Friday, September 15, 2017

Baseload from wind and solar

Those who are still wedded to coal (with an occasional foray into nuclear) keep on insisting that we need baseload electricity, and that wind and solar are too "variable" to allow us to rely on them.

This chart, which references the new Kennedy wind plus solar plus storage project by Windlab in Queensland's far north, shows how closely you can produce near baseload output from a combination of these two different renewables.

Source: Windlab Prospectus

The pine green shows measured daily output averaged over one year from wind, the dark blue from solar, and the top of the green shows the combined averaged daily output from wind and solar.  The black line shows averaged daily demand in Queensland.  The times of day when supply from wind plus solar is "above" the black line (they are on different scales) more power is being produced than is needed, while when the top of the pine green line is "below", it's the reverse.  So storage is needed; the batteries charging up overnight and discharging from midday to about 10 p.m., with the maximum "gap" at about 6 p.m. when everybody gets home and turns on the aircon and the kettle, before wind speeds have picked up.

Well, say the baseload promoters, you don't need storage with baseload, because it's always there.  Have another look at the chart.  The red line on the chart represents hypothetical baseload output from, say, a coal-fired power station.   There would be too much electricity produced between 10 p.m. and six a.m., and too little between 1 p.m. and 10 p.m.  You would still need storage. Or you would need peaking gas plants from 1 p.m. to 10 p.m., and you would have to wastefully vent steam from your coal power station to prevent the grid overloading between 10 p.m. and 6 a.m., because you can't quickly dial down (or up) the heat output of a coal power station to meet variations in demand.  Both very expensive.  In fact batteries would be really useful to a grid with 100% baseload electricity generation too.  The only difference between a 100% baseload and a 100% renewable grid is that you you would need roughly twice as much storage capacity.

The three key objectives to consider are cost, reliability and carbon emissions.  A 100% renewable grid with a mixture of wind and solar produces no carbon, and will (with enough storage capacity) be completely reliable.   All good, but will it be cheaper?

That depends on the cost of storage and how much you need.  The new Tesla Powerpack utility-scale battery costs US$109/MWh (back of the envelope calcs, using the costings revealed by Elon Musk for the South Australia battery bank)  Reading off the chart, you'd need storage equal to half of demand for 3 hours and 20% of demand for 5 hours, or a total storage of 2 and a half hours of daily demand.  Let's say 3 hours, to be safe.  Wind and solar are half the cost of coal, even ignoring coal's CO2 emissions and lethal pollution.  Wind is A$55/MWh, solar (at that latitude) A$65/MWh. Assuming capacity factors of 40% for wind and 25% for solar, about 80% of the output will be wind, 20% solar, so roughly A$57 for the electricity output.  Cost of storage in A$ is $145/MWh (US$109 /0.75)  You'll need 3 hours total storage to cover 24 hours demand, so that's $145/24*3, or  A$18/MWh.  Total cost = A$57 + A$18, or $75/MWh⧫.  New coal is A$110/MWh before a carbon tax, and that ignores the storage needs of baseload ($9/MWh?) or the cost of gas peaking power plants, and venting steam at night (X$/MWh??)  So, A$75 vs A$120, or a 40% discount for renewables over coal, with zero carbon emissions and complete reliability.

A combination of wind plus solar with storage can provide baseload, more cheaply and much more cleanly than coal can.  There is no case for baseload coal.  Existing coal power stations will keep going until they wear out, but there will be no need for new ones to be built.





⧫ Incidentally, the same price that the new CSP plant in South Australia will be selling its output to the SA government.  But the CSP plant will provide 8 hours of storage.  CSP still a lot cheaper than batteries.






2 comments:

  1. Time to review your opinion about the meaning of "baseload". It is not now and never was just a rectangular block of power. Every baseload power station in Australia is flexible enough to act as standby, to run part-loaded, to follow loads up and down and to follow pre-determined system instructions to ramp up or down at nominated rates, eg 5MW/minute.

    Baseload capability is exactly what is needed to provide as much power as needed, when it is needed, except for actual peaks - which, by using annualised graphs as above, have been averaged out.

    So, perhaps the best approach is to consider those annualised demands as being close to the "base load" and for peaking plant to be required to cover the very short or even several day long spikes and troughs.

    Wind and solar can go some way towards achieving this, but in order to provide for 24/7/365 power to match the loads, additional hydro, batteries, GT's or other peaking plant (at least) are essential.

    ReplyDelete
    Replies
    1. Just seen your comment, John. Thank you. Good points--I'll take them on board.

      Delete