Yesterday Ontario’s system operator released a number of reports. Personally these reports provide opportunities to check the performance of the system compared to my expectations, and estimates, and they also provide an indication of how well the IESO adjusted to taking over the responsibilities of the former Ontario Power Authority (OPA) after 2014.
The data released includes:
- Updated Ontario Energy Report website with 2016 1st quarter information, with related electricity report (.pdf), and data file (.xlsx)
- 2016 Q1 Progress Report on Contracted Electricity Supply (.pdf)
- 18 month outlook (.pdf) and related date file (.xlsx)
On first flip through the reports, the graphic that most caught my eye was the 18-month outlook’s “Table 4.1 Existing Generation Capacity as of…” This table shows not only what the IESO considers the capacity, by fuel type, participating in their market, but also what I will call “Capacity Value” and they call “Forecast Capability at Outlook Peak.” This is an important number because it’s used to measure the system’s ability to meet anticipated peak demand. The numbers that caught my attention were a 280 megawatt (MW) capacity of solar, with the forecast capability at peak of 28 MW. The 10% capacity value that indicates is sharply reduced from the 18 month forecast of June 2015.
Let’s ignore the low installed value for solar for a fleeting moment, and concentrate on the reduced capacity value. This June’s 18 month report explains it:
The grid demand profile has been changing, with summer peaks being pushed later in the day. While the contribution of the embedded solar generation displaces load during the daylight hours reducing demand at the grid level, solar power output at the new peak hours is considerably lower. Therefore, the [Solar Capacity Contribution] values at peak dropped significantly and are expected to stay at these levels over the peak period in the foreseeable future. However, solar generation continues to contribute to meeting energy needs during daylight hours.
Solar capacity has now pushed out beyond sunlight hours the peak demand for supply from the grid-connected suppliers. These suppliers were the only ones in the IESO’s world prior to 2015, and there’s little in reporting to suggest the IESO has adjusted its actions based on the growth of distributed generators. Embedded solar has pushed the peak need for supply from the IESO’s grid out of the afternoon “On-Peak” rate hours the Ontario Energy Board (OEB) forces on consumers from May to October, and often beyond the mid-peak hours that are regulated to end at 7 pm year round. This won’t be a surprise to readers of my March 2015 Failed time-of-use electricity pricing in Ontario, but it would, hoefully, be a surprise to the people asking for more solar.
In the past 12 months the IESO’s 18 month forecast changed to show an additional 240 megawatts of solar which delivers only 14 more megawatts for firm supply at the expected hours of peak demand. A solar panel added to Ontario, today, has a capacity value of less than 6% and the reality is if solar continues to be added peak requirement for new supply will move to winter evenings and instead of new solar’s capacity value being close to nothing it will be absolutely zero.
Without a capacity value, the value of output from solar generation will be the market value – which the Ontario Energy Report yesterday revealed to have an arithmetic average below 1 cent per kilowatt hour during the 1st quarter of 2016.
This major change in circumstance is not due to the 280 MW the IESO’s 18-month outlook has, but largely to the solar capacity embedded in distribution networks that impacts the numbers by lowering demand (as less supply is required from the grid). The Ontario Electricity Report, as well as the Progress Report on Contracted supply, show “at end of Q1”, 1,876 MW of embedded solar.
I was curious to check my estimates on supply curtailment against the graphics that occasionally show on the 18-month outlook, and was pleased by the outcome.
From the IESO:
and here are my estimates for nuclear and wind curtailment for the same months:
The IESO’s reporting disguises increased curtailment, visually, by showing only 12-months of history, but the text of the 18 month outlook explains the proportion of wind in curtailment has grown:
The lower demands, high nuclear availability and the increasing amount of wind and solar generation in the system resulted in a high volume of curtailment starting in the fall of 2015. Recent changes were made to the floor prices of solar and wind resources that changed the dispatch order during [Surplus Baseload Generation] conditions. Since wind and solar are now dispatched down before nuclear, increases in the proportion of wind curtailments are observed.
What if there were less nuclear?
The 18-month outlook’s have a set of figures, in Table 4.9, which are from their simulation of, “production to supply Ontario energy demand by fuel type for the entire duration of the outlook.” I find it an interesting table as it indicated how much of a generation type’s output the IESO expects to service Ontario demand, and therefore how much it expects to be dumped on export markets or curtailed altogether. Unfortunately I’ve not figured out how to simplify the jargon here as I show the June 2015 18-month Outlook’s Table 4.9 simulation of supply for the following year (2016), the same for this June’s Table 4.9 (for 2017) and the last reported full year totals (for 2015).
The biggest change in the IESO’s simulations is a 7 terawatt-hour drop in nuclear, which should be due to the refurbishment of a unit at Darlington. Despite the fact wind capacity has continued to grow the IESO still doesn’t expect it will need in 2017 as much as it was required to take in 2015 – and in 2015 it curtailed another 733.5 gigawatt-hours (GWh).
Only one generation type is anticipated to contribute more to meeting Ontario demand in 2017 than it generated in 2015: solar – and I’ve already shown incremental solar capacity will have no value.
The IESO’s simulations anticipate what will replace nuclear output, during refurbishment, is natural gas-fired power, despite the current high curtailment levels, and dumping, of generation from industrial wind turbines.
A last point from the recent data release, which I’ll be expanding on in a future post on my cold air blog. For those who have not yet entirely dismissed the suggestion that the province – and anywhere else – can run on wind, water and sun, and without significantly altering our hydro (water) capacity:
- as 4,103 megawatts of wind and solar capacity are only expected to produce 502 megawatts during peak demand,
- to replace the 11,031 expected from 12,978 megawatts of nuclear capacity the implied need is for another 90,160 megawatts of wind and solar capacity and,
- to replace the 8,530 expected from 9,942 megawatts of natural gas-fired capacity the implied need is for another 69,718 megawatts of wind and solar capacity.
There are problems with this simplistic look at system needs, but I’ll argue it’s overly generous to wind and solar generators to imply 160,000 megawatts of their products could replace 23,000 megawatts of firm supply – but I’ll back up the argument another day.