OzEA -- Modelling Cover Page

The main strand here is to construct and examine models that match a supply (composed of renewables, dispatchable fuelled generators and storage) against given demand scenarios.

Abstractions and Simplifications

To get started we must abstract and simplify brutally. Once we have a working model we can incrementally develop it to incorporate necessary aspects that were butchered in these beginnings. This is our approach to building useful models.

Simplification #1: First, in the real-world operation of the electricity grid / market, it is necessary to predict into the future; to predict the demand in the coming hours and days, and to predict the wind resource. On this site, and into the foreseeable future, we dispense with this real world complication. We will work with 'known' demand, wind farm output, and any other such curves as necessary. This will generally be achieved by using historical data, especially 2003 - 2010.

Simplification #2: The real electricity grid is dispersed in space (i.e. geographically) and involves a large and complex transmission network, which limits how much electricity can move along certain wires. We often ignore the transmission, but of course it much be accounted in any final reckoning.

Simplification #3: The real electricity grid must manage changes in demand and supply more-or-less instantaneously. At the level of seconds and minutes we assume that the technology and infrastructure required to do this is a seamless black-box. We work here at the hour-to-hour level.

OzEA_MI0002

Neil Howes
Subject: modelling hydro input and wind data
Date: 2010-05-24 (at 10:00:17)

The summary of data at this site;
http://www.windfarmperformance.info/
has data for Sep2009 to April2010.
It is easy to only include the 868MW from SA, but realistically with a lot more farms being built especially over a larger geographic area the SA performance is likely to be more like today's 16 wind farms connected to the NEM grid. Ramping changes for the aggregate wind farm outputs seem to be slow(12-24 hours from 0 to 100%)

Presently SA is getting hydro and pumped hydro storage via the Snowy in which SA owns equity(10%?). On this basis surely we should be considering for modelling at least 380MW hydro(10% of snowy capacity) and 60MW pumped storage(10% of present Tumut3 pumping capacity ). Similarly hydro would be contributing about 6% of SA yearly demand.

fc - yes, it's a nice site. Yes, we surely are considering many things - but we start with the most simplified of models and work up.

OzEA_MI0003

Neil Howes
Subject: hydro
Date: 2010-05-24 (at 11:31:19)

Oops when I checked with snowy hydro, SA doesn,t have any equity. However hydro does contribute 5-6% of total electricity used by NEM so we probably should be considering this a starting point, ie need to supply the other 95% from within SA or contribute to building in other states/territories.

OzEA_MI0005

Neil Howes
Subject: capacity factor for wind
Date: 2010-07-12 (at 14:41:14)

The largest Australian wind farm operator, Infigen, reports in its 2010 annual report to ASX, that its 508MW of Australian farms( including in SA and NSW) has a capacity factor of 36%. This is similar to their US wind farms( 1000MW). I was wondering where the 30% capacity factor figure being used came from?

OzEA_MI0006

Barry Brook
Subject: re: capacity factor for wind
Date: 2010-07-12 (at 15:33:26)

Neil, it's described briefly here:

ANALYSIS

The SA demand for 2009 averaged 1.5 GW (1,538 MW).

The wind farm data constitutes 340 MW of installed (nameplate) capacity, and for 2009 produced a mean of 104 MW (indicating a Capacity Factor of around 31%). We multiply this up to simulate higher levels of installed wind power.

OzEA_MI0007

Neil Howes
Subject: capacity factor for wind
Date: 2010-07-15 (at 10:40:12)

Barry and Francis,
If all the modeling is based on 30% capacity factor ( derived from the 5 wind locations for one yeas data ) that OK for our first modeling, but surely we should look at either all of the sites operating in SA or all of the sites in Australia.
We also need to understand that capacity factor depends upon design criteria, turbine height and price as well as location. We have some flexibility in swept area/generator capacity, location( the first farms were built close to electrical transmission lines) and future turbines are likely to have permanent magnets and higher hub heights( capacity factor in US has been rising over last 20 years).

It may seem trivial to argue about capacity of 30% or 31% or 35% but this has a big effect(x3 to x5) on peak load shedding above demand, relative to total output, the lower the capacity the higher shedding required.

OzEA_MI0008

Francis
Subject: Re: capacity factor for wind
Date: 2010-07-15 (at 11:59:49)

Yes, absolutely. The data we have been working with is "Round 1"; we are working on "Round 2" data now, including especially further Wind Farm Generation data (as noted on that data page). When the Round 2 data is ready we will update the various analyses.

Also I agree that sensitivity to the Capacity Factor is a worthwhile consideration (analysis), but am not game to offer a time frame given the state of the OzEA 2do list at the moment.

Name:*
Email:*
Website:
Subject:*
What is the longhand for 'Oz'? :*

MODELS

Abstractions and Simplifications

Post Comment: