Volume 4, Issue 2

This paper presents a new approach to detection of high impedance faults in distribution systems using artificial neural networks. The proposed neural network, was trained by data from simulation of a distribution system under different faults conditions, and tested by data with different system conditions. The proposed neural network has been implemented on a digital signal processor board and its behavior is investigated on a computer power system model. Details of the design procedure, implementation and the results of performance studies with the proposed relay are given in this paper. Performance studies results show that the proposed algorithm performs very well in detecting a high impedance fault with nonlinear arcing resistance. It is clearly shown that with this integrated approach, the accuracy in fault detection is significantly improved compared to other techniques based on conventional algorithms.

In this paper, a novel neural network controlled Hybrid Parallel Active Power Filter has been presented. The neural network controller comprises two similar adaptive linear neurons (ADALINE), and it has been designed to extract fundamental frequency components from non-sinusoidal and unbalanced currents. It has been shown that the proposed neural controller is simple, fast and accurate. The proposed system is implemented using Digital Signal Processor (DSP). Experimental results with different nonlinear loads are presented to confirm the validity of the proposed technique.

The preset work tested a freely domain software for solving index zero and one systems of differential-algebraic equations, named as DASSL. The code encompasses an efficient variable step size and variable order based on BDF methods to solve a system of DAEs or ODEs. The code was applied in power systems time domain studies, i.e., synchronous machine angular transient stability and long-term voltage stability, using the Brazilian South-Southeast Equivalent Power System. Using this real power system model including fast and slow response control devices, it was possible to investigate the code capability in simulating different stability phenomena in the same run. The variable step size and variable order algorithm implemented in DASSL results in a very powerful tool for power system time domain computer simulation.

In this paper a novel control strategy for a real synchronous generator is presented. It is based on linear, diagonal, low order, minimum-phase, fixed, and stable controllers. The controller was obtained via individual channel design (ICD), a framework that allows analysis and synthesis of multivariable control systems by means of classical techniques based on the Bode and Nyquist plots. The generator is part of a one machine – infinite bus system. A linearised model around an operating point in a rotating coordinate frame is used. The theoretical principles behind this control strategy are summarised for completeness. The results using this novel control strategy are obtained throughout simulations in MATLAB®.

Flicker is an important power quality disturbance and has received an increasing concern from power system researchers. Interharmonics are the non-integral frequencies other than harmonic frequencies. Nowadays, research has shown that interharmonics and flicker seem to be closely related. To clarify this relationship, flicker is characterized in the frequency domain. The traditional Fourier-based methods have shown some drawbacks in representing non-stationary, non-periodic power signals and therefore other methods should be investigated for accomplishing this task. This paper introduces the most common signal processing approaches to assess the problem, power spectrum estimation methods and linear transforms. The wavelet transform has shown superior performance comparing to other methods and circumventing the problem time-frequency resolution.