Why are power systems so large and interconnected? For example, what technical obstacles prevents the US eastern interconnection from being 8 isolated islands? Why not separate them by ISO/RTO? Why does every power grid in the world strive to be as large as geography allows?
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Why are Power Grids so Large? 5
- Thread starter Mbrooke
- Start date
Size equals stability.
AND
It allows power contracts to reach the largest number of remote cheap sources.
Keith Cress
kcress -
AND
It allows power contracts to reach the largest number of remote cheap sources.
Keith Cress
kcress -
- Thread starter
- #3
Overvoltage
Electrical
Actually quite the opposite. With many rotating machines (generators and motors), there is a lot of inertia to the system. It takes a lot to "move" that inertia one way or the other, so a rather large disturbance or series of disturbances to have any widespread impact on a large power system, such as the eastern interconnection.
BrianPetersen
Mechanical
If someone at a steel mill in Ohio turns on a furnace that suddenly starts drawing 100 MW of load, rotating generating equipment in Quebec helps keep the frequency and voltage steady throughout the grid. Having everything connected to everything else acts like a much bigger flywheel.
- Thread starter
- #6
davidbeach
Electrical
One advantage of large grids is that different areas have their peak loads at different times. If every Distribution company had to have dedicated generation necessary to cover its peak load there would have to be far more installed generation than if it can all be pooled.
Then there's the need for reserves; large grids mean that reserve pools are possible. For small grids each would have to have sufficient reserves and the total amount of reserves would be considerably more than is necessary when a regional pool can cover many different company's needs.
To me, the missed lesson of August 2003 is the need for "control joints" in the interconnected grids. Instead of trying to hold everything together at all costs, as the present regulatory environment tries to do, there should be a robustness target that covers all reasonable contingencies but then allows separation of the impacted area from the rest of the grid before the trouble cascades. The western interconnection takes a step in that direction with a RAS that will separate north from south. Not sure of the history, but I think it may have been a response to the 1996 disturbances and I don't think it's ever activated. I wonder if those two pieces are still too large.
Then there's the need for reserves; large grids mean that reserve pools are possible. For small grids each would have to have sufficient reserves and the total amount of reserves would be considerably more than is necessary when a regional pool can cover many different company's needs.
To me, the missed lesson of August 2003 is the need for "control joints" in the interconnected grids. Instead of trying to hold everything together at all costs, as the present regulatory environment tries to do, there should be a robustness target that covers all reasonable contingencies but then allows separation of the impacted area from the rest of the grid before the trouble cascades. The western interconnection takes a step in that direction with a RAS that will separate north from south. Not sure of the history, but I think it may have been a response to the 1996 disturbances and I don't think it's ever activated. I wonder if those two pieces are still too large.
-
1
- #8
Ahem...
As politely as I can, I must point out that this is not in fact the case...
The Hydro-Québec / Trans-Énergie system is not synchronously tied to the rest of the Eastern Interconnection. Asynchronous ties are in place between HQ/TÉ and the Eastern Interconnection to facilitate power exchanges/sales between them, and generators on both sides are occasionally synchronously connected to the other system, but standing synchronous connections will go unstable in a matter of minutes [I've seen the graphs].
If you had mentioned rotating equipment other than in Québec...
CR
"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]
SlideRuleEra
Structural
Mbrooke said:
There is another side of the issue, disaster recovery. Consider a power system to be composed of:
Generation
Transmission
Distribution
Generation is, by far, the strongest link, failure of distribution and/or transmission may result in plenty of power and no way to send it out to customers. This is exactly what happened to our system in September 1989... direct hit on our entire system from Hurricane Hugo.
With transmission / distribution going down in an unpredictable manner, there was not choice but to break all ties to the grid and have our generation fleet go dark. The day after the storm passed, some generation was restarted (that's another story) then ties to the grid were reestablished.
Bottom line: The grid connections allowed power to brought into our system is a somewhat backwards manner to supply customers who could still receive power or customers who could have transmission/distribution quickly restored. Took weeks / months to get our own transmission/distribution fully restored to where our own generation could supply all customers as intended.
