China: Boldly Going Where No Grid Has Gone Before

[anorlunda]

Here is an interesting item I would like to share. It seems that China has drastically pushed the boundaries of power transmission grid operations. They boldly go where no man has gone before.

How and why?

• The combination of geography and demographics leads them to place large generation resources very far away from the loads where consumers live.
• Excessively long transmission lines make HVDC attractive as a competition for AC.
• Excessive number of HVDC lines brings new operational challenges and opportunities, which they seem to be managing just fine.

They are doing things that old timers like me never imagined. Specifically, they challenged the catechism that increased interconnectivity of the AC grid is more reliable; a principle we used for more than 100 years. Hats off to them.

[/PLAIN]
An abundance of high-voltage DC makes big AC grids unstable

No country has gone as far with HVDC as China has. It operates more than 20 HVDC lines that deliver hydro, coal, and wind power from the nation’s interior to its eastern megacities. In southern China, five HVDC lines carry about 26 gigawatts of hydropower from mountainous Yunnan province to the coastal factories of Guangdong, meeting more than one-quarter of that province’s electricity demand.

It was this concentration of HVDC transmission that prompted the regional grid operator, Guangzhou-based http://eng.csg.cn/home/ Co. (CSG), to take an unprecedented step: breaking up its AC grid.

Before last July, Yunnan and Guangdong were the western and eastern flanks of a CSG grid that served 248 million people living and working in a million-square-kilometer area. It was one large AC zone, augmented by HVDC lines, and it worked. System reliability was consistently improving, according to Mo Weike, a CSG control center engineer pursuing a Ph.D. at South China University of Technology, in Guangzhou. However, says Mo, the hybrid AC-DC system harbored a “unique risk” of systemwide blackouts.

Illustration: Erik Vrielink
When HVDC Attacks: This entire area was once one synchronous AC grid, but the concentration of HVDC lines to the east meant it was safer to separate Guangdong.
Essentially, the HVDC lines converging on Guangdong were too big for the AC grid. When an HVDC line from Yunnan tripped off-line, up to 6.4 GW of power instantly surged onto the underlying mesh of AC lines. To counter this, CSG used preprogrammed security schemes to quickly reduce output from Yunnan.

But if those countermeasures had failed, the AC lines could overload and disrupt the electronic power switching in other HVDC converters. The latter threatened to knock more transmission off-line and collapse the entire CSG grid.

In July, CSG neutralized this threat by shutting off Yunnan’s AC links to the rest of its grid, turning the province into its own distinct synchronous zone. Power exchanges continue between Yunnan and the rest of the CSG grid (in fact, they have increased since July) via the HVDC lines and newly built back-to-back HVDC links on Yunnan’s eastern border.

CSG’s breakup marks the first reversal in AC’s inexorable expansion trend in over 40 years, according to international power experts.“I haven’t heard of anyone splitting grids in that way,” says Ian Dobson, an expert in grid stability at Iowa State University. But such compartmentalization could become a trend, according to Dobson and others. An optimization study that Dobson coauthored in 2014 suggested that, for the biggest grids, AC connectivity is a net liability in terms of reliability.

 

[Bystander]

make HVDC

How "High?"

 

[anorlunda]

How "High?"

There is no hard rule, but 100 kV up to 1100 Kv.

 

[houlahound]

Fascinating, can someone in the field demonstrate with simple calculations how the transmission losses compare to HVAC.

 

[dlgoff]

Been "my experience", having worked with a lot of Chinese people, ... they're a smart people. My daughter is visiting with her Chinese friend for the Thanksgiving holiday. I'll pass this along him. Thanks @anorlunda.

 

[anorlunda]

Fascinating, can someone in the field demonstrate with simple calculations how the transmission losses compare to HVAC.

