Short answer = it could be done but there are limitations which need proper engineering.
The above answers the question. Following is just a layman's terms explanation for those who want to know:
Looking at existing HVDC (High Voltage Direct Current) transmission, there are already links up to ~2400km operating successfully in China and some of those transfer serious amounts of power eg 8GW. To put that into perspective, 8GW is not far short of the all time record peak demand for electricity in Victoria (10.4GW) and is about twice the generating capacity of the present Snowy scheme.
For an AC power system, the longest one in the world is rather close to home - that being the one which stretches from north Queensland to just west of Ceduna (SA) and supplies practically all electricity across Qld, NSW, ACT, Vic and SA. Tasmania's system is separate from an AC power perspective but is connected via a HVDC link.
If we're going to be building things with any significant portion under water then HVDC is the practical means unless we're talking about short distances. For anything that's completely on land well you can build big AC power systems although that does then require that you operate it as one big AC power system. In contrast, if the only connection is DC then you've got two separate AC systems in practice. Depending on circumstances, either wins.
For the record, Qld - NSW are presently linked by one DC and two AC circuits and the same between Vic and SA. For other states NSW and Vic are linked at AC only and Vic and Tas are linked at DC only.
There are however a number of complexities in all of this which means it's all a case of bespoke design. It's not a "plug and play" situation like consumer devices are these days. Rather, it's all down to "real" engineering and the big issues relate to the existing system at both ends.
Trying to keep this to layman's terms, if you want to transfer what most people think of as "power", that is true power, then that's not particularly difficult. Once you want to transfer other technical characteristics, there's more to electricity than simply the true power aspect, well then having a long transmission line in the middle is problematic since it changes (AC) or completely nullifies (DC) those attributes. You're then reliant upon recreating them at the other end in order for it to work - that's doable so long as you've first worked out exactly what and how much you need to be doing which will be a dynamic thing it won't be constant.
As a practical example of that, there's a need to run synchronous generating plant (in practice gas but if it existed then coal, hydro, nuclear etc could also do it) in SA at all times even if the true power it produces isn't actually needed. That's for reasons of inertia (also known as "system strength"), reactive power (that's getting into the technical stuff.....) and so on and comes down to the limitations of non-synchronous plant (wind and solar) in that regard plus the limitations of long distance transmission at transferring those things from Victoria.
There are workarounds to that which are being built but point is you first have to know you need them (that point was missed in SA until the state ended up in dark.......) then you need to build them and then operate as required. In the meantime, running some gas-fired generation at all times is the workaround not because it's gas as such but because it's already there and can do what's required in purely electrical terms.
The other issue is a practical and economic one about scale. Scale it up and the unit cost comes down since it doesn't cost twice as much to build twice the capacity on an otherwise identical project. Trouble is, the bigger it is the more difficult it is to deal with failure when the inevitable happens. And once you get a "too big" fault on a system, well then in the best case you're going to black out a lot of customers and in the worst case the entire show comes to a halt real quick. One moment everything's normal, a few seconds later and an entire city, state or country is in the dark.
As background for that, if we look at Australia well we run the frequency at 50Hz and for other countries it's either 50 or 60Hz. In layman's terms frequency = speed.
Now think of what happens if you put a sudden load on an engine. Push the mower into long grass for example. The engine slows down and it slows down rather a lot yes.
Now back to the power grid, well if that 50Hz drops by 5% that's about the point where it's all over. At 48Hz the system should hold up but not without shedding quite a bit of load, at 47Hz I sure wouldn't bet on it staying energized at all, at 45Hz no we'll be in the dark before we get to that point.
The infamous big blackout in SA was an event of that nature. Loss of some supply > more supply tripped off as frequency and voltage fell > rest slowed down even further > all over and the only option is restart everything from scratch and all in a matter of seconds. That's the nightmare scenario with power systems but for clarity, SA certainly isn't the only place where it has ever occurred.
