Oz-Energy-Analysis.ORG -- The First Story

We live in time of great concern about the impacts of human activity on Mother Earth, and for the future of humanity in this technological age.

There will soon be some seven billion people living on the planet, and this is reliably predicted to level off at around nine billion later this century. Given that less than a third of us live at least moderately affluent lives, and given that affordable electricity is central to prosperity, it can be roughly estimated that later this century the world may need to be making around four times more electricity than we are now, perhaps more.

Current world electricity use averages at around 2 TW (2,000 GW; 2,000,000,000,000 Watts). Four times this is 8 TW. To achieve 8 TW by later this century, any of the following, or some combination, would suffice:

Let us focus in on a more specific problem - one that we can at least try to address in a scientific way.

First, it is important to recognise that looking ahead forty or fifty years is really difficult. Changes in technology, changes in the way people think and live, changes in the political landscape - these can all happen in ways that change everything. While many technological developments progress at a reasonably predictable pace, over time there are breakthroughs, there are 'left-field' technologies and solutions that change the problem. The further ahead one attempts to look, the more likely that one is simply engaging in crystal ball gazing.

Second, trying to solve the whole problem for the whole world in one go has a certain appeal, but it is not very practical.

Our Problem

Our problem is to describe what would be involved in running Australia (in particular South Australia) predominantly on renewables.

In order to focus on this problem, we exclude others from consideration. While we have various views on climate change and nuclear power (to mention the two big ones), you will not find these laboured here. There are other places for those discussions, including here.

The Answer

Of course we don't have the answer right now. But we can describe what the path we are following looks like from where we are now.

The big problem with renewables, particularly wind and solar, is that 'the sun don't always shine and the wind don't always blow'. That is to say, these are variable sources of power, and we want to introduce them into a system that is currently "electricity on demand".

The variability of renewable power can be managed in various ways: (i) by having different renewable technologies in different places (so that if one source isn't making much electricity, others might be), (ii) storage (such as pumped hydro, thermal storage, and chemical conversions), (iii) demand management, and (iv) backup systems (gas turbines, probably).

But wait, I hear you cry, there are renewables such as geothermal and ocean currents that -are- regular. Which is (sort of) true, but are these technologies tested and ready to roll? And how much power can they produce?

The next big problem is cost. While people might be prepared to pay a bit more for electricity (and for the things that use electricity in their production - which is just about everything), massive price hikes are unlikely to be acceptable to society at large. While we do not start with a full economic analysis, in all the work we do on this site we have these economic aspects in mind. Before policy makers can make decisions, before investors put money into infrastructure, options need to be costed.

So, for this project, on this website, we are developing and communicating the best understanding we can of how a predominantly renewables electricity system might be built in (South) Australia.

And don't forget the Transmission lines

In talking about electricity systems it is easy to forget about the network of transmission lines that moves the electricity around. These are costly infrastructure that need to be included in any serious thinking about how the grid might be configured.

The issue of transmission can cut both ways in relation to renewables. For example, placing solar farms out in the desert requires that suitable transmission lines be built to transport the electricity to the city (or wherever else it will be used). Another example: placing solar panels on the rooftops of suburban homes (which is an expensive way to make electricity in any case) does not load up the transmission system as the electricity is being generated near to where it is used.

Open Science

Whatever else we achieve here, we are performing an experiment in open science. You can see and question the data we are working with, the analysis of the data, and the other aspects of this work as it proceeds. You can ask questions, challenge assumptions, make suggestions and contribute in whatever way your skills and experience allow. Of course some discussions will be technical, but it is our goal and our commitment to be as clear and uncomplicated as we can.

Most of the pages on this site are in a blog format, where discussion and comment follow the material presented. There are Commenting Rules that you need to appreciate and respect; these are necessary to keep the conversations focused and useful.

How this story unfolds ... will be an example of open science as best we can manifest it, and dependent especially on your support. Just keeping an eye on what happens here and talking with others about what matters to you is a valuable contribution. Making suggestions and asking questions where and when you can contribute is also important.

OzEA_S10002

Neil Howes
Subject: congratulations and scope
Date: 2010-05-17 (at 10:15:59)

Barry,
I congratulate you on setting up this site. I am not sure what you are thinking of though in the scope, are you wanting to consider renewable energy solutions for all of Australia, those regions presently connected to the NEM grid or just what can be done in SA?
Since we have some good data on 16 wind farm locations connected to the NEM grid this would seem to be a good starting point, allowing for estimates of solar output anywhere reasonably close to this grid as well as existing hydro dams that are also connected.
If the immediate goal is to see what would be required to supply 20% or 100% of SA's stationary energy needs, surely we should be considering a grid wide solution, and not confining solution that can only be built in SA. Is this what you were thinking?

