Neural network for current transformer saturation correction

Abstract

Current transformer saturation can cause protective relay misoperation or even prevent tripping. This paper presents use of artificial neural networks (ANN) to correct CT secondary waveform distortions. The ANN is trained to achieve the inverse transfer function of iron-core toroidal current transformers which are widely used in protective systems. The ANN provides a good estimate of the true (primary) current for a saturated transformer. The neural network is developed using MATLAB and trained using data from EMTP simulations, and data generated from actual CTs. In order to handle large dynamic ranges of fault currents, a technique of employing two sets of network coefficients is used. Different sets of network coefficients deal with different fault current ranges. The algorithm for running the network was implemented on an Analog Devices ADSP-2 101 digital signal processor. The calculating speed and accuracy proved to be satisfactory in real-time application.

Redes neurais artificiais recorrentes aplicadas na correção de sinais distorcidos pela saturação de transformadores de corrente

“…Estas publicações envolvem apenas a estimação das correntes primárias dos TCs, como as técnicas propostas em (Pihler et al, 1997;Yu et al, 1999;Cummins et al, 2000).…”

“…Assim como em (Yu et al, 2001), o trabalho apresentado por (Yu et al, 1999) emprega dois conjuntos de coeficientes para as RNAs a fim de lidar com diferentes faixas de correntes de falta.…”

“…Embora muitos trabalhos abordem o problema da saturação dos transformadores de corrente, (Kang et al, 1996;Kang et al, 1997;Jamali and Radji, 1997;Yu et al, 1999;Sidhu et al, 1999;Cummins et al, 2000;Yu et al, 2001), os métodos disponíveis até agora apresentam certas limitações. Engenheiros de proteção têm, portanto, que considerar essa inconveniência quando estabelecerem as especificações em termos do desempenho dos relés.…”

An alternative approach using artificial neural networks for power transformer protection

“…Some methods are used to avoid it, but some of the solutions can affect the cost of such a piece of equipment. The use of ANNs appears as an alternative to the correction of distorted signals by the CT saturation [20][21][22]. Figure 7 shows an energization case where the distortions caused by saturation can be clearly observed.…”

Development and hardware implementation of a reliable protective relay data acquisition system

“…They possess excellent features such as generalization capability, noise immunity, robustness, and fault tolerance. AI-based techniques have been used in power system protection and encouraging results are obtained [9][10][11][12][13][14][15][16][17][18][19].…”