It is far simpler than you imagine. For a given conductor, there is a resistance R per foot of length. Power losses are $P_{loss}=I^2R$. Power transmitted is $P_{tran}=VI$. Those equations are the same for AC and DC as long as we use RMS measurements for the AC. So the first answer to your question is that power losses for the same power transmitted are identically equal for AC and DC.

If you want to reduce losses for the same size conductors, just use higher voltage. If you double the voltage (and halve the current for the same power) losses are reduced by a factor of 4. Ten times more voltage provides a 99% reduction in losses. But there are limits to how high we can make the voltage. In AC transmission, the combination of high voltages and long distances results in unmanageable high capacitance to ground.

See the PF Insights article, AC Power Analysis: Part 2, Network Analysis. Below is an equivalent circuit for an AC power line from the article. The R, L, and C parameters in the circuit are all per unit length of the line, so for a 1000 km line those values are 10 times higher than for a 100 km line. The distributed capacitance is approximated by putting half at each end. The same article discusses VARs and Voltage Control problems, and why too much C causes big problems.

Since DC has no capacitance to ground, the capacitance limit does not apply. The operating voltages over long distances can be pushed higher and higher. That is why HVDC becomes more attractive than AC for long distances, the HV for high voltage is the key part of the phrase. For example:

Xinjiang[/PLAIN] - Anhui HVDC line in China, 3333 km, 1100kV, 10000 MW.

p.s. I know that there are different numbers of wires in three-phase AC and DC, and that there are complexities regarding power factor, and so on. Those effects are secondary. The primary effects are what I described above.
https://www.physicsforums.com/insights/ac-power-analysis-part-2-network-analysis/

 

[houlahound]

I can't tell in the diagram if C is between lines or between line and the ground?

Because P = VI the tiny currents in HVAC are going to be less lossy than HVDC. Where is HVDC advantage?

If it is in all the other frequency related losses I would have to see the calculations to believe it. Will view link.

I thought this was settled in the 1800's hence HVAC now.

The HVDC currents must be frikkin huge at kilovolts.

 

[anorlunda]

I can't tell in the diagram if C is between lines or between line and the ground?

Because P = VI the tiny currents in HVAC are going to be less lossy than HVDC. Where is HVDC advantage?

If it is in all the other frequency related losses I would have to see the calculations to believe it. Will view link.

I thought this was settled in the 1800's hence HVAC now.

The HVDC currents must be frikkin huge at kilovolts.

Sorry, the line at the bottom of the diagram is meant to represent ground. The capacitances are to ground.

You got it backwards. P=VI, therefore I=P/V. If V is twice as large for the same P, then I is half as large. So HVDC currents are not frikking huge, they are lower.

Yes it was settled in the 1890s, or so we thought. The huge advantage of AC was the ability to change the voltage using transformers. But the problem with too high C discussed in #6 didn't come up until nearly 100 years later as we tried to make really high voltage AC lines for very long distances.

HVDC can't be used to completely replace AC. We still need many voltages for transmission, distribution and household uses. Therefore, AC and transformers will never go away, and most of the 1890 conclusions still apply. But AC no longer has a monopoly for the longest lines.

There are numerous other AC versus HVDC advantages and disadvantages. The wiki article is pretty good.
https://en.wikipedia.org/wiki/High-voltage_direct_current#Advantages_of_HVDC_over_AC_transmission
https://en.wikipedia.org/wiki/High-voltage_direct_current#Disadvantages

 

[houlahound]

I get it, the DC is being stepped up an down now like AC which they could not do in the old days.

 

[mfb]

Interesting news. Do you expect that to become a common trend as HVDC gets more common?

I get it, the DC is being stepped up an down now like AC which they could not do in the old days.

That is the key point. Power semiconductors give a high conversion efficiency, but they are a quite recent development. The demand for power transmission over 1000+ km didn't exist 100 years ago either.

 

[houlahound]

My town still has its two functioning power generators (only started for tourists now). Every other town had the same. We are 5 minutes drive from a modern generator that supplies 25 percent of the states power thru the national grid. Every year we have power failures and some of our power comes from 1000's of km's away.

