The Energy Collective Group

This group brings together the best thinkers on energy and climate. Join us for smart, insightful posts and conversations about where the energy industry is and where it is going.

9,853 Members

Post

For Eastern Europe, Controllable Renewable Power Is a Good Alternative for New Nuclear Power

Temelín nuclear power station, the Czech Republic

Poland, Slovakia, the Czech Republic and Hungary are all planning to build new nuclear power plants. But according to a new study by Energy Brainpool, commissioned by Greenpeace Energy, they could also opt for controllable renewable power plants. These are cost-competitive with nuclear, at least as reliable, and also allow for energy independence, write Philipp Heidinger, Fabian Huneke and Simon Göß from Energy Brainpool.

As a result of the decommissioning of coal-fired and nuclear power plants resulting from either political reasons or end of lifetime considerations, European power markets are in need for capacity replacement. Especially during the next decade, the need for controllable yet flexible, power generation will grow.

The Visegrád countries of Eastern Europe have ambitious plans for the construction of new nuclear power plants in order to replace older generators. In Hungary, two reactors with a total net capacity of 2.4 GW are to be installed at the Paks site by 2026. The Czech Republic is also planning the construction of two new reactors, also 1,200 MW each, at the existing Temelin and Dukovany sites. Slovakia wants to replace its Bohunice reactor (1,200 MW) in the mid-2020s and is already building two small new reactors, Mochovce 3 and 4 (total 900 MW), which are supposed to come online this year and the next.

In order to meet the demand for electricity at all times, a cRE power plant system not only consists of wind and solar plants, but also of electrolysers and methanisation facilities connected to gas power plants

Slovakia is also planning a new plant at Kecerovce (1,200 MW). Poland has plans for a 3 GW nuclear power plant to go online by 2029 and another 3 GW by the mid-2030s, though no location has been selected yet. In total the four countries aim to install 15.6 GW of nuclear power plant capacity in the coming decades.

Figure 1 depicts existing and planned nuclear power plants in the four countries, where the circular area indicates net capacity of planned reactors.

Figure 1: Status of nuclear power plant projects in the Visegrád countries. [1]

The alternative: controllable renewable energy power plants

Could the capacity needs also be met through renewable energy at comparable costs? In our study, we designed a cost-optimized system of controllable renewable energy (cRE) power plants that have at least the same security of supply and energy independence levels as the proposed nuclear power stations.

As intermittent renewables can only meet the demand for power at times when there is enough solar radiation or wind speeds, they alone cannot reliably cover immediate electricity needs. In order to meet the demand for electricity at all times, a cRE power plant system not only consists of wind and solar plants, but also of electrolysers and methanisation facilities connected to gas power plants. Figure 2 depicts the concept of such a cRE power plant in comparison to a nuclear power plant.

Figure 2: The concept of the cRE power plant compared to a nuclear power plant.

How do the two solutions compare in cost?

Our cost analysis shows a range between 87 and 126 EUR/MWh (all costs are based on 2016 values) for current European nuclear power projects. The actual levelised cost of electricity of the nuclear power plant Flamanville III in France is evaluated at 126 EUR/MWh and the state subsidy for Hinkley Point C in the UK stands at 119 EUR/MWh. Note that these costs are considerably higher than usually given in the literature or in project plans, where figures vary between 55 and 80 EUR/MWh.

Total cost includes capital costs (CAPEX) and the significantly lower operating and maintenance costs. The range of capital costs shown in Figure 3 is largely driven by the planned/actual CAPEX and the financing structure. This includes expected returns and risk premiums for construction. With values between 38 and 100 EUR/MWh, these represent a wide range of fixed costs for the initial investment.

Once built, the nuclear power plant is one of the cheapest generating technologies. However, capital costs from the initial investment can result in high total and levelised cost of electricity generation.

