The Summer of 2018 has been a wild one for Class A customers trying to reduced their demand and capture those ever-growing Global Adjustment savings. Summer started early with the heat arriving in early June and it keep creeping back throughout the entire summer. As we head into the fall we at VPPI have taken a few moments to look at the data and make a few observations:
1 Comment
What should you consider before trying curtailment as a strategy?
The cost to curtail building systems, operations or production is unique to each facility. Direct curtailment costs can be calculated relatively easily and for some, such as commercial buildings, it can be relatively cheap to scale back elevators, lighting, cooling, etc. On the other end of the spectrum, a manufacturing operation needs to consider downtime costs, labour costs, restart time and potential delays in customer deliveries. Indirect costs usually include staff time to plan, implement, monitor and report on curtailment activities. Once the costs have been determined then you should assign a cost per curtailment event to help in the decision-making process keeping in mind that you will need to curtail far more than 5 times per year to catch all 5 peaks. How Often Will We Need to Curtail Activities? Global Adjustment peak events are dependent on the weather so the weather patterns of any given year will determine the number of times that you will need to curtail to be reasonably certain of hitting all 5 peaks. Why? The top 5 peaks can be thought of as a moving baseline with the 5th peak being the point of reference. As shown below, the baseline, currently at 19,300 MW, will stay in place until a higher demand is seen. I.e. a baseline of 21,500 MW. With this new baseline any day projected to be less than this will not need curtailment. Years will a relatively cool summer and warm winter will have fewer days that have a chance of pushing a peak out of the top 5. Hotter summers (especially late summer heat) will have more days that have a chance of pushing into the top 5. It is reasonable to expect to curtail between 18 and 30 times per year and increases of extreme heat events due to climate change could push this to 50 or more. How Long do we need to Curtail Activities during a Peak Event? Each hour of curtailment will cost money so the length of curtailment will have a direct impact on costs and will largely depend on: 1. How quickly you can respond to a peak alert warning; and 2. How good your Peak Prediction service is. This will also give you the information you need to accurately assign a cost of curtailment.
2. The accuracy of the Peak Prediction Service is also key. On certain days the demand profile is relatively flat and peaks are a little more difficult to predict so the accuracy of the software that you select is vital. A good service will be able to give you a 3-4 hour window with updates to keep the curtailment window small. Ones with a less robust software or regression algorithm will give you a long 6+ hour window, forcing you to increase the length of curtailment. How Fast Can We Respond to a Peak Warning? As noted above, peaks can change and do so quickly due to weather events which makes peak prediction much more difficult for curtailment than it is for a dispatchable technology such as energy storage.If you have the ability to curtail quickly to a real-time Peak Alert Warning then both the number of times you need to curtail and the length of the curtailment will be shorter. If your facility needs a longer time to start curtailment then you will end up have a larger curtailment window. Summary
Once you've examined all of these factors. You should be able to assess the cost per curtailment event or cost per curtailment hour. Take this number and multiply it by the number of expected events to get a range of expected curtailment costs for the year. Once you know the costs then the decision to curtail or take another path should be an easier decision to make. What’s The Alternative? If curtailment ends up being too costly or only manages part of your peak load, there are alternatives. The most flexible and useful alternative is Energy Storage coupled with a good Peak Prediction Software. It can respond in real-time to changes in weather or demand and can even react within the peak hour. Energy storage can manage your cost without needing to curtail or can work in conjunction with curtailment. Energy Storage can also improve power quality, grid reliability, and help balance supply and demand for the province without adding GHGs. For more information or a free assessment in how curtailment and energy storage can help you reduce energy costs please reach out to us, we’re happy to help! Monday, June 18th was a very unusual day when it comes to Ontario Peak Demand. The IESO predicted peak demand to occur at 12:00 pm EDT but it actually happened 2 hours earlier at 10 am, which was a first for Ontario. Where did the top of the peak go? Peaks occur at times for very specific reasons so why did this one happen so early and unexpectedly? The demand data for the 10 am to 4 pm period is shown below. Looking at the predicted (dotted lines) versus actual (solid lines) data it’s clear that demand for electricity dropped off for a period of about 2 hours between 11 am and 1 pm. This is atypical of what has occurred in the past, so what caused this? There are two likely scenarios:
If the Curtailment Scenario is unlikely, then what happened between hours 11 and 13 to drop the demand? Let’s look at the Weather Radar: June 18, 12:00 pm EDT June 18, 12:30 pm EDT Looking at the Radar Images from the time period of 11 am to 1 pm we see a strong band of thunderstorms and/or rain pass first through York Region (11:00 am), then the eastern GTA at 11:30 am, followed by Toronto (12:00 pm) and finally Hamilton-Brantford-London and the Kitchener Waterloo area at 12:30 pm. The path of the storm/rain follows the drop in demand with the largest drop occurring between 12-1 pm when it went through the larger load centers. Based on the weather at the time, it’s most likely that the storm/rain event passing through the GTA and SW Ontario caused a drop in demand resulting in the disappearance of the predicted peak. Case Solved!