The locals would prefer the old generators were still uaed.

What is the logic of a power grid? The infrastructure is insanely expensive and is subject to environmental disasters and price manipulation.

What is the argument for a grid, many homes here generate their own power and put it into the grid and get paid by the energy supplier who in turn charge people relying on the grid more to make up for lost revenue from homes putting power into the grid.

Tear the grid down so the minority don't have to subsidise the rest and increase reliability and local economy.

IMO we should be planning for independent, isolated power generation like what worked in the past. Renewables make local power generation even more logical.

 

[mfb]

You would have more power failures with a purely local grid. I don't know where you live: Here in Europe we have a well-connected grid, and power outages are basically nonexistent. A few local short-term ones after severe storms or whatever, but nothing large and nothing long-lasting.

A local grid doesn't work well together with large power plants - but those are the most efficient and cheapest ones. Burning stuff locally is worse for the environment, is more expensive, and can get a logistics nightmare quickly.

Tear the grid down so the minority don't have to subsidise the rest and increase reliability and local economy

It is exactly the opposite.

Renewables make local power generation even more logical.

They need the grid as well, and they need a better grid to handle the fluctuating power. You don't want power only if the sun is shining in your town.

 

[houlahound]

Well that's the corporate line. Not sure if its the physics. Euro sure, millions of people all stacked on top of each other you need lots of instantaneous power. Isolated rural communities of a few thousand people not so much.

 

[mheslep]

I don't see any mention of unit cost for the Chinese HVDC projects in the IEEE article.

One of the most recently completed HVDC projects in N. America (2015) was the Western Alberta Transmission line, 1 GW over 217 miles for $1.65B, or $7.6 million per GW-mile. That seems cost prohibitive for moving something like enough wind generation from the US midwest over a thousand miles to displace a new east coast nuclear plant.

http://www.marketwired.com/press-re...Alberta-transmission-line-project-1840556.htm

 

[dlgoff]

That seems cost prohibitive for moving something like enough wind generation from the US midwest over a thousand miles to displace a new east coast nuclear plant.

You may like this .pdf: Five keys to wind farm bankability

 

[anorlunda]

Interesting news. Do you expect that to become a common trend as HVDC gets more common?

Not necessarily HVDC, but power electronics yes. There is plenty of room to rethink grid operations given the new abilities of power electronics, but two points.

1. In the USA and Europe, the bulk transmission is already highly reliable. Improvements in local power distribution are where gains can be made.
2. The power industry is very conservative, thank God. Replacing a section of bare wire with active components and software is anti-KISS and the first approximation is that it decreases reliability; especially in a world full of cyber attacks. That sets the bar high for the new things to prove that their benefits outweigh the disadvantages. Advocates and salesmen are motivated to put the bar lower, "Spend the money first and hope for benefits later. Heck, if government pays we don't even care if the benefits never materialize; it stimulates jobs."

Tear the grid down so the minority don't have to subsidise the rest and increase reliability and local economy.

IMO we should be planning for independent, isolated power generation like what worked in the past. Renewables make local power generation even more logical.

The frustration in your post comes through loud and clear. It is likely that the outages you are experiencing are because of the local power distribution, and not the cross country transmission grid. Here's how to tell. If the outage blacks out a whole city or region, it is a transmission grid failure. If it blacks out a neighborhood or two, it is local distribution.

There is already a strong trend for distributed generation. Rooftop solar is leading that, but neighborhood level mini-grids are actively investigated. That may improve reliability. But for the millions of people who live in high-rise apartments, there is no room for them to do their own solar. The city of Niagara Falls can't consume all that power themselves, we need a grid to ship it elsewhere. Ditto for North Sea wind farms (unless you are willing to live on a houseboat moored to the wind tower.)

