Home // INTELLI 2012, The First International Conference on Intelligent Systems and Applications // View article

Functional Roots and Manufacturing Tasks

Authors:
Ingo Schwab
Norbert Link

Keywords: Symbolic Regression; Manufacturing; Functional Roots; Machine Learning

Abstract:
Many production processes consist of repetitive, almost identical sub-processes. Process models are needed for state estimation and control purposes. Models are frequently formed from an analysis of input-output data relations of the overall process. For a repetitive process, the model of the repeated process is a functional root of the relation. Functional roots are introduced and symbolic approaches are presented. We propose to find functional roots via Symbolic Regression to model repetitive processes. As a first proof of principle we show the suitability of this approach with two basic and well-known problems in the scientific field of physics and nonlinear dynamics. The exact solutions of these problems are available from textbooks and can be used to assess the results of our approach. The first step in our project work therefore is to develop suitable concepts and technologies. The next steps will include analyzing real world data in cooperation with our project partners.

Pages: 35 to 40

Copyright: Copyright (c) IARIA, 2012

Publication date: April 29, 2012

Published in: conference

ISSN: 2308-4065

ISBN: 978-1-61208-224-0

Location: Chamonix, France

Dates: from April 29, 2012 to May 4, 2012