An Online Dynamic Security Assessment in Power Systems Using RBF-R Neural Networks

ABSTRACT

In this paper, the dynamic security of a large power system against any critical contingency is predicted by a new type of radial basis function neural network, RBF-R NN, as it classifies the system’s transient stability status online. In order to keep the number of measurements limited, as well as to reduce the complexity of the NNs used, the minimum redundancy maximum relevance is adopted as a feature selection method. Moreover, the classification performance of the RBF-R NNs is improved by eliminating the training set instances that are close to the security boundary. The proposed method is applied on a 16-generator-68-bus test system and the performance of the adopted RBF-R NNs is compared with RBF NNs, as well as with multilayer perceptrons. The simulation results show that a significant improvement in prediction accuracy is obtained by the RBF-R NNs together with the feature selection and the elimination of boundary instances.

KEYWORDS:

• Feature selection
• Instance selection
• Neural networks
• Power system security assessment
• Radial basis function
• Transient stability

Additional information

Funding

This work was supported by The Scientific and Technical Research Council of Turkey (TUBITAK) project no. 114E157.

Notes on contributors

S. Jafarzadeh

Sevda Jafarzadeh received the BSc degree in electrical engineering from Tabriz University in Tabriz, Iran in 2015 and MSc degree in electrical engineering from Istanbul Technical University in 2017. Currently, she is a PhD student at the same department. She has worked on the security assessment of large power systems with machine learning approaches during her MSc program. Her research interest is dynamic security assessment. Email: [email protected]

V. M. I. Genc

V M Istemihan Genc received the BSc degree in electrical engineering from Istanbul Technical University, the MSc degree in electrical engineering, systems and control engineering, and systems science and mathematics from Istanbul Technical University, Bogazici University, and Washington University, respectively. He received the DSc degree in 2001 from Washington University. He is currently an Associate Professor in the Department of Electrical Engineering at Istanbul Technical University. His research interests include power system dynamics, stability and control.

Z. Cataltepe

Zehra Cataltepe received the BSc degree in computer engineering from Bilkent University in Ankara, Turkey in 1992 and MSc and PhD degrees from Caltech in computer science in 1994 and 1998, respectively. Sheworked at Bell-Labs and Siemens Corporate Research and then joined the Istanbul Technical University Computer Engineering Department. Currently, she is a Professor at the same department and also the co-founder of tazi.io which is a startup company with the purpose of developing online machine learning products. Her research interests include online machine learning and feature selection. Email: [email protected]