New electricity technologies
In the long-run model investments in new power plant capacity is possible apart from in lignite and waste power plants. Nuclear power investments can be endogenous or exogenous depending on the scenario.
For gas power, steam coal power, oil power and nuclear power it is assumed that all agents are in a position to invest in the most efficient technology (in the long-run model). The main source for efficiencies and investment costs for new power plants is “Projected Costs of Generating Electricity, Update 2010” from the OECD (2010). All estimates from this publication are for plants coming online in 2015.
OECD (2010) provides a range of estimates for investment costs for different types of coal-fired plants from different countries. The model distinguishes between steam coal and lignite power plants, however it is only possible to invest in new steam coal plants. The cost data for an ultra-super critical (USC) pulverised coal plant is used. For natural gas the majority of the estimates from OECD (2010) are for combined cycle gas turbine (CCGT) plants. The estimates differ between the reporting countries. In the model the cost estimate from Belgium (957 €/kW) has been used, which is very close to the average of all the CCGT-estimates in the publication.
When investment in nuclear plants is allowed, OECD (2010) has been used as the source for investment costs and efficiencies. The publication provides cost estimates for several different types of reactors; European pressurised reactor (EPR), pressurised water reactor (PWR) and VVER, which is the Russian version of the PWR. An EPR plant at the lower end of the range; 3228 €/kW, has been used. There are generally few cost estimates for new oil-fired power plants (Tyma 2010 and Schröder et al. 2013). After assessing the available sources an investment cost of 1411 €/kW is assumed.
The investment cost for new wind power plants was based on an assessment of various sources (IRENA 2012, IPCC, OECD 2010 Mott MacDonald 2010, NREL 2012 and NVE 2011). Offshore wind power potential is not included in the model. The cost estimates for onshore wind in OECD (2010) range from 1419 €/kW to 2742 €/kW. In the model the investment cost of a new wind power plant is 2298 €/kW. It is assumed that the investment cost falls over time at a rate of 1 % per anno towards 2050.
Numerous sources were reviewed for the cost of solar PV (IEA ETSAP 2011/E11, OECD 2010, IRENA 2012, IEA 2011, Bazilian et al. 2013, Schröder et al. 2013 and IPCC 2011). An estimate of 2520 €/kW is used, which is towards the lower end of the estimates of the sources mentioned. The reason for this is partly that some of the publications are several years old, and that the cost of solar PV installations has been dropping dramatically in recent years. It is assumed that the investment costs for solar power falls at a rate of 3 % per anno towards 2050.
A variety of technologies come under the term bio power, which means that deciding on a cost estimate for a generic plant is problematic. The cost of biomass-based power generation depends on type of feedstock used, boiler technology, plant capacity and type of plant. The estimates from OECD (2010) vary considerably from country to country, mainly due to differences in the reported technologies. IEA ETSAP has a range for typical values for a biomass CHP plant in 2010 and an estimate for expected costs in 2020. For new plants it seems reasonable to go with the lower end of what IEA ETSAP estimates today, assuming that the cost of a new biomass power plant is 2181 €/kW.
Grid connection costs
When investment is made in a new power plant, one of the cost components involved will be to get the plant connected to the grid. This additional cost of grid connection is not included in the cost estimates mentioned above, apart for reservoir hydro and run-of-river plants. So for all other technologies it is assumed that there is an additional grid connection cost when investing in a new power plant. This cost is made up of two elements; the new transmission lines to connect the plant to the grid and reinforcement of the grid as a result of a new plant coming online. Both of these aspects are often discussed with regards to wind power plants, as the good wind power resources may be located far from the load centres and because introduction of intermittent power sources requires the grid to be more robust.
In the model it is assumed that all technologies have an extra cost component linked to grid connection. According to IRENA (2012), this cost makes up between 9 and 14 % of total investment costs for onshore wind. It is therefore assumed that if the average wind power capacity in the model countries increases with 10 %, then the extra cost of grid connection (including the cost of upgrading the existing network due to congestion as a result of more plants coming online) for the marginal wind power plant is 10 % of the investment costs for wind power. This is the same for all countries. We then assume that 80 % of this cost is linked to actually connecting to the grid, whereas the remaining cost is linked to upgrading the grid.
For the other technologies we have assumed that the location of the plant is more flexible , so hence they have a smaller cost component for grid connection. This is supported by EIA (2012), which shows different cost components in the levelised cost of new electric generating technologies in 2017.