![[idea]](/data/assets/smilies/idea.gif "[idea] [idea]")
BrianPetersen
Mechanical
crshears - I did not know that! Learn something every day. Star for you!
Regional Interconnection of wide-area network in the US is more historical than planning.
The following is a partial historical background that leads to the wide-area network interconnection in the US.
1880-1920: Utilities start developing individually some producing DC and other AC that allowed utilities to grow the electric grids over larger areas guaranteed by a monopoly status granted by the state.
1920-1980: Utilities were “vertically structured”. Small interconnections between neighboring utilities existed to increase reliability and share excess generation. The electric demand grew to create opportunities to extend the grid even to larger areas. In 1941, the Southwest Power Pool (SPP) was formed with 11 regional utilities entering into an inter-company agreement to provide the electricity demand in the early days of WWII. In 1966, the Northeast Power Coordinating Council (NPCC) was formed, as a successor to the CA–USA Interconnection (CANUSE).
NERC was formed in 1968 in the aftermath of the 1965 Northeast Blackout. in the 1970's the energy crisis ended with a 400% increase in oil price. After that, the US Congress tried to changed the vertical structure to allow wholesale competition in electricity production; facilities that produced power more efficiently or used renewable energy could enter the marketplace, while the transmission operators (ISOs and RTOs) maintained a monopoly over the management of the grid – a change known as “restructuring.”
The primary reason for developing an electricity grid interconnection is to reduce the overall combined economic costs of supplying electricity. Energy trading between regions and nations offers significant direct economic benefits. According to a DOE study in early 2000, the transmission system facilitates wholesale electricity markets that saved consumers’ electricity bills by nearly $13 billion annually. Interconnecting utilities also benefit system reliability and provides more flexibility for O&M.
The following is a partial historical background that leads to the wide-area network interconnection in the US.
1880-1920: Utilities start developing individually some producing DC and other AC that allowed utilities to grow the electric grids over larger areas guaranteed by a monopoly status granted by the state.
1920-1980: Utilities were “vertically structured”. Small interconnections between neighboring utilities existed to increase reliability and share excess generation. The electric demand grew to create opportunities to extend the grid even to larger areas. In 1941, the Southwest Power Pool (SPP) was formed with 11 regional utilities entering into an inter-company agreement to provide the electricity demand in the early days of WWII. In 1966, the Northeast Power Coordinating Council (NPCC) was formed, as a successor to the CA–USA Interconnection (CANUSE).
NERC was formed in 1968 in the aftermath of the 1965 Northeast Blackout. in the 1970's the energy crisis ended with a 400% increase in oil price. After that, the US Congress tried to changed the vertical structure to allow wholesale competition in electricity production; facilities that produced power more efficiently or used renewable energy could enter the marketplace, while the transmission operators (ISOs and RTOs) maintained a monopoly over the management of the grid – a change known as “restructuring.”
The primary reason for developing an electricity grid interconnection is to reduce the overall combined economic costs of supplying electricity. Energy trading between regions and nations offers significant direct economic benefits. According to a DOE study in early 2000, the transmission system facilitates wholesale electricity markets that saved consumers’ electricity bills by nearly $13 billion annually. Interconnecting utilities also benefit system reliability and provides more flexibility for O&M.
-
2
- #12
Blessed Christmas, all.
Here it is, 7 pm local time on Christmas Day, and I'm @ work monitoring Ontario's power system while Paul McCartney's "Ecce Cor Meum" plays softly in my headphones so as to not wash out all ambient sound...
I have the time to expand somewhat on the above quote, so here goes.
One of my previous work locations was at the R. H. Saunders Generating Station just west of Cornwall, Ontario, Canada. The electrical configuration of the station is such that each of the four pairs of four units each export their power to one GOT and thence onto the grid "radially," to use our terminology, meaning via four express circuits, each of which terminates in its own "diameter" at a transformer station ~ 4 km away. The two circuits for units 9-12 and 13-16 are equipped with a somewhat convoluted transfer bus which allows reconnection of either four units or eight from the Ontario system onto the Trans-Énergie system - and as I type this, units 9-12 are "exporting" to Québec. This is accomplished by disconnecting a portion of one of our 230 kV circuits from our own grid so that it can be placed on potential from the Trans-Énergie system [historically, we local GS and TS operators referred to this as our French Connection]. The applicable busses at the TS [and the circuit or circuits to Saunders] are then placed on pot from HQ to an open synchronizing breaker at the GS, rated for sustained operation at 2 pu withstand voltage, at which time auto-synchronization is enabled and takes place and the hydraulic units can be loaded up.
