Detection and localization of disk-to-disk short circuits in transformer HV windings using an improved model

Abstract

Summary This paper presents a new improved detailed model for transformer windings. This model accurately represents the transient behaviour of transformer winding better than does the conventional detailed model. The dominant resonance frequencies can be identified using the conventional model; however, nondominant resonance could not be detected. The proposed model can detect both dominant and nondominant resonance frequencies. Frequency response analysis was applied to a 20/0.4 kV 1600 kVA transformer. The simulation results obtained by the conventional and proposed models were compared with the measurement results. The results show the ability of the improved version to model dominant and nondominant resonance. A heuristic method is also proposed to overcome the problem of estimating parameters in large‐scale models.

Evaluation of effectiveness of second-order modal interaction in design procedure of synchronous machines controllers

“…To find the best parameter set for controllers, an intelligence algorithm may be used to determine the optimum values that results the most stable inter‐area mode in addition to minimum NI. Genetic algorithm (GA) as a widely used and tested method in optimization problem is used in this step. The objective function is formed by the weighted combination of SI and NI (F SINI ).…”

Reproducing Transformers’ Frequency Response from Finite Element Method (FEM) Simulation and Parameters Optimization

“…The equivalent RLC circuits use lumped elements such as resistances, self and mutual inductances and capacitances estimated from analytical calculations, or FEM simulations [16,[19][20][21][22]. Lumped-circuit elements are widely used in various transformers/inductors modelling applications [23,24].…”