Application of fuzzy logic control for small-signal stability enhancement of power systems

Abstract

Fuzzy logic controllers can handle imprecise, vague or "fuzzy" information to a significant extent and capable of producing good results under changing operating conditions and uncertainties in system parameters or observed variables. Hence, fuzzy logic control techniques are suitable for large, interconnected power systems which usually possess a significant amount of uncertainties and experience frequent changes in operating conditions. An investigation of the application of fuzzy logic control techniques to enhance small-signal stability of electric power systems is undertaken in this thesis. Attention is given to development and identification of methodologies that can be utilised to apply fuzzy logic control for large, interconnected power systems. The difficulties in identifying a suitable rule base (relationships between input and control variables that are usually represented as a set of if ... then ... statements) and a set of parameters to realise desired performance are the main limitations in applying fuzzy logic control for large systems. A methodology, which uses the response of power system variables to pulse test signals applied at chosen instants, is proposed in this thesis to identify the suitable relationships that are needed to develop the rule base of fuzzy logic controllers. The advantage is that either the actual system or a simulated model can be used to derive the rule base using this technique. In addition, a simple and efficient iterative technique based on gradient search methods is proposed to determine the fuzzy logic controller parameters to achieve a desired control objective. Further, approaches that can be utilised to develop fuzzy logic controllers to realise a certain stabilisation objective without adverse side effects, and to coordinate several control options for enhanced system performance while avoiding detrimental interactions, are presented in this thesis. The Prony signal and transfer function identification techniques are utilised in this thesis to identify the critical oscillation modes, the suitable damping control options and feedback signals. These approaches along with sequential de-centralised controller design are proposed and successfully utilised in this thesis to develop fuzzy logic damping controllers for large interconnected power systems.