OzEA_S10003

Barry Brook
Subject: Re: Scope
Date: 2010-05-17 (at 13:43:33)

Neil, the idea is to model the NEM, but this will be a later stage. For the first step, in trying to reduce the problem to its fundamentals, we are going to look at SA as an entity, in which it has generators, demand, and interconnectors. So Vic and NSW will effectively be treated as 'black boxes' that can supply or receive energy from SA, without having their components modelled. The limit is currently determined by interconnector capacities, and so one of the early questions we want to ask is the trade-off between expanded IC capacity and more storage/backup within SA itself.

So what you're proposing is definitely on our short-term radar, but first, we'd like to get our 'eye in' with a simpler (albeit more abstract) problem.

OzEA_S10004

Gene Preston
Subject: adequacy of solar and wind energy
Date: 2010-05-18 (at 00:49:43)

I was asked to comment on where we stand on renewables. I have been in power system planning for about 40 years. I currently do transmission system studies for wind generators. Here is my take on where we stand at the current time with solar and wind power.

A recent headline in the LA Times stated that solar could supply 25% of the world's energy by 2050. I would estimate wind will do about the same. However, we have a need for even more power than just 25% from solar and about the same from wind. Why are we not forecasting even higher percentages from these sources?

Dr Chu, US energy secretary, has recently stated that current non subsidized prices for solar power is about four times too costly. He expects the next factor of two drop to occur as the solar industry matures. However, Dr Chu says solar needs to drop by another factor of two to really make an impact on the energy supply. The high cost of solar is why its development has been slow. At this point there is much uncertainty whether solar power will ever come down to a price level that will be popular and affordable without subsidies. If such a low price for solar ever happens, you can think of it as a technological miracle, as Bill Gates calls it, when he is referring to other technological miracles that have happened in our lifetimes. Large amounts of economical solar power being constructed soon would require a miracle. I would not bet my future on it happening.

Wind power is reaching its technological maturity and would grow much more quickly (while wind still has US tax subsidies) if the transmission systems were adequate in windy areas. Building the necessary transmission network is going to take time, cost a lot of money, and receive a lot of opposition from land owners and environmentalists. Wind also creates an electrical stability problem as it becomes larger and larger with respect to the rest of the system. These problems with wind power are likely to limit wind to probably no more than 30% of our total energy by 2050. If affordable energy storage becomes widely available, then wind could increase to higher percentage levels. Energy storage may or may not be affordable in the future. Can we bet the world on it being available soon. What happens to us if it doesn't develop as we hope it will?

So where does that leave us? Neither solar nor wind is projected to produce enough energy to simultaneously address the climate change problem and the peal oil problem. This means we will continue to have a high reliance on coal under the current solar and wind plans and projections. We need to do more in developing long term energy resources. What do you think will happen if the world becomes energy deficient? I think the current renewables plans are already causing the world to become energy deficient and the economic consequences are starting to be felt and will eventually be devastating if do not correct this energy shortage problem soon.

OzEA_S10005

Francis
Subject: Re: adequacy of solar and wind energy
Date: 2010-05-18 (at 06:17:55)

Nice comment - thanks Gene. But, the last paragraph... this is tricky. I agree that the world needs energy, and perhaps, in the context of this high level scoping discussion, the point is well made. However, I really want to keep away from the "climate change problem", and the here unstated 'need for nuclear'. Both of these issues are immensely problematic, discussed in length elsewhere, and can only act as distractions here. This is stated in the story above, and in the commenting etiquette; I'm not sure how we could make this point much clearer (other than by getting into exactly those discussions). This site is focused on renewables, and on developing an Open Science model. While I have no particular argument with the 'conventional wisdom' on renewable penetration limits, it is just talk to me at the moment; what we really want here is to get quantitatively at the guts of this issue.

OzEA_S10006

Sam
Subject: Scope and objectives, and OS format
Date: 2010-05-18 (at 11:20:36)

First of all, congratulations on setting up this site - a fantastic initiative.

I'd like to see some discussion that sets the objectives for this research in a more concrete fashion. As far as I can tell, at the moment the goal is just to "describe what would be involved in running Australia (in particular SA) on renewables". This is a fine goal, but it isn't a quantitative target or a defined scenario, which makes it difficult to make a valuable contribution to this site instead of just a general comment.

