In recent years, a new industrial revolution - also known as Industry 4.0 - is emerg- ing and refers to the cyber revolution and specifically about moving towards Smart Manufacturing. Smart Manufacturing is constituted by a network of decentralised machines that collaborate and coordinate their tasks to increment the productivity, quality and reduce the overall cost of production. With the increase of new tech- nologies, multiple structures were implemented to simulate the systems. System simulations are mostly problem-tailored and not capable to represent the system in order to perform different analysis in the same simulation environment. In the case of manufacturing systems, the most frequently used analysis are - from the industrial engineering point of view - the Discrete EVent Systems (DEVS) models, based on probability distributions. The structure of the system is however way more complicated and includes also other specifications, given that - from the control en- gineering point of view - manufacturing systems are essentially industrial automated systems that came with well-known control strategies. This thesis aims to develop a library (SFCLib) to model Discrete-Event Systems (DEVSs) that can be connected to their Dynamic Models (DMs), with the aim of adding the physic of the systems, including also logic and modulating control. A consistent representation has been identified in an existent structure, the Sequential Function Charts (SFC). Connect- ing together the Discrete-Event Systems representation with the related Dynamic Models, The SFC library implementation developed in this thesis is intended to satisfy three main aspects of such systems, which are the scalability in the levels of detail, the integration with the control strategy and the dependency on Dynamic Models. When modelling large systems, their complexity may be elevated and the decomposition of such systems into subsystems may be essential. Every single sub- system is then modelled individually and characterised by its internal functioning. The approach that best matches these features is found in Objected-Oriented Mod- elling (OOM). In particular, in this thesis, the Object-Oriented languages chosen is Modelica, a high-level declarative language for modelling and simulation of physical systems that describes their mathematical behaviour.
Negli ultimi anni, una nuova rivoluzione industriale - conosciuta anche come In- dustria 4.0 - sta emergendo e si riferisce al concetto di cyber revolution e più pre- cisamente alla continua evoluzione verso lo Smart Manufacturing. Con il termine Smart Manufacturing si intende un sistema caratterizzato da una rete costituita da macchine decentralizzate che collaborano e coordinano i loro compiti al fine di incrementare la produttività, la qualità e di ridurre i costi di produzione. Con lo sviluppo di nuove tecnologie, diverse strutture sono state realizzate per la simu- lazione di sistemi. Tali strutture sono generalmente fatte su misura per il sistema e non permettono di rappresentare il sistema in modo da poter svolgere differenti anal- isi nello stesso ambiente di simulazione. Nel caso di sistemi manifatturieri, le analisi più frequentemente usate - dal punto di vista dell’ingegneria industriale - sono i mod- elli di Sistemi ad Eventi Discreti (DEVS), basati su distribuzioni di probabilità. La struttura del sistema è tuttavia ben più complicata ed include anche altre specifiche, dato che - dal punto di vista dell’ingegneria del controllo - i sistemi manifatturieri sono sostanzialmente sistemi industriali automatizzati che hanno strategie di con- trollo ben definite. Questa Tesi punta allo sviluppo di una libreria (SFCLib) per un’unica rappresentazione di Sistemi ad Eventi Discreti e controllo logico in modo tale che entrambi possano essere connessi al controllo modulante che usa la model- lizzazione dinamica del sistema per rappresentarne la fisica. Una rappresentazione coerente è il Sequential Function Chart (SFC), basato sullo standard di implemen- tazione dei PLC, IEC 61131. La libreria SFC sviluppata ha lo scopo di soddisfare tre fattori principali, che sono la scalabilità nel livello di dettaglio, l’integrazione con le strategie di controllo e la dipendenza dai Modelli Dinamici. Quando si modellano ampi sistemi, la loro complessità può essere elevata e la loro scomposizione in sotto- sistemi potrebbe risultare essenziale. Ogni singolo sottosistema è quindi modellato individualmente ed è caratterizzato dal suo funzionamento interno. La strategia che più si adatta a questi aspetti è quella della Modellizzazione Orientata agli Oggetti (OOM). In particolare, in questa Tesi, il linguaggio Orientato agli Oggetti scelto è Modelica, un linguaggio dichiarativo ad alto livello per la modellizzazione e simu- lazione di sistemi fisici e la descrizione matematica del loro comportamento.
A Modelica library for the efficient representation of IEC logic controls
TOMASI, GIACOMO
2019/2020
Abstract
In recent years, a new industrial revolution - also known as Industry 4.0 - is emerg- ing and refers to the cyber revolution and specifically about moving towards Smart Manufacturing. Smart Manufacturing is constituted by a network of decentralised machines that collaborate and coordinate their tasks to increment the productivity, quality and reduce the overall cost of production. With the increase of new tech- nologies, multiple structures were implemented to simulate the systems. System simulations are mostly problem-tailored and not capable to represent the system in order to perform different analysis in the same simulation environment. In the case of manufacturing systems, the most frequently used analysis are - from the industrial engineering point of view - the Discrete EVent Systems (DEVS) models, based on probability distributions. The structure of the system is however way more complicated and includes also other specifications, given that - from the control en- gineering point of view - manufacturing systems are essentially industrial automated systems that came with well-known control strategies. This thesis aims to develop a library (SFCLib) to model Discrete-Event Systems (DEVSs) that can be connected to their Dynamic Models (DMs), with the aim of adding the physic of the systems, including also logic and modulating control. A consistent representation has been identified in an existent structure, the Sequential Function Charts (SFC). Connect- ing together the Discrete-Event Systems representation with the related Dynamic Models, The SFC library implementation developed in this thesis is intended to satisfy three main aspects of such systems, which are the scalability in the levels of detail, the integration with the control strategy and the dependency on Dynamic Models. When modelling large systems, their complexity may be elevated and the decomposition of such systems into subsystems may be essential. Every single sub- system is then modelled individually and characterised by its internal functioning. The approach that best matches these features is found in Objected-Oriented Mod- elling (OOM). In particular, in this thesis, the Object-Oriented languages chosen is Modelica, a high-level declarative language for modelling and simulation of physical systems that describes their mathematical behaviour.| File | Dimensione | Formato | |
|---|---|---|---|
|
thesis_main.pdf
accessibile in internet per tutti
Descrizione: Thesis
Dimensione
3.43 MB
Formato
Adobe PDF
|
3.43 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/165406