But you did put your finger on a social problem that might play a bigger role in the future -- groups subsidizing other groups. The unit costs of electric power are lowest with the highest population density. That puts inner city customers in the position of subsidizing everyone else. In the USA, we cherish and celebrate those subsidies. The Rural Electric Adminstration still exists. Someday, that might blow up politically. Europe already leads the USA in control of population density. For example, Jacksonville Florida 370 $people/km^2$ versus Dusseldorf Germany 61000 $people/km^2$ (if I did the arithmetic right).

However, if the power line on your street is knocked down by a falling tree or chewed up by a squirrel, it doesn't matter where the power is generated. (Squirrels are the #1 problems in parts of the USA.) Before advocating tearing things down, you owe it to yourself to learn whether that would improve or worsen your service..

I don't see any mention of unit cost for the Chinese HVDC projects in the IEEE article.

That seems cost prohibitive for moving something like enough wind generation from the US midwest over a thousand miles to displace a new east coast nuclear plant.

I agree if the reference case is Midwest wind delivered to East Coast loads.

In the OP, I said that the motivation in China is a combination of geography and demographics, so naturally China is not directly comparable to North America or Europe. A better analogy would be if 65% of US power generation was located in Alaska. In Europe, visualize Norway powering the whole continent from Scandinavia down to Italy. Then America and Europe would have a much bigger interest in HVDC. But we don't have that so China will probably continue to lead in HVDC.

 

[OmCheeto]

I don't see any mention of unit cost for the Chinese HVDC projects in the IEEE article.

One of the most recently completed HVDC projects in N. America (2015) was the Western Alberta Transmission line, 1 GW over 217 miles for $1.65B, or $7.6 million per GW-mile. That seems cost prohibitive for moving something like enough wind generation from the US midwest over a thousand miles to displace a new east coast nuclear plant.

http://www.marketwired.com/press-re...Alberta-transmission-line-project-1840556.htm

That seems really cheap, considering how long it will last, based on my recent googling, and calculations: $0.0031/kwh

longevity of the system: 60 years
power: 1,000,000,000 watts
energy: 60,000,000,000 watt years = 525,948,768,000 kwh
dollars: 1,650,000,000
$/kwh: 0.0031

What I pay for electricity: $0.11

That looks like 2 orders of magnitude, to me.

A grand example of "power" vs "energy" vs "time".

Thank you.

ps. A brief foray of my "stream of consciousness" morning's googlings:

https://en.wikipedia.org/wiki/Pacific_DC_Intertie

The line capacity is 3,100 megawatts, which is enough to serve two to three million Los Angeles households and represents almost half (48.7%) of the Los Angeles Department of Water and Power (LADWP) electrical system's peak capacity.​

https://www.bpa.gov/news/newsroom/Pages/BPA-green-lights-DC-Intertie-improvements.aspx

The improvements, estimated to cost approximately $428 million, would modernize equipment that was cutting edge when installed more than 40 years ago but has aged to where BPA now in extreme circumstances locates parts on Ebay.

"Some of the equipment at our Celilo Substation is so old, finding replacement parts has become a challenge. Some of this equipment should be in the Smithsonian," Silverstein said.

Increased costs for the upgrade will only be charged to users of the California Intertie.​

mheslep

One of the most recently completed HVDC projects in N. America (2015) was the Western Alberta Transmission line, 1 GW over 217 miles for $1.65B, or $7.6 million per GW-mile. That seems cost prohibitive for moving something like enough wind generation from the US midwest over a thousand miles to displace a new east coast nuclear plant.

time 2015 year
cap 1 gigawatt
length 217 miles
cost $1.65 $billion

my calculations:
see above​

https://www.bpa.gov/news/newsroom/Pages/Direct-current-line-still-hot-after-40-years.aspx

5/26/2010 12:00 AM

"When you help build a region, you help build your nation."

That's what President John F. Kennedy said in a letter to the Bonneville Power Administration on its 25th anniversary. It was 1962.

