4 ms·
The question never was can wind generate enough power - just build more windmills. The question is can it be reliably integrated into a stable electrical system
by irixusr 11y ago
The question never was can wind generate enough power - just build more windmills. The question is can it be reliably integrated into a stable electrical system?
Unfortunately, largely no. At least not still. Without getting into technical details note that in jurisdictions with a legal mandate to prefer windpower (when available) and large installed windbase the price of power goes negative. That is, there are situations when power companies pay to have anyone dump their power. Otherwise they'll damage their equipment.
the problem never was the turbines, but the grid!
- 7952 11y agoI agree the grid issues are a major constraint on new wind turbine. But is there actually evidence that they cannot be reliably integrated in the grid? Is Denmark's grid actually unstable? At least in the UK the grid operators simply refuses connections if their is not enough grid capacity in the locality of a proposed wind farm. It is surely a problem, but one that is well understood. And paying people to dump power is a method of maintaining the grid in unusual conditions; it is just a bit unpleasant. Excess capacity is an unintended consequence of having free fuel!
- irixusr 11y agoBefore I rant; I dunno the specifics about Denmark, but as a maritime nation they have a unique advantage for windpower. Off-shore windpower is far less stochastic than on-shore windpower (see last paragraph for why). So all the problems below are significantly mitigated. In power engineering we're not talking about switches that readers in this forum are more acquainted with (in computer architecture). Currents are in the thousands of amps, not nano, and volts are in the hundreds of thousands (or millions in one Soviet era line). Furthermore the grid is very strict with quality. The frequency must be within a certain tolerance at all times, and no more than a maximum allowed of phase shift in a period of time. The sinusoid must be quite pure, with strict maximums allowed of harmonics. The voltage has to be right. The power has to flow out, not in. Keep in mind that the grid is unforgiving with the python philosophy of asking for forgiveness. A mistake costs millions. Every time you add or remove a load/source the frequency changes, the voltages change, the flows change. To stabilize the grid you have a bulldozer (the throttle) and a little scalpel (capacitor banks). A few jurisdictions are blessed by the engineering gods and have an instrument that works in-between (hydroelectric storage). Now imagine you don't have to slowly integrate a single large and lethargic source in the middle of the night when ppl are sleeping. Instead you have thousands of stochastically generating sources being integrated and removed as the load is changing. The proof are in the prices. Wind power is expensive, but ironically the spot price sometimes drops to negative! Wind power needs a storage revolution. Then most of these problems go away. There is one cavet to all of this, off-shore wind. Since maritime flows are very constant and the relative speed at an interface is zero, offshore wind power if far more reliable and useful the on-shore.
- 7952 11y agoIn this kind of discussion I think most of us would assume that "instability" means a threat to continuity of supply. High prices are hardly evidence of that. I understand that balancing supply from numerous turbines is difficult, but is that a solved problem? Is there evidence of customer supply problems or not?
- irixusr 11y agoAbout the differences w/ UK, some regulators force you to use wind if its available (i.e. Germany. Ontario too, I think)
- ams6110 11y agoWhy not just feather some or all of the windmills when the supply exceeds the demand? Those adjustments should be possible in a matter of minutes. In the USA I have often driven through wind farms where only a fraction of the windmills are actually turning.
- snowwindwaves 11y agoIn Ontario, Alberta and British Columbia, Canada and Washington, USA at least, I am certain that company that builds a wind farm gets a contract where some government-backed entity guarantees they will buy all of the power the wind farm will produce at a pre-determined rate(s). Otherwise nobody would put up the money to build the wind farm. So while they could feather the blades they have no incentive to. In Ontario if they were forced to not generate they would still be paid for the power they weren't allowed to produce ("being constrained off").