Obviously the intent is to have a set of scenarios with varying levels of renewable supply (mentioned on the models page). Any chance we can have a forum set up where contributors can debate what those scenarios should be? Step one for me would be to set quantitative objectives (e.g. 20% renewable by 2050), and also start to identify the key constraints (financial and institutional) and uncertainties associated with each scenario. We also need to narrow down which renewable technologies will be considered to achieve these targets, and which will be left out for the purposes of this study (is this the purpose of the "generation technologies" page in the discussion section?).

This process would also help with data collection, as it should inform us where the gaps in knowledge are, and how data can feed into the models as they are created. Some of this has probably already been done, but a webpage keeping us up to date with the scenarios would prevent us all from re-inventing the wheel.

The open science format might also benefit from a 'what's hot' sort of page, where contributors can help out particularly with what your group is currently wrangling with. In that way we could be useful when we are needed most, rather than just making general comments with no real idea of whether they are of any use. I see there is a "what's new" page that tells us what is trending in the discussion, but it would also be good to know what is currently needed at your end, rather than just what is popular at our end, or else we'll degenerate into an endless cycle of coal/gas/nuclear/wind/solar/ etc debate.

fc - yes, yes and I agree -- working on it all!

OzEA_S10007

Neil Howes
Subject: Targets for SA renewable energy
Date: 2010-05-19 (at 14:34:26)

According to AEMO site, SA has a range in demand of 900MW to 3300MW in summer and 1000MW to 2300MW in winter, with an average demand of about 1600MW.
Since SA has 900MW installed wind capacity this means that wind power at present is contributing 300/1600 or about 19% of on grid stationary energy, so we probably need to look at a more ambitious target of say 45% and 90% for renewables.
A mix of 65% av wind(35% capacity factor), 25% av solar CSP(33% capacity factor with 24h storage) and 10% av OCGT/CCGT(natural gas at 15% capacity factor) would seem to be a good match for SA summer peak demand( and peak solar output). For the 90% target this would require no new NG capacity, a tripling of present wind capacity and a small increase(or perhaps none) in transmission to Snowy and TAS hydro( total of about 1000MW).
Do these numbers seem reasonable?

OzEA_S10009

Francis
Subject:
Date: 2010-05-19 (at 15:52:28)

Hello Neil; the SA demand for 2009, with the wind output overlaid, can be examined here on our data viewer.
We are not approaching this work in the sort of way you suggest, at least not yet. This work starts with a very high level view -- the basic equation is: demand - generated_supply - change_in_storage = 0. Let's strike out and call this ozEA's fundamental equation. So, at every time step, we are in one of two situations; either the demand is being satisfied by a combination of generated_supply plus drawing on storage resources, OR, the supply is being taken up by the demand plus the recharging of storage resources. Of course a whole gamut of real world issues is being abstracted away here -- that is the physicist side of me.

In the system we have now, demand is pretty much doing its own thing, and it is the job of NEM to make sure that the other two variables (generated_supply and use of storage) are maintained and adjusted in a way that results in the uninterrupted satisfaction of demand. There are a whole lot of factors involved here; especially (a) the slew rates (how quickly resources can be turned on / off / up / down) of generators, and (b) cost.

We want to model an increasing level of renewables (mostly wind, but also will be considering some solar) in a more-or-less continuous fashion. At each point along the way our ability to balance the system with the generated_supply is compromised a little further. I realise this is all very simple sounding. Here is where it gets starts getting more difficult. We have three tools available for enforcing ozEA's fundamental equation; we have generators with various slew rates and efficiencies; we have storage, and we have demand management. Any one of these can fill the gaps -- in principle (well, it'd be a brutal form of demand management without the other two). What we need to model and explore is what the possibilities, problems, and costs are AT increasing levels of renewable penetration. And we mustn't forget the transmission aspects.

So, we are getting going on these questions without including the spatial aspects, and then moving onto more explicit treatment of these transmission and spatial smoothing issues later in the year. I could write out more and more paragraphs, becoming more and more speculative about where we might end up, but will stop here. It's a big problem we are working on; we will solve it by building up from these basics, and following the paths that the data, logic and analysis dictate.

OzEA_S10010

BraveNewClimate
Subject: Abstraction and simple stories
Date: 2010-05-19 (at 17:50:05)

The key point that Francis is getting at, I think, is that we have to abstract this problem before we can get to grips with it. Quite brutally abstract it, in fact - build a skeleton, and then gradually add flesh and bones. Staying anchored to the fundamental equation helps keep focus.

