EIEIOTECHNOVATION
← Training programmes

EIEIO / AUTOMATION & EMERGING TECHNOLOGY

Digital Twin Systems

Course outline

2 days
Discuss this programme →

Duration

Digital Twin Systems

2 days

Overview

This training addresses a crucial need in Malaysian manufacturing industries to develop digital twin systems for automated manufacturing systems to expand process design, monitoring and optimization. Participants will first learn different frameworks that guide the specification of digital twins and gain exposure to the different components and their roles in a digital twin system. Next, participants will review use-cases of digital twins. After the broad-based learning above, participants learning will be focused in three thematic areas namely (1) digital twin software architecture & solutions (2) digital twin models and (3) a conceptual and theoretical overview of physical to virtual (P2V) and virtual to physical (V2P) applications of digital twin systems such as state estimation and model predictive control. Theoretical and conceptual learning is reinforced with hands-on experiential learning using Siemens Tecnomatix Plant Simulation software for process modelling and interaction with selected software components of the OpenTwin framework.

Training Objectives

To help employees skilled in engineering and information technology in Malaysian manufacturing

companies bridge the knowledge and competency gap, the objectives of this training are:

  • To develop systems level thinking in the specification and design of digital twin systems architecture.
  • To develop critical analysis and decision-making ability in the specification, design, selection or adaption of digital twin models.
  • To develop awareness of algorithms in advanced applications of digital twin models.

Target Audience

In house program for engineers / computer scientists/ technologists

Pre-Requisite

Bachelors of engineering degree in Electronics, Computer Engineering or Computer Science or equivalent. Working practical knowledge with containers (docker) and linux system administration.

Learning Outcomes

By the end of this training, trainees will be able to

  • Select, specify and evaluate software components to assemble a digital twin system.
  • Select, specify or adapt digital twin models for manufacturing process and supply chain
  • Explain and apply algorithm to analyze state estimation and model predictive control

Assessment Methods

Oral knowledge assessment, group presentation / demonstration

Agenda

Day 1

Day

Time

Content

Practical

8.15am-

9.45am

Introduction to Digital Twin Systems

  • History & evolution
  • Digital Twin Systems 3D & 5D Framework
  • Cyberphysical systems & Digital Twin Networks
  • Use cases by industry, function & life cycle

Case study 30 min

9.45am-

10.15am

Digital Twin Software Architecture

Survey of commercial software

  • nVidia Omniverse
  • Azure Digital Twin
  • Siemens Tecnomatix

Video & Discussion

30 min

10.15am-

1030am

Break

1030am-

1130am

Digital Twin Software Architecture

ISO23241 Digital Twin for Manufacturing Standard ISO30141 IoT Standard cursory view

ISO23241 OME and Attributes

Group activity

30 mins

1130am-

1230pm

Digital Twin Models (Process Model) Introduction to Queueing Theory Elements of Queue System

Kendall Notation

Discrete Event Simulation & Review of Probability & Random Processes

20 min

1230-

1.30pm

Lunch Break

1.30-

3.00pm

Digital Twin Models (Process Model) Useful Results from Queueing Theory

  • Little’s Law
  • M/M/1 queue
  • M/M/m queue
  • M/M/1/k queue

20 mins

Modelling Manufacturing Stations

Integer Programming Use cases

  • Capital budgeting
  • Warehouse location

-Scheduling

3.00-3.30

Digital Twin Software Architecture Introduction to Siemens Tecnomatix Plant Simulation Software

Initiate Group Project

Handson 30 mins

3.30-

345pm

Break

345pm- 545pm

Digital Twin Software Architecture Introducing the Open Twins Software Framework Eclipse Ditto (Digital Twin State) experiential learning

Apache Kafka (message stream aggregator) experiential learning

Eclipse Hono (IoT aggregator) experiential learning

Grafana (Visualization) experiential learning

100 mins

Day 2

Day

Time

Content

Practical

8.15am-

9.30am

Digital Twin Models (Physics Model) Fundamental physics equations & Finite Element solvers

Introduction to Supervised Learning of Neural Networks

Introduction to Physics Informed Neural Networks Hands-on exploration with PINN

15min

9.30am-

10.45am

P2V Algorithm

State transition & Measurement model Noisy state estimation using drone example Key idea to solve noisy state estimation

Quiz to reinforce comprehension

15min

10.45am-

11.00am

Break

1100am-

1200pm

P2V Algorithm Example

Bayesian Inference & Particle Filtering

Hands on exploration with battery state estimation using particle filtering

30 min

1200am-

1230pm

P2V Algorithm

Bayesian Networks for Multistate Estimation Fault Diagnostics and Failure Prognosis

1230-

1.30pm

Lunch & Prayer Break

1.30-

3.30pm

Group project time to develop process model of manufacturing line with Siemens Plant Simulator Instructor Q&A session

3.30-4.30

V2P Algorithm Example

Model predictive control with drone example Hands with battery state optimization

20min

4.30-

445pm

Break

445pm- 545pm

Project Presentation & Sharing Discussion & Wrap-Up

60 min

Programme scope, delivery arrangements and suitability are confirmed during enquiry.