Australasian Science Magazine

Credit: Kletr/Adobe

By Andrew Blakers

Despite the rapid uptake of solar and wind energy worldwide, fossil fuels are still required when the wind doesn’t blow and the sun doesn’t shine. However, a cheap and proven storage option, in combination with wind and solar energy, could replace the need for fossil fuels within 15 years.

A renewable energy revolution is in progress, driven by rapidly decreasing prices. Together, new photovoltaic (PV) and wind electricity generation capacity is being installed at a greater rate worldwide than the combined amount of new coal, gas, oil and nuclear (Fig. 1). Within a few years, new PV and wind generation capacity installed each year worldwide may each be larger than the rest of the electricity generation industry combined.

PV and wind constitute nearly all new generation capacity in Australia. Tasmania obtains all of its electricity from hydro and wind. South Australia obtains 40% of its electricity from wind and PV, and this figure will be above 50% by 2020. The ACT will source all of its electricity from wind and PV by 2025.

Figure 2 shows the retirement schedule for Australia’s fleet of coal-fired power stations, assuming a 50-year service life. PV and wind can be installed at a sufficient rate to yield a largely renewable Australian electricity system by 2030, smoothly replacing existing coal and gas generators as they reach the end of their working lives.

Solar and wind resources are much larger than all other available energy resources combined. PV and wind systems utilise very common materials, have minimal security and military risks, are available nearly everywhere in vast quantities, and have minimal environmental impact over unlimited time scales.

However, the sun and wind are variable energy sources. When PV and wind constitute more than about half of electricity generation, storage is needed to provide a reliable electricity supply.

Pumped Hydroelectric Energy Storage

Pumped hydroelectric energy storage (PHES) involves pumping water to an upper reservoir when energy is plentiful and cheap. During periods of peak demand, when energy is expensive, water is released through a turbine to recover electrical energy. About 80% of the electricity used to pump the water uphill is recovered, and 20% is lost.

PHES constitutes 99% of energy storage worldwide (>160 GW) because it is much cheaper than alternatives such as batteries. In Australia, existing PHES systems include the Tumut 3, Wivenhoe and Shoalhaven power stations.

PHES is an ideal method of storage: response time (from off to fully-on) is less than a minute, the operational lifetime is 50–100 years, operational costs are low, losses are modest, and PHES provides “spinning reserve” and “black start” capability to help with grid management.

Short-Term Storage

A key point in relation to energy storage in a grid dominated by PV and wind generation is that a few hours of energy storage is usually sufficient. Short-term storage covers high-demand events such as hot summer afternoons and cold winter mornings and evenings, covers peak usage at night, offsets periods of low supply such as wind lulls and cloud, offsets plant and transmission line failures, and covers the time required to bring a low duty cycle biomass-fired power station online or implement demand management if the supply shortfall is likely to be extended. Short-term storage also improves the load factor of constrained power lines, overcoming the need to duplicate powerlines that connect the national grid to wind and solar farms in rural regions.

Currently, traditional hydroelectric and low-duty cycle gas power stations balance any discrepancies between energy demand and supply. As PV and wind energy capacity increases, PHES will increasingly take on this role, supplemented by the use of low duty cycle biomass and demand management to meet occasional shortfalls.

Away from Rivers

Conventional hydroelectric systems are located in river valleys, requiring lake systems spanning thousands of hectares along with expensive and extensive flood control measures to cope with once-in-1000-year floods. As many of the existing hydroelectric power stations are in mountainous national parks, new river-based PHES proposals in these areas would draw strong public resistance to both reservoir construction and extended powerline easements. However, PHES located away from rivers offers low-cost short-term storage, avoids community conflict, reduces transmission costs, and has an unlimited number of potential sites.

Off-river PHES takes advantage of the vastly larger area of land available beyond river systems, with many good sites existing near roads and transmission powerlines. Off-river PHES comprises pairs of small, hectare-scale “turkey nest”reservoirs whose walls are made from spoil scooped from the centre. The two reservoirs are connected by pipes (or tunnels) that incorporate a pump or turbine. There is no river and no need for expensive flood control.

Essentially, the reservoirs are “oversized farm dams”. They can be located in hilly farming country near roads and power transmission networks. In contrast to standard hydroelectric systems, in an off-river PHES system the same water circulates indefinitely between the upper and lower reservoirs, thus eliminating the need for a river.

