NEURAL-LEXICAL CLASSIFICATION OF POWER QUALITY INSTRUMENTATION MULTIPLICITY

Maurizio Caciotta, Michela Cristiani

Abstract

The problem of perceived electrical power quality by users of the Little Medium Enterprises (LME) is settled and matched with the difficulty to individuate the correct instrumentation that is able make their satisfaction. The exorbitant quantity of instruments that are on the market, that, the builders assure, measure power quality, cannot be completely managed by operators. The paper
present a system based on automatic lexical procedure that read the manuals and extract the information that can be properly managed by a Kohonen non supervised neural network to obtain an optimal inner classification in view to cross data with an inquiry on effective perceived quality by users.

Keywords

Power Quality, Kohonen Neural Network, Lexical Analysis

Download

PWC-2003-TC4-028.pdf

DOI

-

IMEKO TC

TC4 - Measurement of Electrical Quantities

Back to the proceedings