The DC line came to life May 21, 1970

The colossal DC line spans 846 miles across 4,200 towers from Celilo near The Dalles, Ore., to the Sylmar Converter Station in Los Angeles.

Thanks to the intertie, electricity consumers on the West Coast enjoy a unique power-sharing arrangement that takes advantage of seasonal weather differences.

In spring and early summer, Northwest rivers usually provide more water for power generation than the region needs. At the same time, temperatures – and air-conditioning needs – in the Southwest climb. That is when the DC intertie sends power south. (It is offered in the region first and only sent south if it is surplus to regional needs.) The revenues from sales of this surplus help keep Northwest rates lower.

At other times, such as in winter and at night, California power plants generate more electricity than local consumers need. When temperatures in the Northwest are low and heating needs rise, the power flows north.

The Pacific Northwest Consumer Power Preference Act of 1964 concerned Northwest governors that the Intertie would not siphon Northwest power. The Preference Act required that BPA sell firm energy first to electric utilities in the Northwest and that only power surplus to Northwest needs would be offered for sale outside the region.

Because the two regions can back each other up, thanks to the intertie, less power has been generated at fossil-fuel power plants. And the regions do not have to build expensive power plants to meet peaking needs: plants that would be idled much of the time. Fish benefit from the intertie too. Money that southwestern utilities pay for power from BPA helps finance fish and wildlife restoration projects in the Columbia River Basin.​

https://www.nwcouncil.org/media/27759/2001_11.pdf
May 2001

California enjoys the greatest benefit from the Intertie.
In the last 15 years, for example, the Northwest
has sent more electricity to the Southwest -- mostly to
California -- every year than it has received. Power
generated in the Northwest has saved Southwest utilities
more than 2.8 trillion cubic feet of natural gas that would
have been burned in power plants.

Earlier that August, on the 14th, Congress approved
the Public Works Appropriations Bill of 1965 -- the
coming fiscal year -- with $42.2 million for Bonneville
and $3.3 million for the Bureau of Reclamation to construct
the federal portion of the Intertie, which is in
Oregon. The California portion would be built by California
utilities, primarily the Los Angeles Department of
Water and Power. In all, the construction cost would top
$700 million.

​

 

[mheslep]

That seems really cheap,

My midwest to east coast scenario: five times greater distance ie 1000 miles, not across the sparse plains of Canada but across the US, fed by 35% capacity factor wind, 5% discount rate, maintenance of 1000 miles of line.

 

[mheslep]

You may like this .pdf: Five keys to wind farm bankability

LCOE is a cost to deploy caculation. An investor needs to know both cost *and* return on investment. For wind generation, after reaching 10% of load or so, as in Texas, the answer to the investor is increasingly, 'zero, or maybe you must pay people to take your wind power'. The reason investors continue to invest is that the government pays them $23 per MWh, for now, which covers the losses.

 

[dlgoff]

In spring and early summer, Northwest rivers usually provide more water for power generation than the region needs. At the same time, temperatures – and air-conditioning needs – in the Southwest climb. That is when the DC intertie sends power south. (It is offered in the region first and only sent south if it is surplus to regional needs.) The revenues from sales of this surplus help keep Northwest rates lower.

At other times, such as in winter and at night, California power plants generate more electricity than local consumers need. When temperatures in the Northwest are low and heating needs rise, the power flows north.

I know this thread is about DC ties but to be sure there's no misconceptions when talking about "power flow directions"; power will "flow" where it's consumed through what ever wire paths that are available. Cost of this consumed energy is all done by regulators.
e.g.

See how these regulators work from United States Electricity Industry Primer - Department of Energy

 

[houlahound]

In theory ...While energy is bid or sold for the last 2percent of coal/gas powered energy can be put up to 1000's of times more than what it would normally sold. Its extortion.

The coal price death spiral I think has commenced.