This is due to the range of weighted average cost of capital (WACC)[2] between 7 and 10 per cent on the one hand and the high divergence of investment costs in the literature and current planned and actual values of European nuclear power projects on the other hand. While the inflation-adjusted maximum value of CAPEX from the literature examined leads to a maximum cost of 54 EUR/MWh, the current actual CAPEX value from the French new construction project Flamanville III with a WACC of 10 per cent results in a base for capital costs of 100 EUR/MWh already. In addition, the operation and maintenance costs range from 17 to 25 EUR/MWh.

In other words: once built, the nuclear power plant is one of the cheapest generating technologies. However, capital costs from the initial investment can result in high total and levelised cost of electricity generation. Costs incurred during dismantling or risk premiums during operation are often borne by states and thus considered external costs. They are not taken into account in the numbers below.

Figure 3: The range of cost components for nuclear power plants in current European projects, derived from the relevant literature based on 6,500 hours of full load hours and a lifetime of 50 years.

How does a controllable renewable energies power plant work?

A cRE power plant uses surpluses from the generation of variable renewable energy sources (vRES) in the electrolysis process. Subsequent enrichment of the separated hydrogen with carbon dioxide yields synthetic gas. This can be fed into the existing gas grid or stored in gas storage facilities, or it can be used to generate electricity with various gas-fired power plant technologies.

Figure 4 shows the hourly residual load (renewable generation subtracted from electricity demand) over the period of one year. The different components of a cRE power plant can be dimensioned based on a country’s potential in wind and solar resources. In the exemplary case in Figure 4, the cRE is expected to provide a constant generation of 1 GW due to the possibility of shifting electricity generated in negative residual load situations via electrolysis to situations where a positive residual load occurs.

Figure 4: The hourly residual load of base load power demand when supplied by intermittent renewable energies and visualisation of options for dimensioning cRE power plant components.

So what are the total costs of the cRE power plant?

Even with the expensive financing conditions for renewable energies that currently prevail and without a joint optimisation of the Visegrád countries among themselves, the costs are comparable to those of nuclear power plants as Table 1 shows.

In Poland levelised cost of electricity of cRE power plants are around 112 EUR/MWh, in the Czech Republic 119 EUR/MWh and in Hungary 129 EUR/MWh. In Slovakia, the potential is still unclear. Since there is still little experience with wind power, initial analyses show high costs of 167 EUR/MWh due to poor wind conditions.

The average levelised costs of electricity for such a power plant system converting excess electricity into electrolysis gas are significantly lower when the electrolysis gas is distributed across all Visegrád countries. The distribution can be enabled on the basis of joint market and balancing group agreements, i.e. via the existing European gas grid. In this case the costs are assumed to be 120 EUR/MWh in 2027 and 100 EUR/MWh in 2035 on the assumption of uniformly declining financing conditions in the four countries.

Table 1: Cost-optimised dimensioning of the cRE power plants in the Visegrád countries for two selected years. Costs in 2016 EUR value. (Source: own calculation in April 2018). Due to very limited experience with wind power in Slovakia, actual wind potential has not been sufficiently studied and a very low level of potential has been assumed in these calculations.

Which factors determine the total costs of a cRE power plant?

In order for a cRE power plant to be economically optimized, the individual components must be dimensioned to optimize overall costs. The national wind and solar potential, but also the investment conditions and technical parameters influence this dimensioning. The total costs in EUR/MWh of the cRE power plant, therefore, consist of two parts, which is also depicted in Table 1.

Firstly, the minimum electricity generation costs in EUR/MWh of the vRES are calculated by varying the ratio of installed PV and wind power. This takes into account the national hourly wind and solar potential as well as the respective technology costs. The modelling shows that an optimal share of 70 to 80 per cent for wind onshore minimises overall costs. This is not due to particularly low wind power generation costs, because those of PV are about the same or even lower. Rather, the modelled ratio of wind and PV leads to cost-optimized direct electricity consumption without the need for efficiency-reducing intermediate storage in electrolysis gas.

