Basslink Data Prior to Outage

There have been some recent reports in the media about the possibility that Basslink was operated too close to its physical limitations, so I’ve looked at the 5-minute MW flow data across Basslink on the weeks leading up to the outage on 20th December 2015, to see if they shed any light on the matter.

First of all here is a figure from a previous post, showing the daily minimum, mean and maximum number of MW generated in Tasmania since May 2015:

The typical demand cycle in Tasmania can be seen in the data from late December 2015, a period when Tasmania generated all its own electricity. In the months leading up to the loss of Basslink Tasmania was a major importer of electricity, but there were times when supply came close to meeting (or exceeding) demand, and one period just before the Basslink outage when supply greatly exceeded demand.

The following figure shows Basslink flow data for 1st to 21st December 2015 at 5-minute resolution:

Summing Tasmanian generation over all generators gives the local supply, and subtracting Basslink flow gives demand, both shown in the following figure:

On most days of the period shown above Tasmania maintained a fairly steady supply that was consistently around 450 MW short of demand, with that shortfall coming in via Basslink. In the early days of the month there were brief periods, usually at times of peak demand, when local supply came close to meeting demand. A possible explanation for these supply peaks was Tasmania avoiding spikes in mainland wholesale prices at times of mainland peak demand.

Something different happened at the times of the afternoon/evening peak demand on 15th – 19th December, when Tasmanian supply greatly exceeded its demand, and Basslink flow reversed, with Tasmania becoming a major exporter. A likely explanation for this was the attractive prices on offer during periods of extreme demand experienced on the Southern mainland during a summer heatwave, see the figure below for South Australian demand from a previous post:

Discussion

Basslink was operated near its electrical limits in December 2015, but those limits were not obviously exceeded, though there were very rapid transitions between high import and high export.

Another comment to make is that at that time Hydro Tasmania was not too worried about the drought that Tasmania had suffered since September 2015, all available hydro generators were run at near maximum capacity to obtain those lucrative exports of electricity, maybe giving an indication of where the balance of power lay between the hydrologists and the economists. Maybe those in charge should have been less guided by wholesale prices and more guided by the rapid depletion of the already low major hydro reservoirs at the end of 2015:

The Return of Basslink Part 2

The previous post looked at the electricity data from Tasmania on the first full working day after the restoration of Basslink. This post looks at the electricity data for the first full working week of the newly restored Basslink.

The following figure shows total supply (blue), demand (red dots, obtained by subtracting the MW flow of Basslink from total supply), the outputs of the key hydro basins (those with large but depleted water storage reservoirs), and total wind power:

The restoration of Basslink can be seen on the first day (Monday 13th June), after which Tasmania sustained a modest export of “surplus” electricity, in fact it was so determined to export something, and/or to avoid importing, that the Gordon hydro basin (amongst others) was used for brief periods to meet (and sometimes exceed) demand peaks.

The following electrical characteristic of Basslink may be one reason why Hydro Tasmania is reluctant to switch backwards and forwards between import and export:

“Due to the minimum hold on current characteristic of the thyristor based converters the Interconnector has a minimum power flow threshold of 40 MW and the submarine cable requires a 2 minute deionisation time when reversing power flow.”

Source of above: http://www.basslink.com.au/basslink-interconnector/operations/

This issue with low and reversing Basslink flow may partly explain the large jumps between import and export flows seen in the data shown below, and in other days looked at in 2015.

The use of the Gordon hydro generators in the current “wet” season is a bit worrying, as their associated lakes still have highly depleted (though growing) reserves, but so far the use of Gordon has been very modest compared with what happened a year ago, a representative example of which is shown again below for 18th June 2015:

The heavy use of the Gordon and Esk generators in previous “wet” seasons prevented their associated lakes from being properly replenished, a major contributory factor to them having very low levels today.

Maybe the hydrologists at Hydro Tasmania should turn off Gordon and Esk and hide the keys until the spring, now that Basslink is available to import any shortfall in electricity.

The Return of Basslink

The Basslink interconnector between Tasmania and the mainland was restored to service at 12 noon EST on 13th June 2016, and immediately started exporting electricity to the mainland. Is Hydro Tasmania repeating the mistakes of recent years, making money by running down hydro storage reserves? This post says no, at least not yet.