 

[mfb]

Euro sure, millions of people all stacked on top of each other

The Eastern US (everything east of the Mississippi) has 180 million people living in 2.4 million square kilometers, for an average density of 75/km², nearly the same as Europe (73/km²).

For example, Jacksonville Florida 370 $people/km^2$ versus Dusseldorf Germany 61000 $people/km^2$ (if I did the arithmetic right).

Düsseldorf has 2800/km². Those comparisons can be misleading. Düsseldorf is in a densly populated area - its area is really just the town and nearly nothing outside. Jacksonville has a lot of fields that are included in the area, but don't contribute notably to the population. Miami has a population density of 4500/km², for example - it is surrounded by other cities.

In theory ...While energy is bid or sold for the last 2percent of coal/gas powered energy can be put up to 1000's of times more than what it would normally sold. Its extortion.

The coal price death spiral I think has commenced.

What?

 

[anorlunda]

üsseldorf has 2800/km2. Those comparisons can be misleading.

Maybe my arithmetic was off, or confused by the difference between a city and its metro area.

I believe it is true that European medium or small size cities typically have much higher population densities than comparable American ones, primarily via multi-family dwellings in Europe versus single-family houses surrounded by grass lawns in the USA. Perhaps another PF member can assist me to find the stats to back that up.

All this is a side issue to my real point, that the cost of utilities (including electric) is strongly influenced by population density. Where rates are uniform across the state, people living in high density areas subsidize those who live in low density areas. I believe that the unbalance is politically unstable in the long term.

 

[mfb]

Nearly everything is easier to install in high-density areas. Mail delivery is cheaper, you need fewer road kilometers per person, high-speed internet access is easier to install, ...

In some cases (mail, roads). the state guarantees that everyone has proper access to it, in other cases (internet speed) people in rural areas have worse service.

 

[OmCheeto]

I know this thread is about DC ties...

I disagree.

They are doing things that old timers like me never imagined. Specifically, they challenged the catechism that increased interconnectivity of the AC grid is more reliable; a principle we used for more than 100 years. Hats off to them.

I take this comment, as a; "hmmmmm... Old school worked for quite a while, but things are changing, and now we need to think about this some more."

but to be sure there's no misconceptions when talking about "power flow directions"; power will "flow" where it's consumed through what ever wire paths that are available. Cost of this consumed energy is all done by regulators.
e.g.
View attachment 109542

See how these regulators work from United States Electricity Industry Primer - Department of Energy

Thank you for the link.
But that, along with @Astronuc 's post from last week, has my head about to explode, trying to figure this all out.

ps. It's always fun, to see how far I can push my old brain, without popping a blood vessel.

 

[dlgoff]

Thank you for the link.
But that, along with @Astronuc 's post from last week, has my head about to explode, trying to figure this all out.

This pdf my help the explosion factor: http://www.nerc.com/docs/oc/rs/NERC%20Balancing%20and%20Frequency%20Control%20040520111.pdf
From the section Balancing and Frequency Control:

Area Control Error (ACE) Review
The Control Performance Standards are based on measures that limit the magnitude and direction of the Balancing Authority’s Area Control Error (ACE). The equation for ACE is:
ACE = (NI_A- NI_S) - 10B (F_A - F_S) - I_ME
Where:
NI_A is Net Interchange, Actual
NI_S is Net Interchange, Scheduled
B is Balancing Authority Bias
F_A is Frequency, Actual
F_S is Frequency, Scheduled
I_ME is Interchange (tie line) Metering Error

NI_A is the algebraic sum of tie line flows between the Balancing Authority and the Interconnection. NI_S is the net of all scheduled transactions with other Balancing Authorities. In most areas, flow into a Balancing Authority is defined as negative. Flow out is positive.

The combination of the two (NI_A - NI_S) represents the ACE associated with meeting schedules, without consideration for frequency error or bias, and if used by itself for control would be referred to as “flat tie line” control.