While value creation in case of nuclear power plants could include domestic processing and thus a highly qualified and skilled workforce over decades, in the case of cRE power plants other new opportunities arise

The levelised cost of electricity of the vRES ranges between 73 and 90 EUR/MWh. For comparison: according to the current tender results, new wind and PV electricity in Germany is only remunerated with 40 to 50 EUR/MWh. Electricity from these renewables could thus already be significantly cheaper. The higher values in the V4-states can be explained by the prevailing poor financing conditions and thus high capital costs there.

Secondly, the additional costs for controllability in EUR/MWh are calculated by varying the optimum capacity of electrolysers in MW and by determining a cost-optimal composition of the gas-fired power plant capacity. These additional costs are strongly dependent on the cost degression of electrolysers and the efficiency rate assumed at 70 per cent including methanisation. In the analysis, the specific costs for electrolysis including methanisation in EUR/MW per year are expected to decline by 55 per cent from 2027 to 2035.

What are concrete steps to implement the cRE power plant politically?

Successful implementation of the cRE power plant concept can be achieved by adapting the regulatory framework for the expansion of vRES and by continuous investment in electrolysis technology. The latter must be transferred to industrial series production to reach the assumed cost degression along with further technological development.

Up to now, the expansion of vRES in South-East Europe is not economical for project planners due to high capital costs (WACC). In a paper, Agora Energiewende proposes a concept based on contractual agreements between the EU Commission, member states and project planners in order to create planning security for investments in vRES. In addition, the European Commission has announced that it will step up support for storage research, which includes the production of synthetic gas by electrolysers.

Furthermore, the existing grid connections and the available area of nuclear power plant sites can be used for the expansion of vRES, the gas grids should be maintained and, if necessary, modernised. An example of the continued use of existing grid connections is France, where a tender of 300 MW for PV will be launched at the Fessenheim nuclear power plant site by the end of 2018.

Clearly, the V4 states will experience a structural transformation of its economies in a scenario including substantial amounts of vRES and cRE power plants. While value creation in case of nuclear power plants could include domestic processing and thus a highly qualified and skilled workforce over decades, in the case of cRE power plants other new opportunities arise. These include the generation and storage of synthetic gas, the manufacturing of key components of the cRE power plants, along with a beneficial economic development of rural areas due to the decentralized character of the plants.

Notes:

[1] The reactor blocks at Mochovce, already advanced in construction, are categorized as operational.

[2] Weighted average cost of capital WACC (real) depict weighted interest rates, which are calculated from interest rates for debt, interest rates for equity depending on the expected return and the inflation rate. With their help, long-term investments with future cash flows are converted to annual values and thus become comparable.

By , and 

Philipp Heidinger (philipp.heidinger@energybrainpool.com), Fabian Huneke (fabian.huneke@energybrainpool.com) and Simon Göß (simon.goess@energybrainpool.com) are experts at Energy Brainpool, a Berlin-based consultancy offering independent energy market expertise with a focus on market design, price development and trade in Germany and Europe.

In 2003, Tobias Federico founded the company with one of the first spot price forecasts on the market. Today, the offer includes fundamental modeling of the electricity prices with the software Power2Sim as well as diverse analyses, forecasts and scientific studies. Energy Brainpool advises on strategic and operational issues and offers expert training since 2008.

Original Post

Content Discussion

David Hervol's picture
David Hervol on May 16, 2018

Philipp, is Germany building a lot of these cRE plants since they appear to be cost effective? Do they require government subsidies?

Engineer- Poet's picture
Engineer- Poet on May 16, 2018

Going through this piece one detail at a time, I am struck by how it starts with unfavorable assumptions about the proven alternative and degenerates into utter fantasy.

The range of cost components for nuclear power plants in current European projects, derived from the relevant literature based on 6,500 hours of full load hours and a lifetime of 50 years.

This is a capacity factor of just 74%, which is absurd.  US plants already hit capacity factors over 90%.  Given the storage technology posited for the cRE scheme, nuclear plants could run breaker-to-breaker at almost 100% power between refuelings.  Running 2-year cycles with e.g. Lightbridge fuel and 30-day refueling outages, plants would be able to hit 96% capacity factor.  Compared to 74% CF, this would cut the amortization cost per kWh by 23%.

