DIGITAL LIBRARY
USING A NEURO-FUZZY MODEL TO CALCULATE THE OPTIMAL RATIO BETWEEN DISCIPLINES FROM VARIOUS FIELDS FOR MECHATRONICS SPECIALIZATION
1 Steinbeis Hochschule Berlin (GERMANY)
2 Lucian Blaga University of Sibiu (ROMANIA)
About this paper:
Appears in: INTED2009 Proceedings
Publication year: 2009
Pages: 3419-3426
ISBN: 978-84-612-7578-6
ISSN: 2340-1079
Conference name: 3rd International Technology, Education and Development Conference
Dates: 9-11 March, 2009
Location: Valencia, Spain
Abstract:
Mechatronics is a highly multidisciplinary specialization and its curricula should include disciplines from various fields, as well as disciplines from the mechatronic domain itself.

A survey of the literature reveals the fact that there are no mathematical relations to describe the optimal ratio between the disciplines within these domains.

Mechanical science, electric and electronic sciences, computer science together with mechatronic sciences are the main domains taken into consideration within this research.

A neuro-fuzzy approach (using both artificial neural networks and fuzzy logic) was proposed by the authors in order to build a mathematical model for calculating the mathematical dependencies between the percentages of disciplines from each of above mentioned fields. The model will allow the user to calculate the necessary changes within the remaining domains when a percentage of disciplines within a domain has to be changed, in order to preserve the equilibrium of the curricula.

The neuro-fuzzy approach involves in the first stage a comprehensive process of data gathering, in order to collect the necessary data for building the model. After collecting the data, which will be done by analyzing curricula from some representative universities within the world, the model will be generated by a semi-automatic process, using the Matlab &
Simulink software package. In this stage, the authors have only to input the data and to select the number and type of the membership functions to be used.

Further steps have to be performed in order to fine tune the generated model. The variables used within the automatic generation of the neuro-fuzzy model have to be replaced by linguistic variables and the intervals of variations have to be manually altered.

The process of fine tuning the model has a great importance and will be one of the major contributions of this work. Without this process, the neuro-fuzzy model could not describe the dependencies between the variables in a correct manner and also, the main advantage of fuzzy logic, making use of the human reasoning and linguistic variables could not be used.
Keywords:
mechatronics, curricula, multidisciplinary, neuro-fuzzy.