Why Time Has Arrived For Flexible Demand
- Dec 30, 2019 9:15 pm GMT
- 944 views
Traditionally the power sector predicted demand and dispatched generation to meet it. Utilities, most of whom were vertically integrated and regulated monopolies, maintained a portfolio of plants – baseload, intermediate and peaking – to meet demand as it varied from hour to hour, day to day, and across the seasons. Demand was a given, namely the sum of the load from all electricity using devices in the network. All consumers bought all the kWhs they consumed from the network and paid a regulated bundled tariff.
California’s Duck Curve: It’s getting worse
In the future, we will increasingly be predicting variable renewable generation and scheduling demand to match it. This is likely to happen in more places as the percentage of renewables reaches higher levels, exceeding 25%, 50% or higher levels in many networks in some periods.
South Australian Duck Curve?
Source: Jenny Riesz, AEMO, presented at Future El. Mkt. Summit, Sydney, 18 Nov 2019
It is not uncommon to reach 100% or higher levels during extremely windy and/or sunny periods especially on weekends or when demand is low, in which case the network experiences over-generation – with prices going negative. When this happens, the excess generation must be exported, stored or – as a last resort – curtailed. As described in article on page 14, in places like Texas and California, renewable generation must be curtailed because it exceeds load, because there is congestion on the transmission network and/or due to lack of sufficient storage.
Up to now, grid operators have done their best to use the traditional tools at their disposal, namely turning flexible thermal plants up or down to accommodate the variable renewable generation. This works up to a point but is approaching limits that are either too expensive and/or operationally challenging to maintain reliability.
A well-known example of the phenomenon is the famous California Duck curve, where 13 GW of ramping is required to meet the evening’s peaks on many sunny days. Many versions of the California duck are now routinely experienced in other parts of the world with similar implications. For example, South Australia’s duck curve (left) is already difficult to manage and is likely to get worse.
Terna, the grid operator in Italy, is expected to face a similar fate as the percentage of renewables increases in Italy. Unsurprisingly, Italy’s duck curve (visual below) looks surprisingly similar to those of California and So. Australia.
The traditional method of ramping thermal plants up and down is approaching the limits of its usefulness. Exporting the excess energy and/or storing it are gaining traction but can only go so far. As countries adopt ever more ambitious renewable targets, new ways must be found to balance supply and demand – but what?
One answer is flexible demand. As time goes on new methods must be found to explore how flexible demand can be aggregated and managed to meet variable generation. Concepts such as demand response (DR) and price responsive demand (PRD) as well as schemes to aggregate and manage flexible demand are being developed. These ideas are not new but were historically considered “nice-to-do.” They are now becoming “must-do.”
As it happens, 3 developments are enabling their implementation on scale that was not imaginable in the past:
- Smart meters are becoming the norm in many places;
- Smart tariffs provide incentives to respond to price signals; and
- Smart connected devices are easier to monitor and manage.
The first two are essential, even if slow in the making. The third, sometimes called digitalization, is expected to make it easier for individual devices within the customers’ premises to be remotely monitored and managed. Moreover, enterprises are emerging who can aggregate large portfolios of behind-the-meter devices and using sophisticated algorithms – artificial intelligence and machine learning among them – to optimize how, when and how much energy is used.
Italian duck curve
Source: Giacomo Terenzi, Terna, presented at AIEE Conf, Rome 12 Dec 2019
A large number of electric vehicles, for example, can soak up some of the excess generation during sunny hours of the belly of the duck curve. Perhaps, some of the stored energy in the batteries can be discharged back into the network during the peak demand hours after the sun has set, a practice called vehicle-to-grid (V2G).
Similar things can be done with multitude of other devices with flexible demand including electric water heaters, pre-cooling or pre-heating buildings, making ice and storing it for later use, pumping or desalinating water, shifting electricity-intensive processes to low price hours of the day, and many more.
With a few exceptions, virtually all customer demand has some flexibility. The industry traditionally ignored this either because it was difficult to manage – it still is – or because it was easier to adjust generation. In the future, demand has to learn to tango with variable generation simply because the alternatives are impractical, difficult or too expensive.
This article originally appeared in the January 2020 issue of EEnergy Informer, a monthly newsletter edited by Fereidoon Sioshansi who may be reached at email@example.com