There is a rising interest in hybrid PV-wind power plants, i.e. where the PV and wind assets are co-located and share a common point of interconnection (POI) with the grid. These systems can be oversized in terms of total PV and wind capacity relative to the POI capacity, without experiencing any significant energy losses. This is due to the anti-correlated PV and wind generation which reduces the need for curtailment, i.e., when the power output of the plant is purposely reduced when the total generation exceeds the POI capacity. Additionally, retrofitting existing wind power plants into PV-wind hybrids is an effective way of increasing the renewable energy penetration in areas with plenty of existing wind power plants, and where getting access to grid for new PV capacity can be a challenge. Such retrofitting will likely lower the PV investment costs, reduce the environmental impact, and accelerate the permitting processes, in comparison to standalone PV projects.

This research proposes a methodology to find which existing wind power plants in a given region (here: the Nordics) are most suitable for being retrofit with PV capacity.

Initially, we identify why some wind power plants are more techno-economically suited for retrofitting than others. For this analysis, we use five years of actual wind power generation data on hourly resolution from 128 wind power plants in Norway and Sweden, in addition to modeled PV generation data for the same locations and time period. Using a techno-economic model, 97 of these plants are deemed suitable for such retrofitting. Using multiple linear regression (MLR) analysis, it's concluded that having high mean PV and low mean wind capacity factors, in addition to strong anti-correlation between the hourly PV and wind generation, results in more profitable retrofitting projects.

Then, based on the MLR analysis, the retrofitting projects' net present value (NPV) can be derived as a linear function of a site's mean PV and wind capacity factor and the anti-correlation factor between the hourly PV and wind generation. This information can, through a few steps, aid in estimating which other existing wind power plants (where we don't have access to hourly generation data) in the Nordics should be retrofitted. Firstly, geographical maps showing the distributions of the potential PV and wind capacity factors across the Nordics are developed, in addition to a map showing the anti-correlation between these resources across the region. Secondly, based on the derived linear function and the three maps, a fourth map is calculated showing the retrofitting project's NPV across the whole of the Nordics. Thirdly, adding the coordinates of the existing wind power plants in the region, we can thus see which of them are situated in the areas where retrofitting results in a high project NPV.

This methodology can thus help developers identifying which wind power plants in their region they should retrofit into PV-wind hybrids.