Busting the baseload power myth

Extract from ABC News

Baseload power is not a fundamental requirement of modern energy production, argues Dr David Mills.

Research by Mills and colleagues shows that a mix of wind and solar technologies could have entirely powered the US in 2006. (Source: iStockphoto)

Related Stories

• Only carbon price makes options viable: report, News Online, 01 Dec 2010
• Solar thermal surge possible by 2050, Science Online, 26 May 2009
• David Mills on the photon economy, Big Idea ABC1 23 October 2009

For years, opponents of renewable energy have argued that traditional energy technologies — coal, gas and nuclear — are essential because they provide continuous baseload power.
Baseload sources of power operate day and night for most of the year, and this efficient use of turbines and generators reduces the cost of energy production.
Many traditional engineers insist that baseload is an absolute requirement for a comprehensive and cheap system, complemented by intermediate peaking and fast peaking plants to meet demand throughout the day.
But baseload output is not a fundamental requirement of modern energy production. It is rather a characteristic of certain fossil, geothermal and nuclear plants that are operated continuously to lower their relative capital expenditure versus fuel cost.
More fundamental to meeting our energy demands is the ability to match inflexible sources of power — those that can only generate energy at certain times such as wind — with flexible sources of power — those that can generate and store energy such as solar.

Dissecting the baseload argument

My colleagues, Weili Cheng and Phillippe Larochelle, and I recently showed that 100 per cent of the 2006 USA electrical load could have been covered on an hour-by-hour basis for the whole year solely from wind and solar energy. No baseload power required.
To understand how this is possible, we need to compare the traditional model of meeting energy demand based on baseload versus a mix of flexible and inflexible sources.

Diagram A (above) shows the traditional baseload model. The blue baseload section is nearly flat except when load drops at night below the baseload power output. The advantage of baseload is that the generator is working flat out most of the time and makes better economic use of its equipment. The middle orange section is called 'intermediate peaking' and it means a system that rises and fall slowly during the day and night to roughly match the rise and fall of electricity grid demand. It uses its equipment for fewer hours than baseload and this has a higher kWh cost.
In the USA, most natural gas combined cycle plants are used for intermediate peaking. If we didn't care about the cost of the gas fossil fuel, we would probably run natural gas combined cycle for both intermediate and baseload sections, because it delivers about half the emissions of coal. If you ran this system with every natural gas combined cycle plant following the load, there would be no baseload in the system but it would work perfectly well.
Diagram B shows our work using load data from 2006 to run models on an hour-by-hour basis. The model shows that demand can be met using a combination of inflexible technologies (in this case wind) topped up with flexible technologies (in this case solar).
The wind is mostly uncontrolled (the only thing you can do is turn the wind generator on or off) and the concentrating solar thermal plants are equipped with low cost thermal (heat) storage like the plants recently installed in Spain, which can follow the grid load.
The solar output in the figure rises and falls to balance the total output with the grid demand and takes the place of both intermediate peaking and fast peaking. The essential load-matching function is performed by the solar thermal storage.
In such a wind/solar scenario, wind does not fit the traditional baseload paradigm. Wind is, in fact, cheaper as an uncontrolled variable source rather than as a baseload source because adding electrical storage like an expensive battery to make the output flat would increase its cost per kWh hugely. But — and this is important — wind, like baseload, is still inflexible because the output cannot be changed short of shutting down the plant; neither wind nor coal nor nuclear can follow the load without changing the technology and increasing the kWh cost.

Matching inflexible and flexible technologies

Following from our our work, we believe that a modern complete system is likely to be composed of inflexible lower cost components, and flexible higher cost components, and the numbers of these must be carefully matched to prevent power blackouts. The components in each of these two buckets — inflexible technologies and flexible technologies — must compete in price with each other, not with elements from the other bucket. They do different jobs.
Inflexible technologies include not only technologies that can or must be run as baseload, such as coal, nuclear, and geothermal, but non-baseload technologies like non-storage photovoltaic, non-storage concentrating solar thermal, and of course, wind.
It is logical that these technologies must compete with each other for the lowest price and best environmental outcome, but wind is already competitive against coal in much of the USA and is very low in emissions. An important outcome is that nuclear, geothermal, coal with sequestration and photovoltaic without storage have to compete against wind per kWh because they provide the same product — inflexible electricity output. This might be a real challenge for these technologies.
Flexible technologies include intermediate peaking natural gas combined cycle plants, gas turbines, concentrating solar with thermal storage, peaking hydro, and photovoltaic with battery storage. Again these must compete with each other on price and environmental factors.
Costs are dropping so quickly that we may be able to very soon construct an inflexible plus flexible combination from solar and wind at much the same levelised cost as current coal plus natural gas combined cycle systems in the USA, and perhaps for Australia as well. Wind is already there.

A new way of thinking

We have seen many cases of solar being compared against coal, but this is clearly not appropriate if the solar has energy storage. Wrong bucket. It is, however, highly appropriate to compare concentrating solar thermal with storage against a combination of natural gas combined cycle and peaking gas turbines, and gas peaking plants are extremely expensive.
This new way of thinking is quite general: it is actually possible to construct a generating system using any inflexible system partnered with any flexible system, such as nuclear with concentrating solar thermal, or wind with gas turbines. However, baseload is no longer a relevant prerequisite for a modern generating system.
Matching of inflexible and flexible low emissions technologies is much more a core requirement, and something that is not reflected in current government policy. With the huge resources of wind and solar available, and the need for low carbon solutions, it is clear what the pillars of the new system must be.
Dr David Mills is known worldwide for pioneering Compact Linear Fresnel Reflector (CLFR) technology and for his work in non-imaging optics, solar thermal energy, and photovoltaic systems over 32 years. He co-founded solar-thermal company Ausra. He is now retired and has no formal affiliation with Ausra, which was bought by French nuclear group Areva earlier this year. This article is based on unpublished research work presented at Solar 2010, the 48^th annual conference of the Australian Solar Energy Society.