English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Feed forward artificial neural networks (multilayer perceptron) allow solving a wide range of regression (approximation) and classification (data splitting into subsets) problems. The corresponding algorithms are applied in electrical and power engineering. The peculiarity of such an artificial neural network is that the training sample can be submitted to the input in an arbitrary sequence. Therefore, the signals themselves during artificial neural network training should be formed taking into account their time form. The paper proposes the use of artificial neural network in a sliding time window. Using simple examples of transients and relay protection, the paper analyses the possibilities of the signals’ time form approximation and the problems of recognition using the artificial neural network of the signal parameters near zero and threshold values. It is shown that errors in the operation of the artificial neural network can be compensated. The choice of the duration of the sliding window must necessarily take into account the need for additional processing of data from the output of the neural network in the form of their statistical analysis, the obtained dependencies approximation or smoothing.

FEED FORWARD NEURAL NET SIGNAL PROCESSING: APPROXIMATION AND DECISION MAKING