Abstract
Distributed renewable electricity generators facilitate decarbonising the electricity network, and the smart grid allows higher renewable penetration while improving efficiency. Smart grid scenarios often emphasise localised control, balancing small renewable generation with consumer electricity demand. This research investigates the applicability of proposed decentralised smart grid scenarios utilising a mixed strategy: quantitative analysis of PV adoption data and qualitative policy analysis focusing on policy design, apparent drivers for adoption of the deviation of observed data from the feed-in tariff impact assessment predictions. Analysis reveals that areas of similar installed PV capacity are clustered, indicating a strong dependence on local conditions for PV adoption. Analysing time series of PV adoption finds that it fits neither neo-classical predictions, nor diffusion of innovation S-curves of adoption cleanly. This suggests the influence of external factors on the decision making process. It is shown that clusters of low installed PV capacity coincide with areas of high population density and vice versa, implying that while visions of locally-balanced smart grids may be viable in certain rural and suburban areas, applicability to urban centres may be limited. Taken in combination, the data analysis, policy impact and socio-psychological drivers of adoption demonstrate the need for a multi-disciplinary approach to understanding and modelling the adoption of technology necessary to enable the future smart grid.