The term 10B (F_A - F_S) is the Balancing Authority’s obligation to support frequency. B is the Balancing Authority's frequency bias stated in MW/0.1Hz (B’s sign is negative). The “10” converts the Bias setting to MW/Hz. FS is normally 60 Hz but may be offset ± 0.02 Hz for time error corrections. Control using “10B (F_A - F_S)” by itself is called “flat frequency” control.

I_ME is a correction factor for meter error. The meters that measure instantaneous⁶ flow are not always as accurate as the hourly meters on tie lines. Balancing Authorities are expected to check the error between the integrated instantaneous and the hourly meter readings. If there is a metering error, a value should be added to compensate for the estimated error. This value is I_ME. This term should normally be very small or zero.

 

[OmCheeto]

LCOE is a cost to deploy caculation. An investor needs to know both cost *and* return on investment. For wind generation, after reaching 10% of load or so, as in Texas, the answer to the investor is increasingly, 'zero, or maybe you must pay people to take your wind power'. The reason investors continue to invest is that the government pays them $23 per MWh, for now, which covers the losses.

That negative pricing, may be their own fault.

The Night They Drove the Price of Electricity Down

First, Texas is an electricity island. The state often behaves as if it is its own sovereign nation, and indeed it was an independent republic for nearly 10 years. Alone among the 48 continental states, Texas runs an electricity grid that does not connect with those that serve other states. The grid is run by Electric Reliability Council of Texas, or ERCOT.​

I'm still fascinated by the BPA live graphs.
It looks as though the wind farms nearly supplied BPA's domestic load requirements, at a few points.
The flat nature of the "thermal" producers is also interesting.

 

[OmCheeto]

This pdf my help the explosion factor: http://www.nerc.com/docs/oc/rs/NERC%20Balancing%20and%20Frequency%20Control%20040520111.pdf
From the section Balancing and Frequency Control:

I'm pretty sure, that this will only make matters worse.

"The more I know, the less I understand".

I'm still trying to get a grip on what would happen if the BPA "circuit breakers" to Los Angeles were to "trip".

LA, gets half of its power from the Pacific Northwest, and, it's A LOT!
What the heck do you do with gigawatts of power, when there's nowhere for it to go?

Gob loads of science going on here, and I haven't even started thinking about it yet.

Oh, what's this. I pushed the emoticon buttons, and...

 

[dlgoff]

What the heck do you do with gigawatts of power, when there's nowhere for it to go?

Protective Relaying for one thing.

https://en.wikipedia.org/wiki/Protective_relay]In[/PLAIN] electrical engineering, a protective relay is a relay device designed to trip a circuit breaker when a fault is detected.[1]:4 The first protective relays were electromagnetic devices, relying on coils operating on moving parts to provide detection of abnormal operating conditions such as over-current, over-voltage, reverse power flow, over-frequency, and under-frequency.[2]

edit: here's a little overview of these devices - http://ecmweb.com/content/what-know-about-protective-relays

 

[OmCheeto]

Protective Relaying for one thing.edit: here's a little overview of these devices - http://ecmweb.com/content/what-know-about-protective-relays

Ummm... I've operated mere megawatt turbine generators, and tripped them off line, under full load.
It wasn't pretty.
But that isn't the problem I'm seeing.
I'm looking at, um, a lot of inertia, 3 orders of magnitude greater, with nowhere for it to go.
And that scares the bejezits out of me.

 

[dlgoff]

Ummm... I've operated mere megawatt turbine generators, and tripped them off line, under full load.
It wasn't pretty.
But that isn't the problem I'm seeing.
I'm looking at, um, a lot of inertia, 3 orders of magnitude greater, with nowhere for it to go.
And that scares the bejezits out of me.

It's spread over a really big grid that's somewhat like a bunch of rubber bands. The dumped power isn't dissipated instantaneously.