Second is the absurdly low 50-year lifespan.  40-year-old reactors are being cleared to run to 60 years, and some are already looking to relicense to 80 years.  Current designs are aiming at 60 years for the first license interval and are likely capable of running for 100 years.

In Poland levelised cost of electricity of cRE power plants are around 112 EUR/MWh, in the Czech Republic 119 EUR/MWh and in Hungary 129 EUR/MWh.

These numbers are based on pure speculation, as nobody has actually built a full-scale system.  The cost figures for the pilot plants so far make energy something only the wealthy could afford.

In this case the costs are assumed to be 120 EUR/MWh in 2027 and 100 EUR/MWh in 2035 on the assumption of uniformly declining financing conditions in the four countries.

Why not assume the same financing costs for nuclear?

Those financing costs would be anything but small.  If the Visegrad countries have similar wind CF to Germany, they’d be getting just 21.3% of nameplate output on average.  This runs smack into your little admission against interest:

the modelled ratio of wind and PV leads to cost-optimized direct electricity consumption without the need for efficiency-reducing intermediate storage in electrolysis gas.

Speaking of that efficiency reduction, it takes on the order of 46 kWh to produce 1 kg of hydrogen in today’s best commercial electrolyzers; this is about 73% efficient.  There’s a further loss in methanation.  Methanation reacts 4 moles of H2 with 1 mole of CO2 to yield 1 mole of CH4 and 2 moles H2O.  Hydrogen has a HHV of 286 kJ/mol, while CH4 has a HHV of just 891 kJ/mol; this represents a cold-gas efficiency of 78%.  There’s 10% loss between HHV and LHV for non-condensing devices like gas turbines, and the best CCGTs peak out at about 62% efficiency on the LHV basis.  Ergo, at BEST you get a busbar-to-busbar efficiency of slightly less than 32%.  If the power feed into the storage system costs the V4 projection of €60.36/MWh, the energy cost alone for power coming OUT of storage is €189.6/MWh.  To this you must add amortization and O&M.  Given the low capacity factors for wind and PV in the region, a very large fraction of total power demand must go through storage.

Nuclear at €125.6/MWh looks very good compared to “renewables”, especially unreliable renewables which are subject to the vagaries of weather and have never been shown running an industrial economy’s grid anywhere, ever.

Nathan Wilson's picture
Nathan Wilson on May 16, 2018

A crucial omission in the economic argument here is that there are actually four separate deployment phases which have to be cost justified, not a single project:
1) renewable generators whose output can be used directly with little curtailed energy.
2) renewable generators whose output will mostly be sold at deeply discounted prices to syn-fuel plants (most such proposal imagine prices in the 0-2 cent/kWh range).
3) syn-fuel plants which have low capacity factor due to variable electric supply.
4) methane-fueled generators which operate at low capacity factor. These must include CO2 capture, otherwise their CO2 emissions are as if they ran on fossil fuel.

Even if phase one is economical (and in most places today it is not, unless generous susidies are available) , the next three are even more difficult.

One might also ask why the selected syn-fuel is methane instead of carbon-free hydrogen? In power plant applications, hydrogen can be used nearly as easily as methane, and of course hydrogen needs no expensive CO2 capture systems (in a market with a price on carbon emission, a cost will still be incurred even if with re-used CO2 since the cost is only incurred on emission.

Unfortunately, even with the best of initial intentions, and an attractive average cost, such as a system has a high likehlihood of stopping at phase 1, with most electricity coming from fossil fuel fired power plants.

Willem Post's picture
Willem Post on May 17, 2018

EP,
Thank you totally debunking this article with real-world facts and figures. You beat me to it.
Germany, etc., likely would be limited to at most 10% solar, because of Duck Curve issues.

