Wednesday, August 17, 2011

Different Strokes for Different Folks...

One aspect of being "fair" is treating people in similar circumstances "the same" but TLC's demand charging fails this test in a number of areas.

One problem with the way TLC's demand charging has been implemented to date is that load control for different customers is applied via different channels and those channels are typically controlled at somewhat different times. This is true even within a single GXP related area - "Taumarunui" for example, or more precisely, all those customers whose supply comes through the Ongarue sub station. The end result is that some of the measurements to find "peak demand" have occurred across different periods for different customers, even though they could well be neighbours.

While there is typically a fairly high degree of agreement between different channels, the small differences can still easily lead to situations where customers are not treated equitably, accentuated by the fact that the underlying TOU meter readings are only at a half hour resolution. Therefore in my own recent submission to TLC (following publication of the Sapere and NZIER reports) I have recommended that only qualifying periods of control common to all relevant channels in a single GXP area be used in any future system based on measuring demand during load controlling.

The reason is that for a customer with a demand (TOU) meter, their final demand value is determined from the periods that were fully controlled and also only those periods that were long enough. For the 2009 and 2010 winters this meant they had to be at least three hours long. The TOU meters record consumption data every half hour, aligned with clock half hours (unless the meter clock is inaccurate for some reason). It also wouldn't be fair or correct to include half hours that were only partially controlled as this would mean some controllable load (e.g. water heating for many customers) could be included which is contrary to the underlying philosophy of TLC's approach - controllable load is not meant to contribute to your demand. So this means that at least six consecutive fully controlled half hours were needed before a potential "peak demand" could be calculated by calculating the average load over each such three hour block. Only periods passing that test (and also falling in the 1 June through 30 September period) are looked at when determining a customer's single highest average demand over a three hour period.

One example is shown below, comparing two channels from the same area. What is displayed are only the qualifying load control periods for the two channels, aligned side by side by date and time so you can see at a glance where the main differences are. In other words, the blocks of time shown are only those remaining after discarding all periods of full control less than three hours long. So while there was actually a lot of other load controlling on these two channels, (including across September) none of that appears here because it didn't qualify.

The dates and times of qualifying periods for the first channel are on the left, aligned where possible with the corresponding period on the second channel on the right (with "<>" separately them). When one or the other side is blank then we know that there was either no controlling at all (probably not likely), or whatever controlling did that occur wasn't both long enough and continuous. These cases are, in my opinion, the largest problem as customers on one channel are being "measured" for that "network peak" but those on the other aren't.

Of course, even when there are corresponding qualifying periods on both channels, there are still often small differences in the start and end times. For example, on June 02, the qualifying period on the first channel shown ended at 21:00 (9 pm) but for the channel shown on the right the latest time measured was 20:30 (8:30 pm). Probably less than half match "exactly" - those are highlighted in green to make it easier for you to look at those that differ. The largest difference appears to be on July 11 where the first channel has an extra 1.5 hours for some reason, all the way through to 22:30 (10:30 pm).

When processing load control periods that have different start or end times (as per the June 02 example) the extraneous half hours would be ignored for purpose of calculating demand values (kW Load) under the modification to the rules I have proposed. That is to say, only the fully controlled half hours common to all relevant channels would be considered and that would mean all demand meter customers in a given GXP area would be measured against exactly the same set of half hours.

