Upravljanje protočnom sušarom pomoću neuronske mreže

undergraduate thesis

• Title (english): Flow dryer control based on neural network
• Author: Ivan Radić
• Mentor: Tomislav Stipančić (mentor)
• Granter: University of Zagreb; Faculty of Mechanical Engineering and Naval Architecture; Zagreb
• Defense date and country: 2021-02-25, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Mechanical Engineering

Abstract

Tema ovo završnog rada je unaprijeđene rada protočne sušare furnira. Trenutno ručno podešavanje parametara unutar sušare ne zadovoljava zahtjeve željene kvalitete proizvoda što uzrokuje sve veće probleme zbog nezadovoljstva klijenata. Prevelike oscilacije od zadanih parametara za temperature i vlažnost zraka untar sušare događale su se zbog neadekvatne regulacije izvršnih članova. Ovim radom opisati će se projekt automatizacije rada sušare sa dodatkom predviđanja izlazne vlažnosti furnira pomoću neuronske mreže. U početnim poglavljima biti će opisane osnove procesa sušanja i neuronskih mreža. Praktičnim dijelom biti će prikazani ladder dijagrami u TIA Portalu koji preko Siemens PLC-a upravljaju ventilima za dovod pregrijane pare u izmjenjivače topline te za regulaciju klapne za odzračivanje, kako ne bi došlo do zasićenja zraka s vodenom parom. U zadnjem dijelu rada biti će objašnjena izrada neuronskih mreža pomoću MATLAB-ovog alata „Neural Network Toolbox“ te mogućnost predviđanja izlazne vlažnosti furnira temeljem uvjeta unutar sušare.

Abstract (english)

The subject of this thesis is to improve the work of a veneer flow dryer. Currently, manually
adjusting the dryer parameters does not meet the requirements of the desired quality of the
product. It causes problems because of customer satisfaction. Excessive oscillations from the
default parameters for ambient air temperature and humidity were caused by inadequate
regulation of actuators. This thesis describes the dryer automation project with the addition of
the forecast of the output moisture of the veneer using the neural net. The basics of the suture
and neural networks will be described in the initial chapters. The practical part will be shown
by the ladder diagrams in TIA Portal which, through Siemens PLC, will control valves for the
supply of overheated steam to heat exchangers and for the regulation of aeration rods, so as to
avoid saturation of air with water vapour. In the last part of the work, it will be explained by
the development of neural networks, using the matrix of "Neural Network Toolbox" tools and
the ability to predict the output moisture of the veneer based on the conditions inside the
dryer.

• Keywords: Sušenje; sušare; neuronske mreže; automatizacija; digitalizacija; Neural Network Toolbox; furnir
• Language: croatian
• URN:NBN: urn:nbn:hr:235:236940
• Study programme: Title: Mechanical Engineering; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka strojarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka strojarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2021-02-19 15:47:59