New way of pursuing the MPPT voltage

[ilario]

In perovskite solar cells the MPPT is not easy to get neither doing small scans and alternating scan direction, as implemented in #9
A new way is needed.
The best solution I could find in the literature is something that I suspect needs specialized hardware: http://onlinelibrary.wiley.com/doi/10.1002/pip.2894/abstract
I would rather go for a step, dwell, measure approach, in which the PCE at a new voltage is measured after a dwell time after keeping the voltage constant.
An algorithm is needed for deciding the direction and the amplitude of voltage steps based on previous PCE measurements, keeping into account that MPPT voltage can change over time.

[ilario]

In case of the "step, dwell, measure approach", a way to reduce the time length for each steady current measurement could be to fit (with a biexponential?) the current profile: http://pubs.rsc.org/en/Content/ArticleLanding/2017/TC/C7TC03482B