Monday 13th June 2016 was a public holiday, so I present here the data for 14th June, a working day with a relatively normal demand cycle. The following figure shows 5-minute demand and supply data for the 14th June 2016. The demand was obtained by summing electricity supply over all Tasmanian generators, and subtracting the metered MW flow of Basslink. For comparison, the same data are also shown for roughly the same day in 2015.

14th June 2016:

Tasmanian total supply (in blue) exceeded demand (in red), so Tasmania was an energy exporter, especially at night. The figure above also shows the outputs of key hydro basins, each of which has large but currently highly depleted water storage reservoirs. The Derwent basin appears to be operated in a “baseload” mode, probably with most water coming from its small storage reservoirs, many of which are spilling at this rainy time of the year. The Gordan and Esk generators were mostly OFF, as they should be at this time of year, in order to allow their storage lakes to replenish, in preparation for the forthcoming “dry” season (spring/summer).

18th June 2015:

The data shown above are typical of working days in May and June 2015, showing the daily roller-coaster ride that the Tasmanian electricity generators were subjected to in those (and other) months. The Derwent generators operated in what appears to be their normal “baseload” mode, but the Gordon and Esk generators were turned on to a high output at around 6 am, as must have been several others, as Tasmania switched from being an importer at night to a major exporter for the rest of the day. On that day (and many others) Tasmania unnecessarily traded its water storage in the Esk and Gordon reservoirs for income.

Some commentators have tried to defend Hydro Tasmania, pointing out that Tasmania switched to being a net electricity importer later in 2015, but importing electricity in the “dry” season does not replenish water storage reservoirs. The problem does not lie with the net import/export of electricity, it lies with the net gain/loss of water. Water management should be the boss of money management for the hydro-dominated electricity system of Tasmania.

Conclusions

So far, Tasmania has used the re-connection of Basslink to export a modest surplus of electricity, probably due to a combination of relatively mild (lowish demand), recently wet (spilling small hydro storages) and currently windy weather, and is acting prudently to replenish its highly depleted large water storage reservoirs.

Misuse of Basslink in 2015

Basslink has just been repaired and re-connected ( http://www.hydro.com.au/about-us/news/2016-06/basslink-return-service ), hence this short review of how Basslink was used in 2015.

From an electricity systems engineering point of view the connection via Basslink of Tasmania and the mainland is a marriage made in heaven. The mainland has excess generating capacity in winter, exactly when Tasmania has its peak demand, together with the need to conserve as much water as possible for the forthcoming “dry” season. The mainland also benefits from access to Tasmanian electricity in summer, allowing it to have less of its own reserve capacity to deal with heatwave spikes in demand and/or unexpected loss of supply equipment.

Thus, one would expect mostly low flows of electricity via Basslink in spring/summer, and high flows of electricity from the mainland to Tasmania in autumn/winter. The reality in 2015 was very different.

Archived electricity data is available to the public back to May 2015, allowing us to see the extraordinary pattern of electricity generation and Basslink flow that occurred between May 2015 and the time of the break in December 2015. The following figure shows the daily average, minimum and maximum number of Megawatts, summed over all Tasmanian generators, since May 2015:

After Basslink Loss

The loss of Basslink can be seen clearly in December 2015, an event which forced Tasmania to operate in “island” mode, meeting the daily demand cycle between relatively narrow minimum and maximum limits.

Before Basslink Loss

Prior to the loss of Basslink the Tasmanian electricity system performed a daily roller-coaster ride, especially between May and August, with daily maximum generation being around double the daily minimum. Note also that the daily maxima of around 2000 MW can only have come about from every available generator operating at maximum output capacity, an extraordinary thing to do during the “wet” season, when the large water storage reservoirs should be allowed to replenish, with their associated hydro generators turned off. Tasmania was generating a substantial surplus of electricity for export between May and August, when it should have been importing.

From October 2015 Tasmania switched to being a major importer of electricity, but still had to generate around 700 MW on average of its own electricity, further depleting its hydro reservoirs, not helped by the relatively dry spring weather.

Conclusions

The data shown above are a smoking gun, revealing how basic principles of electricity systems engineering were abandoned during 2015.

Status of the Derwent Hydro Basin in Tasmania

The previous post gave a high level summary of the current status of the Tasmanian electricity system, as it deals with the continued absence of Basslink, and with the approach of winter, the peak demand period. The Derwent hydro basin was identified as an area where there might be unnecessary over-consumption of water, as gas and diesel generators lie idle. This post looks at the Derwent hydro basin in more detail.

