Wednesday 30 March 2016

Why some unit trips result in dropping customers

Events on the electrical system happen fairly regularly and for the most part no one notices them unless the event results in customers being dropped from the system. A trip of Cat Arm a few days ago resulted in customers being dropped in what's called an under frequency load shed. This is an automatic protection on the system which dumps load when generation and load get too far out of balance. Dropping load helps stabilize the system thus avoiding a total collapse. A trip at Granite Canal yesterday didn't trip anyone. Let's look at why.

On March 26th, 2016 at 11:02 hours, the Cat Arm plant tripped offline when the transmission line into that facility tripped off during severe weather in the area. Cat Arm is at the end of the line so when the line goes so to does the plant. Cat Arm is rated for 127 MW and when it tripped was supplying approximately 100 MW.



For whatever reason, Granite Canal tripped off line on March 29th at 17:41 hours. Granite Canal is rated for 40 MW and was supplying something less than that at the time of the trip.

Granite Canal was carrying less load than Cat Arm and that's why one resulted in an under frequency load shed and the other didn't. In the figure below  I have overlaid frequency trends of both events. It's clear that when Cat Arm tripped, taking the 100 MW with it, a greater imbalance was created in the system resulting in load shedding being activated at 58.8 Hz. The relative magnitude of each imbalance can be seen in the fact that the frequency dropped further for the Cat Arm event.


To finish up, whether a unit trip will result in a customer outage or not will depend on the amount of load it is carrying at the time of the trip but also the total amount of load and generation available on the system at that time. The system is typically more stable in winter as opposed to summer as most all generators are on line during the winter.

Monday 22 February 2016

That Darned Third Line

If only NL Hydro had built that third line from Bay D'Espoir Terminal Station to Western Avalon Terminal Station we wouldn't be running out of power here on the Avalon Peninsula as we could use all that spare hydroelectric power not being used off the Avalon...

Or so goes the argument whenever one of these damned Power Watches is called. Actually, it was this same discussion during #DarkNL that got me started digging into our local power system in some detail and spending far too much time on Twitter. Something about it seemed too easy and too foolish for words to be honest.

So, any truth to the claim?

Maybe. It depends on what time of the year you are talking about.

There's two problems with transmitting power to the Avalon in the existing, 2 line, system.
  1. High load on the Avalon drags the voltage down to the point that the system will collapse if it isn't propped up. This is what happened on the morning of March 4, 2015. I was asleep in a camp on the Snare River in the North West Territories and awoke by my vibrating phone. Twitter had exploded after the power went out.
  2. The lines over heat (thermally constrained) when the outdoor temperature gets up around 15 Celsius and load picks up on the Avalon. Think spring and fall of the year, not winter.
Item 2 dictates when the first unit at Holyrood gets started or stopped and item 1 dictates when the second and third units get started at Holyrood [1].

Now, let's toss in Vale's Long Harbour facility. In order to add that load to the system, it was necessary to relocate a capacitor bank from the newly decommissioned Long Harbour Terminal Station to Come By Chance Terminal Station. A good size cap bank was added, 153 MVAr versus 24 MVAr in the old one. MVAr, or reactive power is used to magnetize the power system. Without it, the active power, or MW, can't do any work like heating your house.

The installation of the Come By Chance capacitor bank provided extra MVAr to the Avalon 230 kV transmission system, propped up the voltage, and delayed the necessary start up of the second and third units at Holyrood as per item 1 above. This capacitor bank, installed in 2012, enabled the import of an extra 120 MW onto the Avalon thus saving fuel and improving efficiency at Holyrood [1, 2].

What this capacitor bank doesn't solve is item 2 above and item 2 is what dictates when the first Holyrood unit is started. For that you do need new transmission. In the spring and fall the new line would help delay the start up of the first Holyrood unit per item 2. That is a savings alright, no doubt, but it's not going to help us during the winter peaks. Unfortunately, there is no unused mystery capacity off the Avalon during peaks as the island is capacity constrained during these periods, i.e., we need new generation or smaller peaks. Also, starting and stopping Holyrood frequently is hard on the gear as the equipment is heated and cooled repeatedly. Thermal operators try to avoid that.

When the Grand Falls and Stephenville mills shut down there was some extra generation capacity and energy opened up as the power was no longer being used by the mills. That was 7 - 12 years ago. During peak times now we're using that extra power capacity as peak load has increased sufficiently to use it up. Doesn't help any that during the peaks we also end up losing a Holyrood unit or Hardwoods, etc. 