                            <> Jun 01 17:30 - Jun 01 21:00
Jun 02 17:30 - Jun 02 21:00 <> Jun 02 17:30 - Jun 02 20:30
Jun 03 07:30 - Jun 03 10:30 <>
Jun 03 17:30 - Jun 03 21:30 <> Jun 03 17:30 - Jun 03 21:00
Jun 04 07:30 - Jun 04 11:30 <>
Jun 04 17:30 - Jun 04 21:00 <> Jun 04 17:00 - Jun 04 21:30
Jun 05 07:30 - Jun 05 11:00 <> Jun 05 08:00 - Jun 05 11:00
Jun 05 17:30 - Jun 05 21:00 <> Jun 05 17:30 - Jun 05 20:30
Jun 08 17:00 - Jun 08 21:00 <> Jun 08 17:00 - Jun 08 21:00
Jun 16 17:30 - Jun 16 20:30 <>
Jun 17 17:30 - Jun 17 21:00 <> Jun 17 17:30 - Jun 17 21:00
Jun 18 07:30 - Jun 18 10:30 <>
Jun 18 17:30 - Jun 18 21:00 <> Jun 18 17:30 - Jun 18 21:00
Jun 19 07:30 - Jun 19 10:30 <> Jun 19 07:30 - Jun 19 10:30
Jun 19 17:00 - Jun 19 21:00 <> Jun 19 17:00 - Jun 19 21:00
Jun 22 17:30 - Jun 22 21:00 <> Jun 22 17:30 - Jun 22 21:00
Jun 23 07:30 - Jun 23 10:30 <> Jun 23 07:30 - Jun 23 10:30
Jun 23 17:00 - Jun 23 21:30 <> Jun 23 17:30 - Jun 23 21:00
Jun 24 17:30 - Jun 24 21:30 <> Jun 24 17:30 - Jun 24 21:30
Jun 25 07:30 - Jun 25 10:30 <> Jun 25 07:30 - Jun 25 10:30
Jun 25 17:00 - Jun 25 20:30 <> Jun 25 17:00 - Jun 25 20:30
Jul 01 17:30 - Jul 01 21:00 <> Jul 01 17:30 - Jul 01 21:00
Jul 02 07:30 - Jul 02 10:30 <> Jul 02 07:30 - Jul 02 10:30
Jul 02 17:00 - Jul 02 21:30 <> Jul 02 17:00 - Jul 02 21:30
Jul 03 17:30 - Jul 03 21:30 <> Jul 03 17:30 - Jul 03 21:00
Jul 06 17:00 - Jul 06 21:00 <> Jul 06 17:00 - Jul 06 21:30
Jul 07 17:30 - Jul 07 21:00 <> Jul 07 17:30 - Jul 07 21:00
Jul 08 17:00 - Jul 08 21:00 <> Jul 08 17:00 - Jul 08 21:00
Jul 09 08:00 - Jul 09 11:00 <>
Jul 09 17:30 - Jul 09 21:30 <> Jul 09 17:30 - Jul 09 21:30
Jul 10 07:30 - Jul 10 11:00 <> Jul 10 07:30 - Jul 10 10:30
Jul 10 17:00 - Jul 10 21:30 <> Jul 10 17:00 - Jul 10 21:30
Jul 11 17:00 - Jul 11 22:30 <> Jul 11 17:00 - Jul 11 21:00
Jul 13 07:30 - Jul 13 10:30 <>
Jul 13 17:30 - Jul 13 21:30 <> Jul 13 17:00 - Jul 13 21:00
Jul 14 07:30 - Jul 14 10:30 <> Jul 14 07:30 - Jul 14 10:30
Jul 14 16:30 - Jul 14 21:00 <> Jul 14 16:30 - Jul 14 21:00
Jul 15 17:30 - Jul 15 21:00 <> Jul 15 17:00 - Jul 15 20:30
Jul 16 17:30 - Jul 16 20:30 <> Jul 16 17:30 - Jul 16 21:00
Jul 21 17:00 - Jul 21 20:30 <> Jul 21 17:30 - Jul 21 20:30
Jul 22 17:30 - Jul 22 20:30 <> Jul 22 17:30 - Jul 22 20:30
Jul 23 17:00 - Jul 23 20:30 <> Jul 23 17:00 - Jul 23 20:00
Jul 24 17:30 - Jul 24 20:30 <> Jul 24 17:30 - Jul 24 20:30
Jul 27 07:30 - Jul 27 10:30 <> Jul 27 07:30 - Jul 27 10:30
Jul 27 17:00 - Jul 27 21:00 <> Jul 27 17:30 - Jul 27 21:00
Jul 28 07:30 - Jul 28 10:30 <> Jul 28 07:30 - Jul 28 10:30
Jul 28 17:00 - Jul 28 20:30 <> Jul 28 17:00 - Jul 28 20:30
Jul 29 17:30 - Jul 29 20:30 <> Jul 29 17:30 - Jul 29 21:00
Jul 30 07:30 - Jul 30 11:00 <> Jul 30 07:30 - Jul 30 10:30
Jul 30 17:30 - Jul 30 20:30 <> Jul 30 17:30 - Jul 30 21:00
                            <> Jul 31 07:30 - Jul 31 10:30
Jul 31 17:30 - Jul 31 20:30 <>
Aug 03 17:30 - Aug 03 20:30 <> Aug 03 17:30 - Aug 03 21:00
Aug 04 17:00 - Aug 04 20:30 <> Aug 04 17:00 - Aug 04 20:30
                            <> Aug 05 07:30 - Aug 05 10:30
Aug 05 17:30 - Aug 05 21:00 <> Aug 05 17:30 - Aug 05 20:30
Aug 07 07:30 - Aug 07 10:30 <>
Aug 07 17:30 - Aug 07 21:00 <> Aug 07 17:30 - Aug 07 21:00
Aug 10 07:30 - Aug 10 10:30 <>
Aug 10 17:00 - Aug 10 20:30 <> Aug 10 17:00 - Aug 10 21:00
Aug 11 17:30 - Aug 11 20:30 <>
Aug 11 17:30 - Aug 11 21:00
Aug 12 17:30 - Aug 12 20:30 <>
                            <> Aug 13 17:30 - Aug 13 20:30
Aug 19 17:30 - Aug 19 20:30 <>
Aug 20 18:00 - Aug 20 21:00 <>
Aug 24 18:00 - Aug 24 21:00 <>
Aug 25 17:30 - Aug 25 21:00 <>
Aug 27 17:30 - Aug 27 20:30 <>

Friday, June 24, 2011

TLC offtake from Tokaanu and Ohakune Grid Exit Points over winter 2010

For those who like graphs, click on the links to download a full set of daily graphs showing TLC's offtake at the Tokaanu (Turangi area) and Ohakune GXP's for the 1 June through 30 September 2010 period.

The half hour data is extracted from the Electricity Authority's Central Dataset. As noted in my previous post I can't guarantee the data used is correct or that I haven't made errors, so please bear that in mind before using these graphs for anything important. Use at your own risk!

One interesting example is shown below, showing the 23 September for Tokaanu. I believe there was a major outage late in the morning that caused this unusual situation. This meant TLC were load controlling more or less continuously from that point onwards for the next 8 or 9 hours - presumably to minimise the amount of power they needed to redirect to or from an alternative source (National Park or Kurutau perhaps?) until the problem was fixed.