The Derwent hydro basin is divided into an Upper region, with substantial water reservoirs feeding four power stations in that region, and a Lower region with six power stations along the River Derwent. The water reservoirs form three groups, Lakes St. Clair and King William in the West, a number of small lagoons and ponds in the centre, and Lake Echo in the East. The following figure shows the water storage history for the past six years:

The figure above shows that two of the groups of water storage have relatively normal levels for this time of year, the exception being Lake Echo, which has only around half the amount of water it should have.

With the aim of understanding the Lake Echo water level I have extracted the daily average number of MW generated by Lake Echo Power Station, and the downstream Tungatinah Power station, since May 2015. The data are converted to capacity factors, and plotted in the following graph:

The Lake Echo Power Station discharges its water into Dee Lagoon, which has the following history of water level back to July 2015:

Source: http://www.hydro.com.au/system/files/water-storage/Web_Lakes_DEE.pdf

There is a very clear correlation between the Lake Echo Power Station MW output and the level of Dee Lagoon, suggesting that the power station only draws water from Lake Echo when it is generating electricity, at least since July 2015. The recent jump in the level of the Dee Lagoon matches in time a recent spell of heavy rainfall in the area.

Dee Lagoon feeds water into Tungatinah Lagoon (the local water storage for Tungatinah Power Station), which also gets water from other sources, and which has the following history of water level back to July 2015:

Source: http://www.hydro.com.au/system/files/water-storage/Web_Lakes_TUNGATINAH.pdf

The MW output of Tungatinah Power Station appears to be driven by the level of Tungatinah Lagoon, especially when water is nearly or actually spilling.

Conclusions

There does not appear to be a single major event or cause of the very low level of Lake Echo, but rather it is probably due to the cumulative effect of a bit too much consumption relative to inflow over the past few years. In particular one has to ask why electricity was generated from Lake Echo for export via Basslink during July 2015, from a reservoir that was in need of replenishment?

The output from Tungatinah Power Station, which gets some of its water from small reservoirs, has the expected seasonal variation, high in the “wet” season and low in the “dry” season.

 

Tasmanian Electricity System Status Report #01

Date: 8th June 2016

Purpose: This post provides a status report on the electricity system of Tasmania, mainly from a systems engineering point of view.

Summary

• Basslink is still down
• The Gordon and Esk hydro basins have at last begun the long process of reservoir recovery, after the reckless and unnecessary over-consumption of water of recent years
• The Derwent basin is still generating large amounts of electrical energy, whilst gas and diesel generators are idle, slowing the recovery in its lake levels
• Peak demands of almost 1600 MW have been seen in recent days as the weather has turned cold and frosty, and the AEMO had to issue a formal notice of low reserve capacity on 1st June

Recent Generator Output

The following series of figures shows daily average MW outputs of each hydro basin and of wind, gas and diesel generators from 1st March 2016, together with (for the hydro basins) the same data for 2015 (from 2nd May only). The fourth figure in the series shows the daily average, minimum and maximum demands, deduced from 5-minute generator data.

Figure above: The Gordon and Esk reservoir generators have at last been able to go to almost zero output, which they should do during every rainy season; the 2015 data (dots) shows how badly this “rule” was violated last year. In contrast, the Derwent basin is generating around 350 MW, preventing any strong recovery in its highly depleted reservoirs (see later for hydro storage data).

Figure above: The three small hydro basins are generating substantial amounts of electricity, probably reflecting high water inflow in recent weeks.

Figure above: Gas and diesel generators played a substantial role during March and April, but have been largely idle in recent weeks. Why have they not been used more in May to allow the Derwent hydro basin storage to recover more quickly?

Figure above: Demand rose as expected as the weather cooled, with peak demands of nearly 1600 MW in recent days. The AEMO issued a low reserve notice on 1st June (a low wind day): http://www.aemo.com.au/AEMO%20Home/Market%20Notices/0053574

Hydro Basin Storage

The following figure shows the data from Hydro Tasmania on the lake water storage levels, summed over the lakes in each basin. The figure also shows how dangerously low the hydro storage levels have become, by projecting them forward in time with no water inflow (as happens during droughts) and 40% capacity factor generation of electricity, giving around 1GW of electricity, but only until spring of this year, after which only the Esk reservoir would remain:

Conclusions

High water inflow, and new gas and diesel generation have allowed the Tasmanian electricity system to continue to operate without Basslink, with some recovery in reservoir storage levels, but why install gas and diesel generators and then not use them fully to give more reservoir recovery, in the Derwent system in particular?