Back when I worked there, there was a paper frequency recording chart in the control room, the potential source of which could be selected from any of the four GOTs. I deliberately selected T3 one night prior to the commencement of export, and noted the quite narrow frequency trace of the Eastern Interconnection. The span of the frequency trace once T3 was connected to the Trans-Énergie system was notably wider, meaning Québec's system frequency that night was definitely less stable than that of the EI, just as it is most of the time...yet for all that, none of the Québecois ever seem to notice anything amiss.
I quite agree, but have no clue as to what such a macro-system instability detector might look like, what intelligence it would need to perform its task, what it would be armed to do, etc., etc.; by way of example, the graph traces of one of those inadvertent EI/Québec parallels, actually a load graph of the circuit used to export power from Saunders to Québec, showed a slow oscillation of gradually increasing magnitude as real power first flowed into and then out of Québec...and it was a sharp-eyed operator taking readings who noticed this trace. Grasping almost immediately the significance of this trace and observing how perilously close the circuit was to tripping on line protection due to load encroachment, the TS operator, without consulting the Regional or System operators, promptly proceeded to the circuit breaker control for that circuit, opened one breaker in the breaker-and-a-half scheme, then opened the second as the circuit real power flow transited through zero, thereby separating the EI and Québec systems from each other without incident.
That was just one inadvertent interconnection; designing, building and commissioning a scheme to do this sort of thing at multiple normally-closed grid tie points is not likely Mission: Impossible, but it would certainly take some doing...and the complexities and sophistication of its discriminatory abilities would have to be quite something. For example, Ontario has connection points to the EI at the Manitoba/Ontario border, into Minnesota, at the Ontario/Michigan Interface, and New York/Ontario interfaces at both Niagara/Buffalo and Cornwall/Massena. Vectorially analyzing the summed flows of all these ties in real time, with the ties being at considerable geographic distance from one another, cross comparing them to scheduled tie line flows, factoring in Area Control Error, and correctly deducing when grid instability is in fact occurring would seem to me to an almost monumental task.
Interesting thread so far!
CR
"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]
Here it is, 7 pm local time on Christmas Day, and I'm @ work monitoring Ontario's power system while Paul McCartney's "Ecce Cor Meum" plays softly in my headphones so as to not wash out all ambient sound...
I have the time to expand somewhat on the above quote, so here goes.
One of my previous work locations was at the R. H. Saunders Generating Station just west of Cornwall, Ontario, Canada. The electrical configuration of the station is such that each of the four pairs of four units each export their power to one GOT and thence onto the grid "radially," to use our terminology, meaning via four express circuits, each of which terminates in its own "diameter" at a transformer station ~ 4 km away. The two circuits for units 9-12 and 13-16 are equipped with a somewhat convoluted transfer bus which allows reconnection of either four units or eight from the Ontario system onto the Trans-Énergie system - and as I type this, units 9-12 are "exporting" to Québec. This is accomplished by disconnecting a portion of one of our 230 kV circuits from our own grid so that it can be placed on potential from the Trans-Énergie system [historically, we local GS and TS operators referred to this as our French Connection]. The applicable busses at the TS [and the circuit or circuits to Saunders] are then placed on pot from HQ to an open synchronizing breaker at the GS, rated for sustained operation at 2 pu withstand voltage, at which time auto-synchronization is enabled and takes place and the hydraulic units can be loaded up.
Back when I worked there, there was a paper frequency recording chart in the control room, the potential source of which could be selected from any of the four GOTs. I deliberately selected T3 one night prior to the commencement of export, and noted the quite narrow frequency trace of the Eastern Interconnection. The span of the frequency trace once T3 was connected to the Trans-Énergie system was notably wider, meaning Québec's system frequency that night was definitely less stable than that of the EI, just as it is most of the time...yet for all that, none of the Québecois ever seem to notice anything amiss.
