(88e) Developing Simulation-Based, Student-Centred Educational Software in Process-Control Education and Training

Laboratory experiments demonstrating important underlying principles of process control are often challenging in their capacity for scalability towards larger-numbers sessions for students, and are often very time- and resource-consuming. Indeed, the recent rush to embrace online teaching and learning in view of the COVID-19 pandemic has accelerated very substantially since Spring 2020 an already-emerging theme of software-based exploration of key concepts in process control for students by means of virtual experiments.

The present contribution has as its central hypothesis and goal the ascertaining of how the complete implementation of virtual-laboratory environments (VLE) can effect positive learning outcomes of a student via constructivist pedagogical approaches. In so doing, a modularised MATLAB-and-Simulink VLE platform has been developed, accompanied by appropriate web-tools, which is focussed specifically on this constructivist approach to student-centred learning. To assess the effectiveness of this VLE, volunteers were split into two groups - the reference and VLE group, both of which took part in the same temperature-control exercise (via PID control). The reference group were exposed to the more traditional behaviouristic practical mode of learning, where they would pay close attention to the actual real-world experimental system and results at UCD Chemical Engineering laboratories; here, the material is gone through in a more prescribed manner, where the student has a certain level of engagement in their own learning, as per classical theory. In contrast, the VLE group encountered the equivalent specifically-designed virtual environment, where they worked in groups and individually to engage with a virtual version of the PID temperature-control experiment, based on the real-world system, where they can vary input parameters and controller settings and explore themselves the system response in time and Laplace domains, with trial-and-error approaches lending themselves to determining optimal settings.

Using equal time allocations to both approaches, it was set up - by design - that the only possible factor between the outcomes for both groups was the pedagogical approaches taken to training. Upon completion, quick MCQ examinations were undertaken to assess the groups’ learning outcomes, and it was found that the VLE cohort displayed a generally superior level of understanding; the level of active self-engagement with the topic was found to correlate reasonably well with the degree of understanding. This suggests that VLEs, designed carefully and done well with appropriate software examples, can lead to positive and effective training experiences.