Design of a Power System Stabilizer Using Adaptive Neuro-Fuzzy Logic

Abstract

Power systems are large, complex and nonlinear systems, and often exhibit low-frequency power oscillations due to insufficient damping. PSSs are widely used to suppress the electromechanical oscillations of generators and thus, enhance the overall stability of power systems. The conventional PSS which uses lead-lag compensation and gain settings for particular operating conditions provides poor performance under different loading conditions. To mitigate the shortcomings of conventional PSS, Artificial Intelligence (AT) based techniques such as Fuzzy Logic Controller (FLC) and Adaptive Neuro-Fuzzy Inference System (ANFIS) have been introduced. FLC based PSS does not require mathematical model, possess lesser computational time, and ensures good performance under all conditions. However, speed and robustness are the most significant properties in comparison to the other classical schemes, there are no practical systematic procedures for the FLPSS design, so the rules and the membership functions of the controller are tuned subjectively, making the design laborious and a time-consuming task. On the other hand, an ANFIS based techniques is a promising method, which adjusts membership functions and rules adaptively to improve a systems performance. In literature, ANFIS-PSS is implemented for multi-machine power system having constant impedance load. The main contribution of this thesis is to introduce a PSS for a multi-machine interconnected power system comprising a dynamic load grounded on adaptive-neuro fuzzy logic technique. Load dynamics is included because it has significant impact on power system dynamic analysis. Transient stability analysis is more concerned with dynamic behaviors of loads. The robustness of ANFIS-PSS is being tested by applying single line to ground faults and symmetrical three phase faults. Another contribution of this thesis is to analyze the performance of using various input signals to PSS. It is found that the performance of an ANFIS-PSS depends on the input signal to PSS and for speed deviation and accelerating power as an input signal, the performance is quite far better compared to the selection of other inputs utilized in this work.