MODELING AND OPTIMIZATION OF A PHOTOVOLTAIC-ELECTROLYSIS SYSTEM FOR HYDROGEN GENERATION

Abstract

Hydrogen production from water electrolysis is widely accepted to be the most sustainable source of hydrogen production, especially when integrated with renewable energy sources such as solar or wind energy. Nevertheless, considerable improvements still are needed for renewable hydrogen to be fully competitive with the current fossil fuel energy resources. In this work, we develop and apply computational tools to model hydrogen production from solar powered water electrolysis systems. A kinetic and mechanistic investigation of the oxygen evolution (OER) and hydrogen evolution (HER) reactions on the active nickel-iron layered double hydroxide (NiFe LDH) electrode is presented. Both linear sweep voltammetry and electrochemical impedance spectroscopy measurements were combined with theoretical models describing the electrode kinetics to evaluate the OER and HER rate constants. The rate determining step and reaction mechanism of the OER and HER were identified as a result of this analysis. A computational algorithm to model an integrated PV-electrolysis-battery system is presented with the goals of identifying the system’s optimal design that maximizes the hydrogen production rate, minimizes the levelized cost of energy and total system’s cost, while targeting a net-zero grid energy operation. The coupled system is connected to the electric grid to ensure uninterrupted operation of the electrolyzer. The model is simulated over one year to include both diurnal and seasonal weather variations. Over 2 million different design configurations were evaluated, 13 of which were chosen as the Pareto Front for this optimization problem. Finally, we develop and apply computational tools to identify optimal design configurations when integrating a large number of PV cells under shady or faulty conditions. Monte Carlo simulation is used to introduce the stochastic effect and uncertainty generated by shading, and the model is simulated under different shading intensities. With the inclusion of a cost attribute, a multi-objective optimization problem is developed. Information entropy weight and the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) methods were combined to identify the optimal number of bypass diodes for each shading case. This computational platform can be extended towards developing simulation-based design tools for the integration of nano-scale energy devices.