Bayesian Psychometrics Modeling based on the Correlation Between Inductor Saturation and Current Limitation

To avoid the disadvantages of strong subjectivity and one-sided evaluation in traditional assessment methods, we propose a cognitive diagnosis assessment method based on Bayesian psychometrics model in the field of power electronics experimental education. Firstly, take the influence of inductance saturation on the current limit as an example, the cognitive diagnosis model is built based on the evidence-centered design framework. Secondly, the conditional probability of the Bayesian network intermediate nodes is calculated based on the graded response model. Finally, with the example of the magnetic core material knowledge, cognitive diagnosis is achieved using WinBUGS based on the experimental data generated from examinees, verifying the correctness and effectiveness of this method.