[Added: 26 June] Seems the outage behind this event was more to do with the National Park GXP. See the graph for TLC's offtake from the National Park GXP on the same date below.

Don't know about you but if I was a TLC customer with a demand meter and my peak 3 hour demand came from a completely unexpected and "out of bounds" period of load controlling such as this one in the afternoon then I'd be less than happy. If you want to check, it's easy enough to do now that TLC have (from late May) made individual demand meter customer graphs available on their site. Dig out your ICP code and click here to see your graph!

Tuesday, May 3, 2011

I have disputed my April invoice from TLC...

My peak demand (kW load) reduced in the winter of 2010 as compared to the winter of 2009. Many other TLC customers - most probably some thousands of them - will be in the same boat.

TLC have previously clearly indicated that new demand charges take effect each April and stay in place for a full year.

Therefore I think my reduced lines charges should have effect from 1 April - not 1 May 2011 as TLC seem to think is acceptable. It's only a difference of $9.53 in my case and for one month but there's a principle at stake as far as I'm concerned. Obviously, to be fair, any remedy should also generally apply to all other TLC customers in a similar situation.

I've made my position clear to TLC - this sort of thing just isn't acceptable - and didn't pay the extra $9.53. Maybe not the best way to "win friends and influence people" but this really offends me.

Here's the earliest statement from TLC that I've found with their position on this - it's the 6 December 2006 letter from John Anderson to all TLC customers, describing the new system (then called "FairPay", which is kind of ironic really). In particular, read item 2 in the numbered list of "key points". (You can click on the image to get a larger version - I've smudged out most of the address details and the red date/time at upper right is from my camera.)

Seems pretty clear to me and there's more in the same vein if you care to read through the various newsletters and other documents that have come out of TLC since that time.

[Update: 10 June 2011] TLC's board has decided to credit all customers affected by the extra month at the old rate according to TLC's Customer Services Manager David Hearn. I was informed via email on Tuesday morning this week and the message was repeated at TLC's presentation in Kakahi in the evening. The credit is to be applied with the June invoices, and all affected customers will receive a letter. The total of the credits applied was $180,000 according to David Hearn at the Kakahi meeting. I've since received my June invoice with an adjustment shown, but no letter has arrived yet.

Wednesday, April 27, 2011

This would be nice!

I really like the information that Orion publish on their site about their load controlling. Lots of information, updated every five minutes, plus options to get text messages and emails informing you when load control starts and stops. A big thumbs up from me.

Here's just one example of the kind of information you can get from their site:

Orion are the equivalent to TLC in the central Canterbury area. It's almost enough to make me move down there. John, Roger, are you listening? :-)

Wednesday, April 13, 2011

TLC's Peak 2010 Winter Load in Ohakune

I've been playing around with half hour GXP data provided by the Electricity Authority (which replaced the Electricity Commission in November 2010). Thank you EA! You can visit their site here:

The data came with a disclaimer that they can't guarantee it's 100% correct - and I could also have made errors when processing it. [Update: 27 April - I DID make an error and the first version of the graphs that I posted had demand value labels that were all double what they should have been. Sorry about that - all fixed now! The "shape" was correct - just the labelling on the y-axes was wrong.]

Bearing in mind those caveats (I make no guarantees whatsoever - this is provided just for interest), here in graphical form are some of TLC's highest three hour "kW load" values from Ohakune. Note the values are actually megawatts (MW), 1000 times larger than the usual kilowatts (kW) that apply to domestic installations and businesses.

Not surprisingly, these three peaks are all in the July school holidays so no doubt Ohakune had a few extras in town over that period.

The Electricity Authority also provide a web interface to some of their historical data (which they call the Central Dataset, or CDS). That's also pretty useful if you're don't need the detailed data. The web site version is aggregated to day or month for example, while the more detailed half hour data comes on a DVD and needs quite a bit of effort (plus the right kind of background/skills) to make much sense of.

[27 April 2011: For the more adventurous, half hour demand data (by Grid Exit Point such as Ohakune - code OKN0111) is also available from the "demand" section of the WITS "Free to Air" site. This data is updated daily and doesn't require sending away to the Electricity Authority for a DVD, although it doesn't show how the offtake from Ohakune is split between TLC and PowerCo, so is slightly less useful in that respect.]

Monday, April 11, 2011

Processing my "kW Load"

My kW load (demand) figure for the next 12 months is 0.891 kW (rounded by TLC to 0.89 kW for billing purposes).

That's fairly low as these things go but to achieve that I had to be very careful, not to mention choosy about who I let come and stay with me over that the critical period. Heaters and that kind of thing are all locked away out of reach!

On a more serious note it probably also helped that there were only three relevant periods of load controlling for my meter last winter. The number of qualifying measurement periods was a lot higher (43) in the previous year and my demand from that period was also correspondingly higher at 1.135 kW.

By comparison, "the formula" (2011 version at least) comes up with a 1.84 kW load based on my 1 June to 30 September 2010 consumption. However the previous year's formula (based on higher winter consumption from the preceding winter) actually produced a lower demand than that at 1.70 kW.

Yes, you read that right!