What the devil does that all mean? The installation of the capacitor bank at Come By Chance TS enabled the import of some extra power onto the Avalon but now we've gotten to the point during peaks that there is no extra power left. Would a third line enable the import of extra power from off the Avalon? Yes it would, but during the peaks there isn't any. During the peaks the island is maxed out and running with low reserve power.

It's a bit different during the Spring to Fall. That's when the system is not in to the winter peak loads and so there would be some generation off the Avalon that the third line would help get to the major load centre on the Avalon. It has nothing to do with water spilling though, that would be base generation. So the third line would help during these periods as it would delay the start up of the first Holyrood unit and enable the import of some of that off Avalon energy.

I asked Hydro to provide spill data for the last 5 years and they were good enough to forward it to me, see the figure below. In the years 2011, 2012, 2013, and 2014 there was little to no spilling in the months of January, February, or March. As for this past year, Hydro hasn't spilled anything since the summer. That's the problem with trying to rely on spilled water, it isn't consistent, and our reservoirs are not large enough to store it all when it does happen.

NL Hydro 5 Year Spill Data

 Once the Maritime Link is built, the spilled energy, along with whatever else is available, is going to Nova Scotia and New England where it will be sold at market rates.

Hydro was going to build the third line back in 2012 but withdrew that application when the province decided it was going to go all in on Muskrat Falls. Back then the line was needed to get power in from a few small hydro plants that they would end up building instead of Muskrat Falls, and to improve upon items 1 and 2 above [5]. After sanction of Muskrat the reasons for the new line changed. I won't get into that here; read reference [4] if you're curious or need some material to help you sleep.

Would a third line from Bay D'Espoir to the Avalon help during the winter peaks? No.

Would a third line help reduce use of Holyrood in the Spring and Fall? Yes.

If you have any questions or comments leave them here or send me a message on Twitter, @kurtbsullivan. If you have some evidence to show otherwise, send it along.


References:
  1. Island Transmission System Outlook, System Planning Department, NL Hydro , December 2010, Pg 4 - 5., http://www.pub.nl.ca/applications/MuskratFalls2011/files/exhibits/Exhibit24-IslandTransmissionOutlook.pdf
  2. Squires, Erin. “Come By Chance Capacitor Bank” Email to Kurt Sullivan. February 10, 2016.
  3. Report on Island Interconnected System to Interconnection with Muskrat Falls addressing Newfoundland and Labrador Hydro, Liberty Consulting, December 17, 2014. (http://www.pub.nl.ca/applications/IslandInterconnectedSystem/files/reports/The-Liberty-Consulting-Group-Hydro-Report-12-17-2014.pdf
  4. http://www.pub.nl.ca/applications/NLH2014Capital/NLHCBSUPP2014/TransmissionLine/application/Application-2014-04-30.pdf
  5. http://www.pub.nl.ca/applications/NLH2012Capital/files/application/NLH2012Application-VolumeII-Report10.pdf
  6. http://www.pub.nl.ca/applications/NLH2013GRA/files/rfi/PUB-NLH-016.pdf
  7. Newfoundland and Labrador Hydro 2013 Amended General Rate Application, Table 2.14, Section 2.5.1. http://www.pub.nl.ca/applications/NLH2013GRA-Amended/files/application/NLH-2013-GRA-Application-Volume-1-AMENDED-2014-11-10.pdf

Low Reservoir Levels, Really?

I'm throwing this together rather quickly, so I apologize in advance if it doesn't read well.

Des Sullivan (no relation as far as I know) has posted over at his Uncle Gnarley blog an article calling on the PUB to refuse NL Hydro's request for a rate increase to cover the increased cost of providing service to the ratepayers. This increase is, according to Hydro, due to increased use of the island's thermal generation which is needed to offset reduced hydro output which is due to reduced reservoir levels across the island.

Reservoir levels are down due to low inflows from rain and snow. According to Hydro anyway.

I'm just going to comment on the data provided, as Mr Sullivan is entitled to his opinions, no matter how much I may disagree with them.

Mr Sullivan starts out with Figure 1 below. This is pretty cool data. I didn't know it was available and you can access it from Environment Canada's website, here.

Figure 1

The blog then goes on to say:

"The following Real-Time Hydrometric Graph shows precipitation data for the area around the Car (sic) Arm Reservoir, a significant source of hydro power. The time parameters cover a period longer than the one for which Nalcor claims low precipitation. The period shown in the Graph is August 2014 to February 2016. Changes in precip are consistent with last year."

If you take the elevation data and overlay this years data with last years data they are in no way consistent as Mr Sullivan claims. I did this and include my plot here as Figure 2.