Now for a practical example with a link between A and B consider that the load on the SWIS (South West Interconnected System) in WA is running at about 2800 MW presently and should get down to about 2000 MW overnight. Meanwhile the total load in the eastern states (the NEM including SA and Tas) is currently about 21,000 MW (that's fairly low given the time of day).
Now suppose that we wanted to link the SWIS with the NEM (National Electricity Market). It could be done but a major consideration is what happens when the inevitable trip of that line occurs at some random unknown time?
What happens to the WA system if it's 5am and half the supply is coming from interstate and that suddenly disappeared? As "seen" by the power stations in WA, that's akin to pushing the mower straight into long grass! Frequency would drop and either generation takes up that lost supply real quick, or some loads are cut immediately, or south-west WA (including Perth) will be in the dark.
Or the opposite direction. What if there's say 3000 MW of load in WA and 1000 MW being sent from the SWIS to the NEM and the line trips? Loss of that load on the grid in WA means there's now too much supply and frequency will be rapidly rising as a result. Can generation cut output quickly enough? If not then we'll see it outright trip due to frequency rising out of spec and worst case that results in an over-reaction which ends up tripping the whole lot.
Now there are solutions to that but it's the sort of thing that needs to be properly engineered and looking at both ends of the line. Whichever is the smaller one (electrically) will pose the biggest issues but that's not to say there won't be any problems on the other end too. It needs a proper examination of the existing power system and crunching the numbers on what works and what doesn't. That's certainly something which can be done but it's real, proper sort of engineering not something that's done in an afternoon.
For an Australian project currently being looked at, that is Marinus Link (aka Project Marinus - building additional HVDC lines between Vic and Tas), there's an obvious financial incentive to go as big as possible. The limits are being pushed there but to cut a long story short it seems that around 750 MW is going to be about it give or take a bit. That doesn't mean 3000 MW can't be built, just that it'll need to be done as 4 x 750 MW in order to keep the maximum size of any failure within manageable limits.
An exception to all this is if one end of the link exists only to supply the link itself. That is, the "grid" at one end isn't really an actual power grid but is simply a power station of whatever sort (eg a solar farm plus some batteries, wind, pumped hydro etc but point is it's only there to supply the link, it's not also supplying a local city etc). In that case well there's no real need to have that side remain energized (electrically live) if the link trips, and you're never going to send power in the reverse direction since there's nothing to use it, so you've only got to worry about the receiving end. Eg for a link from Australia to Singapore, the main concern is about what happens in Singapore if it trips, there's no need to ensure the sending (Australian) end remains live following a trip so long as nothing's physically damaged by that.
That aspect of only needing to worry about one direction of power transfer and only needing to worry about maintaining stability at the receiving end does obviously reduce the amount of engineering work required. It also means that things which wouldn't be acceptable for two way flow will be acceptable in practice given that the concern is really only about one end.
So how likely really is this concern about lines tripping?
Short answer is "it happens". It's 3 weeks to Christmas and odds are something, either transmission or generation, somewhere in the NEM will unexpectedly trip between now and Christmas Day. I can't tell you what or when, that's anyone's guess, but it happens most certainly.
For a recent example generating unit number 1 at Torrens Island B (largest power station in SA) tripped on Monday afternoon and a couple of weeks ago there was a trip of both AC transmission lines between SA and Vic.
That neither of those incidents became newsworthy was because everything worked pretty much as it should work. Eg with the trip at Torrens Island other sources of generation in the NEM immediately took up the load, then units 1 - 3 at Dry Creek and unit 5 at Quarantine power stations (which were previously idle) in SA were started up to restore the system to a secure state (so that's back to being ready just in case anything else goes wrong), then in due course the more efficient but much slower to start Pelican Point station ramped up and Dry Creek + Quarantine were shut down as no longer needed given low demand at the time.
That's how it's supposed to work. Unless you're in the industry or happened to be near Dry Creek power station, which is a peaking and backup plant that's normally idle, when all 3 generators suddenly roared into action then you'd be unaware that anything even happened. For ordinary consumers be they homes or business, life carried on business as usual.