There is still an open question of how much of a role (if any) curtailment played in the demand drop. With the role weather played in this demand event, it is tough to tell right now but this should become clearer as the summer progresses. There are a few key points to take away from this unusual occurrence:
At Virtual Power Plants Inc. we offer a variety of energy storage service plans with peak prediction service to help ensure that your peak usage is kept as low as possible. Contact us for more information.
Now you’ve decided that an energy storage system makes sense, you’ve picked the system so the question now is: Do we Buy, Lease or enter into a Storage as a Service (Shared Savings) agreement?
Before looking at the details of each let’s examine the risk profile of each option
It should be highlighted that Ontario has a unique pricing mechanism for part of the cost of electricity called Global Adjustment (GA) and because of this many of the points raised in this series will not apply to other jurisdictions.
So, you’ve been given a proposal to install an energy storage system for your business that promises large savings. What do we need to look to see if this is the right system and whether it can deliver the proposed savings? For the energy system itself, we will look at 3 key parts:
What Are the Main Points to Consider
When deciding on an energy storage solution there are four main factors which should be fully examined before deciding to proceed.
Does our Facility Have Enough Load? This summer has been an unusually, but not unprecedented, cool and wet summer. In this article, we will be looking at how this has influenced the peak power consumption and increased the likelihood of having a winter peak (or peaks) this year.
To date we have only observed 3 days where the maximum daily temperature in Toronto has exceeded 30 degrees Celsius with two of these being in June. Strangely enough Ottawa has had 4 days and London only 2 days exceeding 30 degrees Celsius, just another indicator of how strange the weather has been this summer. This has had a noticeable effect on peak demand in Ontario with the current top 5 peaks being: Chasing the Peak – Vol. 2 The Cost of Power Interruptions and What Your LDC isn’t Telling You7/6/2017 In this article, we will discuss the magnitude of power interruption problems, what your Local Distribution Company isn’t reporting and the most cost-effective option to mitigate disruptions.
The most interesting thing which I discovered during the writing of this article is that “momentary power interruptions” (lasting less than 5 minutes) are not tracked or reported in Ontario (or the rest of Canada either). This is a little disturbing given that these can be very costly for facilities with sensitive operations such as manufacturing, food processing or data centres. In Ontario, the two main reported reliability statistics are the Average Number of Hours that Power to a Customer is Interrupted (CAIDI) and the Average Number of Times that Power to a Customer is Interrupted (CAIFI). Looking at the data for a couple of LDCs, our system looks quite reliable. This is the first in a series of posts about managing electricity costs in Ontario for Industrial and Commercial customers. To learn more about the Industrial Conservation Initiative (ICI), whether you qualify and how to manage costs please see my earlier post. This post will deal with predicting peaks and how the expansion of the ICI is going to make this harder. Enough of the preamble, let’s have a look at the numbers.
With the expansion of the ICI, the number of potential qualifying buildings is estimated to be over 3500. Every year the number active participants will increase , having significant effect hourly demand. How will this affect the peak? Let’s have a look at the top peak of 2016. |
Terry RasmussenRenewable Energy and Energy Storage Nerd Archives
September 2018
Categories |