Title :
Evolutionary improvement of neural classifiers for generator out-of-step protection
Author :
Rebizant, Waldemar ; Szafran, Janusz ; Feser, Kurt ; Oechsle, Fred
Author_Institution :
Inst. of Electr. Power Eng., Wroclaw Univ. of Technol., Poland
Abstract :
In the paper the results of investigation are presented on application of artificial neural networks to protection of synchronous machines against out-of-step conditions. The new ANN-based protection ensures faster and more secure detection of generator loss of synchronism when compared to traditional solutions. To determine the most suitable network topology for the recognition task a genetic algorithm is proposed. The final ANN structure is found using the rules of evolutionary improvement of the characteristics of individuals by concurrence and heredity. The proposed genetic optimisation principles have been implemented in Matlab programming code. The initial as well as further consecutive network populations were created, trained and graded in a closed loop until the selection criterion was fulfilled. The results of thorough testing of the designed protection scheme with ATP-generated power system signals are described and discussed
Keywords :
machine protection; neural nets; power engineering computing; relay protection; synchronous generators; ANN-based protection; ATP-generated power system signals; Matlab programming code; artificial intelligence; closed loop; concurrence; consecutive network populations; evolutionary improvement; generator out-of-step protection; genetic algorithm; genetic algorithms; heredity; inference mechanisms; neural classifiers; neural networks; protective relaying; selection criterion; synchronous generator stability; synchronous machines; transient analysis; Artificial intelligence; Artificial neural networks; Genetics; Impedance; Power generation; Power system protection; Power system relaying; Power system stability; Protective relaying; Synchronous machines;
Conference_Titel :
Power Tech Proceedings, 2001 IEEE Porto
Conference_Location :
Porto
Print_ISBN :
0-7803-7139-9
DOI :
10.1109/PTC.2001.964835