I quite agree, but have no clue as to what such a macro-system instability detector might look like, what intelligence it would need to perform its task, what it would be armed to do, etc., etc.; by way of example, the graph traces of one of those inadvertent EI/Québec parallels, actually a load graph of the circuit used to export power from Saunders to Québec, showed a slow oscillation of gradually increasing magnitude as real power first flowed into and then out of Québec...and it was a sharp-eyed operator taking readings who noticed this trace. Grasping almost immediately the significance of this trace and observing how perilously close the circuit was to tripping on line protection due to load encroachment, the TS operator, without consulting the Regional or System operators, promptly proceeded to the circuit breaker control for that circuit, opened one breaker in the breaker-and-a-half scheme, then opened the second as the circuit real power flow transited through zero, thereby separating the EI and Québec systems from each other without incident.
That was just one inadvertent interconnection; designing, building and commissioning a scheme to do this sort of thing at multiple normally-closed grid tie points is not likely Mission: Impossible, but it would certainly take some doing...and the complexities and sophistication of its discriminatory abilities would have to be quite something. For example, Ontario has connection points to the EI at the Manitoba/Ontario border, into Minnesota, at the Ontario/Michigan Interface, and New York/Ontario interfaces at both Niagara/Buffalo and Cornwall/Massena. Vectorially analyzing the summed flows of all these ties in real time, with the ties being at considerable geographic distance from one another, cross comparing them to scheduled tie line flows, factoring in Area Control Error, and correctly deducing when grid instability is in fact occurring would seem to me to an almost monumental task.
Interesting thread so far!
CR
"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]
- Thread starter
- #13
Crshears- wow! Not only do I vote this post of the year in my book, but that sounds like the experience of a lifetime! ![[smile]](/data/assets/smilies/smile.gif "[smile] [smile]")
I've seen the stations on Google earth and always wondered about that one with shared common south bus.
If its not to much to ask- how well would that generating station function if it served only radial load in its immediate geographic area at 80% loading? How well would that station tolerate a 5 cycle fault under a radial feed condition? Fault induced delayed voltage recovery resulting?
I've seen the stations on Google earth and always wondered about that one with shared common south bus.
If its not to much to ask- how well would that generating station function if it served only radial load in its immediate geographic area at 80% loading? How well would that station tolerate a 5 cycle fault under a radial feed condition? Fault induced delayed voltage recovery resulting?
Hi MBrooke,
I wish that had been me, but it wasn't; the highlight of my career [to date!] has been the re-starting of the Ontario power grid in the 2003 blackout, another story entirely.
The last two of your questions I'm uncertain how to answer, but the answer to the first one, if you mean by "radial load in its immediate geographic area at 80% loading" that it would be in the islanded state, I'd have to answer, "Probably not very well at all."
This is not for electrical reasons, but because the Saunders/FDR power development operates as a run-of-the-river plant whose primary function is to regulate the outflow of the St. Lawrence River from Lake Ontario as specified from time to time by the International St. Lawrence River Board of Control, a sub-group of the International Joint Commission; the production of electricity is "merely" a by-product - a very valuable one, but secondary nonetheless. By international statute/agreement, ponding [storage of water over the weekend for release through the generators on the following weekdays] is prohibited; only peaking [varying the water flows so as to attain the average 24-hr release rate specified] is allowed. [And whenever the target river flow is 7,930 cubic metres per second or greater, no peaking at all is allowed.]
As a consequence of this, the stations' output profiles generally do not track the profiles of the loads in the area, meaning that a robust group of circuits are needed to handle the resultant swing flows whenever the generation and load are unbalanced.
One result of this was that during the 1998 Ice Storm, when the entire Cornwall/Massena area, the two GSs, a GM engine casting plant and some other local loads remained tied to what remained of the EI by only one skinny 115 kV circuit, the Saunders/FDR generation had to be constantly tweaked so the flow on that ckt remained within its limits, lest that ckt melt...and as a consequence, a great deal of fancy footwork was required to make the hourly adjustments to the spillage at the Long Sault Dam so as to still pass the required amount of water down the river [since whatever water wasn't being passed through the units had to be released some other way].