If I had been on the formula, I would have been another of TLC's customers who reduced their average daily consumption in the winter of 2010 as compared to the previous winter but who still ended up with an increased kW load estimate. The reverse kind of silliness can also happen at the higher end where you can still be "rewarded" with a lower kW load value despite actually increasing your average daily consumption in the 2010 winter. This can happen because of the way "the formula" has been changed this year. Doesn't really make sense does it? However I will definitely be interested to hear from TLC if it makes more cents to them!

On the other hand, if my peak load came via the formula I'd be a lot more relaxed about the occasional blow-out and visitors. In fact I could go completely "crazy" with my electricity usage in the middle of a big fat load controlling period and it wouldn't make diddly squat difference to my formula based demand/kW load (as long I also didn't do that too often).

Anyway, here's a graph showing the half hour data for the full day from where this 0.89 kW peak 3 hour load (also during a load controlling period, June through September 2010) came from.

Each bar represents a half hour period - one recording from the meter. The green bars are those that were completely "uncontrolled" so these really have nothing to do with any peak kW load calculations. The orange bars are those where load controlling started or stopped partway through. These don't get used either, otherwise some "controllable load" from your water cylinder heating and similar could be included and that's not meant to happen.

You may have noticed a period of high load just after midnight. Okay, so you're curious. Turns out I'm a night owl and usually shower just before going to bed. So, my hot water cylinder then has to do its thing for a few hours to heat up the very cold water that will have just flowed in. You may also note some fairly regular "spikes" across the day. Yep, it's that inefficient old water cylinder again, still trying to keep the water nice and hot because a fair bit of the heat has leaked out into my clothes drying cupboard instead!

The red bars indicate the "danger zone" as these are the times that fell completely within a load controlled period (and also meaning in my case that my hot water cylinder should be turned off). The real "danger" is when there are six of those in a row as that means there is a three hour long block that will be processed to get an average load. The six half hours that correspond to my single highest three hour average load (kW) are overlaid with a larger shaded bar and the dotted horizontal line level with the top of that shows the average load over that three hour period, 0.89 kW in this case.

Note also how the times on the x-axis start one minute later than what you might expect. That's because a one minute clock correction was applied to the "raw data" from the meter. If you want to know more there's more on this and other "technical details" below.

Following this paragraph you can read the section of the "log file" generated by my software as it processed the data from my meter - the only change is that I have removed my ICP number for privacy reasons. The log file is generated mainly for diagnostic purposes but it's also a good way to illustrate some of what goes on "underneath the hood" if you happen to be interested in that kind of thing.

Processing meter 2902959
Meter 2902959 (controlled) is assigned channel 6 which maps to 156 using ../ongarue-map.csv
We will allow for a maximum time error of 5 minutes.
Discarding data before a pre-winter clock adjustment (2010-06-01 12:09:00 -> 2010-06-01 12:11:00) for meter 2902959.
Time adjustment +0:01:00 (2010-12-02 13:22:00 -> 2010-12-02 13:23:00) will be applied for meter 2902959.
Ignoring "On" without a matching "Off" at 2010-01-04 06:07:35
Ignoring "On" without a matching "Off" at 2010-03-20 10:31:20
Ignoring "On" without a matching "Off" at 2010-04-21 09:02:26
Ignoring "On" without a matching "Off" at 2010-04-21 14:50:32
Ignoring "On" without a matching "Off" at 2010-05-12 11:48:29
Ignoring "On" without a matching "Off" at 2010-05-12 12:00:06
Ignoring "On" without a matching "Off" at 2010-06-16 00:32:11
Ignoring "On" without a matching "Off" at 2010-07-20 00:37:56
378 controlled periods read from relay times file (../control/cntrl-156.csv)
Using raw controlled load times:
Longest controlled period was 4:08:19
Total controlled time was 12 days, 11:39:31
There are 12 qualifying controlled periods
After clipping:
Longest clipped controlled period was 3:30:00
Total clipped controlled time was 5 days, 1:30:00
There are 3 qualifying clipped controlled periods
13524 records used from usage file (2902959.csv)
1 records skipped (power factor data, etc.)
Adding one hour to times for the week starting 2010-09-26 02:00:00 for meter 2902959.
Subtracting one hour from records for the week starting 2010-04-04 02:00:00 for meter 2902959.
Total usage is 2658.767 kWh
Winter usage is 1844.162 kWh (load detected on 122 different dates)
Incomplete data! (Covers 2010-06-01 12:31:00 to 2010-12-02 11:31:00)
243 controlled intervals, resulting in 5 blocks of time for demand calculations
Max winter 30 minute demand ended at 2010-08-26 19:01:00 and was 3.406 kW.
Max 24/7 winter 180 minute demand ended at 2010-06-12 03:31:00 and was 1.849 kW.
Max controlled winter 180 minute demand ended at 2010-07-26 20:31:00 and was 0.891 kW.
Min controlled winter demand value : 0.369
Median controlled winter demand value: 0.738
Mean controlled winter demand value : 0.666
Standard deviation of winter demands : 0.219

Sorted list of top non-overlapping demand figures:
2010-07-26 20:31:00: 0.891 kW (180 minutes from 2010-07-26 17:31:00)
2010-07-13 20:31:00: 0.738 kW (180 minutes from 2010-07-13 17:31:00)
2010-07-14 21:01:00: 0.369 kW (180 minutes from 2010-07-14 18:01:00)

There were only three times (last winter, in my area, on my channel) where there was a period suitable for measurement according to the rules being applied and the highest average three hour load from each of those appears at the end of log file entries above.

