Examination arrangement

Course content

The course will introduce the students to Structural Health Monitoring (SHM) supported by Digital Twins (DTs). The basic idea with DTs is to minimize the use of physical sensors on a real structure (asset). The concept is to create a simulation model (analytical, FE or big data model) that replicates the physical asset. The DT is then instrumented with virtual sensors (robust and free of charge) that provide additional information used in predictive maintenance and decision support. The DT can either be implemented as a (fast) local edge, a (slow) IoT cloud solution, or as a combination of both for maximum flexibility. After finishing the course, the students shall be able to develop and implement a Digital Twin (DT) for SHM of a Hardrocx bicycle frame during hard riding (student competition).

Course topics are:

1. INTRODUCTION to Structural Health Monitoring by Digital Twins
   • Why DT Modelling and simulation?
   • A review of various DT models (static versus dynamic)
   • The FMU/FMI concept (how to integrate solvers and DT models)
   • How to implement a SHM solution with DTs (Python and low-code tools.)

1. ANALYTICS:
   • Why and how to turn data > information > knowledge>actions (decision support)
   • Machine learning versus analytics

Learning outcome

Knowledge: The student will learn the main principles of Digital Twin modeling and simulation for Structural Health Modeling. This includes knowledge about sensors, data filtering, inverse methods for load identification and data processing for decision support. The student will get basic knowledge about edge and IoT solutions as well as DT modeling and simulation tools.

The student shall learn about current pitfalls due to sensor noise and drifting as well as limitations in DT modeling using inappropriate boundary conditions, incorrect loads, and simulation settings. The student shall be able to identify the best candidates for SHM.

Skills: The student shall be able to setup and run DTs and simple IoT solutions for SHM. Important skills are python programming, instrumentation, load identification, real time process modeling, simulation and visualization. The students will also be trained in low-code app (Mendix) and IoT dashboard development (MindSphere).

General Competence: The student shall be able to master multidiscipline modeling, instrumentation, data filtering, and simulation of Digital Twins. The student will have a general competence on various IoT software solutions for structural health monitoring and visualization.

Learning methods and activities

Lectures, videos, exercises and examples from real applications will be used. There will be individual mandatory multiple-choice assignments and one project. The project will be related to structural health monitoring of a mountain bike from Hardrocx or the Palfinger offshore crane on Gunnerus.

Assignments, project work and field exercises (Hardrocx bike competition).

Tj @ YouTube is my digital twin (teaching assistant)

Compulsory assignments

• Mandatory assignments

Further on evaluation

Evaluation is done according to NTNU guidelines, with feedback from the class and meetings with the reference group. A course survey will be prepared for all students on Blackboard. A final evaluation report is done at the end of the course.

Specific conditions

Recommended previous knowledge

The students should have some basic knowledge of mechatronics, programming, finite element analysis and physics.

Required previous knowledge

-

Course materials

Suggested teaching material and project:

• Hand-out slides
• Papers on Structural Health Monitoring
• Training videos on YouTube
• Python software templates and Mendix tutorials (app and process design)
• Hardrocx bikes used in Structural health Monitoring (project)

Credit reductions