CR
"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]
I wish that had been me, but it wasn't; the highlight of my career [to date!] has been the re-starting of the Ontario power grid in the 2003 blackout, another story entirely.
The last two of your questions I'm uncertain how to answer, but the answer to the first one, if you mean by "radial load in its immediate geographic area at 80% loading" that it would be in the islanded state, I'd have to answer, "Probably not very well at all."
This is not for electrical reasons, but because the Saunders/FDR power development operates as a run-of-the-river plant whose primary function is to regulate the outflow of the St. Lawrence River from Lake Ontario as specified from time to time by the International St. Lawrence River Board of Control, a sub-group of the International Joint Commission; the production of electricity is "merely" a by-product - a very valuable one, but secondary nonetheless. By international statute/agreement, ponding [storage of water over the weekend for release through the generators on the following weekdays] is prohibited; only peaking [varying the water flows so as to attain the average 24-hr release rate specified] is allowed. [And whenever the target river flow is 7,930 cubic metres per second or greater, no peaking at all is allowed.]
As a consequence of this, the stations' output profiles generally do not track the profiles of the loads in the area, meaning that a robust group of circuits are needed to handle the resultant swing flows whenever the generation and load are unbalanced.
One result of this was that during the 1998 Ice Storm, when the entire Cornwall/Massena area, the two GSs, a GM engine casting plant and some other local loads remained tied to what remained of the EI by only one skinny 115 kV circuit, the Saunders/FDR generation had to be constantly tweaked so the flow on that ckt remained within its limits, lest that ckt melt...and as a consequence, a great deal of fancy footwork was required to make the hourly adjustments to the spillage at the Long Sault Dam so as to still pass the required amount of water down the river [since whatever water wasn't being passed through the units had to be released some other way].
CR
"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]
There you go Mbrooke! One specific case of a "water flow regulator" that happens to spew electrical power that can be used as long as it's connected to a large system that can regulate around it's essentially non-load based output.
Fascinating! Thanks CR!
Keith Cress
kcress -
Fascinating! Thanks CR!
Keith Cress
kcress -
- Thread starter
- #16
![[glasses]](/data/assets/smilies/glasses.gif "[glasses] [glasses]")
One other stricture: when the specified average flow is < 7930 cms and peaking is allowed, a ceiling of 7930 cms on river flow applies [this is for vessel navigation reasons in the St. Lawrence Seaway].
Answering this one re-defines too many of the applicable real-life paramters, leaving me uncertain as to how to respond, as I'd be way off into some theoretical weeds somewhere.
CR
"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]
Answering this one re-defines too many of the applicable real-life paramters, leaving me uncertain as to how to respond, as I'd be way off into some theoretical weeds somewhere.
CR
"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]
- Thread starter
- #18
Vast difference between coal and nuclear generation, despite the fact they both use a thermal cycle; the main one is that nuclear units [at the least the CANDUs in my background] are base load units, and admirably accord with Keith's wording of being a facility "... that happens to spew electrical power that can be used as long as it's connected to a large system that can regulate around its essentially non-load based output." I consider the fact that they don't release carbon dioxide in great gouting quantities to be very much in their favour.
Coal plants, despite being the greenhouse gas producers they are, can be designed to have a quite respectable turndown ratio, and can therefore follow load quite well; the ones I worked with had TDRs of seven or so to one. Unfortunately they were two-shifted a great deal more than they had been designed for, and it was this that caused economizer drum cracks - and this because of the thermal cycling due to top-ups once the units had been shut down, and not during normal operation; in retrospect, perhaps once the two-shifting operations began in a big way, consideration should have been given to adding a low-capacity positive-displacement triple cylinder topping pump run via variable-speed drive so as to match the shrinkage rate of the water in the boiler as it cooled down...but alas, that did not happen.
CR
"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]
- Thread starter
- #20
- Status
Similar threads
- Locked
- Question
- Locked
- Question