We pretty sure that the first systems models we come up with are going to be grossly simplified, but we aim to learn-by-doing with these. That is, if we can't understand how to make a 'toy' example, consisting of a 'black box' set of generators, an untunable demand, and then some increasing penetration of wind, modelled without spatial detail, we haven't got a hope of really understanding the core messages of more complex analyses.

So Neil, what you propose is a good storyline, and it would be great if you could flesh it out a bit more. Actually, perhaps, at this stage, it is better viewed as more of an endpoint - for which it would be interesting to see how we might arrive there.

But perhaps we all need to be more constrained in our thinking about the first stories. That is, what is the simplest story we can imagine that still tells us something useful, if we work through its plotline?

OzEA_S10011

Neil Howes
Subject: wind data
Date: 2010-05-20 (at 12:42:19)

Surely we should at least look at all of the data we have for SA wind farms, I get 953MW capacity in operation and 312MW under construction; MtMillar(70MW),Cathederal(66),Clemens(57),Snowtown(99),Hallet1(95), Hallet2(71),Wattle Pt(91),Starfish(34),Lake Bonney1(80), L Bonney2(159),Canunda(46) and Pt Lincoln(75) with Lake Bonney3(39), Waterloo(111) and Hallet4(132) under construction.

The data from Heat of Bight may give a good idea of what to expect if a larger farm is built >500Km to the west, although this is a small turbine(3kw) that may have different performance.

I think the key issue is not what minimum to expect( as we will need essentially 97% back-up, which we have as OCGT and CCGT. The key issue for wind variability is the ramping rates and the capacity level that can be load shed. With the 18 farms connected to AEMO, we could reject any output above 70% capacity with about 0.5% loss of yearly output. This sets an upper limit of storage required (70% wind capacity-lowest demand) of about 1,100MW. If we need this grid capacity to meet peak demand ( at present) this doesnt mean any more interstate grid infrastructure to accommodate 3000MW of wind capacity.

I dont see any point in modeling wind output, grid capacity and economics unless we have some assumptions of what wind capacity can be load shed.
We also have to decide what is a good value for lowest demand, about 900MW seems to be the lowest but if we use the figure of 1,000MW how many MWh will have to be load shed( ie how often would wind output be >70% capacity and demand below 1000MW?).For the AEMO grid connected farms spikes above 70% capacity output seem rare(1-2/month) and of low duration (3-18h). Since the SA potential wind sites cover about the same distance as present wind farms( 1200Km) may expect similar situation.
It may make sense to be prepared to load shed >60% capacity when demand is <1000MW if this only represents (0.1 or 0.5%??) of output, as this would allow a much larger wind capacity to be used.

OzEA_S10012

Francis
Subject: Re: wind data
Date: 2010-05-20 (at 14:41:05)

Neil, I think I'm with you now. To recap; the model here is (i) lots of wind - so much that sometimes there is more than we can use, (ii) more-or-less full backup with gas (both open cycle and closed cycle) to cover the (many) periods when wind does not cover the demand.

As you indicate, this is a cornerstone scenario -- others are extensions to it. So, I agree, this needs to be fleshed out, both as a story and as a model. I'll get on with this shortly; if you, or anyone else, wants to be involved in writing / editing / constructing the head post for this, then please get in touch with me.

OzEA_S10013

Neil Howes
Subject: pumped storage
Date: 2010-05-21 (at 12:15:07)

Peter Lang's analysis of a 9GW(7.6GW pumping) project in Snowy( estimated cost 6-12Billion) would be a good starting point. SA would at most require 2GW storage( but probably <1100MW), so would need 11-22% of such a project(1.3-2.6 Billion, about the same capital cost as new CCGT, but a lower fuel cost.
Another alternative would be a second Bass-link(500-1000MW) to access TAS hydro.

Since SA is already a partner in Snowy Hydro the first option would seem more likely, especially as NSW and VIC are going to also need storage for their wind capacity as it grows, and should be cheaper.

fc - these sorts of numbers don't mean much to most; not really "story".

OzEA_S10014

Peter Lang
Subject: re Neil Howes' comment
Date: 2010-05-22 (at 18:29:17)

Two comments relating to Neil Howes' comment at 12:15 on 2010-05-21:

1. I am wondering if Australia's pumped hydro energy storage capacity has any relevance in the model your are building, at least in the early stages of it, because the energy storage is in the 'black box' beyond the interstate interconnectors. If my understanding is correct, the pumped-hydro energy storage capacity inside the black box would be no more relevant than the generating capacity of coal, gas and wind inside the black box.

