Data mining of remote monitored stand-alone solar PV systems for State of Health estimation

Abstract

In this paper, we use data mining techniques and formulate suitable assessment metrics to derive estimates of the State of Health (SOH) of stand-alone solar home systems. Data is provided from a company with significant numbers of such systems in Africa. The systems in question contain a PV panel, lead-acid battery and a series of DC loads. Data mining allows us to not only estimate the SOH of the battery, but also infer the health of other system components.