Saturday, May 14, 2022

More "renewable" energy, more pricey energy

A climate change hystericist annoyed me enough that I dug up this OECD study on how the more you use "renewable" energy, the more expensive power gets:

 

The Costs of Decarbonisation: System Costs with High Shares of Nuclear and Renewables | en | OECD

"The study highlights the impacts the variability of wind and solar PV production have on electricity system costs, which appears as costly adjustments to the residual system...

Renewable energies have enjoyed in recent years both popular and political support. While average costs per MWh of wind and solar PV are still somewhat higher than those of nuclear energy, the cost gap at the level of plant-level generation costs (as calculated with the levelised cost of electricity (LCOE) methodology as set out in the 2015 OECD study on the Projected Costs of Generating Electricity) no longer seems insurmountable. However, as spelled out in the first OECD Nuclear Energy Agency (NEA) study on system costs, Nuclear Energy and Renewables: System Costs in Decarbonising Electricity Systems (2012), VRE [variable renewable energy] technologies such as wind and solar PV cause a number of additional costs to the system, which are referred to as system costs.

The most important categories of the system costs of VREs are increased outlays for distribution and transmission due to their small unit size and distance from load centres, balancing costs to prepare for unpredictable changes in wind speed and solar radiation and, perhaps, most importantly, the costs for organising reliable supplies through the residual system during the hours when wind and sun are not fully available or not available at all. Variability also induces significant changes in the composition of the remainder of dispatchable technologies that ensure round-the-clock security of supply in the power system. When deploying VREs, one observes, in particular, a shift from technologies with high fixed cost, such as nuclear power to more flexible technologies with low fixed cost such as gas-fired power generation. While the latter will be able to better absorb the loss of operating hours due to VRE infeed, the overall costs of the residual system will increase, an effect known as “profile costs”. In addition, deploying VREs does not automatically translate into carbon emission reductions. For instance, when nuclear power is substituted by a mix of VREs and gas-fired generation that produces electricity when VREs are not available, overall carbon emissions will increase.

All technologies have system costs. Nuclear, for instance, requires particularly strong network connections and access to reliable cooling sources. However, these costs turn out to be an order of magnitude lower than those imposed by the variability of renewable energies. The key advantage of nuclear power in the economic competition with wind and solar PV is the fact that nuclear power plants are dispatchable, i.e. they can produce large amounts of carbon free baseload power in a reliable and predictable fashion...

Since the load factor and the capacity credit of VRE is significantly lower than that of conventional thermal power plants, a significantly higher capacity is needed to produce the same amount of electricity. While about 98 GW are installed in the base case scenario without VRE, the deployment of VRE up to penetration levels of 10% and 30% increases the total capacity of the system to 118 and 167 GW, respectively. The total installed capacity would more than double to 220 GW if a VRE penetration level of 50% must be reached. More than 325 GW, i.e. more than three times the peak demand, are needed if VRE generate 75% of the total electricity demand. In other words, as the VRE penetration increases vast excess capacity, thus investment, is needed to meet the same demand...

Profile costs result from the de-optimisation of the residual system due to the variability of VRE. Total system costs, expressed in USD per unit of net electricity delivered by VRE to the grid are shown in Figure ES6 for the four scenarios of 10%, 30%, 50% and 75% VRE as well as for the two sensitivity scenarios. These system costs must be understood as the increase of the total costs to provide the same service of electricity supply above the costs of the least-cost scenario without any VRE...

System costs vary between less than USD 10 per MWh of VRE for a share of 10% of wind and solar PV to more than USD 50 per MWh of VRE for a share of 75% of wind and solar PV...

These values need to be compared to the plant-level generation costs of VRE, which range, depending on the scenario, from USD 60 per MWh for onshore wind to up to USD 130 per MWh for solar PV. It should also be noted that the system costs are largely unaffected by any declines in plant-level costs as long as the share of VRE remains exogenously imposed. Indeed, all four components of system costs (balancing, profile, connection and grid costs) increase with the deployment of VRE resources, but at different rates. By adding system costs to the costs of plant-level generation as assessed in LCOE calculations, one can calculate the total system costs of electricity provision for the eight scenarios analysed in this study (see Figure ES7 above).