Figure 2: Year Over Year Comparison

From Figure 2 it's clear that the reservoir level at Cat Arm has dropped 26 days sooner than last year and was roughly 1.3 meters (4.3 feet) lower than last year as of January 12. The situation has not improved, specifically I looked at the elevation levels on February 1, 2015 (391.6 m) versus February 1, 2016 (390.04 m).

I don't believe you can conclude that the water levels at Cat Arm are consistent, year over year.

The next figure provided by Mr Sullivan showed the East End of Granite Lake, Figure 3. The blog included another Granite Lake level as well which I won't reproduce here, you can see it in the UG blog. Granite Lake has control structures on both ends. It receives water from Victoria Lake, which is the start of the Bay D'Espoir water supply, and then dumps it into Meelpaeg Reservoir, which is the larger storage space in the region. Hydro can manage the level of water in this lake by moving water through it from Victoria Lake and down into Meelpaeg Reservoir.


Figure 3

That region of the province did have a fair amount of rain recently and you can see the increase in water level that resulted, Figure 3. It's about 0.3 m or 1 foot which is a relatively modest increase. There is a level gauge at Victoria Lake as well. I've reproduced the data as Figure 4. This lake is down 4m (12 feet) since the summer. That's substantial.

Figure 4

There's a little blip of an increase at the end there. That would likely be as a result of the rainfall and melt in February, 2016, and shown in Figure 3.

As part of another blog post I have been writing which discusses the third line from Bay D'Espoir to the Avalon, I sent a request to Hydro for data on how much water has been spilled on the island over the past 5 years. They were good enough to send it along just today actually. The data shows that Hydro has not spilled water since June of 2015. I don't think they have wasted the water.

Finally, Hydro has indicated in its application to the PUB that Newfoundland Power and Corner Brook P&P are also seeing reduced water levels in their reservoirs. So this not just Hydro's issue. If that's not the case then I expect to see strong objections from both during the PUB process.

I had a few issues with some other items in today's Uncle Gnarley blog but will leave it at that for today.


If you see any problem with what I have hastily prepared here, you can let me know by leaving a comment or sending me a message on Twitter, @kurtbsullivan.


Edit: February 23rd: Thanks to Tom Baird @BairdTom for pointing out an error in my discussion on Figure 1. I had originally posted 392.73 m for February 1, 2015

Edit: February 24, 2016: I cleaned up the discussion on Figure 1 a bit more to make it clearer.




Thursday 4 February 2016

Our Heat Pump Energy Savings

I wrote this last year at the end of our first winter with the heat pump in service and posted it to Facebook. 

Finally had a chance to evaluate how much less energy we used to heat the house this past winter versus the winter 2013/2014 now that we have the heat pump installed. I evaluated over the period November to May in each winter.
In order to normalize the energy numbers based on weather variations I downloaded heating degree day (HDD) data from Environment Canada which accounts for weather variations.
So, the energy used per heating degree day (kwh/HDD) for this year was 19% lower than last year. In terms of $/HDD, our energy bill was 16% lower though I did not adjust for changes in the rate Newfoundland Power Inc. charges.

Heat Pumps and Capacity

This is too long for Twitter.

It's not clear whether the need for new generation capacity could be eliminated by installing heat pumps in everyone’s homes and businesses. I know from my own experience and from others who have installed them that there is a significant reduction in the energy required to heat your home over time.

From a capacity perspective, our system is getting close to capacity limited in the winter peak, hence the need for new generation. This is also when air-exchange heat pump systems, like the mini-split system I have, would be working their hardest to draw sufficient energy from the cold outside air. Some systems also have energy hungry electric furnace elements which kick in to supplement the heat pump energy when needed, again at peak times. My system does not have such an element and I believe ground based systems would be more immune to this.

Based on the above I don’t believe heat pumps will help a whole lot at peak times though they do help in the long run. There was a report from the provincial government last year on residential energy efficiency which stated this but I can’t find it anymore. If I find it I'll link it here; otherwise this just ends up being my opinion.

As for cost, my mini-split cost ~$10k installed as a retrofit onto my existing house which has baseboard heaters. Others that I know have paid roughly the same for comparable systems. Rough estimate on retrofitting say half of Newfoundland Power’s 250,000 customers would be $1.5 Billion. You can argue with me on that price, it's just off the top of my head.

Ground based systems are considerably more expensive up front and ducted systems lose a fair amount of energy in the duct making them less effective than ductless systems like mine.

Here's my system.