Now, you may want to check that my working is correct! Actually I really do hope so because I don't want to be accusing of "diddling the data", "feathering my own nest", or anything else in that vein! The more eyes that look the more likely it is that any bugs or mistakes will be found. Please let me know if you can't manage to reproduce my results and think I'm not on the level!

The data is in CSV files and is fairly straightforward to understand so it's quite possible to "manually" check if you're reasonably familiar with a suitable spreadsheet program or similar. Even without that you could open the files in any text editor (e.g. Notepad) and at least check the date/times and kW load values listed above are correct with a few simple calculations.

There are some things to bear in mind though:

1. There are two files, 2902959.csv contains the half hour data and cntrl-156.csv contains the times my ripple control channel was turned off and on according to TLC's systems. There are some "On" statements without a matching "Off". These are apparently caused by resets of some kind and are ignored by my processing software (apart from spitting out a warning to the log file).

2. The dates and times in the half hour data file from the meter refer to the end time of the relevant half hour. There are a number of fields in each record but the only ones that really matter are the date, time, and kWh consumption value.

3. Someone from TLC came and checked the clock on my meter on 1 June 2010 and made a small adjustment. (My records show it was about 25-30 seconds slow before this adjustment and about 14 seconds fast afterwards. This doesn't actually match very well with the clock adjustment information I got from TLC but in any case data from before that point is discarded if it fell in the 1 June to 30 September period. This applies to all cases where such an adjustment was made late (the checks were meant to be done before winter) but it didn't make any difference to my results anyway.

4. The clock in my meter was checked (and perhaps adjusted again) when the data was downloaded (on 2 Dec 2010 in my case). My own records show no adjustment was actually made at all although the information received from TLC later indicated the meter clock was adjusted forward by one minute. I believe this small discrepancy was due to the way the person checking and recording the information did it - including only recording times to a one minute precision as displayed on the meter. (I watched over his shoulder. :-)

5. In any case the processing also allows for a five minute residual error in the times so small remaining inaccuracies in the various clocks shouldn't have any effect. The same five minute allowance was applied to all processing of 2010 winter data from TOU meters. Effectively this means the start and end times of each controlled period are trimmed back by five minutes to create a small "safety buffer".

6. As noted in my last post, my meter is also one of the many that has incorrect daylight saving information programmed into it and this means that in 2010 (and this year also) it transitioned to and from daylight saving time a week later than it was supposed to. The dates and times from the meter (and in the load control switching data) are all in "local time" so therefore a correction has to be made for the one week following the official start and end of daylight saving. This didn't make any difference in my case but it could in others. If you look at the half hour data in the early morning of 11 April and 3 October you will see the meter doing its version of daylight saving time adjustments (albeit a week late). Stupid smart meter! :-)

7. After trimming the five minutes from the start and end times of controlled periods, correcting for the daylight saving glitch and applying any clock adjustment recorded from after the winter to all the remaining times in the data, the processing then looks at all the three hour periods (six consecutive half hour blocks from the corrected meter data) that fall completely within a single period of load controlling. For each it adds up all the consumption data (kWh) and divides by three (duration in hours) to get an average load (kW) over that period. The highest such period is where my 0.891 kW figure comes from.

Thursday, April 7, 2011

What does a "Time Of Use" meter" look like anyway?

A Time Of Use meter is a TOU meter, is a demand meter, is a half-hour meter, is an interval meter...

They come in different shapes and sizes but the one at my house is a Landis+Gyr EM1000 model, installed on 13 May 2009 just before 2pm. (I know this because I have the data that was downloaded later and that is the end time of the first half hour with a normal numeric consumption value - before that point they are all just hyphens, which is what appears in the data I get for these meters when they don't have power, including during outages when some kind of internal backup battery seems to keep the internal clock ticking over.)

Here it is earlier today, on the 7th of April.

Note how the meter is showing the time (in HH MM format on it's LCD at the top left) but is an hour "fast" compared to the camera time at bottom left - which I know is accurate because I set it myself! This difference is because my particular meter is one of the many from earlier batches that was incorrectly configured in terms of the daylight saving dates and so won't switch over to standard time this year until at least next Sunday! This is one of a number of date and time issues that need to be considered when the data from these meters is processed. Their internal clocks are also not automatically updated or synchronised with any external time source. This can be important as the internal clock also determines the date/time stamp associated with each half hour data record stored by the meter and that has to be  reconciled with the load control times from TLC later on. The EM1000 is meant to be able to store 284 days of data and after that the oldest data is discarded or overwritten as required.

The EM1000 model is the most basic model or TOU meter, only suitable for smaller single phase installations as far as I know. TLC currently also use other later (and perhaps more capable) models. I believe these include the EM51000 and EM5300. Click here for more technical details! All of these have a unique meter ID - that's the 2902959 number in the picture above. Then there is another smaller group of older TOU meters that I don't have manufacturer or model numbers for but as far as I know they all have eight digit meters numbers that start with the digit '9'.

The EM1000 has a LCD display that cycles automatically between the date (DDMMYYYY format, or 07042011 for today), time (HH MM), total day ("A" - 6am to midnight, as far as I know) consumption in kWh and total night ("B" - midnight to 6am) consumption. Pressing the big red "scroll" button lets you flick through those four display modes more quickly.

