Application of Artificial Neural Networks - Digital Signal Processing in Power System Protection and Control - page 245

Chapter 12
Application of Artificial Neural Networks
The signal processing methods and algorithms described in preceding chapters
were expressed in form of explicit equations, transfer functions and/or logic rules,
either in crisp or in fuzzy versions. There are, however, specific tasks and power
system operation conditions when, especially for the problems that are complex
and difficult to express in terms of traditional means, other solutions should be
applied. In such situations, both for signal processing and decision-making Arti-
ficial Neural Networks may constitute a good solution.
The Artificial Neural Network ANN is a set of processing units called neurons.
The ANN can be used to approximate the relationship between input and output
signals of the system (process), which is illustrated schematically in Fig. 12.1 . The
required approximation abilities of the ANN are acquainted through so called
training, with examples of the system (process) operation. After training the ANN
is ready to serve as a predictor of the system performance and thus deliver
information e.g. for control purposes.
The neural networks are believed to possess a generalization feature, which
similarly to human reasoning makes them good tools for identification of pat-
terns, even if not all representative features of the patterns are well defined or
when some data is missing. Due to built-in neuron activation functions the
ANNs are well suited to represent non-linear problems and bring answers to
difficult protection and control questions, some of them being outlined in this
chapter.
Below the information on the single neuron features as well as commonly used
neural network structures is presented. The training procedures applied for par-
ticular ANN structures are described, with special emphasis put to the problems
with ANN design. The theoretical part is followed by selected examples of ANN
applications for protection and control tasks in power systems.