Controllable series compensation (CSC) is one of the main flexible AC transmission systems (FACTS) devices, which have the ability to improve the utilisation of existing transmission systems. However, the implementation of this technology changes the apparent line impedance, which is controlled by the firing angle of thyristors, and is accentuated by other factors. This poses problems for conventional protection schemes. The paper proposes an adaptive protection scheme, based on neural networks, with special emphasis on the analysis of the zone-1 performance. The paper describes, in detail, the feature extraction, sampling rate, data window length and training of the designed artificial neural networks (ANNs). The main idea of the protection scheme is to employ an artificial neural network (ANN) to make a decision based on extracting useful features in the desired spectra within a certain frequency range under fault conditions. System simulation and test results are presented and analysed in this paper to indicate the feasibility of using an ANN-based protection scheme in CSC transmission systems.
[1]
A. T. Johns,et al.
Adaptive protection for series compensated EHV transmission systems using neural networks
,
1994
.
[2]
G. Swift,et al.
The Spectra of Fault-Induced Transients
,
1979,
IEEE Transactions on Power Apparatus and Systems.
[3]
Khashayar Khorasani,et al.
Fault identification in an AC-DC transmission system using neural networks
,
1991
.
[4]
Mark W. White,et al.
A neural network approach to the detection of incipient faults on power distribution feeders
,
1990
.
[5]
Zoran Obradovic,et al.
Training an artificial neural network to discriminate between magnetizing inrush and internal faults
,
1994
.