Robust techniques for PV modeling, simulation, visualization and design could catalyze future advances in photovoltaics and a new study has proposed a code-based modeling (CBM) approach to guide such steps.
Current approaches tend to focus on the electrical energy dimension of PV installations with software packages considering design and parametric and financial analysis. However, the modeling of aspects such as solar cell physics, electrochemistry, photochemistry, material characteristics and thermodynamics could also improve PV conversion efficiency, a new study has argued.
Researchers from the University of Manchester, in the U.K., and the Alex Ekwueme Federal University Ndufu-Alike Ikwo, in Nigeria, have coded a CBM approach using programming language MATLAB and trained the model with synthesized data from peer-reviewed literature. The model was validated on Solarex MSX-60 and Shell S140 commercial solar modules to examine how accurately it predicted the parameters stated by the manufacturers.
Results showed the model could repeatedly and reliably predict short circuit current, maximum power point and open-circuit voltage with 0%, less-than-2% and less-than-10% deviation, respectively. The developers of the system said such high-precision modeling and simulation of photovoltaics could help reduce development costs and design turnaround time and facilitate better techno-economic decision-making.
The researchers then demonstrated instances where their CBM approach could be applied to the science and engineering of photovoltaics, such as in investigating the thermodynamics of PV, solar cell material characterization, PV system design and power monitoring.
The CBM approach appears to be robust, according to its developers, because it allows user-defined functions such as inputting customized equations, thus presenting new opportunities for scientists and engineers to advance model-based investigations of PV beyond the current state-of-the-art.
The goal of the paper published in Elsevier is to facilitate the adoption of the CBM approach by other researchers and developers using object-oriented languages such as C++, MATLAB, FORTRAN or Python.