Radial Basis Function Neural Networks for Transient Stability Assessment Part I: Theory of Radial Basis Function Neural Networks

Abstract

Noteworthy strides are currently being taken in the realm of online transient stability assessment (TSA) with the world's first online implementation in a major electric utility in the past year. Such a tool allows the determination of critical contingencies in a short enough time period for use in online conditions. Recent developments in neural network technology have opened up even newer possibilities for real-time assessment and control. Several neural network architectures are introduced in this paper for the online TSA. However, specific types of the radial basis-function network (RBFN) viz., the probabilistic and the general regression neural networks appear to give the most accurate results. In Part I of this paper a detailed theoretical background of the RBFNs is provided.

Department(s)

Electrical and Computer Engineering

Keywords and Phrases

Dynamic security; General regression neural network; Probabilistic neural network

International Standard Serial Number (ISSN)

1078-3466

Document Type

Article - Journal

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2024 Elsevier, All rights reserved.

Publication Date

01 Dec 2001

This document is currently not available here.

Share

 
COinS