To the left of the scroll button is a small LED (red when lit) that basically flashes at a rate that indicates the "demand" your installation is generating at that time. As the labelling says, one flash means you've used 1 watt hour (Wh, 1/1000th of a kWh or "unit"). This means if your home is using a constant 1 kW of power then that LED will flash steadily at a rate of 1000 times per hour, once every 3.6 seconds, or 10 times in 36 seconds. Therefore I can test how much kW load (aka demand) my house is generating by dividing 36 by how long 10 flashes take (in seconds). If ten flashes take about 10 seconds - but remember to count from zero so you include 10 gaps also - then the average load or demand over that period was about 3.6 kW. Other meters might have different details and some models might even show you the load or demand on the LCD but you can usually use your meter to find out the current load by one method or another.

It was actually quite a bit easier to measure the demand on my old style total consumption meter with a rotary dial (as I used to have before my first "demand meter" was installed, but that's another story all on it's own).

Below the flashing LED is a strange looking roughly circular plate with some smaller round structures within that in an upside down V arrangement. All very mysterious and the whole thing reminds me vaguely of those crop circles the aliens leave behind, but instead that's actually where the optical reader device "connects" when the half hour data is downloaded or the meter needs to be reconfigured in some way.

One thing this this meter knows nothing about though is the load control signals arriving at my house. That job goes to the "ripple control receiver" instead, a completely different device in my meter box. Here's what my one looks like:

At the lower left you can see three panels with some strange plus signs and numbers written vertically besides positions 1:, 2: and 3:. These correspond to the three spaces for actual "relays" immediately to the right behind the transparent plastic with screw and orange anti-tamper seal. My one only has one relay installed, in the leftmost position, corresponding to the leftmost position:code label on the left, 1:+101+6. This apparently means that relay is on channel 6 and as the label indicates is controlling the power supply to my hot water cylinder.

The green switch on the relay should tell us whether TLC is load controlling on the relevant channel or not. The up position is "On", meaning power is available to my hot water cylinder, and the down position is "Off", meaning TLC are load controlling and my hot water is gradually becoming cold water again!

There's a saying, however, that what can go wrong, will go wrong!

So what could go wrong with all of this? Well, for starters, perhaps the meter's internal clock will do something weird. Perhaps it won't handle daylight saving transitions properly! Perhaps the backup power system for the meter clock will fail if it has no external power for a long enough period and then the clock will be out. Perhaps some power surge or other external event will cause it to lose data or whatever. Perhaps the ripple control receiver will fail or not detect some control signal properly meaning it is on when it should be off or off when it should be on. Perhaps the channel information recorded by TLC doesn't match what is actually in your meter box. Murphy is everywhere and I like to go looking for him!

So, I carefully record various details every so often, meaning I can do some checks against the data downloaded from the meter when that eventually happens and also against the load control timing information that TLC supply me. Some of us are just born this way and there's really no point in trying to change it!

I record the true date and time using my very accurately set wristwatch. I check the meter date display is correct and then I cycle rapidly through the different modes on the meter until the time display flips over to the next minute so I can record how far out the meter clock is when compared to my watch, to a precision of a second or two. (Since TLC checked and adjusted it in June last year it has held good time, being a more or less constant 14 seconds fast.) I also record the two displays of total units used, A (for day) and B (for night). Finally I note whether the green switch on the ripple control receiver is in the up or down position. And if it's down, I try to avoid using high demand appliances because this could be one of those periods of load controlling that turns out to be long enough to contribute to the "kW load" or "demand" calculations later on!

Other people buy "centameters" and "switchits" and other such newfangled hi-tech gadgets, but I like to go out in the rain on cold, dark and windy nights, trying to see how fast the red light is flashing and which way the little green switch is pointing... :-)

Tuesday, April 5, 2011

The 2011 formula (using data from the 2010 winter)

Recently TLC sent out letters with details of the new charges to be levied on each of their customers and they started arriving in mailboxes on or about the 2nd of April. The information in these letters that I am most interested in talking about is the "kW load" - in other words, the measured (if you have a TOU meter) or estimated highest average load (kW) that each home or business generated over a three hour period while TLC were also load controlling on the relevant channel.

TLC pointed out in the letter that there was an online calculator on their web site for checking your new kW load value or to see how it would change with different inputs (different meter readings). Unfortunately it seems they stuffed up rather badly in this respect because up until last evening (Monday 4th April) the online calculator was actually using the wrong formula - last year's version. I know this because I watched my brother enter his readings from the letter and get a much lower value than what was printed on the letter. Oops! Anyway, this seems to have been corrected by early today (5 April) even though there is nothing on the page (or anywhere else on TLC's site as far as I can see) to indicate there was any error to start with or that a correction has been made. Hmmm - can't say I like that way of doing things. So, if you happened to try to check earlier than today (5 April) you may have been wondering what was going on!

After discovering this discrepancy between letters and the online calculator I did some checking of my own and initially created a spreadsheet that duplicated the 2010 formula just to work out what was going on. Then I got a bit carried away and expanded that to include the different formulae used over the last three years. You can download it here if you want to play with it but please note you do this at your own risk. I make no promises whatsoever about its accuracy or correctness or suitability for anything you may try to do with it. Also, note that the 2011 version of the formula used in the first version of the spreadsheet was only something I "reverse engineered" from a sample of results generated from the updated online calculator earlier today - TLC had not published the precise details of that formula when I looked. Still, I'm confident I got it right and all the results agreed with the online calculator - at least the last time I used it!

