Cognitive Diagnostic Assessment of Power Supply Magnetic Components Design Based on Bayesian Network

The power electronics engineering education is aimed at helping students in becoming qualified electronics engineers. However, the existing evaluation method cannot reflect the students’ performance in knowledge-structure and skillsets objectively and accurately. To address these issues and improve the effectiveness of the current evaluation method in the field, we propose a Bayesian network model-based cognitive diagnostic assessment method and demonstrate it to evaluate students’ knowledge and skills condition with the switched-mode power supply (SMPS) magnetic components design task. The paper starts with a brief introduction to the SMPS inductor design. It continues with the Bayesian network model-based inductor proficiency model, inductor evidence model, and the task model for power magnetics volume and weight in the aerospace SMPS. Then we identify the parameters in the graded response model, relations among variables, and calculate the conditional probability between variables. Finally, we use Markov Chain Monte Carlo estimation method to get the posterior probability distribution of proficiency variables by OpenBUGS. The results show that this cognitive diagnostic assessment system can effectively reflect the students’ study performances, scientifically advise their future study plans, and effectively achieve the education goals.