 

 

Tasmanian Electricity Generation: May to October 2015

Previous posts have dealt with the storage levels of hydro reservoirs in Tasmania, a subject of great interest due to them being at very low levels at the same time as the non-availability of the Basslink interconnector. This post attempts to shed some light on the question of why the reservoir levels are so low, by looking in detail at what happened during the 2015 “rainy” season, May to October.

In the autumn/winter rainy season the large reservoirs are normally allowed to be replenished, with most electricity coming from the small reservoir and run-of-river hydro generators. The replenishment allows the large reservoirs to take over the generation of most electricity during the “dry” season, spring/summer, in particular during periods of drought.

The water inflow to the reservoirs in 2015 can be seen from the following figure, produced by Hydro Tasmania:

Figure source: http://www.hydro.com.au/system/files/www.hydro.com.au/News/Energy_Supply_Plan_Update_-_07_March_2016.pdf

Water inflow was abnormally low in springtime, especially from October to December 2015, but was mostly higher than average earlier in the year, especially in May. The healthy water inflow to August 2015 should have allowed substantial replenishment of the large hydro reservoirs, but the data on water storage and electricity generation tell a different story.

The significance of the reservoir storage data has been discussed in previous posts and is shown again in the following figure:

The following set of figures shows the daily average number of MW generated by each hydro basin, and by wind and gas generators, for each day from 2nd May to 31st October 2015, with the total output over all generators shown in the last one:

Gordon Basin/Power Station

This large reservoir generator averaged around 200 MW output during most of the 6 month period, and produced a total of 880 GWh of energy, whilst its reservoir level remained almost constant. The figure of 880 GWh is important as it indicates roughly how much storage the reservoir might regain in 2016, if the power station produces no electricity during the 2016 “wet” season. The Gordon reservoir could have been replenished by 880 GWh in 2015 if it had been allowed to, there is plenty of reserve generating capacity on the mainland between May and September (a low demand period), with Basslink available at that time to export it to Tasmania.

Esk Basin (Poatina Power Station)

This basin also produced an average of around 200 MW, and a total of 800 GWh of energy, with a small rise in reservoir storage. The Esk reservoir could have been further replenished by 800 GWh in 2015 if it had been allowed to.

King Basin (John Butters Power Station, 143 MW nameplate)

This small reservoir system produced mostly a steady average output of around 100 MW, before falling away in the dry months of September and October.

Pieman Basin

This small reservoir system has 484 MW of nameplate capacity hydro generators, which produced at almost full capacity until the dry period at the end.

Derwent Basin

This basin has intermediate sized storage and managed to produce around 200 MW until the dry period at the end, whilst at the same time increasing its water storage significantly.

Mersey Basin

This basin has mostly run-of-river generators, which produced around 150 MW on average until the dry period at the end.

Wind

The nameplate capacity of the two wind farms in Tasmania is 308 MW, which came close to being achieved on several days, but the “roaring forties” were often the “whispering forties”, with very low wind output on many days.

Gas Generators

The gas generators (Bell Bay and Tamar Valley) were resting during this period, before their starring role after the loss of Basslink in December 2015.

TOTAL GENERATION

The plot of total generation reveals the mistake made by Hydro Tasmania: the large reservoir (Gordon and Esk) systems produced more electricity than was needed by Tasmania between May and August, so in effect a prudent replenishment of their reservoirs when it was raining/snowing was traded for income from the export of electricity to the mainland. The Basslink capacity was there at the time to allow zero electricity output from Gordon and Esk. The water storage history suggests that the same mistake was made at Lake Gordon in the 2014 “wet” season.

At the (dry) end of the period there was a substantial shortfall in local supply, with the balance coming from Basslink imports, until its loss in December, which forced further depletion of reservoirs.

THE FUTURE

The 2015 “wet” season figures of 880 and 800 GWh of energy production for Gordon and Esk provide estimates of how much gain in water storage is achievable during the 2016 “wet” season. The actual gains in water storage in 2016 will set the figures for the maximum amount of electricity that can be generated from those reservoirs in the subsequent “dry” season, on the assumption that a decision is made to stop further depletion of the reservoirs (see the previous post).

Future posts here will report on what is happening in Tasmania in 2016.