With 10% of VRE in the electricity mix, total costs increase only about 5% above the costs of a reference system with only conventional dispatchable generators, which in a mid-sized system such as the one modelled corresponds to additional costs of about USD 2 billion per year. At 30% VRE penetration, costs increase by about USD 8 billion per year, i.e. by 21% with respect to the base case. Reaching more ambitious VRE targets leads to considerably higher costs. Total costs increase by more than USD 15 billion per year if 50% of electric energy generation is provided by variable renewable resources, which corresponds to an additional 42% of costs compared to the base case. Reaching a 75% VRE target finally implies almost doubling the costs for electricity provision to almost USD 70 billion per year, representing more than USD 33 billion above the base case.

A striking effect of the deployment of low marginal cost variable resources on the electricity market is the appearance of hours with zero prices, a substantial increase in the volatility of electricity prices and the commensurate increase in capital cost (not modelled here). Such zero prices are not observed in the two scenarios with no or low VRE deployment but start appearing for 60 hours per year when VRE reach a penetration level of 30%. The number of occurrences increases dramatically with the VRE penetration level; at 50%, more than 1 200 hours in a year feature zero-price levels, i.e. about 14% of the time. When VREs produce 75% of the demand, zero prices occur during 3 750 hours, i.e. more than 43% of the time (see Figure ES8). Since the model works under a financing constraint, the higher frequency of hours with zero prices is compensated by an increase in the number of hours with high electricity prices, which increases volatility. At 75% VRE penetration, the number of hours with prices above USD 100 per MWh is more than double that at zero or low VRE penetration rate.

Last but not least, the generation by VRE as a function of the availability of natural resources such as wind speed or solar radiation, is not only more variable than that from dispatchable plants but also more concentrated during a limited number of hours. Periods with high generation are followed by periods with lower or zero output. Because they all respond to the same meteorological conditions, wind turbines and solar PV plants tend to auto-correlate, i.e. produce disproportionally more electricity when other plants of the same type are generating and to produce less when other wind and solar PV plants are also running at lower utilisation rates. In combination with the zero short-run marginal costs of VRE resources, this causes a decrease in the average price received by the electricity generated by VRE as their penetration level increases, a phenomenon often referred to as self-cannibalisation...
 
The average price received by solar PV and wind resources in the electricity market declines significantly and non-linearly as their penetration level increases, and this price decrease is much steeper for solar PV than for wind as its auto-correlation is higher. The value of the solar PV generation is almost halved even when a penetration rate of only 12.5% is reached. Further deployment of solar PV capacity to a penetration level of 17.5% would further halve its market value to below USD 20 per MWh. Thus, even if the generation costs of solar PV were divided by five, its optimal penetration level would not exceed 17.5%. A similar trend, although less pronounced, is observed for onshore wind, which has a higher load factor than solar PV and whose generation spans over a larger time period. At a penetration level of 22.5%, the value of a megawatt-hour of wind is reduced by 25%. For penetration levels above 30%, the market value of wind electricity is below USD 50 per MWh compared to an average price of all electricity of USD 80 per MWh."
 
 
Besides the headline result as summarised earlier, the true cost of "renewable" energy is higher than the cost of generating it (which is the sexy headline figure that environmentalists love to cite in claiming it is very cheap), once you take into account the cost of distributing it, how unreliable it is and above all covering base load (providing power when the sun isn't shining and/or the wind isn't blowing)

In addition, if you replace nuclear energy with "renewables", carbon emissions go up.
 
Also, you need to build a lot of redundant infrastructure to generate "renewable" energy due to how unreliable it is (so much for it being "green").

High costs aside, we know that markets hate uncertainty, so the unreliability of green energy has even higher costs than are apparent from just looking at these dollar values alone.
 
 
The tragedy is that most consumers who see higher energy prices won't make the connection to renewables, and will just blame "greedy" companies.

blog comments powered by Disqus