The operating characteristics of individual renewable electricity-generating technologies are becoming better understood. Whether powered by wind, waves, tides or the sun, the fundamental characteristics of these renewable technologies are the inherent variability of their output and the fact that they are non-dispatchable; when the required forces of nature are present, electricity is generated and vice versa. As the penetration of renewable energies increases, this variability is likely to have a significant impact on the market and has consequences for supply security.

Diversifying the variable renewables capacity, of for example wind power, with marine technologies could yield significant benefits in dampening the variability of output from the renewables fleet. The UK has huge potential to harness the power of waves and tides around its coastline: wave power has much lower correlations with wind, while both tidal range and tidal stream have no correlation at all.

The following is an overview of some of the economic benefits of a more diversified mix of variable renewable electricity-generating technologies in terms of the costs of generation to meet demand. The mix and location of variable renewables plant also has implications for the costs of transmission but these are not considered here.

Reduced requirement for back-up capacity

One measure used to determine the contribution to system reliability of intermittent renewable energy generation is the capacity credit, which traditionally measures the percentage of maximum potential output that can statistically be shown to contribute to security of supply. Often expressed as the amount of generation from conventional sources that could effectively be ‘replaced’ without any reduction in security of supply, the capacity credit can be viewed as a like-for-like comparison of the contribution to security of supply of different types of generation with their different output patterns and availability.

However the capacity credit for any given renewable energy plant is not constant: it depends on a number of factors including total amount of plant in operation, the geographical distribution of different intermittent technologies, the relationship between output levels at different locations and the proportion of each technology in the energy mix.

The average capacity credit of a particular renewable technology such as wind declines as the amount of installed capacity of that technology increases, since there is an increasing risk that periods of low output from variable renewables plant could coincide with periods of high demand. If wind is the only source of variable generation then significant back-up capacity will be required to cover the reducing contribution to security of supply. The amount of conventional capacity required to back up each additional MWh of output increases as more variable renewable plant is built.

However, this comes with an important caveat. It is perhaps not unreasonable to expect that adding wave and tidal capacity would lead to an increase in the total capacity credit of the entire renewables mix and hence lead to a reduction in the amount of back-up capacity required. As more wind is replaced by marine technologies, the incremental capacity credit of the wind plant being replaced increases. At the same time, the incremental credit of the wave and tidal capacity decreases. This suggests that there should be an optimal mix of variable technologies in terms of its aggregate capacity credit.

Reduced volume and cost of reserve and balancing capacity
The accuracy in forecasting changes in wind power output several hours out is a growing area of research, and one that has a significant impact on the amount of reserve capacity that must be provided. Reserve capacity is required to fill sudden and unforeseen shortfalls in supply or increases in demand; improving the ability of the system operator to predict such changes will reduce the amount of the reserve that is needed.

Although some degree of variability still applies, output from a wave power plant is less variable than that of a wind plant, and can be forecasted with more accuracy. What’s more, output from a tidal plant may be predicted accurately years in advance. This ability to increase the predictability, combined with a reduction in maximum hourly changes in output, may reduce the requirement and the costs of reserve. However, tidal schemes such as the Severn Barrage may require additional reserve and system balancing arrangements due to their size and the consequent risk if there were an outage but these are not considered here.

Reduced frequency and quantity of spilled power

If it is very windy there is going to be significant output at certain times, to the point where demand may be exceeded. Unless other plant can be turned down, demand increased or electricity exported, this excess demand will be spilled or wasted. For the purposes of this analysis to illustrate the benefits of renewables diversification, we have assumed limited increases in flexibility in these areas. As a result, if the output from variable renewables comes solely from wind, then we can expect around 5% of it to be spilled. This means that of the 120 TWh of output that comes from wind generation in our example case, 6 TWh will be lost. That volume of spill could be reduced by greater interconnection to other systems, more dynamic demand response, for example charging electric cars to absorb the excess output, or by diversifying the mix of renewables.

For example, when wind power provides only 60% of the total variable renewables output – or 72 TWh – the amount spilled can be reduced by half to around 3 TWh. Although, interestingly, introducing tidal range capacity would be less beneficial.

Looking at this in a little more depth, if the 120 TWh of wind generation is part of an overall renewable energy target, and 5% is being spilled, only the remaining 114 TWh would count towards the target. The capacity that produces this lost output is effectively ‘redundant’ investment, and must be made up in some other way thus increasing the costs of meeting the target.

Reduction in CO2 emissions and fuel usage

Increasing diversification in intermittent renewable supply also reduces carbon dioxide emissions for several reasons. Firstly, less zero-carbon renewable generation is spilled which in turn requires less generation from conventional generation to meet annual demand. Secondly, reducing the part-loading of conventional plant for reserve and system balancing creates a significant cut in emissions. And finally, flattening the ‘peakiness’ in the combined output of renewable plant requires less generation from the more expensive, less efficient and hence high-carbon conventional generation.

To put all this into context, increasing diversity could reduce the emissions assumed in the example case by approximately 6%. Fossil fuel consumption would also reduce for similar reasons.

Levels of subsidy required under the Renewables Obligation

The correlated output of wind generation that causes the problems associated with spill also has an impact on the returns to be secured from investment in wind plant.

At current levels of use, wind plant in Great Britain should in theory be able to capture a price slightly better than that of the baseload. However, as penetration levels rise, the impact of wind on power prices starts to become more apparent. Significant volumes of wind generation in any given period will displace conventional generation, reduce the cost of marginal plant and lower market prices. This has already been observed in countries such as Spain, Germany and Denmark, where levels of wind capacity are relatively high.

As a result, there begins to be a negative correlation between wind output and the price that generators receive: the higher the wind capacity, and the more correlated the output levels of that plant, the lower the prices that capacity will receive for each MWh of power generated. Conversely, prices will be higher when wind generation levels are low. A self-defeating situation is created whereby initial investment in wind power produces returns that can only diminish over time as more and more plant is incorporated into the mix.

Diversity of wind locations, and diversity of the renewable portfolio as a whole, will therefore bring a positive economic benefit to intermittent plant, since it reduces the negative correlation between output and price captured. This in turn could lead to lower levels of subsidy, in the form of the Renewables Obligation (RO), that would otherwise be needed to promote further investment.

Conclusion

The analysis suggests that annual cost savings from a diversified variable renewables mix could be very significant, as much as 3.3% of the annual wholesale cost of electricity, based on the example case where nuclear plant is relatively inflexible and opportunities for export and response on the demand side remain limited. Furthermore, there may be additional savings in transmission costs since peaks in output may be flattened, possibly requiring less transmission capacity and/or reducing the costs of managing transmission congestion.

Some of the benefits of smoothing variability in wind output could be captured through greater interconnection and a more dynamic demand side enabled by smart grids, smart metering and other new technologies, such as electric cars. It is also true to say that marine technologies will initially be more expensive than wind. Nonetheless, the analysis suggests that marine technologies can complement wind, increasing the cost effectiveness of variable renewables, and ultimately expanding the potential share for renewables in the overall generation mix.

There is the additional potential benefit to consumers in terms of reduced costs of the RO, but this may be offset to a greater or lesser extent by increased wholesale electricity prices. Since both the ROC price and wholesale electricity price represent transfers between producers and consumes, rather than real resource costs, these are not included in the tables above.