Dynamic Black-Box Modeling of Proton Exchange Membrane Fuel Cell Based on Core Vector Regression

This study proposes a new dynamic Core Vector Regression (CVR)-based model for Proton Exchange Membrane Fuel Cell (PEMFC) which can be used for both dynamic andsteady-state studies. So far, most of conventional mathematical PEMFC models have been presented based on conversion lawswhich are time-consuming and need large amount of memory to be applied for controller design, generation and load predictions, optimization and other real-time studies due to their complexmathematical equations. To solve the problems, some black-box identification techniques such as Artificial Neural Network(ANN) and Support Vector Machine (SVM) are also proposed for PEMFC modeling. In this paper, in order to model dynamicmultivariable behavior of PEMFC a CVR-based black-box model is proposed. In this model, decision tree (DT)-based featureselection approach is used to reduce the size of inputs of the CVR. The presented model needs little training time and small amount of memory in comparison with existing black-box models. Moreover, the proposed model has the least errors and the best squared correlation coefficient in comparison with aforementionedblack-box models. In order to illustrate the effectiveness of the proposed modeling procedure, it is applied toa 500W PEMFC stack which shows satisfactory results for both steady-state and dynamic studies.

Keywords:

component , Proton Exchange Membrane Fuel Cell , Core Vector Regression , Decision Tree , Dynamic Modeling , Black-Box Modeling