Busting myths about renewable energy
Need something to help with those difficult conversations over the holiday period? Mark Diesendorf busts some recurring myths about the viability of renewable energy.
As renewable electricity continues to grow rapidly, the proponents of coal and nuclear power in parliaments, big business and the media are spreading a number of false myths about renewable energy and the electricity supply system, repeating them endlessly until some people accept them without thinking. This article busts the most frequently repeated myths. I hope that readers of Renew will use it to set the record straight whenever the opportunity arises.
Denigration of success stories
Two of the greatest success stories in the implementation of renewable energy (in regions without any conventional hydro-electricity) are Denmark and South Australia. Both are now obtaining about half their annual electricity generation from variable renewables, from wind in Denmark and wind supplemented by solar PV in South Australia. In both cases the principal putdown by renewable energy critics is to say that these regions have the most expensive electricity in Europe/Australia respectively.
This is an example of using a true statement to imply a false conclusion. Denmark has a high electricity price because it has a high tax on electricity. This tax goes into consolidated revenue—it is not a subsidy to Danish wind power. If electricity prices without tax are compared across European countries, then Denmark is in the middle.
South Australia had higher electricity prices than the other states long before renewable generation became significant. The main reason is that it has a large fraction of gas in its electricity generation mix, and gas prices are much higher than coal.
Hiding the price reductions
Renewable energy critics claim incorrectly that renewables push up the price of electricity. This was once true, but nowadays wind and solar farms actually reduce the wholesale price of electricity by means of the merit order effect. Every five minutes, power stations bid a price for generating electricity and the grid operator gives top ranking (known as ‘top of the merit order’) for dispatching power into the grid to the lowest bidder, then dispatches the next lowest bidder, and so on until electricity demand for that period is satisfied. The highest electricity price dispatched is paid to all the successful bidders. Wind and solar have no fuel costs and so they can bid a price of zero—they are top of the merit order for dispatching power. Since they displace some gas and coal, which are more expensive to operate, the renewables bring down the price paid to all bidders. Unfortunately, this price reduction can be swamped by other factors, such as increasing gas prices and gaming of the market by large utilities. Even when the wholesale price reduction from renewables is obvious, it isn’t always passed on to retail customers.
The reduction in wholesale electricity price by the merit order effect offsets (approximately) the small increase in retail electricity price to pay for renewable energy certificates (RECs). RECs are the only remaining subsidy to renewable energy in Australia and are due to expire in 2020, although the federal government has discussed (but recently rejected) bringing that date forward.
The baseload myth
Critics claim incorrectly that our electricity system needs ‘baseload’ power stations. These are power stations that can operate continuously at their rated power (aka generating capacity), except when they break down or undergo planned maintenance. In mainland Australia, most baseload power stations burn coal to heat water in a boiler. South Australia has a single gas-fired baseload power station, Torrens Island, that is no longer able to operate as baseload and is heading for retirement.
As wind and solar farms proliferate, the federal government wants to keep baseload coal-fired power stations operating, either by forcibly extending the lifetimes of clapped-out old coal stations such as Liddell in NSW, or by building an unnecessary new, taxpayer-subsidised coal station. This, the proponents claim falsely, is required for maintaining the reliability of the generating system.
This claim has been contradicted by studies published by five different research groups, include our own at UNSW, who have performed hourly computer simulation modelling of the operation of Australia’s large-scale electricity system with 100% renewable energy. Every hour the models balance actual demand against actual wind and solar data recorded at the same time over the region. They find that the required level of generation reliability can be achieved even though the majority of electricity generation comes from variable renewable sources. Furthermore, baseload thermal power stations and large hydro are unnecessary. However, when the electricity grid has more than 40% to 50% of annual generation coming from variable renewables, some storage is needed.
Storage can be provided in several different ways. Conventional hydro, based on large dams, provides about 7% of annual electricity generated by the National Electricity Market, the interconnected system of the eastern and southern states of Australia. According to the UNSW simulation modelling, existing hydro is sufficient to balance variable renewables that provide up to 70% of annual electricity. Beyond that, additional storage, not necessarily hydro, is needed to ensure the reliability criterion is met.
South Australia has no conventional hydro potential and has only a weak transmission link to a neighbouring state, Victoria. It has installed a large battery, has other batteries under development and may soon have a concentrated solar thermal (CST) power station with thermal storage in tanks of molten salts. In addition, several off-river pumped hydro power plants are being planned by private developers.
Pumped hydro is a well-established technology for storing electricity. During periods of low demand and/or excess supply, water is pumped from a lower to a higher reservoir. During periods of high demand and/or low supply, water is released from the upper reservoir, generating electricity while returning to the lower reservoir. In a dry region, an off-river pumped hydro system can use the ocean or a mineshaft as the lower reservoir and so is very suitable for Australia. Only a small amount of water is lost from the system (by evaporation).
Other commercially available forms of ‘storage’, in which storage is via a fuel, are open-cycle gas turbines (OCGTs), essentially jet engines, and reciprocating engines (e.g. diesel generators). These technologies have relatively low capital costs (in dollars per kilowatt of generating capacity) and high fuel costs (in dollars per kilowatt-hour of energy generated). If they are only operated for infrequent periods of a few hours to a few days, to fill the occasional small shortfalls in wind plus solar, the annual fuel bill is small, and so they can be an inexpensive form of reliability insurance. Initially they may be fuelled with fossil fuels such as natural gas or diesel oil, but they can operate on renewable fuels—biofuels, hydrogen made by using renewable electricity to split water or ammonia made by combining renewable hydro with nitrogen from the air—when these become available commercially.
