Naslov
Functional data analysis and its application in production processes

Autori
Lela, Branimir ; Duplančić, Igor ; Šupe, Miroslav

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Conference Proceedings ''Mechanical Technologies and Structural Materials'' / Dražen Živković - Split : Hrvatsko društvo za strojarske tehnologije (HDST), 2011, 93-98

Skup
2nd Interantional Conference MTSM 2011

Mjesto i datum
Split, Hrvatska, 29.09.2011. - 30.09.2011

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Functional data analysis; Mathematical modelling; Algebraic and differential equation; Production processes

Sažetak
Modern manufacturing facilities are with no exception equipped with measurement devices that are able to produce high-resolution discrete data for relevant production parameters. Data measured that way can be described by smooth functions having the derivatives of desired order. Each such function is considered as a data in functional data analysis (FDA) method. FDA is an extension of traditional methods for data analysis to data that can be identified as functions. The objective of this work is to describe the possibility of solving inverse problems in production processes by FDA. In this work algebraic and differential equations are introduced through mathematical models describing the behaviour of some production parameters in the hot rolling process of aluminium strip. The results of the computer simulation show that FDA methodology can be successfully applied in modelling of production systems and in solving problems of parameter estimation in dynamic mathematical models.

Izvorni jezik
Engleski

Znanstvena područja
Strojarstvo

POVEZANOST RADA