A bad but not disastrous response to the same incidents would be a major deviation in frequency, some consumers blacked out but the system remains alive as such.
A disastrous response to the same would be frequency falls too far, generation trips and down goes the whole show thus requiring a complete restart from scratch. Avoiding that is the first and foremost focus of it all really.
As a concept I'm enthusiastic.
I'm not aware of any proper calculations for losses on that project (it might have been done, just not something I'm following) but as an order of magnitude we're talking about 15% not 50%. Don't take that 15% literally, it's an educated guess not based on any actual design, but point is yes there will be losses but most of the power gets from A to B.
An issue there is that HVDC losses aren't linear. There's a loss at both ends, so a line twice as long won't have double the losses, and also losses go up as a % of total power transmitted as power throughput increases. Run the same line at higher capacity and loss goes up in % terms. End result is there's no single figure for losses.
There's also more than one approach to the design and construction itself which will again affect losses. Bearing in mind that the electrical current always needs a return path to form a circuit, there's more than one way of doing that when it comes to cable under the ocean.
One is to have two conductors to form the circuit. The other is to have one conductor only and use the ocean as the second one.
The latter is far cheaper but does create concerns relating to corrosion since the surest way to outright ruin anything metallic is to have stray DC currents going into and out of it in sea water. Do that and anything metal will be full of holes in no time.
The ocean return approach was originally going to be used with Basslink (Tas - Vic) but in short everyone lost their nerve as to exactly where current might end up flowing. Through the water yes but through what else? Oil rigs and gas pipelines for example and that could end very, very badly for obvious reasons. With that in mind the idea was dropped like a hot potato but at very considerable expense financially and the cable was built with a metallic return instead.
It has been done elsewhere overseas though using just a single pole and the ocean as the return.
So the detail of the design will affect losses and costs and without knowing that I'm really just putting 15% forward as a ballpark sort of number.
As for whether or not that matters, at that point it's purely business like any other situation. If 15% of the stock from your shop routinely gets stolen or 15% of your crops aren't of suitable quality for sale well then it really just comes down to whether or not it's still a viable business based on revenue from the other 85% or not? In the case of energy, it's just business at that point - does the project stack up financially or not?
A point of relevance there is that about 95% of Singapore's electricity is presently produced from gas and the cost of the gas itself, as distinct from the cost of machinery etc, is the most important driver of the cost of that operation.
As such, the profitability of generating renewable energy in Australia and sending it to Singapore is primarily a function of the cost of building the overall scheme (everything - cable, solar farm, etc) and the price of natural gas which is variable subject to market forces. Presumably various contracts would underpin an actual development, but ultimately someone is going to be making a profit / loss on it to the extent that the gas price moves up or down.
A related factor there is how will investors (whoever) perceive the whole thing? There's no real precedent for locating a power station in another country over 3000km away from the load to be supplied with virtually all that transmission being under water. There are certainly power stations built on the other side of a border, eg Mexico and the USA or plenty of examples in Europe, but that's a comparatively simple operation compared to having it thousands of km away separated by water.
It will happen.
Short answer = it could be done but there are limitations which need proper engineering.
The above answers the question. Following is just a layman's terms explanation for those who want to know:
Looking at existing HVDC (High Voltage Direct Current) transmission, there are already links up to ~2400km operating successfully in China and some of those transfer serious amounts of power eg 8GW. To put that into perspective, 8GW is not far short of the all time record peak demand for electricity in Victoria (10.4GW) and is about twice the generating capacity of the present Snowy scheme.
For an AC power system, the longest one in the world is rather close to home - that being the one which stretches from north Queensland to just west of Ceduna (SA) and supplies practically all electricity across Qld, NSW, ACT, Vic and SA. Tasmania's system is separate from an AC power perspective but is connected via a HVDC link.
If we're going to be building things with any significant portion under water then HVDC is the practical means unless we're talking about short distances. For anything that's completely on land well you can build big AC power systems although that does then require that you operate it as one big AC power system. In contrast, if the only connection is DC then you've got two separate AC systems in practice. Depending on circumstances, either wins.
