The cost of generation from variable renewable electricity sources such as wind and solar has reduced dramatically in recent years. This is clearly very good news, but it does throw into sharper focus the additional costs of integrating variable renewable electricity (VRE) into electricity systems.
Unfortunately, the impacts that give rise to the additional costs are sometimes counter-intuitive, often complex, and almost always dependent on the characteristics of the system to which the VRE is connected. This leads to wide ranges of cost estimates, depending on the system being assessed, the impact being costed, and the projections and assumptions that are made about the future system.
The analysis and debate around the cost of integrating VRE have been running for several decades. The UKERC Technology and Policy Assessment (TPA) team undertook its first systematic review of this subject back in 2006, with a follow-up review in 2016.
In recent years, many countries have dramatically increased the contribution of VRE to total electricity generation, and have aspirations to grow this resource further. In response to this, there have been many different analyses of the potential impacts, and in our paper published in Nature Energy this week, we show there is a rich and evolving evidence base for the costs of integrating VRE.
Our paper draws upon the data set created from the 2016 review and updates it to incorporate more recent analyses, and to reflect the development of how these impacts are characterised and assessed. The dataset has been deposited with the UKERC Energy Data Centre so that it is publicly available.
For lots of good reasons, many analyses start by breaking the complex and often interdependent set of impacts into more easily understood categories. This inevitably involves a degree of simplification which frequently leads to vigorous debate around the classification of impacts. Nevertheless, such categorising does help to explain the impacts in a more accessible way, particularly to those of us who are not experts in the field.
In our paper we explain the range of impacts and summarise the evidence for their cost implications, where such data is available. Below we focus on the findings for the two most important and well characterised impact categories: operating reserve costs and capacity adequacy costs
Operating reserve costs, are costs that result from the unpredictability or incorrect forecasting of VRE output over short timescales (seconds to a few hours).
Capacity adequacy costs are associated with the extent to which VRE output can be relied upon to help meet times of peak demand which occur during the year.
We found that median values for operating reserve costs were less than €5/MWh when VRE contributed up to 35% of annual electricity production, and less than €10/MWh up to 45%. The range is very dependent on the actual or assumed flexibility of the electricity system being observed or modelled.
For capacity adequacy, we found that median cost values did not exceed €10/MWh for any level of contribution from VRE. The range of values was particularly dependent on the assumed to provide capacity adequacy.
Findings from other analyses that have modelled total or aggregated integration costs are broadly consistent with these values. However, at higher shares of VRE (for example over 50%) we found that there is a very wide range of total integration costs. The primary determining factor here is the assumptions around the provision and cost of system flexibility.
There is a lot more detail on the impacts of VRE in our paper and the supporting documents (and in our earlier UKERC reports). The main message is that integration costs are likely to be relatively low when the contribution from VRE to total electricity supply is in the low tens of %. At higher levels, significantly enhanced system flexibility may be required to minimising integration costs. A key task for policymakers and regulators therefore is to ensure that this flexibility is provided in the most economically efficient way.
To view the paper click here.