Loss of Excitation Protection of Medium Voltage Hydro Generators Using Adaptive Neuro Fuzzy Inference System

Abstract

This chapter presents a new method for loss of excitation (LOE) faults detection in hydro-generators using adaptive neuro fuzzy inference system (ANFIS). The investigations were done under a complete loss of excitation conditions, and a partial loss of excitation conditions in different generator loading conditions. In this chapter, four different techniques are discussed according to the type of inputs to the proposed ANFIS unit, the generator terminal impedance measurements (R and X) and the generator terminal voltage and phase current (Vtrms and Ia), the positive sequence components of the generator terminal voltage magnitude, phase current magnitude and angle (│V+ve│, │I+ve│ and ∟I+ve) in addition to the stator current 3rd harmonics components (magnitudes and angles). The proposed techniques' results are compared with each other and are compared with the conventional distance relay response in addition to other techniques. The promising obtained results show that the proposed technique is efficient.