[6 April update: TLC published the 2011 formula around the same time I was writing the first version of this post, so I've now updated my spreadsheet to use the official version. My reverse engineered formula produced practically identical output but was nevertheless very slightly different. The graph below has also been updated but the only visible change should be the different colours. Red is now 2011 to make it stand out a bit more.]

When you look at the graph below (this and others are generated by the spreadsheet) you can see the latest version of the formula generates significantly higher kW load estimates for the lower end of the scale (as compared to the 2010 formula) but somewhat lower estimates at the other end. The cross-over appears to be at about 2.39 kW of peak load which (coincidentally?) happens to be quite close to what TLC have previously indicated is an average domestic installation's three hour peak load. Anyway, the changes from last year's version mean that even if your average daily consumption over the winter of 2010 was identical to that in the winter of 2009, then the new formula would generally produce a quite different kW load estimate, sometimes higher and sometimes lower, and therefore also changing the amount TLC are asking you to pay. The only situation where your payment for "kW load" wouldn't change would be if you happened to fall pretty much right on the pivot point.

The pivot (cross-over) point corresponds to a 92 day uncontrolled consumption of 1600 kWh (or "units") which in turn is about 17.4 kWh (uncontrolled) per day on average. (If you're wondering where that reference to 92 days come from, it's because the formulae seem to have always been constructed with a 92 day uncontrolled usage value as the main "input" and I'm guessing that is because there are 92 days from the start of June to the end of August).

Remember however that this is also all "uncontrolled consumption", so if your home has a ripple control then it actually corresponds to somewhat higher total daily consumption than that. This is because of an earlier factor that is applied to total consumption readings from "mixed" and "limited off peak" meters in an attempt to make an allowance for power used for water heating and other controllable load (because the main formula always expects only uncontrollable load as input).

This is probably clearer if you look at the working for the lead-up to the final formula calculation in the spreadsheet. You will see that 76% of consumption from mixed meters is deemed to be "uncontrolled" (maybe "uncontrollable" is a better word) in the northern areas, 65% in the southern areas, and a fixed 55% is applied to all "limited off peak" meter readings. Yeah, yeah, I know.... too much information! Actually, I must admit never quite got to the bottom of exactly how these percentages were obtained/derived by TLC but I get the general idea. The most simple case is when you have no ripple control receiver and then 100% of your consumption is obviously "uncontrolled" (and also "uncontrollable")!

How do you read this? Well, let's say your "92 day uncontrolled consumption" was about 500 kWh (which is pretty low, but this is just for an example). Looking up from that value on the x-axis (lower left) we first get to the yellow line for the 2009 formula showing that would have estimated your three hour demand as about 0.9 kW. Going up a bit further, the dark blue line shows the 2010 formula says about 1.0 kW. Finally the latest formula for 2011 looks like about 1.4 kW. (Actually I just cheated and checked my spreadsheet so I know the precise values are 0.88 kW, 1.00 kW, and 1.37 kW respectively.)

Introduction and some background


A quick introduction might be a good place to start.

First, this is a probably a good time to point out that everything I write in this blog is my own opinion and does not represent The Lines Company's official position in any way. Talk to The Lines Company (TLC) to get their version or opinion. It may be quite different from mine in many cases. I am an independent software developer/consultant who has been doing some data processing for them (since December 2009) but am not an employee. I'm also a TLC customer.

Of course I will also try to get my facts straight, but no doubt I will sometimes be wrong or make a mistake and in other cases what I write may be somewhat simplified if only to avoid writing a small book. If I find an error or someone points out a mistake then I will try to correct it and make a note to show that has happened when it's something significant.

So, here's some background info in case you aren't familiar with the situation.

"The Lines Company" (henceforth TLC) is the electricity distribution company in the area where I live (in the central North Island of NZ). They are currently the only lines company in New Zealand that bill their customers directly and also have a unique (and rather controversial method) of determining some of their charges.

Changes are likely in the future but for now the official methodology is still based on the idea that a significant portion of each customer's monthly bill will based on their maximum single average three hour load (or demand) in kilowatts (kW) during a period when TLC's network was being "load controlled" . These charges (proportional to that single peak value) are fixed for a year and come from consumption data of one kind or another collected over the previous winter period.

Currently only a small percentage of TLC's customers have a meter than can actually directly measure the load (or "demand") they impose on the system at any particular time. I am one of them. The vast majority still have old style meters that only track total usage, so for all of these some kind of estimate has to used.

Load control is also known as "ripple control". TLC have equipment that can send out signals across relevant parts of their network and if your home has a ripple control receiver (sometimes also referred to as a "relay") that is on the appropriate channel then this would typically turn off the power to your hot water cylinder (and perhaps other things also such as underfloor heating). Then at some later stage TLC send out another signal to turn the power to those appliances back on. Basically it's a remote control on/off switch to some appliances or parts of the electrical system in your home or business. Not all homes or business have a ripple control receiver installed. If not, they are "uncontrolled". TLC are trying to encourage their customers to put as much of their load as is practical onto a a controlled circuit as this means they can more effectively reduce load on their network at peak times. This is also called "load shedding". It can reduce the charges levied on them by Transpower (who run the national grid) and it can also help them defer upgrades.