The Cost of Duck Curves due to Solar
See URL.
http://www.windtaskforce.org/profiles/blogs/wind-and-solar-hype-versus-reality

Duck curves are entirely due to too much solar generation during midday, i.e., a midday Tsunami. This requires the traditional generators to significantly ramp down their outputs. However, in late afternoon/early evening, solar being minimal, these same generators have to significantly ramp up their outputs to meet peak demand. The daily down and up ramping severely stresses 1) the traditional generators (causing more wear and tear, more Btu/kWh, more CO2/kWh) and 2) the grid.

Grid operators have three approaches to deal with the duck curve, all of which are costly. However, the costs will not be charged to solar system owners, as that would impair the fantasy of solar being low-cost:

1) Having adequate traditional plant capacity to enable changing grid operational practices and have more frequent power plant cycling (up and down ramping at part load), and more frequent cold starts and stops, and increased synchronous hot standby, etc.
2) Shifting part of demand to midday so solar can meet parts of the load that would not normally be provided in the middle of the day.
3) Require owners of rooftop solar, mostly residential, and owners of field-mounted solar, mostly utilities, to have adequate battery system capacity to store their midday solar electricity, instead of just dumping it onto the grid for the owners of traditional generators to deal with.

With the possible exception of those in equatorial latitudes, every jurisdiction in the world that commits to include more than, say, 20% solar in its future generation mix likely will reach a duck curve threshold where daily ramping and storage/curtailment/load-shifting requirements become unmanageable and too expensive.

The level at which this threshold is reached will vary depending on local conditions, but it will generally be lower at higher latitudes than at lower latitudes, and could be as low as 10% at high latitudes; German solar in 2017 = 39.9/654.8 = 6%; wind 106.6/654.8 = 16.3%. For now, Germany still manages 6%, because it is “allowed” to spread its excess solar electricity at near zero or negative cost to nearby grids.
http://www.ag-energiebilanzen.de

As much of the world’s electricity is generated and consumed at high latitudes (40 – 60) one has to question whether solar isn’t more trouble than it’s worth.

http://euanmearns.com/the-california-duck-curve-isnt-confined-to-california/
https://www.vox.com/energy-and-environment/2018/3/20/17128478/solar-duck-curve-nrel-researcher

Willem Post's picture
Willem Post on May 17, 2018

EP,

Russia pipeline $4/MMbtu

US gas in Europe (NYMEX) $3.33/MMBtu.
Liquefaction, regasification and transportation $4/MMBtu
Total $7.33/MMBtu

Using Russian pipeline gas in 60% efficient CCGTs would have low-cost electricity and low CO2/kWh, and none of the hassles of variability, intermittency and Duck Curves.

http://energypost.eu/depoliticization-european-gas-market/

Bas Gresnigt's picture
Bas Gresnigt on May 17, 2018

As wind, solar and storage are widely predicted to continue their major cost decreases during next decade, while
nuclear will continue its costs increases (a o. due to increased safety demands),

the renewable alternarive will become more and more cheaper compared te nuclear!

Willem Post's picture
Willem Post on May 18, 2018

EP,

More on Duck Curves in California.

Figure 2 in the URL shows an up-ramp of about 13000 MW in 3 hours, or 40,000 MWh, delivered as AC to high voltage grids.

If all coal, gas, oil and nuclear plants were shut down, and wind and solar were near zero in late afternoon/early evening, and almost all of the electricity were supplied by battery systems, the capital cost would be at least 40000 x 1000 x $400/kWh = $16 billion.

The battery systems have an AC-to-AC round-trip loss of about 20%, and would need to be charged and discharged several times during the other 21 hours. See Appendix in second URL.

https://www.caiso.com/Documents/FlexibleResourcesHelpRenewables_FastFacts.pdf
http://www.windtaskforce.org/profiles/blogs/wind-and-solar-hype-versus-reality

Willem Post's picture
Willem Post on May 18, 2018

Bas,
You need to read this article.
Also the embedded URLs and Appendix.
http://www.windtaskforce.org/profiles/blogs/wind-and-solar-hype-versus-reality

Bob Meinetz's picture
Bob Meinetz on May 18, 2018

Philipp, leave it to renewables advocacy to compensate for intermittency, the primary deficiency of renewable energy, by denying it exists – by spreading the fake news that intermittent power from the sun and wind can now be “controlled”.