 

[OmCheeto]

It's spread over a really big grid that's somewhat like a bunch of rubber bands. The dumped power isn't dissipated instantaneously.

But that seems counter to what anorlunda posted in his OP; "In July, CSG neutralized this threat by shutting off Yunnan’s AC links to the rest of its grid, turning the province into its own distinct synchronous zone."

But, as I've said, this is way too much information for me to process in just a few days, so, I'm going to take a nap.

 

[dlgoff]

But that seems counter to what anorlunda posted in his OP; "In July, CSG neutralized this threat by shutting off Yunnan’s AC links to the rest of its grid, turning the province into its own distinct synchronous zone."

Pretty sure those (disconnections) didn't occur all at once.

 

[anorlunda]

Ummm... I've operated mere megawatt turbine generators, and tripped them off line, under full load.
It wasn't pretty.
But that isn't the problem I'm seeing.
I'm looking at, um, a lot of inertia, 3 orders of magnitude greater, with nowhere for it to go.
And that scares the bejezits out of me.

You are quite right to be scared. I'll explain, but first let's simplify. Take the grid out of the picture. Imagine the circuit breaker closest so the generator at a hydro plant tripping. That is the worst case scenario from the generator's point of view.

There can be a huge amount of kinetic energy in the water traveling through the pipes (penstocks) leading to the turbine. Instantly, we give a trip signal the gates that close off that pipe right in front of the turbine. We can't close them instantaneously, or the force of the water would wreck them and wash the gates and turbines and generators and the whole power plant down the river. While they are closing gradually, the excess energy goes into accelerating the turbine and generator. 200% even 250% peak speed can be expected, so that is one of the limiting design criteria for the turbine generator. Remember, the breaker is open so the electrical behavior is out of the picture.

But there is still an enormous amount of K.E. in the water. As you say, it has to go somewhere. The somewhere is an engineered solution built into many hydro plants called a surge tank. The water surges up into the tank, converting K.E. to P.E. I heard that in some cases the water overflows the tank and creates a geyser shooting up into the sky. I would love it if some PF member could find a video of such a geyser.

Next step of complexity, imagine that a different breaker very remote from the generator trips leaving the generator with no load, but also with 1000 km of transmission line open-circuit on the far end attached. We call it radial load rejection, and it is very very stressful on the plant both electrically and mechanically.

p.s. You are touching on my favorite subject. I used to teach a course to utilities in the USA and abroad called "Generation Dynamics and Control" To me, that is a very fun subject. It incorporates the dynamics of every type of generation, and the grid, and how to orchestrate and control it both locally and system-wide.

 

[jim hardy]

I'm looking at, um, a lot of inertia, 3 orders of magnitude greater, with nowhere for it to go.
And that scares the bejezits out of me.

well it's not as if all the generators had to suddenly stop rotating when the breakers open. They just have to reduce their input power to zero before it accelerates them to dangerous speed. They can idle at synchronous speed if prime mover is available or coast down...
Our overspeed trips were 111% , indeed i saw one of our turbines at 3900 PM a minute into a regional blackout.
Our turbine inlet valves close imperceptibly fast so the energy available to accelerate turbine is only whatever steam is already in the turbine. To prevent acceleration we keep the generator breakers closed for ten seconds after a turbine trip. Should an electrical fault trip the generator from a high power level, an extra set of valves halfway along the multistage turbine snap closed to bottle up its high pressure half, that trapped steam gets released gradually so as to not overspeed kindly Mr Turbine.

Steam that's already en route to the turbine from the reactor gets bypassed to the condenser , dumping copious heat there. That's how the thermal inertia of the reactor system gets dissipated. Reactor system cools something like thirty degrees between full power and zero power.

That's a steam plant response. I defer to folks who've been around hydro for that scenario.
Windmills can feather and brake. But individually they're exceedingly tiny amounts of power . Our feedwater pumps were five megawatts apiece, about same as biggest wind turbine extant.

old jim