There are typically a number of different channels used in each area for load controlling. They can be turned on and off independently so there's no guarantee that any two homes or businesses will always be controlled at the same time. However, as far as I know, when the network load is particularly heavy then it is quite likely that all or most channels will end up being controlled together.

Now, if you are a TLC customer (apart from a few larger customers that have negotiated contracts directly with TLC) then you either have a time of use (TOU) meter (sometimes also known as a "demand meter") or else you or have one or more older types of meter that only record total power used ("units" or kWh). There are also different models of TOU meter - not everybody has the same thing.

A TOU meter records how much power is used over each clock half hour (according to their own internal clock), say from 6pm to 6:30pm, and then from 6:30pm to 7:00pm, etc. The model I have at my house (EM1000) has enough memory to store a little over nine months of half hourly consumption data. These meters typically have a LCD display that shows some information but not (yet) anything like your highest three hour demand. To get that kind information, first the half hour data has to be downloaded from the meter (to a laptop or special hand-held reader device) and then processed later.

The older, non TOU meters are like a simple car odometer. They just track your total usage. They can be used to determine an average kW load or demand over some particular period but to do that you'd typically have to take a reading at the start of the period and then again three hours later, before finally subtracting the first value from the second, and dividing by three (or however many hours elapsed between your readings). That kind of approach is possible if you are keen enough to do it, but it isn't a practical method for TLC when they have (currently) something like 20,000 plus customers still using that kind of meter - plus of course their meter readers probably won't want to be lurking around outside your house for three hours on some cold winter's night just to get such a reading!

So, how does TLC work out a load or demand value for each customer?

If you have a TOU meter (hopefully you know this but it's also possible that you don't!) then in the months following each winter TLC sends out meter readers to download all the half hour data. Eventually that data finds its way to me (for now at least) as I process that data for TLC as an independent contractor. TLC also provide me with additional information required to do that, including the date and time their records show each ripple control channel was being "controlled" over the relevant period (the last winter) plus the load control channel relevant to each installation. Basically my job is to use the periods of load controlling for the channel corresponding to the relay in your house (but only if they are long enough, meaning they will cover at least six consecutive full half hours of data from your meter) and calculate the average load over the three hour period covered by those half hours. This is done for all the qualifying periods of load controlling falling in the 1 June to 30 September winter period. Finally the highest single three hour load is found from that list and that is the "load" (in kW) you should see on your invoice from TLC for the following year (starting in April in theory). One last thing - "uncontrolled installations" (without any ripple control receiver) are treated as if they are on channel 1 when it comes to determining their peak three hour load. So the processing I do looks at the half hour data from those installations at the times that "channel 1" is being controlled i the relevant area. There are some other complications but that's the basic story for TOU meters.

If you don't have a TOU meter then TLC are forced to estimate (and mostly it's a pretty "rough estimate") your maximum three hour load during a controlled period. In other words, it's meant to be their best shot at estimating the peak average three hour load your home or business would produce if it actually had a real TOU meter installed. There are many many potential problems with getting such an estimate because there is very little hard information to work from and many variables. But basically, TLC (or their statistician) determine a "formula" which attempts to produce such an estimate. Once determined, the input to the formula for each home or business is essentially their average daily power consumption over some (hopefully typical) months in or around the peak winter period. That input comes from the normal readings taken recorded by your electricity retailer (King Country Energy and others).

Currently "the formula" is usually updated (perhaps "recreated" is a better description) each year so there isn't really just one. TLC have a sample of homes (only 150 of them which is a pretty small number in my opinion) and currently all from the Waitomo/Otorohanga area that provides the data to do this. Measurements are made in these homes of their total uncontrolled power consumption over some months of the winter plus their corresponding half hour data during load controlling, and then a curve is fitted to that data. Basically the formula for that curve is "the formula" that is then applied to the other 20,000 odd customers who don't have TOU meters installed (yet). For each of these, the formula takes their average daily "uncontrolled" consumption (over some months) as input and outputs an estimated peak three hour load. Here's the graph from TLC's 2010 methodology paper that shows the data from this sample and the curve that has been fitted to it:

So, cutting to the chase, currently something like 95% of TLC's customers have their demand or load values estimated via one of the several variations of the formula. Oops. I forgot to mention there is also a different version for dairy farm milking sheds (and that they are "measured" over the September through December period instead of the standard June through September winter period). TLC does this because dairy farms apparently tend to generate their maximum load over this later period. Of course, this made me wonder about sheep farms (shearing?) and any number of other kinds of businesses with different peak periods but I can only presume TLC doesn't single these other groups out because they don't currently have the same kind of identifiable impact on TLC's network as a whole.

Finally I should also point out that TLC's whole network is actually generally viewed as five or six separate regions when it comes to the details of billing and load controlling and possibly other technical stuff that I know very little or nothing about. These mostly correspond to certain "Grid Exit Points" (GXPs) and are: Waitomo/Otorohanga (Hangatiki GXP), around Taumarunui (Ongarue GXP), National Park and Ohakune (each with their own GXP's), Turangi (Tokaanu GXP) and Whakamaru/Arohena (which has some more complicated set-up that I think we can safely ignore for now). A Grid Exit Point is a substation where Transpower provides a connection from the main grid to one or more lines companies in that area so they can then distribute the electricity across their own smaller networks to people like you and me.

Phew. That's a bit longer than I thought it would be when I started and definitely enough for now. I hope at least some of it makes sense!