It wouldn’t be so objectionable if renewables promoters knew the difference between power and energy, two distinct entities precisely defined within physics. But a lie by any other name is still a lie, and a lie is what “controllable renewable power” is and always will be.

Bas Gresnigt's picture
Bas Gresnigt on May 18, 2018

The authors of the post are still thinking in power plants which is gradually becoming obsolete due to the ongoing paradigm change with upcoming VRE’s.
Reliable electricity supply is gradually becoming a matter of a good grid organization which combines & utilizes many thousands of small generators (wind, solar, CHP’s, different types of storage, hydro, biomass, geothermal, etc),

as well as consumption adaptations as we see with German alu smelters, and will see with consumer washing-, drying-, dishwashing machines being controlled by cheaper price indications from the smart meter.

Big and baseload fossil and nuclear power plants are far too expensive in the future, considering that they require substantial staff and expensive rotating backup as they fail sometimes in a second, etc.

Bas Gresnigt's picture
Bas Gresnigt on May 19, 2018

The Germans do it more smart. They distribute production in many different facilities.

They have already ~20 major (MW scale) pilot PtG facilities. Those use many different methods producing different gas (from H2 to car regular fuel) etc.
Regular rollout planned to start in 2024.

With increasing emission costs it will become economical to fly with renewable H2 driven planes!

Bas Gresnigt's picture
Bas Gresnigt on May 19, 2018

Willem,
You forget taxes, etc.
We in NL enjoy our own major gas source in Slochteren, lower cost price than Russian gas. Still our. gas is far more expensive. So much that using imported coal was cheaper for baseload power plants.

Bas Gresnigt's picture
Bas Gresnigt on May 20, 2018

@EP,
Your objections:
Nuclear load factor and life span
In the coming higher VRE environment NPP’s score lower load factors as they try not to produce when the market price is below their marginal costs, which is then rather frequent.

No NPP suceeded to operate longer than 50yrs. Considering the ongoing fundamental changes its highly unlikely any new NPP will last longer.
Av. age of present NPP’s when they stop is ~35yrs!

Levelized costs of cRE
The posts uses ~2016 costs of cRE. Those are now already ~15% lower due to developments such low windspeed wind turbines, etc.

The post should have used the installation costs in 2025-2030, the period those scheduled NPP’s become online. Then the installation costs of cRE are widely predicted to become another ~40% lower.

So all in all nuclear will deliver those countries electricity against cost prices roughly twice those in high renewable countries.

Bas Gresnigt's picture
Bas Gresnigt on May 21, 2018

Willem,
The duck curve is also a price curve which allows storage conpanies to earn money each day!
So when the duck curve becomes serious entrepeneurs will jump in and start storage facilities (pumped storage, batteries, etc) as well as simple peakers.

Mark Heslep's picture
Mark Heslep on May 21, 2018

Many US reactors are now licensed to 60 years.

Commercial nuclear power is only 58 years old (Yankee Rowe). The modern generation II design that dominates the world of operating reactors is ~ten years younger still.

Bas Gresnigt's picture
Bas Gresnigt on May 21, 2018

And many US reactors licensed to 60yrs stop before becoming 50yrs….

Bas Gresnigt's picture
Bas Gresnigt on May 22, 2018

And industrial consumers will adapt consumption such that they pay less.
Same for consumers once they have smart meters which allow for hourly following of the wholesale market prices. Especially when those meters can controle the start of (dish)washing and drying machines.

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

We are always happy to have new members join us in that mission. If you have the knowledge to share, it is easy to post on Energy Central. Once your post is approved, it will be published on our website (with more than 100,000 users/month) and may also be published in one of our newsletters (which have between 10,000 and 60,000 circulation).

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »