stabilizer, a linearized
model of the Gilgel Gibe-I hydropower plant was developed in MATLAB/Simulink. The
performance of the existing CPSS was compared with that of the proposed ANFIS-based
PSS through time-domain simulations. Rotor speed deviation, power angle deviation,
accelerating torque deviation, and electrical torque deviation were used as performance
indicators to evaluate system response, settling time, and steady-state behavior.simulation
results show that the proposed ANFIS-based PSS significantly improves system
performance compared with the CPSS. The settling time of rotor speed, rotor angle, and
electrical torque responses was reduced by 52.57%, 26.23%, and 37.16%, respectively.
These improvements were further confirmed using time-domain performance indices for
the Gilgel Gibe-I power plant model. The hybrid learning algorithm used in ANFIS
enables automatic adjustment of fuzzy rules and membership functions during training
using logged input data. By combining the learning capability of neural networks with the
uncertainty-handling ability of fuzzy logic, the proposed controller effectively enhances
damping of low-frequency oscillations and improves the small-signal stability of the
hydropower plant |
en_US |