A new method for estimating insolation based on PV-module currents in a cluster of stand-alone solar systems

Abstract

In order to evaluate the performance of solar home systems (SHSs), data on local insolation is a prerequisite. We present a new method to estimate insolation if direct measurements are unavailable. This method comprises estimation of daily irradiation by correlating photovoltaic (PV) module currents from a number of SHSs, located a few kilometres apart. The method was tested with a 3-year time series for nine SHS in a remote area in Indonesia. Verification with reference cell measurements over a 2-month period showed that our method could determine average daily irradiation with a mean bias error of 1·3%. Daily irradiation figures showed a standard error of 5%. The systematic error in this method is estimated to be around 10%. Especially if calibration with measurements during a short period is possible, the proposed method provides more accurate monthly insolation figures compared with the readily available satellite data from the NASA SSE database. An advantage of the proposed method over satellite data is that irradiation figures can be calculated on a daily basis, while the SSE database only provides monthly averages. It is concluded that the new method is a valuable tool to obtain information on insolation when long-term measurements are absent.