Skip navigation
The Australian National University

Solar insolation data mining using existing PV systems

There has been rapid uptake of solar photovoltaic (PV) micro-generators within Australia: total installed capacity now exceeds 1 GW. Continued rapid falls in PV module cost price and increases in grid power costs are expected to lead to continued vigorous growth of Australia’s network of microgenerators. This is a clear indicator that the challenges of integrating high penetrations of microgenerators – and their inherent intermittency – into the electrical grid can no longer be ignored.

No reliable method currently exists for estimating the contributions of PV systems to the power grid at the spatial and temporal scales required for grid operation. This leaves grid operators vulnerable to power quality issues caused by an inherently variable solar resource (primarily attributable to clouds). Methods of dealing with solar intermittency include storage and bringing fast response alternative stationary energy generation (eg, hydro or gas) on line. However, forecasts of microgenerator contribution are essential to ensure optimised implementation of such responses.

To address these challenges ANU and NICTA, in collaboration with their US and Australian partners, are exploring real-time data-mining of some of the hundreds of thousands of very widely distributed residential PV systems. The output of existing PV systems, when mapped in real time, allows precise detection of cloud location, cloud motion and cloud opacity, perfectly matched to the characteristics of PV systems. Additionally, we are developing an experimental network of low cost all-sky cameras to diversify the cloud detection methods available for synthesis.

The large amount of heterogeneous data to be captured from existing PV systems and low cost all-sky cameras, and the need to process it at low cost in real time to be useful, requires sophisticated strategies quite different from traditional attempts to predict cloud (eg from satellite pictures).

Updated:  7 June 2012 / Responsible Officer:  JavaScript must be enabled to display this email address. / Page Contact:  JavaScript must be enabled to display this email address. / Powered by: Snorkel 1.4