For the record, Qld - NSW are presently linked by one DC and two AC circuits and the same between Vic and SA. For other states NSW and Vic are linked at AC only and Vic and Tas are linked at DC only.
There are however a number of complexities in all of this which means it's all a case of bespoke design. It's not a "plug and play" situation like consumer devices are these days. Rather, it's all down to "real" engineering and the big issues relate to the existing system at both ends.
Trying to keep this to layman's terms, if you want to transfer what most people think of as "power", that is true power, then that's not particularly difficult. Once you want to transfer other technical characteristics, there's more to electricity than simply the true power aspect, well then having a long transmission line in the middle is problematic since it changes (AC) or completely nullifies (DC) those attributes. You're then reliant upon recreating them at the other end in order for it to work - that's doable so long as you've first worked out exactly what and how much you need to be doing which will be a dynamic thing it won't be constant.
As a practical example of that, there's a need to run synchronous generating plant (in practice gas but if it existed then coal, hydro, nuclear etc could also do it) in SA at all times even if the true power it produces isn't actually needed. That's for reasons of inertia (also known as "system strength"), reactive power (that's getting into the technical stuff.....) and so on and comes down to the limitations of non-synchronous plant (wind and solar) in that regard plus the limitations of long distance transmission at transferring those things from Victoria.
There are workarounds to that which are being built but point is you first have to know you need them (that point was missed in SA until the state ended up in dark.......) then you need to build them and then operate as required. In the meantime, running some gas-fired generation at all times is the workaround not because it's gas as such but because it's already there and can do what's required in purely electrical terms.
The other issue is a practical and economic one about scale. Scale it up and the unit cost comes down since it doesn't cost twice as much to build twice the capacity on an otherwise identical project. Trouble is, the bigger it is the more difficult it is to deal with failure when the inevitable happens. And once you get a "too big" fault on a system, well then in the best case you're going to black out a lot of customers and in the worst case the entire show comes to a halt real quick. One moment everything's normal, a few seconds later and an entire city, state or country is in the dark.
As background for that, if we look at Australia well we run the frequency at 50Hz and for other countries it's either 50 or 60Hz. In layman's terms frequency = speed.
Now think of what happens if you put a sudden load on an engine. Push the mower into long grass for example. The engine slows down and it slows down rather a lot yes.
Now back to the power grid, well if that 50Hz drops by 5% that's about the point where it's all over. At 48Hz the system should hold up but not without shedding quite a bit of load, at 47Hz I sure wouldn't bet on it staying energized at all, at 45Hz no we'll be in the dark before we get to that point.
The infamous big blackout in SA was an event of that nature. Loss of some supply > more supply tripped off as frequency and voltage fell > rest slowed down even further > all over and the only option is restart everything from scratch and all in a matter of seconds. That's the nightmare scenario with power systems but for clarity, SA certainly isn't the only place where it has ever occurred.
Now for a practical example with a link between A and B consider that the load on the SWIS (South West Interconnected System) in WA is running at about 2800 MW presently and should get down to about 2000 MW overnight. Meanwhile the total load in the eastern states (the NEM including SA and Tas) is currently about 21,000 MW (that's fairly low given the time of day).
Now suppose that we wanted to link the SWIS with the NEM (National Electricity Market). It could be done but a major consideration is what happens when the inevitable trip of that line occurs at some random unknown time?
What happens to the WA system if it's 5am and half the supply is coming from interstate and that suddenly disappeared? As "seen" by the power stations in WA, that's akin to pushing the mower straight into long grass! Frequency would drop and either generation takes up that lost supply real quick, or some loads are cut immediately, or south-west WA (including Perth) will be in the dark.
Or the opposite direction. What if there's say 3000 MW of load in WA and 1000 MW being sent from the SWIS to the NEM and the line trips? Loss of that load on the grid in WA means there's now too much supply and frequency will be rapidly rising as a result. Can generation cut output quickly enough? If not then we'll see it outright trip due to frequency rising out of spec and worst case that results in an over-reaction which ends up tripping the whole lot.