2. Neil's reference to my analysis of the potential for a pumped hydro energy storage facility in the Snowy Mountains contains a couple of minor misunderstandings. The project considered should be considered to have a rated power of about 8 GW not 9 GW and the cost is likely to be in the range $10 to $15 billion (see the reviewers' comments). I have done some more work on the analysis for the project and I believe the $10 to $15 billion in probably more realistic than my early estimates. I also agree with one of the commenters on the BNC web site who stated the power station would have to be located underground.

fc - yes, storage is a black box for now - we just need an overview understanding. Some meat in the models page soon.

OzEA_S10015

Neil Howes
Subject: black box
Date: 2010-05-24 (at 11:56:47)

I would think its important to know where the additional grid connection have to be made, is it the NSW or VIC/SA boarder or to Snowy and TAS?
This submission by Diesendorf

http://www.aemo.com.au/planning/0419-0008.pdf

Proposes modifications and some new links between SA and VIC/NSW increasing capacity from 520MW to 4500MW. Some are minor( increasing voltage) others would require new transmission lines. This would seem to be more than enough to export any surplus wind power( up to 5400MW capacity).

Diesendorf is envisaging SA being a net exporter of renewable energy so this is considerably more than is needed for SA to be 90-100% renewable or non-FF.

Would it make sense to use as a starting point the existing 520MW linked at a black box( the rest of NEM) and see what wind and solar could be accommodated by this and how much OCGT/CCGT capacity would have to be retained?

OzEA_S10020

Stephen Gloor (Ender)
Subject: Setting Goals
Date: 2010-06-03 (at 12:21:15)

I think that it is important at the outset to set goals that the renewable solution will be measured by. As far as I can see in reading the site there is no measure of success yet.

Right at the start before any analysis is done I feel that it is important to all agree on what would constitute an acceptable and successful renewable solution with specific measurable goals that renewables would either achieve or not achieve.

At least in this way at the conclusion of the analysis renewables can either be said to be a viable way to power the state or Australia or not with pre-determined performance goals that we can all agree on.

They might be a measured ELCC of 40% or less then 3 days of no renewables for x number of years. I am not saying that these should be the goals just that we need to set them very early in the piece.

What should the performance goals be to measure a successful renewable solution.

OzEA_S10022

Stephen Gloor (Ender)
Subject: Goals
Date: 2010-06-03 (at 16:48:04)

Peter has had the suggestion of going for the lowest cost solution that meets the demand however the lowest solution compared to what? We are excluding nuclear etc so there is nothing to compare the renewable solution with.

I am not sure the lowest cost can be measureable without alternatives to compare it with.

Perhaps we can set the initial goal of 70% renewables for a price that is no more than say 25% greater per kWhr than the current electricity price in SA?

Does anyone else agree that measureable goals are important?

OzEA_S10028

Neil Howes
Subject: measurable goals
Date: 2010-06-08 (at 08:04:49)

Before we can consider possible costs we need to know what is required to reach specific targets such as 50% power low carbon or 90% low carbon. Starting with SA wind and solar seems a good place to begin.

Costs will depend upon what technology is used when it is built where it is built. None of these future costs are certain, its very difficult to predict the cost of a new bridge, road tunnel, a ******* power plant -snip-

rules of language please.

OzEA_S10029

Arthur Robey
Subject: Assumptions
Date: 2010-07-29 (at 00:13:25)

All solutions have assumptions. These need to be exposed, before we start.
Here are some that I have found.
1 Australians will be making the decisions about their energy future.
2 The economy will be able to extend credit for any changes that we make.
3 (subset of 2) We will be able to feed ourselves, if not, capital will have to move from industry into agriculture.
4 Raw material procurement will not be a problem
5 We will be able to get peoples attention, without spooking them into panic. In which case they will go for the quickest solution, coal.

OzEA_S10030

francis
Subject: reply to Arthur #29
Date: 2011-12-29 (at 00:35:57)

It's taken me a while Arthur -- thanks very much for your thinking.

You say at the end:
"5 We will be able to get peoples attention, without spooking them into panic. In which case they will go for the quickest solution, coal."

It's not clear whose 'assumption' this is.

Certainly I agree with the sentiment of the first part, but think your last part is a wild conclusion. Indeed, it's pretty much in contradiction with your fist 'assumption' "1 Australians will be making the decisions about their energy future."

My view is of strong public support for the development and deployment of renewable energy.

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

#1. Scoping the Problem

In the beginning

Our Problem

The Answer

And don't forget the Transmission lines

Open Science

Post Comment: