Abstract:

In the education of mechanical engineering undergraduates, control systems labs are usually an indispensable part. Traditionally, students are allotted a timeslot of only a couple of hours to carry out a given experiment. In these exclusively hands-on labs, participants are required to be on-campus. We have extended the hands-on labs by providing students in advance access to virtual twins which are high-fidelity digital copies of their real counterpart. By using these twins, all lab experiments can be carried out from any location, i.e. from e.g. home or from the university library, days before the assigned timeslot for the on-campus lab. This gives students the benefit of self-determined learning speed, independent of location. Since virtual twins behave almost exactly as their real counterpart, virtualization supports students to prepare better for the corresponding on-campus lab.

As an example we show the level control of a two-tank system that is fed by a pump and a single inflow. This experiment is regularly conducted by fifth semester students of mechanical engineering at our university. The control models are designed in Simulink which is extended by an add-on for real-time control. Simulink models that have been developed by lab participants off-campus may be saved to a memory stick and later plugged into the real hardware, where they are runnable with only small configuration changes. A 3D rendering engine gives students a realistic look-and-feel of the tank system under operation, where even the time-varying fill level may be read off the scale.

The virtual twins run in a centralized university infrastructure where a server is capable to export the virtualization to as many as 30 concurrent users. These users are running only a virtualization client, which is simple to install and is available for almost any operating system. Centralization of the virtualization infrastructure has the benefit that students do not need to install a bundle of software on their end devices which otherwise, according to experience, would lead to installation issues and may consequently cause student dissatisfaction.

Keywords:

Control systems labs, virtual twins, self-determined learning speed, location-independent learning.