Controle de um processo de pelotização: realimentação por imagem

Abstract

Pellets are bolls produced with iron ore concentrates of differentchemical and mineralogical, with specific properties and great application on steel manufactures. The palletizing plants around world do not have an automatic process control, being the visual operator experience responsible for the pellet quality. This thesis describes the practical application of a Nebulous control system to an industrial iron ore palletizing plant, with feedback by a digital image process. It is composed with the process and the equipment description and the information about the area where the control is applied. The objective was to develop a model to detect quality pellet size standard on the process, and with the identification of the standard, execute the pellet drum rotation control. The model developed aimed to improve the pellet formation yield, making better the production with quality. This thesis is composed of the process, pellet equipment and industrial plant description. The concepts and studies made on the work for definition of the strategy of the digital process image and the validation of the results are shown. On the mathematical modeling of the pellets formation process is utilizedthe method of identification through the step response to determine theresponse of time of the drum. The fuzzy control system design considered the tests results to define the interactions between the variables that causes changes on the pellets formation. The results obtained with the application of the control system is shownand discussed.