Importantly, the upper reservoir in an off-river PHES can be on top of a hill rather than in a river valley, allowing a much larger “head” (height difference) between the reservoirs. This is an advantage since both energy storage capacity and power capacity are proportional to the head, so the energy storage cost scales inversely with the head.

Figure 3 shows a Google Earth synthetic image of the Tumut 3 pumped hydroelectric system located in a deep river valley. Extensive flood control measures have been constructed. The upper reservoir (Talbingo) has an area of almost 2000 hectares, and the head between it and the lower reservoir (Jounama) is 151 metres. Water can be pumped to the upper reservoir for energy storage and returned to the lower reservoir for power generation at a later time.

Now consider a hypothetical off-river PHES system using a pipe running along a powerline easement indicated by the red line. The head is 700 metres, nearly five times larger than for Tumut 3, and there is no need for flood control. Such a system could operate at the same power rating as Tumut 3 (1500 MW) for 4 hours utilising twin 20-hectare reservoirs – each just 1% of the area of Talbingo. Alternatively, it could operate for 24 hours at 250 MW output.

Where Could Off-River PHES Be Located?

There are thousands of potential off-river PHES sites outside national park sites along the Great Dividing Range, and also in many other hilly regions. Google Earth is an interesting way you can find off-river sites in your local area (Fig. 4). You should look for sites outside national parks with a head of 200 metres or more (300–800 metres would be better), with steep and short pipes (preferably with a slope of more than 150 metres/km), and within 20 km of a high voltage powerline. Wind farms are often located in hilly farmland close to suitable sites.

Remarkably, many studies of storage opportunities overlook PHES in the mistaken belief that there are few remaining development opportunities. This is due to mistakenly overlooking the very large areas available in off-river sites. Recent comments have included “Australia is flat and arid and has little potential for more hydro”, “Pumped hydroelectric storage is at a mature stage of development, but there are limited locations where these facilities can be built” and “further deployment of pumped hydro is severely limited by geographical and environmental site requirements as well as project size requirements to achieve economies of scale”.

Old mining sites can also be suitable for PHES. For example, Genex Power has proposed a 330 MW PHES system at an old gold mine at Kidston in northern Queensland (Fig. 5). The difference in water level between adjacent pits is about 200 metres.

Indicative Cost

Reductions in PV costs have seen the installation of 1.5 million rooftop systems, and much attention has been paid to the need for batteries to store the excess energy that is returned to the grid. Since PV and PV/battery systems operate “behind the meter”in competition with relatively high retail tariffs, even expensive residential battery units have a large potential market.

However, high penetration of wind and solar energy will be facilitated by low-cost mass storage in the wholesale market, in conjunction with load management. This is where off-river PHES has a key role to play.

Most of the costs of off-river and conventional (on-river) PHES are similar. The main difference is that off-river PHES uses relatively tiny and inexpensive reservoirs that have a much larger head and do not require expensive flood control measures. The costs of off-river PHES systems are relatively predictable because each off-river PHES site looks much like another, whereas river valleys vary greatly. The power costs – pipe, pump, turbine, generator, transformer, control and transmission – comprise most of the expense. The energy cost – the reservoirs – is relatively small.

The indicative costs of off-river PHES are $1 million per MW, translating to $20–30 per MWh of storage. This is much cheaper than batteries (around $200 per MWh).

Environmental and Water Issues

The environmental impact of off-river PHES systems is minorbecause the reservoirs are small and located outside national parks and sensitive areas. The area of reservoirs required for a 100% renewable energy Australian electricity grid is a few thousand hectares distributed across the country. For comparison, the area of Lake Eucumbene in the Snowy Mountains is 14,500 hectares.

Sourcing of water is a significant but not major consideration in site selection. The ongoing water requirement is equal to evaporation minus rainfall. Calculations show that this is a tiny fraction of the water annually traded in Australia, even if off-river PHES is used extensively, and is a tiny fraction of what is required to operate the cooling towers of a coal-fired power station with comparable power output.

Conclusion

Pumped hydroelectric energy storage constitutes 99% of all energy storage around the world because it is cheap compared with alternatives such as batteries. It is likely to continue its dominance in the wholesale storage market. Off-river PHES can facilitate very high (50–90%) penetration of wind and PV at modest cost through the provision of low-cost, short-term mass-storage of energy.

Andrew Blakers is Professor of Engineering at the Australian National University, specialising in photovoltaics and renewable energy integration.