Power stations that can provide power on demand are known as ‘dispatchable’. Hydro with dams (including pumped hydro), concentrated solar thermal with thermal storage, batteries, open-cycle gas turbines and diesels using renewable fuels are examples of dispatchable renewable power plants. They all incorporate some form of storage.
Thus, in a reliable 100% renewable electricity system, the vast majority of electricity can come from variable renewables such as wind and solar PV, provided their variations are balanced by a minority of dispatchable renewables and/or other forms of storage. Reliability can be further increased by having a broad mix of renewable energy technologies, geographic dispersion of wind and solar farms, a few new transmission links and demand response. Baseload power stations are unnecessary and the quantity of storage required in Australia for 100% renewable electricity is relatively small.
The myth that all power stations must be ‘dispatchable’
Some supporters of coal power confuse ‘dispatchable’ with ‘baseload’ power stations. However, the two concepts have different technical definitions, given above. Some baseload power stations are not dispatchable. i.e. cannot always supply power when required. This can be due to age or weather, e.g. several old Australian coal stations and French nuclear stations have to be shut down during heatwaves. Conversely, some dispatchable power stations, such as the renewable ones discussed in the previous section, are not baseload. Let’s examine the concept of dispatchability more closely.
In a large-scale electricity system, supply and demand must be matched continuously. If they get out of balance, the voltage and frequency of alternative current change from the required values and the system can temporarily collapse, so blackouts occur. Dispatchable power stations allow the generating system to respond quickly to sudden changes in supply and demand that can upset the balance. Such changes include the breakdown of a large baseload power station, the failure of a transmission line and the sudden increase in demand during a commercial break in a popular TV program when millions turn on their kettles.
The key property required of a dispatchable power station is flexibility in operation, including rapidity of response; therefore, it must be controllable. Flexibility entails that a power station can be started up rapidly from ‘cold’ and, when operating, its output can be rapidly increased or decreased to a large degree. An optional extra property is the ability to supply power for a long enough period of time to fix a system problem and allow adjustment of the supply or demand to maintain the balance between them.
In a conventional power system based on fossil fuels, most power stations are theoretically dispatchable. However, despite the rhetoric, baseload power stations generally have very limited flexibility in operation—they can take up to a whole day to go from cold to full power and, even when hot, cannot rapidly and economically vary their outputs to meet the peaks in demand. Therefore, they have to be ranked low in terms of dispatchability. The much more flexible peakload power stations (hydro and open-cycle gas turbines) have traditionally been used to supply the daily peaks in demand and to provide rapid response to sudden changes in supply and demand. Also, one or more baseload power stations are kept hot and ready to be connected to the grid to cover for the breakdown of another baseload power station that may need weeks or months for repair.
When the grid is dominated by variable renewables, the conventional model breaks down. The failure of a single wind turbine in a wind farm or a single solar module in a solar farm does not affect the farm output significantly. Shortfalls in wind and solar PV generation are weather-related and most last only a few hours during peaks in demand or overnight. A minority of shortfalls may last through a bad weather period of four to seven days; longer gaps would be very rare. So the emphasis must be on fast, flexible, dispatchable power plants as partners for variable renewables.
Batteries are the fastest, responding in a fraction of a second. However, if they have to provide energy storage for more than several hours, their capital cost is very high. Fortunately, other dispatchable renewable power stations can take over when batteries are emptied. Pumped hydro can respond within several seconds and operate from several hours to months, depending on the size of its storage. For environmental and other reasons, most off-river pumped hydro systems would have small upper reservoirs that can store energy for days to weeks. Concentrated solar thermal can respond within several seconds and run for several hours to overnight, provided its thermal storage has been charged up on a sunny day. Open-cycle gas turbines and diesel plants, when cold, may take up to about 10 minutes to warm up, can respond within several seconds when already hot, and can operate economically for up to several days before fuel costs become painful.
Demand response, once organised, can decrease or increase demand within seconds. However, most forms of demand response—such as switching off a fridge, air conditioner or aluminium smelter—can only last for up to an hour.
Variable renewables such as wind and PV, can reduce their power within seconds—they are ‘dispatchable downwards’.
To sum up, a large-scale renewable electricity system where the bulk of energy is provided by variable renewables and has limited or no conventional hydro based on large dams will have a mix of several types of dispatchable renewables and other forms of storage with different response times and operating periods, e.g. the mix under development in South Australia.
Is coal power dispatchable?
In Germany some fairly new coal-fired power stations are being modified to become more flexible, so that they can help balance the variations in wind and PV. But this is expensive and they will never be able to match the flexibility and response speeds of batteries, dispatchable renewables or peakload power stations generally. Modification of Australia’s old coal stations is not economically viable and a new coal station will never be built without huge subsidies from taxpayers. As renewable energy replaces fossil fuels, there may be a case for keeping one of the least ancient coal stations in good repair as potential long-term backup for a rare long-term weather event or, more likely, failure of a major transmission line. But coal power in Australia is too inflexible and too slow to respond to be called ‘dispatchable’.