Now there are solutions to that but it's the sort of thing that needs to be properly engineered and looking at both ends of the line. Whichever is the smaller one (electrically) will pose the biggest issues but that's not to say there won't be any problems on the other end too. It needs a proper examination of the existing power system and crunching the numbers on what works and what doesn't. That's certainly something which can be done but it's real, proper sort of engineering not something that's done in an afternoon.
For an Australian project currently being looked at, that is Marinus Link (aka Project Marinus - building additional HVDC lines between Vic and Tas), there's an obvious financial incentive to go as big as possible. The limits are being pushed there but to cut a long story short it seems that around 750 MW is going to be about it give or take a bit. That doesn't mean 3000 MW can't be built, just that it'll need to be done as 4 x 750 MW in order to keep the maximum size of any failure within manageable limits.
An exception to all this is if one end of the link exists only to supply the link itself. That is, the "grid" at one end isn't really an actual power grid but is simply a power station of whatever sort (eg a solar farm plus some batteries, wind, pumped hydro etc but point is it's only there to supply the link, it's not also supplying a local city etc). In that case well there's no real need to have that side remain energized (electrically live) if the link trips, and you're never going to send power in the reverse direction since there's nothing to use it, so you've only got to worry about the receiving end. Eg for a link from Australia to Singapore, the main concern is about what happens in Singapore if it trips, there's no need to ensure the sending (Australian) end remains live following a trip so long as nothing's physically damaged by that.
That aspect of only needing to worry about one direction of power transfer and only needing to worry about maintaining stability at the receiving end does obviously reduce the amount of engineering work required. It also means that things which wouldn't be acceptable for two way flow will be acceptable in practice given that the concern is really only about one end.
So how likely really is this concern about lines tripping?
Short answer is "it happens". It's 3 weeks to Christmas and odds are something, either transmission or generation, somewhere in the NEM will unexpectedly trip between now and Christmas Day. I can't tell you what or when, that's anyone's guess, but it happens most certainly.
For a recent example generating unit number 1 at Torrens Island B (largest power station in SA) tripped on Monday afternoon and a couple of weeks ago there was a trip of both AC transmission lines between SA and Vic.
That neither of those incidents became newsworthy was because everything worked pretty much as it should work. Eg with the trip at Torrens Island other sources of generation in the NEM immediately took up the load, then units 1 - 3 at Dry Creek and unit 5 at Quarantine power stations (which were previously idle) in SA were started up to restore the system to a secure state (so that's back to being ready just in case anything else goes wrong), then in due course the more efficient but much slower to start Pelican Point station ramped up and Dry Creek + Quarantine were shut down as no longer needed given low demand at the time.
That's how it's supposed to work. Unless you're in the industry or happened to be near Dry Creek power station, which is a peaking and backup plant that's normally idle, when all 3 generators suddenly roared into action then you'd be unaware that anything even happened. For ordinary consumers be they homes or business, life carried on business as usual.
A bad but not disastrous response to the same incidents would be a major deviation in frequency, some consumers blacked out but the system remains alive as such.
A disastrous response to the same would be frequency falls too far, generation trips and down goes the whole show thus requiring a complete restart from scratch. Avoiding that is the first and foremost focus of it all really.
As a concept I'm enthusiastic.
In practice I'm cautious in view of the troubled history of making what should be easy happen in practice. Think "politics" there rather than "engineering" or "finance".
I say that based on observation of actual occurrences rather than cynicism. We're having a lot of trouble changing some timers which are already installed, all that needs to happen is change the settings, so needless to say I'm cautious about anything even slightly more ambitious.
That's not wanting to sound negative, it's just having seen the degree of politics standing in the way of all this.
Hello and welcome to Aussie Stock Forums!
To gain full access you must register. Registration is free and takes only a few seconds to complete.
Already a member? Log in here.
